VOLUME 6 DISEASE CONTROL PRIORITIES • THIRD EDITION Major Infectious Diseases DISEASE CONTROL PRIORITIES • THIRD EDITION Series Editors Dean T. Jamison Rachel Nugent Hellen Gelband Susan Horton Prabhat Jha Ramanan Laxminarayan Charles N. Mock Volumes in the Series Essential Surgery Reproductive, Maternal, Newborn, and Child Health Cancer Mental, Neurological, and Substance Use Disorders Cardiovascular, Respiratory, and Related Disorders Major Infectious Diseases Injury Prevention and Environmental Health Child and Adolescent Health and Development Disease Control Priorities: Improving Health and Reducing Poverty DISEASE CONTROL PRIORITIES Budgets constrain choices. Policy analysis helps decision makers achieve the greatest value from limited available resources. In 1993, the World Bank published Disease Control Priorities in Developing Countries (DCP1), an attempt to systematically assess the cost-effectiveness (value for money) of interventions that would address the major sources of disease burden in low- and middle-income countries. The World Bank’s 1993 World Development Report on health drew heavily on DCP1’s findings to conclude that specific interventions against noncommunicable diseases were cost-effective, even in environments in which substantial burdens of infection and undernutrition persisted. DCP2, published in 2006, updated and extended DCP1 in several aspects, including explicit consideration of the implications for health systems of expanded intervention coverage. One way that health systems expand intervention coverage is through selected platforms that deliver interventions that require similar logistics but deliver interventions from different packages of conceptually related interventions, for example, against car- diovascular disease. Platforms often provide a more natural unit for investment than do individual interventions. Analysis of the costs of packages and platforms—and of the health improvements they can generate in given epidemiological environments—can help to guide health system investments and development. DCP3 differs importantly from DCP1 and DCP2 by extending and consolidating the con- cepts of platforms and packages and by offering explicit consideration of the financial risk protection objective of health systems. In populations lacking access to health insurance or prepaid care, medical expenses that are high relative to income can be impoverishing. Where incomes are low, seemingly inexpensive medical procedures can have catastrophic financial effects. DCP3 offers an approach to explicitly include financial protection as well as the distribution across income groups of financial and health outcomes resulting from policies (for example, public finance) to increase intervention uptake. The task in all of the DCP volumes has been to combine the available science about interventions implemented in very specific locales and under very specific conditions with informed judgment to reach reasonable conclusions about the impact of intervention mixes in diverse environments. DCP3 ’s broad aim is to delineate essential intervention packages and their related delivery platforms to assist decision makers in allocating often tightly constrained budgets so that health system objectives are maximally achieved. DCP3 ’s nine volumes are being published throughout 2015–18 in an environment in which serious discussion continues about quantifying the sustainable development goal (SDG) for health. DCP3’s analyses are well-placed to assist in choosing the means to attain the health SDG and assessing the related costs. Only when these volumes, and the analytic efforts on which they are based, are completed will we be able to explore SDG-related and other broad policy conclusions and generalizations. The final DCP3 volume will report those conclusions. Each individual volume will provide valuable, specific policy analyses on the full range of interventions, packages, and policies relevant to its health topic. More than 500 individuals and multiple institutions have contributed to DCP3. We convey our acknowledgments elsewhere in this volume. Here we express our particular gratitude to the Bill & Melinda Gates Foundation for its sustained financial support, to the InterAcademy Medical Panel (and its U.S. affiliate, the Institute of Medicine of the National Academy of Sciences), and to World Bank Publications. Each played a critical role in this effort. Dean T. Jamison Rachel Nugent Hellen Gelband Susan Horton Prabhat Jha Ramanan Laxminarayan Charles N. Mock VOLUME 6 DISEASE CONTROL PRIORITIES • THIRD EDITION Major Infectious Diseases EDITORS King K. Holmes Stefano Bertozzi Barry R. Bloom Prabhat Jha © 2017 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 20 19 18 17 This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Holmes, K. K., S. Bertozzi, B. Bloom, and P. Jha, editors. Major Infectious Diseases. Disease Control Priorities, third edition, volume 6. Washington, DC: World Bank. doi:10.1596/978-1-4648-0524-0. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. ISBNs and DOIs: Softcover: Hardcover: ISBN: 978-1-4648-0524-0 ISBN: 978-1-4648-0526-4 ISBN (electronic): 978-1-4648-0525-7 DOI: 10.1596/978-1-4648-0524-0 DOI: 10.1596/978-1-4648-0526-4 Cover photo: © Jim Pridgeon/University of Washington. Used with the permission of Jim Pridgeon/University of Washington. Further permission required for reuse. Cover design: Debra Naylor, Naylor Design, Washington, DC Chapter opener photos: chapter 1: © European Union/ECHO/Isabel Coello. Used via a Creative Commons license (https://creativecommons.org/licenses /by/2.0/). Original photo was cropped to fit template; chapter 2: © Joerg Boethling/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 3: © Joerg Boethling/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 4: © Joerg Boethling/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 5: © Joerg Boethling/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 6: © Charles Pieters. Used via a Creative Commons license (https://creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 7: © International Women’s Health Coalition. Used via a Creative Commons license (https://creativecommons. org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 8: © Joerg Boethling/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 9: © Joerg Boethling/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 10: © International Women’s Health Coalition. Used via a Creative Commons license (https://creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 11: © Bhopal Medical Appeal. Used via a Creative Commons license (https://creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 12: © European Commission DG ECHO. Used via a Creative Commons license (https://creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 13: © European Commission DG ECHO. Used via a Creative Commons license (https://creativecommons.org /licenses/by/2.0/). Original photo was cropped to fit template; chapter 14: © Albert González Farran, UNAMID. Used via a Creative Commons license (https:// creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 15: © Md. Khalid Rayhan Shawon. Used with permission; further permission required for reuse; chapter 16: © Penny Tweedie/Alamy Stock Photo. Used with permission; further permission required for reuse; chapter 17: © Knut-Erik Helle. Used via a Creative Commons license (https://creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template; chapter 18: © European Commission DG ECHO. Used via a Creative Commons license (https://creativecommons.org/licenses/by/2.0/). Original photo was cropped to fit template. Library of Congress Cataloging-in-Publication Data has been requested. Contents Foreword xiii Abbreviations xv 1. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 1 King K. Holmes, Stefano Bertozzi, Barry R. Bloom, Prabhat Jha, Hellen Gelband, Lisa M. DeMaria, and Susan Horton PART 1 HIV/AIDS AND OTHER SEXUALLY TRANSMITTED INFECTIONS 2. Global Mortality and Morbidity of HIV/AIDS 29 Kristen Danforth, Reuben Granich, Danielle Wiedeman, Sanjiv Baxi, and Nancy Padian 3. HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 45 Corey Casper, Heidi Crane, Manoj Menon, and Deborah Money 4. HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 67 Katherine Harripersaud, Margaret McNairy, Saeed Ahmed, Elaine J. Abrams, Harsha Thirumurthy, and Wafaa M. El-Sadr 5. Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 91 Charles B. Holmes, Timothy B. Hallett, Rochelle P. Walensky, Till Bärnighausen, Yogan Pillay, and Myron S. Cohen 6. Prevention of Mother-to-Child Transmission of HIV and Syphilis 113 Grace John-Stewart, Rosanna W. Peeling, Carol Levin, Patricia J. Garcia, David Mabey, and John Kinuthia ix 7. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 137 Geoff P. Garnett, Shari Krishnaratne, Kate L. Harris, Timothy B. Hallett, Michael Santos, Joanne E. Enstone, Bernadette Hensen, Gina Dallabetta, Paul Revill, Simon A. J. Gregson, and James R. Hargreaves 8. Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 157 David Wilson and Jessica Taaffe 9. Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 179 James G. Kahn, Lori A. Bollinger, John Stover, and Elliot Marseille 10. Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 203 Harrell W. Chesson, Philippe Mayaud, and Sevgi O. Aral PART 2 TUBERCULOSIS 11. Tuberculosis 233 Barry R. Bloom, Rifat Atun, Ted Cohen, Christopher Dye, Hamish Fraser, Gabriela B. Gomez, Gwen Knight, Megan Murray, Edward Nardell, Eric Rubin, Joshua Salomon, Anna Vassall, Grigory Volchenkov, Richard White, Douglas Wilson, and Prashant Yadav PART 3 OTHER MAJOR INFECTIONS 12. Malaria Elimination and Eradication 315 Rima Shretta, Jenny Liu, Chris Cotter, Justin Cohen, Charlotte Dolenz, Kudzai Makomva, Gretchen Newby, Didier Ménard, Allison Phillips, Allison Tatarsky, Roly Gosling, and Richard Feachem 13. Malaria Control 347 Fabrizio Tediosi, Christian Lengeler, Marcia Castro, Rima Shretta, Carol Levin, Tim Wells, and Marcel Tanner 14. Febrile Illness in Adolescents and Adults 365 John A. Crump, Paul N. Newton, Sarah J. Baird, and Yoel Lubell 15. Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 385 Joseph B. Babigumira, Hellen Gelband, and Louis P. Garrison Jr. 16. Viral Hepatitis 401 Stefan Z. Wiktor x Contents 17. An Investment Case for Ending Neglected Tropical Diseases 411 Christopher Fitzpatrick, Uzoma Nwankwo, Edeltraud Lenk, Sake J. de Vlas, and Donald A. P. Bundy 18. Drug-Resistant Infections 433 Molly Miller-Petrie, Suraj Pant, and Ramanan Laxminarayan DCP3 Series Acknowledgments 449 Volume and Series Editors 451 Contributors 453 Advisory Committee to the Editors 457 Reviewers 459 Index 461 Contents xi Foreword Since the publication of the second edition of Disease and improved delivery of interventions through health Control Priorities in 2006, we have experienced some of systems and other sectors. Community engagement is the most substantial progress in infectious disease–caused the key to success in many cases; a community buy-in to mortality and morbidity. The number of annual deaths very simple, non-technological prevention mechanisms attributable to human immunodeficiency virus/acquired was instrumental in the sharp decline in dracunculiasis immune deficiency syndrome (HIV/AIDS) declined cases, from 130,000 in 2000 to only 22 in 2015 and 50 percent between 2004 and 2015, thanks to an unprece- 0 cases at the time of writing in 2017. However, dogs, dented expansion of life-saving antiretroviral therapy to which act as alternative hosts for the worm, present a over 18 million people (UNAIDS 2016); since 2006, threat to total eradication and remind us of the impor- mother-to-child transmission of HIV has been reduced tance of a “One Health” approach. to low levels, even in generalized epidemic settings At the same time, several epidemics and new patho- (AVERT 2017). Similarly, fewer children and adults die gens have emerged, including the swine flu (H1N1) from malaria, diarrheal diseases, and lower respiratory pandemic; the Middle East Respiratory Syndrome infections. Two infectious diseases are close to eradica- (MERS); the largest Ebola outbreak ever known in the tion: polio and dracunculiasis (Guinea worm disease). West African region where it had never caused an out- This third edition of the Disease Control Priorities break; and an epidemic of Zika and associated neurolog- (DCP3) comes at a pivotal moment for infectious disease ical disorders. In particular, the collective failure to control and research. Its chapters clearly demonstrate respond to the Ebola outbreak in a timely and coordi- that, despite the remarkable progress, infectious diseases nated fashion before it spiraled out of control—infecting remain a major threat to health worldwide—particularly over 28,000 people and causing over 11,000 deaths—was in South Asia and Sub-Saharan Africa—but that an a wake-up call for the world. The disastrous impact of increasing range of highly cost-effective interventions the Ebola epidemic prompted an urgent rethinking of is available. how governments, nongovernmental organizations, and As this volume amply illustrates, innovations in the international organizations can better work to contain prevention, diagnosis, and treatment of infectious dis- emerging disease threats in an increasingly intercon- eases have been impressive. They include preexposure nected world. prophylaxis (PrEP) to prevent HIV infection, new forms It is, however, noteworthy that almost as many people of computer-based education for clinicians to manage in the three Ebola-affected West African countries died sexually transmitted infections, HPV vaccines to pre- from the disease’s disruption on increasing mortality vent cervical cancer, and a cure for hepatitis C. The new from HIV/AIDS, tuberculosis, and malaria as from attention to viral hepatitis in this volume is most Ebola itself (Parpia and others 2016). These three dis- welcome, as greatly improved control is now technically eases, as well as diarrheal diseases and lower respiratory feasible—although the history of tuberculosis illustrates infections, continue to exact a heavy burden, particularly that a cure alone is insufficient to bring a disease under in Sub-Saharan Africa, where infectious diseases remain control. Much of the progress is due to political the leading cause of death. In 2015, over 1.8 million and technical leadership, greatly increased funding, people worldwide died from tuberculosis (including xiii 0.4 million among people with HIV) (WHO 2017); whether people will accept and use the interventions, 1.1 million people from AIDS (UNAIDS 2016); and an whether the interventions are affordable and work in estimated 429,000 people died from malaria (WHO various parts of the real world, and what the best way is 2016). In spite of real achievements in improved access to deliver them. to HIV treatment, over 2 million new infections occur If we are to reach the ambitious targets under the each year, with hardly any decline in new infections over Sustainable Development Goals, we must focus not only the past five years, and several subpopulations continue on delivery of innovation but also on “innovation of to be heavily affected. A critical review of current delivery.” One example might be new systems of com- HIV strategies may be needed to achieve the United munity-based treatment for tuberculosis to minimize Nations goals of ending the AIDS epidemic. Lower respi- transmission in health care settings. DCP3 helps us to ratory infections remain a major persistent cause of think about improving health care delivery models death in children. through its unique focus on packages of interventions, Many of these infectious diseases have sophisticated and on the interrelationships among different kinds of vaccines, diagnostics, and therapeutics available, but interventions, at both the policy level and in terms of the political, economic, and social factors limit the extent to outcomes across populations. which populations can benefit. Furthermore, in a world DCP3 is to be lauded for its focus on equity, recog- of growing resistance to antimicrobials and drug-resistant nizing that cost-effective intervention is not cost- infections, we need to continue to develop innovations effective if the financial burden falls on the poor. With in biomedicine. We also need to improve incentives for this DCP3 volume on major infectious diseases, we rational antibiotic use, antimicrobial stewardship, and have a highly pragmatic addition to the literature that increased acceptance of the importance of prevention to will help policy makers across the world make smarter avoid infection. decisions to improve health sustainably and equitably The global health agenda is an increasingly crowded in the ongoing fight against infectious disease threats, space, and the cost-effectiveness of interventions is old and new. under growing scrutiny. While there is more informa- tion than ever regarding the cost-effectiveness of differ- Peter Piot, MD, PhD ent interventions in a growing spectrum of contexts, Director, London School of Hygiene & Tropical Medicine hard choices remain in terms of allocating scarce fund- London, United Kingdom ing to infectious diseases, especially in light of the com- plexities of fragile health systems, comorbidities with other infections and NCDs, structural factors that can REFERENCES undermine disease prevention, and treatment pro- AVERT. 2017. “Prevention of Mother-to-Child Transmission grams. One particularly valuable facet of DCP3 is that it (Pmtct) of HIV.” AVERT. https://www.avert.org/professionals demonstrates that some of the most effective steps we /hiv-programming/prevention/prevention-mother-child. can take to reduce the burden of infections are not nec- Parpia, A. S., M. L. Ndeffo-Mbah, N.S. Wenzel, and A. P. Galvani. essarily expensive, as exemplified by the low cost per 2016. “Effects of Response to 2014–2015 Ebola Outbreak on Deaths from Malaria, HIV/AIDS, and Tuberculosis, disability-adjusted life year averted of condoms for West Africa.” Emerging Infectious Diseases 22 (3): 433–41. female sex workers or insecticide-treated bednets. Often, UNAIDS. 2016. “AIDS by the Numbers.” UNAIDS, Geneva. the key is not just more, but smarter, investment, for WHO (World Health Organization). 2016. “10 Facts on example, better integration of services, strong commu- Malaria.” WHO, Geneva. http://www.who.int/features nity engagement, and targeted interventions based on /factfiles/malaria/en/. the population most in need in specific locations. ———. 2017. “Tuberculosis: Fact Sheet.” WHO, Geneva. In addition to cost-effectiveness, key questions are http://www.who.int/mediacentre/factsheets/fs104/en/. xiv Foreword Abbreviations ACTs artemisinin-combination therapies ADCs AIDS-defining cancers AEM AIDS Epidemic Model AIDS acquired immune deficiency syndrome AIM AIDS Impact Model ANC antenatal clinic ART antiretroviral treatment BCC behavior change communication CBT community-based testing CDC Centers for Disease Control and Prevention (U.S.) CHMI controlled human malaria infection CHW community health worker CI confidence interval CQ chloroquine CRP C-reactive protein CSF cerebrospinal fluid CVD cardiovascular disease DAA direct-acting antiviral DALYs disability-adjusted life years DBS dried blood spots DCP2 Disease Control Priorities in Developing Countries, second edition DCP3 Disease Control Priorities (third edition) DDT dichloro-diphenyl-trichloroethane DMPPT Decision Makers’ Program Planning Tool EID early infant diagnostic ELISA enzyme-linked immunosorbent assay EMOD Epidemiological Modeling EMTCT elimination of mother-to-child transmission EPP Epidemic Projection Package FBC facility-based care FI febrile illness FSW female sex worker xv G6PD glucose-6-phosphate dehydrogenase GBD Global Burden of Disease GDP gross domestic product GHD Global Health Decisions GMAP Global Malaria Action Plan GMEP Global Malaria Eradication Program GRADE Grading of Recommendations, Assessment, Development and Evaluation GRNE Global Resource Needs Estimates GTS Global Technical Strategy HAI health care-associated infections HAT Human African trypanosomiasis HBC home-based care HBV hepatitis B virus HCC hepatocellular carcinoma HCV hepatitis C virus HDL high-density lipoprotein cholesterol HHV-8 human herpes virus 8 HICs high-income countries HIV/AIDS human immunodeficiency virus/acquired immune deficiency syndrome HIVAM HIV-associated malignancies HIV human immunodeficiency virus HPV human papillomavirus HRQL health and health-related quality of life HSV herpes simplex virus HTC HIV testing and counseling HTS HIV/AIDS testing service ICER incremental cost-effectiveness ratio IDU injecting drug user IMAI Integrated Management of Adolescent and Adult Illness IMCI Integrated Management of Childhood Illness IRS indoor residual spraying ITN insecticide-treated net IVM integrated vector management JEV Japanese encephalitis virus KS Kaposi sarcoma LDL low-density lipoprotein cholesterol LGBT lesbian, gay, bisexual, and transgender LICs low-income countries LLIN long-lasting insecticide-treated nets LMICs low- and middle-income countries LTFU loss to follow-up MCH maternal and child health MDA mass drug administration MDG Millennium Development Goal MDR-TB multidrug-resistant tuberculosis MMC medical male circumcision MERS Middle East respiratory syndrome MCC mobile clinic care xvi Abbreviations MRSA Methicillin-resistant Staphylococcus aureus MSM men who have sex with men MTCT mother-to-child transmission NADCs non–AIDS-defining cancer NBS National Bureau of Statistics NCC noncommunicable chronic comorbidities NCDs noncommunicable diseases NECT nifurtimox-eflornithine combination therapy NHL non-Hodgkin lymphoma NNRTI non-nucleoside reverse transcriptase inhibitors NRTI nucleoside reverse transcriptase inhibitors NSP needle and syringe program NTDs neglected tropical diseases NT-NMFI nontreatable nonmalaria febrile illness OOP out of pocket PCR polymerase chain reaction PEPFAR President’s Emergency Plan for AIDS Relief (United States) PMI President’s Malaria Initiative PMTCT prevention of mother-to-child transmission POC point-of-care POCT point-of-care test PPT periodic presumptive treatment PrEP preexposure prophylaxis PROMISE Promoting Maternal-Infant Survival Everywhere QALY quality-adjusted life year RNM Resource Needs Model RBM Roll Back Malaria (Partnership) RPR rapid plasma reagin RCT randomized controlled trial RDTs rapid diagnostic tests SAC school-age children SARS severe acute respiratory syndrome SDG Sustainable Development Goal SMART Strategies for Management of Antiretroviral Therapy STDs sexually transmitted diseases SP sulfadoxine/pyrimethamine STIs sexually transmitted infections STH soil-transmitted helminthiases SW sex workers TasP treatment as prevention TB tuberculosis TCPs target candidate profiles T-NMFI treatable nonmalaria febrile illness UI uncertainty interval UNAIDS Joint United Nations Programme on HIV/AIDS UN United Nations Abbreviations xvii VCT voluntary counseling and testing VIMT vaccines that interrupt malaria transmission VMMC voluntary male medical circumcision WHO World Health Organization WWR What Works Reviews YLDs years lived with disability YLLs years of life lost xviii Abbreviations Chapter 1 Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition King K. Holmes, Stefano Bertozzi, Barry R. Bloom, Prabhat Jha, Hellen Gelband, Lisa M. DeMaria, and Susan Horton INTRODUCTION Global under-five mortality fell by almost two-thirds Infectious diseases were responsible for the largest (from 14 percent to 5 percent) between 1970 and 2010 global burden of premature death and disability until (Norheim and others 2015). In 1980, smallpox, respon- the end of the twentieth century, when that distinction sible for 300 million–500 million deaths in the twentieth passed to noncommunicable diseases. Over the previ- century, was declared to be the first disease eradicated ous centuries, global pandemics of infectious diseases, from the planet following a global immunization cam- such as smallpox, cholera, and influenza, periodically paign led by the WHO. Wild Poliovirus has been elimi- threatened the survival of entire populations. At least as nated from all but three countries (Afghanistan, Nigeria, early as the late 1800s, improved living conditions and Pakistan) and currently is the focus of a major (such as better sanitation and piped water supplies), eradication program. particularly in high-income countries (HICs), began to The decline of the vaccine-preventable diseases has drive down the infectious disease burden. contributed to a recognition of the potential for using By the mid-twentieth century, safe, effective, and vaccines to prevent other infectious diseases, including affordable vaccines and the increasing availability of human immunodeficiency virus/acquired immune defi- antibiotics had further reduced the toll of infectious ciency syndrome (HIV/AIDS), tuberculosis (TB), diseases in HICs. Not until the second half of the malaria, hepatitis C, and a variety of neglected tropical twentieth century did large-scale efforts begin to better diseases (NTDs). Hepatitis B and C substantially increase control infectious diseases in low- and middle-income the risk of death from cirrhosis and liver cancer. The countries (LMICs), where the infectious disease effect of viral hepatitis is significant. Indeed, an impor- burden was greatest and highly varied. These efforts tant recent study (Stanaway and others 2015) found that included a global commitment to immunize the viral hepatitis led to an estimated 0.9 million deaths in world’s children against the major infections for which 1990 (including hepatitis-caused deaths from cirrhosis vaccines are available and global campaigns to control and liver cancer). Furthermore, this number has been malaria and diarrheal disease. The International Health increasing rapidly—to an estimated 1.5 million deaths Regulations of the World Health Organization (WHO) in 2013—despite the fact that hepatitis B is a vaccine- represent a key agreement among 196 countries to preventable disease and that hepatitis B and C are implement metrics and measures to detect and control both treatable. outbreaks of infectious diseases and to prevent pan- Emerging pandemic viral infections remain a con- demics (World Health Assembly 2005). stant threat, many entering the human population from Corresponding author: King K. Holmes, Departments of Global Health and Medicine, University of Washington, Seattle, Washington, United States; kkh@uw.edu. 1 contact with animals. The most recent such infections APPROACHES TO INFECTIOUS DISEASE include SARS (severe acute respiratory syndrome), MERS CONTROL IN THE TWENTY-FIRST CENTURY (Middle East respiratory syndrome), and Ebola and Zika viruses (Madhav and others 2018) as well as, perennially, Vaccines and curative treatments for some of the major influenza and chikungunya infections. Compared with infectious diseases have existed for decades. Many of antibiotics to treat bacterial infections, relatively few anti- them are relatively inexpensive and highly cost-effective, viral drugs have been developed to treat these emerging yet many are underused because of cost and lack of viral infections. Therefore, the most important interven- access attributed to poorly functioning health care sys- tion is to break the chain of transmission. A global increase tems. New drugs and vaccines will continue to be the in antibiotic-resistant bacteria includes a small but grow- mainstays in preventing and treating infections, but ing number that are resistant to most or essentially all of delivery of such interventions will be critical to driving the available antimicrobials. down the burden of infection. Spectacular progress has been made in reducing mor- An ultimate goal for selected infections is eradication. tality from most infectious diseases (table 1.1). For To date, only two diseases—smallpox in humans and example, in low-income countries (LICs) from 2000 to rinderpest in cattle and other ruminant animals—have 2010, the number of deaths before age 70 years from been eradicated. Elimination of polio, yaws, and Guinea HIV/AIDS, TB, and malaria fell by 46 percent, 35 percent, worm infections is being pursued. This is a more distant and 36 percent, respectively (Norheim and others 2015). but still possible goal for malaria (Shretta and others 2017). Rapid progress was also reported in other country A handful of other infections—such as measles, income groups. However, table 1.1 shows also that if the mumps, rubella, lymphatic filariasis, and cysticercosis— death rates of 2010 remain static, about 5.1 million peo- are candidates for elimination because of disease char- ple will still die in 2030 from these three conditions and acteristics or the available means to control them (CDC from other communicable diseases, many of which are 1993). Those infectious diseases that persist require concentrated in LMICs. In contrast, mortality in HICs continued effort to develop new drugs and vaccines for from these conditions (except for HIV/AIDS) will be treatment and prevention as well as strategies that relatively small, although major pandemics of other allow such treatments to be used most effectively pathogens are not predictable. Hence, infectious diseases across the globe. Despite the development of new will remain a major threat to humankind, especially in drugs to combat infectious diseases, antimicrobial LMICs, requiring vigilance, surveillance, and new inter- resistance is threatening to remove many of the tools ventions of all types. in our current armamentarium. Table 1.1 Projected 2030 Mortality and 10-Year Trends for Selected Infections, by Country Income Group Deaths Change Deaths Change Deaths Change Deaths Change Deaths in 2030 (% per in 2030 (% per in 2030 (% per in 2030 (% per in 2030 (millions) decade)a (millions) decade)a (millions) decade)a (millions) decade)a (millions) Cause Low income Lower middle income Upper middle income High income Globalc HIV/AIDS 0.92 −46 0.76 0 0.41 17 < 0.01 — 2.10 Tuberculosis 0.32 −35 0.65 −43 0.14 −52 < 0.01 — 1.12 Malaria 0.37 −36 0.33 −28 0.02 — < 0.01 — 0.73 Other 0.35 −23 0.59 −15 0.14 −30 0.05 — 1.13 communicable diseasesb All causes 8.62 −24 18.11 −16 11.60 −23 3.00 −16 41.33 Source: Norheim and others 2015. Note: — = not available; HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome. Table estimates the number of deaths before age 70 years (age 0–69 years) that would occur in 2030 if the rate of change (percentage per decade) during 2000–10 were to continue for standardized death rates of those under age 70 years in each country income group (as classified by the World Bank). a. “Change (% per decade)” = percentage change during 2000–10. b. “Other communicable diseases” mostly align with other infectious conditions covered in this volume (such as hepatitis, sexually transmitted infections, and neglected tropical diseases) but not completely for some diseases (for example, meningitis). c. Global totals are by summation of each of the four regions. 2 Major Infectious Diseases Infection Control: Targeting and Integration of Importance of Rapid Differential Diagnosis in Specialized Services Infection Control One approach—targeting settings and populations Even for infectious diseases requiring specialized services, with the highest transmission rates—is being used to many infections will initially be diagnosed or suspected improve HIV/AIDS control. The continued emergence at the primary care level or first-level hospital and then of new pandemics, like Ebola, will be addressed with referred to a second- or third-level hospital. Many infec- similarly targeted approaches. However, for the ongoing, tious illnesses are caused by pathogens that can be life highly prevalent infections in LMICs—TB, hepatitis, threatening. This makes differential diagnosis—based in sexually transmitted infections (STIs), malaria, typhoid, part on symptom assessment, clinical manifestations, and other febrile illnesses—the future lies in improving physical exam, medical history, history of exposures, age and integrating services at the primary care level and up and gender, laboratory testing where available, and avail- the chain to the highest levels of hospital care. ability of treatment—the key to population infection The high incidence of comorbidities—such as TB control (Burnett and others 2016). and viral hepatitis in immunocompromised persons The widespread adoption of rapid tests for malaria with HIV infection—calls for integration of HIV, TB, diagnosis is an example of an easy-to-use diagnostic and viral hepatitis diagnosis and treatment. Patients who that has vastly improved malaria treatment in many are seen in STI clinics and their sex partners have an places. Rapid point-of-care tests are available for HIV, elevated risk of having another STI, such as HIV infec- hepatitis C, influenza, and syphilis but are still in devel- tion. Therefore, integration of HIV testing, care, and opment for some other infections. In addition, conven- evaluation of sex partners into all STI clinic settings tional microbiology is being transformed by molecular offers opportunities for efficiently managing this set of testing, which could be available even in LICs within the infections. Selected populations for whom specialized decade. A series of publications illustrates the significant services are already the norm (such as pregnant women) effect of integrated infectious disease training on diag- can receive additional attention. This may include nosis (in Uganda) and infectious disease management screening for HIV and syphilis (see chapter 6 of this (Imani and others 2015; Weaver and others 2014) volume, John-Stewart and others 2017). This volume focuses on major infectious diseases that Health care service integration at this level requires are common in LMICs, particularly among adults strategic planning. For example, integration of HIV and (see box 1.1). Unlike most of the serious infections that TB diagnosis and management must be done in ways predominate among children, many of these are long- that are safe: crowded clinics with long waiting times lived chronic infections (including some acquired as create a perfect opportunity for TB transmission from children). The perspective includes an emphasis on what someone with active TB to an immunosuppressed HIV has changed since the first edition of Disease Control patient. Integrated, population-level intervention pack- Priorities in Developing Countries in 1993 (box 1.2). ages can focus not only on interventions that are We first review the major interventions for priority infec- financed mainly by the health sector, such as increased tious diseases, namely HIV/AIDS, other STIs, TB, malaria immunization and treatment, but also on interventions and other febrile illnesses, hepatitis, and NTDs. We then related to the agriculture or infrastructure sector (and address the cross-cutting issues of antimicrobial resistance. financed mainly by those ministries, not the health min- istry), such as the following: • Improve access to sanitation, clean water, and hygiene. Box 1.1 • Reduce population growth and crowding. • Decrease day-to-day close contact with animals. • Change the environments that sustain vectors of Volume Focus: Infectious Disease Control important pathogens. This volume focuses on control of the major infectious diseases. Infectious disease control A related cross-sectoral priority is preventing antimi- involves not only prevention of transmission and crobial resistance through the development, availability, spread of infectious disease at the population and use of affordable diagnostics to guide appropriate and individual levels, but also effective treatment antimicrobial use in humans, while also enforcing poli- and cure of infectious diseases in individuals. cies to prevent nontherapeutic use of antimicrobials as growth promoters in livestock. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 3 Box 1.2 Comment by the Series Editors of Disease Control Priorities, Third Edition Budgets constrain choices. Policy analysis helps care, medical expenses that are high relative to decision makers achieve the greatest value from income can be impoverishing. Where incomes are limited resources. In 1993, the World Bank pub- low, seemingly inexpensive medical procedures can lished the first edition of Disease Control Priorities have catastrophic financial effects. DCP3 considers in Developing Countries (DCP1), which sought to financial protection and the distribution across assess systematically the cost-effectiveness (value income groups of the outcomes from policies (for for money) of interventions that would address example, public financing of health care) to increase the major sources of disease burden in LMICs intervention uptake and to improve delivery quality (Jamison and others 1993). The World Bank’s 1993 (Verguet, Laxminarayan, and Jamison 2015). All of World Development Report on health drew heavily the volumes seek to combine the available science on the findings in DCP1 to conclude that specific about interventions implemented in specific locales interventions against noncommunicable diseases and conditions with informed judgment to reach were cost-effective, even in environments where reasonable conclusions about the effects of interven- substantial burdens of infection and undernutrition tion mixes in diverse environments. DCP3’s broad persisted (World Bank 1993). aim is to delineate essential intervention packages— such as those, in this volume, for major infectious Disease Control Priorities in Developing Countries, diseases—and their related delivery platforms. This second edition (DCP2), published in 2006, updated information will assist decision makers in allocating and extended DCP1 in several respects, explicitly often tightly constrained budgets and achieving considering the implications for health care systems health care system objectives. of expanded intervention coverage (Jamison and others 2006). One way that health care systems can Four of DCP3’s nine volumes were published in 2015 expand coverage of health interventions is through and 2016, and the remaining five will appear in 2017 selected delivery platforms for those interventions and 2018. The volumes appear in an environment in that require similar logistics but address heteroge- which serious discussion continues about quantifying neous health problems. Platforms often provide a and achieving the Sustainable Development Goal more natural unit for investment than do individual (SDG) for health (UN 2015b). DCP3’s analyses are interventions, but conventional health economics well placed to assist in choosing the means to attain has offered little understanding of how to make the health SDG and assessing the related costs. These choices across platforms. Analysis of the costs of volumes, and the analytic efforts on which they are packages and platforms—and of the health improve- based, will enable researchers to explore SDG- ments they can generate in given epidemiological related and other broad policy conclusions and environments—can help guide health care system generalizations. The final volume will report those investments and development. conclusions. Each volume will provide specific pol- icy analyses on the full range of interventions, pack- This third edition of Disease Control Priorities ages, and policies relevant to its health topic. (DCP3) introduces the notion of packages of inter- ventions (Jamison and others 2015–18). Whereas Dean T. Jamison “platforms” refer to logistically related sets of inter- Rachel Nugent ventions, “packages” comprise conceptually related ones. (The 21 packages of interventions developed Hellen Gelband in the nine volumes of DCP3 include those targeting Susan Horton surgery and cardiovascular disease, for example.) In Prabhat Jha addition, DCP3 explicitly considers the financial risk Ramanan Laxminarayan protection objective of health care systems. In popu- lations lacking access to health insurance or prepaid Charles N. Mock 4 Major Infectious Diseases We provide updated estimates of the cost-effectiveness of number of deaths annually from HIV/AIDS has the major sets of interventions, recognizing that there are declined from 2 million in 2005 to 1.1 million in 2015, large knowledge gaps concerning the economics of many the lowest number since 1998 (UNAIDS 2015, 2016). conditions in LMICs. We conclude by outlining future Yet AIDS still ranks sixth among the global causes of strategies that are relevant to continued progress against death—and first in Sub-Saharan Africa. these major infectious diseases. Building on the progress to date, UNAIDS has set two important goals: (a) a 75 percent reduction in new HIV infections (compared with 2010) by 2030 and (b) suc- MAJOR INFECTIOUS DISEASES IN THIS cessful achievement of the UNAIDS 90-90-90 campaign, VOLUME which seeks to have 90 percent of all people living with HIV knowing they have HIV, 90 percent of those diag- HIV/AIDS and Other Sexually Transmitted Infections nosed with HIV receiving treatment, and 90 percent of HIV/AIDS, the worst human pandemic since the 1918 those on treatment having an undetectable viral load influenza epidemic, has accounted for more than (virally suppressed). Furthermore, UNAIDS seeks to 25 million deaths since it was first identified in 1981 and eliminate mother-to-child transmission of both HIV it has hit Sub-Saharan Africa the hardest. However, the and syphilis. We now have the tools to attain these goals, tide is beginning to turn as life-extending antiretroviral even despite the remaining challenges of needing an HIV treatment (ART) and preventive interventions are scaled vaccine for prevention (as we have vaccines to prevent up and as sexual behaviors may have become less risky in hepatitis B virus [HBV] and human papillomavirus many settings. [HPV] infection); an effective cure for HIV infection (as Antiretroviral drugs are now widely available in we have for the hepatitis C virus [HCV] infection); and most settings and are highly affordable at US$315 per an effective suppressive therapy for hepatitis B. person per year (UNAIDS 2015). Worldwide, 17 million In addition to effective medical interventions (table 1.2), HIV-infected people are receiving these life-extending national legislative and policy frameworks are needed to drugs—an impressive number given that only 2.2 million enable effective deployment of these interventions. people were on ART in 2005. However, this number is Mother-to-child transmission of HIV will not be elimi- still far short of the Joint United Nations Programme nated by 2030, the current goalpost, without effective on HIV/AIDS (UNAIDS) target to treat the 37 million national policies to support prevention. Even more people currently living with HIV. The estimated important, laws and policies to protect and reduce Table 1.2 Essential HIV/AIDS Intervention Package, by Delivery Platform Delivery platformsa First- Second- and Intervention Nationwide policies and Community health post or level third-level type regulations pharmacy Primary health center hospital hospitals Prevention Legal and human 1. Laws and policies to protect and 2. Gender-based violence rights reduce stigma for key populations, counseling and rape-response with full decriminalization of LGBT referral (medical and justice) population* Structural 3. Universal access to HIV testing, interventions with immediate linkage to care and treatment and intensified outreach to populations at higher risk of infection 4. Universal access to drug substitution therapy for addiction 5. Brothels: Condoms required* 6. Needle exchange encouraged* table continues next page Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 5 Table 1.2 Essential HIV/AIDS Intervention Package, by Delivery Platform (continued) Delivery platformsa First- Second- and Intervention Nationwide policies and Community health post or level third-level type regulations pharmacy Primary health center hospital hospitals Direct (biological) 7. PrEP for discordant prevention couples 8. Male circumcision service provision* 9. PMTCT (Option B+) Behavioral 10. HIV education and counseling interventions: for pregnant women, sex Prevention workers, IDUs, GBT males, and HIV+ persons and their partners* 11. Access to needle exchange for IDU* 12. Condom distribution* 13. Partner notification* 14. Adherence support for high- risk or failing patients Social marketing: 15. Promotion of condoms, VMMC, Information, and testing at national and facility- education, and based levels* communication Treatment Treatment 16. Policies and guidelines to support 17. Community-based HIV testing 20. Provider-initiated all steps of HIV care continuum, and counseling (for example, counseling and HIV including expanded testing through mobile units or venue- testing (as well as TB through diverse strategies; linkage based testing)* and STI testing) for all to care; ART initiation with support 18. Household HIV testing and in contact with health for adherence and retention; counseling in high-prevalence care system in high- and performance and efficiency settings prevalence settings, optimization through data-driven including prenatal care* management, task shifting, and 19. Referral and navigation of HIV+ individuals to HIV care 21. ART initiation decentralization, as appropriate for level of epidemic sites to ensure linkage 22. Support for adherence and retention 23. Laboratory viral load monitoring Behavioral 24. Adherence support including 26. Case manager and structural adherence clubs, community- interventions: based ART groups, text Care reminders, and other means 25. Nutrition, transportation, and financial reimbursement Note: Interventions shown in orange indicate areas that are relatively neglected by governments. ART = antiretroviral treatment; GBT = gay, bisexual, or transgender; HIV = human immunodeficiency virus; HIV+ = HIV-positive; IDU = injection drug users; LGBT = lesbian, gay, bisexual, and transgender; Option B+ = a three-drug ART regime in pregnancy and postpartum for HIV-positive mothers; PMTCT = prevention of mother-to-child transmission; PrEP = preexposure prophylaxis; STI = sexually transmitted infection; TB = tuberculosis; VMMC = voluntary male medical circumcision. a. All interventions listed for lower-level platforms can be provided at higher levels. Similarly, each facility level represents a spectrum and diversity of capabilities. The column in which intervention is listed is the lowest level of the health care system in which it would usually be provided. Interventions marked with an asterisk (*) should be closely integrated with STI prevention and treatment interventions. 6 Major Infectious Diseases stigma for key populations are urgently needed in many reduce the number of newborns infected during the countries. Indeed, in recent years, lesbian, gay, bisexual, birth process and the mother-to-child transmission of and transgender (LGBT) rights have regressed in some HIV. Substantial progress has been made in this regard: settings, and criminalization of these populations has new pediatric HIV infections declined by 50 percent increased. Two chapters in this volume provide useful from 2010 to 2015 (UNAIDS 2016). detail for policy makers who are considering such issues: Interventions to offer household-based testing in chapter 8 (Wilson and Taaffe 2017) outlines the factors high-prevalence settings will contribute to the first “90” to consider when tailoring a response to a local epi- of the UNAIDS 90-90-90 goals (90 percent of all people demic, and chapter 9 (Kahn and others 2017) presents living with HIV know they have HIV). Interventions to various models that can help guide decisions regarding effectively and promptly link newly diagnosed persons the cost-effectiveness of the different interventions. living with HIV to services and treatment contribute to Optimal HIV management requires managing people the second “90” (90 percent of HIV-diagnosed people across the continuum of care, from testing to counseling receive treatment) and are critical across all settings. and from ART to adherence support. Sociocultural bar- Finally, multiple strategies for promoting adherence to riers in gaining access to care include the following: treatment and retention in care—ranging from commu- nity support groups to mobile health interventions—are • Fear of diagnosis, complicated by a culture of stigma critical to ensuring that treatment is effective and con- and discrimination in many countries tinuous, thus achieving the third “90” (90 percent of • Structural barriers such as distance to health clinics those being treated have an undetectable viral load). • System-level barriers such as clinic hours, coordination among clinics, and shortages of health care workers Burden of STIs other than HIV/AIDS In addition to HIV, another 40 bacterial, viral, and Biomedical interventions that have come to the fore- parasitic pathogens have been identified as primarily front since the publication of Disease Control Priorities in sexually transmitted, or as potentially sexually transmis- Developing Countries, second edition (DCP2) by the sible (see annex 1A). The common curable bacterial STIs World Bank (Jamison and others 2006) and that have include trichomoniasis, chlamydia infection, gonorrhea, proven highly effective at preventing HIV transmission and syphilis. In 2012, the WHO estimated the global include treatment as prevention (Gomez and others incidence of these four curable STIs among men and 2013); preexposure prophylaxis (PrEP); male circumci- women ages 15–49 years: 131 million new cases of sion; and new treatment regimens for prevention of chlamydia infection, 78 million of gonorrhea, 143 million mother-to-child transmission (PMTCT). Furthermore, of trichomoniasis, and 6 million of syphilis (WHO these interventions can be successfully delivered at first- 2016b). These estimates mean that approximately level care facilities, thereby increasing accessibility. 1 million new infections could be cured with existing PrEP—using a once-a-day tablet, the current version treatments each day (Newman and others 2015). of which includes two antiretroviral drugs—provides a Other common sexually transmitted pathogens are method beyond condoms for at-risk people to prevent herpes simplex virus (HSV-1 and HSV-2, both of which becoming infected with HIV (Baeten 2016; Jenness and cause genital herpes) and HPV.2 In 2012, the global others 2016). PrEP access is still limited in LMICs and prevalence of HSV-2 among men and women ages does not prevent other STIs. 15–49 years was 417 million, with higher prevalence in Voluntary male medical circumcision (VMMC) women than in men. An estimated 19.2 million individ- significantly reduces sexual acquisition of HIV by men uals ages 15–34 years were newly infected with HSV-2 in and is most cost-effective in settings where HIV is highly 2012 (Looker and others 2015). prevalent. Recent studies have shown that demand is Extensive studies of the prevalence of oncogenic genital high for VMMC, which can be offered at some first-level HPV infections included a global systematic review of health care facilities and at health centers. In some coun- age-specific prevalence of oncogenic types of HPV infec- tries, VMMC has even been delivered effectively in tion in males (Smith and others 2011) and in females mobile vans. (Winer and others 2012). In general, these studies show Advances have increased the effectiveness of PMTCT high prevalence of oncogenic HPV types among those with treatment as well. For an HIV-infected mother not yet new sex partners or a high number of lifetime partners. receiving ART, the recommendations are to start ART Common STIs may cause significant complications at the first prenatal visit (regardless of the mother’s to women’s reproductive health, including pelvic inflam- CD4 cell count or WHO clinical stage)1 and to continue matory disease, tubal pregnancy infertility, cervical can- lifelong ART. Use of this protocol could significantly cer, perinatal and neonatal morbidity, mother-to-child Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 7 transmission of syphilis or HIV, and a host of other role of key populations in the epidemiology and control conditions (Chesson, Mayaud, and Aral 2017, chapter 10 of HIV and other STIs in LMICs has become increasingly in this volume). However, for morbidity and mortality, clear (Baral and others 2007; Baral and others 2012). years of life lost, disability-adjusted life years (DALYs), These key populations include, in particular, female sex and costs of medical care, the major STIs are as follows: workers; men who have sex with men (MSM), who are understudied and underserved in most LMICs; and • HIV infection injection drug users, who are at risk not only for HIV • HBV and HCV infection but also for other blood-borne STIs such as syphilis • HPV infection, with HPV-related genital, anal, and and hepatitis viruses. Patterns of sexual networks oropharyngeal cancers linking MSM with heterosexual populations warrant • Syphilis, with its related perinatal and pediatric mor- future research. bidity and mortality Until recently, HCV was repeatedly described as not • HSV-1 and HSV-2 infection, with related central sexually transmitted, and its transmission had been associ- nervous system and pediatric morbidity ated with injection drug use, blood transfusions, and iatro- genic exposures but not with heterosexual transmission. In aggregate, these major pathogens cause extensive However, HCV recently has been found in the semen of morbidity and mortality attributable to unsafe sex. men with HCV viremia, and rectal HCV shedding was Moreover, the consequences of STIs disproportionately found in 20 of 43 (47 percent) HIV-infected MSM who affect women and children. STIs, including HIV/AIDS, also had HCV infection. The presence of HCV in rectal are one of the leading causes of morbidity and mortality, fluid was associated with high blood levels of HCV (Foster as measured by DALYs, for reproductive-age women in and others 2017). Most important, co-infections with HCV LMICs (Owusu-Edusei and others 2014). and HIV have been commonly found in Australia, Europe, In addition to the mortality and morbidity attribut- and North America. Thus, screening for HCV—a curable able to the major STI pathogens listed earlier, other STI infection—is now being recommended for HIV-infected pathogens account for severe morbidity, including infer- MSM in high-income countries (Harrison and others tility, ectopic pregnancy, epididymitis, neonatal eye infec- 2017; Kratz and others 2015; Nanduri and others 2016). tion, and other common diseases. These other pathogens that can be transmitted sexually include the Zika and STI Interventions: Prevention, Treatment, and Ebola viruses and group C Neisseria meningitidis. Sexual Education transmissions of these pathogens have been documented Prevention and treatment are both important to STI but are not yet well studied (Hader 2017). control, and the HIV epidemic has influenced changes in the approach to STI prevention in general. During Unsafe Sex as a Global Risk Factor for Death and the 1980s and 1990s, behavioral prevention dominated Disability in Adolescents and Young Adults the HIV world and gained prominence in the STI The Global Burden of Disease (GBD) study recently domain. However, since the turn of the century, recog- reported annual assessments of risk factors for death nition has grown that behavioral interventions (heav- and DALYs in adolescents and young adults in ily weighted toward condom use) have not decreased 188 countries for 2013 (Mokdad and others 2016). STI incidence sufficiently and sustainably (Aral 2011; Among adolescent males ages 15–19 years, unsafe sex Kippax and Stephenson 2012). was the second most common risk factor for death. Among adolescent females ages 15–19 years, unsafe sex STI Prevention was the number one risk factor. Among young adults Concurrently, remarkable progress has been made in bio- ages 20–24 years (males and females combined), medical approaches to preventing HIV/AIDS, including unsafe sex was the second most common risk factor. male circumcision, PrEP, and ART (Baeten and others 2012; As for the risk of disability (as measured by DALYs), Dodd, Garnett, and Hallett 2010; Grant and others 2010; unsafe sex was the second most common risk factor in Katz and Wright 2008; Pretorius and others 2010). Given 2013. Important to the global burden, the number the success of these biomedical approaches, the field of STI and proportion of the worldwide population who are ado- prevention is increasingly drawing on them, reinforced by lescents are also steadily growing (Hader 2017). development of effective biomedical interventions for pre- venting STIs other than HIV. More specifically, these inter- Key Populations for STI Control in LMICs ventions include promotion and provision of the HPV Although adolescents and young adults experience and HBV vaccines to females and males, early detection a large proportion of STIs, including HIV infection, the and curative treatment of HCV infection, point-of-care 8 Major Infectious Diseases diagnostic tests for syphilis, dual tests for syphilis and HIV, However, this practice, linked to the increasing avail- and an understanding of the effects of male circumcision ability of new antimicrobials in LMICs, may be contrib- for preventing certain STIs other than HIV. uting to emerging antimicrobial resistance in LMICs In addition, clinical platforms offering STI-related (Miller-Petrie, Pant, and Laxminarayan 2017, chapter 18 reproductive health services are playing a key role in of this volume). Although the common curable STIs can screening patients for HIV and HCV. They also emphasize be managed effectively in LMICs with widely available outreach to sex partners for HIV and other STI screening. antibiotics, global development of antibiotic resistance Table 1.3 provides an assessment of the platforms and has eroded the success of treatment of some infections, essential interventions for preventing and treating STIs. including gonorrhea. Canchihuaman and others (2011) have also demon- Pharmacy Treatment of STIs and Clinician Online strated the feasibility and effectiveness of using computer- Education based education to reach out to clinicians and midwives Individual treatment of STIs in LMICs is largely based to vastly expand and improve the scope and effect of on syndromic management, which is often provided by online continuing education of STI management. This pharmacies without clinical examination. Provision of approach is a critical and effective step to guide large guidelines and training to pharmacy workers can signifi- groups of clinicians and communities, even in remote cantly improve STI management by pharmacy workers rural areas, to better health care in general but especially (García and others 2012). regarding infectious diseases. Table 1.3 Essential STI Intervention Package, by Delivery Platform Platforms for intervention delivery Nationwide, regional, Primary and local health health and Second-, and systems, policies and reproductive First-level third-level regulations Community health posta Pharmaciesb health clinicsc hospitals hospitalsd Structural 1. Organize, coordinate, 3. Linkages to clinical 6. Training on and where possible, services for FSW, MSM, syndromic integrate programs sex partners of persons treatment for STI and HIV/AIDS with STI/HIV* of STIs by into one national 4. Training for police pharmacists center and into to ensure access regional centers, with to services for key essential funding and populations, especially system support for needle exchange* local programs* 5. Home-based services 2. National policies to and Internet use for enable prevention partner notification, HIV and treatment efforts diagnosis, and initiating for key populations* HIV treatment in patients with HIV infection* Behavioral 7. Social marketing 9. School-based sexual prevention linked to education health education (STI/ on STI/HIV risks, and HIV risks, condom use, on sexual health, substance abuse, key including condom and vaccines, VMMC)* safe sex promotion*e 10. Condom promotion; 8. National curriculum Needle exchange for and policy regarding IDU*f sexual health education (including online education)* table continues next page Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 9 Table 1.3 Essential STI Intervention Package, by Delivery Platform (continued) Platforms for intervention delivery Nationwide, regional, Primary and local health health and Second-, and systems, policies and reproductive First-level third-level regulations Community health posta Pharmaciesb health clinicsc hospitals hospitalsd Biomedical 11. Guidelines, funding, 12. School-based and health 16. Vaccine prevention and social marketing post provision of HPV and provision for HPV and HBV HBV vaccines, and linkage (HPV, HBV) vaccines; and for to or provision of VMMC (females VMMC (adolescents, services and males) adults, infants) 13. Access to needle 17. VMMC* exchange for IDU* 18. Visual 14. Screening and treatment inspection for major STIs among with acetic prison populations acid for 15. Preexposure cervical antimicrobial prophylaxis dysplasia in high risk populations* Diagnosis 19. Guidelines for 20. Syndromic- 21. Diagnosis and 22. Diagnosis and and expedited partner based treatment of treatment treatment therapy via treatment suspected pelvic of anal, pharmacies of STIs inflammatory oropharyngeal, disease; viral and liver hepatitis; ART, cancers; and plus detection other HIV and treatment comorbidities. or referral of comorbidities, and some HIV comorbidities Note: Interventions shown in orange indicate areas that are relatively neglected by governments. Interventions marked with an asterisk (*) should be closely integrated with HIV prevention and treatment interventions. FSW = female sex workers; HBV = hepatitis B virus; HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome; HPV = human papillomavirus; IDU = injection drug users; MSM = men who have sex with men; PrEP = preexposure prophylaxis; STD = sexually transmitted disease; STI = sexually transmitted infection; VMMC = voluntary male medical circumcision. a. This platform involves extension of health services beyond conventional clinical platforms to reach high-risk populations. b. Pharmacies are very accessible (proximity, short wait times, low cost) and provide much of the treatment for STI syndromes. Yet, adherence to STD treatment guidelines in pharmacies has been dismal (Chalker and others 2000). However, training of physicians, midwives, and pharmacy workers can lead to greatly improved STD syndromic management (Garcia and others 2012). After training of pharmacy workers, pharmacy-based STD syndromic management was cost-effective, when only program costs are used, and cost saving from the societal perspective (Adams and others 2003). c. For this volume, we are assuming that most clinical service delivery at the primary care and reproductive health clinics level is provided by nurses. Primary health clinics in LICs and MICs tend to lack diagnostic testing but also have lower costs and are more accessible than hospitals. d. Service delivery by physicians, physician assistants, or nurses. Specialist expertise includes reproductive health, laboratory capacity, obstetrics and gynecology, and pediatrics. e. Curricula should include information on condoms, safe-sex promotion and provision, warning signs, and accessing care. f. Sanchez and others 1998. Tuberculosis the respiratory route and most commonly affects the TB is arguably the world’s leading cause of death from lungs but can damage any tissue. Only a minority an infectious agent.3 The WHO estimates that 10.4 million (approximately 10 percent) of individuals infected with new cases and 1.5 million deaths occur from TB each year M. tuberculosis progress to active TB disease, while the (WHO 2016a). One-third of TB cases remain unknown remainder may maintain a latent infection that serves as to the health care system. For those accessing treatment, a reservoir. TB has special challenges, including (a) a however, prevalence and mortality have declined signifi- substantial number of patients with active disease are cantly, and millions of lives have been saved. asymptomatic, capable of transmitting infection without TB is caused by the bacterium Mycobacterium tuber- knowing it; (b) patients must maintain compliance with culosis, which is transmitted between humans through treatment for six to nine months; and (c) the pathogen 10 Major Infectious Diseases persists in many infected individuals in a latent state for • Expand preventive therapy to reduce transmission many years but can be reactivated over a lifetime to cause from TB patients to contacts, especially to children disease and become transmissible. and HIV-positive individuals. People at every rung of the socioeconomic ladder are at • Emphasize community-based delivery of TB treat- risk, although TB disproportionately affects the poor. ment and services wherever possible to improve Approximately 80 percent of patients reside in 22 high- treatment completion, reduce the dangers of hospital burden countries. Treatment of TB disease requires multi- transmission, decrease costs, and improve patient ple drugs for many months. These lengthy drug regimens quality of life. are challenging for both patients and health care systems— • Improve hospital and clinic infection control. especially in LMICs, where the disease burden often far • Enhance drug supply chains for access to TB outstrips local resources. For TB susceptible to first-line treatments that have small markets. drugs (the least expensive), cure rates greater than • Expand information technology and electronic 90 percent are expected at a cost of US$200 to US$500. The medical records to enable more effective disease control. increasing incidence of multidrug-resistant TB (MDR-TB), which requires even longer treatment regimens with The need is urgent for new tools, including inexpen- expensive and difficult-to-tolerate drugs, represents an sive and sensitive point-of-care diagnostic tests, rapid emerging threat, not least to hospital and clinic personnel. tests for drug resistance, new and shorter drug regimens The United Nations’ (UN) Sustainable Development for both drug-susceptible and drug-resistant TB, and a Goal (SDG) 3 seeks to end the TB epidemic altogether by more effective vaccine to prevent the disease. 2030, but the decline in incidence of TB has been slow, only about 1.5 percent per year.4 Without new tools, the UN targets are unlikely to be met even by 2050. The current Malaria and Other Adult Febrile Illnesses policy of passive case finding (waiting for patients to be ill Febrile illnesses are major causes of morbidity and mor- enough to seek treatment) is suboptimal in high-burden tality in LMICs for children and adults, and most are countries. Faster rates of progress on TB will require ear- largely indistinguishable on clinical presentation. lier, more accurate case detection; rapid commencement Simple rapid diagnostic tests (RDTs) are lacking for the of and adherence to effective treatment; and, where possi- common, serious causes of fever except malaria, making ble, preventive treatment of latent TB (table 1.4). appropriate treatment uncertain for most febrile Durable control will require new strategies and tools patients, only a minority of whom have malaria. that are more effective than those now in use—for exam- ple, new, shorter drug regimens that are effective for both Malaria drug-sensitive and drug-resistant TB. These must be not The massive investment in malaria control over the only cost-effective but also affordable and capable of past decade has been successful in greatly reducing being effective on a large scale. In addition to new tools, malaria prevalence, but eliminating malaria is a very effective TB control requires the strengthening of weak distant goal in most of Sub-Saharan Africa and much of health care systems (including improvements in surveil- Asia. Continued progress depends on maintaining lance, information technology, logistics, and drug supply) and increasing the use of effective preventive measures and strengthening of community health care systems to (such as insecticide-treated nets, indoor residual spray- be more responsive and effective. ing, and intermittent preventive therapy for pregnant Within the context of current knowledge, Bloom women and infants); widespread use of RDTs; and and others (2017) in chapter 11 in this volume advocate treatment with effective artemisinin-combination ther- for optimizing the approaches known to be effective, apies (ACTs) to bring the countries with the highest including the following: endemic rates to preelimination levels (Shretta and others 2017, chapter 12 in this volume). Table 1.5 sum- • Identify high-transmission countries and hot spots marizes the essential interventions for prevention and within countries where targeted efforts can be more treatment of malaria. effective and cost-effective. Continued surveillance and substantial expenditures • Increase early TB detection and diagnosis, particu- over many years will be needed to eventually eradicate larly in selected high-burden countries, by introduc- malaria, and whether it can be done globally without tion of new tools for active case finding. adding at least one more effective tool to the set of inter- • Rapidly provide appropriate and better maintenance ventions in widespread use is unclear. The currently for patients diagnosed with either drug-susceptible TB or available vaccines may or may not be effective enough MDR-TB, enabling higher levels of completion and care. to boost results sufficiently. In April 2017, the WHO Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 11 Table 1.4 Essential Tuberculosis Intervention Package, by Delivery Platform Delivery platform Second- and Intervention Nationwide policies Community health post third-level type and regulations or pharmacy Primary health center First-level hospital hospitals Surveillance and 1. Passive case finding 3. Symptomatic disease detection 2. Active case finding surveillance in high-burden 4. Active contact tracing countries of TB-positive patients Data collection 5. Information systems and patient tracking Diagnosis and 6. National guidelines 7. Symptomatic 9. Sputum smears 12. GeneXpert/RIFa or drug sensitivity promoting diagnosis, local sputum 10. Testing of children and culture for diagnosis testing the provision smears household members of drug-susceptible TB Relapse and of diagnostic 8. Referral for diagnosis and HIV+ individuals reinfection labs; diagnostic and drug-susceptible for case finding in both diagnosis technology including TB tests drug-susceptible and GeneXpert or culture MDR-TB cases for drug-susceptible TB; fixed/mobile 11. Availability of fixed/ X-ray; and training mobile X-ray for diagnosis Treatment 13. WHO guidelines: 14. Provision and 16. Treatment of drug- of drug- four-drug regimen observation of treatment susceptible TB susceptible TB for two months, then after one month at first- until transmission two drugs-regimen level hospital is reduced (one for four months 15. Use of cell-phone SMS month), then transfer to support treatment of treatment to adherence community level Treatment of 17. WHO guidelines: 18. Provision of 20. Treatment until 21. Specialized drug-resistant TB Multiple-drug appropriate second- sputum is negative treatment regimen after line drugs, monitoring or GeneXpert is for treatment drug-susceptible 19. INH preventive therapy negative; treatment failures, TB testing for nine as outpatients after MDR-TB, months to two years sputum is negative surgery Coinfection 22. Provider incentives 23. Referral or provision 25. Separate areas in with HIV to improve quality of of HIV treatment as health facilities TB care appropriate for TB to avoid 24. Information systems to transmission to link diagnostic hospital AIDS patients care to outpatient and community care Note: Interventions shown in orange indicate areas that are relatively neglected by governments. HIV = human immunodeficiency virus; HIV+ = HIV-positive; INH = isoniazid; MDR-TB = multidrug-resistant tuberculosis; SMS = short message service (text messaging); TB = tuberculosis; WHO = World Health Organization. a. GeneXpert/RIF refers to a new test that simultaneously detects Mycobacterium TB complex (MTBC) and resistance to rifampin (RIF). announced that Ghana, Kenya, and Malawi will partici- cannot be reliably applied and malaria prevalence is high pate in a pilot malaria vaccine implementation program (Babigumira, Gelband, and Garrison 2017, chapter 15 in in select areas, beginning in 2018 (WHO 2017). this volume). When the test for malaria is negative, patients Despite global guidelines to the contrary, presumptive with severe disease should receive an antimicrobial regi- treatment of undifferentiated febrile illness as malaria is men tailored to locally important nonmalarial pathogens still appropriate in places where RDTs (or microscopy) (Crump and others 2017, chapter 14 in this volume). 12 Major Infectious Diseases Table 1.5 Essential Malaria Intervention Package, by Delivery Platform Delivery platforma Second- Population- and based health Health First-level third-level Intervention type interventions Community center hospital hospitals All malaria-endemic countries Case management: 1. Prophylaxis 2. Diagnosis with RDTs or microscopy, Uncomplicated for travelers including parasite species malaria (or fever) 3. Treatment with ACTs (or current first- line combination) for malaria-positive individuals where diagnosis is available 4. Where both RDTs and microscopy are unavailable and malaria is common, presumptive treatment with ACTs for nonsevere suspected malaria; if severe, ACTs plus antibiotics 5. Plasmodium vivax: Chloroquine alone or chloroquine plus 14-day course of primaquine (for G6PD normal individuals) 6. Case investigation, reactive case detection, proactive case detection (including mass screening and treatment) Case management: 7. Single-dose rectal artesunate, then 8. Parenteral Severe malaria referral to first-level hospital artesunate, then full-course ACTs Vector control: ITNs 9. ITNs available in health centers and antenatal clinics and via social marketing Malaria elimination countries 10. Mass drug administration to high-risk groups in geographic or demographic clusters 11. Single low-dose primaquine added to first- line treatment Malaria control countries Vector control: IRS 12. IRS in selected areas with high transmission and entomologic data on IRS susceptibility Vector control: 13. Larviciding and water management in Larviciding and water specific circumstances where breeding management sites can be identified and regularly targeted Mass drug 14. IPTp, IPTi, and SMC Sahel region administration Note: Interventions shown in orange indicate areas that are relatively neglected by governments. ACTs = artemisinin-combination therapies; G6PD = glucose-6-phosphate-dehydrogenase; IPTi = intermittent preventive treatment in infants; IPTp = intermittent preventive treatment of pregnant women; IRS = indoor residual spraying; ITN = insecticide-treated net; RDT = rapid diagnostic test; SMC = seasonal malaria chemoprevention. a. All interventions listed for lower-level platforms can be provided at higher levels. Similarly, each facility level represents a spectrum and diversity of capabilities. The column in which an intervention is listed is the lowest level of the health care system in which it would usually be provided. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 13 Where an understanding of locally important blood- (Crump and others 2017, chapter 14 in this volume). stream infections and other pathogens is lacking, stan- Concomitantly, research to identify priorities for dardized fever etiology research is needed to inform improvements in management, such as selection of management. The development of accurate point-of-care empiric antimicrobial therapies, should be undertaken diagnostic or biomarker tests would improve targeting in the same countries. of antimicrobials. The laboratory methods that can be used for research are impractical at the bedside in low-resource settings. For such settings, accurate RDTs are needed—first, to dis- Nonmalarial Fever tinguish viral from bacterial (and potentially easily treat- A diverse set of pathogens contributes to nonmalarial able) infections; and second, to provide pathogen-specific fever. Prevention efforts may target pathogen reservoirs tests for major causes of treatable nonmalarial fevers, based (for example, by vaccinating livestock for brucellosis); tar- on surveillance and other local research. get sources of infection (such as through vector control to Finally, cost and outcome data are needed to develop reduce arbovirus infections); interrupt transmission (for credible estimates of the total burden of nonmalarial example, by reducing occupational exposure to Coxiella febrile illnesses and to enable accurate cost-effectiveness burnetii among abattoir workers); and provide immuno- analyses related to fever in order to strengthen logic protection (such as through typhoid vaccines).5 resource-stratified approaches to the adoption and A lack of knowledge and a lack of tools hamper prog- integration of interventions (summarized in table 1.6). ress in combating nonmalarial fevers. The predominant This information is particularly important because causes of fever in LMICs are largely unknown because decisions on services to include in universal health cov- research on fever etiology has not been done. National erage are being made. surveillance or sentinel site studies, preferably coordi- nated globally, are urgently needed to identify major Viral Hepatitis causes of severe febrile illness, especially bloodstream infections and pathogens with specific treatments Five mostly unrelated viruses—hepatitis A, B, C, D, and (for example, brucellosis, rickettsioses, and Q fever) E—infect the liver, with varied routes of infection: Table 1.6 Essential Intervention Package for Adult Febrile Illness, by Delivery Platform Delivery platforma Community Second- and Intervention Nationwide policies health post third-level type and regulations or pharmacy Primary health center First-level hospital hospitals Case 1. Standard practice 3. Evaluation for malaria with 5. Clinical history and 8. Reference management: guidelines RDT or microscopy (see examination to identify source diagnostics for All fevers 2. Essential medicines, malaria interventions) of fever major causes including relevant 4. If negative for malaria, 6. Evaluation for malaria and HIV of nonmalarial antibacterials referral if fever persists fever 7. Treatment for the apparent beyond seven days cause and reevaluation after one week Case 9. Prereferral antimicrobial 10. Emergency management of 14. Reference management: according to standard septic shock with intravenous diagnostics for Severe practice guidelines (for fluids, supplemental oxygen, major causes febrile illness example, extended- and antimicrobial according to of nonmalarial spectrum cephalosporin) standard practice guidelines fever 11. Clinical history and physical examination to identify source of fever 12. Blood culture before antimicrobial; hemoglobin and glucose measurement 13. Treatment of apparent cause table continues next page 14 Major Infectious Diseases Table 1.6 Essential Intervention Package for Adult Febrile Illness, by Delivery Platform (continued) Delivery platforma Community Second- and Intervention Nationwide policies and health post third-level type regulations or pharmacy Primary health center First-level hospital hospitals Prevention: 15. National policy on Vaccines typhoid vaccines 16. National policy on control of brucellosis and leptospirosis in livestock Prevention: 17. National policies on Nonvaccine control of sources of measures nationally important causes of nonmalarial fever (such as vector control for arbovirus infections) 18. National policies on interruption of transmission of nationally important causes of nonmalarial fever (for example, management of occupational exposure to Coxiella burnetii among abattoir workers) Surveillance 19. Nationwide or sentinel site surveillance to identify major causes of severe febrile illness, especially bloodstream infections 20. Assurance that national recommendations for antimicrobial management of severe febrile illness match etiologic findings Note: HIV = human immunodeficiency virus; RDT = rapid diagnostic test. a. All interventions listed for lower-level platforms can be provided at higher levels. Similarly, each facility level represents a spectrum and diversity of capabilities. The column in which an intervention is listed is the lowest level of the health care system in which it would usually be provided. • Hepatitis A and E are transmitted by the fecal-oral reuse of medical equipment and substandard appli- route through contaminated water and food; they can cation of infection control measures) or through also be transmitted sexually. unsafe practices among people who inject drugs. • Most hepatitis B (HBV) infections occur through Sexual transmission of hepatitis C is rare in mother-to-child and early-life horizontal transmis- heterosexual couples but more common among sion between family members, among adults through HIV/AIDS-infected MSM. sexual intercourse, and through unsafe injection • Hepatitis D is transmitted by blood and bodily fluids. practices and transfusion of unscreened blood. • Most hepatitis C (HCV) infections occur through Most hepatitis deaths (96 percent) are caused by HBV unsafe injections, either in medical settings (from and HCV, which cause chronic, lifelong infection Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 15 resulting in progressive liver damage leading to cirrhosis Hepatitis Treatment and hepatocellular carcinoma. Mortality rates from hep- Chronic HBV and HCV infections can be treated effec- atitis are highest in West Africa and parts of Asia; in tively. The new direct-acting antiviral medicines for absolute numbers, East Asia and South Asia account for hepatitis C can cure more than 90 percent of individu- just over half of hepatitis deaths, which totaled 1.45 mil- als with chronic infection with a two- to three-month lion globally in 2013. An estimated 250 million people course of treatment, although the current costs of treat- live with chronic HBV infection; 80 million have chronic ment are very high. Hepatitis C treatment could also HCV infection (Gower and others 2014; Schweitzer and reduce HCV transmission because people who have others 2015). been cured do not transmit the infection. There is no In some West African countries, more than 8 percent cure for chronic hepatitis B, but effective antiviral treat- of the population is infected with hepatitis. The ments can suppress viral replication and prevent disease regions with the highest prevalence of HCV infection progression. Table 1.7 summarizes both the preventive are West and Central Africa, Eastern Europe, and and the treatment interventions for hepatitis. Central Asia. Hepatitis C prevalence is extremely high in a few other countries as well, most notably the Arab Republic of Egypt and Pakistan, where high incidence Neglected Tropical Diseases persists largely because of weak preventive measures, NTDs affect more than 1 billion of the poorest, most such as reuse of syringes and needles in health care marginalized people of the world. These infections are settings. a consequence of the environmental and socioeco- nomic conditions in which the poor live, and the ill Hepatitis Prevention health and disability they cause are a primary factor Hepatitis A and E infections can be prevented through locking the poor into poverty. At least 18 diseases are improved sanitation. Although no reliable estimates are recognized as NTDs by World Health Assembly resolu- available, the incidence of hepatitis A and E has declined tions.6 Although not covered further here, the WHO likely as part of the overall reduction in the number of has recently added snakebite deaths to the NTD list. deaths owing to diarrhea. An effective hepatitis A vac- Snakebite causes about 50,000 deaths in India per year cine exists, and 18 countries have introduced universal and an estimated 100,000 deaths globally (Mohapatra childhood hepatitis A vaccination. and others 2011). The most notable achievement in hepatitis preven- The NTD concept was developed to draw attention tion is the reduction in incidence of acute and chronic to a disease control opportunity that had been HBV infection as a result of universal childhood overlooked by the Millennium Development Goals. hepatitis B vaccination. At the end of 2013, 183 of The ending of NTD epidemics is now embedded 194 countries had introduced universal childhood within the SDGs for 2030, under target 3.3, reflecting vaccination; global coverage with three doses of the UN’s High-Level Political Forum on Sustainable hepatitis B vaccine is estimated to be 81 percent effec- Development 2016 promise of “ensuring that no one is tive (WHO 2015). Universal infant vaccination with left behind.”7 Chapter 17 of this volume (Fitzpatrick high coverage levels has led to major reductions in the and others 2017) focuses on specific WHO targets for prevalence of chronic HBV infection among children. control, elimination, and eradication of a subset of In China, the prevalence of chronic HBV infection these diseases. declined from approximately 8 percent in 1992 to 1 percent in 2006 among children ages one to four years (Liang and others 2009). Interventions to End NTDs However, challenges remain in achieving further Three key interventions address a large share of the reductions in incidence. Full protection for children burden of disease caused by this set of diseases. requires that they receive the first vaccine dose within In recognition of the increasingly integrated delivery 24 hours of birth, which is a logistical challenge and a of interventions to the poorest, most remote, and barrier to further progress. otherwise most marginalized communities of the Other proven interventions for hepatitis prevention world, we consider them by intervention rather than that have not been fully implemented around the world by disease, as follows: (for various technical and political reasons) are universal safe injections, blood supply screening for HBV and HCV, • Preventive chemotherapy by mass drug administration and harm reduction for injection drug users (for exam- • Innovative and intensified disease management ple, provision of sterile needles and opioid substitutes). • Vector ecology and management 16 Major Infectious Diseases Table 1.7 Essential Hepatitis Intervention Package, by Delivery Platform Delivery platforma Second- Community and third- Nationwide policies and health post or level Intervention type regulations pharmacy Primary health care First-level hospital hospitals Hepatitis B 1. Policy for universal 2. Delivery of hepatitis B vaccination newborn and childhood vaccination including vaccination birth dose Interventions to 3. Policy for hepatitis B 4. Vaccination of health reduce hepatitis vaccination of health care workers transmission care workers in health care settings Harm-reduction 5. Policy for the provision services for IDU of harm-reduction services (including injection equipment and opioid substitution therapy) to IDU; use of this wording for HIV or STI safe injection* 6. Community services: IDU-friendly harm reduction with sufficient coverage Hepatitis testing 7. National testing policy 8. Hepatitis 9. Referral of persons services identifying priority testing of with hepatitis infection groups for testing individuals as to care and setting a testing identified in strategy the national testing policy Hepatitis treatment 10. Treatment guidelines 11. Referral of persons 13. Treatment of with hepatitis infection hepatitis B and C to assessment for for eligible persons treatment eligibility 14. Mentoring of 12. Assuming sufficient primary-care training, initiation of personnel involved hepatitis treatment and in treatment follow-up initiation and follow-up 15. Screening blood transfusion for hepatitis B and C Note: Interventions shown in orange indicate areas that are relatively neglected by governments. HIV = human immunodeficiency virus; IDU = injection drug users; STI = sexually transmitted infection. Interventions marked with an asterisk (*) should be closely integrated with HIV/AIDS and STI prevention and treatment interventions. a. All interventions listed for lower-level platforms can be provided at higher levels. Similarly, each facility level represents a spectrum and diversity of capabilities. The column in which an intervention is listed is the lowest level of the health care system in which it would usually be provided. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 17 The interventions are discussed in detail in chapter 17 The NTDs for which the primary intervention is disease in this volume (Fitzpatrick and others 2017) but are management are Buruli ulcer, Chagas disease, human summarized as follows: African trypanosomiasis (HAT), leishmaniasis, leprosy, Preventive chemotherapy by mass drug administration and yaws. is effective against lymphatic filariasis, onchocerciasis, Vector ecology and management aims to control the schistosomiasis, soil-transmitted helminthiases, and tra- transmission of the causative pathogens of insect-borne choma. The specific drugs and regimens vary by disease, NTDs with proven interventions that are applied in an and many populations are affected by more than one of ecologically friendly manner. The main NTDs for which these conditions. Mass campaigns can be combined to this is an important strategy are Chagas disease, dengue, target several pathogens at once. chikungunya, visceral leishmaniasis (kala azar), and Zika Innovative and intensified disease management refers to virus. Table 1.8 summarizes the essential interventions a shift from passive management to active surveillance, for preventing and treating NTDs. early diagnosis, and treatment, with the aim to eliminate or control, not just to manage. Treatment of Buruli ulcer, Recent Progress against NTDs for example, has evolved from late-stage surgical removal Since the NTD concept took hold, substantial successes of infected or dead tissue and correction of deformity have been recorded, including a reduction in deaths to the early-stage use of antibiotics. The gains go caused by visceral leishmaniasis, rabies, schistosomiasis, beyond health benefits to include reductions in hospital- HAT, Chagas disease, and soil-transmitted helminthiases ization costs to health care systems and to individuals. (among which, for example, ascariasis is estimated to Table 1.8 Essential Intervention Package for Neglected Tropical Diseases, by Delivery Platform Delivery platforma Second- and Intervention Community health post or Primary First-level third-level type Nationwide policies or regulations pharmacy health care hospital hospitals Preventive 1. Integrated guidelines and strategy 2. Mass drug administration chemotherapy on the coordinated use of preventive for lymphatic filariasis, chemotherapy for NTDs onchocerciasis, schistosomiasis, soil- transmitted helminthiases, trachoma, and food-borne trematodiases as appropriate Innovative and 3. Integrated guidelines and strategy 4. Lymphedema management 7. Hydrocele intensified disease for skin-related NTDs including (in 5. Early detection and treatment of and trichiasis management addition to those listed elsewhere Chagas disease, human African surgery in this table) Buruli ulcer and trypanosomiasis, leprosy, and mycetoma leishmaniases 6. Total community treatment for yaws Vector ecology 8. Integrated vector management 9. Sustained vector management and management guidelines and strategy for Chagas disease, dengue, and visceral leishmaniasis Veterinary public 10. Not covered in DCP3 chapter; for interventions for the control of echinococcosis and rabies, see World Bank (2012). health services Water, sanitation, 11. See interventions in chapter 9 of DCP3 volume 7 (Hutton and Chase 2017). and hygiene Note: DCP3 = Disease Control Priorities, third edition (Jamison and others 2015–18); NTDs = neglected tropical diseases. a. All interventions listed for lower-level platforms can be provided at higher levels. Similarly, each facility level represents a spectrum and diversity of capabilities. The column in which an intervention is listed is the lowest level of the health care system in which it would usually be provided. 18 Major Infectious Diseases have caused 142,000 deaths in 2012, down from about Antibiotic consumption increased by an estimated 220,000 in 2000) (WHO 2014). In addition, the follow- 30 percent or more in 71 countries between 2000 and ing results were recorded: 2010, reaching approximately 70 billion standard units (single-dose units) in 2010 (Van Boeckel and others • New HAT cases have fallen by 80 percent between 2014). This increase was primarily in first-line classes of 2000 and 2014, to an estimated total of fewer than antibiotics, including penicillins and cephalosporins, 4,000 cases per year. which together make up more than half of global con- • The number of cases of visceral leishmaniasis sumption. Use of last-resort antibiotic classes, especially (kala azar) in Bangladesh, India, and Nepal fell by carbapenems and polymyxins, also increased. 75 percent between 2005 (when a regional program Despite the recent increases in antibiotic consump- was launched) and 2014, to a reported 10,209 cases. tion worldwide, with few exceptions, per capita con- • In 2000, more than 130,000 cases of dracuncu- sumption in LMICs is much lower than in HICs. liasis (Guinea worm disease) were reported; in Alongside increasing consumption and rising rates of 2015, only 22 cases were reported, reflecting near antibiotic resistance, lack of access to antibiotics is still a eradication. serious concern for most LMICs. Each year, pneumonia kills approximately 1 million children under age Much of the burden of NTDs occurs with morbid- five years, and an estimated 445,000 could be saved with ity rather than mortality—and here, too, the progress the universal provision of antibiotics for community- has been good, albeit somewhat less dramatic: the total acquired pneumococcal infections (Miller-Petrie, Pant, number of DALYs decreased by 19 percent between and Laxminarayan 2017, chapter 18 in this volume). 2000 and 2012, from 1.0 percent of the GBD to When they are available, first-line antibiotic treatments 0.8 percent (WHO 2014). are still relatively affordable, but newer antibiotics needed to treat resistant infections may be out of reach for LMICs (Miller-Petrie, Pant, and Laxminarayan 2017, ANTIMICROBIAL RESISTANCE chapter 18 in this volume).8 Every use of an antibiotic, whether appropriate or inap- propriate, exerts selection pressure, giving resistant bac- Interventions to Ensure Appropriate Antibiotic Use teria an advantage and accelerating the development of Certain interventions are effective at reducing antibiotic resistance. Bacterial resistance to first-line, second-line, use or increasing appropriate use, but their effects on anti- and last-resort antibiotics is growing wherever it has biotic resistance rates are difficult to determine because of been monitored (CDDEP 2016). Increased travel, trade, the long timeline for effects to become apparent. Therefore, and migration mean that resistant bacteria can spread recommendations are based largely on success in changing faster than ever (Du and others 2016; Johnson and patterns of use. Interventions aim to both reduce the need Woodford 2013). for antibiotics by preventing infections and reduce the The burden of antimicrobial resistance falls heavily inappropriate or unnecessary use of antibiotics (in both on LMICs. They typically have high burdens and humans and animals). The broad categories of interven- rapid spread of infectious disease, poor nutrition, tions are as follows (summarized from Miller-Petrie, Pant, and increasing rates of antibiotic consumption in humans and Laxminarayan 2017, chapter 18 of this volume): and animals, in addition to weaker health care systems and sparse standards and regulations governing access, • Reduce and eventually phase out subtherapeutic anti- use, and quality of antibiotics (Okeke and others 2005). biotic use in agriculture. Improved sanitation and hygiene at the farm level would reduce the need for prophylactic antibiotics. Antibiotic use in ani- Drivers of Increased Antibiotic Use mal agriculture should be reduced, focusing on the The increase in antibiotic use is driven by the burden of involvement of farmers and the agricultural industry infectious disease as well as by economic, behavioral, in carefully phasing out the use of growth promot- environmental, and structural factors. For instance, ers and premixed animal feeds (Laxminarayan, Van expanded insurance coverage and increased physician Boeckel, and Teillant 2015). density intensify the consumption of antibiotics (Klein • Adopt incentives that encourage antibiotic stewardship and others 2015; Zhang, Lee, and Donohue 2010). and discourage overuse. Ensuring that payments are Decision fatigue and patient demand also increase anti- not linked to prescribing and introducing rewards for biotic prescribing. compliance may improve prescribing patterns. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 19 • Improve hospital infection control and antibiotic stew- project, which provides results in international dollars ardship. Antibiotic stewardship programs, infection of a WHO region)9 could not be converted and were prevention and control, and especially handwashing omitted. with soap can reduce infections, antibiotic use, and Cost-effectiveness results depend on context. resistance while also improving patient outcomes. The cost-effectiveness of the same intervention in two • Educate health care professionals, policy makers, and different countries may vary depending on local costs; the public about sustainable antibiotic use. Although health interventions, on average, cost more in countries public awareness is growing that antibiotic resistance with higher income because of higher salaries. Vaccines presents a threat, there is little awareness of individ- generally cost more in countries not eligible for bulk ual actions to reduce antibiotic use. Patients, parents, purchasing discounts (such as prices of Gavi, the Vaccine health care providers, stakeholders, and hospital Alliance).10 In addition, the cost-effectiveness of vaccina- heads all need to be aware of what they can do to tion and screening programs (and some other interven- reduce unnecessary use. tions) vary according to prevalence of the condition; • Reduce the need for antibiotics through improved prevention programs are often more cost-effective where water, sanitation, and immunization. Disease preven- prevalence is higher. Cost-effectiveness may vary with tion achieves the dual purposes of keeping people comorbidities and with opportunities to deliver the healthy and saving antibiotic doses. Water, sanitation, intervention synergistically with other interventions hygiene, and vaccination should be core components (and therefore at lower cost). Usual care (the usual com- of any response, with financing from infrastructure parator of cost-effectiveness) may also vary. and health sectors. In some cases, the interventions are subdivided by • Ensure political commitment to meet the threat of anti- study location (for example, southern Africa or biotic resistance. Without national commitment in Southeast Asia) or by low-income or middle-income the form of implemented action plans, the long-term country designation. Where neither is specified, the sustainability of efforts to curb antibiotic resistance results apply to low- and lower-middle-income coun- will be weakened. Although international efforts to tries, which are the main emphases of this third edition curb antibiotic resistance have focused largely on of Disease Control Priorities (Jamison and others national action, international support is also needed. 2015–18). Fewer results for upper-middle-income countries are included. HIV/AIDS interventions account for almost half of the interventions and stud- COST-EFFECTIVENESS OF INTERVENTIONS ies, consistent with its share of funding relative to other health conditions. FOR ADULT INFECTIOUS DISEASE Figure 1.1 represents a reductionist view of the large The substantial mobilization of donor resources by literature on cost-effectiveness of infectious disease organizations including The Global Fund to Fight AIDS, interventions and should be interpreted with caution, Tuberculosis, and Malaria; The President’s Emergency especially when comparing results of two different stud- Plan for AIDS Relief; and the President’s Malaria ies that rely on inconsistent underlying assumptions. Initiative has been accompanied by efforts to ensure However, more than half of the interventions listed in value for money, which has led to a substantial literature figure 1.1 cost less than US$100 per DALY prevented, on the costs and cost-effectiveness of interventions to suggesting that they could be cost-effective even in the combat some major infectious diseases (but little beyond poorest countries. These interventions include some that these diseases). are preventive, such as providing female condoms to sex Figure 1.1 summarizes various estimates of workers in South Africa (although in practice, wide- cost-effectiveness, measured per DALY prevented—the spread use has been difficult to achieve); undertaking metric most commonly used in economic studies to voluntary male circumcision in high-incidence African compare cost-effectiveness across different health inter- countries; and supplying insecticide-treated nets in ventions. The estimates are summarized from expert Africa to prevent malaria. Biomedical treatment inter- searches of the literature undertaken for the chapters in ventions costing less than US$100 per DALY prevented this volume. For full details of the individual studies include treating severe malaria with artesunate; screen- used, along with bibliographic references, see annex 1B. ing and treating pregnant women for syphilis; treating All cost-effectiveness results have been translated into malaria (with ACTs); treating TB (with first-line drugs); 2012 U.S. dollars for comparability. A few studies and, for various NTDs, providing detection and treat- (particularly publications from the WHO-CHOICE ment and preventive biomedical therapy for some con- (CHOsing Interventions that are Cost-Effective) ditions in endemic areas. 20 Major Infectious Diseases Figure 1.1 Estimated Costs of Selected Infectious Disease Interventions Hepatitis C treatment, UMICs PrEP-ARV for noninfected partner, serodiscordant couples, S Af Online sex education to prevent STI, LAC BCC alone, sex establishments, LAC Scale up ART to all infected, lower-mid income, Af PrEP during pregnancy and breastfeeding, S Af Use Xpert to diagnose TB, MICs Give female condom to sex workers, S Af Vector control for dengue BCC plus regulation, sex establishments, LAC PMTCT Option A HIV versus no treatment, SE Asiaa Eradicate yaws (detect and treat) Screen and treat for syphilis, UMICs Treat TB with second-line drugs, MICs Scale up ART to all infected, S Af Add syphilis screen to HIV screen and treat, UMICs Home presumptive treatment malaria, Africa Detect and treat human African trypanosomiasis Supply ITNs for malaria, Africa Add Xpert to smear to diagnose TB, LMICs Treat smear -ve TB with first-line drugs, LICs Comprehensive malaria mgmt (spray, nets, treat), Africa IRS for malaria, Africa Detect and treat leprosy Intermittent preventive treatment malaria in pregnancy, Africa Hepatitis B vaccination Preventive chemotherapy for trachoma Intermittent preventive treatment malaria in infants, Africa PMTCT Option B HIV versus no treatment, Africaa Preventive chemotherapy for schistosomiasis and STH Treat malaria with ACT, Africa Detect and treat visceral leishmaniasis Treat smear +ve TB with first-line drugs, LICs Screen and treat for syphilis, LICs Preventive chemotherapy, lymphatic filariasis Add syphilis screen to HIV screen and treat, LICs Preventive chemotherapy for onchocerciasis Treat severe malaria with artesunate, Africa and SE Asia Voluntary male circumcision 1 10 100 1,000 10,000 100,000 Cost range per DALY prevented, US$ (2012) HIV & STI TB Malaria Hepatitis Neglected tropical diseases Sources: Estimates based on sources listed by subject in annex 1B of this chapter. Note: ACT = artemisinin-combination therapy; ART = antiretroviral treatment; ARV = antiretroviral; BCC = behavior change communications; DALY = disability-adjusted life year; HIV = human immunodeficiency virus; IRS = indoor residual spraying; ITNs = insecticide-treated nets; LAC = Latin America and the Caribbean; LICs = low-income countries; LMICs = low- and middle-income countries; MICs = middle-income countries; PrEP = preexposure prophylaxis (provision of ART to noninfected individuals at risk); PMTCT = prevention of mother-to-child transmission; S Af = southern Africa; SE Asia = Southeast Asia; STH = soil-transmitted helminthiases; STI = sexually transmitted infection; TB = tuberculosis; UMICs = upper-middle-income countries. a. “Option A” is a two-drug ART regimen in pregnancy and postpartum for HIV-positive mothers. “Option B” is a three-drug ART regime in pregnancy and postpartum for HIV-positive mothers. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 21 Another group of interventions costs US$100–US$399 virus outbreaks have reminded us over the past couple per DALY prevented and thus would be considered of years. Through basic and translational research, cost-effective (less than the per capita annual income) of some of the most devastating diseases of human- all but the poorest four to five countries. This second kind—polio, diphtheria, measles, and tetanus—have group includes providing ART for people with been dramatically reduced, and smallpox has been HIV/AIDS, with pregnant women being a particularly eradicated. Meanwhile, new pathogens emerge, and high priority (to prevent transmission to their children newly drug-resistant organisms represent continuing while also treating HIV in the mother). This group of and unpredictable threats. interventions also includes intermittent preventive treat- Four main challenges will need to be met to achieve ment of malaria in infants and pregnant women as well meaningful progress in the fight against the diseases as treatment of MDR-TB with second-line (more expen- addressed in this volume. sive) drugs. Some interventions, such as those requiring behavioral change or those being implemented in Latin America, 1. Focusing and Targeting of Intervention Strategies cost more than US$400 per DALY prevented. These If we have relearned something in the decade since include programs aiming to change sexual behavior as DCP2, it is that infectious diseases are not distributed well as vector control interventions for dengue. PrEP, uniformly—not across continents, not across coun- which includes provision of ART to the uninfected part- tries, not even in communities. Strategies should be ner in an HIV-serodiscordant couple, varies considerably designed to understand and respond appropriately to in cost-effectiveness, ranging from being cost saving to disease hot spots and key populations, thus ensuring costing more than US$5,000 per DALY prevented, access to the most effective interventions for the right depending on the context. populations in the right places at the right time, espe- Interventions can be cost-effective according to global cially in LMICs, where the disease burden is greatest. norms but still too expensive for most LICs to provide to Given the prodigious heterogeneity in the distribution everyone in need, especially where prevalence of the of diseases (both geographically and across population condition is high. Examples include the provision of subgroups) and the scarcity of resources, efficiency is ART in low-income African countries (Alistar, Grant, increased by matching those resources to the popula- and Bendavid 2014) and the treatment of drug-resistant tions that would benefit the most. Globally, infectious TB (Fitzpatrick and Floyd 2012). diseases disproportionately affect people in LMICs, Certain other interventions are likely to be cost- which, at current rates of progress, will bear the bulk of effective, but no studies could be identified for the con- premature deaths from infectious disease in 2030. In text. For example, needle exchange programs for LMICs, the poorer, more marginalized, and often stig- injection drug users are expected to be cost-effective matized populations are most at risk and the hardest to given that HIV prevalence among this group world- reach. Whereas the need for treatment may become wide is 19 percent; such programs could also prevent obvious as it is demanded by the sick, the need and hepatitis B and C (Wilson and others 2015). However, demand for prevention are often poorly matched. cost-effectiveness estimates for needle exchange pro- Prevention efforts can track incidence only if we know grams were not identified for low- or lower-middle- where the incidence is highest, which requires purpose- income countries. ful, population-based surveillance. Reaching marginal- In yet other cases, studies evaluated interventions that ized populations will require not only dedicating have been superseded by more effective measures, or resources to them but also working to remove the several studies of the same intervention had widely stigma, discrimination, and taboos that hamper effec- divergent results. These have not been included among tive prevention and treatment. studies used for the cost-effectiveness analysis of infec- tious disease interventions (for example, PMTCT Option A has now been supplanted by Option B/B+). 2. Scale-Up of Interventions against Major Infectious Diseases Although we have evidence of interventions that work to prevent and treat disease, in many cases, those ser- CONCLUSIONS vices have not been implemented at the scale necessary The variety and distribution of infectious diseases to sufficiently reduce incidence and the resulting mor- have evolved over time, and they will continue to chal- bidity and mortality. Scaling up these interventions lenge the global community—as the Zika and Ebola requires that they be tailored for distribution at the 22 Major Infectious Diseases lowest appropriate level of health care service delivery and key inputs (drugs, diagnostics, and vaccines) but and that staff is adequately trained and supervised. also on financing research and development and Areas ripe for this approach include training of phar- operational research to create new tools or to make macists in syndromic treatment of STIs and malaria far better use of existing tools (Hecht and others RDTs at the lowest health care level. However, scale-up 2012). In particular, as antimicrobial resistance con- will not be possible unless we address the next chal- tinues to increase, diseases once thought to be highly lenge: integration. treatable could, without significant investment in development of new drugs, pose a much more serious threat to global health. 3. Integration of Services More Effectively across A generation ago, as antibiotics were cheating death Disease Areas and vaccines were clearing diseases from entire conti- Reaching the world’s poor with an entire arsenal of nents, the world’s population believed that it was mov- specialized clinics is impossible. People are frequently ing into the postinfectious disease era. HIV seemed to affected by more than one condition: those who be the exception that proved the rule and, at any rate, are at risk for one NTD are also likely at risk for a would quickly be tamed with a vaccine. Instead, these number of others; sex workers are at risk for multiple genetically facile microorganisms, which have survival STIs, and HIV patients are especially susceptible to TB as their evolutionary goal, have taught us humility. infection. Moreover, treatment of infectious disease We will still be struggling against infectious diseases in has grown increasingly complex (for example, ART future generations, but the struggle will be far less and TB treatment) as well as costly (as evidenced by costly in lives and resources if we invest today in their the new hepatitis C treatment). Consequently, efforts control. to combat certain major diseases that have similar modes of transmission (such as HIV, STIs, and hepati- tis) would benefit from shared strategies for their ANNEXES prevention and diagnosis. Although specialized care is The following annexes to this chapter are available at clearly needed for especially rare or difficult-to-treat http://www.dcp-3.org/infectiousdiseases: conditions, standardizing prevention and treatment protocols and making information and communica- tion technologies available enable care to be pushed • Annex 1A: Forty Sexually Transmitted and Sexually out to integrated service points close to where patients Transmissible Pathogens live, study, and work. Proximity and integration of ser- • Annex 1B: Sources of Cost-Effectiveness Analysis for vices help ensure continuity of care from prevention or Selected Infectious Disease Interventions diagnosis through treatment and follow-up care. Addressing each of these challenges requires strength- ening of health care systems, including communica- ACKNOWLEDGMENTS tion; information technology; logistics, drug and The editors of this volume wish to thank Desiree vaccine supplies; and training of health care providers, Bernard, Elisabeth Gunningham, Varsha Malhotra, and which includes community health workers in LMICs. Alicair Peltonen for their valuable assistance on this Additionally, improving health care systems for infec- effort. They especially thank Brianne Adderley for her tious diseases, particularly those such as HIV, TB, and hard work keeping this large endeavor well organized. hepatitis that require extended medical treatment and monitoring, provides a basis and could serve as a model for improving care and treatment for noncom- NOTES municable diseases. World Bank Income Classifications as of July 2014 are as fol- lows, based on estimates of gross national income (GNI) per capita for 2013: 4. Development of New Technologies—Drugs, Vaccines, Diagnostics, Behavioral Interventions, and Delivery Methods—to Prevent and Treat These Diseases • Low-income countries (LICs) = US$1,045 or less • Middle-income countries (MICs) are subdivided: Most urgent is the need for the global community (a) lower-middle-income = US$1,046 to US$4,125 to invest in developing new antimicrobials. This (b) upper-middle-income (UMICs) = US$4,126 to approach includes rethinking global development US$12,745 assistance to focus not only on providing services • High-income countries (HICs) = US$12,746 or more. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 23 1. The U.S. Centers for Disease Control and Prevention interventions and programs that maximize health for (CDC) assesses the severity of HIV disease by cluster the available resources (“Cost-Effectiveness and Strategic of differentiation 4 (CD4) cell counts and the presence Planning (WHO-CHOICE),” WHO website, http://www of specific HIV-related conditions. The WHO Clinical .who.int/choice//en/). “An international dollar would Staging and Disease Classification System—which can buy in the cited country a comparable amount of goods be used readily in resource-constrained settings without and services a U.S. dollar would buy in the United access to CD4 cell count measurements or other diagnostic States (“What Is an ‘International Dollar’?” World Bank or laboratory testing methods—classifies HIV disease on Knowledge Base, https://datahelpdesk.worldbank.org the basis of clinical manifestations that clinicians can rec- /knowledgebase/). ognize and treat (U.S. Department of Health and Human 10. Gavi, the Vaccine Alliance is an international organization Services 2014). established in 2000 to bring together the public and private 2. For a comprehensive list of sexually transmitted and sex- sectors with the shared goal of creating equal access to new ually transmissible pathogens, see annex 1A at http://www and underused vaccines for children living in the poorest .dcp-3.org/infectiousdiseases. countries. For more information about Gavi’s vaccine 3. Whether TB or HIV causes more deaths depends on how pricing strategy, see Gavi, the Vaccine Alliance (2011). one allocates the deaths in coinfected individuals. There is no clear, correct answer. 4. SDG 3, titled “Good Health and Well-Being,” aims to REFERENCES “Ensure healthy lives and promote well-being for all at all ages.” It sets nine primary targets, including target Adams, E. J., P. J. García, G. P. Garnett, W. J. Edmunds, and 3.3: “By 2030, end the epidemics of AIDS, tuberculosis, K. K. Holmes. 2003. “The Cost-Effectiveness of Syndromic malaria and neglected tropical diseases and combat Management in Pharmacies in Lima, Peru.” Sexually hepatitis, water-borne diseases and other communicable Transmitted Diseases 30 (5): 379–87. diseases.” Alistar, S. S., P. M. Grant, and E. Bendavid. 2014. “Comparative 5. See table 14.2. “Nonmalarial Febrile Diseases: Exposure, Effectiveness and Cost-Effectiveness of Antiretroviral Diagnosis, Prevention, and Treatment” (Crump and others Therapy and Pre-Exposure Prophylaxis for HIV Prevention 2017, chapter 14 in this volume). in South Africa.” BMC Medicine 12 (1): 46. 6. The 18th World Health Assembly–recognized NTDs are Aral, S. O. 2011. “Utility and Delivery of Behavioural Buruli ulcer, Chagas disease, dengue and chikungunya, Interventions to Prevent Sexually Transmitted Infections.” dracunculiasis (Guinea worm disease), echinococcosis, Sexually Transmitted Infections 87 (Suppl 2): ii31–33. food-borne trematodiases, human African trypanosomi- Babigumira, J., H. Gelband, and L. P. Garrison Jr. 2017. “Cost- asis (sleeping sickness), leishmaniasis, leprosy (Hansen’s Effectiveness of Strategies for the Diagnosis and Treatment disease), lymphatic filariasis, mycetoma, onchocerciasis of Febrile Illness in Children.” In Disease Control Priorities (river blindness), rabies, schistosomiasis, soil-transmitted (third edition): Volume 6, Major Infectious Diseases, edited helminthiases, taeniasis/cysticercosis, trachoma, and yaws by K. K. Holmes, S. Bertozzi, B. R. Bloom, and P. Jha. (endemic treponematosis) (“Neglected Tropical Diseases,” Washington, DC: World Bank. WHO website, http://www.who.int/neglected_diseases Baeten, J. M. 2016. “Making an Impact with Preexposure /diseases/en/). Prophylaxis for Prevention of HIV Infection.” Journal of 7. The High-Level Political Forum on Sustainable Infectious Diseases 214 (12): 1787–89. Development is the UN’s central platform for the Baeten, J. M., D. Donnell, P. Ndase, N. R. Mugo, J. D. follow-up and review of the 2030 Agenda for Sustainable Campbell, and others. 2012. “Antiretroviral Prophylaxis for Development (UN 2015a), adopted at the United Nations HIV Prevention in Heterosexual Men and Women.” New Sustainable Development Summit on September 25, England Journal of Medicine 367 (5): 399–410. 2015 (“High-Level Political Forum on Sustainable Baral, S., C. Beyrer, K. Muessig, T. Poteat, A. L. Wirtz, and oth- Development 2016: Ensuring that No One Is Left Behind,” ers. 2012. “Burden of HIV among Female Sex Workers in UN Sustainable Development Knowledge Platform, Low-Income and Middle-Income Countries: A Systematic https://sustainabledevelopment.un.org/hlpf/2016). Review and Meta-Analysis.” The Lancet Infectious Diseases 8. As is represented by the photo on the cover of this volume, 12 (7): 538–49. the high global burden of infectious diseases (both in Baral, S., F. Sifakis, F. Cleghorn, and C. Beyrer. 2007. “Elevated humans and in animals), combined with the increasing Risk for HIV Infection among Men Who Have Sex with availability, use, and dispensing of antimicrobials (often Men in Low- and Middle-Income Countries 2000–2006: based upon syndromic diagnosis rather than on diagnos- A Systematic Review.” PLoS Medicine 4 (12): e339. tic tests), contribute to the global acceleration of antimi- Bloom, B. R., R. Atun, T. Cohen, C. Dye, H. Fraser, and crobial resistance. others. 2017. “Tuberculosis.” In Disease Control Priorities 9. The CHOICE (CHOsing Interventions that are Cost- (third edition): Volume 6, Major Infectious Diseases, edited Effective) project is a WHO initiative developed in 1998 by K. K. Holmes, S. Bertozzi, B. R. Bloom, and P. Jha. to provide policy makers with evidence for deciding on Washington, DC: World Bank. 24 Major Infectious Diseases Burnett, S. M., M. K. Mbonye, R. Martin, A. Ronald, S. Zawedd- Gavi Pledging Conference, London, June 13. http://www Muyania, and others. 2016. “Effect of On-Site Support on .gavialliance.org/library/publications/pledging-conference Laboratory Practice for Human Immunodeficiency Virus, -for-immunisation/6a--market-shaping--access-to-low-and Tuberculosis, and Malaria Testing.” American Journal of -sustainable-vaccine-prices/. Clinical Pathology 146 (4): 469–77. Gomez, G. B., A. Borquez, K. K. Case, A. Wheelock, A. Vassall, Canchihuaman, F. A., P. J. García, S. S. Gloyd, and K. K. Holmes. and C. Hankins. 2013. “The Cost and Impact of Scaling Up 2011. “An Interactive Internet-Based Continuing Education Pre-Exposure Prophylaxis for HIV Prevention: A Systematic Course on Sexually Transmitted Diseases for Physicians and Review of Cost-Effectiveness Modelling Studies.” PLoS Midwives in Peru.” PLoS One 6 (5): e19318. Medicine 10 (3): e1001401. CDC (U.S. Centers for Disease Control and Prevention). Gower, E., C. Estes, S. Blach, K. Razavi-Shearer, and H. Razavi. 1993. “Recommendations of the International Task Force 2014. “Global Epidemiology and Genotype Distribution for Disease Eradication.” Morbidity and Mortality Weekly of the Hepatitis C Virus Infection.” Journal of Hepatology Report 42 (RR-16): 8. 61 (1): S45–57. CDDEP (Center for Disease Dynamics, Economics and Policy). Grant, R. M., J. R. Lama, P. L. Anderson, V. McMahan, A. Y. Liu, 2016. “Resistance Map.” CDDEP, Washington, DC. http:// and others. 2010. “Preexposure Chemoprophylaxis for HIV www.cddep.org/projects/resistance-map. Prevention in Men Who Have Sex with Men.” New England Chalker, J., N. T. K. Chuc, T. Falkenberg, N. T. Do, and Journal of Medicine 363 (27): 2587–99. G. Tomson. 2000. “STD Management by Private Pharmacies Hader, S. 2017. “(Preventing) the Coming Epidemic: HIV in Hanoi: Practice and Knowledge of Drug Sellers.” Sexually in Youth.” Presentation to the annual Conference on Transmitted Infections 76 (4): 299–302. Retroviruses and Opportunistic Infections (CROI), Seattle, Chesson, H. W., P. Mayaud, and S. O. Aral. 2017. “Sexually WA, February 15. Transmitted Infections: Impact and Cost-Effectiveness of Harrison, O. B., K. Cole, J. Peters, F. Cresswell, G. Dean, Prevention.” In Disease Control Priorities (third edition): and others. 2017. “Genomic Analysis of Urogenital and Volume 6, Major Infectious Diseases, edited by K. K. Holmes, Rectal Neisseria Meningitidis Isolates Reveals Encapsulated S. Bertozzi, B. R. Bloom, and P. Jha. Washington, DC: World Hyperinvasive Meningococci and Coincident Multidrug- Bank. Resistant Gonococci.” Sexually Transmitted Infections 2017 Crump, J., P. N. Newton, S. J. Baird, and Y. Lubell. 2017. (January): 1–7. doi:10.1136/sextrans-2016-052781. “Febrile Illness in Adolescents and Adults.” In Disease Hecht, R., D. T. Jamison, J. Augenstein, G. Partridge, and Control Priorities (third edition): Volume 6, Major Infectious K. Thorien. 2012. “Vaccine Research and Development.” Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, In Rethink HIV: Smarter Ways to Invest in Ending HIV and P. Jha. Washington, DC: World Bank. in Sub-Saharan Africa, edited by B. Lomborg, 299–320. Dodd, P. J., G. P. Garnett, and T. B. Hallett. 2010. “Examining New York: Cambridge University Press. the Promise of HIV Elimination by ‘Test and Treat’ in Hyper-Endemic Settings.” AIDS 24 (5): 729–35. Hutton, G., and C. Chase. 2017. “Water Supply, Sanitation, Du, L., J. You, K. M. Nicholas, and R. H. Cichewicz. 2016. and Hygiene.” In Disease Control Priorities (third “Chemoreactive Natural Products that Afford Resistance edition): Volume 7, Injury Prevention and Environmental against Disparate Antibiotics and Toxins.” Angewandte Health, edited by C. N. Mock, O. Kobusingye, R. Nugent, Chemie International Edition 18 (55): 4220–25. and K. R. Smith. Washington, DC: World Bank. Fitzpatrick, C., U. Nwankwo, E. Lenk, S. de Vlas, and D. Bundy. Imani, P. 2015. “Effect of Integrated Infectious Disease Training 2017. “An Investment Case for Ending Neglected Tropical and On-Site Support on the Management of Childhood Diseases.” In Disease Control Priorities (third edition): Volume Illnesses in Uganda: A Cluster Randomized Trial.” BMC 6, Major Infectious Diseases, edited by K. K. Holmes, S. Bertozzi, Pediatrics 5 (13). B. R. Bloom, and P. Jha. Washington, DC: World Bank. Jamison, D. T., J. G. Breman, A. R. Measham, G. Alleyne, Fitzpatrick, M. C., and K. Floyd. 2012. “A Systematic Review M. Claeson, D. B. Evans, and P. Musgrove, eds. 2006. Disease of the Cost and Cost-Effectiveness of Treatment for Control Priorities in Developing Countries, second edition. Multidrug-Resistant Tuberculosis.” Pharmacoeconomics Washington, DC: World Bank and Oxford University Press. 30 (1): 63–80. Jamison, D. T., W. H. Mosley, A. R. Measham, and Foster, A. L., M. M. Gaisa, R. M. Hijdra, S. S. Turner, T. J. Morey, J. L. Bobadilla, eds. 1993. Disease Control Priorities in and others. 2017. “Shedding of Hepatitis C Virus into the Developing Countries. Washington, DC: World Bank and Rectum of HIV-Infected Men Who Have Sex with Men.” Oxford University Press. Clinical Infectious Diseases 64 (3): 284–88. Jamison, D. T., R. Nugent, H. Gelband, S. Horton, P. Jha, García, P. J., C. P. Carcamo, G. P. Garnett, P. E. Campos, and R. Laxminarayan, and C. N. Mock, eds. 2015–18. Disease K. K. Holmes. 2012. “Improved STD Syndrome Management Control Priorities. Third edition. 9 volumes. Washington, by a Network of Clinicians and Pharmacy Workers in Peru: DC: World Bank. The PREVEN Network.” PLoS One 7 (10): e47750. Jenness, S. M., S. M. Goodreau, E. Rosenberg, E. N. Beylerian, Gavi, the Vaccine Alliance. 2011. “Market Shaping: Access to K. W. Hoover, and others. 2016. “Impact of the Centers for Low and Sustainable Vaccine Prices.” Document 6a for the Disease Control’s HIV Preexposure Prophylaxis Guidelines Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 25 for Men Who Have Sex with Men in the United States.” The Miller-Petrie, M., S. Pant, and R. Laxminarayan. 2017. “Drug- Journal of Infectious Diseases 214 (12): 1800–07. Resistant Infections.” In Disease Control Priorities (third Johnson, A. P., and N. Woodford. 2013. “Global Spread of edition): Volume 6, Major Infectious Diseases, edited by Antibiotic Resistance: The Example of New Delhi Metallo- K. K. Holmes, S. Bertozzi, B. R. Bloom, and P. Jha. β-Lactamase (NDM)-Mediated Carbapenem Resistance.” Washington, DC: World Bank. Journal of Medical Microbiology 62 (Pt 4): 499–513. Mohapatra, B., D. A. Warrell, W. Suraweera, P. Bhatia, John-Stewart, G., R. W. Peeling, C. Levin, P. J. Garcia, D. Mabey, N. Dhingra, and others. 2011. “Snakebite Mortality in and J. Kinuthia. 2017. “Prevention of Mother-to-Child India: A Nationally Representative Mortality Survey.” PLoS Transmission of HIV and Syphilis.” In Disease Control Neglected Tropical Diseases 5 (4): e1018. Priorities (third edition): Volume 6, Major Infectious Mokdad, A. H., M. H. Forouzanfar, F. Daoud, A. A. Mokdad, C. Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, El Bcheraoui, and others. 2016. “Global Burden of Diseases, and P. Jha. Washington, DC: World Bank. Injury, and Risk Factors for Young People’s Health during Kahn, J. G., L. Bollinger, J. Stover, and E. Marseille. 2017. 1990–2013: A Systematic Analysis for the Global Burden “Improving the Efficiency of the HIV/AIDS Policy Response: of Disease Study 2013.” The Lancet 387 (10036): 2383–401. A Guide to Resource Allocation Modeling.” In Disease Nanduri, S., C. Foo, V. Ngo, C. Jarashow, R. Civen, and others. Control Priorities (third edition): Volume 6, Major Infectious 2016. “Outbreak of Serogroup C Meningococcal Disease Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, Primarily Affecting Men Who Have Sex with Men— and P. Jha. Washington, DC: World Bank. Southern California.” Morbidity and Mortality Weekly Katz, I. T., and A. A. Wright. 2008. “Circumcision—A Surgical Report 65 (35): 939–40. Strategy for HIV Prevention in Africa.” New England Newman, L., J. Rowley, S. Vander Hoorn, N. S. Wijesooriya, Journal of Medicine 359 (23): 2412–15. M. Unemo, and others. 2015. “Global Estimates of the Prevalence and Incidence of Four Curable Sexually Kippax, S., and N. Stephenson. 2012. “Beyond the Distinction Transmitted Infections in 2012 Based on Systematic Review between Biomedical and Social Dimensions of HIV and Global Reporting.” PLoS One 10 (12): e0143304. Prevention through the Lens of a Social Public Health.” Norheim, O., P. Jha, K. Admasu, T. Godal, R. Hum, and others. American Journal of Public Health 102 (5): 789–99. 2015. “Avoiding 40% of the Premature Deaths in Each Klein, E. Y., M. Makowsky, M. Orlando, E. Hatna, N. P. Braykov, Country, 2010–30: Review of National Mortality Trends to and others. 2015. “Influence of Provider and Urgent Help Quantify the UN Sustainable Development Goal for Care Density across Different Socioeconomic Strata on Health.” The Lancet 385 (9964): 239–52. Outpatient Antibiotic Prescribing in the USA.” Journal of Okeke, I. N., K. P. Klugman, Z. A. Bhutta, A. G. Duse, P. Jenkins, Antimicrobial Chemotherapy 70 (5): 1580–87. and others. 2005. “Antimicrobial Resistance in Developing Kratz, M. M, D. Weiss, A. Ridpath, J. R. Zucker, A. Geevarughese, Countries. Part II: Strategies for Containment.” The Lancet and others. 2015. “Community-Based Outbreak of Neisseria Infectious Diseases 5 (9): 568–80. meningitidis Serogroup C Infection in Men Who Have Sex Owusu-Edusei, K., Jr., G. Tao, T. L. Gift, A. Wang, L. Wang, with Men, New York City, New York, USA 2010–2013.” and others. 2014. “Cost-Effectiveness of Integrated Routine Emerging Infectious Disease 21 (8): 1379–86. Offering of Prenatal HIV and Syphilis Screening in China.” Laxminarayan, R., T. Van Boeckel, and A. Teillant. 2015. “The Sexually Transmitted Diseases 41 (2): 103–10. Economic Costs of Withdrawing Antimicrobial Growth Pretorius, C., J. Stover, L. Bollinger, N. Bacaer, and B. Williams. Promoters from the Livestock Sector.” Food, Agriculture, 2010. “Evaluating the Cost-Effectiveness of Pre-Exposure and Fisheries Papers 78, Organisation for Economic Prophylaxis (PrEP) and Its Impact on HIV-1 Transmission Co-operation and Development, Paris. in South Africa.” PLOS ONE 5(11). Liang, X., S. Bi, W. Yang, L. Wang, G. Cui, and others 2009. Sánchez, J., E. Gotuzzo, J. Escamilla, C. Carrillo, L. Moreyra, “Evaluation of the Impact of Hepatitis B Vaccination and others. 1998. “Sexually Transmitted Infections in among Children Born during 1992–2005 in China.” Journal Female Sex Workers: Reduced by Condom Use but Not of Infectious Diseases 200 (1): 39–47. by a Limited Periodic Examination Program.” Sexually Looker, K. J., A. S. Magaret, K. M. E. Turner, P. Vickerman, Transmitted Diseases 25 (2): 82–89. S. L. Gottliev, and others. 2015. “Global Estimates of Schweitzer, A., J. Horn, R. T. Mikolajczyk, G. Krause, and Prevalent and Incident Herpes Simplex Virus Type 2 J. J. Ott. 2015. “Estimations of Worldwide Prevalence of Infections in 2012.” PLoS One 10 (10): e0140765. Chronic Hepatitis B Virus Infection: A Systematic Review Madhav, N., B. Oppenheim, M. Galivan, P. Mulembakani, E. of Data Published Between 1965 and 2013.” The Lancet 386 Rubin, and N. Wolfe. 2018. “Pandemics: Risks, Impacts, and (10003): 1546–55. Mitigation.” In Disease Control Priorities (third edition): Shretta, R., J. Liu, C. Cotter, J. Cohen, C. Dolenz, and others. Volume 9, Disease Control Priorities: Improving Health 2017. “Malaria Elimination and Eradication.” In Disease and Reducing Poverty, edited by D. T. Jamison, R. Nugent, Control Priorities (third edition): Volume 6, Major Infectious H. Gelband, S. Horton, P. Jha, R. Laxminarayan, and Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, C. N. Mock. Washington, DC: World Bank. and P. Jha. Washington, DC: World Bank. 26 Major Infectious Diseases Smith J. S., P. A. Gilbert, A. Melendy, R. K. Rana, and WHO (World Health Organization). 2014. “Global Health J. M. Pimenta. 2011. “Age-Specific Prevalence of Human Estimates 2014 Summary Tables: DALY by Cause, Age, Sex Papillomavirus Infection in Males: A Global Review.” by World Bank Income Category, 2000–2012.” WHO, Geneva. Journal of Adolescent Health 48 (6): 540–52. http://www.who.int/healthinfo/global_burden_disease/en. Stanaway, J. D., A. D. Flaxman, M. Naghavi, C. Fitzmaurice, ———. 2015. Investing to Overcome the Global Impact of T. Vos, and others. 2015. “The Global Burden of Viral Neglected Tropical Diseases: Third WHO Report on Neglected Hepatitis from 1990 to 2013: Findings from the Global Tropical Diseases. Geneva: WHO. Burden of Disease Study 2013.” The Lancet 388 (10049): ———. 2016a. Global Tuberculosis Report 2016. Geneva: WHO. 1081–88. ———. 2016b. “Sexually Transmitted Infections (STIs).” Fact UN (United Nations). 2015a. Resolution Adopted by the sheet, WHO, Geneva. General Assembly on 25 September 2015. Transforming ———. 2017. “Ghana, Kenya and Malawi to Take Part in Our World: The 2030 Agenda for Sustainable Development. WHO Malaria Vaccine Pilot Programme.” News Release, Resolution A/RES/70/1, United Nations General Assembly, WHO, Geneva, April 24. New York. Wilson, D. P., B. Donald, A. J. Shattock, D. Wilson, and ———. 2015b. The Millennium Development Goals Report N. Fraser-Hurt. 2015. “The Cost-Effectiveness of 2015. New York: UN. Harm Reduction.” International Journal of Drug Policy UNAIDS (Joint United Nations Programme on HIV/AIDS). 26 (Suppl 1): S5–11. 2015. AIDS by the Numbers 2015. Booklet, UNAIDS, Wilson, D. P., and J. Taaffe. 2017. “Tailoring the Local HIV/AIDS Geneva. Response to Local HIV/AIDS Epidemic.” In Disease Control ———. 2016. “Global HIV Statistics.” Fact sheet, November, Priorities (third edition): Volume 6, Major Infectious UNAIDS, Geneva. Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, U.S. Department of Health and Human Services. 2014. and P. Jha. Washington, DC: World Bank. Guide for HIV/AIDS Clinical Care. Clinical reference Winer, R. L., J. P. Hughes, Q. Feng, L. F. Xi, S. K. Lee, and guide, HIV/AIDS Bureau, Health Resources and Services others. 2012. “Prevalence and Risk Factors for Oncogenic Administration, U.S. Department of Health and Human Human Papillomavirus Infections in High-Risk Mid- Services, Rockville, MD. Adult Women.” Sexually Transmitted Diseases 39 (11): Van Boeckel, T. P., S. Gandra, A. Ashok, Q. Caudron, B. T. Grenfall, 848–56. and others. 2014. “Global Antibiotic Consumption 2000 to World Bank. 1993. World Development Report 1993: Investing 2010: An Analysis of National Pharmaceutical Sales Data.” in Health. Washington, DC: World Bank and Oxford The Lancet Infectious Diseases 114 (8): 742–50. University Press. Verguet, S., R. Laxminarayan, and D. T. Jamison. 2015. ———. 2012. People, Pathogens and Our Planet: Volume 2, “Universal Public Finance of Tuberculosis Treatment in The Economics of One Health. Economic and Sector Work, India: An Extended Cost-Effectiveness Analysis.” Health Report 69145-GLB. Washington, DC: World Bank. Economics 24 (3): 318–32. World Health Assembly. 2005. International Health Regulations Weaver, M.R., S.M. Burnett, I. Crozier, S.N. Kinoti, I. Kirunda, (2005). Geneva: World Health Organization. and others. 2014. “Improving Facility Performance in Zhang, Y., B. Y. Lee, and J. M. Donohue. 2010. “Ambulatory Infectious Disease Care in Uganda: A Mixed Design Study Antibiotic Use and Prescription Drug Coverage in with Pre/Post and Cluster Randomized Trial Components.” Older Adults.” Archives of Internal Medicine 170 (15): PLoS One. 9(8). 1308–14. Major Infectious Diseases: Key Messages from Disease Control Priorities, Third Edition 27 Chapter 2 Global Mortality and Morbidity of HIV/AIDS Kristen Danforth, Reuben Granich, Danielle Wiedeman, Sanjiv Baxi, and Nancy Padian INTRODUCTION 2015; INSIGHT START Study Group 2015; Kitahata The HIV/AIDS epidemic has seen dramatic shifts since and others 2009; Lopez-Cortes, Gutierrez-Valencia, and the first cases were described in 1981. Initially perceived Ben-Marzouk-Hidalgo 2016; Lundgren, Babiker, and as a disease among gay men or Haitians in Western Neaton 2016; Médecins Sans Frontières 2013; Montaner countries, HIV transmission has been reported in virtu- and others 2006; SMART Study Group and others 2006; ally all parts of the world. Prevalence levels in the 1990s Sterne and others 2009; Violari and others 2008). In reached more than 30 percent among adults in many high-income countries (HICs), access to early treatment Sub-Saharan African cities, and no accessible, effective has led to near-normal life expectancy for persons living treatment was available. Although treatment was avail- with HIV/AIDS (Johnson and others 2013; May and able for a limited number of people in wealthier settings others 2014; Rodger and others 2013; Samji and others shortly after the studies on triple therapy in 1996, mor- 2013). As a result, the focus of clinical care of HIV/AIDS tality nevertheless soared, particularly in Sub-Saharan in these settings has shifted from treatment of a usually Africa, slashing the hard-won gains in life expectancy fatal infectious disease with multiple comorbidities (see resulting from social and economic development and chapter 4 [Harripersaud and others 2017] and chapter 11 advances in medical technology and nutrition (United [Bloom and others 2017] of this volume) to manage- Nations Population Division 2004) by more than a ment of a chronic condition and prevention of illness, decade within a few years. death, and transmission for individuals who remain Since 2000, remarkable progress has been made in the adherent to treatment (Attia and others 2009; Cohen and diagnosis and treatment of persons living with HIV/AIDS. others 2011; Das and others 2010; Fang and others 2004; With medications now affordable at a cost of approxi- Montaner and others 2010). mately US$129–$568 per person per year even in the hardest hit countries (Bendavid and others 2010; Menzies, Berruti, and Blandford 2012; PEPFAR 2013; UNAIDS New Focus on Treatment and Care 2015a; Walensky and others 2013), 17 million people Treatment access and care recommendations have dra- were receiving antiretroviral therapy (ART) in 2015. matically changed all over the world since the second The international targets are to treat nearly three-quarters edition of Disease Control Priorities in Developing of those living with HIV/AIDS by 2030 (UNAIDS 2012, Countries (Jamison and others 2006) (Bertozzi and oth- 2014b, 2015b, 2016b; WHO 2013c). ers 2006; WHO 2010, 2013a, 2015b). Accumulating evi- ART has reduced HIV/AIDS morbidity and mortality dence definitively demonstrates that treatment reduces significantly (Cohen and others 2011; Danel and others morbidity and mortality, irrespective of disease stage or Corresponding author: Kristen Danforth, School of Public Health, University of Washington, Seattle, Washington, United States; danfortk@uw.edu. 29 immunological competence, for example, CD4 level recent recommendations by the U.S. Centers for Disease (Danel and others 2015; INSIGHT START Study Group Control (CDC) (AIDS.gov 2017). In addition, these tar- 2015; Kitahata and others 2009). Treatment simultane- gets are subject to meaurement challenges (many of which ously prevents onward transmission (chapter 5 in this are described subsequently); however, they are critical to volume, Holmes and others 2017; Attia and others address success in epidemic control. Although measure- 2009). Accordingly, the dream of ending the epidemic as ment of disease burden is informative, it falls short of a public health threat by 2030 (UNAIDS 2015d) no lon- meeting needs in public health for which the influences of ger seems impossible. positive health—such as successes in care and treatment— must also be tracked (Thacker and others 2006). In addition to monitoring regional burdens, more pre- cise surveillance tools, including geospatial mapping and Expanded Surveillance targeted surveillance, have uncovered microepidemics The drive to end the epidemic has resulted in the expan- concentrated in small regions and in key, vulnerable pop- sion of surveillance. In addition to tracking the burden ulations, which heretofore might have been missed. The of incidence, prevalence, and mortality, programs now availability of such detailed surveillance data at subnational track success in meeting the 90-90-90 targets as part of and smaller local levels has revealed the microepidemics their efforts to monitor and evaluate the continuum of defined by locality or risk group that fuel generalized epi- care (IAPAC 2016; UNAIDS 2014e, 2015c, 2016b). demics in Sub-Saharan Africa (Tanser and others 2014). The new focus on care in treatment is reflected in the Although we address this phenomenon in this chapter, recent changes in the World Health Organization’s (WHO) more extensive detail about the concentration of infection guidelines (2015b) and demonstrated by the 90-90-90 among key populations and risk groups is provided in campaign of the Joint United Nations Programme on chapter 8 of this volume (Wilson and Taaffe 2017). HIV/AIDS (UNAIDS), which recommend treating infected individuals as soon as possible. These 90-90-90 targets propose achievement of the following by 2020: Chapter Content This chapter is divided into two sections. The first • Ninety percent of all people living with HIV/AIDS describes the distribution of surveillance indicators for will know their HIV/AIDS status effective monitoring of national HIV/AIDS programs. • Ninety percent of people with diagnosed HIV/AIDS These indicators include the conventional, key outcomes infection (or 81 percent of all people living with HIV) of mortality and morbidity—incidence, prevalence, and will receive sustained ART disability-adjusted life years (DALYs)—as well as more • Ninety percent of all people receiving ART will be recent indicators that reflect the pivot to ending the virally suppressed—that is, will achieve 73 percent epidemic: tracking 90-90-90 targets and examining sur- population-based suppression in people living with veillance in smaller units of analysis, including microep- HIV (UNAIDS 2014a). Achieving 90-90-90 is the idemics and key populations. The second part of the first step to ensuring access to treatment for nearly chapter addresses challenges in the measurement of all everyone by 2030, which could lead to ending AIDS of these indicators. as a public threat, as well as the virtual elimination of Subsequent chapters in this volume address current HIV transmission in many settings (Granich 2016). cost-effective approaches for treatment (chapter 5, Holmes and others 2017), prevention of mother-to- The UNAIDS Fast-Track Targets for 2020 (Stover and child transmission (chapter 6, John-Stewart and others others 2016) include reducing by 75 percent the number 2017), and combination prevention (chapter 7, Garnett of people newly infected annually (compared with and others 2017). The burden, prevention, and manage- 2010), with zero new infections among children, and ment of HIV/AIDS-related comorbidities, including reducing the annual number of people dying from other sexually transmitted infections and tuberculosis, HIV/AIDS-related causes to fewer than 500,000. These are discussed in chapter 10 (Chesson and others 2017) targets are the next steps to the even more challenging and chapter 11 (Bloom and others 2017) of this volume. yet achievable 95-95-95 goals for 2030, when annual The goal of this chapter is not to provide a complete deaths related to HIV/AIDS should be fewer than review of available data on burden, but rather to situate 200,000 and incidence should be reduced by 90 percent this volume—volume 6, Major Infectious Diseases—of (compared with 2010). the third edition of Disease Control Priorities (DCP3) in Some may argue that this is an aspirational slogan, but the context of an HIV/AIDS epidemic and response that assessing accomplishments in care is consistent with is at a turning point. 30 Major Infectious Diseases DISTRIBUTION OF KEY EPIDEMIC prevention. Deaths peaked at just more than 2 million MEASURES: MORTALITY, INCIDENCE, per year from 2004 to 2005 (figure 2.1) and have been steadily declining since, driven primarily by gains in PREVALENCE, AND DALYS Sub-Saharan Africa. The 1.1 million individuals who Given the outsized importance of HIV/AIDS among lost their lives to AIDS in 2015 represent the lowest donors and global health organizations, a variety of number since 1998, and this number was 45 percent sources for burden of disease estimates at global and lower than at the peak of the epidemic (UNAIDS 2016a, national levels exist, such as UNAIDS, the WHO, and the 2016b). Despite this progress, high-burden countries Institute for Health Metrics and Evaluation (UNAIDS will need to accelerate access to ART treatment to avert 2016a; Wang and others 2016; WHO 2013b). To maintain millions of premature AIDS deaths and new HIV infec- consistency with the other volumes in the DCP3 series tions (Granich and others 2015). and the chapters that follow in this volume, we focus Despite these gains, AIDS remains a significant cause primarily on data from two sources, UNAIDS and the of death, and global trends mask persistent regional WHO’s Global Health Estimates, supplemented with and subregional variation. AIDS is the sixth-leading country-specific studies, where relevant, for illustrative cause of death globally and the leading cause in Sub- purposes. No estimate is without limitations, and Saharan Africa, a fact that has not changed since 2000, the UNAIDS and WHO figures are no exception. These despite the 41 percent decline in the region’s AIDS- limitations are discussed in a later section, titled related mortality rate. AIDS was responsible for one in “Measurement: Challenges in Surveillance.” nine deaths in the WHO’s African region in 2012 (WHO 2013b). In contrast to declining rates in Sub- Saharan Africa, AIDS-related mortality rates per Global and Regional Trends in Mortality 100,000 population from 2000 to 2012 increased from Global trends in AIDS-related mortality reveal the 3.6 to 10.2 in Europe, from 2.7 to 5.6 in the Eastern remarkable success of HIV treatment and other Mediterranean Region, and from 1.6 to 3.2 in the Figure 2.1 Number of Deaths Related to AIDS, 1990–2015 2,000 1,500 Number of deaths (thousands) 1,000 500 0 1990 1995 2000 2005 2010 2015 Year Global East and Southern Africa West and Central Africa Western and Central Europe and North America Latin America and the Caribbean Asia and the Pacific Eastern Europe and Central Asia Middle East and North Africa Source: Based on UNAIDS 2016a. Note: AIDS = acquired immune deficiency syndrome. Global Mortality and Morbidity of HIV/AIDS 31 Western Pacific Region. The numbers of AIDS-related 2001, three years earlier than AIDS-related mortality deaths in Europe and the Eastern Mediterranean Region peaked, and they have been slowly declining or have pla- have also been steadily increasing and have tripled since teaued over the past decade in most regions (figure 2.2). 2000 (WHO 2013b). The trends in prevalence reveal that HIV/AIDS remains a predominantly East and Southern Africa health challenge, with a 2014 prevalence of 7.4 percent, more than triple Morbidity: Incidence and Prevalence that of the western and central part of the continent. Trends in HIV/AIDS prevalence are similar to those of However, low prevalence rates in low disease burden set- mortality, although one must note that by definition, tings still present major challenges, even, for example, in increased treatment results in increased prevalence. the United States (Del Rio 2015). More to the point, In the current era of massive global antiretroviral scale although prevalence rates are currently low in Eastern up, with plans for reaching millions more (UNAIDS Europe and Central Asia, they are the only regions where 2012, 2014b; WHO 2013c), there are critical drivers of prevalence rates are still rising, with rates increasing from changing morbidity in persons living with the disease. 0.1 percent in 1990 to 0.8 percent in 2014. Most important, with earlier initiation of ART and A central focus of the new UNAIDS goals is a improved access to care, persons with HIV are living 75 percent reduction in new infections (compared with longer (Kitahata and others 2009; Sterne and others 2010) by 2030. Models of incidence suffer from signifi- 2009; SMART Study Group and others 2006; U.S. cant methodological limitations, which are discussed National Institutes of Health 2015; Violari and others later in this chapter in the section on measurement chal- 2008) and therefore are experiencing the health conse- lenges. Incident infections have fallen substantially since quences of aging. This is an outcome of improved treat- their peak of an estimated 3.5 million per year in 1997 to ment options and increased access to interventions that 2.1 million in 2015. However, current projections from lead to longer lives. Global prevalence rates peaked in UNAIDS and the Institute of Health Metrics and Figure 2.2 Trends in HIV Prevalence, by Region, 1990–2015 10 HIV prevalence (percent) 1 0.1 1990 1995 2000 2005 2010 2015 Year East and Southern Africa West and Central Africa Global Latin America and the Caribbean Western and Central Europe and North America Asia and Pacific Eastern Europe and Central Asia Middle East and North Africa Source: Based on UNAIDS 2016a. Note: HIV = human immunodeficiency virus. 32 Major Infectious Diseases Evaluation indicate that, in general, the world is not yet nevertheless representing a 9.6 percent decrease from on track to meet the UNAIDS goal (UNAIDS 2016a; 2000. In 2012, by region, Africa accounted for 66.8 mil- Wang and others 2016); new infections declined lion DALYs lost (72.7 percent); South-East Asia, 11.8 14 percent globally between 2010 and 2014 (UNAIDS million (12.8 percent); Europe, 4.5 million (4.9 percent); 2016a). Although HIV prevalence remains heavily con- Eastern Mediterranean, 2.0 million (2.2 percent); centrated in Africa, other regions have also emerged as Western Pacific, 3.3 million (3.6 percent); and the important sources of new infections (figure 2.3). Between Americas, 3.5 million (3.8 percent). HIV/AIDS is still the 1990 and 2015, the proportion of incident infections in leading cause of morbidity in Africa, but DALYs lost East and Southern Africa steadily declined from declined 19.2 percent since 2000 (WHO 2013b). 59 percent to 46 percent, and those in Eastern Europe The Global Burden of Disease data present lower abso- and Central Asia and Asia and the Pacific steadily lute values for DALYs (69.4 million in 2013), but show a increased from 1 percent to 9 percent and from 10 percent similar decline from 2000 to 2013 (IHME 2016b). to 15 percent, respectively. While these relative relation- As with mortality and incidence, these regional num- ships are informative, the assumptions underlying inci- bers mask significant subregional variation. In the African dence models, and thus current and future projections, region, 31 percent of all DALYS lost in Botswana were are likely to be revised as new empirical data improve due to HIV/AIDS (324,000), compared to 6.9 percent of our knowledge about the effect of ART scale-up on HIV- DALYs in Ethiopia (3,353,000) and 4.3 percent of DALYs transmission rates in these regions. in Eritrea (126,000) (WHO 2013b). The trends in global DALYs show both the promise of ART and the gap that remains in getting effective treat- Morbidity: Disability-Adjusted Life Years ment to all who need it and ensuring adherence among In 2012, 91.9 million DALYs were lost worldwide because those receiving it. The decline in DALYs is driven largely of HIV/AIDS, second only to diarrhea in terms of mor- by reductions in AIDS deaths, and thus the pattern of bidity from infectious disease and seventh overall, but DALYs lost because of HIV/AIDS parallels that for Figure 2.3 Proportion of Incident HIV Infections, by Region, 1990 and 2015 thousands a. 1990 b. 2015 Western and Central Europe and Western and Central Europe and North America North America 100 91 Asia and the Pacific West and Asia and West and 200 Central Africa the Pacific Central Africa 410 300 400 Middle East and North Africa Middle East and 3.3 North Africa East and 21 East and Latin America Southern Africa Southern Africa and the 1,200 Latin America 960 Caribbean and the 130 Caribbean 100 Eastern Europe and Eastern Europe Central Asia and 13 Central Asia 190 Source: Based on UNAIDS 2016a. Note: HIV = human immunodeficiency virus. Global Mortality and Morbidity of HIV/AIDS 33 mortality (Wang and others 2016). Although deaths As such, regional cascade results should be viewed with decreased 18 percent between 2000 and 2012, morbidity considerable caution. Nevertheless, even these suboptimal declined by only 9.6 percent (UNAIDS 2016a; WHO estimates reveal critical trends. Perhaps more important, 2013b). One contributing factor is that a small but especially with regard to comparative estimates across important portion of morbidity is due to years lived countries or regions or over time, the data enforce the with disability, which has plateaued since 2005, driven in value of monitoring the health outcomes that are essential part by the estimated 46 percent of HIV-positive indi- for epidemic control. viduals globally who are not currently on ART (IHME Preliminary estimates from Levi and others (2016), 2016a; UNAIDS 2016b; Wang and others 2016). Among based on data from 69 countries for which data were those who are, fewer than half remain virally suppressed available, show that progress is uneven. three years after initiating treatment (UNAIDS 2015b; Progress in achieving the first 90 percent in pre- WHO 2015a). A better understanding of the structure dominantly high-income regions is encouraging; in six and composition of epidemics, including global resource out of nine countries across North America, Australasia, allocation for AIDS (Granich and others 2016) within and Western Europe, 80 percent or more of individuals countries, is an essential next step as the world shifts know their HIV status. In the majority of these coun- from trying to manage the epidemic to trying to end it. tries, the most significant gap is the proportion of HIV-positive individuals who currently receive ART. Conversely, in lower-income regions, the need for scaled-up testing services coupled with demand cre- REACHING THE 90-90-90 TARGETS ation to meet the first 90 percent target is great. For Tracking progress toward the achievement of the 90-90-90 example, data from Levi and others (2015) for coun- targets is a central challenge for surveillance efforts, partic- tries where full treatment cascades were available in ularly in measurement, given the lack of individual cohort the African and Asian regions suggest that fewer than data from most regions. Moreover, the methods used to half of all HIV-positive individuals are aware of their determine the national cascades included in estimating status (figure 2.4). In Sub-Saharan Africa, the majority regional cascades often vary and often do not follow the of those who know they are HIV-positive are success- WHO recommendations. The lack of viral load data nec- fully initiated on ART, and roughly 75 percent achieve essary for estimating the final component of the cascade viral suppression. For the Kyrgyz Republic, the Russian is particularly problematic because the data are not avail- Federation, and Vietnam, fewer than 35 percent of able to most people living with HIV (UNAIDS 2014e). those who know they are HIV-positive are successfully Figure 2.4 Cascade of HIV Care in Africa and Asia a. Asia b. Sub-Saharan Africa (Kyrgyz Republic, Russian Federation, and Vietnam) 100 100 1,628,000 23,500,000 83 80 80 75 10,000,000 7,500,000 60 55 60 Percent Percent 51 41 127,000 12,000,000 40 34 40 673,000 231,000 20 20 0 0 HIV+ people Diagnosed On ART Viral load HIV+ people Diagnosed On ART Viral load suppression suppression HIV care cascade HIV care cascade Sources: Adapted from data from Levi and others 2015; UNAIDS 2015e. Note: ART = antiretroviral therapy; HIV = human immunodeficiency virus. The numbers above each bar represent the number of individuals at that stage as a proportion of the number of individuals at the preceding stage. Sub-Saharan Africa viral load data from Botswana, Burkina Faso, Cameroon, Côte d’Ivoire, Kenya, Malawi, Mali, Mozambique, Nigeria, Senegal, South Africa, Tanzania, Uganda, Zambia, and Zimbabwe. 34 Major Infectious Diseases initiated on ART, and among these, only about half are Similarly, although the epidemic is disproportion- virally suppressed. Data such as these, which are essen- ately concentrated in countries of East and Southern tial to inform program planners where to effectively Africa, pockets of high rates of transmission within invest resources, illustrate the uneven progress of countries are driving the spread of the disease. Data regions toward achievement of the 90-90-90 targets. from 1,724 sites supported by the President’s Emergency These data also reinforce how success cannot be real- Plan for AIDS Relief (PEPFAR) and community-based ized without consideration of the complete cascade. services in Zimbabwe show that from October 2013 to Although rates of viral suppression among those on September 2014, approximately 80 percent of all newly ART in the 15 Sub-Saharan African countries with diagnosed people living with HIV/AIDS were identified data available are high, the very low rates of diagnosis by only 30 percent of sites (figure 2.5). mean that, overall, fewer than 32 percent of HIV- One of the earliest studies to investigate subnational positive individuals in those countries are virally analyses examined geospatial data on HIV/AIDS in suppressed. KwaZulu-Natal, South Africa (MEASURE Evaluation 2016; Tanser and others 2009). Within a relatively homogeneous population, where age-adjusted preva- REGIONAL MICROEPIDEMICS AND KEY lence was 27 percent for women and 14 percent for POPULATIONS men, local prevalence had notable spatial variation, with three very-high-prevalence clusters (approxi- Regional Microepidemics mately 36 percent) along the main national road and Global and regional estimates conceal significant three relatively low-prevalence clusters (6 percent) country-level heterogeneity and within-country (Tanser and others 2009). In another study, Magadi variability that is characterized by microepidemics— (2013) examined spatial distribution of HIV/AIDS pockets of illness concentrated among specific popula- infection in relation to various demographic factors. tions and smaller subnational geographic regions. In She found that the urban poor in Sub-Saharan Africa Brazil, most AIDS cases and HIV infections occur in had significantly higher rates of infection than did their fewer than 10 percent of the country’s 5,570 municipal- urban nonpoor counterparts and that the well- ities (UNAIDS 2014c). In India, national prevalence documented higher risk among women was amplified was only 0.4 percent in 2011, but 71 of 672 districts among the urban poor. had a prevalence of ≥ 1.0 percent. Three-quarters of Anderson and others (2014) modeled and demon- these 71 districts are located in the southern and north- strated the importance of targeting prevention services, eastern parts of the country (National AIDS Control including early treatment, to microepidemics. For exam- Organisation 2012). ple, in Kenya, just 9 of 47 counties represented an Figure 2.5 HIV Diagnoses Yield from PEPFAR-Supported Testing Sites, Zimbabwe, 2013–14 6,000 120 Cumulative % of new HIV cases identified Number of people living with HIV 5,000 100 4,000 80 3,000 60 2,000 40 1,000 20 0 0 0 10 20 30 40 50 60 70 80 90 100 Cumulative % of total sites HTC HIV + yield Cumulative percentage of HIV-positive people Source: UNAIDS 2015d. Note: HIV = human immunodeficiency virus; HTC = HIV testing and counseling; PEPFAR = President’s Emergency Plan for AIDS Relief. Global Mortality and Morbidity of HIV/AIDS 35 Map 2.1 Estimated New Infections in Kenya, 2014 mobile migrant populations in Sub-Saharan Africa. IBRD 42545 | OCTOBER 2016 Given that issues related to stigma, discrimination, and punitive legislations make tracking these groups extremely challenging, some patterns are clear. High rates of transmission among people who inject drugs and among sex workers are the main drivers of new infections in the Middle East and North Africa, and MSM is the main contributor in Latin America and the Caribbean. In Pakistan, transmission to female spouses of HIV-positive injection drug users and bisexual men, and subsequently to children through mother-to-child transmission, is a critical source of new infections in these regions. Patterns vary substantially across coun- tries in Eastern Europe and Central Asia with respect to the relative contribution of such key populations NAIROBI (Gouws and Cuchi, on behalf of the International Estimated new HIV Collaboration on Estimating HIV Incidence by Modes infections, 2014 >250 of Transmission 2012), which is limited by the severe 250–499 lack of data and signficant issues in surveillance dis- 500–999 cussed subsequently. Similarly, the pattern within Africa 1,000–4,999 is mixed. In Kenya, HIV/AIDS overall prevalence among >5000 key populations is extremely high at 5.3 percent: Counties accounting for the majority of 29 percent among sex workers, 18 percent among MSM, new infections and 18 percent among people who inject drugs (UNAIDS 2015d, 2016a). Source: UNAIDS 2015d. Other key populations include prisoners and individ- uals in the military, for whom regional statistics are estimated 65 percent of all new infections, and HIV sorely lacking. The risk of infection among transgender prevalence varied substantially across counties, from less people has recently emerged as a public health emer- than 1 percent to 22 percent in 2014 (map 2.1) (UNAIDS gency. Although the global picture of HIV/AIDS among 2015d). Similar uneven regional variation can also be transgender people is varied—with HIV prevalence seen in the United States (map 2.2) and in many other ranging from 8 percent to 68 percent—transgender peo- countries. Using these data to model the rollout of pre- ple are among the groups most affected by HIV/AIDS, vention programs, Anderson and others (2014) demon- particularly in the Latin America and the Caribbean and strated that a focused approach using local epidemiologic Asia and the Pacific regions (WHO 2011). data to direct prevention programs would achieve greater These studies highlight the assumption that effect than would a uniform approach for the same significant proportions of new HIV infections, even amount of investment. Such focused surveillance is nec- in Sub-Saharan Africa, may occur among key popula- essary for a more targeted response to the epidemic and tions (Hirnschall 2015). This means that no way is the hallmark of PEPFAR 3.0 and the pivot to a data- exists to fully end the epidemic without addressing driven approach that strategically targets microepidem- infections in these key populations, even though ics, hot spots, and key populations. stigma, social norms, and legal restrictions present formidable challenges to identifying and engaging them in programs for prevention or care (Wilson and Key Populations Taaffe 2017). Defining Key Populations A deeper dive into the statistics in the United States The emergence of smaller, subnational regions as sub- reveals that minority race, especially African Americans, stantial contributors to incident infections is due in part constitutes another key population. African Americans to the presence of key populations, traditionally defined are overrepresented among people living with HIV in as people who inject drugs, commercial sex workers, every region in the United States. In contrast, Asians and men who have sex with men (MSM), (addressed and Caucasians constitute the proportion of the popula- more fully in chapter 8 of this volume [Wilson and tion with the lowest infection rates. Notably, adjusting Taaffe 2017]), and more recently, young women and for poverty reduces the magnitude of these differences, 36 Major Infectious Diseases Map 2.2 Rates of Persons Ages 25–34 Years Living with HIV/AIDS Diagnosis, by County, United States, 2013 IBRD 42546 | OCTOBER 2016 0 50 100 125 175 200 275 350 475 750 Data not shown Data not released to AIDSVu Source: AIDSVu 2016. Note: AIDS = acquired immune deficiency syndrome; HIV = human immunodeficiency virus. but it did not change the trends based on race and eth- populations. In one study in the United States, 73 percent nicity (Del Rio 2015). These same trends are mirrored in of transgender women who tested HIV-positive were AIDS cases and in access to HIV-specific health services previously unaware of their status (U.S. CDC 2011), a (Del Rio 2015) and show that many of the challenges figure that can be far greater in other parts of the world. and surveillance needs in LMICs discussed in this chap- Similarly, in the Asia and Pacific region, fewer than half of ter are also relevant to HICs, and substantial opportuni- the key populations know their HIV status (figure 2.6). ties exist for cross-learning. Reaching 90-90-90 Targets among Key Populations MEASUREMENT: CHALLENGES IN Not surprisingly, the success in meeting 90-90-90 targets SURVEILLANCE also varies by key populations; factors such as stigma and discrimination present even greater obstacles to linking Prevalence and Incidence these populations to care. One study of people living with Any summary report of morbidity and mortality is only HIV/AIDS in the United States found that only 59 percent as good as available data and the methodologies used for of transgender participants, compared to 82 percent of collection. The accuracy of estimates of prevalence, inci- participants with a birth-assigned gender, were accessing dence, and disease-specific mortality are challenging ART (Melendez and others 2006). HIV/AIDS-related for any health outcome, especially in low-income, stigma also creates a barrier to getting tested for many key high-mortality countries where vital registration and Global Mortality and Morbidity of HIV/AIDS 37 Figure 2.6 HIV Testing Coverage among Key Populations, Asia and Moreover, regardless of the accuracy of surveillance Pacific Region, 2007–12 tools, population-based prevalence estimates invariably suffer from survey methodological limitations, such as 100 representativeness and potential nonresponse bias HIV testing coverage (%) 80 (Flaxman, Vos, and Murray 2015). These challenges can be particularly salient for HIV/AIDS because of social 60 reasons, such as the persistent stigma and discrimination 42 41 39 that come with a diagnosis (Tanser and others 2014) or 40 with identification as a person in a key population that is 28 at exceptionally high-risk (see chapter 8 in this volume 20 [Wilson and Taaffe 2017]) and with the methodological 0 challenges of assessing an asymptomatic infection for Female sex Male sex Men who have People who which representative sampling is imperative. workers workers sex with men inject drugs Most surveillance is primarily designed to assess mor- Key population tality and prevalence. However, epidemics change; Source: UNAIDS 2014a. although incidence is associated with prevalence, other Note: HIV = human immunodeficiency virus. factors, such as migration, mortality, survival rates, and the inherent epidemic trajectory, attenuate the legitimacy of using prevalence as a measure of incidence (Brookmeyer cause-of-death data may be lacking or incomplete 2010). Incidence rates modeled from prevalence are also (Lopman and others 2006; Mathers and others 2005; constrained by the time between surveys, as well as in-and- Murray and Lopez 2013). As a result, almost all global or out population mobility and migration that is especially large regional estimates of morbidity and mortality critical for men in Sub-Saharan Africa who often migrate require some modeling, the results of which can vary for work (Brookmeyer 2010; Busch and others 2010; depending on assumptions and sources of data (Flaxman, Hallett and others 2008; Marston, Harriss, and Slaymaker Vos, and Murray 2015; UNAIDS 2014d). For example, 2008). Although incidence rates obtained from prospec- for generalized epidemics, UNAIDS estimates are based tive data are most robust, these rates are subject to the on routine surveillance of antenatal clinics, augmented potential for nonrandom loss to follow-up where sicker with results from population-based household surveys, individuals are more likely to be lost. Newer surveillance where available (UNAIDS 2013). These estimates can methods using a geospatial approach to monitoring, com- lead to oversampling of urban populations and provide bined with more reliable incidence assays, can provide no direct information on HIV/AIDS prevalence rates in more robust estimates to identify weaker performance men. These data are supplemented by models of trans- sites or regions where enhanced efforts are warranted. mission from Spectrum (UNAIDS 2016c) and models of Data from PEPFAR-supported, ongoing, population-based progression from the International Epidemiologic HIV/AIDS impact assessments—a multicountry initiative Databases to Evaluate AIDS. These data sources tend to to measure the reach and impact of programs in PEPFAR- underrepresent high-risk groups within the general pop- supported countries through population-based surveys to ulation, and UNAIDS estimates for concentrated epi- estimate incidence, prevalence, and viral load suppression demics rely on extrapolating from individual studies of among adults and children in Malawi, Uganda, Zambia, key populations (Mahy 2016; UNAIDS 2013). and Zimbabwe—should be well-suited to this purpose A salient example of the limitations and fragility of (ICAP 2016). these surveillance methods is the revised estimate by UNAIDS of infection and mortality among adolescents presented at the 2016 meeting of the International AIDS Loss to Follow-Up Society (Mahy 2016). On the basis of more robust Loss to follow-up also affects facility-based HIV/AIDS empirical data on the relative demographic growth of surveillance estimates. Even at care and treatment clinics adolescents, the effect of prevention of mother-to-child that have individual-level patient data, high rates of loss to transmission programs, and ART coverage among ado- follow-up undermine estimates of facility-based survival lescents, estimates of adolescent HIV/AIDS-related mor- and mortality (Egger and others 2011; Geng, Bangsberg, tality were revised. Significantly, HIV/AIDS-specific and others 2010; Geng, Glidden, and others 2010; Geng, causes of death are now ranked eighth in the leading Nash, and others 2010; Geng and others 2008; Geng causes of death in this age group, down from second and others 2012; Geng and others 2013). Attempts to val- before these revisions. idate HIV/AIDS-related mortality have been attempted 38 Major Infectious Diseases through verbal autopsies (Lopman and others 2006) and devote resources, and design programs that clearly actual autopsies (Coulibaly and others 1994; Domoua and match local epidemic realities and to determine whether others 1995; Greenberg and others 1995; Lucas, Diomande, HIV/AIDS services are appropriately matched, suffi- and others 1994; Lucas, Hounnou, and others 1994). cient, and best packaged. However, verbal autopsies lack external validity and must To facilitate these kinds of analyses, the UNAIDS be tailored to specific context, and actual autopsies pres- Reference Group on Estimates, Modelling and ent a major challenge to scale-up. Unfortunately, the costs Projections is reconsidering assumptions underlying the and efforts required to intensively track all missing Spectrum model that drives surveillance estimates. The patients is likely to be prohibitive in most settings. group is working toward standardizing the data collec- Geng and others (2015) offer a viable alternative: inten- tion that underlies spatial analysis, as well as the optimal sively trace a manageable random sample of the individu- methods and the frequency of use of such methods, als lost to follow-up and incorporate their weighted recognizing the need for user-friendly programs to make outcomes into the available clinic sample. The power of these estimates possible (UNAIDS Reference Group on this approach is evident in a review they conducted of Estimates, Modelling and Projections 2013). Other mod- mortality from 14 clinics in Eastern Africa. Sample- eling efforts will also be helpful to provide a critique of corrected estimates of three-year mortality in each clinic the standard approaches as we learn more about the ranged from 2 times to more than 10 times higher com- effect on incidence and prevalence of key interventions pared to the naïve (that is, unadjusted) estimates (Geng, such as treatment, circumcision, and other prevention Odeny, and Lyamuya 2016; Geng and others 2015). Similar methods. Supplemental phylogenetic studies assess the results are apparent from additional analyses in the same contribution of high-risk groups and provide critical clinics also by Geng, Odeny, and Lyamuya (2016) examin- knowledge about transmission dynamics and the accu- ing retention in care. Using only routine clinic data with- racy of targeting those at risk. However, the cost and out supplementation by tracing, after two years of ART, technical capacity for such studies currently prohibit they found 26 percent of patients were reported as lost to their widespread use. A related ongoing challenge is the follow-up; sampled corrected estimates revealed that best method for empirical estimation of population sizes 14 percent of the clinic population who were presumed of hidden high-risk groups (Tanser and others 2014). In lost had actually transferred their care elsewhere. Although settings nearing HIV elimination, future efforts will such data can be used to correct facility or regional esti- likely rely on the most current phylogentic studies to mates, they require individual data collection. identify new cases in clusters combined with tradi- Cohort approaches allow for accountability for each tional public health outbreak control methods to end patient started on treatment, and with the push toward ongoing transmission. 90-90-90, programs that use unique identifiers for those diagnosed with HIV coupled with a national cohort will be able to better account for retention—including trans- MEASUREMENT: CHALLENGES IN fers from clinic to clinic and other outcomes, regardless ASSESSING THE 90-90-90 TARGETS of geographic location. As in other transmissible infec- tious diseases programs that are responsible for provid- The tremendous benefits in early treatment and viral ing access to successful treatment, this approach will suppression, both for improving individual health and ensure access to life-saving treatment, prevent transmis- stemming transmission, have led to the 90-90-90 fram- sion, and allow follow-up for everyone on ART. ing of the HIV/AIDS response, led by UNAIDS, PEPFAR, and other organizations. Surveillance efforts now aim to track the success in meeting these targets; methodologies Surveillance at Smaller Units of Analysis: for doing so have been standardized to the extent possi- Microepidemics and Key Populations ble to permit regional and national comparisons using Regional and subnational variation discussed through- numerous population-based household surveys— out this chapter is in addition to mapping infection by primarily the Demographic and Health Surveys for key populations (Wilson and Taaffe 2017). The principle 24 countries conducted between 2007 and 2013 and for of “knowing your epidemic” (UNAIDS 2008) has shifted the first 90 percent. to “know your local epidemics,” and the practice of mon- itoring data at the country or province level has shifted • Given the methodological issues and potential biases to collecting data at the community and facility levels related to survey data, getting a robust estimate of (UNAIDS 2014c). These more precise pictures of the the true percentage of people in a population who epidemic permit local governments to set priorities, are infected remains a challenge, although the use of Global Mortality and Morbidity of HIV/AIDS 39 biologic sampling and careful attention to regional responsive to and informed by effective interventions for variations have improved newer surveys. treatment and other prevention interventions. • The first 90 percent, HIV diagnosis, is estimated Notably, however, the critical nature of this approach using the number of people diagnosed with HIV is not new. It parallels the need for targeted active case- divided by the estimate of people living with HIV. finding, long recognized as a critical aspect of epidemic UNAIDS produces annual estimates for the denomi- control for diseases such as smallpox (Kerrod and others nator (number of people living with HIV). 2005), severe acute respiratory syndrome (Cheng and • The second 90 percent, treatment coverage, is esti- others 2013), Ebola virus (Tom-Aba and others 2015), mated using the annual country figure for the num- and tuberculosis (Yuen and others 2015). The ability to ber of people on treatment (N) divided by the accomplish disease control is a global good and requires number of people diagnosed with HIV. UNAIDS an ongoing international effort; neither risk nor infec- produces annual and midyear estimates for treatment tion respects political boundaries, and our ability to coverage (Stover and others 2014; UNAIDS 2016c). improve public health is directly dependent on the weak- • The third 90 percent is derived from the universal est link in the chain. viral load indicator that is defined in the Global AIDS Response Progress reports and is the percentage of people on ART who are virally suppressed, but it is NOTE the estimate for which the data are weakest. Viral load World Bank Income Classifications as of July 2014 are as fol- data are often not available for everyone on treatment, lows, based on estimates of gross national income (GNI) per and surrogate measures such as the proportion of capita for 2013: samples in the national lab that meet viral suppression criteria are often used for estimates (UNAIDS 2015c). • Low-income countries (LICs) = US$1,045 or less • Middle-income countries (MICs) are subdivided: As noted previously, accurate estimates can only be (a) lower-middle-income = US$1,046 to US$4,125 derived using a cohort approach. Valid measurement of (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 these targets requires that individuals be followed longi- • High-income countries (HICs) = US$12,746 or more. tudinally from the time they are tested until the time they are virally suppressed. This level of granularity would also provide more robust estimates of microepi- REFERENCES demics and key populations in both prevalence and AIDS.gov. 2017. “National HIV/AIDS Strategy: Overview.” incidence. This level is essential for monitoring long- https://www.aids.gov/federal-resources/national-hiv-aids term follow-up. -strategy/overview/. AIDSVu. 2016. “Maps—AIDSVu.” Rollins School of Public Health, Emory University. http://aidsvu.org/map/. CONCLUSIONS Anderson, S. J., P. Cherutich, N. Kilonzo, I. Cremin, D. Fecht, and others. 2014. “Maximising the Effect of Combination As HIV/AIDS programmatic efforts transition to a uni- HIV Prevention through Prioritisation of the People and fied focus on ending the epidemic by 2030 (UNAIDS Places in Greatest Need: A Modelling Study.” The Lancet 2015d), new demands, including accurate monitoring of 384 (9939): 249–56. 90-90-90 goals, are being placed on monitoring and sur- Attia, S., M. Egger, M. Muller, M. Zwahlen, and N. Low. 2009. veillance strategies. Broad measures of national preva- “Sexual Transmission of HIV According to Viral Load lence and mortality are no longer sufficient to guide and Antiretroviral Therapy: Systematic Review and Meta- interventions that will need to increase in efficiency to Analysis.” AIDS 23 (11): 1397–404. effectively target hard-to-reach populations. Data needed Bendavid, E., E. Leroux, J. Bhattacharya, N. Smith, and to target key populations and microepidemics will need G. Miller. 2010. “The Relation of Price of Antiretroviral to be collected through subnational surveillance meth- Drugs and Foreign Assistance with Coverage of HIV Treatment in Africa: Retrospective Study.” BMJ 341: C6218. ods, which have both time and budget implications. Bertozzi, S., N. S. Padian, J. Wegbreit, L. M. Demaria, Expanding resources for surveillance and monitoring of B. Feldman, and others. 2006. “HIV/AIDS Prevention and a disease that is already one of the best documented in Treatment.” In Disease Control Priorities in Developing history may not be appealing in an era of ever-increasing Countries (second edition), edited by D. T. Jamison, competing priorities, but it will be necessary if programs J. G. Breman, A. R. Measham, G. Alleyne, M. Claeson, are to be targeted in order to maximize their effect. D. B. Evans, P. Jha, A. Mills, and P. Musgrove. Washington, Improved surveillance systems and techniques must be DC: Oxford University Press and World Bank. 40 Major Infectious Diseases Bloom, B. R., R. Atun, T. Cohen, C. Dye, H. Fraser, and Flaxman, A. D., T. Vos, and C. J. Murray. 2015. An Integrative others. 2017. “Tuberculosis.” In Disease Control Priorities Metaregression Framework for Descriptive Epidemiology. (third edition): Volume 6, Major Infectious Diseases, edited Publications on Global Health, Institute for Health by K. K. Holmes, S. Bertozzi, B. R. Bloom, and P. Jha. Metrics and Evaluation. Seattle, WA: University of Washington, DC: World Bank. Washington Press. Brookmeyer, R. 2010. “Measuring the HIV/AIDS Epidemic: Garnett, G. P., S. Krishnaratne, K. Harris, T. B. Hallett, Approaches and Challenges.” Epidemiologic Reviews 32: 26–37. M. Santos, and others. 2017. “Cost-Effectiveness of Busch, M. P., C. D. Pilcher, T. D. Mastro, J. Kaldor, G. Vercauteren, Interventions to Prevent HIV Acquisition.” In Disease and others. 2010. “Beyond Detuning: 10 Years of Progress Control Priorities (third edition): Volume 6, Major Infectious and New Challenges in the Development and Application Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, of Assays for HIV Incidence Estimation.” AIDS 24 (18): and P. Jha. Washington, DC: World Bank. 2763–71. Geng, E. H., D. R. Bangsberg, N. Musinguzi, N. Emenyonu, Cheng, V. C., J. F. Chan, K. K. To, and K. Y. Yuen. 2013. “Clinical M. B. Bwana, and others. 2010. “Understanding Reasons Management and Infection Control of SARS: Lessons for and Outcomes of Patients Lost to Follow-Up in Learned.” Antiviral Research 100 (2): 407–19. Antiretroviral Therapy Programs in Africa Through a Chesson, H. W., S. O. Aral, and P. Mayaud. 2017. “Sexually Sampling-Based Approach.” Journal of Acquired Immune Transmitted Infections: Impact and Cost-Effectiveness of Deficiency Syndromes 53 (3): 405–11. Prevention.” In Disease Control Priorities (third edition): Geng, E. H., M. B. Bwana, W. Muyindike, D. V. Glidden, Volume 6, Major Infectious Diseases, edited by K.K. Holmes, D. R. Bangsberg, and others. 2013. “Failure to Initiate S. Bertozzi, B. R. Bloom, and P. Jha. Washington, DC: Antiretroviral Therapy, Loss to Follow-Up and Mortality World Bank. among HIV-Infected Patients during the Pre-ART Period in Cohen, M. S., Y. Q. Chen, M. Mccauley, T. Gamble, Uganda.” Journal of Acquired Immune Deficiency Syndromes M. C. Hosseinipour, and others. 2011. “Prevention of HIV-1 63 (2): e64–71. Infection with Early Antiretroviral Therapy.” New England Geng, E. H., N. Emenyonu, M. B. Bwana, D. V. Glidden, Journal of Medicine 365 (6): 493–505. and J. N. Martin. 2008. “Sampling-Based Approach to Coulibaly, G., M. N’Dhatz, K. Domoua, E. Aka-Danguy, Determining Outcomes of Patients Lost to Follow-Up F. Traore, and others. 1994. “[Prevalence of Pneumocystosis in Antiretroviral Therapy Scale-Up Programs in Africa.” in HIV-Infected Patients in a Pneumology Unit. Autopsy Journal of the American Medical Association 300 (5): 506–7. Study Performed in Abidjan (Côte d’Ivoire)].” Revue de Geng, E. H., D. V. Glidden, D. R. Bangsberg, M. B. Bwana, Pneumologie Clinique 50 (3): 116–20. N. Musinguzi, and others. 2012. “A Causal Framework Danel, C., R. Moh, D. Gabillard, A. Badje, J. Le Carrou, and for Understanding the Effect of Losses to Follow-Up on others. 2015. “A Trial of Early Antiretrovirals and Isoniazid Epidemiologic Analyses in Clinic-Based Cohorts: The Case Preventive Therapy in Africa.” New England Journal of of HIV-Infected Patients on Antiretroviral Therapy in Medicine 373 (9): 808–22. Africa.” American Journal of Epidemiology 175 (10): Das, M., P. L. Chu, G. M. Santos, S. Scheer, E. Vittinghoff, and 1080–87. others. 2010. “Decreases in Community Viral Load Are Geng, E. H., D. V. Glidden, N. Emenyonu, N. Musinguzi, Accompanied by Reductions in New HIV Infections in San M. B. Bwana, and others. 2010. “Tracking a Sample Francisco.” PLoS One 5 (6): e11068. of Patients Lost to Follow-Up Has a Major Impact on Del Rio, C. 2015. “Health Disparities: What Do They Have to Understanding Determinants of Survival in HIV-Infected Do with CFAR?” 19th Annual CFAR Meeting, Center for Patients on Antiretroviral Therapy in Africa.” Tropical Applied Rationality, Seattle, WA, November 4–6. Medicine and International Health 15 (Suppl 1): 63–69. Domoua, K., M. N’Dhatz, G. Coulibaly, F. Traore, Geng, E. H., D. Nash, A. Kambugu, Y. Zhang, P. Braitstein, J. B. Konan, and others. 1995. “[Autopsy Findings and others. 2010. “Retention in Care among HIV-Infected in 70 AIDS Patients Who Died in a Department of Patients in Resource-Limited Settings: Emerging Insights and Pneumology in Ivory Coast: Impact of Tuberculosis].” New Directions.” Current HIV/AIDS Reports 7 (4): 234–44. Medecine Tropicale: Revue du Corps de Sante Colonial 55 Geng, E. H., T. A. Odeny, and R. E. Lyamuya. 2016. “Retention (3): 252–54. in Care and Patient-Reported Reasons for Undocumented Egger, M., B. D. Spycher, J. Sidle, R. Weigel, E. H. Geng, and Transfer or Stopping Care among HIV-Infected Patients others. 2011. “Correcting Mortality for Loss to Follow-Up: on Antiretroviral Therapy in Eastern Africa: Application A Nomogram Applied to Antiretroviral Treatment of a Sampling-Based Approach.” Clinical Infectious Diseases Programmes in Sub-Saharan Africa.” PLoS Medicine 8 (1): 62 (7): 935–44. e1000390. Geng, E. H., T. A. Odeny, R. E. Lyamuya, A. Nakiwogga- Fang, C. T., H. M. Hsu, S. J. Twu, M. Y. Chen, Y. Y. Chang, and Muwanga, L. Diero, and others. 2015. “Estimation of others. 2004. “Decreased HIV Transmission after a Policy Mortality among HIV-Infected People on Antiretroviral of Providing Free Access to Highly Active Antiretroviral Treatment in East Africa: A Sampling Based Approach in Therapy in Taiwan.” Journal of Infectious Diseases 190 (5): an Observational, Multisite, Cohort Study.” The Lancet HIV 879–85. 2 (3): e107–16. Global Mortality and Morbidity of HIV/AIDS 41 Gouws, E., and P. Cuchi, on behalf of the International INSIGHT START Study Group. 2015. “Initiation of Collaboration on Estimating HIV Incidence by Modes of Antiretroviral Therapy in Early Asymptomatic HIV Transmission. 2012. “Focusing the HIV Response Through Infection.” New England Journal of Medicine 373 (9): Estimating the Major Modes of HIV Transmission: A 795–807. Multi-Country Analysis.” Sexually Transmitted Infections 88 Jamison, D. T., J. G. Breman, A. R. Measham, G. Alleyne, (Suppl 2): i76–85. M. Claeson, D. B. Evans, P. Jha, A. Mills, and P. Musgrove, Granich, R. 2016. “Adoption of New Guidelines: A Work editors. 2006. Disease Control Priorities in Developing In Progress.” Presentation at the UN 90-90-90 Target Countries (second edition). Washington, DC: Oxford Workshop, Durban, South Africa, July 17. University Press and World Bank. Granich, R., S. Gupta, B. Hersh, B. Williams, J. Montaner, and Johnson, L. F., J. Mossong, R. E. Dorrington, M. Schomaker, others. 2015. “Trends in AIDS Deaths, New Infections and C. J. Hoffmann, and others. 2013. “Life Expectancies of ART Coverage in the Top 30 Countries with the Highest South African Adults Starting Antiretroviral Treatment: AIDS Mortality Burden; 1990–2013.” PLoS One 10 (7): Collaborative Analysis of Cohort Studies.” PLoS Medicine e0131353. 10 (4): e1001418. Granich, R., S. Gupta, J. Montaner, B. Williams, and J. M. Zuniga. John-Stewart, G., R. Peeling, C. Levin, P. Garcia, D. Mabey, 2016. “Pattern, Determinants, and Impact of HIV Spending and J. Kinuthia. 2017. “Prevention of Mother-to-Child on Care and Treatment in 38 High-Burden Low- and Transmission of HIV and Syphilis.” In Disease Control Middle-Income Countries.” Journal of the International Priorities (third edition): Volume 6, Major Infectious Association of Providers of AIDS Care 15 (2): 91–100. Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, Greenberg, A. E., S. Lucas, O. Tossou, I. M. Coulibaly, and P. Jha. Washington, DC: World Bank. D. Coulibaly, and others. 1995. “Autopsy-Proven Causes of Kerrod, E., A. M. Geddes, M. Regan, and S. Leach. 2005. Death in HIV-Infected Patients Treated for Tuberculosis in “Surveillance and Control Measures during Smallpox Abidjan, Côte d’Ivoire.” AIDS 9 (11): 1251–54. Outbreaks.” Emerging Infectious Diseases 11 (2): 291–97. Hallett, T. B., B. Zaba, J. Todd, B. Lopman, W. Mwita, and Kitahata, M. M., S. J. Gange, A. G. Abraham, B. Merriman, others. 2008. “Estimating Incidence from Prevalence in M. S. Saag, and others. 2009. “Effect of Early versus Generalised HIV Epidemics: Methods and Validation.” Deferred Antiretroviral Therapy for HIV on Survival.” PLoS Medicine 5 (4): e80. New England Journal of Medicine 360 (18): 1815–26. Harripersaud, K., M. McNairy, S. Ahmed, E. J. Abrams, Levi, J., A. Raymond, A. Pozniak, P. Vernazza, P. Kohler, H. Thirumurthy, and W. M. El-Sadr. 2017. “Management and others. 2015. “Can the UNAIDS 90–90–90 Target Be of HIV/AIDS and Co-Morbidities in Children and Adults: Reached? Analysis of 12 National Level HIV Treatment Cost-Effectiveness Considerations.” In Disease Control Cascades.” Journal of the International AIDS Society 18 Priorities (third edition): Volume 6, Major Infectious (5 Suppl 4). Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, Levi, J., A. Raymond, A. Pozniak, P. Vernazza, P. Kohler, and P. Jha. Washington, DC: World Bank. and others. 2016. “Can the UNAIDS 90-90-90 Target Be Hirnschall, G. 2015. “Key Populations: Demographics, Achieved? A Systematic Analysis of National HIV Treatment Epidemiology, Epidemic Drivers.” Presentation given at the Cascades.” BMJ Global Health 1 (2). IAS-ILF Thematic Roundtable on Key Populations, Seattle, Lopez-Cortes, L. F., A. Gutierrez-Valencia, and O. J. Ben- WA, February 24. Marzouk-Hidalgo. 2016. “Antiretroviral Therapy in Early Holmes, C. B., T. Hallett, R. P. Walensky, T. Bärnighausen, HIV Infection.” New England Journal of Medicine 374 (4): 393. Y. Pillay, and others. 2017. “Effectiveness and Cost- Lopman, B. A., R. V. Barnabas, J. T. Boerma, G. Chawira, Effectiveness of Treatment as Prevention for HIV/AIDS.” In K. Gaitskell, and others. 2006. “Creating and Validating Disease Control Priorities (third edition): Volume 6, Major an Algorithm to Measure AIDS Mortality in the Adult Infectious Diseases, edited by K. K. Holmes, S. Bertozzi, Population Using Verbal Autopsy.” PLoS Medicine 3 (8): e312. B. R. Bloom, and P. Jha. Washington, DC: World Bank. Lucas, S. B., M. Diomande, A. Hounnou, A. Beaumel, IAPAC (International Association of Providers of AIDS Care.) C. Giordano, and others. 1994. “HIV-Associated Lymphoma 2016. “Global HIV 90-90-90 Watch.” http://www.hiv90-90 in Africa: An Autopsy Study in Côte d’Ivoire.” International -90watch.org/. Journal of Cancer 59 (1): 20–24. ICAP (International Center for AIDS Care and Treatment Lucas, S. B., A. Hounnou, C. Peacock, A. Beaumel, A. Kadio, Program). 2016. “Population-Based HIV Impact and others. 1994. “Nocardiosis in HIV-Positive Patients: Assessments.” Columbia Mailman School of Public Health, An Autopsy Study in West Africa.” International Journal of Columbia University, New York. http://icap.columbia.edu Tuberculosis and Lung Disease 75 (4): 301–7. /global-initatives/the-phia-project/. Lundgren, J., A. G. Babiker, and J. D. Neaton. 2016. IHME (Institute for Health Metrics and Evaluation). 2016a. “Antiretroviral Therapy in Early HIV Infection.” “GBD Compare.” http://vizhub.healthdata.org/gbd New England Journal of Medicine 374 (4): 394. -compare/. Magadi, M. A. 2013. “The Disproportionate High Risk of HIV ———. 2016b. “Data and Tools.” IHME, University of Infection among the Urban Poor in Sub-Saharan Africa.” Washington, Seattle. http: //www.healthdata.org/data-tools. AIDS and Behavior 17 (5): 1645–54. 42 Major Infectious Diseases Mahy, M. 2016. “Global Adolescent Mortality Data: Where Samji, H., A. Cescon, R. S. Hogg, S. P. Modur, K. N. Althoff, and Do the Numbers Come from and How Can We Improve others. 2013. “Closing the Gap: Increases in Life Expectancy Them?” Powerpoint presentation at AIDS 2016: 21st among Treated HIV-Positive Individuals in the United International AIDS Conference, Durban, South Africa, July. States and Canada.” PLoS One 8 (12): e81355. Marston, M., K. Harriss, and E. Slaymaker. 2008. “Non- SMART (Strategies for Management of Antiretroviral Therapy) Response Bias in Estimates of HIV Prevalence Due to Study Group, W. M. El-Sadr, J. Lundgren, J. D. Neaton, the Mobility of Absentees in National Population-Based F. Gordin, and others. 2006. “CD4+ Count-Guided Surveys: A Study of Nine National Surveys.” Sexually Interruption of Antiretroviral Treatment.” New England Transmitted Infections 84 (Suppl 1): i71–77. Journal of Medicine 355 (22): 2283–96. Mathers, C. D., D. M. Fat, M. Inoue, C. Rao, and A. D. Lopez. Sterne, J. A., M. May, D. Costagliola, F. de Wolf, A. N. Phillips, 2005. “Counting the Dead and What They Died from: An and others. 2009. “Timing of Initiation of Antiretroviral Assessment of the Global Status of Cause of Death Data.” Therapy in AIDS-Free HIV-1-Infected Patients: A Bulletin of the World Health Organization 83 (3): 171–77. Collaborative Analysis of 18 HIV Cohort Studies.” The May, M. T., M. Gompels, V. Delpech, K. Porter, C. Orkin, and Lancet 373 (9672): 1352–63. others. 2014. “Impact on Life Expectancy of HIV-1 Positive Stover, J., K. Andreev, E. Slaymaker, C. Gopalappa, K. Sabin, Individuals of CD4+ Cell Count and Viral Load Response and others. 2014. “Updates to the Spectrum Model to to Antiretroviral Therapy.” AIDS 28 (8): 1193–202. Estimate Key HIV Indicators for Adults and Children.” MEASURE Evaluation. 2016. PLACE: Priorities for Local AIDS AIDS 28 (Suppl 4): S427–34. Control Efforts. Chapel Hill, NC: University of North Stover, J., L. Bollinger, J. A. Izazola, L. Loures, P. DeLay, Carolina. http://www.cpc.unc.edu/measure/resources/tools and others. 2016. “What Is Required to End the AIDS /hiv-aids/place. Epidemic as a Public Health Threat by 2030? The Cost Médecins Sans Frontières. 2013. Untangling the Web of and Impact of the Fast-Track Approach.” PLoS One 11 (5): Antiretroviral Price Reductions, 16th ed. Geneva: Médecins e0154893. Sans Frontières. Tanser, F., T. Bärnighausen, G. S. Cooke, and M. L. Newell. Melendez, R. M., T. A. Exner, A. A. Ehrhardt, B. Dodge, 2009. “Localized Spatial Clustering of HIV Infections in R. H. Remien, and others. 2006. “Health and Health Care a Widely Disseminated Rural South African Epidemic.” among Male-to-Female Transgender Persons Who Are HIV International Journal of Epidemiology 38 (4): 1008–16. Positive.” American Journal of Public Health 96 (6): 1034–37. Tanser, F., T. de Oliveira, M. Maheu-Giroux, and T. Bärnighausen. Menzies, N. A., A. Berruti, and J. M. Blandford. 2012. “The 2014. “Concentrated HIV Subepidemics in Generalized Determinants of HIV Treatment Costs in Resource Limited Epidemic Settings.” Current Opinion in HIV and AIDS 9 Settings.” PLoS ONE 8 (5): http://dx.doi.org/10.1371 (2): 115–25. /journal.pone.0048726. Thacker, S. B., D. F. Stroup, V. Carande-Kulis, J. S. Marks, Montaner, J. S., R. Hogg, E. Wood, T. Kerr, M. Tyndall, and K. Roy, and others. 2006. “Measuring the Public’s Health.” others. 2006. “The Case for Expanding Access to Highly Public Health Reports 121 (1): 14–22. Active Antiretroviral Therapy to Curb the Growth of the Tom-Aba, D., A. Olaleye, A. T. Olayinka, P. Nguku, N. Waziri, HIV Epidemic.” The Lancet 368 (9534): 531–36. and others. 2015. “Innovative Technological Approach to Montaner, J. S., V. D. Lima, R. Barrios, B. Yip, E. Wood, and Ebola Virus Disease Outbreak Response in Nigeria Using others. 2010. “Association of Highly Active Antiretroviral the Open Data Kit and Form Hub Technology.” PLoS One Therapy Coverage, Population Viral Load, and Yearly New 10 (6): e0131000. HIV Diagnoses in British Columbia, Canada: A Population- UNAIDS (Joint United Nations Programme on HIV/AIDS). Based Study.” The Lancet 376 (9740): 532–39. 2008. UNAIDS Annual Report: Knowing Your Epidemic. Murray, C. J., and A. D. Lopez. 2013. “Measuring the Global Geneva: UNAIDS. Burden of Disease.” New England Journal of Medicine 369 ———. 2012. UNAIDS Report on the Global AIDS Epidemic (5): 448–57. 2012. Geneva: United Nations. National AIDS Control Organisation. 2012. HIV Sentinel ———. 2013. “Methodology: Understanding the HIV Surveillance 2010–2011: A Technical Brief. New Delhi: Estimates.” UNAIDS, Geneva. Department of AIDS Control, Ministry of Health and ———. 2014a. 90–90–90—An Ambitious Treatment Target to Family Welfare, Government of India. Help End the AIDS Epidemic. Geneva: UNAIDS. PEPFAR (President’s Emergency Plan for AIDS Relief). 2013. ———. 2014b. The Gap Report. Geneva: UNAIDS. “2013 Report on Costs of Treatment in the President’s ———. 2014c. Local Epidemics Issues Brief. Geneva: UNAIDS. Emergency Plan for AIDS Relief (PEPFAR).” PEPFAR, ———. 2014d. “Methodology: Understanding the HIV Washington, DC. Estimates.” UNAIDS, Geneva. Rodger, A. J., R. Lodwick, M. Schechter, S. Deeks, J. Amin, and ———. 2014e. “Ambitious Treatment Targets: Writing the others. 2013. “Mortality in Well Controlled HIV in the Final Chapter of the AIDS Epidemic.” UNAIDS discussion Continuous Antiretroviral Therapy Arms of the SMART paper, UNAIDS, Geneva. and ESPRIT Trials Compared with the General Population.” ———. 2015a. “‘15 by 15’: A Target Achieved.” UNAIDS, Geneva. AIDS 27 (6): 973–79. ———. 2015b. “AIDS by the Numbers 2015.” UNAIDS, Geneva. Global Mortality and Morbidity of HIV/AIDS 43 ———. 2015c. Global AIDS Response Progress Reporting 2015. 1980–2015: The Global Burden of Disease Study 2015.” Geneva: UNAIDS. The Lancet HIV 3 (8): e361–87. ———. 2015d. On the Fast Track to End AIDS by 2030: Focus WHO (World Health Organization). 2010. Antiretroviral Therapy on Location and Population. Geneva: UNAIDS. for HIV Infection in Adults and Adolescents: Recommendations ———. 2015e. How AIDS Changed Everything. Geneva: for a Public Health Approach. Geneva: WHO. UNAIDS. ———. 2011. Prevention and Treatment of HIV and Other ———. 2016a. “AIDSinfo.” http://aidsinfo.unaids.org/. Sexually Transmitted Infections among Men Who Have Sex ———. 2016b. “Global AIDS Update 2016.” UNAIDS, Geneva. with Men and Transgender People: Recommendations for a ———. 2016c. “Spectrum/EPP.” UNAIDS, Geneva. http:// Public Health Approach. Geneva: WHO. www.unaids.org/en/dataanalysis/datatools/spectrumepp. ———. 2013a. Consolidated Guidelines on the Use of UNAIDS Reference Group on Estimates, Modelling and Antiretroviral Drugs for Treating and Preventing HIV Projections. 2013. “Identifying Populations at Greatest Risk Infection: Recommendations for a Public Health Approach. of Infection: Geographic Hotspots and Key Populations.” Geneva: WHO. UNAIDS, Geneva. ———. 2013b. “Global Health Estimates.” http://www.who United Nations Population Division. 2014. World Population .int/healthinfo/global_burden_disease/en/. Prospects: The 2004 Revision. New York: United Nations. ———. 2013c. Global Update on HIV Treatment 2013: Results, U.S. CDC (United States Centers for Disease Control and Impact and Opportunities. Geneva: WHO. Prevention). 2011. “HIV among Transgender People.” ———. 2013d. “WHO Methods and Data Sources for Global http://www.cdc.gov/hiv/group/gender/transgender/. Burden of Disease Estimates 2000–2011.” WHO, Geneva. U.S. National Institutes of Health. 2015. “Starting Antiretroviral ———. 2015a. Global Health Sector Response to HIV, Treatment Early Improves Outcomes for HIV-Infected 2000–2015: Focus on Innovations in Africa. Geneva: WHO. Individuals.” News release, May 27. ———. 2015b. Guideline on When to Start Antiretroviral Therapy Violari, A., M. F. Cotton, D. M. Gibb, A. G. Babiker, J. Steyn, and and on Pre-Exposure Prophylaxis for HIV. Geneva: WHO. others. 2008. “Early Antiretroviral Therapy and Mortality Wilson, D., and J. Taaffe. 2017. “Tailoring the Local HIV among HIV-Infected Infants.” New England Journal of Response to Local HIV Epidemics.” In Disease Control Medicine 359 (21): 2233–44. Priorities (third edition): Volume 6, Major Infectious Walensky, R. P., E. L. Ross, N. Kumarasamy, R. Wood, F. Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, Noubary, and others. 2013. “Cost-Effectiveness of HIV and P. Jha. Washington, DC: World Bank. Treatment as Prevention in Serodiscordant Couples.” Yuen, C. M., F. Amanullah, A. Dharmadhikari, E. A. Nardell, New England Journal of Medicine 369 (18): 1715–25. J. A. Seddon, and others. 2015. “Turning Off the Tap: Wang, H., T. M. Wolock, A. Carter, G. Nguyen, H. H. Kyu, Stopping Tuberculosis Transmission through Active Case- and others. 2016. “Estimates of Global, Regional, and Finding and Prompt Effective Treatment.” The Lancet 386 National Incidence, Prevalence, and Mortality of HIV, (10010): 2334–43. 44 Major Infectious Diseases Chapter 3 HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health Corey Casper, Heidi Crane, Manoj Menon, and Deborah Money INTRODUCTION The first portrayals of the epidemic in the early 1980s The total number of people living with human in the United States described a surge of cancer cases immunodeficiency virus (HIV) worldwide continues among men who have sex with men (MSM). Since the to grow annually, attributable to both new infections initial reports from the early epidemic, it has become clear and the increased longevity of infected persons that HIV threatens the reproductive health of women treated with potent antiretroviral therapy (ART). This across income settings and populations. Paradoxically, in growing population bears the burden of associated an age of new hope for prolonged lifespan stemming from health conditions that complicate long-term HIV the success of ART, increased longevity is also bringing a infection. Specifically, secondary epidemics of cancer; host of noncommunicable chronic comorbidities (NCCs). reproductive ill health; and noncommunicable Among cancers—the focus of the first section of diseases, such as cardiovascular disease, diabetes, this chapter—the pandemic initially saw an annual rise renal dysfunction, and liver damage, have been through 1996 in what came to be known as acquired observed across the globe. This wide spectrum of ill- immune deficiency syndrome (AIDS)–defining cancers nesses complicating ongoing HIV infection is a chal- (ADCs), including Kaposi sarcoma (KS), non-Hodgkin lenging global health threat and underscores the need lymphoma (NHL), and cervical cancer (CDC 1992). for a greater understanding of these comorbidities, These cancers decreased in incidence with the widespread broader access to treatment, and increasingly sophis- availability of ART in high-resource settings but never fell ticated treatment to avoid widespread preventable to the levels seen before HIV; they continue with little morbidity and death. change in incidence in low- and middle-income countries This chapter provides an overview of some of the (LMICs). Additionally, a troubling rise in other, non-AIDS most common, most rapidly increasing, or most morbid defining cancers (NADCs)—such as anal cancer, hepatoc- complications of persistent HIV infection; it is not ellular carcinoma (HCC), and lung cancer—continues to meant to be exhaustive. Additionally, many comorbidi- be observed globally despite access to ART. ties of long-term HIV infection are addressed in detail in The next section of the chapter considers the multi- other chapters of this volume. faceted impact of HIV on women’s reproductive health. Corresponding author: Corey Casper, Fred Hutchinson Cancer Research Center, Seattle, Washington, United States; ccasper@fredhutch.org. 45 Girls and women who are at risk of HIV or who are and cervical cancer. These malignancies, all associated already infected are vulnerable to problems of access with a viral oncogenic infection, were frequently found in to adequate reproductive choice and contraceptive individuals with low CD4 T-cell counts, and therefore options. Yet, the evidence suggests that preventing common before ART became widely available. mother-to-infant transmission of HIV can effectively be Over time, additional cancers were observed to be achieved through the prevention of unplanned or more common in people with HIV, but these were not unwanted pregnancies. Barrier contraceptive methods, considered to be AIDS defining (Patel and others although suboptimal at preventing pregnancy, do pro- 2008; Shiels and others 2011; Silverberg and others tect against HIV. The section also examines the compli- 2015). These NADCs—which are significantly more cated influence of HIV on infertility in women and the common in people with HIV than in peers without it transmission of human papillomavirus (HPV), herpes in the same population—include anal cancer, HCC, simplex virus (HSV), pelvic inflammatory disease, bac- and lung cancer. terial vaginosis, and others. Today, incident cancer cases are roughly equally The final section of the chapter highlights the wide divided between ADCs and NADCs. In the United States, range of NCCs associated with long-standing HIV infec- approximately one person will develop cancer for every tion, including cardiovascular and metabolic illnesses. 100 people with HIV each year (Riedel and others 2013). Taken together, the data from many parts of the world One study estimates a similar incidence of cancer among clearly show that despite the dramatic decrease in deaths HIV-infected South Africans receiving ART (0.87 cases due to HIV with the continued expansion of access to of cancer developing for every 100 people with HIV fol- successful treatment, access to ART alone will not pre- lowed over a year) (Sengayi and others 2016). vent, and in some cases may precipitate, a wide spectrum The time that has passed since initial infection with of complications of long-term HIV infection. These HIV also influences which cancer arises; ADCs are more challenges require the attention of health care providers, common soon after diagnosis, and NADCs are more policy makers, and researchers, empowered by access to common after five or more years (Robbins and others accurate information on the burden of these diseases 2015). Survival after a diagnosis of cancer accompanying and the growing number of potential solutions. HIV appears to be considerably shorter, on average, espe- cially for NADCs, compared with similar cancers in peo- ple without HIV (Achenbach and others 2011; Coghill HIV AND CANCER and others 2015). Less is known about the epidemiology of cancer At the beginning of the global HIV pandemic, an outbreak among people with HIV in LMICs. Cancer registries that of cancer among MSM in the United States was described cover the entire population are few in these areas and in 1981 (Hymes and others 1981). Since then, the evidence do not routinely capture HIV status of cancer cases. suggests that the risk of cancer is 2–3,000 times higher A retrospective analysis from 13 clinical sites in the East among people with HIV than those who are not infected Asia and Pacific region, as part of the TREAT Asia HIV (Grulich and others 2007); up to 9 percent of people living Observational Database study, revealed data comparable with long-term HIV infection will develop cancer over the to that seen among the non-Asian HIV populations. In course of their HIV care (Shiels and others 2011). Cancer that analysis, CD4 T-cell counts greater than 200 cells per is the leading non-AIDS cause of death worldwide among cubic millimeter (mm3) were found protective against people with HIV (Smith and others 2014). ADCs; older patients and those not on potent ART were more likely to be diagnosed with an NADC (Petoumenos and others 2010). Epidemiology of Cancer among People with HIV Attempts to match cancer registries with data on In high-income countries (HICs), the incidence of cancer individual HIV status have been conducted in Uganda in people with HIV rose annually until 1996, when access (Mbulaiteye and others 2006), Nigeria (Akarolo-Anthony to potent ART became widely available and observers and others 2014), and South Africa (Sengayi 2016), and noted a sharp decline in new cancer cases (Shiels and found the cancers to be more common among people others 2011). with HIV, similar to the United States. Hospital-based Since 1998, however, the incidence of cancer among cancer registries in India also show a similar spectrum of people with HIV has been gradually rising annually in the cancer in people with HIV, except for a notably lower United States. Initially, the overwhelming number were frequency of KS (Venkatesh and others 2012). ADCs (CDC 1992), including KS; NHL; Burkitt, immuno- Few studies have assessed survival after diagnosis of blastic, and primary central nervous system lymphomas; HIV-associated malignancies (HIVAMs) in resource-limited 46 Major Infectious Diseases settings, but the odds of death appear to be significantly In a meta-analysis involving nearly 200 studies and higher among people with HIV (Coghill and others 2013). more than 1 million women who had a normal cervical Cohort studies conducted in high-income regions have cytologic examination, HPV DNA was detected in nearly found cancer to be among the leading causes of death of 12 percent of women globally, although with marked people with HIV (Bonnet and others 2004; Kowalska and geographic variation. Regions of frequent HPV detec- others 2012; Smith and others 2014). tion include the Caribbean (35 percent) and Eastern Africa (34 percent), yet frequency is only 10 percent or less in Northern Europe (10 percent), Northern Africa Pathogenesis of HIV-Associated Malignancies (9 percent), Western Europe (9 percent), Southern People with HIV are predisposed to a higher risk of Europe (9 percent), Southern Asia (7 percent), North cancer through a variety of mechanisms. First, the America (5 percent), and Western Asia (2 percent) immunosuppression that accompanies CD4 T-cell (Bruni and others 2010). depletion may lead to the development of cancer when More than 100 types of HPV exist, but HPV 16 and the immune system fails to seek out early cancer occur- 18 cause 70 percent of cancers. HIV may increase cervi- rences and destroy them. This helps explain why the risk cal cancer risk by increasing the rates of persistent HPV for many HIVAMs is inversely related to CD4 T-cell infection, in contrast to HPV in women without HIV count (Biggar and others 2007). whose immune systems often clear the infection (Ahdieh The observation that some cancers (cervical cancer and others 2000). and Hodgkin lymphoma) in people with HIV were not The risk of cervical cancer in women with HIV associated with lower CD4 T-cell counts—coupled with increases with age and has no direct association with CD4 epidemiologic studies suggesting that the risk for cancer T-cell count (Yanik and others 2013); the risk, however, is was greater among people with HIV than among highly reduced in women receiving ART (Blitz and others 2013). immunosuppressed recipients of organ transplantation For all women, comprehensive screening and treatment (Grulich and others 2007)—led to the exploration of programs dramatically reduce deaths from this disease. mechanisms beyond CD4 T-cell depletion that could lead to cancer in people with HIV. Kaposi Sarcoma HIV replication itself may foster the development of KS, the most common HIV-associated malignancy cancer, potentially through the induction of angiogenic, worldwide, is caused by infection with human herpes anti-apoptotic, or proliferative signaling. That failure virus 8 (HHV-8). In the United States and Europe, KS is to suppress HIV replication (or time spent with HIV 10 times more common among men than women; in detectable in the plasma) is independently associated Sub-Saharan Africa, the rates are nearly equal. In part, with risk of cancer (Bruyand and others 2009) supports KS incidence mirrors HHV-8 prevalence, which is high this hypothesis. Other mechanisms being explored that among MSM in HICs and endemic in many parts of cause cancer in persons with HIV include immune acti- Sub-Saharan Africa (Nguyen and Casper 2010). vation, immune exhaustion, and HIV integration. The KS incidence declined nearly tenfold by 1996, the year individual components of ART may also reduce cancer after potent ART became available in the United States risk beyond their effect on immune reconstitution. (Eltom and others 2002); it has declined more slowly in Gantt, Casper, and Ambinder (2013) found that protease Sub-Saharan Africa since ART was rolled out (Mutyaba inhibitors have antineoplastic properties, and nucleoside and others 2015). Mortality from KS is unusual in HICs, reverse transcriptase inhibitors (NRTIs) suppress viruses although the disease persists in up to 50 percent of that cause cancer in people with HIV. patients treated with ART and chemotherapy (Achenbach and others 2011; Nguyen and others 2008). In LMICs, KS mortality is high and response to treatment is poor Specific HIV-Associated Malignancies (Gondos and others 2005; Mosam and others 2012). Cervical Cancer Taken together, these factors mean that prevention of KS Globally, cervical cancer, caused by HPV (Forman and is paramount, calling for new therapies. others 2012), is the fourth most common cancer and third leading cause of cancer-related death among Lymphoma women; the incidence in LMICs is second only to breast The pathogenesis of AIDS-related lymphomas is often cancer (De Vuyst and others 2013; Ferlay and others attributable to tumor-causing viruses, including Epstein- 2013). In Sub-Saharan Africa, cervical cancer is the lead- Barr virus, HHV-8, and hepatitis C virus. Although the ing cause of cancer-related death among women (Bosch incidence of such NHLs has significantly decreased since and others 2013; Lozano and others 2012). the widespread implementation of ART, incidence among HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 47 people with HIV remains approximately 10 times greater limiting widespread global implementation (Jacobson than in the general population (Robbins and others 2014; and others 2011; Poordad and others 2011). Chapter 16 van Leeuwen and others 2009). Patients with more severe of this volume (Wiktor 2017) provides additional infor- immunodeficiency, as measured by CD4 T-cell count, mation on viral hepatitis. are at an even greater risk (Guiguet and others 2009). The incidence of Hodgkin lymphoma is significantly Anal Cancer increased among people with HIV, and its incidence has Anal cancer, like other urogenital cancers, is caused pri- not declined in parallel with the falling incidence of NHL. marily by HPV infection. Anal cancer is nearly 30 times Data suggest, however, that prolonged ART use does more common among people with HIV (Grulich and reduce the risk of Hodgkin’s lymphoma (Kowalkowski others 2007). Despite long-term use of ART, however, and others 2014). the incidence of anal cancer among people with HIV has People with HIV who also have lymphoma are more not declined (Piketty and others 2012). One large cohort likely to present with symptomatic extranodal and cen- study in the United States revealed that rates of anal tral nervous system involvement (Carbone and Gloghini cancer had increased fivefold over those before 1996, 2005). Independent of these factors, cancer-specific that is, before ART became potent (Crum-Cianflone and mortality among these patients is higher compared with others 2010). patients without HIV (Coghill and others 2015). Although anal-cancer-specific mortality among patients with HIV does not appear elevated (Coghill and Hepatocellular Carcinoma others 2015), prevention strategies to reduce morbidity, HCC is the seventh most common cause of all cancers including anal cytology and high resolution anoscopy, globally and the third leading cause of cancer-related merit further consideration (Chiao and others 2006) death; nearly 85 percent of cases of HCC and of death and are being evaluated. secondary to HCC occur in LMICs (Ferlay and others 2013). Although the pathogenesis is not entirely clear, Lung Cancer most cases of HCC worldwide are secondary to HIV infection is associated with approximately a three- hepatitis B, a double-stranded DNA virus that can be fold increased risk of lung cancer, which is the most transmitted sexually and can lead to chronic infection. common NADC and third most common cancer overall A significant proportion of the remaining burden is sec- (Engels and others 2006; Kirk and others 2007; Robbins ondary to hepatitis C (Arzumanyan, Reis, and Feitelson and others 2015). Although the higher incidence of lung 2013; McGlynn, Petrick, and London 2015). cancer among HIV patients appears to be in part inde- HIV routes of transmission are similar to those of pendent of tobacco use, the prevalence of tobacco use hepatitis B virus and hepatitis C virus, so co-infection is is several-fold higher among patients with HIV in the high. In a cohort study of more than 3,000 patients, United States compared with the general population approximately one-third of people with HIV were (Mdodo and others 2015). Smoking is associated with co-infected with hepatitis C virus (Fultz and others lung-cancer-specific mortality, which has increased in the 2003). The prevalence of hepatitis C virus infection era of ART (Kirk and others 2007). Mortality from lung tends to be higher, and the aggressiveness of hepatitis C cancer has not been associated with either immuno- virus disease is greater, among people with HIV (Puoti suppression or HIV viral load (Clifford and others 2012). and others 2004). A separate analysis documented that nearly 10 percent of HIV patients were chronic carriers of hepatitis B virus. Accordingly, cancer prevention Strategies for Prevention of HIV-Associated efforts need to target the treatment of hepatitis B and Malignancies hepatitis C before the development of HCC. Vaccines Persons with HIV and chronic viral hepatitis do not Vaccines against HPV and hepatitis B virus offer out- progress more rapidly to HCC than do HIV negative standing opportunities to reduce the burden of cancer persons, in marked contrast to other viral-associated in people with HIV. The most current HPV vaccine malignancies. However, the use of antiviral medications targets nine HPV types and is capable of eliminating in the treatment of HIV or associated diseases, including more than 90 percent of cervical cancer cases. The interferon and nucleoside or nucleotide analogs, mark- effect of the vaccine on other HPV-related cancers and edly reduces the risk of developing HCC in both HIV the efficacy among people with HIV are yet to be positive and HIV negative persons (Sung and others determined. 2008). Although current treatment options for hepatitis The World Health Organization (WHO) recom- C virus are effective, the high cost of treatment is mends universal vaccination of girls ages 9–13 years. 48 Major Infectious Diseases Uptake has been limited globally by gaps in funding and, acetic acid) are being widely adopted, although the accu- in the United States, by sensitivities around vaccines racy of these tests in high-risk populations needs to be for sexually transmitted diseases. Furthermore, this age more completely defined (Campos and others 2015; group is not covered by the routine Expanded Program Chung and others 2013; Moses and others 2015). Novel on Immunization, and alternative delivery strategies such approaches, including self-swabbing testing for oncogenic as school-based vaccination need to be further explored HPV, are being evaluated for use in LMICs. High-risk (Watson-Jones and others 2015). Although HPV vaccines lesions detected during screening can be locally ablated are approved for three doses administered over a six- to prevent the development of invasive cervical cancer month period, a two-dose HPV vaccine schedule in HIV- (WHO 2013b). negative populations has been shown to be non-inferior Few other HIV-associated cancers have been shown (Dobson and others 2013; Kreimer and others 2015). to be preventable with screening efforts. Because of the The use of the HPV vaccine among people with exist- biological similarities between cervical and anal cancer, ing HIV infection has been the subject of several studies; screening for preinvasive anal cancer has been evaluated. preliminary data show good rates of seroconversion and When performed by highly trained and experienced lower peak antibody levels, but efficacy data are pending clinicians and laboratories, screening techniques such as (Money and others 2016). A highly effective vaccine high-resolution anoscopy can detect high-grade dyspla- against hepatitis B has been seen to reduce the incidence sia (Dalla Pria and others 2014). However, evidence is of HCC in hepatitis B–endemic regions (Chang and oth- still lacking from prospective randomized trials that ers 1997; Chang and others 2009; Hsu and others 1988). ablation of these high-grade lesions reduces the risk of The WHO recommends universal vaccination against invasive anal cancer. hepatitis B with the primary infant vaccination series; Screening for lung cancer with computed tomogra- this vaccine is now included in national infant immuni- phy has been evaluated among patients with HIV and zation programs in more than 90 percent of countries may be effective. However, the incidence of the disease is (Kane 1995, 2012). low, and strategies for targeting screening efforts are Although vaccination against hepatitis B virus in sus- needed (Hulbert and others 2014). ceptible people with HIV is part of national and interna- tional guidelines (CDC 2006; WHO 2013a), the hepatitis Chemoprevention of Cancer B virus vaccine has reduced immunogenicity in people Increasing evidence suggests that the risk of cancer in with HIV (Landrum and others 2009). Programs for people with HIV can be reduced through the use of ART. catch-up vaccinations among groups at high risk of HIV In a randomized trial of early or delayed initiation of infection are also absent in low-resource settings. ART in people recently diagnosed with HIV, the risk of cancer overall was reduced by nearly two-thirds among Screening people who initiated therapy at a CD4 T-cell count of Women in the general population in the United States, more than 500 cells/mm3, compared with those who ages 21–65 years, are recommended to have cervical can- initiated at CD4 T-cell of less than 350 cells/mm3 cer screening with cytology every three years (Moyer (Klingman and others 2015). The findings included a 50 2012); the recommendation for women with HIV, in percent reduction in NADCs, a 91 percent reduction in contrast, is screening twice in the first year after initial KS, and a 70 percent reduction in lymphoma. The use of diagnosis and then annually for the rest of their lives ART is also active in preventing the acquisition of some (Panel on Opportunistic Infections in HIV-Infected viral oncogens, such as hepatitis B, which in turn reduces Adults and Adolescents 2016). The WHO, however, rec- the risk of cancer (Heuft and others 2014). ommends that if a woman with HIV has an initial nega- Finally, treatment with antiviral therapy of chronic tive screening test (that is, normal cytology, negative infections that cause cancer, such as hepatitis B or hepa- visual inspection with acetic acid, or negative cervical titis C, has been shown to reduce the risk of HCC (Shen HPV screening test), then she should undergo subse- and others 2012), although the efficacy of this strategy in quent screening within three years (WHO 2013b). people with HIV has not been further evaluated. Such screening recommendations, independent of the screening modality, may not be feasible in low-income Other Approaches to Prevention of HIV-Associated countries (LICs) with stressed health care systems. Malignancies Additionally, the economic costs and the human resources Behavioral interventions to reduce smoking among peo- required for standard cervical screening through either ple with HIV have been moderately effective and may cytology or HPV testing are limited in LICs. Low-cost reduce a broad spectrum of HIV comorbidities (Keith screening techniques (for example, visual inspection with and others 2016). HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 49 HIV and Cancer Conclusions Long-acting, injectable progestin or progestin-based Cancer is an increasingly common complication and is a implants offer effective contraception. However, con- leading cause of death in people with HIV. Strategies for cerns have arisen about declining bone mineral density cancer prevention—including vaccination, screening, with long-term progesterone use in women with HIV, and early ART initiation—can reduce the cancer burden, who are already more susceptible to bone loss (Lopez but additional strategies for reducing the burden of and others 2014). HIVAMs are needed. Furthermore, a prospective trial of preexposure prophylaxis (PrEP) with ART for serodiscordant couples found that women using injectable hormonal contracep- IMPACT OF HIV AND OTHER SEXUALLY tion were more likely than other women to acquire or TRANSMITTED INFECTIONS ON FEMALE transmit HIV infection (Heffron and others 2012). Formal studies of this association are underway. REPRODUCTIVE HEALTH Other relevant factors include the regional differences The ongoing burden of HIV in women is substantial, in cost, acceptability, and access to contraception. More with 1,000 new infections per day (UNAIDS 2014). This recently, data suggest that intrauterine devices are safe in concentration of the HIV epidemic has therefore had women with HIV and are highly effective, yet affordability substantial impact on women’s reproductive health, and limits their uptake (Wanyenze and others 2013). intersects with the ongoing and long-standing burden of other sexually transmitted infections (STIs) in women. Infertility HIV affects fertility in women in a variety of ways. Women with HIV may wish to prevent transmission by Impact of HIV Infection on Family Planning Methods avoiding sexual contact with partners; additionally, the and Conception biology of HIV itself may have an impact on fertility. Family Planning Some studies have found higher infertility in women Girls and women at risk of HIV or who are already with HIV (Yaro and others 2001), although most such infected are particularly vulnerable to the global prob- studies involved cohorts with inadequately managed lems of access to adequate reproductive choice and con- HIV disease; no data support the notion that infertility traceptive options (WHO 2006). HIV often magnifies is higher among women with HIV and who are in good the lack of personal control over reproductive decisions. health on ART. Beyond the importance of reproductive health for For women with HIV whose partners do not have women themselves, transmission of HIV from mother to HIV, timed self-insemination with partner sperm is a infant at a population level can most effectively be pre- simple and effective way to prevent risk of transmission to vented by providing improved protection from HIV to the male partner. However, many serodiscordant couples women of reproductive age and preventing unplanned will choose to have unprotected sex if the female partner and unwanted pregnancies (WHO 2006). is on adequate suppressive ART (Loutfy and others 2012). Among contraceptive methods, barrier methods such Suppressive ART has been shown to eliminate the risk as male condoms, female condoms, and cervical caps are of HIV transmission between serodiscordant couples, at suboptimal in preventing pregnancy, but they do protect least in one study (Rodger and others 2016). against HIV acquisition and transmission (Weller and For women without HIV whose male partners have Davis 2002). Dual methods of protection and contracep- HIV, prevention of acquisition from the male partner ide- tion, such as condoms and oral contraceptives, are not ally includes ensuring that the male partner is on fully acceptable to many couples, resulting in inferior meth- suppressive ART. However, with individuals not engaged ods of contraception being used in partnerships in in care or in regions where access to ART is limited, alter- which one partner is HIV positive and the other partner native approaches are important (Loutfy and others is HIV negative (serodiscordant couples) (Heffron and 2012). One option is for the woman to use PrEP. Among others 2010; Nieves and others 2015). women given oral tenofovir and emtricitabine PrEP who Concerns about the potential interactions between became pregnant, the PrEP was discontinued when the antiretroviral drugs and the estrogen component in com- pregnancy became known. Of note, in this study, preg- bination oral contraceptives notwithstanding (Robinson, nancy outcomes and infant growth did not differ in Jamshidi, and Burke 2012; Thurman, Anderson, and women taking PrEP compared with those taking a placebo Doncel 2014), recent data suggest that contraception (Mugo and others 2014). efficacy for women with HIV is not inferior to that for Overall, most HIV serodiscordant couples cannot other women (Pyra and others 2015). afford the expensive assisted reproductive treatments for 50 Major Infectious Diseases reduction of HIV transmission. In vitro fertilization times more likely to have genital warts. This risk can be and intracytoplasmic sperm injection have been used reduced in individuals with suppressed HIV (Blitz and to decrease male-to-female HIV transmission; more others 2013). recently, sperm washing and intrauterine insemination Among women undergoing in vitro fertilization, provides a less complex, less costly alternative and appears detection of HPV in the cervix was associated with sub- to be safe, with no reported cases of HIV transmission stantial reduction in pregnancy (Depuydt and others (Barnes and others 2014; Ohl and others 2005). 2016). Detection of HPV in semen was associated with Although no data suggest that couples in which both substantially lower rates of pregnancy after intrauterine partners have HIV have higher rates of infertility than insemination, and also associated with more miscar- other couples, they may face infertility treatment issues if riages (Garolla and others 2016). they coincidentally have other fertility issues. In addition to issues of access and affordability, women with HIV Herpes Simplex Virus may also face discrimination in infertility treatment. A Global distribution, incidence, and prevalence of infec- cross-sectional study in the United States found that tion with HSV—HSV-1 and HSV-2—vary widely by public attitudes toward people with HIV who seek infer- country, region, and population subgroup. Women are tility treatment are typically negative; only 38 percent of more susceptible than men to HSV-2 infection. Primary respondents favor offering it (Mok-Lin and others 2011). HSV infection contracted in the third trimester of However, health care providers have generally been sup- pregnancy may result in neurocognitive problems in the portive of women with HIV who want children; in a fetus, developmental delays, or death if the infant facility in Ghana, 94.3 percent of health care workers becomes infected (Watts and others 2003). For women providing care to women with HIV were supportive of with recurrent HSV or primary HSV infection occurring their rights to reproduction (Laar 2013). before 34 weeks gestation, prevention of transmission to Regarding pregnancy outcomes after infertility treat- the infant can be achieved with acyclovir or valacyclovir ment, a case-control study of women who received in suppression therapy during pregnancy (Workowski and vitro fertilization treatment showed that well-controlled Bolan 2015). HIV had no impact on fertility outcomes for women In the United States, the prevalence of HSV-2 among with HIV compared with those without HIV (Nurudeen adult men and women with HIV is three times higher and others 2013). Beyond fertility, greater risks of still- than in the general population (Patel and others 2012). birth, premature birth, and low birth weight have been HSV-2 may also accelerate HIV disease progression in observed in pregnancies of women with HIV (Turner co-infected individuals. However, in a randomized clini- and others 2013). cal trial involving couples in which the HIV-1–infected HIV prevention strategies aimed at men have included partner was co-infected with HSV-2, acyclovir did not voluntary male circumcision, most widely offered in reduce the risk of HIV-1 transmission, despite a reduc- high-endemic areas in Sub-Saharan Africa. A review of tion in plasma HIV-1 RNA and a 73 percent reduction in published literature in this area suggests that opportuni- the occurrence of genital ulcers due to the HSV-2 virus ties for more extensive engagement of adolescents in (Celum and others 2010). sexual and reproductive health care have been missed (Kaufman and others 2016). Pelvic Inflammatory Disease Pelvic inflammatory disease is classically associated with the ascension of microorganisms, including Neisseria Other Sexually Transmitted Infections gonorrhoeae, Chlamydia trachomatis, Mycoplasma geni- Human Papillomavirus talium, from the lower genital tract to the upper genital HPV infection is associated with the great majority of tract and of bacterial vaginosis-associated organisms squamous cell cancers of the cervix, vulva, vagina, penis, from the vagina or cervix into the endometrium and anus, and oropharynx. Women with HIV are at greater fallopian tubes (Cohen and others 1998). Long-term risk of contracting a persistent HPV infection and are consequences of pelvic inflammatory disease can include often infected by a broader range of HPV genotypes ectopic pregnancy, chronic pelvic pain, tubal infertility, (Salters and others 2016). adnexal tenderness, tubo-ovarian abscesses, fallopian Infection with HPV may also serve as a marker of tube dysfunction, difficult or painful sexual intercourse increased risk for HIV infection because both infections (dyspareunia), pelvic adhesions, and recurrent pelvic may be sexually transmitted. Massad and others (2004) inflammatory disease. found that women with HIV were four times as likely Among women with pelvic inflammatory disease, to have vulvar intraepithelial neoplasia and almost five those with HIV present with tubo-ovarian abscess HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 51 more often than women without HIV. Among Kenyan IMPACT OF NONCOMMUNICABLE CHRONIC women with lower abdominal pain and suspected pel- COMORBIDITIES IN PEOPLE WITH HIV vic inflammatory disease, histologically confirmed endometritis was three times more common in those Expanded access to ART and accompanying increases in with HIV than among those without (Cohen and oth- longevity of people with HIV have led to an increase in ers 1998). Nevertheless, treatment of pelvic inflamma- NCCs, including cardiovascular disease (CVD), diabetes tory disease has been found to be equally successful and other metabolic conditions, renal disease, liver dis- in all women, regardless of HIV status (Bukusi and ease, cancers, and mental illness. Data demonstrating others 1999). the importance of these conditions among people with HIV are often predominantly from HICs. Unfortunately, Bacterial Vaginosis much less of the data and research advances for these A meta-analysis of 23 studies found a significantly conditions are from LMICs, where most people with increased risk of HIV incidence among women with HIV live (Narayan and others 2014). In many regions, bacterial vaginosis (relative risk 1.6, 95 percent confi- such as Sub-Saharan Africa, HIV care, including with dence interval 1.2, 2.1) (Atashili and others 2008). In ART, is more widely available, but is not accompanied by a case-control analysis of 5,110 women in South care for NCCs, resulting in preventable morbidity and Africa, bacterial vaginosis at baseline enrollment was death (Narayan and others 2014). associated with double the risk of acquiring HIV infection during 36 months’ follow-up, after adjust- Cardiovascular Disease ing for demographic characteristics, other STIs, and Burden and Epidemiology of CVD in People with HIV sexual behavior (Myer and others 2005). Bacterial CVD is one of the most important causes of NCC among vaginosis is also associated with higher risk of people with HIV for several reasons. First, CVD, specifi- transmission of HIV to a male partner (Cohen and cally ischemic heart disease, was the number one cause of others 2012). death, years of life lost, and disability-adjusted life years In addition, bacterial vaginosis has been associ- in 2010 in the United States, and the number one cause ated with increased risk of acquiring other STIs, with of disability-adjusted life years globally (Murray and developing pelvic inflammatory disease, with several Lopez 2013). CVD and its risk factors are increasing adverse outcomes of pregnancy (for example, fetal loss, in many LMICs as an emerging epidemic, even among spontaneous abortion, stillbirth, preterm delivery, low those without HIV (Mensah 2008). birth weight, and disease in the offspring), and with Second, rates of CVD, particularly myocardial infarc- infertility. A large and growing number of possible tion, are much higher among people with HIV than mechanisms by which any vaginal dysbiosis may con- among those without, most likely due in part to chronic tribute to these complications have been identified inflammation and immune activation (Freiberg and (Brotman 2011; Hillier and others 1995). others 2013; Silverberg and others 2014; Triant 2014). The clinical classification of myocardial infarction was Sexually Transmitted Intestinal and Enteric Infections divided into five types in 2007 as part of the universal A study early in the HIV epidemic identified diverse myocardial infarction definition (Thygesen and others pathogens, often of a polymicrobial nature, associated 2007); type 1 and type 2 constitute almost the entirety. with proctitis, proctocolitis, and enteritis (Quinn and Type 1 or primary myocardial infarction events result others 1983). Sexually associated proctitis, proctocolitis, spontaneously from atherosclerotic plaque instability. or enteritis, which occur more commonly in MSM, but Type 2 myocardial infarction events are secondary events also in heterosexual women and men through unpro- due to other illnesses or causes resulting in myocardial tected anal intercourse, may increase susceptibility ischemia from increased oxygen demand or decreased to HIV (Fleming and Wasserheit 1999). Such infec- supply, as can occur in the setting of hypotension or tions are occurring with greater frequency (Cone and hypoxia. Little is known about myocardial infarction Whitlow 2013). types in people with HIV in either HICs or LMICs. However, in the United States, type 2 myocardial infarc- tion events make up close to half of all myocardial HIV and Women’s Reproductive Health Conclusions infarction events among people with HIV, a much higher HIV and other STIs have been associated with major proportion than in the general population. Understanding direct and indirect harm to the reproductive health of myocardial infarction types among people with HIV women in many settings and populations. may help clarify unanswered questions regarding risk 52 Major Infectious Diseases factors and higher prevalence among people with HIV didanosine use (Sabin and others 2008), although both (Crane and others 2014). are controversial (Cutrell and others 2008). Third, the effect of CVD among people with HIV The Strategies for Management of Antiretroviral may be even more profound in LMICs than in HICs, Therapy (SMART) study found a trend toward increased given that most people with HIV live in LMICs and CVD with delayed ART or ART interruptions (El-Sadr may have additional CVD risk factors unique to these and others 2006), suggesting that despite potential meta- areas (Bloomfield and others 2014). In a comparison of bolic impacts of ART, the overall effect is protective. This NCCs among two cohort studies of people with HIV on finding is several years old; in most regions, current regi- ART, one from Botswana and one from Tennessee, event mens now have even lower negative metabolic impacts. rates were higher in Botswana in comparisons stan- Several studies have found that low recent and nadir dardized to the U.S. population; the largest discrepan- CD4 T-cell counts are associated with CVD (Drozd and cies were for CVD disease (Wester and others 2011). As others 2014; Lang and others 2012; Silverberg and others a result, the major public health impact of CVD among 2014). Elevated or detectable HIV RNA levels (viremia) people with HIV has been increasingly recognized in have also been associated with CVD (Drozd and others HICs and LMICs (Bloomfield and others 2014; Currier 2014; Freiberg and others 2013; Lang and others 2012; and others 2003). Silverberg and others 2014). These studies, as well as findings from the SMART trial, support the hypothesis Factors Associated with CVD among People with HIV that CVD risk is reduced by control of HIV itself, and Causes of CVD among people with HIV are multifacto- they provide additional support for recommendations rial and include long-term HIV exposure, consequences for earlier ART initiation (Silverberg and others 2014; of ongoing inflammation, progressive immune dysfunc- Triant 2014). However, overall risk and relative contribu- tion, and possible adverse effects associated with ART tions of specific risk factors require more study in HICs (Aberg 2009). These factors are compounded by the aging and LMICs, including assessment of the effects of inter- of the population of people with HIV in many regions ventions that address these risk factors. and by higher rates of traditional CVD risk factors such as smoking, diabetes, and dyslipidemia (Silverberg and Other CVD Outcomes others 2014; Triant 2014). Although much of the CVD and HIV literature has focused While modifying traditional CVD risk factors may be on myocardial infarction, this is not the only CVD out- important in preventing CVD, these factors have not come of relevance to people with HIV. In the United States, been shown to explain the entire CVD risk increase people with HIV have 1.8 times the risk of heart failure among people with HIV, highlighting the importance of (Butt and others 2011); a meta-analysis suggested a preva- novel and HIV-specific factors influencing CVD risk lence of 8 percent for systolic dysfunction and 43 percent (Triant 2014). HIV may also accelerate CVD through for diastolic dysfunction (Cerrato and others 2013). chronic inflammation. The increase in myocardial infarc- Data on cardiac dysfunction are more limited in other tion risk in people with HIV is similar to that in inflam- parts of the world; several studies present such data, but matory diseases like rheumatoid arthritis (Solomon and they precede the widespread availability of ART. In others 2003). Inflammatory and coagulation markers Zimbabwe, 50 percent of people with HIV had cardiac have been shown to predict CVD (Ford and others 2010; dysfunction, including 22 percent with left-ventricular Triant, Meigs, and Grinspoon 2009), but the most appro- dysfunction, 6 percent with isolated right-ventricular priate intervention targets remain unclear. dilation, and 9 percent with dilated cardiomyopathy (Hakim, Matenga, and Siziya 1996). A Rwanda study Impact of ART on CVD Risk showed dilated cardiomyopathy in 18 percent of people ART has greatly reduced morbidity and mortality with HIV who were not receiving ART (Twagirumukiza (Palella and others 1998); paradoxically, however, ART and others 2007). may theoretically increase CVD risk (Friis-Moller and More recently, in South Africa, cardiomyopathy was the others 2007) because of altered metabolism or athero- most common cardiac disease manifestation (38 percent) genic effects (Behrens and others 1999; Holmberg, among people with HIV with newly diagnosed heart Moorman, and Greenberg 2004; Holmberg and others disease (Sliwa and others 2012). These findings suggest 2002). CVD risk may be higher with longer ART or pro- that heart failure and cardiomyopathy are important com- tease inhibitor duration (Currier and others 2003; DAD plications among people with HIV. However, the data are Study Investigators 2004; Friis-Moller and others 2007; insufficient to address whether these outcomes are more Holmberg and others 2002), or with recent abacavir or common among people with HIV in LMICs than in HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 53 HICs (Bloomfield and others 2014), how accurate these than among ART-naive individuals (Jantarapakde and estimates might be in the current treatment era, and how others 2014). these outcomes may have improved in LMICs with more In Uganda, people with HIV had slightly higher mean widely available ART or in HICs with earlier initiation glucose levels over time, as evidenced by elevations in of ART. HbA1c measurements, compared with those without Factors associated with CVD outcomes, such as HIV (Dillon and others 2013). heart failure, among people with HIV also include The incidence of diabetes mellitus in people with traditional risk factors such as smoking, prior myocar- HIV was higher in the earlier years of the ART era than dial infarction, hypertension, and higher age (Cerrato in recent years (Capeau and others 2012). The decline in and others 2013). Other factors possibly associated diabetes mellitus incidence during the ART era is likely with cardiomyopathy among patients in Sub-Saharan because ART medications, such as didanosine, were Africa not receiving ART include nutritional factors, more often used earlier and had greater negative meta- low CD4 T-cell counts, higher viral load, and advanced bolic impacts. These more toxic agents are used less HIV stage (Nzuobontane, Blackett, and Kuaban 2002; often today, suggesting the global incidence of diabetes Twagirumukiza and others 2007). mellitus among people with HIV may continue to decline. However, other factors may serve to increase the risk of diabetes mellitus in the contemporary ART era, Type 2 Diabetes Mellitus especially the increasing burden of diabetes in countries Burden and Epidemiology of Diabetes Mellitus in moving from low- to middle-income status. People with HIV In North America and Europe, people with HIV more Factors Associated with Diabetes Mellitus among often have glucose abnormalities and type 2 diabetes People with HIV mellitus than do people without HIV (Brown and others Factors that may predispose people with HIV to diabetes 2005; Guaraldi and others 2011). In analyses adjusted for mellitus are often similar to those of the general popula- age and body mass index, men with HIV who were not tion, including older age (Butt and others 2004; Capeau receiving ART had 2.2 times the prevalence of diabetes and others 2012; Hughes and others 2005), obesity mellitus than men without HIV (Brown and others (Hughes and others 2005), and racial or ethnic minority 2005). The prevalence was more than four times higher group (Butt and others 2004; Hughes and others 2005). among men with HIV who were receiving ART than In many parts of the world, including the United States, among uninfected men (Brown and others 2005). increasing obesity and sedentary lifestyles will contribute Estimated prevalence rates among people with HIV have to increasing diabetes mellitus incidence among people varied based on the patient population, from 3 percent with and without HIV (Samaras 2009). to 21 percent (Hadigan and others 2001; Salehian and Hepatitis C virus is another established risk factor for others 2005; Visnegarwala and others 2005). Diabetes diabetes mellitus in the general population (Fallahi and mellitus prevalence and impaired glucose tolerance rates others 2013). Hepatitis C virus has a higher prevalence in are higher among those older than age 60 years (Guaraldi many populations of people with HIV; as such, it may be and others 2011), ranging from 21 percent to 66 percent a more important risk factor in individuals co-infected (Arama and others 2013; Araujo and others 2014; with HIV and hepatitis C virus (Butt and others 2004; Hadigan and others 2001). Visnegarwala and others 2005). More is known about rates of diabetes mellitus and HIV-specific risk factors for diabetes mellitus include impaired glucose tolerance in HICs than in LICs. The the use of protease inhibitors, particularly indinavir estimated prevalence of diabetes mellitus among people (Capeau and others 2012) or ritonavir (Brown and with HIV is higher in HICs than in LMICs; however, others 2005), and longer duration of exposure to NRTIs the estimated absolute number of people with HIV (Tien and others 2007), particularly didanosine and with diabetes mellitus is greater in LMICs (Ali and stavudine (Capeau and others 2012). However, a meta- others 2014). A study of countries in South America analysis of Sub-Saharan African studies found that estimated diabetes mellitus prevalence rates to be from among people with HIV, ART use was associated with a 0.8 percent in Columbia to 6.5 percent in Brazil (Cahn lower HbA1c value (Dillon and others 2013). People with and others 2010). HIV with a lower CD4 T-cell nadir (< 300 cells/mm3) Although the data on people with HIV in Asia are have a higher incidence of abnormal glucose metabolism particularly limited, a study in Thailand found that than those with a higher CD4 T-cell nadir (Brown and among 580 people with HIV, 4.7 percent had hypergly- others 2005). This finding suggests that current treat- cemia; this rate was higher among ART-experienced ment recommendations that include starting people with 54 Major Infectious Diseases HIV on ART at higher CD4 T-cell counts may contribute South Africa has one of the highest prevalence rates of to a decrease in diabetes mellitus incidence among peo- HIV, but the prevalence and effect of dyslipidemia is ple with HIV. less clear. Black Africans often exhibit lower fasting triglyc- eride, total cholesterol, and LDL levels, and higher HDL levels, than white Africans (Seedat 1999); however, in 300 Dyslipidemia black individuals in South Africa newly diagnosed with Burden and Epidemiology of Dyslipidemia in HIV HIV, HDL levels were found to be lower than levels nor- Dyslipidemia is an important NCC, given both the high mally associated with increased CVD risk (Fourie and prevalence among people with HIV and impact of dys- others 2010). A cross-sectional study of 580 people with lipidemia on CVD risk (Giannarelli, Klein, and Badimon HIV in Thailand found that 41 percent had triglyceride 2011). The condition is common among people with levels greater than 150 milligrams per deciliter (mg/dL), HIV who are untreated and is the most common meta- 40 percent had total cholesterol values greater than bolic abnormality associated with ART (Friis-Moller and 200 mg/dL, and 12 percent had LDL values greater than others 2003). 160 mg/dL (Jantarapakde and others 2014). A study of HIV infection itself, before initiating effective ART, 129 people with HIV from Thailand who had survived has been associated with changes in lipids, including with HIV more than 10 years found that more than reductions in total cholesterol, low-density lipoprotein 50 percent had lipid abnormalities (Kiertiburanakul, cholesterol (LDL), and high-density lipoprotein choles- Luengroongroj, and Sungkanuparph 2012). terol (HDL), as well as increases in triglyceride values, particularly among those with more advanced HIV dis- Factors Associated with Dyslipidemia among People ease (Riddler and others 2003). with HIV Among people with HIV on ART, HIV-associated dys- Dyslipidemia among people with HIV is associated with lipidemia may include decreased HDL, increased LDL, a number of factors, such as gender, older age, race, and increased non-HDL cholesterol, and hypertriglyceri- CD4 T-cell count (Crane and others 2011). However, demia. Some of the lipid changes after ART initiation are ART is likely to be one of the most important factors in due to a return to health with HIV treatment (Liu and dyslipidemia risk among people with HIV. Dyslipidemia others 2013). One large cross-sectional study found ele- is associated with most protease inhibitors; among the vated total cholesterol levels in 27 percent of people with protease inhibitors, darunavir and atazanavir have been HIV receiving protease inhibitors and 23 percent receiving found to have better lipid profiles than older protease non-nucleoside reverse transcriptase inhibitors (NNRTI), inhibitors (Carey and others 2010; Mills and others 2009) compared with 8 percent of ART-naive people with HIV (table 3.1). (Friis-Moller and others 2003). Cohort studies have sug- The NNRTIs tend to have a smaller impact on lipid gested that the effect of ART on lipids may be greatest in levels than the protease inhibitors, except for atazanavir the first six months after ART initiation (Papadopoulos and darunavir (Daar and others 2011). Etravirine and and others 2012). rilpivirine, in particular, are NNRTIs with less impact A large cross-sectional study of people with HIV in New York City, most of whom were receiving ART, found that the prevalence of elevated LDL was 37 percent in Table 3.1 Antiretroviral Medications Associated with men and 31 percent in women (Myerson and others Little or Less Negative Impact on Lipid Levels 2014). Furthermore, 30 percent of black people with HIV, 40 percent of Hispanic people with HIV, and 37 percent of Class Agents white people with HIV had elevated LDL levels (Myerson Protease inhibitors Atazanavir and others 2014). Women with HIV have lower HDL and higher triglyceride values compared with women without Darunavir HIV (Schwartz and others 2014). NNRTI Etravirine Although less is known about dyslipidemia in people Rilpivirine with HIV in LMICs than in HICs, it remains one of the Integrase inhibitors Raltegravir better investigated NCCs in LMICs. Cross-sectional stud- Dolutegravir ies from Cameroon and Tanzania of people with HIV not receiving ART found low HDL levels and elevated Elvitegravir triglyceride levels; high triglyceride levels were particularly NRTI Tenofovir associated with advanced stages of immunodeficiency Note: NNRTI = non-nucleoside reverse transcriptase inhibitors; NRTI = nucleoside (Armstrong and others 2011; Nguemaim and others 2010). reverse transcriptase inhibitors. HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 55 on lipid levels than other NNRTIs (Cohen and others much more notable, with half of the patients having 2013; Fatkenheuer and others 2012) (see table 3.1). elevated triglyceride levels after ART (Pinto Neto and Integrase inhibitors are increasingly important compo- others 2013). nents of both follow-up or salvage regimens (Capetti In Cameroon and Ethiopia, lipid levels were higher and others 2014) and naive regimens (Gunthard and among people with HIV on NNRTI-based ART compared others 2014). Their use will likely continue to increase, with those not on therapy (Abebe and others 2014; Pefura particularly in regions with access to the once-daily Yone and others 2011). Similarly, Nigerian people with integrase inhibitor dolutegravir and the single daily pill HIV on ART had higher total cholesterol levels and lower coformulation of dolutegravir with the NRTIs abacavir HDL levels than those who were ART naive (Muhammad, and lamivudine. Sani, and Okeahialam 2013). An important advantage of the integrase inhibitors Among Sub-Saharan African women randomized is their minimal impact on lipid levels. In ART-naive to regimens containing nelfinavir versus lopinavir/ people with HIV initiating a regimen with raltegravir ritonavir, larger increases in lipid levels except HDL or efavirenz, those on raltegravir had smaller increases were found over time among those on lopinavir/ in total cholesterol, HDL, LDL, and triglycerides with ritonavir (Shaffer and others 2014). Variable increases up to five years of follow-up (Gotuzzo and others in triglyceride, LDL, and HDL values have been reported 2012; Rockstroh and others 2013). The impact of after ART initiation (Ceccato and others 2011). Among dolutegravir is likely similar to raltegravir (Raffi and Ugandan people with HIV with CD4 T-cell < 200/mm3 others 2013). or symptomatic HIV who initiated predominantly The impact of the NRTI class of agents on lipids stavudine/lamivudine/nevirapine, substantial increases is variable. Among the NRTIs, tenofovir has the least in HDL occurred during the initial 24 months of negative impact on lipid levels (see table 3.1). A study of ART, with less frequent elevations in total cholesterol, treatment-naive people with HIV who initiated ataza- LDL, and triglyceride values (Buchacz and others navir/ritonavir or efavirenz with abacavir/lamivudine or 2008). Among patients in Tanzania, one of the largest tenofovir/emtricitabine found that those on abacavir/ longitudinal studies found low HDL and high lamivudine generally had greater increases in lipid levels, triglyceride levels were common before ART initiation; including total cholesterol, LDL, and HDL levels, than after ART was initiated, unfavorable changes were more those on tenofovir/emtricitabine at 48 and 96 weeks (Sax common among those on stavudine and efavirenz and others 2011). A United States–based cohort study compared with zidovudine- and nevirapine-based found that people whose ART regimen contained teno- regimens (Liu and others 2013). fovir (in combination with lamivudine or emtricitabine) had lower concentrations of total cholesterol, LDL, non- Summary of HIV and Dyslipidemia HDL cholesterol, HDL, and triglyceride levels, compared To summarize, HIV itself is associated with changes in with other pairs of NRTIs (Crane and others 2011). lipid levels, particularly elevated triglyceride levels, most Those who received didanosine and stavudine—a notably among those with more pronounced immune combination that is no longer recommended—had the deficiency. Initiating ART is also associated with dyslipi- highest levels of LDL (Crane and others 2011). While demia, but the effect is variable, depending on the regi- stavudine leads to elevated total cholesterol and triglyc- men. Protease inhibitors as a class are associated with eride values, it is rarely used in the United States any increased lipid levels, particularly triglycerides, but longer; however, it has been more widely used in LMICs darunavir and atazanavir have less effect. (Ali and others 2014). Unfortunately, many of the newer Other classes of ART are more variable. NNRTIs raise agents with little or no impact on lipid levels are not as lipid levels, but etravirine and rilpivirine have less widely available in LMICs (Ali and others 2014). impact, as do integrase inhibitors such as raltegravir and Several studies in LMICs have examined the influence dolutegravir. Among the NRTIs, tenofovir has little of ART in general on lipid levels; fewer studies in LMICs effect, while agents such as stavudine have a substantial have focused on individual ART regimens or agents. negative impact on lipids. Data from LMICs on class- Dyslipidemia, such as low HDL levels, was more common and agent-specific impacts, particularly of newer regi- among patients with longer exposure to ART in several mens, are sparse. studies from South America (Albuquerque and others Recommendations for initial ART regimens have 2005; Cahn and others 2010). However, a Brazilian study varied over time but more recently have included regi- found that low HDL was common before ART initiation; mens less likely to cause dyslipidemia than those used a with initiation of ART, increased triglyceride levels were few years ago (Gunthard and others 2014), suggesting 56 Major Infectious Diseases that in HICs the rate of newly developed dyslipidemia availability of ART. In the era of effective therapy for may drop. However, these benefits may not apply to HIV, health professionals will need to deploy more com- other regions of the world where these regimen choices plex strategies and treatments that recognize the rising are less widely available. impact of comorbidities complicating long-term HIV infection. HIV and NCCs Conclusions NCCs among people with HIV present an increasing Cancer Prevention global health challenge in the era of increasing access to Strategies for cancer prevention, including vaccination, ART, which is leading to older populations of people screening, and early ART initiation, can reduce the cancer with HIV. burden, but additional strategies for reducing the burden Even though this chapter briefly summarizes several of HIVAMs are needed. NCCs, many serious NCCs are not addressed here because of space considerations. These include strokes, bone disease, cognitive impairment, renal disease, and Reproductive Health frailty. A very important NCC is liver disease, which is an In the areas of reproductive health beyond fertility area in flux, with increasing treatment options for hepa- issues, evidence has also suggested a greater risk of still- titis C virus (chapter 16 in this volume, Wiktor 2017). birth, premature birth, or low birth weight in pregnan- Those covered here can be considered the tip of the cies of women with HIV. Providing women with access iceberg. As access to ART continues to expand, partic- to appropriate fertility and contraception to ensure ularly in LMICs, access to care for NCCs will also be safe and planned pregnancies will be paramount in needed to avoid widespread preventable morbidity and strategies to reduce these risks associated with HIV and death. Notably, a theme common to many NCCs is the reproduction. benefit of early ART. Many NCCs are worsened by the lack of treatment or late treatment initiation, lead- ing to increased burden of illness. Focused efforts on Noncommunicable Chronic Comorbidities engagement in care, expanded access to ART, and ear- Among NCCs, ART has changed the nature of the global lier initiation of ART are likely to reduce the rates of health challenge. This chapter only brushes the surface of these NCCs, particularly in areas with access to newer a major issue, leaving out other diseases associated with ART agents with less metabolic and other associated HIV. Notably, and as with cancers, a theme common to toxicities. many NCCs is the benefit of early ART. Many of the However, as the population of people with HIV ages, diseases are worsened by lack of treatment or late treat- the burden of many NCCs will continue to increase. The ment. Better care, expanded access, and earlier initiation importance of NCCs among people with HIV as a global of ART are likely to help lower disease rates. Because health challenge cannot be underestimated. populations of people with HIV will continue to rise, Data demonstrating the importance of these condi- however, these comorbidities cannot be underestimated tions among people with HIV are predominantly from as a looming public health challenge. HICs; unfortunately, much less of the data and research The burden of NCCs will be concentrated in areas of advances for these conditions are from LMICs, where the world, such as Sub-Saharan Africa, where the number most people with HIV live (Narayan and others 2014). of HIV infections continue to rise and access to life- In many areas, such as Sub-Saharan Africa, HIV care, prolonging ART is expanding. including ART, is more widely available than it has pre- Taken together, the great success in treating HIV viously been, but this care is not accompanied by care for has been met by a set of considerable challenges for NCCs, resulting in preventable morbidity and death persons with the infection and for the public health (Narayan and others 2014). community in the form of chronic noncommunicable diseases. However, improvements in the treatment of HIV infection, attention to preventive measures for CONCLUSIONS chronic disease in persons at high risk for their compli- This chapter reviews the complex association of HIV cations, and increased research and awareness offer with a host of other diseases and conditions; not least is hope that the same successes enjoyed in the fight against the rise in those that have occurred because people with HIV infection can be brought to bear to reduce the com- HIV have been living longer thanks to the widespread plications of long-term infection. HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 57 NOTE and Risk of Diabetes Mellitus in HIV-Infected Patients Receiving Current Antiretroviral Drugs.” European Journal World Bank Income Classifications as of July 2014 are as fol- of Endocrinology 171 (5): 545–54. doi:10.1530/EJE-14–0337. lows, based on estimates of gross national income (GNI) per Armstrong, C., E. Liu, J. Okuma, D. Spiegelman, C. Guerino, and capita for 2013: others. 2011. “Dyslipidemia in an HIV-Positive Antiretroviral Treatment-Naive Population in Dar es Salaam, Tanzania.” • Low-income countries (LICs) = US$1,045 or less Journal of Acquired Immune Deficiency Syndromes 57 (2): • Middle-income countries (MICs) are subdivided: 141–45. doi:10.1097/QAI.0b013e318219a3d1. (a) lower-middle-income = US$1,046 to US$4,125 Arzumanyan, A., H. M. Reis, and M. A. Feitelson. 2013. (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 “Pathogenic Mechanisms in HBV- and Hepatitis C Virus- • High-income countries (HICs) = US$12,746 or more. Associated Hepatocellular Carcinoma.” Nature Reviews Cancer 13 (2): 123–35. doi:10.1038/nrc3449. Atashili, J., C. Poole, P. M. Ndumbe, A. A. Adimora, and J. S. Smith. 2008. “Bacterial Vaginosis and HIV Acquisition: REFERENCES A Meta-Analysis of Published Studies.” AIDS 22 (12): Abebe, M., S. Kinde, G. Belay, A. Gebreegziabxier, F. Challa, 1493–501. doi:10.1097/QAD.0b013e3283021a37. and others. 2014. “Antiretroviral Treatment Associated Barnes, A., D. Riche, L. Mena, T. Sison, L. Barry, and others. 2014. Hyperglycemia and Dyslipidemia among HIV Infected “Efficacy and Safety of Intrauterine Insemination and Assisted Patients at Burayu Health Center, Addis Ababa, Ethiopia: Reproductive Technology in Populations Serodiscordant A Cross-Sectional Comparative Study.” BMC Research for Human Immunodeficiency Virus: A Systematic Review Notes 7: 380. doi:10.1186/1756–0500–7-380. and Meta-Analysis.” Fertility and Sterility 102 (2): 424–34. Aberg, J. A. 2009. “Cardiovascular Complications in HIV doi:10.1016/j.fertnstert.2014.05.001. Management: Past, Present and Future.” Journal of Acquired Behrens, G., A. Dejam, H. Schmidt, H. J. Balks, G. Brabant, and Immune Deficiency Syndromes 50 (1): 54–64. others. 1999. “Impaired Glucose Tolerance, Beta Cell Function Achenbach, C. J., S. R. Cole, M. M. Kitahata, C. Casper, and Lipid Metabolism in HIV Patients under Treatment with J. H. Willig, and others. 2011. “Mortality after Cancer Protease Inhibitors.” AIDS 13 (10): F63–70. Diagnosis in HIV-Infected Individuals Treated with Biggar, R. J., A. K. Chaturvedi, J. J. Goedert, E. A. Engels, and Antiretroviral Therapy.” AIDS 25 (5): 691–700. doi:10.1097 HIV AIDS Cancer Match Study. 2007. “AIDS-Related Cancer /QAD.0b013e3283437f77. and Severity of Immunosuppression in Persons with AIDS.” Ahdieh, L., A. Munoz, D. Vlahov, C. L. Trimble, L. A. Timpson, and Journal of the National Cancer Institute 99 (12): 962–72. K. Shah. 2000. “Cervical Neoplasia and Repeated Positivity of doi:10.1093/jnci/djm010. Human Papillomavirus Infection in Human Immunodeficiency Blitz, S., J. Baxter, J. Raboud, S. Walmsley, A. Rachlis, and others. Virus-Seropositive and -Seronegative Women.” American 2013. “Evaluation of HIV and Highly Active Antiretroviral Journal of Epidemiology 151 (12): 1148–57. Therapy on the Natural History of Human Papillomavirus Akarolo-Anthony, S. N., L. D. Maso, F. Igbinoba, S. M. Mbulaiteye, Infection and Cervical Cytopathologic Findings in HIV- and C. A. Adebamowo. 2014. “Cancer Burden among HIV- Positive and High-Risk HIV-Negative Women.” Journal of Positive Persons in Nigeria: Preliminary Findings from the Infectious Diseases 208 (3): 454–62. doi:10.1093/infdis/jit181. Nigerian AIDS-Cancer Match Study.” Infectious Agents and Bloomfield, G. S., P. Khazanie, A. Morris, C. Rabadan-Diehl, Cancer 9 (1): 1. doi:10.1186/1750–9378–9-1. L. A. Benjamin, and others. 2014.“HIV and Noncommunicable Albuquerque, E. M., E. C. de Faria, H. C. Oliveira, D. O. Magro, Cardiovascular and Pulmonary Diseases in Low- and Middle- and L. N. Castilho. 2005. “High Frequency of Fredrickson’s Income Countries in the ART Era: What We Know and Best Phenotypes IV and IIb in Brazilians Infected by Human Directions for Future Research.” Journal of Acquired Immune Immunodeficiency Virus.” BMC Infectious Diseases 5: 47. Deficiency Syndromes 67 (Suppl 1): S40–53. doi:10.1186/1471–2334–5-47. Bonnet, F., C. Lewden, T. May, L. Heripret, E. Jougla, and Ali, M. K., M. J. Magee, J. A. Dave, I. Ofotokun, M. Tungsiripat, others. 2004. “Malignancy-Related Causes of Death in and others. 2014. “HIV and Metabolic, Body, and Bone Human Immunodeficiency Virus-Infected Patients in Disorders: What We Know from Low- and Middle-Income the Era of Highly Active Antiretroviral Therapy.” Cancer Countries.” Journal of Acquired Immune Deficiency Syndromes 101 (2): 317–24. 67 (Suppl 1): S27–39. doi:10.1097/QAI.0000000000000256. Bosch, F. X., T. R. Broker, D. Forman, A. B. Moscicki, M. L. Arama, V., C. Tiliscan, A. Streinu-Cercel, D. Ion, R. Mihailescu, Gillison, and others. 2013. “Comprehensive Control of and others. 2013. “Insulin Resistance and Adipokines Serum Human Papillomavirus Infections and Related Diseases.” Levels in a Caucasian Cohort of HIV-Positive Patients Vaccine 31 (Suppl 7): H1–31. doi:10.1016/j.vaccine.2013 Undergoing Antiretroviral Therapy: A Cross Sectional .10.003. Study.” BMC Endocrine Disorders 13: 4. doi:10.1186/1472 Brotman, R. M. 2011. “Vaginal Microbiome and Sexually –6823–13–4. Transmitted Infections: An Epidemiologic Perspective.” Araujo, S., S. Banon, I. Machuca, A. Moreno, M. J. Perez-Elias, Journal of Clinical Investigation 121 (12): 4610–17. doi:10.1172 and J. L. Casado. 2014. “Prevalence of Insulin Resistance /JCI57172. 58 Major Infectious Diseases Brown, T. T., S. R. Cole, X. Li, L. A. Kingsley, F. J. Palella, and others. Journal of Antimicrobial Agents 43 (2): 189–94. doi:10.1016/j 2005. “Antiretroviral Therapy and the Prevalence and Incidence .ijantimicag.2013.10.013. of Diabetes Mellitus in the Multicenter AIDS Cohort Study.” Carbone, A., and A. Gloghini. 2005. “AIDS-Related Lymphomas: Archives of Internal Medicine 165 (10): 1179–84. From Pathogenesis to Pathology.” British Journal of Bruni, L., M. Diaz, X. Castellsague, E. Ferrer, F. X. Bosch, Haematology 130 (5): 662–70. doi:10.1111/j.1365–2141 and S. de Sanjose. 2010. “Cervical Human Papillomavirus .2005.05613.x. Prevalence in 5 Continents: Meta-Analysis of 1 Million Women Carey, D., J. Amin, M. Boyd, K. Petoumenos, and S. Emery. with Normal Cytological Findings.” Journal of Infectious 2010. “Lipid Profiles in HIV-Infected Adults Receiving Diseases 202 (12): 1789–99. doi:10.1086/657321. Atazanavir and Atazanavir/Ritonavir: Systematic Review Bruyand, M., R. Thiebaut, S. Lawson-Ayayi, P. Joly, A. J. Sasco, and Meta-Analysis of Randomized Controlled Trials.” and others. 2009. “Role of Uncontrolled HIV RNA Level Journal of Antimicrobial Chemotherapy 65 (9): 1878–88. and Immunodeficiency in the Occurrence of Malignancy doi:10.1093/jac/dkq231. in HIV-Infected Patients during the Combination CDC (Centers for Disease Control and Prevention). 1992. Antiretroviral Therapy Era: Agence Nationale de Recherche “1993 Revised Classification System for HIV Infection and sur le Sida (ANRS) CO3 Aquitaine Cohort.” Clinical Expanded Surveillance Case Definition for AIDS among Infectious Diseases 49 (7): 1109–16. doi:10.1086/605594. Adolescents and Adults.” MMWR Recommendations and Buchacz, K., P. J. Weidle, D. Moore, W. Were, J. Mermin, and Reports 41 (RR-17): 1–19. others. 2008. “Changes in Lipid Profile over 24 Months ———. 2006. “A Comprehensive Immunization Strategy to among Adults on First-Line Highly Active Antiretroviral Eliminate Transmission of Hepatitis B Virus Infection in the Therapy in the Home-Based AIDS Care Program in Rural United States.” Morbidity and Mortality Weekly Reports 55. Uganda.” Journal of Acquired Immune Deficiency Syndromes Ceccato, M. G., P. F. Bonolo, A. I. Souza Neto, F. S. Araujo, 47 (3): 304–11. and M. I. Freitas. 2011. “Antiretroviral Therapy–Associated Bukusi, E. A., C. R. Cohen, C. E. Stevens, S. Sinei, M. Reilly, and Dyslipidemia in Patients from a Reference Center in Brazil.” others. 1999. “Effects of Human Immunodeficiency Virus 1 Brazilian Journal of Medical and Biological Research 44 (11): Infection on Microbial Origins of Pelvic Inflammatory 1177–83. Disease and on Efficacy of Ambulatory Oral Therapy.” Celum, C., A. Wald, J. R. Lingappa, A. S. Magaret, R. S. Wang, American Journal of Obstetrics and Gynecology 181 (6): and others. 2010. “Acyclovir and Transmission of HIV-1 1374–81. from Persons Infected with HIV-1 and HSV-2.” New Butt, A. A., C. C. Chang, L. Kuller, M. B. Goetz, D. Leaf, and others. England Journal of Medicine 62 (5): 427–39. doi:10.1056 2011. “Risk of Heart Failure with Human Immunodeficiency /NEJMoa0904849. Virus without Prior Diagnosis of Coronary Heart Disease.” Cerrato, E., F. D’Ascenzo, G. Biondi-Zoccai, A. Calcagno, Archives of Internal Medicine 171 (8): 737–43. doi:10.1001 S. Frea, and others. 2013. “Cardiac Dysfunction in Pauci /archinternmed.2011.151. Symptomatic Human Immunodeficiency Virus Patients: A Butt, A. A., S. L. Fultz, C. K. Kwoh, D. Kelley, M. Skanderson, Meta-Analysis in the Highly Active Antiretroviral Therapy and others. 2004. “Risk of Diabetes in HIV Infected Veterans Era.” European Heart Journal 34 (19): 1432–36. doi:10.1093 Pre- and Post-HAART and the Role of Hepatitis C Virus /eurheartj/ehs471. Coinfection.” Hepatology 40 (1): 115–19. Chang, M. H., C. J. Chen, M. S. Lai, H. M. Hsu, T. C. Wu, Cahn, P., O. Leite, A. Rosales, R. Cabello, C. A. Alvarez, and and others. 1997. “Universal Hepatitis B Vaccination in others. 2010. “Metabolic Profile and Cardiovascular Risk Taiwan and the Incidence of Hepatocellular Carcinoma in Factors among Latin American HIV-Infected Patients Children. Taiwan Childhood Hepatoma Study Group.” New Receiving HAART.” Brazilian Journal of Infectious Diseases England Journal of Medicine 336 (26): 1855–59. doi:10.1056 14 (2): 158–66. doi:S1413–86702010000200008 [pii]. /nejm199706263362602. Campos, N. G., P. E. Castle, T. C. Wright Jr., and J. J. Kim. 2015. Chang, M. H., S. L. You, C. J. Chen, C. J. Liu, C. M. Lee, and “Cervical Cancer Screening in Low-Resource Settings: others. 2009. “Decreased Incidence of Hepatocellular A Cost-Effectiveness Framework for Valuing Tradeoffs Carcinoma in Hepatitis B Vaccinees: A 20-Year Follow-Up between Test Performance and Program Coverage.” Study.” Journal of the National Cancer Institute 101 (19): International Journal of Cancer 137 (9): 2208–19. doi:10.1002 1348–55. doi:10.1093/jnci/djp288. /ijc.29594. Chiao, E. Y., T. P. Giordano, J. M. Palefsky, S. Tyring, and H. El Capeau, J., V. Bouteloup, C. Katlama, J. P. Bastard, V. Guiyedi, Serag. 2006. “Screening HIV-Infected Individuals for Anal and others. 2012. “Ten-Year Diabetes Incidence in Cancer Precursor Lesions: A Systematic Review.” Clinical 1046 HIV-Infected Patients Started on a Combination Infectious Diseases 43 (2): 223–33. doi:10.1086/505219. Antiretroviral Treatment.” AIDS 26 (3): 303–14. doi:10.1097 Chung, M. H., K. P. McKenzie, H. De Vuyst, B. A. Richardson, /QAD.0b013e32834e8776. F. Rana, and others. 2013. “Comparing Papanicolau Smear, Capetti, A., P. Meraviglia, S. Landonio, G. Sterrantino, A. Di Visual Inspection with Acetic Acid and Human Papillomavirus Biagio, and others. 2014. “Four Years Data of Raltegravir- Cervical Cancer Screening Methods among HIV-Positive Based Salvage Therapy in HIV-1-Infected, Treatment- Women by Immune Status and Antiretroviral Therapy.” AIDS Experienced Patients: The SALIR-E Study.” International 27 (18): 2909–19. doi:10.1097/01.aids.0000432472.92120.1b. HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 59 Clifford, G. M., M. Lise, S. Francesshi, M. Egger, C. Bouchardy, DAD (Diabetes and Depression) Study Investigators. 2004. and others. 2012. “Lung Cancer in the Swiss HIV Cohort “Cardio- and Cerebrovascular Events in HIV-Infected Study: Role of Smoking, Immunodeficiency, and Persons.” AIDS 18 (13): 1811–17. Pulmonary Infection.” British Journal of Cancer 106 (3): Dalla Pria, A., M. Alfa-Wali, P. Fox, P. Holmes, J. Weir, and 447–52. others. 2014. “High-Resolution Anoscopy Screening of HIV- Coghill, A. E., P. A. Newcomb, M. M. Madeleine, B. A. Richardson, Positive MSM: Longitudinal Results from a Pilot Study.” I. Mutyaba, and others. 2013. “Contribution of HIV Infection AIDS 28 (6): 861–67. doi:10.1097/QAD.0000000000000160. to Mortality among Cancer Patients in Uganda.” AIDS 27 Depuydt, C. E., L. Verstraete, M. Berth, J. Beert, J. P. Bogers, and (18): 2933–42. doi:10.1097/01.aids.0000433236.55937.cb. others. 2016. “Human Papillomavirus Positivity in Women Coghill, A. E., M. S. Shiels, G. Suneja, and E. A. Engels. 2015. Undergoing Intrauterine Insemination Has a Negative Effect “Elevated Cancer-Specific Mortality among HIV-Infected on Pregnancy Rates.” Gynecology and Obstetric Investigation Patients in the United States.” Journal of Clinical Oncology 81 (1): 41–46. doi:10.1159/000434749. 3 (21): 2376–83. doi:10.1200/JCO.2014.59.5967. De Vuyst, H., L. Alemany, C. Lacey, C. J. Chibwesha, Cohen, C. J., J. M. Molina, I. Cassetti, P. Chetchotisakd, V. Sahasrabuddhe, and others. 2013. “The Burden of Human A. Lazzarin, and others. 2013. “Week 96 Efficacy and Safety Papillomavirus Infections and Related Diseases in Sub- of Rilpivirine in Treatment-Naive, HIV-1 Patients in Saharan Africa.” Vaccine 31 (Suppl 5): F32–46. doi:10.1016/j Two Phase III Randomized Trials.” AIDS 27 (6): 939–50. .vaccine.2012.07.092. doi:10.1097/QAD.0b013e32835cee6e. Dillon, D. G., D. Gurdasani, J. Riha, K. Ekoru, G. Asiki, and others. Cohen, C. R., J. R. Lingappa, J. M. Baeten, M. O. Ngayo, 2013. “Association of HIV and ART with Cardiometabolic C. A. Spiegel, and others. 2012. “Bacterial Vaginosis Traits in Sub-Saharan Africa: A Systematic Review and Meta- Associated with Increased Risk of Female-to-Male HIV-1 Analysis.” International Journal of Epidemiology 42 (6): 1754– Transmission: A Prospective Cohort Analysis among 71. doi:10.1093/ije/dyt198. African Couples.” PLoS Medicine 9 (6): e1001251. Dobson, S. R., S. McNeil, M. Dionne, M. Dawar, G. Ogilvie, doi:10.1371/journal.pmed.1001251. and others. 2013. “Immunogenicity of 2 Doses of HPV Cohen, C. R., S. Sinei, M. Reilly, E. Bukusi, D. Eschenbach, and Vaccine in Younger Adolescents vs. 3 Doses in Young others. 1998. “Effect of Human Immunodeficiency Virus Women: A Randomized Clinical Trial.” Journal of the Type 1 Infection upon Acute Salpingitis: A Laparoscopic American Medical Association 309 (17): 1793–802. Study.” Journal of Infectious Diseases 178 (5): 1352–58. doi:10.1001/jama.2013.1625. Cone, M. M., and C. B. Whitlow. 2013. “Sexually Transmitted Drozd, D. R., R. M. Nance, J. A. C. Delaney, G. A. Burkholder, and Anorectal Infectious Diseases.” Gastroenterology Clinics W. C. Mathews, and others. 2014. “Lower CD4 T-Cell Count of North America 42 (4): 877–92. doi:10.1016/j.gtc.2013 and Higher Viral Load Are Associated with Increased Risk .09.003. of Myocardial Infarction.” Conference on Retroviruses and Crane, H. M., C. Grunfeld, J. H. Willig, M. J. Mugavero, S. Van Opportunistic Infections (CROI), Boston, MA, March 3–6. Rompaey, and others. 2011. “Impact of NRTIs on Lipid El-Sadr, W. M., J. D. Lundgren, J. D. Neaton, F. Gordin, Levels among a Large HIV-Infected Cohort Initiating D. Abrams, and others. 2006. “CD4 T-cell+ Count-Guided Antiretroviral Therapy in Clinical Care.” AIDS 25 (2): Interruption of Antiretroviral Treatment.” New England 185–95. doi:10.1097/QAD.0b013e328341f925. Journal of Medicine 355 (22): 2283–96. Crane, H. M., S. R. Heckbert, D. R. Drozd, M. J. Budoff, J. A. C. Eltom, M. A., A. Jemal, S. M. Mbulaiteye, S. S. Devesa, and Delaney, and others. 2014. “Lessons Learned from the R. J. Biggar. 2002. “Trends in Kaposi’s Sarcoma and Non- Design and Implementation of Myocardial Infarction Hodgkin’s Lymphoma Incidence in the United States from Adjudication Tailored for HIV Clinical Cohorts.” American 1973 through 1998.” Journal of the National Cancer Institute Journal of Epidemiology 179 (8): 996–1005. 94 (16): 1204–10. Crum-Cianflone, N. F., K. H. Hullsiek, V. C. Marconi, A. Ganesan, Engels, E. A., M. V. Brock, J. Chen, C. M. Hooker, M. Gillison, and A. Weintrob, and others. 2010. “Anal Cancers among HIV- others. 2006. “Elevated Incidence of Lung Cancer among Infected Persons: HAART Is Not Slowing Rising Incidence.” HIV-Infected Individuals.” Journal of Clinical Oncology 24 (9): AIDS 24 (4): 535–43. 1383–88. doi:10.1200/JCO.2005.03.4413. Currier, J. S., A. Taylor, F. Boyd, C. M. Dezii, H. Kawabata, and Fallahi, P., S. M. Ferrari, M. Colaci, I. Ruffilli, R. Vita, and others. others. 2003. “Coronary Heart Disease in HIV-Infected 2013. “Hepatitis C Virus Infection and Type 2 Diabetes.” Individuals.” Journal of Acquired Immune Deficiency Syndromes Clinical Therapeutics 164 (5): e393–404. doi:10.7417/CT 33 (4): 506–12. .2013.1620. Cutrell, A., C. Brothers, J. Yeo, J. Hernandez, and D. Lapierre. Fatkenheuer, G., C. Duvivier, A. Rieger, J. Durant, D. Rey, and 2008. “Abacavir and the Potential Risk of Myocardial others. 2012. “Lipid Profiles for Etravirine versus Efavirenz Infarction.” The Lancet 371 (9622): 1413. in Treatment-Naive Patients in the Randomized, Double- Daar, E. S., C. Tierney, M. A. Fischl, P. E. Sax, K. Mollan, and Blind SENSE Trial.” Journal of Antimicrobial Chemotherapy others. 2011. “Atazanavir Plus Ritonavir or Efavirenz as 67 (3): 685–90. doi:10.1093/jac/dkr533. Part of a 3-Drug Regimen for Initial Treatment of HIV-1.” Ferlay, J., I. Soerjomataram, M. Ervik, R. Dikshit, S. Eser, and Annals of Internal Medicine 154 (7): 445–56. doi:10.7326/0003 others. 2013. “GLOBOCAN 2012 v1.0; Cancer Incidence –4819–154–7-201104050–00316. and Mortality Worldwide.” IARC CancerBase 11. 60 Major Infectious Diseases Fleming, D. T., and J. N. Wasserheit. 1999. “From Immune Deficiency Syndromes 61 (1): 73–77. doi:10.1097 Epidemiological Synergy to Public Health Policy and /QAI.0b013e318263277e. Practice: The Contribution of Other Sexually Transmitted Grulich, A. E., M. T. van Leeuwen, M. O. Falster, and C. M. Diseases to Sexual Transmission of HIV Infection.” Sexually Vajdic. 2007. “Incidence of Cancers in People with HIV/ Transmitted Infections 75 (1): 3–17. AIDS Compared with Immunosuppressed Transplant Ford, E. S., J. H. Greenwald, A. G. Richterman, A. Rupert, and Recipients: A Meta-Analysis.” The Lancet 370 (9581): 59–67. others. 2010. “Traditional Risk Factors and D-Dimer doi:10.1016/s0140–6736(07)61050–2. Predict Incident Cardiovascular Disease Events in Chronic Guaraldi, G., G. Orlando, S. Zona, M. Menozzi, F. Carli, and HIV Infection.” AIDS 24 (10): 1509–17. doi:10.1097/QAD others. 2011. “Premature Age-Related Comorbidities .0b013e32833ad914. among HIV-Infected Persons Compared with the General Forman, D., C. de Martel, C. J. Lacey, I. Soerjomataram, Population.” Clinical Infectious Diseases 53 (11): 1120–6. J. Lortet-Tieulent, and others. 2012. “Global Burden of doi:10.1093/cid/cir627. Human Papillomavirus and Related Diseases.” Vaccine 30 Guiguet, M., F. Boue, J. Cadranel, J. M. Lang, E. Rosenthal, and (Suppl 5): F12–23. doi:10.1016/j.vaccine.2012.07.055. others. 2009. “Effect of Immunodeficiency, HIV Viral Load, Fourie, C. M., J. M. Van Rooyen, A. Kruger, and A. E. Schutte. and Antiretroviral Therapy on the Risk of Individual 2010. “Lipid Abnormalities in a Never-Treated HIV-1 Malignancies (FHDH-ANRS CO4): A Prospective Cohort Subtype C-Infected African Population.” Lipids 45 (1): Study.” The Lancet Oncology 10 (12): 1152–59. doi:10.1016 73–80. doi:10.1007/s11745–009–3369–4. /s1470–2045(09)70282–7. Freiberg, M. S., C. C. Chang, L. H. Kuller, M. Skanderson, Gunthard, H. F., J. A. Aberg, J. J. Eron, J. F. Hoy, A. Telenti, and E. Lowy, and others. 2013. “HIV Infection and the Risk of others. 2014. “Antiretroviral Treatment of Adult HIV Acute Myocardial Infarction.” JAMA Internal Medicine Infection: 2014 Recommendations of the International 173 (8): 614–22. doi:1659742[pii]10.1001/jamainternmed Antiviral Society-USA Panel.” Journal of the American .2013.3728. Medical Association 312 (4): 410–25. doi:10.1001/jama Friis-Moller, N., P. Reiss, C. A. Sabin, R. Weber, A. Monforte, .2014.8722. and others. 2007. “Class of Antiretroviral Drugs and the Hadigan, C., J. B. Meigs, C. Corcoran, P. Rietschel, S. Piecuch, and Risk of Myocardial Infarction.” New England Journal of others 2001. “Metabolic Abnormalities and Cardiovascular Medicine 356 (17): 1723–35. Disease Risk Factors in Adults with Human Immunodeficiency Friis-Moller, N., R. Weber, P. Reiss, R. Thiebaut, O. Kirk, and Virus Infection and Lipodystrophy.” Clinical Infectious others. 2003. “Cardiovascular Disease Risk Factors in HIV Diseases 32 (1): 130–39. doi:CID000838 [pii]10.1086/317541. Patients—Association with Antiretroviral Therapy: Results Hakim, J. G., J. A. Matenga, and S. Siziya. 1996. “Myocardial from the DAD Study.” AIDS 17 (8): 1179–93. Dysfunction in Human Immunodeficiency Virus Infection: Fultz, S. L., A. C. Justice, A. A. Butt, L. Rabeneck, S. Weissman, An Echocardiographic Study of 157 Patients in Hospital in and others. 2003. “Testing, Referral, and Treatment Patterns Zimbabwe.” Heart 76 (2): 161–65. for Hepatitis C Virus Coinfection in a Cohort of Veterans Heffron, R., D. Donnell, H. Rees, C. Celum, N. Mugo, and with Human Immunodeficiency Virus Infection.” Clinical others. 2012. “Use of Hormonal Contraceptives and Risk of Infectious Diseases 36 (8): 1039–46. doi:10.1086/374049. HIV-1 Transmission: A Prospective Cohort Study.” The Gantt, S., C. Casper, and R. F. Ambinder. 2013. “Insights into the Lancet Infectious Diseases 12 (1): 19–26. doi:10.1016/S1473 Broad Cellular Effects of Nelfinavir and the HIV Protease –3099(11)70247-X. Inhibitors Supporting Their Role in Cancer Treatment and Heffron, R., E. Were, C. Celum, N. Mugo, K. Ngure, and others. Prevention.” Current Opinions in Oncology 25 (5): 495–502. 2010. “A Prospective Study of Contraceptive Use among doi:10.1097/CCO.0b013e328363dfee. African Women in HIV-1 Serodiscordant Partnerships.” Garolla, A., B. Engl, D. Pizzol, M. Ghezzi, A. Bertoldo, and Sexually Transmitted Diseases 37 (10): 621–28. doi:10.1097 others. 2016. “Spontaneous Fertility and In Vitro Fertilization /OLQ.0b013e3181e1a162. Outcome: New Evidence of Human Papillomavirus Sperm Heuft, M. M., S. M. Houba, G. E. van den Berk, T. Smissaert van Infection.” Fertility and Sterility 105 (1): 65–72 e1. doi:10.1016 de Haere, and others. 2014. “Protective Effect of Hepatitis B /j.fertnstert.2015.09.018. Virus–Active Antiretroviral Therapy against Primary Hepatitis Giannarelli, C., R. S. Klein, and J. J. Badimon. 2011. B Virus Infection.” AIDS 28 (7): 999–1005. doi:10.1097/QAD “Cardiovascular Implications of HIV-Induced Dyslipidemia.” .0000000000000180. Atherosclerosis 219 (2): 384–89. doi:10.1016/j.atherosclerosis Hillier, S. L., R. P. Nugent, D. A. Eschenbach, M. A. Krohn, .2011.06.003. R. S. Gibbs, and others. 1995. “Association between Bacterial Gondos, A., H. Brenner, H. Wabinga, and D. M. Parkin. 2005. Vaginosis and Preterm Delivery of a Low-Birth-Weight “Cancer Survival in Kampala, Uganda.” British Journal of Infant. The Vaginal Infections and Prematurity Study Cancer 92 (9): 1808–12. Group.” New England Journal of Medicine 333 (26): 1737–42. Gotuzzo, E., M. Markowitz, W. Ratanasuwan, G. Smith, G. Prada, doi:10.1056/NEJM199512283332604. and others. 2012. “Sustained Efficacy and Safety of Raltegravir Holmberg, S. D., A. C. Moorman, and A. E. Greenberg. 2004. after 5 Years of Combination Antiretroviral Therapy as Initial “Trends in Rates of Myocardial Infarction among Patients Treatment of HIV-1 Infection: Final Results of a Randomized, with HIV.” New England Journal of Medicine 350 (7): Controlled, Phase II Study (Protocol 004).” Journal of Acquired 730–32. HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 61 Holmberg, S. D., A. C. Moorman, J. M. Williamson, T. C. Tong, Therapy: Results from the START Trial (INSIGHT 001).” 15th D. J. Ward, and others. 2002. “Protease Inhibitors and International Conference on Malignancies in AIDS and Other Cardiovascular Outcomes in Patients with HIV-1.” The Acquired Immunodeficiencies, Bethesda, MD, October 26. Lancet 360 (9347): 1747–48. Kowalkowski, M., M. Mims, R. Day, X. Du, W. Chan, and E Chiao. Hsu, H. M., D. S. Chen, C. H. Chuang, J. C. Lu, D. M. Jwo, and 2014. “Longer Duration of Combination Antiretroviral others. 1988. “Efficacy of a Mass Hepatitis B Vaccination Therapy Reduces the Risk of Hodgkin Lymphoma: A Cohort Program in Taiwan. Studies on 3464 Infants of Hepatitis B Study of HIV-Infected Male Veterans.” Cancer Epidemiology Surface Antigen-Carrier Mothers.” Journal of the American 38 (4): 386–92. Medical Association 260 (15): 2231–35. Kowalska, J. D., J. Reekie, A. Mocroft, P. Reiss, B. Ledergerber, Hughes, C. A., R. P. Cashin, D. T. Eurich, and S. Houston. 2005. and others. 2012. “Long-Term Exposure to Combination “Risk Factors for New-Onset Diabetes Mellitus in Patients Antiretroviral Therapy and Risk of Death from Specific Receiving Protease Inhibitor Therapy.” Cancer Journal of Causes: No Evidence for Any Previously Unidentified Infectious Diseases Medical Microbiology 16 (4): 230–22. Increased Risk Due to Antiretroviral Therapy.” AIDS 26 (3): Hulbert, A., C. M. Hooker, J. C. Keruly, T. Brown, K. Horton, 315–323. doi:10.1097/QAD.0b013e32834e8805. and others. 2014. “Prospective CT Screening for Lung Kreimer, A. R., M. E. Sherman, V. V. Sahasrabuddhe, and Cancer in a High-Risk Population: HIV-Positive Smokers.” M. Safaeian. 2015. “The Case for Conducting a Randomized Journal of Thoracic Oncology 9 (6): 752–59. Clinical Trial to Assess the Efficacy of a Single Dose of Hymes, K. B., T. Cheung, J. B. Greene, N. S. Prose, A. Marcus, and Prophylactic HPV Vaccines among Adolescents.” Journal of the others. 1981. “Kaposi’s Sarcoma in Homosexual Men—A National Cancer Institute 107 (3). doi:10.1093/jnci/dju436. Report of Eight Cases.” The Lancet 2 (8247): 598–600. Laar, A. K. 2013. “Reproductive Rights and Options Available to Jacobson, I. M., J. G. McHutchison, G. Dusheiko, A. M. Di Women Infected with HIV in Ghana: Perspectives of Service Bisceglie, K. R. Reddy, and others. 2011. “Telaprevir for Providers from Three Ghanaian Health Facilities.” BMC Previously Untreated Chronic Hepatitis C Virus Infection.” Womens Health 13: 13. doi:10.1186/1472–6874–13–13. New England Journal of Medicine 364 (25): 2405–16. Landrum, M. L., K. Huppler Hullsiek, A. Ganesan, A. C. Weintrob, doi:10.1056/NEJMoa1012912. N. F. Crum-Cianflone, and others. 2009. “Hepatitis B Vaccine Jantarapakde, J., N. Phanuphak, C. Chaturawit, S. Pengnonyang, Responses in a Large U.S. Military Cohort of HIV-Infected P. Mathajittiphan, and others. 2014. “Prevalence of Metabolic Individuals: Another Benefit of HAART in those with Syndrome among Antiretroviral-Naive and Antiretroviral- Preserved CD4 T-Cell Count.” Vaccine 27 (34): 4731–38. Experienced HIV-1 Infected Thai Adults.” AIDS Patient doi:10.1016/j.vaccine.2009.04.016. Care and STDs 28 (7): 331–40. doi:10.1089/apc.2013.0294. Lang, S., M. Mary-Krause, A. Simon, M. Partisani, J. Gilquin, Kane, M. A. 1995. “Global Programme for Control of Hepatitis and others. 2012. “HIV Replication and Immune Status Are B Infection.” Vaccine 13 (Suppl 1): S47–49. Independent Predictors of the Risk of Myocardial Infarction ———. 2012. “Preventing Cancer with Vaccines: Progress in in HIV-Infected Individuals.” Clinical Infectious Diseases the Global Control of Cancer.” Cancer Prevention Research 55 (4): 600–07. doi:10.1093/cid/cis489. 5 (1): 24–29. doi:10.1158/1940–6207.capr-11–0533. Liu, E., C. Armstrong, D. Spiegelman, G. Chalamilla, M. Njelekela, Kaufman, M. R., M. Smelyanskaya, L. M. Van Lith, E. C. Mallalieu and others. 2013. “First-Line Antiretroviral Therapy and and others. 2016. “Adolescent Sexual and Reproductive Health Changes in Lipid Levels over 3 Years among HIV-Infected Services and Implications for the Provision of Voluntary Adults in Tanzania.” Clinical Infectious Diseases 56 (12): Medical Male Circumcision: Results of a Systematic Literature 1820–28. doi:10.1093/cid/cit120. Review.” PLoS One 11 (3): e0149892. doi:10.1371/journal Lopez, L. M., D. A. Grimes, K. F. Schulz, K. M. Curtis, and .pone.0149892. M. Chen. 2014. “Steroidal Contraceptives: Effect on Bone Keith, A., Y. Dong, J. Shuter, and S. Himelhoch. 2016. Fractures in Women.” Cochrane Database of Systematic “Behavioral Interventions for Tobacco Use in HIV-Infected Reviews 6: CD006033. Smokers: A Meta-Analysis.” Journal of Acquired Immune Loutfy, M. R., S. Margolese, D. M. Money, M. Gysler, Deficiency Syndromes 72 (5): 527–33. S. Hamilton, and others. 2012. “Canadian HIV Pregnancy Kiertiburanakul, S., P. Luengroongroj, and S. Sungkanuparph. Planning Guidelines.” Journal of Obstetrics and Gynaecology 2012. “Clinical Characteristics of HIV-Infected Patients 34 (6): 575–90. Who Survive after the Diagnosis of HIV Infection for More Lozano, R., M. Naghavi, K. Foreman, S. Lim, K. Shibuya, and than 10 Years in a Resource-Limited Setting.” Journal of the others. 2012. “Global and Regional Mortality from 235 International Association of Physicians in AIDS Care 11 (6): Causes of Death for 20 Age Groups in 1990 and 2010: A 361–65. doi:10.1177/1545109712449191. Systematic Analysis for the Global Burden of Disease Study Kirk, G. D., C. Merlo, P. O’ Driscoll, S. H. Mehta, N. Galai, and 2010.” The Lancet 380 (9859): 2095–128. doi:10.1016 others. 2007. “HIV Infection Is Associated with an Increased /s0140–6736(12)61728–0. Risk for Lung Cancer, Independent of Smoking.” Clinical Massad, L. S., M. J. Silverberg, G. Springer, H. Minkoff, N. Hessol, Infectious Diseases 45 (1): 103–10. doi:10.1086/518606. and others. 2004. “Effect of Antiretroviral Therapy on the Klingman, K., J. D. Lundgren, A. Babiker, F. Gordin, S. Emery, Incidence of Genital Warts and Vulvar Neoplasia among and others. 2015. “Risk of Both AIDS-Related and Non-AIDS- Women with the Human Immunodeficiency Virus.” American Related Cancers Decreased by Early Initiation of Antiretroviral Journal of Obstetrics and Gynecology 190 (5): 1241–48. 62 Major Infectious Diseases Mbulaiteye, S. M., E. T. Katabira, H. Wabinga, D. M. Parkin, African Medicine 12 (1): 24–28. doi:10.4103/1596–3519 P. Virgo, and others. 2006. “Spectrum of Cancers among .108246. HIV-Infected Persons in Africa: The Uganda AIDS-Cancer Murray, C. J., and A. D. Lopez. 2013. “Measuring the Global Registry Match Study.” International Journal of Cancer Burden of Disease.” New England Journal of Medicine 118 (4): 985–90. 369 (5): 448–57. doi:10.1056/NEJMra1201534. McGlynn, K. A., J. L. Petrick, and W. T. London. 2015. Mutyaba, I., W. Phipps, E. M. Krantz, J. D. Goldman, S. Nambooze, “Global Epidemiology of Hepatocellular Carcinoma: An and others. 2015. “A Population-Level Evaluation of the Emphasis on Demographic and Regional Variability.” Effect of Antiretroviral Therapy on Cancer Incidence in Clinical Liver Disease 19 (2): 223–38. doi:10.1016/j.cld Kyadondo County, Uganda, 1999–2008.” Journal of Acquired .2015.01.001. Immune Deficiency Syndromes 69 (4): 481–86. Mdodo, R., E. L Frazier, S. R. Dube, C. L Mattson, M. Y. Sutton, Myer, L., L. Denny, R. Telerant, M. de Souza, T. C. Wright Jr., and others. 2015. “Cigarette Smoking Prevalence among and L. Kuhn. 2005. “Bacterial Vaginosis and Susceptibility Adults with HIV Compared with the General Adult to HIV Infection in South African Women: A Nested Case- Population in the United States: Cross-Sectional Surveys.” Control Study.” Journal of Infectious Diseases 192 (8): Annals of Internal Medicine 162 (5): 335–44. 1372–80. doi:10.1086/462427. Mensah, G. A. 2008. “Ischaemic Heart Disease in Africa.” Heart Myerson, M., E. Poltavskiy, E. J. Armstrong, S. Kim, V. Sharp, 94 (7): 836–43. doi:10.1136/hrt.2007.136523. and H. Bang. 2014. “Prevalence, Treatment, and Control of Mills, A. M., M. Nelson, D. Jayaweera, K. Ruxrungtham, I. Cassetti, Dyslipidemia and Hypertension in 4278 HIV Outpatients.” and others. 2009. “Once-Daily Darunavir/Ritonavir vs. Journal of Acquired Immune Deficiency Syndromes 66 (4): Lopinavir/Ritonavir in Treatment-Naive, HIV-1-Infected 370–77. doi:10.1097/QAI.0000000000000168. Patients: 96-Week Analysis.” AIDS 23 (13): 1679–88. Narayan, K. M., P. G. Miotti, N. P. Anand, L. M. Kline, doi:10.1097/QAD.0b013e32832d7350. C. Harmston, and others. 2014.“HIV and Noncommunicable Mok-Lin, E., S. Missmer, K. Berry, L. S. Lehmann, and Disease Comorbidities in the Era of Antiretroviral E. S. Ginsburg. 2011. “Public Perceptions of Providing IVF Therapy: A Vital Agenda for Research in Low- and Middle- Services to Cancer and HIV Patients.” Fertility and Sterility Income Country Settings.” Journal of Acquired Immune 96 (3): 722–27. doi:10.1016/j.fertnstert.2011.06.051. Deficiency Syndromes 67 (Suppl 1): S2–7. doi:10.1097/QAI Money, D. M., E. Moses, S. Blitz, S. M. Vandriel, N. Lipsky, and .0000000000000267. others. 2016. “HIV Viral Suppression Results in Higher Nguemaim, N. F., J. Mbuagbaw, T. Nkoa, G. Alemnji, G. Teto, Antibody Responses in HIV-Positive Women Vaccinated and others. 2010. “Serum Lipid Profile in Highly Active with the Quadrivalent Human Papillomavirus Vaccine.” Antiretroviral Therapy–Naive HIV-Infected Patients in Vaccine 34 (40): 4799–806. Cameroon: A Case-Control Study.” HIV Medicine 11 (6): Mosam, A., F. Shaik, T. S. Uldrick, T. Esterhuizen, G. H. Friedland, 353–59. doi:10.1111/j.1468–1293.2009.00784.x. and others. 2012. “A Randomized Controlled Trial of Highly Nguyen, H. Q., and C. Casper. 2010. “The Epidemiology of Active Antiretroviral Therapy versus Highly Active Kaposi Sarcoma.” In Kaposi Sarcoma: A Model of Oncogenesis, Antiretroviral Therapy and Chemotherapy in Therapy-Naive edited by L. Pantanowitz, J. Stebbing, and B. J. Dezube, Patients with HIV-Associated Kaposi Sarcoma in South 197–232. Kerala, India: Research Signpost. Africa.” Journal of Acquired Immune Deficiency Syndromes Nguyen, H. Q., A. S. Magaret, S. E. Van Rompaey, M. M. Kitahata, 60 (2): 150–57. doi:10.1097/QAI.0b013e318251aedd. A. Wald, and others. 2008. “Persistent Kaposi Sarcoma in Moses, E., H. N. Pedersen, S. M. Mitchell, M. Sekikubo, the Era of HAART: Characterizing the Predictors of Clinical D. Mwesigwa, and others. 2015. “Uptake of Community- Response.” AIDS 22 (8): 937–45. Based, Self-Collected HPV Testing vs. Visual Inspection Nieves, C. I., A. Kaida, G. R. Seage, J. Kabakyenga, W. Muyindike, with Acetic Acid for Cervical Cancer Screening in Kampala, and others. 2015. “The Influence of Partnership on Uganda: Preliminary Results of a Randomised Controlled Contraceptive Use among HIV-Infected Women Accessing Trial.” Tropical Medicine and International Health 10 (20): Antiretroviral Therapy in Rural Uganda.” Contraception 1355–67. doi:10.1111/tmi.12549. 92 (2): 152–59. doi:10.1016/j.contraception.2015.04.011. Moyer, V. A. 2012. “Screening for Cervical Cancer: U.S. Nurudeen, S. K., L. C. Grossman, L. Bourne, M. M. Guarnaccia, Preventive Services Task Force Recommendation Statement.” M. V. Sauer, and others. 2013. “Reproductive Outcomes of Annals of Internal Medicine 156 (12): 880–91, W312. HIV Seropositive Women Treated by Assisted Reproduction.” doi:10.7326/0003–4819–156–12–201206190–00424. Journal of Women’s Health 22 (3): 243–49. doi:10.1089/jwh Mugo, N. R., T. Hong, C. Celum, D. Donnell, E. A. Bukusi, and .2012.3855. others. 2014. “Pregnancy Incidence and Outcomes among Nzuobontane, D., K. N. Blackett, and C. Kuaban. 2002. “Cardiac Women Receiving Preexposure Prophylaxis for HIV Involvement in HIV Infected People in Yaounde, Cameroon.” Prevention: A Randomized Clinical Trial.” Journal of the Postgraduate Medical Journal 78 (925): 678–81. American Medical Association 312 (4): 362–71. doi:10.1001 Ohl, J., M. Partisani, C. Wittemer, J. M. Lang, S. Viville, and /jama.2014.8735. R. Favre. 2005. “Encouraging Results Despite Complexity of Muhammad, S., M. U. Sani, and B. N. Okeahialam. Multidisciplinary Care of HIV-Infected Women Using 2013. “Prevalence of Dyslipidemia among Human Assisted Reproduction Techniques.” Human Reproduction Immunodeficiency Virus Infected Nigerians.” Annals of 20 (11): 3136–40. doi:10.1093/humrep/dei185. HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 63 Palella, F. J., Jr., K. M. Delaney, A. C. Moorman, M. O. Pyra, M., R. Heffron, N. R. Mugo, K. Nanda, K. K. Thomas, and Loveless, J. Fuhrer, and others. 1998. “Declining Morbidity others. 2015. “Effectiveness of Hormonal Contraception in and Mortality among Patients with Advanced Human HIV-Infected Women Using Antiretroviral Therapy.” AIDS Immunodeficiency Virus Infection. HIV Outpatient Study 29 (17): 2353–59. Investigators.” New England Journal of Medicine 338 (13): Quinn, T. C., W. E. Stamm, S. E. Goodell, E. Mkrtichian, 853–60. J. Benedetti, and others. 1983. “The Polymicrobial Origin of Panel on Opportunistic Infections in HIV-Infected Adults and Intestinal Infections in Homosexual Men.” New England Adolescents. 2016. “Guidelines for the Prevention and Journal of Medicine 309 (10): 576–82. Treatment of Opportunistic Infections in HIV-infected Raffi, F., A. Rachlis, H. J. Stellbrink, W. D. Hardy, C. Torti, and Adults and Adolescents: Recommendations from the others. 2013. “Once-Daily Dolutegravir versus Raltegravir Centers for Disease Control and Prevention, the National in Antiretroviral-Naive Adults with HIV-1 Infection: 48 Institutes of Health, and the HIV Medicine Association of Week Results from the Randomised, Double-Blind, the Infectious Diseases Society of America.” http://aidsinfo Non-Inferiority SPRING-2 Study.” The Lancet 381 (9868): .nih.gov/contentfiles/lvguidelines/adult_oi.pdf. 735–43. doi:10.1016/S0140–6736(12)61853–4. Papadopoulos, A., N. Pantazis, P. Panagopoulos, S. Kourkounti, Riddler, S. A., E. Smit, S. R. Cole, R. Li, J. S. Chmiel, and others. G. Xylomenos, and others. 2012. “Effects of First Antiretroviral 2003. “Impact of HIV Infection and HAART on Serum Regimen on Lipid Levels in HIV (+) Individuals.” Journal Lipids in Men.” Journal of the American Medical Association of Chemotherapy 24 (1): 38–47. doi:10.1179/1120009X12Z 289 (22): 2978–82. .0000000008. Riedel, D. J., E. I. Mwangi, L. E. Fantry, C. Alexander, M. B. Patel, P., T. Bush, K. H. Mayer, S. Desai, K. Henry, and others. Hossain, and others. 2013. “High Cancer-Related Mortality 2012. “Prevalence and Risk Factors Associated with Herpes in an Urban, Predominantly African-American, HIV- Simplex Virus-2 Infection in a Contemporary Cohort of Infected Population.” AIDS 27 (7): 1109–17. doi:10.1097 HIV-Infected Persons in the United States.” Sexually /QAD.0b013e32835dc068. Transmitted Diseases 39 (2): 154–60. Robbins, H. A., R. M. Pfeiffer, M. S. Shiels, J. Li, H. I. Hall, and Patel, P., D. L. Hanson, P. S. Sullivan, R. M. Novak, A. C. E. A. Engels. 2015. “Excess Cancers among HIV-Infected Moorman, and others. 2008. “Incidence of Types of Cancer People in the United States.” Journal of the National Cancer among HIV-Infected Persons Compared with the General Institute 107 (4). doi:10.1093/jnci/dju503. Population in the United States, 1992–2003.” Annals of Robbins, H. A., M. S. Shiels, R. M. Pfeiffer, and E. A. Engels. Internal Medicine 148 (10): 728–U29. 2014. “Epidemiologic Contributions to Recent Cancer Pefura Yone, E. W., A. F. Betyoumin, A. P. Kengne, F. J. Kaze Trends among HIV-Infected People in the United States.” Folefack, and J. Ngogang. 2011. “First-Line Antiretroviral AIDS 28 (6): 881–90. doi:10.1097/qad.0000000000000163. Therapy and Dyslipidemia in People Living with HIV-1 in Robinson, J. A., R. Jamshidi, and A. E. Burke. 2012. Cameroon: A Cross-Sectional Study.” AIDS Research and “Contraception for the HIV-Positive Woman: A Review of Therapy 8: 33. doi:10.1186/1742–6405–8-33. Interactions between Hormonal Contraception and Petoumenos, K., E. Hui, N. Kumarasamy, S. J. Kerr, J. Y. Choi, and Antiretroviral Therapy.” Infectious Diseases in Obstetrics and others. 2010. “Cancers in the TREAT Asia HIV Observational Gynecology 2012 (890160). Database (TAHOD): A Retrospective Analysis of Risk Rockstroh, J. K., E. DeJesus, J. L. Lennox, Y. Yazdanpanah, Factors.” Journal of the International AIDS Society 13 (51). M. S. Saag, and others. 2013. “Durable Efficacy and Safety Piketty, C., H. Selinger-Leneman, A. M. Bouvier, A. Belot, of Raltegravir versus Efavirenz When Combined with M. Mary-Krause, and others. 2012. “Incidence of HIV- Tenofovir/Emtricitabine in Treatment-Naive HIV-1-Infected Related Anal Cancer Remains Increased Despite Long-Term Patients: Final 5-Year Results from STARTMRK.” Journal Combined Antiretroviral Treatment: Results from the of Acquired Immune Deficiency Syndromes 63 (1): 77–85. French Hospital Database on HIV.” Journal of Clinical doi:10.1097/QAI.0b013e31828ace69. Oncology 30 (35): 4360–66. doi:10.1200/JCO.2012.44.5486. Rodger, A. J., V. Cambiano, T. Bruun, P. Vernazza, S. Collins, and Pinto Neto, L. F., M. B. das Neves, R. Ribeiro-Rodrigues, others. 2016. “Sexual Activity without Condoms and Risk of K. Page, and A. E. Miranda. 2013. “Dyslipidemia and Fasting HIV Transmission in Serodifferent Couples When the HIV Glucose Impairment among HIV Patients Three Years Positive Partner Is Using Suppressive Antiretroviral Therapy.” after the First Antiretroviral Regimen in a Brazilian AIDS Journal of the American Medical Association 316 (2): Outpatient Clinic.” Brazilian Journal of Infectious Diseases 171–81. 17 (4): 438–43. doi:10.1016/j.bjid.2012.12.006. Sabin, C. A., S. W. Worm, R. Weber, P. Reiss, W. El-Sadr, and Poordad, F., J. McCone Jr., B. R. Bacon, S. Bruno, M. P. Manns, others. 2008. “Use of Nucleoside Reverse Transcriptase and others. 2011. “Boceprevir for Untreated Chronic Hepatitis Inhibitors and Risk of Myocardial Infarction in HIV- C Virus Genotype 1 Infection.” New England Journal of Infected Patients Enrolled in the DAD Study: A Multicohort Medicine 364 (13): 1195–206. doi:10.1056/NEJMoa1010494. Collaboration.” The Lancet 371 (9622): 1417–26. Puoti, M., R. Bruno, V. Soriano, F. Donato, G. B. Gaeta, and Salehian, B., J. Bilas, M. Bazargan, and M. Abbasian. 2005. others. 2004. “Hepatocellular Carcinoma in HIV-Infected “Prevalence and Incidence of Diabetes in HIV-Infected Patients: Epidemiological Features, Clinical Presentation Minority Patients on Protease Inhibitors.” Journal of the and Outcome.” AIDS 18 (17): 2285–93. National Medical Association 97 (8): 1088–92. 64 Major Infectious Diseases Salters, K. A., A. Cescon, W. Zhang, G. Ogilvie, M. Murray, Syndrome Epidemic to De Novo Presentations of Heart and others. 2016. “Cancer Incidence among HIV-Positive Disease in the Heart of Soweto Study Cohort.” European Women in British Columbia, Canada: Heightened Risk of Heart Journal 33 (7): 866–74. doi:10.1093/eurheartj Virus-Related Malignancies.” HIV Medicine 17 (3): 188–95. /ehr398. doi:10.1111/hiv.12290. Smith, C. J., L. Ryom, R. Weber, P. Morlat, C. Pradier, and Samaras, K. 2009. “Prevalence and Pathogenesis of Diabetes others. 2014. “Trends in Underlying Causes of Death in Mellitus in HIV-1 Infection Treated with Combined People with HIV from 1999 to 2011 (DAD): A Multicohort Antiretroviral Therapy.” Journal of Acquired Immune Collaboration.” The Lancet 384 (9939): 241–48. doi:10.1016 Deficiency Syndromes 50 (5): 499–505. doi:10.1097/QAI /s0140–6736(14)60604–8. .0b013e31819c291b. Solomon, D. H., E. W. Karlson, E. B. Rimm, C. C. Cannuscio, Sax, P. E., C. Tierney, A. C. Collier, E. S. Daar, K. Mollan, and L. A. Mandl, and others. 2003. “Cardiovascular Morbidity others. 2011. “Abacavir/Lamivudine versus Tenofovir Df/ and Mortality in Women Diagnosed with Rheumatoid Emtricitabine as Part of Combination Regimens for Initial Arthritis.” Circulation 107 (9): 1303–07. Treatment of HIV: Final Results.” Journal of Infectious Sung, J. J., K. K. Tsoi, V. W. Wong, K. C. Li, and H. L. Chan. Diseases 204 (8): 1191–201. doi:10.1093/infdis/jir505. 2008. “Meta-Analysis: Treatment of Hepatitis B Infection Schwartz, J. B., K. L. Moore, M. Yin, A. Sharma, D. Merenstein, Reduces Risk of Hepatocellular Carcinoma.” Alimentary and others. 2014. “Relationship of Vitamin D, HIV, HIV Pharmacology and Therapeutics 28 (9): 1067–77. doi:10.1111 Treatment, and Lipid Levels in the Women’s Interagency /j.1365–2036.2008.03816.x. HIV Study of HIV-Infected and Uninfected Women in Thurman, A. R., S. Anderson, and G. F. Doncel. 2014. “Effects the United States.” Journal of the International Association of Hormonal Contraception on Antiretroviral Drug of Providers of AIDS Care 13 (3): 250–59. doi:10.1177 Metabolism, Pharmacokinetics and Pharmacodynamics.” /2325957413506748. American Journal of Reproductive Immunology 71 (6): 523–30. Seedat, Y. K. 1999. “Hypertension in Black South Africans.” doi:10.1111/aji.12210. Journal of Human Hypertension 13 (2): 96–10. Thygesen, K., J. S. Alpert, H. D. White, A. S. Jaffe, F. S. Apple, Sengayi, M., A. Spoerri, M. Egger, D. Kielkowski, T. Crankshaw, and others. 2007. “Universal Definition of Myocardial and others. 2016. “Record Linkage to Correct Under- Infarction.” Circulation 116 (22): 2634–53. Ascertainment of Cancers in HIV Cohorts: The Sinikithemba Tien, P. C., M. F. Schneider, S. R. Cole, A. M. Levine, M. Cohen, HIV Clinic Linkage Project.” International Journal of Cancer and others. 2007. “Antiretroviral Therapy Exposure and 139 (6): 1209–16. Incidence of Diabetes Mellitus in the Women’s Interagency Shaffer, D., M. D. Hughes, F. Sawe, Y. Bao, A. Moses, and others. HIV Study.” AIDS 21 (13): 1739–1745. 2014. “Cardiovascular Disease Risk Factors in HIV-Infected Triant, V. A. 2014. “Epidemiology of Coronary Heart Disease in Women after Initiation of Lopinavir/Ritonavir– and Patients with Human Immunodeficiency Virus.” Reviews of Nevirapine–Based Antiretroviral Therapy in Sub-Saharan Cardiovascular Medicine 15 (Suppl 1): S1–8. Africa: A5208 (OCTANE).” Journal of Acquired Immune Triant, V. A., J. B. Meigs, and S. K. Grinspoon. 2009. “Association Deficiency Syndromes 66 (2): 155–63. doi:10.1097/QAI of C-Reactive Protein and HIV Infection with Acute .0000000000000131. Myocardial Infarction.” Journal of Acquired Immune Shen, Y. C., C. Hsu, C. C. Cheng, F. C. Hu, and A. L. Cheng. Deficiency Syndromes 51 (3): 268–73. doi:10.1097/QAI 2012. “A Critical Evaluation of the Preventive Effect of .0b013e3181a9992c. Antiviral Therapy on the Development of Hepatocellular Turner, A. N., S. Tabbah, V. Mwapasa, S. J. Rogerson, S. R. Carcinoma in Patients with Chronic Hepatitis C or B: Meshnick, and others. 2013. “Severity of Maternal HIV-1 A Novel Approach by Using Meta-Regression.” Oncology Disease Is Associated with Adverse Birth Outcomes in 82 (5): 275–89. doi:10.1159/000337293. Malawian Women: A Cohort Study.” Journal of Acquired Shiels, M. S., R. M. Pfeiffer, M. H. Gail, H. I. Hall, J. Li, and Immune Deficiency Syndromes 64 (4): 392–99. others. 2011. “Cancer Burden in the HIV-Infected Population Twagirumukiza, M., E. Nkeramihigo, B. Seminega, E. Gasakure, in the United States.” Journal of the National Cancer Institute F. Boccara, and others. 2007. “Prevalence of Dilated 103 (9): 753–62. doi:10.1093/jnci/djr076. Cardiomyopathy in HIV-Infected African Patients Not Silverberg, M. J., B. Lau, C. J. Achenbach, Y. Jing, K.N. Althoff, Receiving HAART: A Multicenter, Observational, Prospective, and others. 2015. “Cumulative Incidence of Cancer among Cohort Study in Rwanda.” Current HIV Research 5 (1): Persons with HIV in North America: A Cohort Study.” 129–37. Annals of Internal Medicine 163 (7): 507–18. UNAIDS (Joint United Nations Programme on HIV/AIDS). Silverberg, M. J., W. A. Leyden, L. Xu, M. A. Horberg, C. R. Chao, 2014. The Gap Report. UNAIDS, Geneva. and others. 2014. “Immunodeficiency and Risk of Myocardial van Leeuwen, M. T., C. M. Vajdic, M. G. Middleton, A. M. Infarction among HIV-Positive Individuals with Access to McDonald, M. Law, and others. 2009. “Continuing Declines Care.” Journal of Acquired Immune Deficiency Syndromes in Some but Not All HIV-Associated Cancers in Australia 65 (2): 160–66. doi:10.1097/QAI.0000000000000009. after Widespread Use of Antiretroviral Therapy.” AIDS Sliwa, K., M. J. Carrington, A. Becker, F. Thienemann, 23 (16): 2183–90. doi:10.1097/QAD.0b013e328331d384. M. Ntsekhe, and others. 2012. “Contribution of the Human Venkatesh, K. K., S. Saghayam, B. Devaleenal, S. Poongulali, Immunodeficiency Virus/Acquired Immunodeficiency T. P. Flanigan, and others. 2012. “Spectrum of Malignancies HIV/AIDS Comorbidities: Impact on Cancer, Noncommunicable Diseases, and Reproductive Health 65 among HIV-Infected Patients in South India.” Indian Journal among HIV-1-Infected Adults Receiving Combination of Cancer 49 (1): 176–80. doi:10.4103/0019–509X.98947. Antiretroviral Therapy in Resource-Replete versus Resource- Visnegarwala, F., L. Chen, S. Raghavan, and E. Tedaldi. 2005. Limited Urban Setting.” AIDS 25 (12): 1471–79. doi:10.1097 “Prevalence of Diabetes Mellitus and Dyslipidemia among /QAD.0b013e328347f9d4. Antiretroviral Naive Patients Co-Infected with Hepatitis C WHO (World Health Organization). 2006. Sexual and Virus (HCV) and HIV-1 Compared to Patients without Reproductive Health of Women Living with HIV/AIDS. Co-Infection.” Journal of Infection 50 (4): 331–37. Geneva: WHO. Wanyenze, R. K., G. J. Wagner, N. M. Tumwesigye, M. ———. 2013a. Global Policy Report on the Prevention and Control Nannyonga, F. Wabwire-Mangen, and others. 2013. “Fertility of Viral Hepatitis in WHO Member States. Geneva: WHO. and Contraceptive Decision-Making and Support for HIV ———. 2013b. WHO Guidelines for Screening and Treatment Infected Individuals: Client and Provider Experiences and of Precancerous Lesions for Cervical Cancer Prevention. Perceptions at Two HIV Clinics in Uganda.” BMC Public Geneva: WHO. Health 13: 98. doi:10.1186/1471–2458–13–98. Wiktor, S. Z. 2017. “Viral Hepatitis.” In Disease Control Priorities Watson-Jones, D., N. Mugo, S. Lees, M. Mathai, S. Vusha, and (third edition): Volume 6, Major Infectious Diseases, others. 2015. “Access and Attitudes to HPV Vaccination edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, and P. Jha. amongst Hard-to-Reach Populations in Kenya.” PLoS One Washington, DC: World Bank. 10 (6): e0123701. doi:10.1371/journal.pone.0123701. Workowski, K. A., and G. A. Bolan. 2015. “Sexually Watts, D. H., Z. A. Brown, D. Money, S. Selke, M. L. Huang, Transmitted Diseases Treatment Guidelines, 2015.” and others. 2003. “A Double-Blind, Randomized, Placebo- MMWR Recommendations and Reports 64 (RR-03): Controlled Trial of Acyclovir in Late Pregnancy for the 1–137. Reduction of Herpes Simplex Virus Shedding and Cesarean Yanik, E. L., S. Napravnik, S. R. Cole, C. J. Achenbach, S. Gopal, Delivery.” American Journal of Obstetrics and Gynecology and others. 2013. “Incidence and Timing of Cancer in HIV- 188 (3): 836–43. Infected Individuals Following Initiation of Combination Weller, S., and K. Davis. 2002. “Condom Effectiveness Antiretroviral Therapy.” Clinical Infectious Diseases 57 (5): in Reducing Heterosexual HIV Transmission.” Cochrane 756–64. doi:10.1093/cid/cit369. Database of Systematic Reviews (1): CD003255. doi:10.1002 Yaro, S., N. Meda, A. Desgrees Du Lou, I. Sombie, M. Cartoux, /14651858.CD003255. and others. 2001. “[Impaired Fertility in Women Infected Wester, C. W., J. R. Koethe, B. E. Shepherd, S. E. Stinnette, with HIV-1. Implications for Sentinel Serosurveillance].” P. F. Rebeiro, and others. 2011. “Non-AIDS-Defining Events Rev Epidemiol Sante Publique 49 (3): 221–28. 66 Major Infectious Diseases Chapter 4 HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations Katherine Harripersaud, Margaret McNairy, Saeed Ahmed, Elaine J. Abrams, Harsha Thirumurthy, and Wafaa M. El-Sadr INTRODUCTION The chapter includes information from peer-reviewed manuscripts identified through a targeted literature The management of human immunodeficiency virus review focused on publications pertinent to low- and (HIV) infection has evolved substantially since the middle-income countries (LMICs), with a focus on Sub- advent of the HIV/acquired immune deficiency syn- Saharan Africa. Studies conducted in other LMIC regions drome (HIV/AIDS) epidemic in the 1980s. The discov- and high-income countries are referenced when they ery of effective antiretroviral therapy (ART) transformed address a key relevant issue. A table summarizing the lives of persons living with HIV (Deeks, Lewin, and approaches to improving HIV testing, linkage to and Havlir 2013) by achieving a substantial drop in morbid- engagement with HIV care, retention in HIV care, and ity and mortality (Danel and others 2015; START Study adherence to HIV treatment is included in annex 4A. Group 2015). Additionally, evidence supports the effi- cacy of ART in preventing the transmission of HIV infection (Cohen and others 2011). HIV TESTING SERVICES Progress in controlling the HIV epidemic, however, requires the achievement of virologic suppression among Rationale and Coverage all HIV-infected individuals, which, in turn, requires the HIV testing services (HTS) are essential for identifying identification of such individuals and their retention HIV-positive persons in need of care and ART, as well as across the care continuum—from conducting HIV test- for identifying at-risk HIV-negative persons for referral ing, linking HIV-positive individuals to care, retaining and engagement in HIV prevention programs (Celum them in care, and achieving viral suppression (figure 4.1) and others 2013). Despite the importance of HIV (Gardner and others 2011). For each step of the contin- testing, the Joint United Nations Programme on HIV/ uum, this chapter discusses the rationale, relevant guide- AIDS (UNAIDS) estimates that nearly half of the lines, measurements of each parameter, barriers to 36.9 million people living with HIV globally in 2014 achieving successful outcomes, interventions demon- were unaware of their infection (UNAIDS 2015a). strated to be effective, and available data on the costs and Similarly, only 44 percent of pregnant women in LMICs cost-effectiveness of interventions. access HIV testing (WHO 2014c). Recent findings from Corresponding author: Wafaa M. El-Sadr, ICAP, Mailman School of Public Health, Columbia University, New York, New York, United States; wme1@cumc .columbia.edu. 67 Figure 4.1 HIV Care Continuum Test Engage, counsel, Retain, counsel, Adherence support monitor, and support monitor, and support Initiate Link ART Undetectable HIV-positive HIV care HIV treatment viral load Source: McNairy and El-Sadr 2012. Adapted with permission from Wolters Kluwer. Note: ART = antiretroviral treatment; HIV = human immunodeficiency virus. the Population HIV Impact Assessment (PHIA) surveys epidemics (WHO 2013b). Provider-initiated testing and completed in Malawi, Zambia, and Zimbabwe suggest counseling (PITC) is recommended for women as a routine the need to enhance HIV testing coverage; in these component of antenatal care, with retesting before delivery surveys of random samples of households, 70.4 percent for persons with a first negative test (WHO 2015a). of individuals found to be HIV-positive were aware of All infants of HIV-infected pregnant women should HIV infection (ICAP 2016). be tested for infection at age four to six weeks using a HIV testing is especially critical for pregnant women virologic test that directly detects the virus, including and children. HIV-infected pregnant women must be deoxyribonucleic acid (DNA), ribonucleic acid (RNA), identified early during pregnancy to benefit from ART or DNA polymerase chain reaction (PCR) assays (WHO for their own health and for the prevention of vertical 2015a). Infants testing positive should be initiated on transmission; HIV-negative pregnant and postpartum ART, while those testing negative should be tested again women require repeat testing during pregnancy and the with virologic or serologic assays (depending on their postpartum period (Drake and others 2014). age) after the risk period for mother-to-child transmis- HIV testing and ART initiation for HIV-infected sion ends with cessation of breastfeeding. infants in the first months of life has been demonstrated For children ages 18 months or older, the WHO to markedly reduce mortality (Violari and others 2008). recommends serological testing when HIV infection is However, only 46 percent of HIV-exposed infants in suspected or exposure has occurred; and in high-prevalence LMICs received an early infant diagnostic (EID) test for settings, HIV testing should be routinely available to chil- HIV in 2014 (UNICEF 2015). dren attending immunization, under-five, malnutrition, Adolescents are another high-risk group in need of and tuberculosis clinics; all hospitalized children; and increased access to and uptake of HIV testing. In 2015, those receiving services for orphans and vulnerable there were 1.8 million adolescents ages 10–19 years living children (WHO 2015a). In generalized HIV epidemic with HIV (UNAIDS 2015b). Since 2004, mortality settings, HTS, with links to prevention, treatment, and has decreased 30 percent among adults but increased care, are recommended for all adolescents (WHO 2015a). 50 percent among adolescents (Porth and others 2014). For populations at substantial risk for HIV, including In the PHIA surveys cited above, only 46.4 percent of persons who inject drugs (PWID), sex workers (SWs), and young individuals (15–24 years old) found to be HIV- men who have sex with men (MSM), frequent, voluntary, positive in the population surveys were aware of their community-based HTS, linked to prevention, treatment, HIV infection (ICAP 2016). and care, are recommended. Guidelines Measurement The World Health Organization (WHO) recommends Common measures of HIV testing include the num- universal testing in countries with generalized HIV ber of tests conducted and the number of individuals 68 Major Infectious Diseases who are tested, counseled, and receive their results Community-based testing (CBT) strategies, includ- (PEPFAR 2013). However, these measures do not ing testing in homes, schools, the workplace, or other assess the effectiveness of HIV testing in identifying community venues (Bateganya, Abdulwadud, and Kiene all persons with HIV infection in a community. 2010; Coates and others 2014), have demonstrated Others have highlighted the importance of measuring promise for expanding knowledge of HIV status. the proportion of individuals who test positive for Compared with facility-based testing, CBT has been HIV who then link to care (El-Sadr, Gamble, and associated with increased testing, higher proportions of Cohen 2013). first-time testers, higher proportions of HIV-infected individuals identified with CD4 cell count greater than 350 cells per microliter, and an overall lower seropositiv- Barriers ity rate (Suthar, Ford, and others 2013). Home-based Individual-level barriers to testing include perceived testing has been highly acceptable, with 83 percent low risk of infection, anticipated psychological burden accepting testing in a meta-analysis (Sabapathy and of living with HIV, fear of HIV-related stigma, direct others 2012). It was also found to be cost-effective for and indirect financial costs of accessing HIV testing, reaching youth, men, and rural populations (Sweat and and gender inequality (Musheke and others 2013). others 2011; van Rooyen, McGrath, and others 2013; Testing of children and adolescents is complicated by Wachira and others 2014). Home-based testing has also the need to obtain consent from a parent or legal been more effective than facility-based testing at guardian (WHO 2013b, 2014b). System-level barriers diagnosing people with less advanced disease and has include high patient load, test kit shortages, inadequate increased linking of HIV-positive persons to care counseling space, and poor counseling skills (Larsson (Wachira, Kimaiyo, and others 2012). and others 2011; Nuti, Kabengula, and Msuya 2011). CBT has also shown promise for pediatric case find- For pregnant women, low uptake of antenatal services ing (Ahmed and others 2015). Among members of in some settings remains an impediment to achieving households with HIV-infected individuals in Uganda, universal HIV testing. home-based testing increased identification of HIV- infected persons, including children, compared with clinic-based care (Lugada and others 2010). Approaches to Improve HTS Uptake Mass campaigns that provide HTS in conjunction Voluntary counseling and testing (VCT) is dependent on with other health services have also been used. In Kenya individuals’ actively seeking HIV testing at a clinic, hos- and Uganda, HTS were provided along with free bed pital, mobile health unit, or free-standing site (Matovu nets, water purification kits, condoms, and cotrimox- and Makumbi 2007). However, VCT is of limited value azole to those who tested HIV-positive (Lugada and among individuals who do not perceive themselves to be others 2010; Tumwesigye and others 2010). In Kenya, at risk. Moreover, children rarely access VCT services, 80 percent of the population in a catchment area was because caregivers fear stigma and disclosure of their tested for HIV in a period of one week; in Uganda, own serostatus (Ahmed and others 2013). 63 percent of households were tested through such a In 2007, the WHO and UNAIDS released guidance campaign. In Tanzania, a CBT campaign for children recommending PITC for all patients in high-prevalence identified 108 new pediatric cases (3.6 percent) (Shea settings and for all patients, irrespective of epidemic and others 2013). Mass campaigns that include HIV setting, whose clinical presentation suggests HIV infec- testing have been less successful in reaching young tion (Kennedy and others 2013; Roura and others adults and adult males (Chamie and others 2014; 2013; WHO 2007). A systematic review of PITC in Ostermann and others 2011). Sub-Saharan Africa reported increased uptake of HIV Couples’ testing is another approach to identifying testing after PITC was implemented in antenatal and concordant HIV-positive couples in need of ART and family planning, tuberculosis, sexually transmitted serodiscordant couples in need of ART for positive part- infection, and outpatient clinics (Kennedy and others ner or preexposure prophylaxis (PrEP) or voluntary 2013). PITC uptake is lower among men and among medical male circumcision for negative partner (Baeten women who are not pregnant, however (Baggaley and and others 2012; Gray and others 2007). others 2012; Hensen and others 2012; MacPherson and Self-testing for HIV holds promise (Choko and others others 2012). PITC has been effective in identifying 2011; Pant Pai and others 2013; Thirumurthy and others HIV-infected and -exposed infants in inpatient set- 2016). In Malawi, self-testing with oral test kits was tings, but it is not widely implemented for children acceptable among 92 percent of persons, and 99 percent (Kankasa and others 2009). of results were accurate (Choko and others 2011). HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 69 More research is needed to evaluate linking of HIV- In East Africa, VCT can be delivered for US$10–US$30 positive patients to care, as well as disclosure of results to per person tested, and the overall cost-effectiveness of partners among persons who self-test positive. VCT has been favorable in the few available studies Limited access to the specialized tests (nucleic acid (Grabbe and others 2010; Menzies and others 2009). In tests including DNA and RNA PCR) for diagnosis of HIV Tanzania, the cost per disability-adjusted life year (DALY) infection in infants resulted in high mortality among averted was estimated to be US$13–US$18 and the HIV-infected babies (Creek and others 2008). The intro- cost per HIV infection averted to be US$249–US$346 duction of new laboratory technology at centralized lab- (Sweat and others 2000). Offering VCT for free was oratories in many Sub-Saharan African countries, coupled highly cost-effective (Thielman and others 2006). Because with the use of dried blood spots for specimen collection, utilization levels for VCT services are an important driver improved access and uptake of early infant diagnosis of the costs per client, demand creation for VCT is essen- (Essajee and others 2015; Ghadrshenas and others 2013). tial for reducing costs. In addition, many studies underes- The use of text message printers and computerized track- timate the cost-effectiveness of HIV testing by neglecting ing systems has increased the rate of return of EID results the prevention benefits of knowing HIV-positive status, to clinics and families (Essajee and others 2015; which leads to a decrease in risk behaviors, and the Finocchario-Kessler and others 2014). cost-effectiveness of ART provision to individuals In South Africa, couples-based HIV testing in MSM identified as HIV-infected. appears to be effective (Stephenson and others 2013). In The costs of VCT approaches to HIV testing are China, peer-driven counseling sessions significantly higher when calculated per HIV-infected person identi- increased testing rates among MSM (Menzies and fied rather than per individual tested. Costs per others 2009). Text reminders also increased testing rates HIV-infected person identified are often lower for alter- (Bourne and others 2011). For PWID, venue-based test- native testing approaches such as PITC (including ing at methadone clinics and drop-in sites was successful hospital-based testing) than for VCT. In Uganda, in increasing knowledge of HIV status (Xia and others hospital-based testing costs US$12 per client tested and 2013). Additionally, a qualitative study noted that female US$43 per HIV-infected person identified (Menzies and sex workers (FSWs) in Benin would more likely access others 2009). In contrast, VCT costs US$19 per person HIV-testing services if paired with outreach strategies tested and US$101 per HIV-infected person identified. (Dugas and others 2015). A systematic review of seven partner notification Community-Based Testing studies in LMICs to achieve testing of partners described Evidence is emerging on the costs of approaches to CBT. the use of email, text messaging, and social networking, Studies have found the costs of such approaches to be with most clients choosing to send notifications via text lower than or comparable to the costs of VCT, at approx- rather than email to enhance testing of partners of HIV- imately US$8–US$15 per person tested (Grabbe and positive individuals (Hochberg, Berringer, and Schneider others 2010; Menzies and others 2009). In Kenya 2015). In a study from Malawi, only 25 percent of part- and Uganda, mobile HTS had a lower cost per person ners in the passive referral arm returned for HIV testing, tested than VCT. The cost per HIV-infected person iden- compared with 51 percent in both the contract and the tified was lower for mobile HTS than VCT in Kenya, but provider referral arms (p < 0.001) (Brown and others higher in Uganda. A modeling analysis for South Africa 2011). Similarly, a cohort study in Cameroon reported that estimated the incremental cost-effectiveness ratios that 46 percent of partners in the passive referral group (ICERs) per DALY averted of home-based HIV testing returned for testing as compared with 60 percent in the and counseling along with enhanced links to care found provider group and 61 percent in the contract referral a high level of cost-effectiveness (Smith and others arm (Henley and others 2013). 2015). The cost per DALY averted was US$1,340 at an ART threshold of CD4 count less than 200 cells per microliter and US$1,360 at universal access to ART. Cost-Effectiveness Considerations Although not high-prevalence settings, studies Table 4.1 describes cost and cost-effectiveness studies conducted in China and India have provided additional regarding HTS. evidence on the cost-effectiveness of targeted use of VCT among key populations and high-risk groups. In Voluntary Counseling and Testing China, among populations at higher risk of HIV acqui- Evidence suggests that while VCT and CBT have low sition, such as MSM, VCT was cost saving (Wang, costs per person tested, the cost per HIV-infected person Moss, and Hiller 2011). In Indonesia, costs per HIV- tested is lower for PITC. infected person identified were lower in prisons than in 70 Major Infectious Diseases Table 4.1 Cost-Effectiveness of Approaches to the HIV Care Continuum Region or Cost (C) or cost- Currency as Cost in Intervention Study country Cost per outcome effectiveness (CE) Unit of outcome presented (year) 2012 US$ HIV testing and counseling Grabbe and others Sub-Saharan $26.75 CE per HTC client 2007 US$ $41.81 2010 Africa, Kenya $43.69 CE per new HTC client 2007 US$ $68.28 $189.14 CE per HIV-positive client 2007 US$ $295.59 $237.60 CE per new HIV-positive client 2007 US$ $371.32 identified All mobile compared with $14.91 CE per HTC client 2007 US$ $23.30 stand-alone $16.58 CE per new HTC client 2007 US$ $25.91 $157.21 CE per HIV-positive client 2007 US$ $245.69 $178.10 CE per new HIV-positive client 2007 US$ $278.34 identified Community-site mobile compared $8.82 CE per HTC client 2007 US$ $13.78 with stand-alone $9.73 CE per new HTC client 2007 US$ $15.21 HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations Semimobile container compared $17.23 CE per HTC client 2007 US$ $26.93 with stand-alone $20.06 CE per new HTC client 2007 US$ $31.35 Fully mobile truck compared with $20.38 CE per HTC client 2007 US$ $31.85 stand-alone $23.39 CE per new HTC client 2007 US$ $36.55 VCT Sweat and others Sub-Saharan $12.77 CE per DALY averted 1998 US$ $17.47 2000 Africa, Kenya VCT, enrolled as couples $2.75–$3.48 CE per DALY averted; HIV-1- 1998 US$ $3.47–$4.73 positive males and females $19.48–$21.86 CE per DALY averted; HIV-1- 1998 US$ $26.38– negative females and males $29.60 Free VCT campaign Thielman and Sub-Saharan $5.40 CE per DALY averted 2003 US$ $8.09 others 2006 Africa, Tanzania Sustained free VCT $4.72 CE per DALY averted 2003 US$ $7.07 Scaled-up, community-based VCT Tromp and others East Asia and $9.17 CE per DALY averted 2008 US$ $12.59 2013 Pacific, Indonesia table continues next page 71 72 Major Infectious Diseases Table 4.1 Cost-Effectiveness of Approaches to the HIV Care Continuum (continued) Region or Cost (C) or cost- Currency as Cost in Intervention Study country Cost per outcome effectiveness (CE) Unit of outcome presented (year) 2012 US$ VCT Wang, Moss, and East Asia and 1,087,669 yuan CE Incremental cost per HIV 2002 yuan $228,514.82 Hiller 2011 Pacific, China infection averted (including averted cost) in general population 53,317 yuan CE Incremental cost per DALY 2002 yuan $11,201.68 averted (including averted cost) in general population Stand-alone HTC Menzies and Sub-Saharan $19.26 C Cost per client tested 2007 US$ $25.07 others 2009 Africa, Uganda Hospital-based HTC $11.68 C Cost per client tested 2007 US$ $15.20 Household-member HTC $13.85 C Cost per client tested 2007 US$ $18.03 Door-to-door HTC $8.29 C Cost per client tested 2007 US$ $10.79 Stand-alone HTC $100.59 C Cost per HIV-positive 2007 US$ $130.92 individual identified Hospital-based HTC $43.10 C Cost per HIV-positive 2007 US$ $56.09 individual identified Household-member HTC $231.65 C Cost per HIV-positive 2007 US$ $301.49 individual identified Door-to-door HTC $163.93 C Cost per HIV-positive 2007 US$ $213.35 individual identified HTC in prisons, HIV community Siregar and others East Asia and $23.00 C per VCT in prison 2008 US$ $31.58 clinics, and hospitals 2011 Pacific, Indonesia $39.00 C per VCT in HIV community 2008 US$ $53.56 clinics $65.00 C per VCT in STI community 2008 US$ $89.26 clinics $74.00 C per VCT in hospitals 2008 US$ $101.62 VCT Sweat and others Sub-Saharan $249.00–$346.00 CE per infection averted 1998 US$ $473.46– 2000 Africa, Kenya and $722.66 Tanzania Free VCT Thielman and Sub-Saharan $92.00 CE per infection averted 2003 US$ $137.78 others 2006 Africa, Tanzania table continues next page Table 4.1 Cost-Effectiveness of Approaches to the HIV Care Continuum (continued) Region or Cost (C) or cost- Currency as Cost in Intervention Study country Cost per outcome effectiveness (CE) Unit of outcome presented (year) 2012 US$ VCT at community health centers Tromp and others East Asia and $248.00 CE per HIV infection averted 2008 US$ $340.56 2013 Pacific, Indonesia VCT Wang, Moss, and East Asia and 165,067 yuan C Lifetime cost of HIV care and 2002 yuan $34,679.90 Hiller 2011 Pacific, China treatment per person in MSM population and in general population VCT Venkatesh and India $800.00–$1,900.00 CE per YLS 2010 US$ $912.27– others 2013 $2,166.65 Home-based HTC with enhanced Smith and others Sub-Saharan $1,090.00–$1,360.00 CE per DALY averted 2013 US$ — link to care 2015 Africa, South Africa DNA-PCR for EID compared with Menzies and Sub-Saharan $1,489.00 C per infant correctly diagnosed 2007 US$ $1,937.90 use of rapid HIV tests to screen others 2009 Africa, Uganda and informed of result for out HIV-uninfected infants DNA-PCR versus modified algorithm HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations Improvement of retention and adherence HBC, FBC, and MCC for provision Babigumira and Sub-Saharan $2,615.00 CE per QALY, ICER for MCC 2008 US$ $3,200.86 of ART others 2009 Africa, Uganda versus FBC $2,814.00 CE per QALY, ICER for HBC versus 2008 US$ $3,444.44 FBC $2,241.00 CE per life year, ICER for MCC 2008 US$ $2,743.07 versus FBC $2,251.00 CE per life year, ICER for HBC 2008 US$ $2,755.31 versus FBC Interventions to prevent LTFU: Losina and others Sub-Saharan $1,200–$3,100 depending CE per YLS 2006 US$ $1,489.27– $22/intervention/person/year 2009 Africa, Côte on efficacy in reducing $3,847.28 d’Ivoire LTFU (10%–75%) Interventions to prevent LTFU: $1,500–$4,900 depending CE per YLS 2006 US$ $1,861.59– $41/intervention/person/year on efficacy in reducing $6,081.18 LTFU (10%–75%) table continues next page 73 74 Major Infectious Diseases Table 4.1 Cost-Effectiveness of Approaches to the HIV Care Continuum (continued) Region or Cost (C) or cost- Currency as Cost in Intervention Study country Cost per outcome effectiveness (CE) Unit of outcome presented (year) 2012 US$ Interventions to prevent LTFU: $2,000–$8,400 depending CE per YLS 2006 US$ $2,482.12– $77/intervention/person/year on efficacy in reducing $10,424.88 LTFU (10%–75%) Patient tracer to follow up lost Rosen and Sub-Saharan $432.00 C per patient returned to care 2009 US$ $552.10 patients Ketlhapile 2010 Africa, South Africa Patient tracer to follow up lost $18.00 C per patient attempted to be 2009 US$ $23.00 patients traced in the intervention, including those who could not be found through tracing Peer health workers for Chang and others Sub-Saharan $189.00 C per virologic failure averted 2012 US$ $231.34 adherence monitoring and social 2013 Africa, Uganda $1,025.00 C per LTFU averted 2012 US$ $1,254.64 support Case management to improve Marseille 2011 Sub-Saharan $33.00 C per patient served 2009 US$ $42.11 adherence Africa, Ethiopia Case management to improve $84.00 C per “successful exit” from the 2009 US$ $107.20 adherence case management program Note: — = not available; ART = antiretroviral therapy; DALY = disability-adjusted life year; DNA = deoxyribonucleic acid; EID = early infant diagnostic; FBC = facility-based care; HBC = home-based care; HIV = human immunodeficiency virus; HTC = HIV testing and counseling; ICER = incremental cost-effectiveness ratio; LTFU = loss to follow-up; MCC = mobile clinic care; MSM = men who have sex with men; PCR = polymerase chain reaction; QALY = quality-adjusted life year; STI = sexually transmitted infection; VCT = voluntary counseling and testing; YLS = year of life saved. hospitals, underscoring the importance of identifying A systematic review of linkage of HIV-infected high-risk locations for HIV testing (Siregar and others pregnant women revealed a failure to initiate ART 2011). In India, one-time voluntary HIV testing in the among 38 percent to 88 percent of women known to be population was found to be very cost-effective (ICER eligible (Ferguson and others 2012; Psaros and others of US$1,100 per life year saved). The cost-effectiveness 2015). Although infants born to a known HIV-infected of such screening was even greater in high-prevalence mother should be engaged in care, linking HIV-exposed areas (ICER of US$1,100 per life year saved) and high- infants to appropriate follow-up services has been inad- risk groups (US$800 per life year saved) (Venkatesh equate (Ahmed and others 2013; Chatterjee and others and others 2013). A strategy of annual screening in 2011; Ghadrshenas and others 2013). For children found high-prevalence districts and high-risk groups was also to be HIV-infected, linking with and entry into care is found to be cost-effective, as was screening every five similarly difficult (Phelps and others 2013). years in the national population. Early Infant Diagnosis Guidelines In Uganda, one study assessed the cost-effectiveness of The WHO identifies linking to HIV care as necessary to incorporating initial screening with rapid HIV tests into realize the full health and prevention benefits of ART the conventional testing algorithm of DNA-PCR to (WHO 2013a). However, no consistent guidance exists screen out HIV-uninfected infants (Suthar, Ford, and on the optimal timing from receipt of a positive HIV test others 2013). Costs per infant were US$23.47 for DNA- to linking to HIV care. The International Association of PCR screening, compared with US$7.58 and US$22.75 Physicians in AIDS Care and the Centers for Disease for the modified algorithm that used rapid HIV tests. Control and Prevention recommend that individuals The modified algorithm was significantly less costly for testing HIV-positive be linked to care within three infants older than age three months. Cost-effectiveness months of diagnosis (CDC 2013b; Thompson and was assessed by calculating the incremental cost per others 2012). infected infant correctly diagnosed, with parents informed of results. The conventional algorithm had cost-effectiveness ranging from US$539 to US$7,139 per Measurements of Linkage to Care infant correctly diagnosed and with family informed of Measuring successful linkage from HIV testing to care is result, suggesting that screening infants with rapid HIV often not feasible because HTS typically record aggre- tests before DNA-PCR is potentially cost-effective in gate data on number of tests without individual identi- infants older than age three months. fiers (McNairy and El-Sadr 2012). In addition, some studies report links within 30–90 days of an HIV- positive test, while others do not specify the time LINKING TO AND ENGAGEMENT WITH interval (Medley and others 2013; van Rooyen, Barnabas, HIV CARE and others 2013). Some HIV programs require HIV-positive individu- Rationale and Coverage als to register in the HIV clinic and receive a medical Linking to HIV care and treatment services after a posi- record number and an appointment date for a clinic tive HIV test is a critical, but poorly documented, step in visit, but not necessarily documentation of an encounter the HIV care continuum. The verbal or written referral with a clinician (Elul and others 2014). Other programs process is often insufficient, with significant loss to prefer evidence of receipt of clinical evaluation or a CD4 follow-up (El-Sadr, Gamble, and Cohen 2013; Mugglin cell count test (Rosen and Fox 2011). and others 2012). In a systematic review of 28 studies from Sub-Saharan Africa, a median of 59 percent of patients testing HIV-positive linked to care (Rosen and Barriers Fox 2011). It is important to note that based on the One review of 24 studies (21 of which were from PHIA surveys completed and reported, of all HIV- Sub-Saharan Africa) cited the multiple steps needed to positive individuals identified through the populations enroll HIV-positive individuals into care (Bogart and surveys in Malawi, Zambia, and Zimbabwe, a substantial others 2013; Govindasamy and others 2014). Reasons proportion (87%) indicated that they were receiving reported for not linking include fear that HIV-positive ART, suggesting excellent ART initiation (ICAP 2016). status will be purposefully or inadvertently disclosed However, these data do not inform the time from an (Hatcher and others 2012), fear of discrimination and HIV-positive test to linkage to care and treatment. spousal violence or separation (Bogart and others 2013; HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 75 Gari and others 2013), and distance to the health facility and linkage to care for those found HIV-positive, as well and transportation costs (Bogart and others 2013; as prompt ART initiation (Finocchario-Kessler and Hensen and others 2012; Posse and Baltussen 2009). others 2014). System-level barriers include inconvenient clinic hours; long waiting times; shortages of skilled health Structural-Level Approaches care workers; and delays in CD4 cell count, viral load, Point-of-care (POC) CD4 testing has been shown to and EID results. Furthermore, improperly trained or increase likelihood of timely access to care (Wynberg overworked health care workers and clinics where space and others 2014); for those eligible for ART, it has been limits privacy discourage patients from engaging in care shown to increase likelihood of initiating ART (Faal (Fayorsey and others 2013; Hensen and others 2012; and others 2011; Larson and others 2012; Larson and Posse and Baltussen 2009; Tran and others 2012). others 2013; Patten and others 2013; Wynberg and oth- ers 2014). In South Africa, initiating POC CD4 testing at the time of HIV diagnosis more than doubled the Approaches to Improving Linkage to Care likelihood that patients would initiate ART (Faal and Individual-Level Approaches others 2011). A study from Uganda demonstrated that patients who Many countries, particularly in Sub-Saharan Africa, received extended posttest counseling and monthly visits have successfully decentralized HIV care to the primary by peer support workers were 80 percent more likely care level, reducing transport time and costs for patients than other patients to access HIV care (Muhamadi and (Govindasamy, Ford, and Kranzer 2012; Suthar, Hoos, others 2011). Such programs are encouraged among and others 2013). Task-shifting and task-sharing— populations less likely to access and sustain HIV care allowing trained peer health workers, nurses, and other (Wouters and others 2012). Training for counseling that nonphysician cadres to administer HIV services—has emphasizes linkage could be a simple and feasible enabled decentralization and the scale-up of HIV approach to more effectively engage HIV-infected indi- services. These approaches were implemented in Malawi viduals in care. and Uganda with improved linkage to care and minimal Use of community health workers (CHWs) and increases to costs (Arem and others 2011; McCollum peer educators, who are often themselves HIV-positive, and others 2010). to provide support, guidance, and help with naviga- Colocating HIV testing and care services may also tion to HIV-positive individuals has shown consider- enhance linkage to care (Torian and others 2008). The able promise for improving linkage to care (Ackerman effectiveness of this approach has perhaps been best Gulaid and Kiragu 2012; Hatcher and others 2012; demonstrated with the integration of prevention of Kim and others 2012). A study from the United States mother-to-child transmission into antenatal care services, found that newly diagnosed HIV-positive persons with dramatic increases in enrollment into care for HIV- were more successfully linked to care when supported infected pregnant and breastfeeding women (Ferguson by a case manager (Craw and others 2008). In Kenya, and others 2012). 63.2 percent of patients who received home visits by Home-based services, including HIV testing, POC peers were enrolled in ART within three months CD4 testing, and immediate initiation of ART, may (Kohler and others 2011). Similarly, when CHWs were increase linkage to care, especially in rural areas and in assigned to HIV-infected pregnant women in Malawi settings with high stigma (Helleringer and others 2009; at the time of diagnosis, more than 70 percent of eligi- Lahuerta and others 2013; Myer and others 2013). In ble women and eligible children received ART (Ahmed South Africa, home-based HIV testing, followed by POC and others 2015; Kim and others 2012). CD4 testing, counseling, and referral, was associated with Incentives, including food, conditional cash transfers, 86 percent of patients’ initiating ART within three months and vouchers, have been used to encourage linkage to (van Rooyen, Barnabas, and others 2013). In Malawi, a care (Kundu and others 2012; Solomon and others program that offered self-testing and immediate ART ini- 2014). In India among PWID, modest voucher incentives tiation for those testing HIV-positive showed a significant significantly improved linkage to HIV care (Solomon increase in ART initiation (MacPherson and others 2014). and others 2014). Text messages can help remind patients of appoint- ments, testing, and medication adherence (van Velthoven Cost-Effectiveness Considerations and others 2013). The HIV Infant Tracking System in Individual-Level Approaches Kenya improved linkage to HIV services for HIV- Few of the studies assessing individual-level approaches exposed infants, with increased uptake of EID testing to promoting linkage to care have evaluated their 76 Major Infectious Diseases cost-effectiveness. In a study of nonmonetary incentives that 78 percent to 97 percent of HIV-infected children to promote linkage to care in India, an incentive worth had a CD4+ cell count measured; 63.2 percent to US$4 was effective in increasing ART initiation, suggest- 90.7 percent of children were assessed for ART initiation; ing that relatively low-cost interventions are capable of and 39.5 percent to 99.4 percent of eligible children making a difference in this step of the care cascade started ART (Mugglin and others 2013). Loss to (Solomon and others 2014). follow-up and death are significantly higher among Given that HIV-infected patients gain individual children younger than age one year and among those health benefits and generate positive health externalities with advanced disease (McNairy and others 2013). once they initiate ART, allocating resources to approaches Globally, approximately 1.7 million PWID are living that promote linkage to care has the potential to be more with HIV, only 38 percent of whom are estimated to be cost-effective than allocating resources to approaches receiving ART (WHO 2014c). A systematic review that promote HIV testing in the general population. found that loss to follow up among FSWs was only 6 percent, albeit from few available studies (Mountain Structural-Level Approaches and others 2014). In a study from Zimbabwe, an A study of home-based HIV testing and counseling estimated 50 percent to 70 percent of HIV-infected accompanied by POC CD4 testing and lay counselor FSWs reported being enrolled in HIV care, and only follow-up visits in South Africa reported ICERs of 25 percent to 35 percent accessed ART (Cowan and US$1,090–US$1,360 per DALY averted depending on others 2013). Among MSM in LMICs, data on access to the ART initiation criteria used (Smith and others 2015). HIV treatment remains limited (Arreola and others 2012; UNAIDS 2014). RETENTION IN HIV CARE Guidelines Rationale and Coverage The WHO guidelines highlight the importance of reten- Based on evidence of the benefits of ART when initiated tion in care to enable achievement of viral suppression at early stages of HIV disease, it is anticipated that the (WHO 2016), including for adults, children, adolescents, period from diagnosis to ART initiation will be short- and pregnant women. Strategies to increase retention in ened with adoption of the WHO’s 2016 guidelines for care include community-level interventions for adults universal ART (START Study Group 2015; WHO 2015b). and interventions to enhance retention among pregnant A systematic review of 28 studies from Sub-Saharan women during the postpartum period, highlighting the Africa found that mean retention of adult patients importance of follow-up among caregivers for children before ART initiation was only 46 percent, and mean and development of adolescent-friendly services. retention from determination of ART eligibility to ART initiation was 68 percent (Rosen and Fox 2011). A sys- tematic review indicated that retention on ART among Measurement of Retention in Care adult patients was 80 percent, 70 percent, and 65 percent Retention in care is defined as the proportion of patients at 12, 24, and 36 months, respectively (Fox and Rosen who remain in care as evidenced by a clinical visit or 2010). Loss to follow-up and death were more frequent pharmacy visit within a defined period. For example, for among men, adolescents and young adults, and pregnant an HIV program that recommends a clinical visit every women (DeSilva and others 2009; Lamb and others 3 months, a patient is retained at 12 months if the 2014; Lawn and others 2008; Phillips and others 2014). patient has completed a visit within 3 months of the Retention in care remains a major challenge for pre- scheduled 12-month visit. vention of mother-to-child transmission programs, including those implementing the Option B+ approach— universal treatment for all pregnant and breastfeeding Barriers women. Studies demonstrate significant loss to follow-up Barriers to retention in care are multifactorial (Bogart for pregnant women on ART, especially those newly and others 2013; Geng and others 2010; Ware and others diagnosed during antenatal care, those who are diag- 2013). Structural barriers include financial constraints nosed late in pregnancy, younger women, and those at such as transport costs and lost work wages, long wait earlier HIV disease stages (Haas and others 2016; times and inconvenient clinic hours, mobility to seek Tenthani and others 2014). employment, health care worker attitudes, and perceived A systematic review of eight studies from Sub- low-quality care (Geng and others 2010; Maskew and Saharan Africa, with a total of 10,741 children, reported others 2007). Psychosocial and behavioral barriers HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 77 include anxiety and hopelessness, stigma, lack of (Assefa and others 2012; Brennan and others 2011; perceived severity of HIV disease, lack of social support, Emdin, Chong, and Millson 2013; Fairall and others and reluctance to return after a hiatus from clinic atten- 2012; Iwu and Holzemer 2014; Sherr and others 2010; dance (Wringe and others 2009). Biomedical barriers Shumbusho and others 2009; Thurman and others include inadequate opportunistic infection prevention 2010). Evidence suggests improved retention for patients and management that may hinder clinic attendance and who initiate and maintain ART at primary health facili- contribute to deteriorating health (Brinkhof, Pujades- ties (full decentralization) versus patients who initiate at Rodriguez, and Egger 2009). In a meta-analysis of secondary health facilities and are maintained at pri- 17 studies evaluating loss to follow-up in patients on mary health facilities (partial decentralization) (Auld ART, the most common reasons reported were lack of and others 2015; Reidy and others 2014). money, improving or deteriorating health, and transfer In a study from rural Uganda, provision of US$2.50– to another HIV care site (Brinkhof, Pujades-Rodriguez, US$7.00 to patients on ART to cover transportation costs and Egger 2009). was associated with increased retention at 12 months Women are often lost to care when they return to of between 87 percent and 92 percent (Emenyonu, their home villages or towns for delivery and postpar- Thirumurthy, and Muyindike 2010). Several program- tum care, and postdelivery when they make the tran- matic and research studies are now examining how best sition to routine ART services (Colvin and others to optimize retention of HIV-infected pregnant women 2014; Phillips and others 2015; Schnippel and others (Sturke and others 2014). 2015). Retention is particularly challenging for chil- Patient ART groups in Mozambique and South dren, who depend on caregivers to bring them for Africa, in which one individual is designated to pick up clinic visits. Caregiver fear of disclosing HIV status to medications for the group, showed more than 95 percent the child, unstable family structure, and unsympa- retention in care of patients over 12 months, as well as thetic school environments may lead to loss to favorable longer-term outcomes (Luque-Fernandez and follow-up for children (Busza and others 2014; others 2013; Rasschaert and others 2014). Wachira, Middlestadt, and others 2012). For key populations, intensified posttest counseling Engagement and retention in care are particularly combined with follow-up counseling by CHWs signifi- difficult for key populations because of systematic exclu- cantly increased the proportion that were enrolled sion, social and institutionalized stigma, harassment, and retained in HIV care (WHO 2014b; Wouters and and other psychosocial barriers that discourage engage- others 2012). ment in care after an HIV-positive diagnosis (Baral and others 2012; Mtetwa and others 2013; WHO 2014c). Cost-Effectiveness Considerations Few studies have assessed the cost-effectiveness of Approaches to Improving Retention approaches to improve retention in HIV care (table 4.1). Several interventions have been noted to enhance reten- tion in care. Using Treatment Supporters Provision of free cotrimoxazole improved 12-month The cost-effectiveness of approaches that rely on treat- retention by 20 percent among pre-ART patients in ment supporters has been assessed in South Africa, Kenya (Kohler and others 2011), and food assistance was where the costs of using patient tracers to determine the associated with increased clinic attendance in Haiti status of patients lost to follow-up and to assist patients (Ivers and others 2010). Weekly mobile phone commu- in returning to care were determined (Rosen and nication via text messages to encourage retention is Ketlhapile 2010). Although the average cost per patient being evaluated in an ongoing study in Kenya (van der attempted to be traced in the intervention (including Kop and others 2013). HIV treatment programs that those not found through tracing) was reasonably low at include staff or peer workers who conduct outreach for US$18, because information systems to track deaths and patients who fail to attend clinic visits had higher reten- monitor patients who transferred to other sites were not tion, higher estimated mortality (resulting from more available, the cost of the intervention per patient returned accurate ascertainment of outcomes among those lost to to care was high at US$432. follow-up), and lower loss to follow-up (McMahon and others 2013). Eliminating Patient Costs and Providing Incentives Task-shifting from physician- to nurse-led HIV man- A modeling study estimated the long-term clinical ben- agement has been associated with improved patient efits and cost-effectiveness of retention interventions in retention in several studies from Sub-Saharan Africa Côte d’Ivoire (Losina and others 2009), including 78 Major Infectious Diseases eliminating ART copayments, eliminating charges to three surveys completed in Zimbabwe, Malawi and patients for opportunistic-infection-related drugs, Zambia showed that 88.6 percent of HIV-positive improving personnel training, and providing meals and patients who indicated that they were on ART had viral transportation reimbursements for patients. The inter- suppression. vention costs varied from US$22 per person per year to US$77 per person per year. The results suggest that for a US$22 per person per year intervention that reduces Guidelines loss to follow-up by 10 percent, the cost-effectiveness Guidelines from both the WHO and the International ratio of the intervention (compared to no intervention) Association of Providers of AIDS Care recommend a would be US$3,100 per year of life saved. Using the once-daily, fixed-dose regimen, with the goal of facilitat- WHO threshold for cost-effectiveness of 3 × per capita ing adherence (Thompson and others 2012; WHO GDP, such an intervention would be cost-effective if it 2016). The guidelines also recognize the centrality of had an efficacy of at least 12 percent (WHO 2014a). excellent adherence to the success of ART for individual Similarly, the more costly US$77 per person per year as well as population health and the complexity of main- intervention is also cost-effective, with an efficacy of at taining adherence to lifetime treatment. To support least 41 percent. patient adherence, the WHO recommends implementa- tion of evidence-based interventions, including peer counselors, mobile phone text messages, reminder ADHERENCE TO HIV TREATMENT devices, cognitive behavioral therapy, and behavioral skills training. Rationale and Coverage The clinical effectiveness of ART for individuals and to reduce transmission depends on adherence to treat- Measurement ment (Cohen and others 2011; START Study Group Adherence measures include self-reporting, pill counts, 2015). A meta-analysis published in 2006 found that and pharmacy claims, or more reliably, directly observed adherence among patients on ART in Sub-Saharan therapy (Chaiyachati and others 2011; Kabore and Africa and North America was 77 percent and others 2015; Simoni and others 2006). In research con- 55 percent, respectively (Mills, Nachega, Buchan, and texts, measures include determination of drug concen- others 2006). However, a systematic review of findings tration in blood samples and use of Medication Event from 53 countries indicated that 62 percent of adoles- Monitoring System caps on prescription containers cents and young adults (ages 12–24 years) receiving (Bulgiba and others 2013; Liu, Ma, and Zhang 2010; treatment were at least 85 percent adherent to ART Thompson and others 2012). (Kim and others 2014). Among children, adherence varies considerably by age and medication formulation but has been estimated to be 75 percent in Sub-Saharan Barriers Africa (Vreeman and others 2008). A systematic review A systematic review of patient-related barriers found of 51 studies reporting on adherence during and after that fear of disclosure, stigma, concomitant substance pregnancy found that 77 percent of pregnant women abuse, forgetfulness, suspicions of treatment, regimens had adequate adherence, but adherence decreased that are too complicated, high pill burden, decreased during the postpartum period to 53 percent (Nachega quality of life, work and family responsibilities, food and others 2012). insecurity, and limited access to medication are com- Lastly, a systematic review of HIV-infected PWID monly reported barriers (Mills, Nachega, Bangsberg, and found that ART adherence ranged from 33 percent to others 2006; Young and others 2014). System-level barri- 97 percent in LMICs (Feelemyer and others 2015); ers include lack of awareness about ART, stigma, per- another systematic review determined that 76 percent of ceived high costs for antiretrovirals and related services, FSWs globally adhered to ART (Mountain and others lack of financial means, distance and duration of travel 2014). ART adherence rates among MSM populations in to health providers, lack of consistency and coordination LMICs have not been reliably estimated. across services, limited involvement of the community The lack of broad availability of viral load measure- in the program planning process, poor clinical practices ment in LMICs has limited the ability to assess adher- and health care worker attitudes toward patients, and ence through the effect on viral suppression (Lecher stock outs of antiretroviral drugs (Bezabhe and others and others 2015). The recently conducted PHIA sur- 2014; Coetzee, Kagee, and Vermeulen 2011; Kagee and veys provide encouraging findings. Overall, the first others 2011). HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 79 For pregnant women, additional barriers to adher- greater than 95 percent adherence and viral suppression ence include medication side effects, disparate loca- (Horvath and others 2012). Another meta-analysis of tions for delivery of ART (antenatal care versus ART eight studies reported higher adherence among text clinic), and health worker attitudes (Gourlay and oth- message recipients than among controls (Finitsis, ers 2013; Hodgson and others 2014; Thompson and Pellowski, and Johnson 2014). others 2012). For children and adolescents, barriers Community adherence support delivered by peers— include high pill burden, poorly tolerated formulations, peer educators or patient advocates—improved reten- ART side effects, concerns about stigma and discrimi- tion among both adults and children on ART in South nation, and a lack of youth-friendly clinical services Africa and was associated with decreased mortality (Denison and others 2015; Hudelson and Cluver 2015; (Bemelmans and others 2014; Grimwood and others Lall and others 2015). 2012; Root and Whiteside 2013). Among MSM (Beyrer and others 2010), PWID Studies also report significantly improved viral load (Feelemyer and others 2015), and SWs (Mountain and suppression among patients in HIV programs with peer others 2014), mental illness, stigma and discrimina- workers (Chang and others 2010; Pearson and others tion, lack of confidentiality, health worker discrimina- 2007; Taiwo and others 2010). A randomized controlled tion, violence, and lack of tailored services, as well as study in Rakai, Uganda, observed decreased virologic structural barriers such as social and legal critical failure rates among patients at clinics with peer workers enablers, frequently discourage HIV-infected patients compared with those without peer workers (Chang and from adhering to ART care (Grubb and others 2014; others 2010). WHO 2014b). Decentralizing HIV services from secondary and tertiary health facilities to primary care facilities or community-based adherence clubs has improved virol- Approaches to Enhancing Adherence ogic suppression (Chishinga and others 2014; Grimsrud A systematic review of adherence interventions in and others 2015). Adherence clubs implemented in Cape Sub-Saharan Africa identified six interventions that Town, South Africa, that decentralize care to CHWs and demonstrate efficacy: text messages and other reminder include peer support and self-management features devices, treatment supporters, directly observed ther- demonstrated only 6 percent loss to follow-up, and fewer apy, education and counseling, food supplements, than 2 percent of patients experienced viral rebound and different care-delivery models (Bärnighausen and (Grimsrud and others 2015). others 2011). Interventions have been explored to enhance ART Data on patient-reported barriers to adherence sug- adherence among children and adolescents, including gest that efforts to reduce pill burden (with fixed dose counseling, peer support group therapy, medication combinations) and drug-specific side effects may result diaries, directly observed therapy, and improved in higher adherence (Nachega and others 2014). While antiretroviral formulations (Denison and others 2015). data on other strategies targeting HIV-infected pregnant A qualitative study of HIV-infected adolescents in women are limited, adoption of Option B+ may improve Zimbabwe suggested benefit from support group inter- adherence when it consists of a once-daily, fixed-dose ventions (Mupambireyi and others 2014). A study in combination regimen (Ahmed, Kim, and Abrams 2013; Zambia highlighted the importance of family support Vitalis 2013). and life-skills training to enhance adherence to ART for In Sub-Saharan Africa, two randomized trials have adolescents living with HIV (Denison and others shown that text message reminders to patients to take 2015). The introduction of Option B+ with a simpli- their medication can significantly increase adherence. fied once-daily, fixed-dose combination regimen is The overall effect of text messaging was influenced by expected to improve adherence among pregnant and level of education, gender, and timing and interactivity breastfeeding women (CDC 2013a). of the message (Lester and others 2010; Mbuagbaw and Among PWID, those who receive care in supportive others 2013). A recent review of studies evaluating the environments have ART outcomes similar to outcomes effect of text messaging on ART adherence noted one of non-PWID HIV-infected individuals (Wolfe, Carrieri, study that found that weekly one-way text messages to and Shepard 2010). Creating an enabling environment is patients increased the proportion of patients with greater also critical, including supporting legislation, making than 90 percent adherence, while another study found policy and financial commitments, decriminalizing that weekly two-way messages (that is, messages sent to behaviors of key populations, addressing stigma and the patient with provider follow-up based on the patient’s discrimination, empowering specific communities, and response) increased the proportion of patients with addressing violence against people from key populations 80 Major Infectious Diseases (WHO 2014b). Among PWID, opioid substitution CONCLUSIONS therapy was associated with greater ART adherence, sup- porting the need for integration of drug treatment and The global HIV response is at a critical crossroads. HIV treatment services (Malta and others 2008; Milloy, Although declines in the number of new infections and Montaner, and Wood 2012). Among SWs, interventions in HIV-related mortality have been noteworthy, more similar to those mentioned above but tailored to the remains to be done, both to sustain these gains and to specific needs of this population have been noted to be accelerate epidemic control (Piot and others 2015). effective in improving ART adherence, including adher- Achievement of optimal outcomes for HIV-infected ence counseling and monthly support groups (Graham individuals and for the prevention of transmission to and others 2013; Huet and others 2011; Konate and others is dependent on optimizing every step of the HIV others 2011). care continuum. As described in this chapter, many Food incentives provided at scheduled appointments promising and efficacious approaches exist to address have increased ART adherence and have modestly specific gaps. The findings from the PHIA surveys are enhanced nutritional status (Cantrell and others 2008). encouraging and demonstrate, at least for the first three countries surveyed, good progress toward the UNAIDS 90/90/90 targets, with certain gaps identified particularly Cost-Effectiveness Considerations in terms of reaching the first 90 target, engaging men Only a few approaches to promoting adherence have and adolescents and young adults. However, for key pop- been assessed for cost-effectiveness (table 4.1). A num- ulations, large gaps remain in achieving the 90/90/90 ber of studies that evaluated the efficacy of adherence targets and in addressing the gaps in the HIV care interventions did not include cost-effectiveness analyses continuum (ICAP 2016). (Bärnighausen and others 2011). In South Africa, it has It is important to note that enhancing one step in the been estimated that higher ART adherence can reduce continuum will be insufficient to achieve the overall health care costs, particularly hospitalization costs desired outcome of HIV programs. Thus, research (Nachega and others 2010), suggesting that effective efforts should focus on identifying effective combina- adherence approaches could be highly cost-effective and tions of interventions that target multiple steps along the possibly cost saving as well. continuum. Similarly, research studies need to assess the cost-effectiveness of such interventions and packages of Peer Counseling interventions across the care continuum. Having infor- While the cost-effectiveness of using peer health workers mation on cost-effectiveness is critical to motivating has not been determined in studies, one study has reported policy change and resource mobilization. on the costs of this approach. In Uganda, an approach that Now more than ever, identifying cost-effective meth- used peer health workers to provide clinical and adher- ods that enable the achievement of high service coverage ence monitoring and psychosocial support to patients at and quality is essential to controlling the HIV epidemic. clinics and during monthly home visits cost US$189 per virologic failure averted and US$1,025 per patient loss to follow-up averted (Chang and others 2013). ANNEX The annex to this chapter is as follows. It is available at Decentralized Care http://www.dcp-3.org/infectiousdiseases. A study in Uganda that assessed the cost-effectiveness of facility-based care (FBC), home-based care (HBC), and • Annex 4A. Effectiveness of HIV Interventions mobile clinics indicated that facility-based ART provi- sion was the least costly, and the ICER for mobile clinic care relative to FBC was US$2,615 per quality-adjusted life year (Babigumira and others 2009). The ICER for ACKNOWLEDGMENTS HBC relative to FBC was US$2,814 per quality-adjusted Partial funding support was provided to Wafaa M. El-Sadr life year. Thus, though patient outcomes are often better and Katherine Harripersaud through the National with mobile care and HBC, their costs result in cost- Institutes of Health, National Institute of Allergy and effectiveness ratios that in some countries exceed the Infectious Diseases Cooperative Agreement under coop- threshold of three times per capita GDP. Nevertheless, erative agreement #UM1 AI068619; to Elaine J. Abrams these approaches may be warranted in cases in which from the Eunice Kennedy Shriver National Institute for patient populations reside far from facilities or if the Child Health and Human Development under award costs of these approaches can be reduced. 1RO1HD074558; and to Harsha Thirumurthy from the HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 81 Eunice Kennedy Shriver National Institute for Operational Research: Evaluation of Swaziland’s Hub-and- Child Health and Human Development under award Spoke Model for Decentralizing Access to Antiretroviral K01HD061605. Therapy Services.” Journal of Acquired Immune Deficiency Syndromes 69 (1): e1–12. Babigumira, J. B., A. K. Sethi, K. A. Smyth, and M. E. Singer. 2009. NOTE “Cost Effectiveness of Facility-Based Care, Home-Based Care, and Mobile Clinics for Provision of Antiretroviral World Bank Income Classifications as of July 2014 are as Therapy in Uganda.” Pharmacoeconomics 27 (11): 963–73. follows, based on estimates of gross national income Baeten, J. M., D. Donnell, P. Ndase, N. R. Mugo, J. D. (GNI) per capita for 2013: Campbell, and others. 2012. “Antiretroviral Prophylaxis for HIV Prevention in Heterosexual Men and Women.” New • Low-income countries (LICs) = US$1,045 or less England Journal of Medicine 367 (5): 399–410. • Middle-income countries (MICs) are subdivided: Baggaley, R., B. Hensen, O. Ajose, K. L. Grabbe, V. J. Wong, and (a) lower-middle-income = US$1,046 to US$4,125 others. 2012. “From Caution to Urgency: The Evolution (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 of HIV Testing and Counselling in Africa.” Bulletin of the • High-income countries (HICs) = US$12,746 or more. World Health Organization 90 (9): 652–58B. Baral, S., C. Beyrer, K. Muessig, T. Poteat, A. L. Wirtz, and oth- ers. 2012. “Burden of HIV among Female Sex Workers in Low-Income and Middle-Income Countries: A Systematic REFERENCES Review and Meta-Analysis.” The Lancet Infectious Diseases Ackerman Gulaid, L., and K. Kiragu. 2012. “Lessons Learnt 12 (7): 538–49. from Promising Practices in Community Engagement for Bärnighausen, T., K. Chaiyachati, N. Chimbindi, A. Peoples, the Elimination of New HIV Infections in Children by J. Haberer, and others. 2011. “Interventions to Increase 2015 and Keeping Their Mothers Alive: Summary of a Desk Antiretroviral Adherence in Sub-Saharan Africa: A Systematic Review.” Journal of the International AIDS Society 15 (Suppl 2): Review of Evaluation Studies.” The Lancet Infectious Diseases 17390. 11 (12): 942–51. Ahmed, S., M. H. Kim, and E. J. Abrams. 2013. “Risks and Bateganya, M., O. A. Abdulwadud, and S. M. Kiene. 2010. Benefits of Lifelong Antiretroviral Treatment for Pregnant “Home-Based HIV Voluntary Counselling and Testing and Breastfeeding Women: A Review of the Evidence for (VCT) for Improving Uptake of HIV Testing.” Cochrane the Option B+ Approach.” Current Opinion in HIV and Database of Systematic Reviews 7 (July): CD006493. AIDS 8 (5): 474–89. Bemelmans, M., S. Baert, E. Goemaere, L. Wilkinson, M. Ahmed, S., M. H. Kim, A. C. Dave, R. Sabelli, K. Kanjelo, and Vandendyck, and others. 2014. “Community-Supported others. 2015. “Improved Identification and Enrolment into Models of Care for People on HIV Treatment in Sub- Care of HIV-Exposed and -Infected Infants and Children Saharan Africa.” Tropical Medicine and International Health Following a Community Health Worker Intervention in 19 (8): 968–77. Lilongwe, Malawi.” Journal of the International AIDS Society Beyrer, C., S. D. Baral, D. Walker, A. L. Wirtz, B. Johns, and 18 (1): 19305. others. 2010. “The Expanding Epidemics of HIV Type Ahmed, S., M. H. Kim, N. Sugandhi, B. R. Phelps, R. Sabelli, and 1 among Men Who Have Sex with Men in Low- and others. 2013. “Beyond Early Infant Diagnosis: Case Finding Middle-Income Countries: Diversity and Consistency.” Strategies for Identification of HIV-Infected Infants and Epidemiologic Reviews 32 (June): 137–51. Children.” AIDS 27 (Suppl 2): S235–45. Bezabhe, W. M., L. Chalmers, L. R. Bereznicki, G. M. Arem, H., N. Nakyanjo, J. Kagaayi, J. Mulamba, G. Nakigozi, Peterson, M. A. Bimirew, and others. 2014. “Barriers and others. 2011. “Peer Health Workers and AIDS Care and Facilitators of Adherence to Antiretroviral Drug in Rakai, Uganda: A Mixed Methods Operations Research Therapy and Retention in Care among Adult HIV-Positive Evaluation of a Cluster-Randomized Trial.” AIDS Patient Patients: A Qualitative Study from Ethiopia.” PLoS One Care and STDs 25 (12): 719–24. 9 (5): e97353. Arreola, S., P. Hebert, K. Makofane, J. Beck, and G. Ayala. 2012. Bogart, L. M., S. Chetty, J. Giddy, A. Sypek, L. Sticklor, and oth- Access to HIV Prevention and Treatment for Men Who Have ers. 2013. “Barriers to Care among People Living with HIV Sex with Men: Findings from the 2012 Global Men’s Health in South Africa: Contrasts between Patient and Healthcare and Rights Study (GMHR). Oakland, CA: Global Forum on Provider Perspectives.” AIDS Care 25 (7): 843–53. MSM and HIV (MSMGF). Bourne, C., V. Knight, R. Guy, H. Wand, H. Lu, and oth- Assefa, Y., A. Kiflie, B. Tekle, D. H. Mariam, M. Laga, and ers. 2011. “Short Message Service Reminder Intervention others. 2012. “Effectiveness and Acceptability of Delivery Doubles Sexually Transmitted Infection/HIV Re-Testing of Antiretroviral Treatment in Health Centres by Health Rates among Men Who Have Sex with Men.” Sexually Officers and Nurses in Ethiopia.” Journal of Health Services Transmitted Infections 87 (3): 229–231. Research and Policy 17 (1): 24–29. Brennan, A. T., L. Long, M. Maskew, I. Sanne, I. Jaffray, and Auld, A. F., H. Kamiru, C. Azih, A. L. Baughman, H. Nuwagaba- others. 2011. “Outcomes of Stable HIV-Positive Patients Biribonwoha, and others. 2015. “Implementation and Down-Referred from a Doctor-Managed Antiretroviral 82 Major Infectious Diseases Therapy Clinic to a Nurse-Managed Primary Health Care in Rakai, Uganda: A Cluster-Randomized Trial.” PLoS Clinic for Monitoring and Treatment.” AIDS 25 (16): One 5 (6): e10923. 2027–36. Chatterjee, A., S. Tripathi, R. Gass, N. Hamunime, S. Panha, Brinkhof, M. W., M. Pujades-Rodriguez, and M. Egger. 2009. and others. 2011. “Implementing Services for Early Infant “Mortality of Patients Lost to Follow-Up in Antiretroviral Diagnosis (EID) of HIV: A Comparative Descriptive Treatment Programmes in Resource-Limited Settings: Analysis of National Programs in Four Countries.” BMC Systematic Review and Meta-Analysis.” PLoS One 4 (6): Public Health 11 (July): 553. e5790. Chishinga, N., P. Godfrey-Faussett, K. Fielding, and H. Ayles. Brown, L. B., W. C. Miller, G. Kamanga, N. Nyirenda, 2014. “Effect of Home-Based Interventions on Virologic P. Mmodzi, and others. 2011. “HIV Partner Notification Is Outcomes in Adults Receiving Antiretroviral Therapy in Effective and Feasible in Sub-Saharan Africa: Opportunities Africa: A Meta-Analysis.” BMC Public Health 14 (1): 239. for HIV Treatment and Prevention.” Journal of Acquired Choko, A. T., N. Desmond, E. L. Webb, K. Chavula, S. Napierala- Immune Deficiency Syndromes 56 (5): 437–42. Mavedzenge, and others. 2011. “The Uptake and Accuracy Bulgiba, A., U. Y. Mohammed, Z. Chik, C. Lee, and D. Peramalah. of Oral Kits for HIV Self-Testing in High HIV Prevalence 2013. “How Well Does Self-Reported Adherence Fare Setting: A Cross-Sectional Feasibility Study in Blantyre, Compared to Therapeutic Drug Monitoring in HAART?” Malawi.” PLoS Medicine 8 (10): e1001102. Preventive Medicine 57 (Suppl): S34–36. Coates, T. J., M. Kulich, D. D. Celentano, C. E. Zelaya, Busza, J., E. Dauya, T. Bandason, H. Mujuru, and R. A. Ferrand. S. Chariyalertsak, and others. 2014. “Effect of Community- 2014. “‘I Don’t Want Financial Support but Verbal Support.’ Based Voluntary Counselling and Testing on HIV Incidence How Do Caregivers Manage Children’s Access to and and Social and Behavioural Outcomes (NIMH Project Retention in HIV Care in Urban Zimbabwe?” Journal of the Accept; HPTN 043): A Cluster-Randomised Trial.” The International AIDS Society 17 (1): 18839. Lancet Global Health 2 (5): e267–77. Cantrell, R. A., M. Sinkala, K. Megazinni, S. Lawson-Marriott, Coetzee, B., A. Kagee, and N. Vermeulen. 2011. “Structural S. Washington, and others. 2008. “A Pilot Study of Food Barriers to Adherence to Antiretroviral Therapy in a Supplementation to Improve Adherence to Antiretroviral Resource-Constrained Setting: The Perspectives of Health Therapy among Food-Insecure Adults in Lusaka, Zambia.” Care Providers.” AIDS Care 23 (2): 146–51. Journal of Acquired Immune Deficiency Syndromes 49 (2): Cohen, M. S., Y. Q. Chen, M. McCauley, T. Gamble, M. C. 190–95. Hosseinipour, and others. 2011. “Prevention of HIV-1 CDC (Centers for Disease Control and Prevention). 2013a. Infection with Early Antiretroviral Therapy.” New England “Impact of an Innovative Approach to Prevent Mother-to- Journal of Medicine 365 (6): 493–505. Child Transmission of HIV—Malawi, July 2011–September Colvin, C. J., S. Konopka, J. C. Chalker, E. Jonas, J. Albertini, 2012.” Morbidity and Mortality Weekly Report 62 (8): and others. 2014. “A Systematic Review of Health System 148–51. Barriers and Enablers for Antiretroviral Therapy (ART) ———. 2013b. Recommended Prevention Services: Linkage for HIV-Infected Pregnant and Postpartum Women.” PLoS to and Retention in HIV Medical Care. Atlanta, GA: One 9 (10): e108150. CDC. http://www.cdc.gov/hiv/prevention/programs/pwp Cowan, F. M., S. Mtetwa, C. Davey, E. Fearon, J. Dirawo, /linkage.html. and others. 2013. “Engagement with HIV Prevention Celum, C., J. M. Baeten, J. P. Hughes, R. Barnabas, A. Liu, and Treatment and Care among Female Sex Workers in others. 2013. “Integrated Strategies for Combination HIV Zimbabwe: A Respondent-Driven Sampling Survey.” PLoS Prevention: Principles and Examples for Men Who Have One 8 (10): e77080. Sex with Men in the Americas and Heterosexual African Craw, J. A., L. I. Gardner, G. Marks, R. C. Rapp, J. Bosshart, Populations.” Journal of Acquired Immune Deficiency and others. 2008. “Brief Strengths-Based Case Management Syndromes 63 (Suppl 2): S213–20. Promotes Entry into HIV Medical Care: Results of the Chaiyachati, K., L. R. Hirschhorn, F. Tanser, M. L. Newell, Antiretroviral Treatment Access Study-II.” Journal of and T. Bärnighausen. 2011. “Validating Five Questions of Acquired Immune Deficiency Syndromes 47 (5): 597–606. Antiretroviral Nonadherence in a Public-Sector Treatment Creek, T., A. Tanuri, M. Smith, K. Seipone, M. Smit, and others. Program in Rural South Africa.” AIDS Patient Care and 2008. “Early Diagnosis of Human Immunodeficiency STDs 25 (3): 163–70. Virus in Infants Using Polymerase Chain Reaction on Chamie, G., D. Kwarisiima, T. D. Clark, J. Kabami, V. Jain, and Dried Blood Spots in Botswana’s National Program for others. 2014. “Uptake of Community-Based HIV Testing Prevention of Mother-to-Child Transmission.” Pediatric during a Multi-Disease Health Campaign in Rural Uganda.” Infectious Disease Journal 27 (1): 22–26. PLoS One 9 (1): e84317. Danel, C., R. Moh, D. Gabillard, A. Badje, J. Le Carrou, and Chang, L. W., J. Kagaayi, G. Nakigozi, D. Serwada, T. C. Quinn, others. 2015. “A Trial of Early Antiretrovirals and Isoniazid and others. 2013. “Cost Analyses of Peer Health Worker and Preventive Therapy in Africa.” New England Journal of mHealth Support Interventions for Improving AIDS Care Medicine 373 (9): 808–22. in Rakai, Uganda.” AIDS Care 25 (5): 652–56. Deeks, S. G., S. R. Lewin, and D. V. Havlir. 2013. “The End of Chang, L. W., J. Kagaayi, G. Nakigozi, V. Ssempijja, A. H. Packer, AIDS: HIV Infection as a Chronic Disease.” The Lancet and others. 2010. “Effect of Peer Health Workers on AIDS 382 (9903): 1525–33. HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 83 Denison, J. A., H. Banda, A. C. Dennis, C. Packer, N. Nyambe, and Treatment in Five Sub-Saharan African Countries.” and others. 2015. “‘The Sky Is the Limit’: Adhering to Journal of Acquired Immune Deficiency Syndromes 62 (5): Antiretroviral Therapy and HIV Self-Management from e124–30. the Perspectives of Adolescents Living with HIV and Their Feelemyer, J., D. Des Jarlais, K. Arasteh, and A. Uuskula. 2015. Adult Caregivers.” Journal of the International AIDS Society “Adherence to Antiretroviral Medications among Persons 18 (1): 19358. Who Inject Drugs in Transitional, Low, and Middle Income DeSilva, M. B., S. P. Merry, P. R. Fischer, J. E. Rohrer, C. O. Countries: An International Systematic Review.” AIDS and Isichei, and others. 2009. “Youth, Unemployment, and Behavior 19 (4): 575–83. Male Gender Predict Mortality in AIDS Patients Started on Ferguson, L., A. D. Grant, D. Watson-Jones, T. Kahawita, HAART in Nigeria.” AIDS Care 21 (1): 70–77. J. O. Ong’ech, and others. 2012. “Linking Women Who Test Drake, A. L., A. Wagner, B. Richardson, and G. John-Stewart. HIV-Positive in Pregnancy-Related Services to Long-Term 2014. “Incident HIV during Pregnancy and Postpartum HIV Care and Treatment Services: A Systematic Review.” and Risk of Mother-to-Child HIV Transmission: Tropical Medicine and International Health 17 (5): 564–80. A Systematic Review and Meta-Analysis.” PLoS Medicine Finitsis, D. J., J. A. Pellowski, and B. T. Johnson. 2014. “Text 11 (2): e1001608. Message Intervention Designs to Promote Adherence Dugas, M., E. Bedard, G. Batona, A. C. Kpatchavi, F. A. Guedou, to Antiretroviral Therapy (ART): A Meta-Analysis of and others. 2015. “Outreach Strategies for the Promotion Randomized Controlled Trials.” PLoS One 9 (2): e88166. of HIV Testing and Care: Closing the Gap between Health Finocchario-Kessler, S., B. J. Gautney, S. Khamadi, V. Okoth, Services and Female Sex Workers in Benin.” Journal of K. Goggin, and others. 2014. “If You Text Them, They Will Acquired Immune Deficiency Syndromes 68 (Suppl 2): Come: Using the HIV Infant Tracking System to Improve S198–205. Early Infant Diagnosis Quality and Retention in Kenya.” El-Sadr, W. M., T. R. Gamble, and M. S. Cohen. 2013. “Linkage AIDS 28 (Suppl 3): S313–21. from HIV Testing to Care: A Positive Test Often Leads Fox, M. P., and S. Rosen. 2010. “Patient Retention in Nowhere.” Sexually Transmitted Diseases 40 (1): 26–27. Antiretroviral Therapy Programs up to Three Years on Elul, B., M. Lahuerta, F. Abacassamo, M. R. Lamb, L. Ahoua, Treatment in Sub-Saharan Africa, 2007–2009: Systematic and others. 2014. “A Combination Strategy for Enhancing Review.” Tropical Medicine and International Health Linkage to and Retention in HIV Care among Adults Newly 15 (Suppl 1): 1–15. Diagnosed with HIV in Mozambique: Study Protocol for Gardner, E. M., M. P. McLees, J. F. Steiner, C. Del Rio, and a Site-Randomized Implementation Science Study.” BMC W. J. Burman. 2011. “The Spectrum of Engagement in Infectious Diseases 14 (1): 549. HIV Care and Its Relevance to Test-and-Treat Strategies for Emdin, C. A., N. J. Chong, and P. E. Millson. 2013. “Non- Prevention of HIV Infection.” Clinical Infectious Diseases Physician Clinician Provided HIV Treatment Results in 52 (6): 793–800. Equivalent Outcomes as Physician-Provided Care: A Meta- Gari, S., C. Doig-Acuna, T. Smail, J. R. Malungo, A. Martin- Analysis.” Journal of the International AIDS Society 16 Hilber, and others. 2013. “Access to HIV/AIDS Care: (July): 18445. A Systematic Review of Socio-Cultural Determinants in Emenyonu, N., H. Thirumurthy, and W. Muyindike. 2010. Low- and High-Income Countries.” BMC Health Services “Cash Transfers to Cover Clinic Transportation Costs Research 13 (May): 198. Improve Adherence and Retention in Care in a HIV Geng, E. H., D. Nash, A. Kambugu, Y. Zhang, P. Braitstein, Treatment Program in Rural Uganda.” Prepared for the and others. 2010. “Retention in Care among HIV-Infected Seventeenth Conference on Retroviruses and Opportunistic Patients in Resource-Limited Settings: Emerging Insights Infections, San Francisco, CA, February 16–19. and New Directions.” Current HIV/AIDS Reports 7 (4): Essajee, S., L. Vojnov, M. Penazzato, I. Jani, G. K. Siberry, and 234–44. others. 2015. “Reducing Mortality in HIV-Infected Infants Ghadrshenas, A., Y. Ben Amor, J. Chang, H. Dale, G. Sherman, and Achieving the 90-90-90 Target through Innovative and others. 2013. “Improved Access to Early Infant Diagnosis Approaches.” Journal of the International AIDS Diagnosis Is a Critical Part of a Child-Centric Prevention Society 18 (Suppl 6): 20299. of Mother-to-Child Transmission Agenda.” AIDS 27 Faal, M., N. Naidoo, D. K. Glencross, W. D. Venter, and R. Osih. (Suppl 2): S197–205. 2011. “Providing Immediate CD4 Count Results at HIV Gourlay, A., I. Birdthistle, G. Mburu, K. Iorpenda, and Testing Improves ART Initiation.” Journal of Acquired A. Wringe. 2013. “Barriers and Facilitating Factors to the Immune Deficiency Syndromes 58 (3): e54–59. Uptake of Antiretroviral Drugs for Prevention of Mother- Fairall, L., M. O. Bachmann, C. Lombard, V. Timmerman, to-Child Transmission of HIV in Sub-Saharan Africa: K. Uebel, and others. 2012. “Task Shifting of Antiretroviral A Systematic Review.” Journal of the International AIDS Treatment from Doctors to Primary-Care Nurses in Society 16 (1): 18588. South Africa (STRETCH): A Pragmatic, Parallel, Cluster- Govindasamy, D., N. Ford, and K. Kranzer. 2012. “Risk Factors, Randomised Trial.” The Lancet 380 (9845): 889–98. Barriers, and Facilitators for Linkage to Antiretroviral Fayorsey, R. N., S. Saito, R. J. Carter, E. Gusmao, K. Frederix, Therapy Care: A Systematic Review.” AIDS 26 (16): and others. 2013. “Decentralization of Pediatric HIV Care 2059–67. 84 Major Infectious Diseases Govindasamy, D., J. Meghij, E. Kebede Negussi, R. Clare of Provider-Initiated Testing and Counselling.” Tropical Baggaley, N. Ford, and others. 2014. “Interventions to Medicine and International Health 17 (1): 59–70. Improve or Facilitate Linkage to or Retention in Pre-ART Hochberg, C. H., K. Berringer, and J. A. Schneider. 2015. “Next- (HIV) Care and Initiation of ART in Low- and Middle- Generation Methods for HIV Partner Services: A Systematic Income Settings: A Systematic Review.” Journal of the Review.” Sexually Transmitted Diseases 42 (9): 533–39. International AIDS Society 17 (August): 19032. Hodgson, I., M. L. Plummer, S. N. Konopka, C. J. Colvin, Grabbe, K. L., N. Menzies, M. Taegtmeyer, G. Emukule, E. Jonas, and others. 2014. “A Systematic Review of P. Angala, and others. 2010. “Increasing Access to HIV Individual and Contextual Factors Affecting ART Initiation, Counseling and Testing through Mobile Services in Kenya: Adherence, and Retention for HIV-Infected Pregnant and Strategies, Utilization, and Cost-Effectiveness.” Journal of Postpartum Women.” PLoS One 9 (11): e111421. Acquired Immune Deficiency Syndromes 54 (3): 317. Horvath, T., H. Azman, G. E. Kennedy, and G. W. Rutherford. Graham, S. M., P. Mugo, E. Gichuru, A. Thiong’o, M. Macharia, 2012. “Mobile Phone Text Messaging for Promoting and others. 2013. “Adherence to Antiretroviral Therapy Adherence to Antiretroviral Therapy in Patients with and Clinical Outcomes among Young Adults Reporting HIV Infection.” Cochrane Database of Systematic Reviews High-Risk Sexual Behavior, Including Men Who Have Sex 3: CD009756. with Men, in Coastal Kenya.” AIDS and Behavior 17 (4): Hudelson, C., and L. Cluver. 2015. “Factors Associated with 1255–65. Adherence to Antiretroviral Therapy among Adolescents Gray, R. H., G. Kigozi, D. Serwadda, F. Makumbi, S. Watya, Living with HIV/AIDS in Low- and Middle-Income and others. 2007. “Male Circumcision for HIV Prevention Countries: A Systematic Review.” AIDS Care 27 (7): 805–16. in Men in Rakai, Uganda: A Randomised Trial.” The Lancet Huet, C., A. Ouedraogo, I. Konate, I. Traore, F. Rouet, and 369 (9562): 657–66. others. 2011. “Long-Term Virological, Immunological, and Grimsrud, A., J. Sharp, C. Kalombo, L. G. Bekker, and L. Myer. Mortality Outcomes in a Cohort of HIV-Infected Female 2015. “Implementation of Community-Based Adherence Sex Workers Treated with Highly Active Antiretroviral Clubs for Stable Antiretroviral Therapy Patients in Cape Therapy in Africa.” BMC Public Health 11 (December): 700. Town, South Africa.” Journal of the International AIDS ICAP (International Center for AIDS Care and Treatment Society 18 (May): 19984. Programs). 2016. “The PHIA [Population HIV Impact Grimwood, A., G. Fatti, E. Mothibi, M. Malahlela, J. Shea, and Assessment] Project.” ICAP, Mailman School of Public others. 2012. “Community Adherence Support Improves Health, Columbia University, New York. http://phia.icap Programme Retention in Children on Antiretroviral .columbia.edu/. Treatment: A Multicentre Cohort Study in South Africa.” Ivers, L. C., Y. Chang, J. Gregory Jerome, and K. A. Freedberg. Journal of the International AIDS Society 15 (2): 17381. 2010. “Food Assistance Is Associated with Improved Grubb, I. R., S. W. Beckham, M. Kazatchkine, R. M. Thomas, Body Mass Index, Food Security, and Attendance at E. R. Albers, and others. 2014. “Maximizing the Benefits Clinic in an HIV Program in Central Haiti: A Prospective of Antiretroviral Therapy for Key Affected Populations.” Observational Cohort Study.” AIDS Research and Therapy Journal of the International AIDS Society 17 (July): 19320. 7 (August): 33. Haas, A. D., L. Tenthani, M. T. Msukwa, K. Tal, A. Jahn, and Iwu, E. N., and W. L. Holzemer. 2014. “Task Shifting of HIV others. 2016. “Retention in Care during the First 3 Years of Management from Doctors to Nurses in Africa: Clinical Antiretroviral Therapy for Women in Malawi’s Option B+ Outcomes and Evidence on Nurse Self-Efficacy and Job Programme: An Observational Cohort Study.” The Lancet Satisfaction.” AIDS Care 26 (1): 42–52. HIV 3 (4): e175–82. Kabore, L., P. Muntner, E. Chamot, A. Zinski, G. Burkholder, Hatcher, A. M., J. M. Turan, H. H. Leslie, L. W. Kanya, Z. Kwena, and others. 2015. “Self-Report Measures in the Assessment and others. 2012. “Predictors of Linkage to Care Following of Antiretroviral Medication Adherence: Comparison with Community-Based HIV Counseling and Testing in Rural Medication Possession Ratio and HIV Viral Load.” Journal Kenya.” AIDS and Behavior 16 (5): 1295–307. of the International Association of Providers of AIDS Care Helleringer, S., H. P. Kohler, J. A. Frimpong, and J. Mkandawire. 14 (2): 156–62. 2009. “Increasing Uptake of HIV Testing and Counseling Kagee, A., R. H. Remien, A. Berkman, S. Hoffman, L. Campos, among the Poorest in Sub-Saharan Countries through and others. 2011. “Structural Barriers to ART Adherence in Home-Based Service Provision.” Journal of Acquired Southern Africa: Challenges and Potential Ways Forward.” Immune Deficiency Syndromes 51 (2): 185–93. Global Public Health 6 (1): 83–97. Henley, C., G. Forgwei, T. Welty, M. Golden, A. Adimora, and Kankasa, C., R. J. Carter, N. Briggs, M. Bulterys, E. Chama, others. 2013. “Scale-Up and Case-Finding Effectiveness and others. 2009. “Routine Offering of HIV Testing to of an HIV Partner Services Program in Cameroon: An Hospitalized Pediatric Patients at University Teaching Innovative HIV Prevention Intervention for Developing Hospital, Lusaka, Zambia: Acceptability and Feasibility.” Countries.” Sexually Transmitted Diseases 40 (12): 909–14. Journal of Acquired Immune Deficiency Syndromes 51 (2): Hensen, B., R. Baggaley, V. J. Wong, K. L. Grabbe, N. Shaffer, 202–8. and others. 2012. “Universal Voluntary HIV Testing in Kennedy, C. E., V. A. Fonner, M. D. Sweat, F. Amolo Okero, Antenatal Care Settings: A Review of the Contribution R. Baggaley, and others. 2013. “Provider-Initiated HIV HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 85 Testing and Counseling in Low- and Middle-Income Care: Experiences of Pregnant Women in Rural Uganda.” Countries: A Systematic Review.” AIDS and Behavior 17 (5): Health Policy and Planning 27 (1): 69–75. 1571–90. Lawn, S. D., A. D. Harries, X. Anglaret, L. Myer, and Kim, M. H., S. Ahmed, W. C. Buck, G. A. Preidis, R. Wood. 2008. “Early Mortality among Adults Accessing M. C. Hosseinipour, and others. 2012. “The Tingathe Antiretroviral Treatment Programmes in Sub-Saharan Programme: A Pilot Intervention Using Community Africa.” AIDS 22 (15): 1897–908. Health Workers to Create a Continuum of Care in the Lecher, S, D. Ellenberger, A. A. Kim, P. N. Fonjungo, S. Agolory, Prevention of Mother to Child Transmission of HIV M. Y. Borget, and others. 2015. “Scale-up of HIV Viral (PMTCT) Cascade of Services in Malawi.” Journal of the Load Monitoring—Seven Sub-Saharan African Countries.” International AIDS Society 15 (Suppl 2): 17389. Morbidity and Mortality Weekly Report 64 (46): 1287–90. Kim, S. H., S. M. Gerver, S. Fidler, and H. Ward. 2014. Lester, R. T., P. Ritvo, E. J. Mills, A. Kariri, S. Karanja, and others. “Adherence to Antiretroviral Therapy in Adolescents Living 2010. “Effects of a Mobile Phone Short Message Service with HIV: Systematic Review and Meta-Analysis.” AIDS on Antiretroviral Treatment Adherence in Kenya (WelTel 28 (13): 1945–56. Kenya1): A Randomised Trial.” The Lancet 376 (9755): Kohler, P. K., M. H. Chung, C. J. McGrath, S. F. Benki-Nugent, 1838–45. J. W. Thiga, and others. 2011. “Implementation of Free Liu, X., Q. Ma, and F. Zhang. 2010. “Therapeutic Drug Cotrimoxazole Prophylaxis Improves Clinic Retention Monitoring in Highly Active Antiretroviral Therapy.” Expert among Antiretroviral Therapy–Ineligible Clients in Kenya.” Opinion on Drug Safety 9 (5): 743–58. AIDS 25 (13): 1657–61. Losina, E., H. Toure, L. M. Uhler, X. Anglaret, A. D. Paltiel, Konate, I., L. Traore, A. Ouedraogo, A. Sanon, R. Diallo, and others. 2009. “Cost-Effectiveness of Preventing Loss and others. 2011. “Linking HIV Prevention and Care for to Follow-Up in HIV Treatment Programs: A Côte d’Ivoire Community Interventions among High-Risk Women in Appraisal.” PLoS Medicine 6 (10): e1000173. Burkina Faso: The ARNS 1222 ‘Yerelon’ Cohort.” Journal Lugada, E., J. Levin, B. Abang, J. Mermin, E. Mugalanzi, of Acquired Immune Deficiency Syndromes 57 (Suppl 1): and others. 2010. “Comparison of Home- and Clinic- S50–54. Based HIV Testing among Household Members of Persons Kundu, C. K., M. Samanta, M. Sarkar, S. Bhattacharyya, and Taking Antiretroviral Therapy in Uganda: Results from a S. Chatterjee. 2012. “Food Supplementation as an Incentive Randomized Trial.” Journal of Acquired Immune Deficiency to Improve Pre-Antiretroviral Therapy Clinic Adherence Syndromes 55 (2): 245–52. in HIV-Positive Children: Experience from Eastern India.” Luque-Fernandez, M. A., G. Van Cutsem, E. Goemaere, K. Journal of Tropical Pediatrics 58 (1): 31–37. Hilderbrand, M. Schomaker, and others. 2013. “Effectiveness Lahuerta, M., F. Ue, S. Hoffman, B. Elul, S. G. Kulkarni, and of Patient Adherence Groups as a Model of Care for Stable others. 2013. “The Problem of Late ART Initiation in Sub- Patients on Antiretroviral Therapy in Khayelitsha, Cape Saharan Africa: A Transient Aspect of Scale-Up or a Long- Town, South Africa.” PLoS One 8 (2): e56088. Term Phenomenon?” Journal of Health Care for the Poor MacPherson, P., D. G. Lalloo, A. T. Choko, G. H. Mann, S. B. and Underserved 24 (1): 359–83. Squire, and others. 2012. “Suboptimal Patterns of Provider- Lall, P., S. H. Lim, N. Khairuddin, and A. Kamarulzaman. 2015. Initiated HIV Testing and Counselling, Antiretroviral “Review: An Urgent Need for Research on Factors Impacting Therapy Eligibility Assessment and Referral in Primary Adherence to and Retention in Care among HIV-Positive Health Clinic Attendees in Blantyre, Malawi.” Tropical Youth and Adolescents from Key Populations.” Journal of Medicine and International Health 17 (4): 507–17. the International AIDS Society 18 (2, Suppl 1): 19393. MacPherson, P., D. G. Lalloo, E. L. Webb, H. Maheswaran, Lamb, M. R., R. Fayorsey, H. Nuwagaba-Biribonwoha, V. Viola, A. T. Choko, and others. 2014. “Effect of Optional Home V. Mutabazi, and others. 2014. “High Attrition before and Initiation of HIV Care Following HIV Self-Testing on after ART Initiation among Youth (15–24 Years of Age) Antiretroviral Therapy Initiation among Adults in Malawi: Enrolled in HIV Care.” AIDS 28 (4): 559–68. A Randomized Clinical Trial.” Journal of the American Larson, B. A., K. Schnippel, A. Brennan, L. Long, T. Xulu, and Medical Association 312 (4): 372–79. others. 2013. “Same-Day CD4 Testing to Improve Uptake Malta, M., S. A. Strathdee, M. M. Magnanini, and F. I. Bastos. of HIV Care and Treatment in South Africa: Point-of- 2008. “Adherence to Antiretroviral Therapy for Human Care Is Not Enough.” AIDS Research and Treatment 2013: Immunodeficiency Virus/Acquired Immune Deficiency 941493. Syndrome among Drug Users: A Systematic Review.” Larson, B. A., K. Schnippel, B. Ndibongo, T. Xulu, A. Brennan, Addiction 103 (8): 1242–57. and others. 2012. “Rapid Point-of-Care CD4 Testing at Marseille, E. A. 2011. “Case Management to Improve Adherence Mobile HIV Testing Sites to Increase Linkage to Care: An for HIV-Infected Patients Receiving Antiretroviral Therapy Evaluation of a Pilot Program in South Africa.” Journal of in Ethiopia: A Micro-Costing Study.” Cost Effectiveness and Acquired Immune Deficiency Syndromes 61 (2): e13–17. Resource Allocation 20 (9): 18. Larsson, E. C., A. Thorson, G. Pariyo, P. Conrad, M. Arinaitwe, Maskew, M., P. MacPhail, C. Menezes, and D. Rubel. 2007. and others. 2011. “Opt-Out HIV Testing during Antenatal “Lost to Follow-Up: Contributing Factors and Challenges 86 Major Infectious Diseases in South African Patients on Antiretroviral Therapy.” South Workers: A Systematic Review and Meta-Analysis.” PLoS African Medical Journal 97 (9): 853–57. One 9 (9): e105645. Matovu, J. K., and F. E. Makumbi. 2007. “Expanding Access to Mtetwa, S., J. Busza, S. Chidiya, S. Mungofa, and F. Cowan. 2013. Voluntary HIV Counselling and Testing in Sub-Saharan “‘You Are Wasting Our Drugs’: Health Service Barriers to Africa: Alternative Approaches for Improving Uptake, HIV Treatment for Sex Workers in Zimbabwe.” BMC Public 2001–2007.” Tropical Medicine and International Health Health 13 (July): 698. 12 (11): 1315–22. Mugglin, C., J. Estill, G. Wandeler, N. Bender, M. Egger, and Mbuagbaw, L., M. L. van der Kop, R. T. Lester, H. Thirumurthy, others. 2012. “Loss to Programme between HIV Diagnosis C. Pop-Eleches, and others. 2013. “Mobile Phone Text and Initiation of Antiretroviral Therapy in Sub-Saharan Messages for Improving Adherence to Antiretroviral Africa: Systematic Review and Meta-Analysis.” Tropical Therapy (ART): An Individual Patient Data Meta-Analysis Medicine and International Health 17 (12): 1509–20. of Randomised Trials.” BMJ Open 3 (12): e003950. Mugglin, C., G. Wandeler, J. Estill, M. Egger, N. Bender, McCollum, E. D., G. A. Preidis, M. M. Kabue, E. B. Singogo, and others. 2013. “Retention in Care of HIV-Infected C. Mwansambo, and others. 2010. “Task Shifting Routine Children from HIV Test to Start of Antiretroviral Therapy: Inpatient Pediatric HIV Testing Improves Program Systematic Review.” PLoS One 8 (2): e56446. Outcomes in Urban Malawi: A Retrospective Observational Muhamadi, L., N. M. Tumwesigye, D. Kadobera, G. Marrone, Study.” PLoS One 5 (3): e9626. F. Wabwire-Mangen, and others. 2011. “A Single-Blind McMahon, J. H., J. H. Elliott, S. Y. Hong, S. Bertagnolio, Randomized Controlled Trial to Evaluate the Effect of and M. R. Jordan. 2013. “Effects of Physical Tracing on Extended Counseling on Uptake of Pre-Antiretroviral Care Estimates of Loss to Follow-Up, Mortality, and Retention in in Eastern Uganda.” Trials 12 (July): 184. Low- and Middle-Income Country Antiretroviral Therapy Mupambireyi, Z., S. Bernays, M. Bwakura-Dangarembizi, Programs: A Systematic Review.” PLoS One 8 (2): e56047. and F. M. Cowan. 2014. “‘I Don’t Feel Shy Because I Will McNairy, M. L., and W. M. El-Sadr. 2012. “The HIV Care Be among Others Who Are Just Like Me…’: The Role of Continuum: No Partial Credit Given.” AIDS 26 (14): Support Groups for Children Perinatally Infected with 1735–38. HIV in Zimbabwe.” Children and Youth Services Review McNairy, M. L., M. R. Lamb, R. J. Carter, R. Fayorsey, G. Tene, 45 (October): 106–13. and others. 2013. “Retention of HIV-Infected Children Musheke, M., H. Ntalasha, S. Gari, O. McKenzie, V. Bond, and on Antiretroviral Treatment in HIV Care and Treatment others. 2013. “A Systematic Review of Qualitative Findings Programs in Kenya, Mozambique, Rwanda, and Tanzania.” on Factors Enabling and Deterring Uptake of HIV Testing Journal of Acquired Immune Deficiency Syndromes 62 (3): in Sub-Saharan Africa.” BMC Public Health 13 (March): 220. e70–81. Myer, L., K. Daskilewicz, J. McIntyre, and L. G. Bekker. 2013. Medley, A., M. Ackers, M. Amolloh, P. Owuor, H. Muttai, and “Comparison of Point-of-Care versus Laboratory-Based others. 2013. “Early Uptake of HIV Clinical Care after CD4 Cell Enumeration in HIV-Positive Pregnant Women.” Testing HIV-Positive during Home-Based Testing and Journal of the International AIDS Society 16 (1): 18649. Counseling in Western Kenya.” AIDS and Behavior 17 (1): Nachega, J. B., R. Leisegang, D. Bishai, H. Nguyen, M. Hislop, 224–34. and others. 2010. “Association of Antiretroviral Therapy Menzies, N., B. Abang, R. Wanyenze, F. Nuwaha, B. Mugisha, Adherence and Health Care Costs.” Annals of Internal and others. 2009. “The Costs and Effectiveness of Four HIV Medicine 152 (1): 18–25. Counseling and Testing Strategies in Uganda.” AIDS 23 (3): Nachega, J. B., J. J. Parienti, O. A. Uthman, R. Gross, D. W. 395–401. Dowdy, and others. 2014. “Lower Pill Burden and Once- Milloy, M. J., J. Montaner, and E. Wood. 2012. “Barriers Daily Antiretroviral Treatment Regimens for HIV Infection: to HIV Treatment among People Who Use Injection A Meta-Analysis of Randomized Controlled Trials.” Clinical Drugs: Implications for ‘Treatment as Prevention.’” Current Infectious Diseases 58 (9): 1297–307. Opinion in HIV/AIDS 7 (4): 332–38. Nachega, J. B., O. A. Uthman, J. Anderson, K. Peltzer, Mills, E. J., J. B. Nachega, D. R. Bangsberg, S. Singh, B. Rachlis, S. Wampold, and others. 2012. “Adherence to Antiretroviral and others. 2006. “Adherence to HAART: A Systematic Therapy during and after Pregnancy in Low-Income, Review of Developed and Developing Nation Patient- Middle-Income, and High-Income Countries: A Systematic Reported Barriers and Facilitators.” PLoS Medicine 3 (11): Review and Meta-Analysis.” AIDS 26 (16): 2039–52. e438. Nuti, K. A., J. S. Kabengula, and S. E. Msuya. 2011. “Perceived Mills, E. J., J. B. Nachega, I. Buchan, J. Orbinski, A. Attaran, and Barriers and Attitudes of Health Care Providers towards others. 2006. “Adherence to Antiretroviral Therapy in Sub- Provider-Initiated HIV Testing and Counseling in Mbeya Saharan Africa and North America: A Meta-Analysis.” Journal Region, Southern Highland Zone of Tanzania.” Pan African of the American Medical Association 296 (6): 679–90. Medical Journal 8 (March): 17. Mountain, E., S. Mishra, P. Vickerman, M. Pickles, C. Gilks, Ostermann, J., E. A. Reddy, M. M. Shorter, C. Muiruri, and others. 2014. “Antiretroviral Therapy Uptake, Attrition, A. Mtalo, and others. 2011. “Who Tests, Who Doesn’t, Adherence, and Outcomes among HIV-Infected Female Sex and Why? Uptake of Mobile HIV Counseling and Testing HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 87 in the Kilimanjaro Region of Tanzania.” PLoS One 6 (1): Services in Central Province, Kenya.” Journal of Acquired e16488. Immune Deficiency Syndromes 67 (1): e34–40. Pant Pai, N., J. Sharma, S. Shivkumar, S. Pillay, C. Vadnais, and Root, R., and A. Whiteside. 2013. “A Qualitative Study of others. 2013. “Supervised and Unsupervised Self-Testing Community Home-Based Care and Antiretroviral for HIV in High- and Low-Risk Populations: A Systematic Adherence in Swaziland.” Journal of the International AIDS Review.” PLoS Medicine 10 (4): e1001414. Society 16 (October): 17978. Patten, G. E., L. Wilkinson, K. Conradie, P. Isaakidis, A. D. Rosen, S., and M. P. Fox. 2011. “Retention in HIV Care between Harries, and others. 2013. “Impact on ART Initiation of Testing and Treatment in Sub-Saharan Africa: A Systematic Point-of-Care CD4 Testing at HIV Diagnosis among HIV- Review.” PLoS Medicine 8 (7): e1001056. Positive Youth in Khayelitsha, South Africa.” Journal of the Rosen, S., and M. Ketlhapile. 2010. “Cost of Using a Patient International AIDS Society 16 (July): 18518. Tracer to Reduce Loss to Follow-Up and Ascertain Pearson, C. R., M. A. Micek, J. M. Simoni, P. D. Hoff, Patient Status in a Large Antiretroviral Therapy Program E. Matediana, and others. 2007. “Randomized Control Trial in Johannesburg, South Africa.” Tropical Medicine and of Peer-Delivered, Modified Directly Observed Therapy International Health 15 (Suppl 1): 98–104. for HAART in Mozambique.” Journal of Acquired Immune Roura, M., D. Watson-Jones, T. M. Kahawita, L. Ferguson, Deficiency Syndromes 46 (2): 238–44. and D. A. Ross. 2013. “Provider-Initiated Testing and PEPFAR (President’s Emergency Plan for AIDS Relief). 2013. Counselling Programmes in Sub-Saharan Africa: A PEPFAR Next Generation Indicators Reference Guide. Systematic Review of Their Operational Implementation.” Washington, DC: PEPFAR. AIDS 27 (4): 617–26. Phelps, B. R., S. Ahmed, A. Amzel, M. O. Diallo, T. Jacobs, and Sabapathy, K., R. Van den Bergh, S. Fidler, R. Hayes, and others. 2013. “Linkage, Initiation and Retention of Children N. Ford. 2012. “Uptake of Home-Based Voluntary HIV in the Antiretroviral Therapy Cascade: An Overview.” AIDS Testing in Sub-Saharan Africa: A Systematic Review and 27 (Suppl 2): S207–13. Meta-Analysis.” PLoS Medicine 9 (12): e1001351. Phillips, T., M. L. McNairy, A. Zerbe, L. Myer, and E. J. Schnippel, K., C. Mongwenyana, L. C. Long, and B. A. Larson. Abrams. 2015. “Implementation and Operational Research: 2015. “Delays, Interruptions, and Losses from Prevention Postpartum Transfer of Care among HIV-Infected Women of Mother-to-Child Transmission of HIV Services during Initiating Antiretroviral Therapy during Pregnancy.” Journal Antenatal Care in Johannesburg, South Africa: A Cohort of Acquired Immune Deficiency Syndromes 70 (3): e102–9. Analysis.” BMC Infectious Diseases 15 (1): 46. Phillips, T., E. Thebus, L.-G. Bekker, J. Mcintyre, E. J. Abrams, Shea, S. R. M., S. Makungu, I. Sultan, M. Minde, B. Anosike, and others. 2014. “Disengagement of HIV-Positive Pregnant and others. 2013. “Know Your Child’s Status Testing and Postpartum Women from Antiretroviral Therapy Events: A Targeted Strategy for Paediatric HIV Case Services: A Cohort Study.” Journal of the International AIDS Identification in the Lake Zone of Tanzania.” Prepared Society 17 (1): 19242. for the Seventh International AIDS Conference, Kuala Piot, P., S. S. Abdool Karim, R. Hecht, H. Legido-Quigley, Lumpur, June 30–July 3. K. Buse, and others. 2015. “Defeating AIDS: Advancing Sherr, K. H., M. A. Micek, S. O. Gimbel, S. S. Gloyd, J. P. Hughes, Global Health.” The Lancet 386 (9989): 171–218. and others. 2010. “Quality of HIV Care Provided by Porth, T., C. Suzuki, A. Gillespie, S. Kasedde, and I. Idele. Non-Physician Clinicians and Physicians in Mozambique: 2014. “Disparities and Trends in AIDS Mortality among A Retrospective Cohort Study.” AIDS 24 (Suppl 1): S59–66. Adolescents Living with HIV in Low- and Middle-Income Shumbusho, F., J. van Griensven, D. Lowrance, I. Turate, Countries.” Paper prepared for the Twentieth International M. A. Weaver, and others. 2009. “Task Shifting for Scale-Up AIDS Conference, Melbourne, Australia, July 20–25. of HIV Care: Evaluation of Nurse-Centered Antiretroviral Posse, M., and R. Baltussen. 2009. “Barriers to Access to Treatment at Rural Health Centers in Rwanda.” PLoS Antiretroviral Treatment in Mozambique, as Perceived by Medicine 6 (10): e1000163. Patients and Health Workers in Urban and Rural Settings.” Simoni, J. M., C. R. Pearson, D. W. Pantalone, G. Marks, and AIDS Patient Care and STDs 23 (10): 867–75. N. Crepaz. 2006. “Efficacy of Interventions in Improving Psaros, C., J. E. Remmert, D. R. Bangsberg, S. A. Safren, and Highly Active Antiretroviral Therapy Adherence and HIV-1 J. A. Smit. 2015. “Adherence to HIV Care after Pregnancy RNA Viral Load: A Meta-Analytic Review of Randomized among Women in Sub-Saharan Africa: Falling off the Cliff Controlled Trials.” Journal of Acquired Immune Deficiency of the Treatment Cascade.” Current HIV/AIDS Reports 12 Syndromes 43 (Suppl 1): S23–35. (1): 1–5. Siregar, A. Y., D. Komarudin, R. Wisaksana, R. van Crevel, Rasschaert, F., T. Decroo, D. Remartinez, B. Telfer, F. Lessitala, and R. Baltussen. 2011. “Costs and Outcomes of VCT and others. 2014. “Sustainability of a Community-Based Delivery Models in the Context of Scaling up Services in Anti-Retroviral Care Delivery Model: A Qualitative Research Indonesia.” Tropical Medicine and International Health Study in Tete, Mozambique.” Journal of the International 16 (2): 193–99. AIDS Society 17 (October): 18910. Smith, J. A., M. Sharma, C. Levin, J. M. Baeten, H. van Rooyen, Reidy, W. J., M. Sheriff, C. Wang, M. Hawken, E. Koech, and and others. 2015. “Cost-Effectiveness of Community-Based others. 2014. “Decentralization of HIV Care and Treatment Strategies to Strengthen the Continuum of HIV Care 88 Major Infectious Diseases in Rural South Africa: A Health Economic Modelling Thirumurthy, H., S. H. Masters, S. N. Mavedzenge, S. Maman, Analysis.” The Lancet HIV 2 (4): e159–68. E. Omanga, and others. 2016. “Promoting Male Partner Solomon, S. S., A. K. Srikrishnan, C. K. Vasudevan, S. Anand, HIV Testing and Safer Sexual Decision Making through M. S. Kumar, and others. 2014. “Voucher Incentives Improve Secondary Distribution of Self-Tests by HIV-Negative Linkage to and Retention in Care among HIV-Infected Female Sex Workers and Women Receiving Antenatal and Drug Users in Chennai, India.” Clinical Infectious Diseases Post-Partum Care in Kenya: A Cohort Study.” The Lancet 59 (4): 589–95. HIV 3 (6): e266–74. START Study Group. 2015. “Initiation of Antiretroviral Thompson, M. A., M. J. Mugavero, K. R. Amico, V. A. Cargill, Therapy in Early Asymptomatic HIV Infection.” New L. W. Chang, and others. 2012. “Guidelines for Improving England Journal of Medicine 373 (August): 795–807. Entry into and Retention in Care and Antiretroviral Stephenson, R., C. Rentsch, P. Sullivan, A. McAdams- Adherence for Persons with HIV: Evidence-Based Mahmoud, G. Jobson, and others. 2013. “Attitudes toward Recommendations from an International Association of Couples-Based HIV Counseling and Testing among MSM Physicians in AIDS Care Panel.” Annals of Internal Medicine in Cape Town, South Africa.” AIDS and Behavior 17 (Suppl 1): 156 (11): 817–33, w-284–94. S43–50. Thurman, T. R., L. J. Haas, A. Dushimimana, B. Lavin, and Sturke, R., C. Harmston, R. J. Simonds, L. M. Mofenson, N. Mock. 2010. “Evaluation of a Case Management Program G. K. Siberry, and others. 2014. “A Multi-Disciplinary for HIV Clients in Rwanda.” AIDS Care 22 (6): 759–65. Approach to Implementation Science: The NIH-PEPFAR Torian, L. V., E. W. Wiewel, K. L. Liu, J. E. Sackoff, and T. R. PMTCT Implementation Science Alliance.” Journal of Frieden. 2008. “Risk Factors for Delayed Initiation of Acquired Immune Deficiency Syndromes 67 (Suppl 2): Medical Care after Diagnosis of Human Immunodeficiency S163–67. Virus.” Archives of Internal Medicine 168 (11): 1181–87. Suthar, A. B., N. Ford, P. J. Bachanas, V. J. Wong, J. S. Rajan, and Tran, D. A., A. Shakeshaft, A. D. Ngo, J. Rule, D. P. Wilson, and others. 2013. “Towards Universal Voluntary HIV Testing others. 2012. “Structural Barriers to Timely Initiation of and Counselling: A Systematic Review and Meta-Analysis Antiretroviral Treatment in Vietnam: Findings from Six of Community-Based Approaches.” PLoS Medicine 10 (8): Outpatient Clinics.” PLoS One 7 (12): e51289. e1001496. Tromp, N., A. Siregar, B. Leuwol, D. Komarudin, A. van der Suthar, A. B., D. Hoos, A. Beqiri, K. Lorenz-Dehne, C. McClure, Ven, and others. 2013. “Cost-Effectiveness of Scaling up and others. 2013. “Integrating Antiretroviral Therapy into Voluntary Counselling and Testing in West-Java, Indonesia.” Antenatal Care and Maternal and Child Health Settings: Acta Medica Indonesiana 45 (1): 17–25. A Systematic Review and Meta-Analysis.” Bulletin of the Tumwesigye, E., G. Wana, S. Kasasa, E. Muganzi, and F. Nuwaha. World Health Organization 91 (1): 46–56. 2010. “High Uptake of Home-Based, District-Wide, HIV Sweat, M., S. Gregorich, G. Sangiwa, C. Furlonge, D. Balmer, Counseling and Testing in Uganda.” AIDS Patient Care and and others. 2000. “Cost-Effectiveness of Voluntary HIV-1 STDs 24 (11): 735–41. Counselling and Testing in Reducing Sexual Transmission UNAIDS (Joint United Nations Programme on HIV/AIDS). of HIV-1 in Kenya and Tanzania.” The Lancet 356 (9224): 2014. UNAIDS: The Gap Report. Geneva: UNAIDS. 113–21. ———. 2015a. How AIDS Changed Everything—MDG6: Sweat, M., S. Morin, D. Celentano, M. Mulawa, B. Singh, and 15 Years, 15 Lessons of Hope from the AIDS Response. others. 2011. “Community-Based Intervention to Increase Geneva: UNAIDS. HIV Testing and Case Detection in People Aged 16–32 ______.2015b. “AIDSinfo.” Geneva: UNAIDS. http://aidsinfo Years in Tanzania, Zimbabwe, and Thailand (NIMH Project .unaids.org/. Accept, HPTN 043): A Randomised Study.” The Lancet UNICEF (United Nations Children’s Fund). 2015. “Situation— Infectious Diseases 11 (7): 525–32. Global Statistics Tables: Children and AIDS 2015 Statistical Taiwo, B. O., J. A. Idoko, L. J. Welty, I. Otoh, G. Job, and others. Update.” UNICEF. http://www.childrenandaids.org/situation. 2010. “Assessing the Viorologic and Adherence Benefits of van der Kop, M. L., D. I. Ojakaa, A. Patel, L. Thabane, Patient-Selected HIV Treatment Partners in a Resource- K. Kinagwi, and others. 2013. “The Effect of Weekly Short Limited Setting.” Journal of Acquired Immune Deficiency Message Service Communication on Patient Retention in Syndromes 54 (1): 85–92. Care in the First Year after HIV Diagnosis: Study Protocol Tenthani, L., A. D. Haas, H. Tweya, A. Jahn, J. J. van Oosterhout, for a Randomised Controlled Trial (WelTel Retain).” BMJ and others. 2014. “Retention in Care under Universal Open 3 (6): e003155. Antiretroviral Therapy for HIV-Infected Pregnant and van Rooyen, H., R. V. Barnabas, J. M. Baeten, Z. Phakathi, Breastfeeding Women (‘Option B+’) in Malawi.” AIDS P. Joseph, and others. 2013. “High HIV Testing Uptake 28 (4): 589–98. and Linkage to Care in a Novel Program of Home-Based Thielman, N. M., H. Y. Chu, J. Ostermann, D. K. Itemba, A. HIV Counseling and Testing with Facilitated Referral in Mgonja, and others. 2006. “Cost-Effectiveness of Free HIV KwaZulu-Natal, South Africa.” Journal of Acquired Immune Voluntary Counseling and Testing through a Community- Deficiency Syndromes 64 (1): e1–8. Based AIDS Service Organization in Northern Tanzania.” van Rooyen, H., N. McGrath, A. Chirowodza, P. Joseph, American Journal of Public Health 96 (1): 114–19. A. Fiamma, and others. 2013. “Mobile VCT: Reaching Men HIV Care Continuum in Adults and Children: Cost-Effectiveness Considerations 89 and Young People in Urban and Rural South African Pilot ———. 2013b. “WHO Guidelines Approved by the Guidelines Studies (NIMH Project Accept, HPTN 043).” AIDS and Review Committee.” In HIV and Adolescents: Guidance for Behavior 17 (9): 2946–53. HIV Testing and Counselling and Care for Adolescents Living van Velthoven, M. H., S. Brusamento, A. Majeed, and J. Car. with HIV; Recommendations for a Public Health Approach 2013. “Scope and Effectiveness of Mobile Phone Messaging and Considerations for Policy-Makers and Managers. for HIV/AIDS Care: A Systematic Review.” Psychology, Geneva: WHO. Health, and Medicine 18 (2): 182–202. ———. 2014a. Choosing Interventions that Are Cost-Effective Venkatesh, K. K., J. E. Becker, N. Kumarasamy, Y. M. Nakamura, (WHO-CHOICE). Geneva: WHO. http://www.who.int K. H. Mayer, and others. 2013. “Clinical Impact and Cost- /choice/en/. Effectiveness of Expanded Voluntary HIV Testing in India.” ———. 2014b. Consolidated Guidelines on HIV Prevention, PLoS One 8 (5): e64604. Diagnosis, Treatment, and Care for Key Populations. Geneva: Violari, A., M. F. Cotton, D. M. Gibb, A. G. Babiker, J. Steyn, and WHO. others. 2008. “Early Antiretroviral Therapy and Mortality ———. 2014c. Global Update on the Health Sector Response to among HIV-Infected Infants.” New England Journal of HIV, 2014. Geneva: WHO. Medicine 359 (21): 2233–44. ———. 2015a. Consolidated Guidelines on HIV Testing Services. Vitalis, D. 2013. “Factors Affecting Antiretroviral Therapy Geneva: WHO. Adherence among HIV-Positive Pregnant and Postpartum ———. 2015b. Guideline on When to Start Antiretroviral Women: An Adapted Systematic Review.” International Therapy and on Pre-Exposure Prophylaxis for HIV. Geneva: Journal of STD and AIDS 24 (6): 427–32. WHO. Vreeman, R. C., S. E. Wiehe, E. C. Pearce, and W. M. Nyandiko. ———. 2016. Consolidated Guidelines on the Use of 2008. “A Systematic Review of Pediatric Adherence Antiretroviral Drugs for Treating and Preventing HIV to Antiretroviral Therapy in Low- and Middle-Income Infection: Recommendations for a Public Health Approach. Countries.” Pediatric Infectious Disease Journal 27 (8): 686–91. Geneva: WHO. Wachira, J., S. Kimaiyo, S. Ndege, J. Mamlin, and P. Braitstein. Wolfe, D., M. P. Carrieri, and D. Shepard. 2010. “Treatment and 2012. “What Is the Impact of Home-Based HIV Counseling Care for Injecting Drug Users with HIV Infection: A Review and Testing on the Clinical Status of Newly Enrolled Adults of Barriers and Ways Forward.” The Lancet 376 (9738): in a Large HIV Care Program in Western Kenya?” Clinical 355–66. Infectious Diseases 54 (2): 275–81. Wouters, E., W. Van Damme, D. van Rensburg, C. Masquillier, Wachira, J., S. E. Middlestadt, R. Vreeman, and P. Braitstein. and H. Meulemans. 2012. “Impact of Community-Based 2012. “Factors Underlying Taking a Child to HIV Care: Support Services on Antiretroviral Treatment Programme Implications for Reducing Loss to Follow-Up among Delivery and Outcomes in Resource-Limited Countries: HIV-Infected and -Exposed Children.” Sahara Journal A Synthetic Review.” BMC Health Services Research 12 9 (1): 20–29. (July): 194. Wachira, J., S. Ndege, J. Koech, R. C. Vreeman, P. Ayuo, and Wringe, A., M. Roura, M. Urassa, J. Busza, V. Athanas, and others. 2014. “HIV Testing Uptake and Prevalence among others. 2009. “Doubts, Denial, and Divine Intervention: Adolescents and Adults in a Large Home-Based HIV Understanding Delayed Attendance and Poor Retention Testing Program in Western Kenya.” Journal of Acquired Rates at a HIV Treatment Programme in Rural Tanzania.” Immune Deficiency Syndromes 65 (2): e58–66. AIDS Care 21 (5): 632–37. Wang, S., J. R. Moss, and J. E. Hiller. 2011. “The Cost- Wynberg, E., G. Cooke, A. Shroufi, S. D. Reid, and N. Ford. Effectiveness of HIV Voluntary Counseling and Testing in 2014. “Impact of Point-of-Care CD4 Testing on Linkage to China.” Asia-Pacific Journal of Public Health 23 (4): 620–33. HIV Care: A Systematic Review.” Journal of the International Ware, N. C., M. A. Wyatt, E. H. Geng, S. F. Kaaya, O. O. Agbaji, and AIDS Society 17 (January): 18809. others. 2013. “Toward an Understanding of Disengagement Xia, Y. H., W. Chen, J. D. Tucker, C. Wang, and L. Ling. 2013. from HIV Treatment and Care in Sub-Saharan Africa: “HIV and Hepatitis C Virus Test Uptake at Methadone A Qualitative Study.” PLoS Medicine 10 (1): e1001369. Clinics in Southern China: Opportunities for Expanding WHO (World Health Organization). 2007. Guidance on Detection of Bloodborne Infections.” BMC Public Health Provider-Initiated HIV Testing and Counselling in Health 13 (September): 899. Facilities. Geneva: WHO. Young, S., A. C. Wheeler, S. I. McCoy, and S. D. Weiser. 2014. ———. 2013a. Consolidated Guidelines on the Use of “A Review of the Role of Food Insecurity in Adherence Antiretroviral Drugs for Treating and Preventing HIV to Care and Treatment among Adult and Pediatric Infection: Recommendations for a Public Health Approach. Populations Living with HIV and AIDS.” AIDS and Behavior Geneva: WHO. 18 (Suppl 5): S505–15. 90 Major Infectious Diseases Chapter 5 Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV Charles B. Holmes, Timothy B. Hallett, Rochelle P. Walensky, Till Bärnighausen, Yogan Pillay, and Myron S. Cohen INTRODUCTION ART reduces the sexual transmission of HIV. Final The beneficial effects of antiretroviral therapy (ART) results of this trial with nearly 10,000 person-years of on individual health are well established, and ART is follow-up with similar conclusions were published in widely used to reduce the morbidity and mortality due 2016 (Cohen and others 2016). to the human immunodeficiency virus (HIV) (WHO This emerging evidence stimulated a range of questions 2016). Recent evidence has strengthened the case for regarding the biological mechanisms of HIV treatment as initiating ART as early in the disease stage as possible prevention (TasP), variations in efficacy across subgroups, (Danel and others 2015; INSIGHT START Study differences in at-risk populations, optimal implementa- Group 2015). Similarly, using ART to prevent mother- tion strategies, and potential implications for public health to-child transmission of HIV is supported with con- (Cohen, Holmes, and others 2012; Delva and others 2012). clusive evidence and has been adopted into clinical Recognition of the dual benefits of treatment has resulted policies worldwide, as discussed in chapter 6 of this in the reevaluation of the cost-effectiveness of ART, as well volume (John-Stewart and others 2017). Years of accu- as of the paradigms of HIV prevention (Garnett and mulating biological and observational evidence also others 2017) and has led to policy discussions about how suggest that ART may reduce sexual transmission of best to value the risks and benefits of treatment for per- HIV, although the field lacked conclusive evidence sonal and public health. until recently (Donnell and others 2010; Nachega and Even as substantial research and evaluation have others 2013). improved the understanding of these trade-offs, clinical The evidence base and attention to “treatment as/ and public health policy and funding decisions are being for prevention” strengthened substantially in 2011 with made at the program, national, and global levels. This the interim results from HIV Prevention Trials Network chapter examines the concept of HIV TasP, focusing on (HPTN) 052, a randomized controlled trial of early the underlying biological mechanisms, effectiveness, and versus delayed use of ART among serodiscordant cou- cost-effectiveness of various strategies and settings and ples (Cohen and others 2011). The trial demonstrated assessing how these factors may influence resource allo- a 96 percent reduction in new infections with earlier cation, policy decisions, and research agendas at the initiation of ART and provided strong evidence that national and global levels. Corresponding author: Charles B. Holmes, Johns Hopkins University, Baltimore, Maryland, United States, and Centre for Infectious Disease Research in Zambia (CIDRZ); Charles.Holmes@cidrz.org. 91 THE BIOLOGY OF TRANSMISSION Susceptibility to infection varies greatly. The only proven relative immunity to HIV results from deletion HIV is transmitted in three ways: from parenteral expo- in the CCR5 receptor, which has been observed in sure to contaminated blood and blood products; from about 1 of 100 Caucasians and less commonly in non- exposure of many mucosal sites to infected genital Caucasians (O’Brien and others 2000). Many studies secretions; and from mother to baby before, during, or have tried to define factors that allow some people to after delivery (Royce and others 1997). These routes of remain exposed and uninfected (McLaren and others transmission have been studied extensively and found 2013). By definition, all HIV-discordant couples include to have different probabilities of transmission, given a partner who is exposed and uninfected (Muessig and exposure. In each case, the biology of transmission is Cohen 2014). Yet, many people in this group will believed to be defined by the infectiousness of the host become infected. and the susceptibility of the person exposed (Cohen There is little evidence to suggest that innate immu- and others 2010; Cope and others 2014; Pilcher and nity, antibodies, or T-cell responses provide durable or others 2004). reliable resistance to HIV. More likely, “apparent resis- HIV-infected fluids contain cells infected and not tance” reflects the absence of factors that amplify trans- infected with HIV. The replication of HIV generates a mission (Pilcher and others 2007). Inflammation from very large number of viral variants—viruses that have any source will cause defects in the mucosa, evoke a large different nucleic acid sequences—that constitute an number of receptive cells, increase the number of recep- infectious swarm. Within the swarm, some viruses are tors expressed, and activate cells that favor HIV replica- capable of producing infection; others are defective or tion. For example, bacterial vaginosis characterized by a less fit for transmission (Ho and others 2013). The like- change in vaginal flora and watery discharge is strongly lihood that HIV will cause infection is governed by the associated with HIV acquisition in women (Taha and number of viruses—the inoculum (Baeten and others others 1998). Unique cytokine profiles may also favor 2011; Donnell and others 2010; Laeyendecker and others these conditions (Olivier and others 2014). 2012; Quinn and others 2000)—and the genotypic and TasP uses ART to reduce the replication of HIV in phenotypic characteristics of HIV in the swarm (Martin the blood and mucosal secretions profoundly, quickly, and others 2014). The probability of HIV transmission and reliably. The hypothesis that treatment could serve in heterosexual couples directly reflects the concentra- as prevention began as soon as the first ART was devel- tion of HIV in the fluid studied (Baeten and others 2011; oped (Henry 1988). Numerous groups have since Quinn and others 2000). In a landmark study of demonstrated the ability of ART to penetrate and sup- heterosexual transmission, Quinn and others (2000) press viral replication in the male and female genital observed no transmission when the blood plasma viral tract (Thompson, Cohen, and Kashuba 2013). Many load was less than 1,500 copies per milliliter, and the antiretroviral agents achieve similar or higher concen- most transmission when the viral load was more than trations in the genital tract as in blood (Kwara and 37,500 copies per milliliter. Unprotected anal intercourse others 2008). However, the body’s ability to metabolize appears to have high risk of transmission per contact and eliminate the medications may compromise the (Baggaley, Dimitrov, and others 2013), explaining the prevention benefits of treatment, and viruses isolated high incidence of HIV in men who have sex with men from areas of low or variable drug levels have demon- (MSM), but the viral load required for transmission by strated site-specific resistance. this route has not been determined. To date, few cases of HIV transmission have been The phenotype of the founder viruses that initiate documented when a person with HIV has been treated infection helps determine the probability of HIV trans- sufficiently to prevent viral replication (Cohen and mission above and beyond the inoculum effects (Carlson others 2016; Rodger and others 2016). However, HIV and others 2014; Parrish and others 2013; Ping and can be found in the male and female genital tract secre- others 2013). HIV variants that cause infection are dual tions even when HIV is suppressed in the blood tropic—that is, they use both CD4 and CCR5 receptors (Anderson and Cu-Uvin 2011; Reichelderfer and (Joseph and others 2014; Ping and others 2013; Shaw others 2000). The implication is that the HIV detected and Hunter 2012). Only one to three founder viruses are in the genital tract under these conditions may be generally transmitted, and the number of variants may defective and incapable of causing infection (Zhang reflect the route of exposure (Keele and others 2008). and others 1998). Transmission from penile-vaginal exposure has the An additional concern is the pharmacology of ART fewest variants, followed by anal exposure, followed by in the genital tract, which is considered a special com- parenteral exposure (Li and others 2010). partment. Studies of the female genital tract 92 Major Infectious Diseases and colorectum have noted that the concentrations of partner with a viral load less than 400 copies per millili- tenofovir and emtricitabine and their active metabo- ter was zero for persons on ART and 0.16 per 100 per- lites vary according to the type of mucosal tissue son-years for persons not on ART, based on five studies (Patterson and others 2011). Differential penetration and one event. A meta-analysis of studies of serodiscor- or metabolism of ART offers further insight into the dant heterosexual couples where the HIV-positive part- highly variable level of protection conferred by these ner was on ART and virally suppressed found zero agents (Hendrix and others 2008). The results from transmissions per 100 person-years (Loutfy and others preexposure prophylaxis (PrEP) clinical trials suggest 2013); a similar review of partners on combination ART that the use of ART for prevention can be optimized by for at least six months found a transmission risk of choosing agents that (1) preferentially penetrate sites between 1 and 13 per 100,000 sex acts. Another of HIV acquisition or transmission or (2) have a long meta-analysis of 50 publications found a 91 percent tissue half-life that might provide a pharmacologic (79 percent to 96 percent) reduction in incidence of buffer for imperfect drug adherence. In summary, HIV-1 per partner among couples when the index case extensive studies of HIV transmission have been com- used ART (Baggaley, White, and others 2013). Supervie pleted, and the results help illuminate the understand- and others (2014) reported at most one HIV transmis- ing of the ways to use ART to maximize the prevention sion over an estimated 113,480 sex acts—of which of transmission. 17 percent were not condom protected—among 1,672 serodiscordant couples where the index partner had been treated for more than six months. EVIDENCE OF THE EFFECTIVENESS OF The PARTNER Study is assessing the occurrence of TREATMENT AS PREVENTION linked transmission among serodiscordant heterosexual and MSM couples who have condomless sex, are not tak- Four lines of complementary evidence support the idea ing PrEP, and have a recent viral load of less than 200 cells that treatment in HIV-infected individuals reduces their per cubic millimeter (cells/mm3). No linked transmis- transmission of HIV to others: sions among MSM couples were observed among 1,238 couple-years of follow-up (Rodger and others • Observational studies of serodiscordant couples 2016), implying that ART treatment prevented transmis- • A randomized controlled trial sion of HIV during unprotected anal intercourse. • Ecologic studies • Population-based studies. The HPTN 052 Randomized Controlled Trial HPTN 052 was a randomized controlled trial designed Observational Studies of Serodiscordant Couples to provide an understanding of the magnitude and dura- As shown in annex 5A, 14 observational studies of sero- bility of ART for prevention. The study enrolled 1,562 discordant couples have been reported (Muessig and serodiscordant couples at 13 sites in nine countries in Cohen 2014). In 11 of these, ART was associated with the Africa, Asia, and the Americas; it randomized infected prevention of HIV transmission. The two studies from men and women to start ART at CD4 T-cell counts of China failed to note a prevention benefit from ART (Lu 200–250 cells/mm3, compared with subjects who started and others 2010; Wang and others 2010). A larger retro- ART at CD4 T-cell counts of 350–550 cells/mm3 (median spective analysis of 38,862 serodiscordant heterosexual cell count of 446 cells/mm3). All participants were couples across China noted a 26 percent relative reduc- offered couples counseling for prevention. In those tion in transmission when the index case received ART receiving delayed ART, the counseling itself appeared to (Jia and others 2012). In most of these studies (including reduce HIV transmission to levels far lower than in ear- studies in China), it is not known either whether the lier studies (less than 2 percent per year). However, the HIV-infected person receiving ART was actually using addition of early ART led to a 96 percent prevention of the agents prescribed or what degree of viral suppression HIV transmission compared with delayed ART in an was achieved. interim analysis. Infected subjects who were treated ear- Several systematic reviews of TasP studies have been lier not only had CD4 T-cell counts that rose quickly but conducted. Attia and others (2009) reviewed 11 cohorts also developed fewer infections (Grinsztejn and others reporting on 5,021 heterosexual couples and 461 trans- 2014). After 8,494 person-years of follow-up, early ART mission events. The transmission rate overall from maintained 93 percent effectiveness in the prevention of patients on ART was 0.46 per 100 person-years, based on new linked infections compared with delayed ART five events. The transmission rate from a seropositive (Cohen and others 2015; Cohen and others 2016). Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 93 Ecological and Population-Based Studies may have less access to health care and a greater risk of As shown in annex 5B, a large number of ecological acquiring HIV (Lopez-Quintero, Shtarkshall, and studies demonstrate the ability of ART to reduce the Neumark 2005; Spielberg and others 2003). Another incidence of HIV (Smith and others 2012). Most of these approach is to use back-calculation from new diagno- are from North America (Castel and others 2012; Das ses (Fang and others 2004), although this approach and others 2010; Katz and others 2002; Montaner and relies on assumptions related to disease progression others 2010; Porco and others 2004; Wood and others markers such as the onset of symptoms and decline in 2009); one is from Taiwan, China (Fang and others CD4 T-cell count to estimate the time of infection 2004); and one is from Australia (Law and others 2011). (Holmes and others 2006; Novitsky and others 2010; Each study used an ecological measure of exposure Wand and others 2009; Wolbers and others 2010). (access to ART), outcome (HIV incidence), or both. The Longitudinal cohort follow-up data have also been reliability of the results lies in the strength of the mea- used to define population incidence and are consid- surements used for exposure and outcome. ered the gold standard of HIV incidence estimation, The exposure of the entire HIV-infected population despite well-known sources of bias (Porco and others to ART can only be measured if every person infected 2004; Wood and others 2009). In a striking example of with HIV can be identified and their treatment and the power of cohort studies, Tanser and others (2013) virological suppression status assessed. Indeed, the enrolled 16,000 HIV-negative people from 2005 to hypothesis that use of ART by the entire population 2011 to receive HIV antibody testing every six months. infected will decrease HIV incidence assumes that An HIV-negative individual living in a community ongoing care will sustain viral suppression, thereby with 30 percent to 40 percent ART coverage was preventing transmission (Cohen and others 2011; 38 percent less likely to acquire HIV than a person Walensky and others 2010). However, in some settings, living in a community with less than 10 percent ART substantial numbers of HIV-infected persons are lost coverage. As noted above, no change in cohort inci- to follow-up along the path from testing to suppressive dence was observed in a cluster randomized controlled treatment (Gardner and others 2011). In the first trial in the same area (Iwuji, Orne-Gliemann, Balestre, population-based ART randomized controlled trial and others 2016). The likely reason for the failure of completed (AAAS 2016; Iwuji, Orne-Gliemann, the trial to show effectiveness was that ART coverage Larmarange, and others 2016), individuals living in was nearly the same in both the intervention and the communities in KwaZulu-Natal, South Africa, receiv- control arms of the trial. Links to care were low, and ing “immediate ART” irrespective of CD4 T-cell count the TasP intervention generally did not induce more did not have lower incidence of HIV than those in people in the intervention arm to take up ART. In con- control (standard of care) communities; however, indi- trast, in the population-based cohort study by Tanser viduals in the immediate ART communities did not and others (2013), ART coverage across different geo- have the anticipated uptake and benefits of ART graphic communities ranged from less than 10 percent because of the difficulty of implementing this strategy. to 30 percent to 40 percent. Several other community randomized TasP trials are Laboratory assays to identify persons with recent underway (Boily and others 2012). HIV infection can be applied to stored biospecimens collected in the course of routine surveillance or epi- demiological research studies. The serologic testing Challenges in the Measurement of Population-Level algorithm for recent HIV seroconversion derives HIV HIV Incidence incidence based on differences in antibodies gener- The ability to detect a benefit of TasP depends on the ated in the weeks after infection (Janssen and others ability to detect changes in HIV incidence. Widely dif- 1998), although logistical challenges in storing and ferent methods have been developed to measure HIV tracking remnant blood can affect the completeness incidence. Perhaps the most commonly used approach of data (Das and others 2010; Katz and others 2002). estimates population-based incidence using informa- Even relatively new laboratory methods misclassify tion on newly identified cases as a proxy for new established and early infections (Le Vu and others infections (Castel and others 2012; Das and others 2008), but other methods are in development (Burns 2010; Montaner and others 2010). Newly diagnosed and others 2014). Currently, surveillance for recent patients acquired HIV at some unknown earlier time, infections in low- and middle-income countries and they are not “incident” in the traditional sense. (LMICs) is limited, which further constrains the abil- Using new diagnoses as a proxy for incidence also ity to track the effect of intervention scale-up on the misses people who do not seek testing; these people incidence of recent infections. However, successful 94 Major Infectious Diseases development of serological detection of incidence eligibility to targeted groups produced small declines in infection would allow cross-sectional detection of HIV incidence: 2 percent in FSWs, 5 percent in MSM, incident HIV infection either in stored samples or in and 5 percent in PWIDs; in contrast, expanding eligibility demographic surveys. to all adults and prioritizing access for all three key pop- ulations resulted in a 30 percent cumulative decline in new infections (Eaton and others 2012). Modeling Population-Level Prevention Effectiveness A model assessing the impact of a test-and-treat strat- Mathematical modeling has been used extensively to egy for urban MSM in New York City estimated a reduc- gauge the potential of ART to reduce or eliminate the tion in new cases of 39.3 percent over 20 years. The annual spread of HIV, and virtually all models report a benefit testing component of this approach provided the major- from ART; the magnitude of the benefit reflects the ity of the projected impact, whereas earlier treatment degree of coverage, model assumptions, and program (at CD4 T-cell counts of less than 500 cells/mm3) itself quality issues such as retention and adherence (Cohen contributed to an 8.5 percent decline in new infections and others 2013; Maddali and others 2015). In a power- over 20 years (Sorensen and others 2012). A test-and-treat ful and controversial analysis of the South African HIV strategy for adults with HIV in British Columbia using a epidemic, Granich and others (2009) projected that model specifically built to include the main drivers of the massive expansion of testing and treatment (“test and local epidemic demonstrated 37 percent to 62 percent treat”) along with best case program quality could sub- reductions in new infections over 25 years (Lima and stantially reduce and potentially eliminate the HIV epi- others 2008). demic in South Africa within 10 years. Wagner and The effectiveness of TasP will be highly dependent on Blower (2012) also demonstrated the theoretical possi- the elements of the HIV cascade, as outlined in detail in bility of HIV epidemic elimination in South Africa using chapter 4 of this volume (Harrispersaud and others a test-and-treat approach. However, they reported that it 2017), including testing frequency and coverage, links to would take 40 years, and the cumulative costs would be care, adherence to treatment, virological suppression, much higher. The differences were partly attributable to and long-term retention in care (Delva and others 2012). differing model assumptions about survival time on Maddali and others (2015) projected that moving to ART and the costs of ART over time. A modeled analysis early treatment in India resulted in a reduction from an of expanded testing and treatment regardless of CD4 estimated 1,285,000 new HIV infections to 1,050,000 T-cell count in Washington, DC, found a more modest infections under existing program conditions over a impact on HIV transmission (Walensky and others 20-year period. However, with enhanced testing, links to 2010). In a comparison of 12 independent mathematical care, and retention in care, the projected number of new models, Eaton and others (2012) reported broad agree- infections with early treatment was projected to fall fur- ment regarding the substantial potential to reduce HIV ther to 517,000. As pointed out by Wilson and Fraser incidence in generalized epidemics in Sub-Saharan (2014), country-level data on virological suppression are Africa, despite large differences in the structures of the variable. For example, 26.1 percent of 266 individuals models. For example, in South Africa and Zambia, reporting ART use in the 2012 Kenya AIDS Indicator expanding ART eligibility to all HIV-positive adults was Survey were found to have a detectable viral load greater projected to avert 9 percent to 40 percent of new infec- than 550 copies (Cherutich and others 2016; National tions over a 20-year time horizon, with greater reduc- AIDS and STI Control Programme 2013). tions attributed to strategies involving increased testing Efforts to project the impact of HIV TasP strategies and links to care. have highlighted the importance of understanding the Multiple investigators have also modeled the effects of relative infectiousness of people with acute and early various strategies incorporating TasP in concentrated infection. HIV transmission is more efficient during epidemics in which the HIV epidemic has the largest acute infection, reflecting higher viral loads and pheno- burden among specific populations, such as persons who typic factors that favor transmission (Cohen, Dye, and inject drugs (PWIDs), MSM, and female sex workers others 2012). A study in Uganda reported that people (FSWs) (Boily and Shubber 2014). In an analysis of the with acute and early infection are 26 times more likely epidemic in Belgaum, India, the expansion of eligibility to to transmit HIV than people with established infection all FSWs resulted in a 13 percent decline in projected HIV (Hollingsworth, Anderson, and Fraser 2008). Viral phy- infections (Eaton and others 2012). Expanding eligibility logenetic results suggest that acute and early infections to all HIV-positive adults, in conjunction with prioritized are responsible for one-third to one-half of new HIV access for FSWs, resulted in 29 percent to 41 percent of cases in MSM (Brenner, Wainberg, and Roger 2013; new HIV infections being averted. In Vietnam, expanding Rieder and others 2010). A modeling study by Eaton Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 95 and Hallett (2014) reported that a higher proportion of In summary, the effectiveness of TasP has been well early infection lessened the impact of ART on esti- established through observational studies and clinical mated incidence in the first year in South Africa but did trials. Ecological studies and projection models further not have an important influence on the long-term demonstrate the substantial potential for population- effect on incidence (figure 5.1). Powers, Kretzschmar, level HIV prevention from expanded treatment across a and Miller (2014) have challenged the conclusions of wide variety of geographies, epidemic types, and popu- this report, and the contribution of people with acute lations. Ongoing population-based studies will further and early infection to the spread of HIV continues to be evaluate the validity of these models and provide addi- debated. The debate turns on numerous assumptions, tional evidence on the impact of TasP strategies and the including levels and distribution of risk behavior and real-world effects of variable program quality along the epidemic patterns. HIV treatment cascade. Regardless of acute infection’s potential impact, diag- nosing and linking people to care as early as possible are crucial. However, it is difficult to detect and diagnose EVALUATING COST-EFFECTIVENESS people with acute and early infection. When acutely infected patients are identified, U.S. guidelines recom- Metrics of Cost-Effectiveness mend immediate treatment to preserve CD4 T-cell Given the effectiveness of treatment for reducing the count, shrink the viral reservoir, and reduce HIV trans- sexual transmission of HIV, it is increasingly important mission (DHHS 2014). The World Health Organization for policy makers to consider the cost-effectiveness of (WHO) has not yet issued specific guidelines related to TasP. Accordingly, analysts have begun to grapple with acute infection, although it does recommend treatment how best to represent the range of effects of ART. for all HIV-infected individuals (WHO 2016). In its simplest form, a narrow definition of cost- effectiveness has been represented as the incremental cost per infection averted by ART (Bärnighausen, Figure 5.1 Reduction in HIV Incidence Rate after 30 Years, by Salomon, and Sangrujee 2012; Ying and others 2015). Relative Infectiousness during Early Infection However, this outcome alone does not value ART’s long-term health and health-related quality-of-life 40 R 2 = 0.00 effects in the denominator of the cost-effectiveness ratio as recommended by consensus guidelines for cost- 30 effectiveness (Gold and others 1996; Weinstein and others 1996). A trial-based cost-effectiveness analysis of Reduction in HIV incidence rate after 30 years, % 20 HPTN 052 and other analyses used joint measures—for example, life-years saved, disability-adjusted life years 10 (DALYs) averted—to value the impacts on both health and prevention (Eaton and others 2014; Walensky and 7x 10x 15x 20x 30x 50x others 2013). In this construct, life-years saved or DALYs Relative infectiousness during early infection (log scale) averted by ART include both direct effects on health and downstream (discounted) effects on the prevention of 40 new infections. Other analysts have reported the cost per death averted, which similarly values deaths directly 30 averted by the therapeutic and preventive effects of ART, although this approach is less common and does 20 not fully account for health-related quality of life (Bärnighausen, Bloom, and Humair 2012). 10 Cost-Effectiveness Estimates of Treatment No increase 9.2x increase 26x increase First-generation studies evaluated the cost-effectiveness Relative infectiousness during early infection of ART versus no ART and generally did not include the Source: Eaton and Hallet 2014. prevention effects of treatment. Second-generation Note: HIV = human immunodeficiency virus; x = times. This model showed no effect of relative cost-effectiveness studies of ART examine circum- infectiousness during early infection on the long-term (30-year) reduction in HIV incidence projected with scale-up of treatment as prevention. Relative infectiousness is increased 9.2 times and 26 stances in which ART is widely used for its health bene- times, neither of which had an effect on projected long-term incidence reduction. fits. Instead of comparisons with no ART, these studies 96 Major Infectious Diseases look at expanding ART to various groups and include of testing and links to the health system. Compared with the effect of ART on sexual transmission (table 5.1) existing conditions, these scenarios were very Most of these studies model the cost-effectiveness of cost-effective over 20 years. However, they were less earlier initiation of treatment compared with later initi- cost-effective over a shorter time horizon, in part because ation and typically include scenarios in which treat- the effect of ART on HIV transmission is initially small ment is started at CD4 T-cell counts greater than but increases (Eaton and others 2014). 500 cells/mm3 or is started promptly regardless of CD4 Walensky and others (2013) reported on the cost- T-cell count; in contrast, the 2010 WHO guidelines effectiveness of earlier treatment in a trial-based analysis (WHO 2010) recommend beginning treatment at CD4 focused on earlier treatment of serodiscordant couples T-cell count of 350 cells/mm3. Some analyses also in South Africa. Treatment of all discordant couples was include related interventions, such as expanded testing cost saving in South Africa over five years; over a lifetime, and links to care, and compare ART with other preven- it cost US$590 per life-year saved in South Africa and tion modalities, such as PrEP and voluntary medical US$530 per life-year saved in India, both considered very male circumcision (VMMC). cost-effective. Another trial-based analysis of earlier These second-generation studies generally demon- treatment in serodiscordant couples in Uganda reported strate more favorable cost-effectiveness than previous a cost per DALY averted of US$1,075 over 10 years (Ying analyses. In a cost-effectiveness analysis that considered and others 2015). the sexual prevention–related effects of ART, Long, In the concentrated epidemic setting of India, Brandeau, and Owens (2010) estimated that expanded Maddali and others (2015) and Eaton and others (2014) treatment using prevailing eligibility criteria of CD4 reported favorable cost-effectiveness ratios for broader T-cell levels < 200 cells/mm3 was very cost-effective in strategies of earlier treatment, ranging from US$199 per the United States and that increased frequency of HIV DALY averted (Eaton and others 2014) to US$512 per testing resulted in a substantial additional decrease in quality-adjusted life year gained (Maddali and others incidence and remained very cost-effective. 2015). Eaton and others (2014) noted that in the city of In generalized epidemics in Sub-Saharan Africa, all Belgaum in southern India, where the epidemic is driven analyses demonstrated the cost-effectiveness of further largely by FSWs, the estimated incidence of new infec- ART expansion, including early ART. Five of the 10 pub- tions has fallen substantially since 2003 as a result of lished analyses focused exclusively on South Africa. In an programs targeting this special population. They found extensive analysis, Eaton and others (2014) assessed the that the incremental cost-effectiveness ratio for expand- cost-effectiveness of earlier treatment using six indepen- ing ART to all, regardless of CD4 T-cell counts, was dent models for South Africa and four for Zambia. The US$131 per DALY averted in the presence of these pro- models incorporated a common costing framework and grams; it was slightly higher (US$241) in the theoretical some common assumptions, although the models case in which these programs did not exist. In Vietnam, retained their individual structural features. The cost- where the epidemic is driven by FSWs, MSM, and IDUs, effectiveness of starting treatment at CD4 T-cell counts the incremental cost per DALY averted was US$289. of 500 cells/mm3 compared with CD4 T-cell counts of To guide resource allocation decisions for different 350 cells/mm3 ranged from US$237 to US$1,691 per interventions in the presence of budget constraints, DALY averted in South Africa and from being cost saving Bärnighausen, Bloom, and Humair (2012) reported a to US$749 per DALY averted in Zambia over a 20-year favorable incremental cost per infection averted of ART time horizon (Eaton and others 2014). These estimates initiated only in persons with CD4 T-cell counts of less were considered likely to be very cost-effective in com- than 350 cells/mm3 and VMMC scale-up (US$1,402), parison with international benchmarks for each country compared with early treatment and VMMC scale-up and were similar to those in South Africa reported by (US$7,325–US$10,083); however, this measure does Alistar, Grant, and Bendavid (2014). However, the not value the health- and quality-of-life-related effects threshold for determining whether an intervention is of these interventions. This difference was less marked likely to be cost-effective is poorly known in many when using the measure of incremental cost- resource-limited settings. Granich and others (2012) effectiveness of deaths averted (US$7,761–US$10,014 reported lower costs per DALY averted over a shorter versus US$6,650). In a budgetary analysis, Bärnighausen, time frame of five years and found earlier treatment to Bloom, and Humair (2012) found that VMMC com- be cost-saving over 40 years, using generally more opti- bined with ART at CD4 T-cell counts of less than mistic measures of program quality. 350 cells/mm3 was cumulatively less expensive over More aggressive public health strategies included 12 years and provided a similar incidence reduction as treatment at all CD4 T-cell counts and greater expansion the expansion of treatment eligibility to those with Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 97 98 Major Infectious Diseases Table 5.1 Studies Estimating the Cost-Effectiveness of Treatment Direct and Indirect Benefits Study Study location Study group Intervention, comparison Outcome Conclusions Alistar, Grant, and South Africa HIV-positive adults ART scale-up at CD4 T-cell level At 20 years, cost per QALY gained: US$362–US$370 Scale-up of test-and-treat ART strategy Bendavid 2014 < 350 cells/mm3 vs. test and (all CD4 T-cell levels); US$481–US$486 (< 350 is very cost-effective. Focused PrEP treat at all CD4 T-cell levels; cells/mm3); US$192–US$270 for ART at CD4 strategies, if feasible, would be highly addition of focused (aimed at T-cell levels < 350 cells/mm3 and focused PrEP; cost-effective combined with ART. highest risk) or general PrEP US$1,078–US$1,125 for ART at CD4 T-cell levels < 350 cells/mm3 and general PrEP (2012 US$) Bärnighausen, Bloom, South Africa HIV-positive persons; ART at CD4 T-cell level < 350 At 12 years, US$1,402 per infection averted for Using cost per infection averted as and Humair 2012 general population cells/mm3 vs. at all CD4 T-cell ART (50% coverage) and VMMC (80% coverage); a measure, ART at CD4 T-cell levels levels; plus VMMC; varying US$7,325–US$10,083 per infection averted for < 350 cells/mm3 with VMMC scale-up levels of coverage of each ART at all CD4 T-cell levels (20–80% coverage) has lowest cost-effectiveness ratios, intervention and 80% VMMC; US$6,650 per death averted whereas ART at all levels has higher (50% ART at CD4 T-cell levels < 350 cells/mm3); ratios.a US$7,761–US$10,014 per death averted (70–80% ART at CD4 T-cell levels < 350 cells/mm3; 20–80% ART at all CD4 levels; 80% VMMC) (2012 US$) Eaton and others 2014 South Africa; HIV-positive adults, ART at CD4 T-cell levels < 350 At 20 years, costs per DALY averted at CD4 T-cell ART at CD4 T-cell levels < 500 cells/ Zambia; Bangalore, including key cells/mm3 vs. < 500 cells/mm3 levels < 500 cells/mm3 vs. < 350 cells/mm3: South mm3 and ART at all levels is very cost- Manipur, and populations and at all CD4 T-cell levels Africa: US$237–US$1,691; Zambia: dominating effective in South Africa, in Zambia, Belgaum, India; to US$749; Vietnam: US$290; at all levels vs. and for concentrated epidemic settings Vietnam CD4 T-cell levels < 350 cells/mm3: South Africa: in India and Vietnam. US$438–US$3,790; Zambia: dominating: US$790; India: US$131 (all) and US$199 (< 500 cells/mm3); Vietnam: US$289 (all) (2012 US$) Granich and others South Africa HIV-positive adults Best-case testing and ART: CD4 At five years, ART at CD4 T-cell levels < 500 Early ART is very cost-effective when 2012 T-cell levels < 200 cells/mm3, cells/mm3 vs. 350 cells/mm3: US$221 per DALY considering a short time frame of five < 350 cells/mm3, and < 500 averted; ART at all CD4 T-cell levels: US$1,728 years, with projected cost savings over cells/mm3 vs. expanded testing per DALY averted; with enhanced prevention (40% a 40-year time horizon. and ART at all levels reduction in HIV incidence): US$233 for CD4 T-cell levels < 500 cells/mm3, and US$1,767 for all CD4 T-cell levels (2012 US$) Hontelez and others Hlabisa, South HIV-positive adults ART at CD4 T-cell levels ≤ 350 Costs of treating patients at CD4 T-cell levels Front-loaded costs of treating at CD4 2011 Africa cells/mm3 or ≤ 200 cells/mm3 ≤ 350 cells/mm3 or ≤ 200 cells/mm3 by 2017: T-cell levels ≤ 350 cells/mm3 vs. ≤ 200 breakeven of cumulative net costs in 2026 cells/mm3 may be offset by model- (2010 US$) projected savings from health and prevention gains after 2026. table continues next page Table 5.1 Studies Estimating the Cost-Effectiveness of Treatment Direct and Indirect Benefits (continued) Study Study location Study group Intervention, comparison Outcome Conclusions Hontelez and others 10 countries in Sub- HIV-positive adults ART at all CD4 T-cell levels Over 35 years, US$269 per life-year saved Treatment scale-up at all CD4 T-cell 2016 Saharan Africab with continued scale-up vs. no levels was cost-effective. further scale-up Long, Brandeau, and United States HIV-positive adults; Expanded ART (75% coverage) At 20 years, cost per QALY gained for expanded Expansion of ART coverage in the Owens 2010 general population vs. status quo, testing and ART: US$21,647; testing (low risk once, high risk United States is very cost-effective, expanded ART annually) and expanded ART: US$22,055; testing and expanded testing increases QALYs (low risk every three years, high risk annually) and gained. Expansion of ART alone expanded ART: US$31,274 (2012 US$) decreases new infections by 10.3%; addition of testing (low risk once and high risk annually) decreases new infections by 17.3% over 20 years. Maddali and others India HIV-positive adults Early (CD4 T-cell levels ≥ 350 At 20 years, cost per QALY gained: US$442 Early treatment is very cost-effective 2015 cells/mm3) vs. delayed (CD4 (ideal program conditions)—US$530 (realistic in India. T-cell levels < 350 cells/mm3) parameters for program performance) for ART at initiation of ART CD4 T-cell levels ≥ 350 cells/mm3 (2014 US$) Nichols and others Macha, Zambia HIV-positive adults; Early (CD4 T-cell levels < 500 Cost per QALY gained: US$62 for ART at CD4 Early ART is very cost-effective in this 2014 sexually active cells/mm3) vs. delayed (CD4 T-cell levels < 500 cells/mm3; ART at CD4 T-cell rural setting. Adding PrEP was not adults; general T-cell levels < 350 cells/mm3) levels < 500 cells/mm3 and general population cost-effective. Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV population ART, with and without PrEP for PrEP: $5,861 (2012 US$) most sexually active or general population Walensky and others South Africa; India HIV-positive partners Early (CD4 T-cell levels < 550 Five-year cost per life-year saved: South Africa: Early initiation was cost saving in 2013 in serodiscordant cells/mm3) vs. delayed (< 350 cost saving; India: US$1,800; lifetime horizon, South Africa over a five-year interval, couples cells/mm3) initiation of ART incremental cost per life-year saved: South Africa: cost-effective in India at five years, and US$590; India: US$530 (2012 US$) very cost-effective in both countries over a lifetime. Ying and others 2015 Kampala, Uganda Partners in Scale-up ART at CD4 T-cell At 10 years: cost per DALY averted: US$1,075 ART scale-up at CD4 T-cell levels serodiscordant levels < 500 cells/mm3 vs. for scale-up ART at CD4 T-cell levels < 500 cells/ < 500 cells/mm3 is very cost-effective couples current ART uptake; PrEP for mm3 (2012 US$); cost per DALY averted: US$5,354 compared with current ART uptake. serodiscordant couples (2012 US$) for ART scale-up at CD4 T-cell levels < The addition of PrEP to ART scale-up 500 cells/mm3 and PrEP averted more DALYs, but was not cost- effective by per capita GDP standards. Source: Commission on Macroeconomics and Health 2001. Note: ART = antiretroviral therapy; DALY = disability-adjusted life year; HIV = human immunodeficiency virus; PrEP = preexposure prophylaxis; QALY = quality-adjusted life year; VMMC = voluntary medical male circumcision. This table follows the World Health Organization–endorsed convention, which classifies interventions in cost-effectiveness studies as (1) “cost-effective” when they avert a DALY (or gain a QALY) at a cost of three times per capita GDP or (2) “very cost-effective,” when they avert a DALY (or gain a QALY) at a cost of one times per capita GDP. Dominating = one intervention dominates another intervention when it provides a greater health benefit at a lower cost (Eaton and others 2014, 26). a. Cost per infection averted does not value ART’s long-term health and health-related quality-of-life effects in the denominator of the cost-effectiveness ratio as recommended by consensus guidelines for cost-effectiveness. b. Ethiopia, Kenya, Malawi, Mozambique, Nigeria, South Africa, Tanzania, Uganda, Zambia, and Zimbabwe. 99 CD4 T-cell counts of more than 350 cells/mm3. Ying The challenge of introducing earlier initiation, along and others (2015) reported that scaling up ART to CD4 with the valuation of the prevention effects of treat- T-cell levels of less than 500 cells/mm3 among serodis- ment, is the scarcity of programmatic data (outside of cordant couples in Uganda was very cost-effective com- trials and well-studied cohorts) at this stage of the pared with status quo uptake at CD4 T-cell levels less disease to guide parameter development. Many new than or equal to 350 cells/mm3, whereas the addition of parameters need to be considered and potentially PrEP raised the cost per DALY averted to more than included, which introduces greater uncertainty three times the GDP per capita of Uganda. When using (Bärnighausen, Salomon, and Sangrujee 2012). a denominator of infections averted that did not Substantial uncertainty exists about whether the include health-related effects, they reported more unit costs of various elements of HIV care and treat- favorable estimates of cost-effectiveness of the addition ment will remain constant as scale-up is accelerated, of PrEP to ART scale-up. The emerging issue of ART as extended to persons with higher CD4 T-cell counts, a component of combination prevention strategies is or both. As Meyer-Rath and Over (2012) noted, many further examined in chapter 7 of this volume (Garnett models rely either on fixed unit costs for a year of and others 2017). treatment or on cost accounting identities, in which In general, these myriad cost-effectiveness analyses each of the constituent costs of treatment is estimated present persuasive model-based projections of the and multiplied by health care utilization figures. These incremental cost-effectiveness of expanding treatment costs may need to be considered in a flexible cost- access across a range of settings and populations in function manner, given the likely nonlinearities of LMICs and high-income countries, especially where the inputs around scaling TasP, such as uneven need costs of outreach are low and a long-term perspective is for new infrastructure (Meyer-Rath and Over 2012). taken. The following sections consider further method- Many models also assume constant antiretroviral ological issues in modeling the cost-effectiveness of prices into the future, but for any given drug, prices TasP and the impact of TasP on global health recom- tend to decline over time (Holmes and others 2010). mendations and policy. Conversely, when newer drugs replace existing drugs in guidelines, abrupt price increases may result (Waning and others 2009). Key Considerations and Limitations of Models Most cost-effectiveness models further assume that Projecting the Cost-Effectiveness of Treatment as ART retention and adherence do not vary by stage of Prevention the disease at initiation, the previous health experience As with the modeling of nearly any intervention, many of individuals, and other determinants of health-care- decisions must be made about model structure, and seeking behavior. However, earlier initiation may lead parameters need to be estimated. In general, model to reduced—or increased—overall retention and adher- parameters can be easily explored in sensitivity analy- ence, due to underlying differences in those facts. ses, whereas structural choices may have large impacts Finally, the preventive effects of TasP are likely to and their influence may be more difficult to ascertain. alter the composition of the HIV-infected population In the case of HIV cost-effectiveness, models have been over time, changing its biological and behavioral char- calibrated, to the extent possible, with increasingly acteristics and leading to different costs and outcomes good inputs and outputs from real-life implementa- (Bärnighausen, Salomon, and Sangrujee 2012; Smit and tion of care and treatment programs. Even so, the HIV others 2015). response in resource-limited settings has only been Risk compensation is another consideration that active for the past 11–12 years, and most modeling could have a positive or negative effect on projected ben- horizons are a lifetime; accordingly, certain parameter efits (Dukers and others 2001; Stolte and others 2004). choices—especially as they relate to evolving drug Changes in HIV risk taking following ART scale-up have costs and future lines of treatment—are likely to entail at times been considered risk compensation—that is, substantial uncertainty. HIV-negative persons take more sexual risks as ART The best approach to calibrating effectiveness mod- coverage lowers the average risk of HIV acquisition and els to underlying HIV epidemic trends has been ART availability reduces the potential health losses. debated and managed differently by various groups However, persons could take fewer sexual risks, includ- (Hallett, Eaton, and Menzies 2014; Okano and Blower ing changes in sexual behavior, substance use, and 2014). Modeling cost-effectiveness of TasP also requires contraceptive use, in response to ART scale-up because examining initiation of treatment earlier in the of improved survival expectations and increased opti- course of disease than has been done in most LMICs. mism for the future (Bor and others 2013; Raifman and 100 Major Infectious Diseases others 2014). Future models of combination prevention South Africa is less clear, and the conclusions from the modalities need to attempt to examine the presence of models are conflicting. The up-front costs of expanding potential behavioral effects. testing and links to care are high, and four of seven mod- els favored simply expanding eligibility; three favored expanding testing and links to care at the current thresh- Select Policy Questions Addressed by old of CD4 T-cell counts of less than 350 cells/mm3 Cost-Effectiveness Models (Eaton and others 2014). In 2012–13, South Africa chose As with nearly all health policy decisions, there are trade- to expand coverage of persons with CD4 T-cell counts of offs among benefits, costs, and risks as national and less than 350 cells/mm3 and to use less toxic antiretrovi- global decision makers consider possible strategies for ral medicines; in 2014–15, the country shifted to a policy their investments in treatment and prevention. Although of treating persons with CD4 T-cell counts of less than most TasP scenarios are considered cost-effective or very 500 cells/mm3; in mid-2016, the government announced cost-effective over reasonable horizons for decision the intention to shift to a “treat all” policy consistent making, they all have higher up-front costs. Therefore, with updated WHO guidelines (WHO 2016). optimal policy choices will vary greatly according to available funding, local HIV response, goals, and other How Might the Timing of the Costs and Benefits elements of feasibility and local preference, such as the of TasP Affect Policy Makers’ Decisions, and What political environment. This section highlights several Further Information Could Be Helpful? critical policy issues and illustrates the trade-offs involved Although numerous strategies are considered cost- in using select modeling and cost-effectiveness analyses. effective, treating greater numbers of people brings greater up-front costs regardless of potential down- What Trade-Offs Are Involved in Expanding Testing stream (discounted) savings. Expanding access to per- and Links to Care under Existing Treatment Guidelines, sons with CD4 T-cell counts of less than 350 cells/mm3 Compared with Expanding Earlier Treatment? results in additional costs that tend to increase over time Eaton and others (2014) considered the comparative after a small initial spike (Eaton and others 2014). This effects on new HIV infections and cost-effectiveness of a increase occurs because increasing numbers of individu- policy decision in South Africa, where estimated ART cov- als live longer and incur costs to the health system. erage under existing guidelines (CD4 T-cell counts of less In this analysis, increasing eligibility to persons with than 350 cells/mm3) was approximately 50 percent at the CD4 T-cell counts of less than 500 cells/mm3, with or time of the analysis, and in Zambia, where reported ART without expanded access, results in a greater initial spike coverage (CD4 T-cell counts of less than 350 cells/mm3) in costs; however, as with the “treat all” strategy, the cost was more than 90 percent. First, they found that expand- curve generally declines each subsequent year, in part ing testing and linking patients to care under existing because these strategies are expected to reduce the num- guidelines in South Africa had a substantial effect on low- ber of new HIV infections (Eaton and others 2014). ering the incidence of new infections (6 percent to Accordingly, later outlays could diminish with larger 28 percent, depending on the model). The approach up-front investments, even when including the effects of averted more infections than changing eligibility to CD4 discounting. Decision making will hinge on the assess- T-cell counts of less than 500 cells/mm3 without expand- ment of potential impacts, relevance of model assump- ing testing and links to care (5 percent to 12 percent of tions to the environment, validity of available model infections averted). In Zambia, where reported coverage inputs, availability of funding and competing invest- was already high, simply expanding ART eligibility averted ments, and other local factors. 21 percent to 40 percent of new infections, an impact These types of models generally do not consider the greater than expanding both testing and links to care (8 financial costs to patients of starting treatment earlier. percent to 17 percent). Given the high estimated coverage These costs are related largely to transport and reported in Zambia, the model assumed that cases were opportunity costs; depending on the extent of decen- being identified earlier and that expanding testing and tralization of HIV services, they could be substantial links to care would have somewhat less of an impact than (Rosen and others 2007). If treatment is more for pre- simply raising the threshold. Since the time of this analysis, vention than for direct health benefits, these higher Zambia has raised the treatment threshold to CD4 T-cell up-front costs may discourage patients from getting counts of less than 500 cells/mm3. care, although this theoretical risk requires further From a cost-effectiveness perspective, Zambia’s deci- empirical data. Several investigators have demonstrated sion to expand eligibility was generally supported by greater productivity with HIV treatment in clinic- and assessments of costs per DALY averted. The picture for population-based cohorts (Bor and others 2012), which Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 101 is excluded in most model-based analyses. The extent to NATIONAL AND GLOBAL GUIDANCE, which productivity gains could offset transport and POLICIES, AND RESOURCE ALLOCATION other costs among people starting treatment while healthier is unknown. The release in May 2011 of the HPTN 052 data on the It is also difficult for models to reflect that, in the remarkable reduction of sexual transmission of HIV context of constrained budgets, additional spending on disease among serodiscordant couples sent ripples not HIV treatment will lead to displaced resources for only through the scientific community, but also through other interventions. If the treatment intervention gen- policy-making circles, including national governments, erates health per cost at a rate greater than a certain the WHO, and leading funders of HIV programs in threshold, then despite that displacement, there is a net low-resource settings (El Sadr and others 2011). After gain in health. However, quantifying that threshold is more than two decades of focusing on the health-related difficult. International guidance has, until recently, sug- effects of ART, experts began to grapple with its role as a gested benchmarks related to the GDP per capita of a tool for the prevention of sexual transmission of HIV. country; there are indications that a more realistic Members of the WHO Guidelines Group on Couples assessment of the opportunity costs of health expendi- HIV Testing and Counseling incorporated the HPTN ture would demand lower cost per health gain for an 052 findings in their review process (WHO 2012a), judg- intervention to be likely to be cost-effective (Woods ing them to be directly and immediately applicable to and others 2015). couples counseling and testing. In addition to other Policy makers in many sectors, but particularly health, potential interventions, the evidence for treatment in face a trade-off between higher up-front costs and this setting was considered to be substantial. After stake- longer-term gains. Unlike problems associated with non- holder reviews, updated guidelines were released 11 communicable diseases, TasP could reduce the intensity months after the HPTN 052 results (box 5.1). and spread of a transmissible pandemic. In this time- In April 2012, the WHO incorporated the findings of dependent context, donors and policy makers have often HPTN 052 and released the programmatic update on leaned toward up-front investments. Option B+ (WHO 2012b). In addition to programmatic data from Malawi that supported full treatment for What Approaches Could Improve the Cost- pregnant women, the potential effects of treatment on Effectiveness of Treatment as Prevention? sexual prevention of HIV was considered as follows: “If Innovations in care delivery, such as task-shifting ele- Option B+ can be supported, funded, scaled up at the ments of the delivery cascade to lower-level staff mem- primary care level and sustained, it will also likely pro- bers, have been widely adopted and have facilitated vide the best protection for the mother’s health, and it reductions in the costs of delivering care. Innovators offers a promising new approach to preventing sexual such as Médecins Sans Frontières and national govern- transmission and new HIV infections in the general ments have further pioneered methods of care delivery, population” (WHO 2012b, 4). now known as differentiated care, which target the In 2013, the WHO combined all of its ART-related intensity of care to the needs of patients (Duncombe and HIV guidance into a single guideline that considered the others 2015; Holmes and Sanne 2015). For example, the sexual prevention effects of treatment demonstrated in formation of community adherence groups among sta- ble patients in Mozambique allowed for substantially less clinic contact and increased retention in care among those opting into these models (Decroo and others Box 5.1 2011). Greater attention to both the models of care and the costs of care delivery is another tool that can be used by in-country stakeholders to strive for greater efficiency Recommendations from the WHO Couples and quality of care delivery. The President’s Emergency HIV Testing and Counseling Guidelines Plan for AIDS Relief (PEPFAR) program’s expenditure analysis approach entails the collection of detailed data People with HIV whose partners do not have that allow country-level decision makers to distinguish HIV and who are started on ART for their between low- and high-cost providers of quality care. A own health should be advised that ART is also report of results from this methodology noted reduc- recommended to reduce HIV transmission to tions in the heterogeneity of the costs of supporting not uninfected partners. This is a strong recommen- only ART, but also HIV testing and other key elements dation based on high-quality evidence. along the TasP cascade (PEPFAR 2012). 102 Major Infectious Diseases HPTN 052. The guidelines committee changed eligibil- others (El Sadr and others 2011). The board also recom- ity criteria to CD4 T-cell counts of 500 cells/mm3, per- mended that “careful evaluations, including assessment suaded by a combination of clinical benefits drawn of benefit/risk/impact/feasibility and modeling exercises mainly from LMIC settings (Kitahata and others 2009), are urgently needed to identify populations that should along with evidence of reduced sexual transmission and be prioritized for this intervention, given local condi- tuberculosis among treated individuals. In addition, tions” (El Sadr and others 2011, 19). Following these and review of the evidence using the Grading of other recommendations, the U.S. government strongly Recommendations Assessment, Development and endorsed accelerating combination prevention in 2011, Evaluation (GRADE) system led to recommendations including a 50 percent increase in persons on treatment for treatment regardless of CD4 T-cell count for preg- over a two-year period. As overall allocations to HIV nant women, serodiscordant couples, persons with reached a plateau, PEPFAR weighted its financial alloca- tuberculosis, and persons with severe liver disease. The tions more heavily toward treatment and several other U.S. and European guidelines, which already counseled high-impact interventions (Cohen, Holmes, and others earlier initiation of treatment, also incorporated the ben- 2012; Goosby and others 2012; Holmes and others 2012; efits of TasP (DHHS 2014; European AIDS Clinical PEPFAR 2014). In 2014, PEPFAR took the further step of Society 2014). Updated WHO guidelines, reviewed in the endorsing the UNAIDS (Joint United Nations Programme setting of further evidence for the health benefits of ear- on HIV/AIDS) 90-90-90 targets, and in 2015 announced lier ART initiation, recommended offering treatment to support for the ongoing expansion of treatment in the all individuals with HIV, regardless of CD4 T-cell count context of further studies on the effectiveness of earlier (INSIGHT START Study Group 2015; TEMPRANO treatment on individual health (PEPFAR 2014, 2015). ANRS Study Group 2015; WHO 2016). Since the release of the WHO documents recom- mending earlier treatment, national governments have ONGOING RESEARCH weighed guidelines changes. As of 2015, the WHO reported that of the 58 WHO focus countries, 34 Even as investigators have advanced the understanding (59 percent) have adopted new guidelines for treatment of the preventive role of treatment, critical questions of serodiscordant couples and 37 (64 percent) have raised remain. Ongoing studies are evaluating the biological, their thresholds to CD4 T-cell counts of 500 cells/mm3. pharmacologic, clinical, and public health elements of Rwanda’s national guidelines support a TasP strategy ART as a prevention modality, and priority areas for (ART regardless of CD4 T-cell count) and universal test- future research continue to emerge. ing. Other countries with limited health infrastructure and financial resources are struggling with high unmet needs at lower CD4 T-cell counts and have legitimate Biological and Pharmacological Studies concerns about crowding out treatment slots for sicker The quest to understand the biology of transmission patients (Linas and others 2006). The prevention effects revolves around the fitness requirements of the viral of treatment have also substantially influenced policy and pathogen (Carlson and others 2014) and its susceptibil- allocation decisions of major payers and funders of HIV ity to innate host defenses (Borrow 2011). Successful programs in LMICs. viral suppression does not prevent intermittent viral The results of HPTN 052 arrived during the latter shedding in the genital tracts of both men (Kalichman stages of a global economic downturn and coincided and others 2010) and women (Cu-Uvin and others with a leveling off of the rapid growth of HIV-related 2010). However, observational studies suggest that the overseas development aid (Kaiser Family Foundation viruses being shed during viral suppression are likely 2015). Despite these challenges, global leaders, advo- compromised and not readily transmitted (Cohen and cates, and public health officials were energized by the others 2013). potential impact of the addition of ART to the combina- Perhaps the most important TasP consideration lies tion prevention armamentarium. in the simplification of treatment itself. Successful sup- The PEPFAR program is the largest bilateral program pression of viral replication virtually eliminates HIV supporting the HIV response in LMICs. Its federally transmission (Muessig and Cohen 2014). Accordingly, chartered Scientific Advisory Board reviewed the HPTN linked transmission events reflect failed treatment or 052 data, discussed its potential applications, and ulti- resistance to the antiretroviral regimen being used. mately recommended that PEPFAR support the use of Failure to adhere to a treatment regimen is the greatest ART in specific populations with CD4 T-cell counts problem. Newer antiviral agents are very well tolerated greater than 350 cells/mm3 to prevent transmission to but still require daily medication. Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 103 Glaxo-Smith-Kline and Janssen are exploring a com- impact of TasP. Four large studies underway are designed bination of injectable agents, with one combination to demonstrate that TasP, as part of a package of preven- being tested in the Phase 2b LATTE (Long-Acting tion interventions, reduces population-level HIV inci- Antiretroviral Treatment Enabling) Trial (Margolis and dence in generalized epidemics (box 5.2). others 2014). The trial includes a run-in of oral agents for safety testing, which has been completed, followed by maintenance injections every month or every two CONCLUSIONS months. Injectable ART may be appropriate for several Treatment has well-known direct effects on the health types of patients, but it is particularly attractive for peo- outcomes of HIV-positive individuals; it has also been ple in serodiscordant sexual relationships. conclusively shown in observational studies and ran- domized clinical trials to prevent sexual transmission of HIV. These prevention effects are backed by years of Clinical, Public Health, and Population Effects Studies basic science and clinical work that have established the Granich and others (2011) identified more than 50 effectiveness of ART in reducing HIV replication in ongoing studies covering the impact of treatment on blood and genital tissues and secretions. The projected prevention among serodiscordant couples and key pop- effectiveness of early treatment on reductions in sexual ulations and the secondary benefits of treatment for transmission of HIV is dependent on the performance individuals infected with both tuberculosis and HIV. of the treatment cascade, including HIV testing and These studies examine HIV incidence and mortality in links to care, retention in care, and virological suppres- the general population and economic outcomes for sion. When the prevention effects of ART are included in patients receiving TasP. analyses of the cost-effectiveness of earlier treatment, Substantial funding has been allocated to research ART is generally found to be a highly cost-effective inter- aimed at understanding the potential population-level vention in diverse settings of varied income levels, HIV Box 5.2 Population-Level Effects of Treatment as Prevention: Select Studies Population Effects of ART to Reduce HIV to establish the causal impact of TasP (treatment as Transmission (PopART), also known as the HIV prevention) on population-level HIV incidence and Prevention Trials Network 071, examines the effect other health, economic, and social outcomes. The of universal testing and treatment compared with trial randomized 34 communities with a total adult treatment at CD4 T-cell counts of less than 350 cells/ population of 34,000 to receive home-based HIV mm3 or the standard of care. PopART includes 21 testing and ART referral under either a TasP strategy communities in two countries—South Africa and (intervention) or the South African standard of care Zambia—with a total population of 1.2 million; with home-based HIV testing (control) (Iwuji and results are expected in 2017–18 (Hayes and others others 2013). The only difference between the inter- 2014). vention and the control arm was whether HIV- positive people were offered immediate ART in early The Botswana Combination Prevention Project stages of the disease (intervention) vs. only in later assesses the provision of treatment to all individuals disease stages (control). with CD4 T-cell counts greater than 350 cells/mm3 The SEARCH (Sustainable East Africa Research on or with a viral load greater than or equal to 10,000 Community Health) Study, based in Kenya and Uganda, copies per milliliter, compared with individuals includes 32 communities of approximately 10,000 indi- receiving the standard of care in the setting of viduals each. The study compares early treatment to scaled-up combination prevention (CDC 2013). standard of care, with multiple health outcomes and a The Africa Centre and the Agence Nationale de prevention outcome of community viral load (Chamie Recherche sur le Sida 12249 TasP Trial was developed and others 2012; Jain and others 2013). 104 Major Infectious Diseases epidemic types, and risk populations. It is projected to Attia, S., M. Egger, M. Müller, M. Zwahlen, and N. Low. 2009. be most effective and cost-effective when paired with “Sexual Transmission of HIV According to Viral Load efforts to identify infected individuals through expanded and Antiretroviral Therapy: Systematic Review and Meta- testing and links to care. Analysis.” AIDS 23 (11): 1397–404. Baeten, J. M., E. Kahle, J. R. Lingappa, R. W. Coombs, S. Delany- Ongoing population-based studies will provide fur- Moretlwe, and others. 2011. “Genital HIV-1 RNA Quantity ther information on the wider prevention impact of Predicts Risk of Heterosexual HIV-1 Transmission.” Science earlier treatment as part of a package of combination Translational Medicine 3 (77): 77ra29. prevention modalities. Baggaley, R. F., D. Dimitrov, B. N. Owen, M. Pickles, A. R. Butler, and others. 2013. “Heterosexual Anal Intercourse: A Neglected Risk Factor for HIV?” American Journal of ANNEXES Reproductive Immunology 69 (1): 95–105. Baggaley, R. F., R. G. White, T. D. Hollingsworth, and The annexes to this chapter are as follows. They are avail- M. C. Boily. 2013. “Heterosexual HIV-1 Infectiousness able at http://www.dcp-3.org/infectiousdiseases. and Antiretroviral Use: Systematic Review of Prospective Studies of Discordant Couples.” Epidemiology 24 (1): • Annex 5A. Studies of TasP in Serodiscordant Couples 110–21. • Annex 5B. Ecological Studies Examining the Effec- Bärnighausen, T., D. E. Bloom, and S. Humair. 2012. tiveness of ART on HIV Incidence “Economics of Antiretroviral Treatment vs. Circumcision for HIV Prevention.” Proceedings of the National Academy of Sciences of the United States of America 109 (52): 21271–76. NOTES Bärnighausen, T., J. A. Salomon, and N. Sangrujee. 2012. “HIV The authors are indebted to Musonda Namuyemba and Megan Treatment as Prevention: Issues in Economic Evaluation.” Wolf, MPH, for their editorial assistance. PLoS Medicine 9 (7): e1001263. This chapter is linked closely with chapters on the HIV care Boily, M. C., B. Mâsse, R. Alsallaq, N. S. Padian, J. W. Eaton, continuum in adults and children (chapter 4, Harrispersaud and others. 2012. “HIV Treatment as Prevention: and others 2017), prevention of mother-to-child transmis- Considerations in the Design, Conduct, and Analysis of sion (chapter 6, John-Stewart and others 2017), and cost- Cluster Randomized Controlled Trials of Combination effectiveness of interventions to prevent HIV acquisition HIV Prevention.” PLoS Medicine 9 (7): e1001250. (chapter 7, Garnett and others 2017). Boily, M. C., and Z. Shubber. 2014. “Modelling in Concentrated World Bank Income Classifications as of July 2014 are as Epidemics: Informing Epidemic Trajectories and Assessing follows, based on estimates of gross national income (GNI) Prevention Approaches.” Current Opinion in HIV and AIDS per capita for 2013: 9 (2): 134–49. Bor, J., A. J. Herbst, M.-L. Newell, and T. Bärnighausen. 2013. • Low-income countries (LICs) = US$1,045 or less “Increases in Adult Life Expectancy in Rural South Africa: • Middle-income countries (MICs) are subdivided: Valuing the Scale-Up of HIV Treatment.” Science 339 (a) lower-middle-income = US$1,046 to US$4,125 (6122): 961–65. (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 Bor, J., F. Tanser, M.-L. Newell, and T. Bärnighausen. 2012. • High-income countries (HICs) = US$12,746 or more. “Nearly Full Employment Recovery among South African HIV Patients on Antiretroviral Therapy: Evidence from a Large Population Cohort.” Health Affairs 31 (7): 10.1377 /hlthaff.2012.0407. REFERENCES Borrow, P. 2011. “Innate Immunity in Acute HIV-1 Infection.” AAAS (American Association for the Advancement of Current Opinion in HIV and AIDS 6 (5): 353–63. Science). 2016. “Entering the Health-Care System: The Brenner, B., M. A. Wainberg, and M. Roger. 2013. “Phylogenetic Challenge of the Universal Test-and-Treat Strategy.” Press Inferences on HIV-1 Transmission: Implications for the release, Eurekalert, July 20. http://www.eurekalert.org Design of Prevention and Treatment Interventions.” AIDS /pub_releases/2016-07/a-eth071916.php#.V5TTdKRGkxo 27 (7): 1045–57. .email. Burns, D. N., V. DeGruttola, C. D. Pilcher, M. Kretzschmar, Alistar, S. S., P. M. Grant, and E. Bendavid. 2014. “Comparative C. M. Gordon, and others. 2014. “Toward an Endgame: Effectiveness and Cost-Effectiveness of Antiretroviral Finding and Engaging People Unaware of Their HIV-1 Therapy and Pre-Exposure Prophylaxis for HIV Prevention Infection in Treatment and Prevention.” AIDS Research and in South Africa.” BMC Medicine 12 (46): 11. Human Retroviruses 30 (3): 217–24. Anderson, B. L., and S. C. Cu-Uvin. 2011. “Clinical Parameters Carlson, J. M., M. Schaefer, D. C. Monaco, R. Batorsky, Essential to Methodology and Interpretation of Mucosal D. T. Claiborne, and others. 2014. “HIV Transmission: Responses.” American Journal of Reproductive Immunology Selection Bias at the Heterosexual HIV-1 Transmission 65 (3): 352–60. Bottleneck.” Science 345 (6193): 1254031. Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 105 Castel, A. D., M. Befus, S. Willis, A. Griffin, S. Hader, and others. Danel, C., R. Moh, D. Gabillard, A. Badje, J. Le Carrou, and 2012. “Use of the Community Viral Load as a Population- others. 2015. “A Trial of Early Antiretrovirals and Isoniazid Based Biomarker of HIV Burden.” AIDS 26 (3): 345–53. Preventive Therapy in Africa.” New England Journal of CDC (Centers for Disease Control and Prevention). Medicine 373 (9): 808–22. 2013. “Botswana Combination Prevention Project Das, M., P. L. Chu, G. M. Santos, S. Scheer, E. Vittinghoff, (BCPP).” Atlanta, GA. https://clinicaltrials.gov/ct2/show and others. 2010. “Decreases in Community Viral Load /NCT01965470. Are Accompanied by Reductions in New HIV Infections Chamie, G., D. Kwarisiima, T. D. Clark, J. Kabami, V. Jain, and in San Francisco.” PLoS One 5 (6): e11068. http://dx.doi others. 2012. “Leveraging Rapid Community-Based HIV .org/10.1371/journal.pone.0011068. Testing Campaigns for Non-Communicable Diseases in Decroo, T., B. Telfer, M. Biot, J. Maikere, S. Dezembro, and Rural Uganda.” PLoS One 7 (8): e43400. others. 2011. “Distribution of Antiretroviral Treatment Cherutich, P., A. A. Kim, T. A. Kellogg, K. Sherr, A. Waruru, through Self-Forming Groups of Patients in Tete Province, and others. 2016. “Detectable HIV Viral Load in Kenya: Mozambique.” Journal of Acquired Immune Deficiency Data from a Population-Based Survey.” PLoS One 11 (5): Syndromes 56 (2): e39–44. e0154318. Delva, W., D. P. Wilson, L. Abu-Raddad, M. Gorgens, Cohen, M. S, Y. Chen, M. McCauley, T. Gamble, and D. Wilson, and others. 2012. “HIV Treatment as Prevention: M. Hosseinpour. 2015. “Final Results of the HPTN 052 Principles of Good HIV Epidemiology Modelling for Randomized Controlled Trial: Antiretroviral Therapy Public Health Decision-Making in All Modes of Prevention Prevents HIV Transmission.” International AIDS Society and Evaluation.” PLoS Medicine 9 (7): e1001239. 2015, 8th Conference on HIV Pathogenesis, Treatment, and DHHS (Department of Health and Human Services). Prevention, Vancouver, British Columbia, July 19–22. 2014. “Guidelines for the Use of Antiretroviral Agents Cohen, M. S., Y. Q. Chen, M. McCauley, T. Gamble, in HIV-1-Infected Adults and Adolescents.” Panel on M. C. Hosseinipour, and others. 2011. “Prevention of HIV-1 Antiretroviral Guidelines for Adults and Adolescents, Infection with Early Antiretroviral Therapy.” New England DHHS, Washington, DC. Journal of Medicine 365 (6): 493–505. Donnell, D., J. M. Baeten, J. Kiarie, K. Thomas, W. Stevens, Cohen, M. S., Y. Q. Chen, M. McCauley, T. Gamble, and others. 2010. “Heterosexual HIV-1 Transmission after M. C. Hosseinipour, and others. 2016. “Antiretroviral Initiation of Antiretroviral Therapy: A Prospective Cohort Therapy for the Prevention of HIV-1 Transmission.” Analysis.” The Lancet 375 (9731): 2092–98. New England Journal of Medicine 375: 830–39. Dukers, N. H., J. Goudsmit, J. B. de Wit, M. Prins, G. J. Weverling, Cohen, M. S., C. Dye, C. Fraser, W. C. Miller, K. A. Power, and and others. 2001. “Sexual Risk Behaviour Relates to the others. 2012. “HIV Treatment as Prevention: Debate and Virological and Immunological Improvements during Commentary; Will Early Infection Compromise Treatment- Highly Active Antiretroviral Therapy in HIV-1 Infection.” as-Prevention Strategies?” PLoS Medicine 9 (7): e1001232. AIDS 15 (3): 369–78. Cohen, M. S., C. L. Gay, M. P. Busch, and F. M. Hecht. Duncombe, C., S. Rosenblum, N. Hellman, C. Holmes, L.Wilkinson, 2010. “The Detection of Acute HIV Infection.” Journal of and others. 2015. “Reframing HIV Care: Putting People at the Infectious Diseases 202 (Suppl 2): S270–77. Centre of Antiretroviral Delivery.” Tropical Medicine and Cohen, M. S., C. Holmes, N. Padian, M. Wolf, G. Himschall, International Health 20 (4): 430–47. and others. 2012. “HIV Treatment as Prevention: How Eaton, J. W., and T. B. Hallett. 2014. “Why the Proportion Scientific Discovery Occurred and Translated Rapidly into of Transmission during Early-Stage HIV Infection Does Policy for the Global Response.” Health Affairs 31 (7): Not Predict the Long-Term Impact of Treatment on HIV 1439–49. Incidence.” Proceedings of the National Academy of Sciences Cohen, M. S., M. K. Smith, K. E. Muessig, T. B. Hallett, of the United States of America 111 (45): 16202–7. K. A. Powers, and others. 2013. “Antiretroviral Treatment of Eaton, J. W., L. F. Johnson, J. A. Salomon, T. Bärnighausen, HIV-1 Prevents Transmission of HIV-1: Where Do We Go E. Bendavid, and others. 2012. “HIV Treatment as Prevention: from Here?” The Lancet 383 (9903): 1515–24. Systematic Comparison of Mathematical Models of the Commission on Macroeconomics and Health. 2001. Potential Impact of Antiretroviral Therapy on HIV Incidence Macroeconomics and Health: Investing in Health for Economic in South Africa.” PLoS Medicine 9 (7): e1001245. Development. Geneva: World Health Organization. Eaton, J. W., N. A. Menzies, J. Stover, V. Cambiano, Cope, A. B., A. M. Crooks, T. Chin, J. D. Kuruc, K. S. McGee, L. Chindelevitch, and others. 2014. “Health Benefits, Costs, and others. 2014. “Incident Sexually Transmitted Infection and Cost-Effectiveness of Earlier Eligibility for Adult as a Biomarker for High-Risk Sexual Behavior Following Antiretroviral Therapy and Expanded Treatment Coverage: Diagnosis with Acute HIV.” Sexually Transmitted Diseases A Combined Analysis of 12 Mathematical Models.” 41 (7): 447–52. The Lancet Global Health 2 (1): e23–34. doi:10.1016 Cu-Uvin, S., A. K. DeLong, K. K. Venkatesh, J. W. Hogan, /S2214-109X(13)70172-4. J. Ingersoll, and others. 2010. “Genital Tract HIV-1 RNA El Sadr, W., M. S. Cohen, K. DeCock, L.-G. Bekker, S. Karim, Shedding among Women with Below Detectable Plasma and others. 2011. PEPFAR Scientific Advisory Board Viral Load.” AIDS 24 (16): 2489–97. Recommendations for the Office of the U.S. Global AIDS 106 Major Infectious Diseases Coordinator: Implications of HPTN 052 for PEPFAR’s Infectious Diseases, edited by K.K. Holmes, S. Bertozzi, B. R. Treatment Programs. Washington, DC: PEPFAR. https:// Bloom, and P. Jha. Washington, DC: World Bank. www.pepfar.gov/documents/organization/177126.pdf. Hayes, R., H. Ayles, N. Beyers, K. Sabapathy, S. Floyd, and European AIDS Clinical Society. 2014. Guidelines, version 7.1. others. 2014. “HPTN 071 (PopART): Rationale and Design Copenhagen: European AIDS Clinical Society. http://www of a Cluster-Randomised Trial of the Population Impact of .eacsociety.org/files/guidelines_english_71_141204.pdf. an HIV Combination Prevention Intervention Including Fang, C.-T., H.-M. Hsu, S.-J. Twu, M.-Y. Chen, Y. Y. Chang, and Universal Testing and Treatment: A Study Protocol for a others. 2004. “Decreased HIV Transmission after a Policy Cluster Randomised Trial.” Trials 15 (1): 57. of Providing Free Access to Highly Active Antiretroviral Hendrix, C. W., E. J. Fuchs, K. J. Macura, L. A. Lee, T. L. Parsons, Therapy in Taiwan.” Journal of Infectious Diseases 190 (5): and others. 2008. “Quantitative Imaging and Sigmoidoscopy 879–85. to Assess Distribution of Rectal Microbicide Surrogates.” Gardner, E. M., M. P. McLees, J. F. Steiner, C. Del Rio, and W. Clinical Pharmacology and Therapeutics 83 (1): 97–105. J. Burman. 2011. “The Spectrum of Engagement in HIV Henry, K. 1988. “Setting AIDS Priorities: The Need for a Closer Care and Its Relevance to Test-and-Treat Strategies for Alliance of Public Health and Clinical Approaches toward Prevention of HIV Infection.” Clinical Infectious Diseases the Control of AIDS.” American Journal of Public Health 52 (6): 793–800. 78 (9): 1210–12. Garnett, G., S. Krishnaratne, K. Harris, T. Hallet, M. Santos, Ho, Y.-C., L. Shan, N. N. Hosmane, J. Wang, S. B. Laskey, and others. 2017. “Cost-Effectiveness of Interventions to and others. 2013. “Replication-Competent Non-Induced Prevent HIV Acquisition.” In Disease Control Priorities Proviruses in the Latent Reservoir Increase Barrier to HIV-1 (third edition): Volume 6, Major Infectious Diseases, edited Cure.” Cell 155 (3): 540–51. by K. K. Holmes, S. Bertozzi, B. R. Bloom, and P. Jha. Hollingsworth, T. D., R. M. Anderson, and C. Fraser. 2008. Washington, DC: World Bank. “HIV-1 Transmission, by Stage of Infection.” Journal of Gold, M. R., J. E. Siegel, L. B. Russell, and M. C. Weinstein. Infectious Diseases 198 (5): 687–93. 1996. Cost-Effectiveness in Health and Medicine: Report Holmes, C. B., J. M. Blandford, N. Sangrujee, S. R. Stewart, of the Panel on Cost-Effectiveness in Health and Medicine. A. Dubois, and others. 2012. “PEPFAR’s Past and Future New York: Oxford University Press. Efforts to Cut Costs, Improve Efficiency, and Increase the Goosby, E., D. Von Zinkernagel, C. B. Holmes, D. Haroz, Impact of Global HIV Programs.” Health Affairs 31 (7): and T. Walsh. 2012. “Raising the Bar: PEPFAR and New 1553–60. Paradigms for Global Health.” Journal of Acquired Immune Holmes, C. B., W. Coggin, D. Jamieson, H. Mihm, R. Granich, Deficiency Syndromes 60 (Suppl 3): S158–62. and others. 2010. “Use of Generic Antiretroviral Agents and Granich, R. M., C. F. Gilks, C. Dye, K. M. De Cock, and Cost Savings in PEPFAR Treatment Programs.” Journal of B. G. Williams. 2009. “Universal Voluntary HIV Testing the American Medical Association 304 (3): 313–20. with Immediate Antiretroviral Therapy as a Strategy for Holmes, C. B., and I. Sanne. 2015. “Changing Models of Care to Elimination of HIV Transmission: A Mathematical Model.” Improve Progression through the HIV Treatment Cascade The Lancet 373 (9657): 48–57. in Different Populations.” Current Opinion in HIV and Granich, R. M., S. Gupta, A. B. Suthar, C. Smyth, D. Hoos, and AIDS 10 (6): 447–50. others. 2011. “Antiretroviral Therapy in Prevention of HIV Holmes, C. B., R. Wood, M. Badri, S. Zilber, B. Wang, and and TB: Update on Current Research Efforts.” Current HIV others. 2006. “CD4 Decline and Incidence of Opportunistic Research 9 (6): 446–69. Infections in Cape Town, South Africa: Implications for Granich, R. M., J. G. Kahn, R. Bennett, C. B. Holmes, N. Garg, Prophylaxis and Treatment.” Journal of Acquired Immune and others. 2012. “Expanding ART for Treatment and Deficiency Syndromes 42 (4): 464–69. Prevention of HIV in South Africa: Estimated Cost and Hontelez, J. A., A. Y. Chang, O. Ogbuoji, S. J. Vlas, Cost-Effectiveness 2011–2050.” PLoS One 7 (2): e30216. T. Bärnighausen, and others. 2016. “Changing HIV Grinsztejn, B., M. C. Hosseinipour, H. J. Ribaudo, S. Swindells, Treatment Eligibility under Health System Constraints in J. Eron, and others. 2014. “Effects of Early Versus Delayed Sub-Saharan Africa: Investment Needs, Population Health Initiation of Antiretroviral Treatment on Clinical Outcomes Gains, and Cost-Effectiveness.” AIDS 30 (150): 2341–50. of HIV-1 Infection: Results from the Phase 3 HPTN Hontelez, J. A. C., S. J. de Vlas, F. Tanser, R. Bakker, 052 Randomised Controlled Trial.” The Lancet Infectious T. Bärnighausen, and others. 2011. “The Impact of the Diseases 14 (4): 281–90. New WHO Antiretroviral Treatment Guidelines on HIV Hallett, T. B., J. W. Eaton, and N. Menzies. 2014. “Beware Epidemic Dynamics and Cost in South Africa.” PLoS One of Using Invalid Transmission Models to Guide HIV 6 (7): e21919. Health Policy: Authors’ Reply.” The Lancet Global Health INSIGHT START Study Group. 2015. “Initiation of 2: e261. Antiretroviral Therapy in Early Asymptomatic HIV Harrispersaud, K., M. McNairy, S. Ahmed, E. J. Abrams, H. Infection.” New England Journal of Medicine 373 (9): Thirumurthy, and others. 2017. “HIV Care Continuum in 795–807. Adults and Children: Cost-Effectiveness Considerations.” In Iwuji, C. C., J. Orne-Gliemann, E. Balestre, J. Larmarange, Disease Control Priorities (third edition): Volume 6, Major R. Thiebaut, and others. 2016. “The Impact of Universal Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 107 Test and Treat on HIV Incidence in a Rural South African Keele, B. F., E. E. Giorgi, J. F. Salazar-Gonzalez, J. M. Decker, Population: ANRS 12249 TasP Trial, 2012–2016.” 21st K. T. Pham, and others. 2008. “Identification and International AIDS Conference (AIDS 2016), Durban, Characterization of Transmitted and Early Founder Virus South Africa, July 18–22. Envelopes in Primary HIV-1 Infection.” Proceedings of the Iwuji, C., J. Orne-Gliemann, J. Larmarange, N. Okesola, National Academy of Sciences of the United States of America F. Tanser, and others. 2016. “Uptake of Home-Based HIV 105 (21): 7552–57. Testing, Linkage to Care, and Community Attitudes about Kitahata, M. M., S. J. Gange, A. G. Abraham, B. Merriman, ART in Rural Kwazulu-Natal, South Africa: Descriptive M. S. Saaq, and others. 2009. “Effect of Early versus Results from the First Phase of the ANRS 12249 TasP Deferred Antiretroviral Therapy for HIV on Survival.” Cluster-Randomised Trial.” PLoS Medicine 13 (8): e1002107. New England Journal of Medicine 360 (18): 1815–26. Iwuji, C. C., J. Orne-Gliemann, F. Tanser, S. Boyer, R. J. Lessells, Kwara, A., A. Delong, N. Rezk, J. Hogan, H. Burtwell, and and others. 2013. “Evaluation of the Impact of Immediate others. 2008. “Antiretroviral Drug Concentrations and HIV Versus WHO Recommendations: Guided Antiretroviral RNA in the Genital Tract of HIV-Infected Women Receiving Therapy Initiation on HIV Incidence; the ANRS 12249 Long-Term Highly Active Antiretroviral Therapy.” Clinical TasP (Treatment as Prevention) Trial in Hlabisa Sub- Infectious Diseases 46 (5): 719–25. district, KwaZulu-Natal, South Africa, Study Protocol for a Laeyendecker, O., R. Brookmeyer, C. E. Mullis, D. Donnell, Cluster Randomized Control Trial.” Trials 14 (1): 230. J. Lingappa, and others. 2012. “Specificity of Four Jain, V., T. Liegler, J. Kabami, G. Chamie, T. D. Clark, and others. Laboratory Approaches for Cross-Sectional HIV Incidence 2013. “Assessment of Community Viral Load in a Rural East Determination: Analysis of Samples from Adults with African Setting Using a Fingerprick-Based Blood Collection Known Nonrecent HIV Infection from Five African Method.” Clinical Infectious Diseases 56 (4): 598–605. Countries.” AIDS Research and Human Retroviruses 28 Janssen, R. S., G. A. Satten, S. L. Stramer, B. D. Rawal, (10): 1177–83. T. R. O’Brien, and others. 1998. “New Testing Strategy to Law, M. G., I. Woolley, D. J. Templeton, N. Roth, J. Chuah, and Detect Early HIV-1 Infection for Use in Incidence Estimates others. 2011. “Trends in Detectable Viral Load by Calendar and for Clinical and Prevention Purposes.” Journal of the Year in the Australian HIV Observational Database.” Journal American Medical Association 280 (1): 42–48. of the International AIDS Society 14 (1): 10. Jia, Z., Y. Mao, F. Zhang, Y. Ruan, Y. Ma, and others. 2012. Le Vu, S., J. Pillonel, C. Semaille, P. Bernillon, Y. Le Strat, “Antiretroviral Therapy to Prevent HIV Transmission in and others. 2008. “Principles and Uses of HIV Incidence Serodiscordant Couples in China (2003–11): A National Estimation from Recent Infection Testing: A Review.” Observational Cohort Study.” The Lancet 382 (9899): Eurosurveillance 13 (36): 11–16. 1195–203. Li, H., K. J. Bar, S. Wang, J. M. Decker, Y. Chen, and others. John-Stewart, G., R. Peeling, C. Levin, P. J. Garcia, D. Mabey, 2010. “High Multiplicity Infection by HIV-1 in Men Who and others. 2017. “Elimination of Mother-to-Child Have Sex with Men.” PLoS Pathogens 6 (5): e1000890. Transmission of HIV and Syphilis.” In Disease Control Lima, V. D., K. Johnston, R. S. Hogg, A. R. Levy, P. R. Harrigan, Priorities (third edition): Volume 6, Major Infectious and others. 2008. “Expanded Access to Highly Active Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, Antiretroviral Therapy: A Potentially Powerful Strategy and P. Jha. Washington, DC: World Bank. to Curb the Growth of the HIV Epidemic.” Journal of Joseph, S. B., K. T. Arrildt, A. E. Swanstrom, G. Schnell, B. Lee, Infectious Diseases 198 (1): 59–67. and others. 2014. “Quantification of Entry Phenotypes of Linas, B. P., H. Zheng, E. Losina, A. Rockwell, R. P. Walensky, Macrophage-Tropic HIV-1 across a Wide Range of CD4 and others. 2006. “Optimizing Resource Allocation in Densities.” Journal of Virology 88 (4): 1858–69. United States AIDS Drug Assistance Programs.” Clinical Kaiser Family Foundation. 2015. “Financing the Response to Infectious Diseases 43 (10): 1357–64. HIV in Low- and Middle-Income Countries: International Long, E. F., M. L. Brandeau, and D. K. Owens. 2010. “The Assistance from Donor Governments in 2015.” Kaiser Cost-Effectiveness and Population Outcomes of Expanded Family Foundation, Menlo Park, CA. http://files.kff.org HIV Screening and Antiretroviral Treatment in the United / attachment/Financing-the-Response-to-HIV-in-Low States.” Annals of Internal Medicine 153 (12): 778–89. -and-Middle-Income-Countries-International-Assistance Lopez-Quintero, C., R. Shtarkshall, and Y. D. Neumark. 2005. -from-Donor-Governments-in-2015. “Barriers to HIV-Testing among Hispanics in the United Kalichman, S. C., C. Cherry, C. M. Amaral, C. Swetzes, L. Eaton, States: Analysis of the National Health Interview Survey.” and others. 2010. “Adherence to Antiretroviral Therapy and AIDS Patient Care and STDs 19 (10): 672–83. HIV Transmission Risks: Implications for Test-and-Treat Loutfy, M. R., W. Wu, M. Letchumanan, L. Bondy, T. Antoniou, Approaches to HIV Prevention.” AIDS Patient Care and and others. 2013. “Systematic Review of HIV Transmission STDs 24 (5): 271–77. between Heterosexual Serodiscordant Couples Where the Katz, M. H., S. K. Schwarcz, T. A. Kellogg, J. D. Klausner, HIV-Positive Partner Is Fully Suppressed on Antiretroviral J. W. Dilley, and others. 2002. “Impact of Highly Active Therapy.” PLoS One 8 (2): e55747. Antiretroviral Treatment on HIV Seroincidence among Lu, W., G. Zeng, J. Luo, S. Duo, G. Xing, and others. 2010. Men Who Have Sex with Men: San Francisco.” American “HIV Transmission Risk among Serodiscordant Couples: Journal of Public Health 92 (3): 388–94. A Retrospective Study of Former Plasma Donors in 108 Major Infectious Diseases Henan, China.” Journal of Acquired Immune Deficiency Influence Local HIV Shedding and HIV Target Cell Panel Syndromes 55 (2): 2332–38 Activation.” Journal of Infectious Diseases 209 (8): 1174–84. Maddali, M. V., D. W. Dowdy, A. Gupta, and M. Shah. Patterson, K. B., H. A. Prince, E. Kraft, A. J. Jenkins, 2015. “Economic and Epidemiological Impact of Early N. J. Shaheen, and others. 2011. “Penetration of Tenofovir Antiretroviral Therapy Initiation in India.” Journal of the and Emtricitabine in Mucosal Tissues: Implications for International AIDS Society 18: 20217. Prevention of HIV-1 Transmission.” Science Translational Margolis, D., C. Brinson, J. Eron, G. Richmond, R. LeBlanc, Medicine 3 (112): 112. and others. 2014. “744 and Rilpivirine as Two-Drug Oral Parrish, N. F., F. Gao, H. Li, E. E. Giorgi, H. J. Barbian, Maintenance Therapy: LAI116482 (LATTE) Week 48 and others. 2013. “Phenotypic Properties of Transmitted Results.” Special Issue: Abstracts from the 2014 Conference Founder HIV-1.” Proceedings of the National Academy of on Retroviruses and Opportunistic Infections. Topics in Sciences of the United States of America 110 (17): 6626–33. Antiviral Medicine International 22 (e-1): 34. PEPFAR (President’s Emergency Plan for AIDS Relief). 2012. Martin, E., J. M. Carlson, A. Q. Le, D. R. Chopera, R. McGovern, “Report on Pilot Expenditure Analysis of PEPFAR Programs and others. 2014. “Early Immune Adaptation in HIV-1 in Six Countries.” Department of State, Washington, DC. Revealed by Population-Level Approaches.” Retrovirology http://www.pepfar.gov/documents/organization/195700.pdf. 11 (64): 1–16. ———. 2014. PEPFAR 3.0 Controlling the Epidemic: Delivering McLaren, P. J., C. Coulonges, S. Ripke, L. van den Berg, on the Promise of an AIDS-free Generation. Washington, DC: S. Buchbinder, and others. 2013. “Association Study of The Office of the U.S. Global AIDS Coordinator. http:// Common Genetic Variants and HIV-1 Acquisition in 6,300 www.pepfar.gov/documents/organization/234744.pdf. Infected Cases and 7,200 Controls.” PLoS Pathogens 9 (7): ———. 2015. “2015 United Nations General Assembly e1003515. Sustainable Development Summit.” Fact Sheet, Department Meyer-Rath, G., and M. Over. 2012. “HIV Treatment as of State, Washington, DC. http://www.pepfar.gov/documents Prevention: Modelling the Cost of Antiretroviral /organization/247548.pdf. Treatment—State of the Art and Future Directions.” PLoS Pilcher, C. D., J. J. Eron Jr., S. Galvin, C. Gay, and M. S. Cohen. Medicine 9 (7): e1001247. 2004. “Acute HIV Revisited: New Opportunities for Montaner, J. S., E. Wood, T. Kerr, V. Lima, R. Barrios, and Treatment and Prevention.” Journal of Clinical Investigation others. 2010. “Expanded Highly Active Antiretroviral 113 (7): 937–45. Therapy Coverage among HIV-Positive Drug Users to Pilcher, C. D., G. Joak, I. F. Hoffman, F. E. A. Martinson, Improve Individual and Public Health Outcomes.” Journal C. Mapanje, and others. 2007. “Amplified Transmission of of Acquired Immune Deficiency Syndromes 55 (1): S5–S9. HIV-1: Comparison of HIV-1 Concentrations in Semen Muessig, K. E., and M. S. Cohen. 2014. “Advances in HIV and Blood during Acute and Chronic Infection.” AIDS 21 Prevention for Serodiscordant Couples.” Current HIV/AIDS (13): 1723–30. Reports 11 (4): 434–46. Ping, L. H., S. B. Joseph, J. A. Anderson, M. R. Abrahams, Nachega, J. B., O. A. Uthman, E. J. Mills, and T. C. Quinn. J. F. Salazar-Gonzalez, and others. 2013. “Comparison of 2013. “Adherence to Antiretroviral Therapy for the Success Viral Env Proteins from Acute and Chronic Infections of Emerging Interventions to Prevent HIV Transmission: with Subtype C Human Immunodeficiency Virus Type A Wake-Up Call.” Journal of AIDS and Clinical Research 1 Identifies Differences in Glycosylation and CCR5 (Suppl 4): 2039–52. Utilization and Suggests a New Strategy for Immunogen National AIDS and STI Control Programme. 2013. “Kenya Design.” Journal of Virology 87 (13): 7218–33. AIDS Indicator Survey 2012.” Ministry of Health, Porco, T. C., J. N. Martin, K. A. Page-Shafer, A. Cheng, Nairobi, Kenya. E. Charlebois, and others. 2004. “Decline in HIV Infectivity Nichols, B. E., R. Baltussen, J. H. van Dijk, P. Thuma, Following the Introduction of Highly Active Antiretroviral J. L. Nouwen, and others. 2014. “Cost-Effectiveness of Therapy.” AIDS 18 (1): 81–88. PrEP in HIV/AIDS Control in Zambia: A Stochastic Powers, K. A., M. E. Kretzschmar, and W. C. Miller. 2014. League Approach.” Journal of Acquired Immune Deficiency “Impact of Early-Stage HIV Transmission on Treatment as Syndromes 66 (2): 221–28. Prevention.” Proceedings of the National Academy of Sciences Novitsky, V., R. Wang, S. Lagakos, and M. Essex. 2010. “HIV-1 of the United States of America 111 (45): 15867–68. Subtype C Phylodynamics in the Global Epidemic.” Viruses Quinn, T. C., M. J. Wawer, N. Sewankambo, D. Serwadda, 2 (1): 33–54. C. Li, and others. 2000. “Viral Load and Heterosexual O’Brien, T. R., D. H. McDermott, J. P. Ionnidis, M. Carrington, Transmission of Human Immunodeficiency Virus Type P. M. Murphy, and others. 2000. “Effect of Chemokine 1. Rakai Project Study Group.” New England Journal of Receptor Gene Polymorphisms on the Response to Potent Medicine 342 (13): 921–29. Antiretroviral Therapy.” AIDS 14 (7): 821–26. Raifman, J., T. Chetty, F. Tanser, T. Mutevedzi, P. Matthews, Okano, J. T., and S. Blower. 2014. “Beware of Using Invalid and others. 2014. “Preventing Unintended Pregnancy and Transmission Models to Guide HIV Health Policy.” The HIV Transmission: Effects of the HIV Treatment Cascade Lancet Global Health 2 (5): e260. on Contraceptive Use and Choice in Rural KwaZulu- Olivier, A. J., L. Masson, K. Ronacher, G. Walzl, D. Coetzee, Natal.” Journal of Acquired Immune Deficiency Syndromes and others. 2014. “Distinct Cytokine Patterns in Semen 67 (Suppl 4): S218–27. Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 109 Reichelderfer, P. S., R. W. Coombs, D. J. Wright, J. Cohn, in Risk of HIV Acquisition in Rural KwaZulu-Natal, D. N. Burns, and others. 2000. “Effect of Menstrual Cycle South Africa.” Science 339 (6122): 966–71. on HIV-1 Levels in the Peripheral Blood and Genital Tract: TEMPRANO ANRS Study Group. 2015. “A Trial of Early WHS 001 Study Team.” AIDS 14 (14): 2101–7. Antiretrovirals and Isoniazid Preventive Therapy in Africa.” Rieder, P., B. Joos, V. von Wyl, H. Kuster, C. Grube, and New England Journal of Medicine 373: 808–22. others. 2010. “HIV-1 Transmission after Cessation of Early Thompson, C. G., M. S. Cohen, and A. D. M. Kashuba. 2013. Antiretroviral Therapy among Men Having Sex with Men.” “Antiretroviral Pharmacology in Mucosal Tissues.” Journal AIDS 24 (8): 1177–83. of Acquired Immune Deficiency Syndromes 63 (2): S240–47. Rodger, A. J., V. Cambiano, T. Bruun, P. Vernazza, S. Collins, and Wagner, B. G., and S. Blower. 2012. “Universal Access to HIV others. 2016. “Sexual Activity without Condoms and Risk Treatment versus Universal ‘Test and Treat’: Transmission, of HIV Transmission in Serodifferent Couples when the Drug Resistance, and Treatment Costs.” PLoS One 7 (9): HIV-Positive Partner Is Using Suppressive Antiretroviral e41212. Therapy.” Journal of the American Medical Association Walensky, R. P., A. D. Paltiel, E. Losina, B. L. Morris, C. A. Schott, 316 (2): 171–81. and others. 2010. “Test and Treat DC: Forecasting the Rosen, S., M. Ketlhapile, I. Sanne, and M. B. De Silva. 2007. Impact of a Comprehensive HIV Strategy in Washington, “Cost to Patients of Obtaining Treatment for HIV/AIDS in DC.” Clinical Infectious Diseases 51 (4): 392–400. South Africa.” South African Medical Journal 97 (7): 524–29. Walensky, R. P., E. L. Ross, N. Kumarasmy, R. Wood, F. Noubary, Royce, R. A., A. Seña, W. Cates, and M. S. Cohen. 1997. “Sexual and others. 2013. “Cost-Effectiveness of HIV Treatment Transmission of HIV.” New England Journal of Medicine as Prevention in Serodiscordant Couples.” New England 336: 1072–78. Journal of Medicine 369 (18): 1715–25. Shaw, G. M., and E. Hunter. 2012. “HIV Transmission.” Cold Wand, H., D. Wilson, P. Yan, A. Gonnermann, A. McDonald, Spring Harbor Perspectives in Medicine 2 (11): a006965. and others. 2009. “Characterizing Trends in HIV Infection Smit, M., K. Brinkman, S. Geerlings, C. Smit, K. Thyagarajan, among Men Who Have Sex with Men in Australia by and others. 2015. “Future Challenges for Clinical Care of an Birth Cohorts: Results from a Modified Back-Projection Ageing Population Infected with HIV: A Modelling Study.” Method.” Journal of the International AIDS Society 12: 19. The Lancet Infectious Diseases 15 (7): 810–18. Wang, L., G. E. Zeng, J. Luo, S. Duo, G. Xing, and others. 2010. Smith, M. K., K. A. Powers, K. E. Muessig, W. C. Miller, and “HIV Transmission Risk among Serodiscordant Couples: M. S. Cohen. 2012. “HIV Treatment as Prevention: The A Retrospective Study of Former Plasma Donors in Henan, Utility and Limitations of Ecological Observation.” PLoS China.” Journal of Acquired Immune Deficiency Syndromes Medicine 9 (7): e1001260. 55 (2): 232–38. Sorensen, S. W., S. L. Sansom, J. T. Brooks, G. Marks, Waning, B., W. Kaplan, A. C. King, D. A. Lawrence, E. M. Begier, and others. 2012. “A Mathematical Model of H. G. Leufkens, and others. 2009. “Global Strategies to Comprehensive Test-and-Treat Services and HIV Incidence Reduce the Price of Antiretroviral Medicines: Evidence among Men Who Have Sex with Men in the United States.” from Transactional Databases.” Bulletin of the World Health PLoS One 7 (2): e29098. Organization 87 (7): 520–28. Spielberg, F., B. M. Branson, G. M. Goldbaum, D. Lockhart, Weinstein, M. C., J. E. Siegel, M. R. Gold, M. S. Kamlet, and A. Kurth, and others. 2003. “Overcoming Barriers to HIV L. B. Russell. 1996. “Recommendations of the Panel on Testing: Preferences for New Strategies among Clients Cost-Effectiveness in Health and Medicine.” Journal of the of a Needle Exchange, a Sexually Transmitted Disease American Medical Association 276 (15): 1253–58. Clinic, and Sex Venues for Men Who Have Sex with Men.” WHO (World Health Organization). 2010. Antiretroviral Journal of Acquired Immune Deficiency Syndromes 32 (3): Therapy for HIV Infection in Adults and Adolescents: 318–27. Recommendations for a Public Health Approach. Geneva: Stolte, I. G., N. H. Dukers, R. B. Geskus, R. A. Coutinho, and WHO. J. B. de Wit. 2004. “Homosexual Men Change to Risky ———. 2012a. Guidance on Couples HIV Testing and Sex When Perceiving Less Threat of HIV/AIDS since Counselling, Including Antiretroviral Therapy for Treatment Availability of Highly Active Antiretroviral Therapy: A and Prevention in Serodiscordant Couples. Geneva: WHO. Longitudinal Study.” AIDS 18 (2): 303–9. ———. 2012b. Programmatic Update: Use of Antiretroviral Supervie, V., J. P. Viard, D. Costagliola, and R. Breban. 2014. Drugs for Treating Pregnant Women and Preventing HIV “Heterosexual Risk of HIV Transmission per Sexual Act Infection in Infants. Geneva: WHO. under Combined Antiretroviral Therapy: Systematic ———. 2016. Consolidated Guidelines on the Use of Review and Bayesian Modeling.” Clinical Infectious Diseases Antiretroviral Drugs for Treating and Preventing HIV 59 (1): 115–22. Infection: Recommendations for a Public Health Approach. Taha, T. E., D. R. Hoover, G. A. Dallabetta, N. I. Kumwenda, 2nd ed. Geneva: WHO. L. A. Mtimavalve, and others. 1998. “Bacterial Vaginosis and Wilson, D., and N. Fraser. 2014. “Controlling the HIV Disturbances of Vaginal Flora: Association with Increased Epidemic with Antiretrovirals: Who Pays and Why? Acquisition of HIV.” AIDS 12 (13): 1699–706. Costs, Effectiveness, and Feasibility of HIV Treatment Tanser, F., T. Bärnighausen, E. Grapsa, J. Zaidi, and M. L Newell. as Prevention.” Clinical Infectious Diseases 59 (Suppl 1): 2013. “High Coverage of ART Associated with Decline S28–31. 110 Major Infectious Diseases Wolbers, M., A. Babiker, C. Sabin, J. Young, M. Dorrucci, and Need for Future Research.” CHE Research Paper 109, Centre others. 2010. “Pretreatment CD4 Cell Slope and Progression for Health Economics, University of York, United Kingdom. to AIDS or Death in HIV-Infected Patients Initiating Ying, R., M. Sharma, R. Heffron, C. L. Celum, J. M. Baeten, Antiretroviral Therapy: The CASCADE Collaboration; A and others. 2015. “Cost-Effectiveness of Pre-Exposure Collaboration of 23 Cohort Studies.” PLoS Medicine 7 (2): Prophylaxis Targeted to High-Risk Serodiscordant Couples e1000239. as a Bridge to Sustained ART Use in Kampala, Uganda.” Wood, E., T. Kerr, B. D. L. Marshall, K. Li, R. Zhang, and others. Journal of the International AIDS Society 18 (Suppl 3): 20013. 2009. “Longitudinal Community Plasma HIV-1 RNA Zhang, H., G. Dornadula, M. Beumont, L. Livornese, B. Van Concentrations and Incidence of HIV-1 among Injecting Uitert, and others. 1998. “Human Immunodeficiency Virus Drug Users: Prospective Cohort Study.” BMJ 338 (b1649). Type 1 in the Semen of Men Receiving Highly Active Woods, B., P. Revill, M. Sculpher, and K. Claxton. 2015. “Country- Antiretroviral Therapy.” New England Journal of Medicine Level Cost-Effectiveness Thresholds: Initial Estimates and the 339 (25): 1803–9. Effectiveness and Cost-Effectiveness of Treatment as Prevention for HIV 111 Chapter 6 Prevention of Mother-to-Child Transmission of HIV and Syphilis Grace John-Stewart, Rosanna W. Peeling, Carol Levin, Patricia J. Garcia, David Mabey, and John Kinuthia INTRODUCTION its elimination potentially more feasible. The rollout of The past decade has yielded enormous progress in the PMTCT of syphilis may be hampered by less political prevention of mother-to-child transmission (PMTCT) will to implement it than PMTCT of HIV, lack of of the human immunodeficiency virus (HIV) (UNAIDS accountability for monitoring PMTCT of syphilis, 2015). Interventions that decrease mother-to-child inconsistent availability of diagnostic tests, and use of transmission (MTCT) of HIV from more than 30 tests that are not POC (WHO 2006). PMTCT of syphilis percent to 1 percent have been identified. Decentralized could readily leverage advances in PMTCT of HIV by point-of-care (POC) approaches for detecting maternal using these programs to enhance the supply chain, lab HIV and immediate provision of comprehensive antiret- testing, and accountability for prompt syphilis diagnosis roviral treatment (ART) have resulted in rapid decreases and treatment. in the number of HIV-infected infants. Indeed, PMTCT The current momentum in PMTCT of HIV offers has been credited with driving observed decreases in a unique opportunity to accelerate PMTCT of HIV HIV incidence overall. and syphilis concurrently. Combining interventions for The guidelines for PMTCT of HIV have been PMTCT of HIV and syphilis adds minimal cost while dynamic: the World Health Organization (WHO) and potentially benefiting twice as many mother-infant pairs Joint United Nations Programme on AIDS (UNAIDS) as interventions focused solely on either HIV or syphilis. have revised their recommendations every two to five Health systems improvements for rapid diagnosis and years, most recently to recommend combination lifelong treatment, partner engagement, and follow-up of ART for all pregnant HIV-infected women (WHO mothers and infants can enhance both types of PMTCT 2014a). The term elimination of mother-to-child trans- programs. Combined PMTCT of HIV and syphilis will mission (EMTCT) was used to further spur global efforts yield sustained and important benefits for women and to virtually eliminate pediatric HIV by 2015 (UNAIDS children worldwide. 2011; WHO 2014a). This chapter reviews the rates, burden, and conse- PMTCT of syphilis (Treponema pallidum) has not quences of mother-to-child transmission of HIV and received the same amount of attention as PMTCT of syphilis; the effectiveness of interventions to decrease HIV, although syphilis is estimated to affect more chil- transmission; the estimated cost-effectiveness of these dren globally than HIV. PMTCT of syphilis requires less interventions; and several successful PMTCT pro- costly and less intensive interventions than HIV, making grammatic approaches. The chapter also highlights Corresponding author: Grace John-Stewart, Department of Global Health, University of Washington, Seattle, Washington, United States; gjohn@uw.edu. 113 opportunities for integrated programming to effi- estimates), and because of varied infant outcomes attrib- ciently decrease the number of infants infected with utable to maternal infection, and differences in the tim- these chronically debilitating pathogens. Because syph- ing and routes of transmission. Diagnosis of congenital ilis testing is already currently recommended by WHO syphilis is clinical and imprecise. In contrast, infant HIV for all pregnant women (WHO 2006) and is at least diagnosis is based on a robust replicable measure— partially implemented in most antenatal clinics, new detection of HIV virus—that is highly sensitive and costs for improving the program may be minimal: specific. adapting programs to incorporate dual POC tests to diagnose syphilis, providing training to improve adher- HIV Transmission ence to guidelines, and increasing accountability for MTCT of HIV occurs either during pregnancy, at deliv- tracking outcomes in PMTCT of syphilis within clinics ery, or through breastfeeding. Without intervention, can be added to existing programs with limited addi- HIV MTCT rates range from 20 percent to 35 percent tional investment. in breastfed infants and from 15 percent to 20 percent In 2015, Cuba became the first country to eliminate in nonbreastfed infants (table 6.1) (De Cock and others perinatal HIV and syphilis (WHO 2015). This experi- 2000). Cofactors of MTCT of HIV include maternal ence demonstrates that the goal of dual elimination viral burden (both systemically and in mucosal com- is attainable and feasible with effective integration of partments to which the baby is exposed, such as geni- PMTCT of HIV and syphilis. Children do not need to tal secretions or breast milk), immunosuppression, suffer from the consequences of either of these devastat- and preterm birth (John and Kreiss 1996; John and ing infections when these PMTCT programs function others 2001). synergistically. Syphilis Transmission MTCT of syphilis predominantly occurs during preg- GLOBAL BURDEN AND CONSEQUENCES OF nancy, but few studies have been conducted on the MTCT OF HIV AND SYPHILIS estimated risk. The precision and accuracy of syphilis MTCT risk estimates are limited by the study design used MTCT Rates and Cofactors (case control), varied diagnostic tests for maternal Comparisons of MTCT of syphilis and MTCT of HIV syphilis, and lack of a good diagnostic marker of infant are difficult because of differences in the precision of infection. In a systematic review, Gomez and others infant diagnosis (and thus in the precision of MTCT risk (2013) screened 3,258 records and identified six studies Table 6.1 Mother-to-Child Transmission of Syphilis and HIV, Selected Findings Metric Syphilis HIV Estimated annual number of 1.36 million (summary from 2008 data 1.45 million (UNAIDS 2013) pregnant women infected worldwide [Newman and others 2013]) Timing of transmission In utero In utero, intrapartum, postnatal (John and Kreiss 1996) Method for detecting infant infection Clinical manifestations, cerebrospinal fluid, Detection of HIV virus by nucleic acid amplification tests and radiological Transmission risk 15.5 percent of infants born to mothers with 20 to 35 percent infant HIV transmission depending untreated syphilis have clinical signs of on breastfeeding duration in the absence of maternal congenital syphilis (Gomez and others 2013) treatment or infant prophylaxis (De Cock and others 2000) Adverse infant outcomes: fetal 4.6-fold increased risk of combined adverse 2-fold to 4-fold increased risk of combined adverse death, preterm birth, stillbirth, outcomes (66.5 percent among mothers with outcomes (Brocklehurst and French 1998) neonatal death, low birth weight syphilis vs. 14.3 percent among mothers without syphilis) (Gomez and others 2013) Cofactors for MTCT Maternal RPR level (Watson-Jones, Maternal HIV viral load, other sexually transmitted Changalucha, and others 2002) infections, CD4 count, route of delivery, infant breastfeeding (John and others 2001) Note: CD4 = cluster of differentiation 4 (blood count); HIV = human immunodeficiency virus; MTCT = mother-to-child transmission; RPR = rapid plasma reagin. 114 Major Infectious Diseases conducted between 1917 and 2000 (all case-control stud- Maternal-Infant Survival Everywhere (PROMISE) ies) to estimate the MTCT risk of syphilis. The estimated study, which enrolled more than 3,500 mother-infant rates of congenital syphilis (diagnosed in infants show- pairs, compared two types of antiretroviral regimens ing signs of clinical infection) ranged from 2.2 percent to for pregnant women who did not meet cluster of dif- 40.9 percent, with a pooled MTCT rate of 15.5 percent ferentiation 4 (CD4) or clinical criteria for triple ART. (table 6.1). Cofactors of syphilis MTCT remain unde- One regimen type, Option A, was a short-course fined (Gomez and others 2013); some evidence suggests antiretroviral regimen that included different single or that more recent maternal active syphilis (with high double drug regimens in pregnancy, delivery, and post- rapid plasma reagin [RPR] titer) is associated with partum.2 Option B was a combined three-drug ART increased adverse infant outcomes (Watson-Jones, regimen administered to all women during pregnancy Changalucha, and others 2002). and postpartum.3 Both regimens provided preventive In contrast to studies of MTCT of HIV, outcomes antiretrovirals for breastfeeding MTCT; Option A of maternal syphilis often emphasize estimated attrib- relied on infant prophylaxis, while Option B relied on utable adverse infant outcomes in addition to infant maternal ART with a six-week period of infant prophy- infections, because of difficulties in infant diagnosis laxis. Although Option A was inferior to Option B for and strong evidence of numerous additional adverse PMTCT in this randomized controlled trial, Option B outcomes. Accordingly, combined adverse infant out- was associated with higher rates of prematurity, low comes prevented by treatment of maternal syphilis are birth weight, and neonatal death.4 Among the two used to estimate cost-effectiveness. Gomez and others’ Option B alternatives, lamivudine-ART was safer for (2013) systematic review attributed multiple adverse infants than tenofovir-ART. Thus, the PROMISE study infant outcomes—including spontaneous abortion, showed that the antiretroviral regimens a woman stillbirth, fetal death, preterm birth, low birth weight, receives can affect both MTCT rates and rates of neonatal death, and congenital syphilis—to untreated adverse infant outcomes. The WHO guidelines recom- syphilis. The pooled frequency of these adverse infant mend Option B+, both for enhanced PMTCT effec- outcomes was 66.5 percent in mothers with syphilis, tiveness and for programmatic feasibility. Future compared with 14.3 percent in mothers without syph- guidelines and programs may adapt specific Option B+ ilis (Gomez and others 2013). The authors noted antiretroviral regimens to concurrently decrease HIV marked heterogeneity and potential for bias in the and other adverse child outcomes. estimates. Newer syphilis molecular diagnostics (poly- merase chain reaction [PCR] assays) have greater than Maternal HIV and Syphilis Coinfection 70 percent sensitivity. To date, however, these newer Some evidence suggests that women coinfected with diagnostic tests for syphilis have not been used to HIV and syphilis may have greater than twofold estimate MTCT syphilis rates (Grimprel and others increased risk of HIV transmission than women with 1991; Palmer and others 2003; Sanchez and others HIV infection alone; however, some studies have not 1993; Shields and others 2012). found this association (Lee and others1998; Mwapasa and others 2006; Schulte and others 2001; Yeganegh Other Infant Impacts from MTCT of HIV and others 2015). Children of coinfected mothers may Estimates of the impact of maternal HIV infection on have HIV-syphilis coinfection and poorer outcomes infants have focused predominantly on HIV transmis- than those with either infection alone (Mwapasa and sion risk itself, without including multiple adverse infant others 2006). outcomes. However, untreated maternal HIV may con- tribute to adverse infant outcomes in addition to infant HIV infection, including fetal death, preterm birth, Maternal and Pediatric Burden of HIV and Syphilis infant mortality, stillbirth, and low birth weight. In a Maternal Burden of HIV and Syphilis systematic review of 31 studies, HIV-infected mothers There are distinct regional and global patterns of had increased risk of spontaneous abortion, perinatal maternal HIV and syphilis. In Sub-Saharan Africa, for mortality, infant mortality, intrauterine growth retarda- example, both HIV and syphilis are highly prevalent; tion, low birth weight, and preterm birth (Brocklehurst however, the distributions of HIV infection and syphilis and French 1998).1 vary distinctly by region. Consistently assessed national In addition, some of the new antiretroviral regi- HIV and syphilis prevalence estimates are not pro- mens for PMTCT may confer increased risk of adverse duced; however, South Africa, which has a higher HIV outcomes unrelated to HIV (Townsend and others prevalence, appears to have lower antenatal syphilis 2007; Townsend and others 2010). The Promoting prevalence than do other southern and eastern African Prevention of Mother-to-Child Transmission of HIV and Syphilis 115 countries. China has higher syphilis but lower HIV 150,000 children younger than age 15 years were prevalence than the Russian Federation. Although syph- newly infected, as shown in map 6.3 (UNAIDS 2016a). ilis and HIV are both sexually transmitted infections This is a modeled estimate with confidence limits (STIs), their distinct regional distribution may stem ranging from 110,000 to 190,000. from differences in transmission epidemiology, sexual Although the annual number of HIV-infected infants networks, treatment program effectiveness and cover- has been decreasing, 1.8 million children younger than age, notification and tracing guidelines, and measure- age 15 years are living with HIV, predominantly acquired ment methods (WHO 2013). through MTCT before the expansion of effective PMTCT Map 6.1 shows the global distribution of HIV programs, as shown in map 6.4 (UNAIDS 2016a). Many prevalence in women ages 15–24 years, and map 6.2 of these children remain undiagnosed and untreated shows the distribution of maternal antenatal syphilis until they become symptomatic. Although additional seroprevalence. children older than age 15 years were infected with HIV in utero or at childbirth, it is difficult to estimate their Pediatric Burden of HIV numbers. Annually, an estimated 1.5 million pregnant women worldwide are HIV infected, which, during the peak Pediatric Burden of Syphilis of the HIV epidemic in the mid-1990s, resulted in Similar maps for congenital syphilis are not available, more than 500,000 infant HIV infections per year although map 6.5 approximates relative burdens by (UNAIDS 2013). However, with active PMTCT pro- country. Children with syphilis are harder to diagnose, grams, infant HIV infections have steadily declined to map, and count than those with HIV. Worldwide between about half that peak level; in 2015, an estimated 1997 and 2003, an estimated 2 million women with Map 6.1 Global HIV Prevalence among Women Ages 15–24 Years, 2014 IBRD 42548 | OCTOBER 2016 >2.03 2.03–3.96 3.96–5.89 5.89–7.82 7.82–9.75 9.75–11.68 11.68–13.61 >13.61 No data Source: World Bank (http://en.actualitix.com/country/wld/prevalence-of-hiv-women-15-24-years.php). Note: HIV = human immunodeficiency virus. 116 Major Infectious Diseases Map 6.2 Global Syphilis Seroprevalence among Pregnant Women, 2012 IBRD 42549 | DECEMBER 2016 Europe: 0.16% Mediterranean: 0.06% Pacific: 0.33% Americas: 0.84% Asia: Africa: 0.62% 2.13% >5% _ >1.0 to 4.9% _ >0.5 to 0.9% _ _0.5% < Data not available, regional median used Source: Reprinted from “Global Health Estimates.” 2013. Online Database. WHO, Geneva. syphilis became pregnant, resulting in 728,000–1,527,000 resulting in 360,000 adverse outcomes, including new cases of congenital syphilis each year (Schmid and 150,000 early fetal deaths or stillbirths, 50,000 preterm others 2007). This broad range in estimates of congenital or low-birth-weight infants, 60,000 neonatal deaths, syphilis is a result of poor ability to measure cases. and 110,000 infants with congenital infection (WHO Other infant complications may also be partially 2013). extrapolated from maternal syphilis rates. The WHO estimated that, in 2008, 1.36 million pregnant women worldwide had active syphilis, of whom 80 percent had Consequences of Pediatric HIV and Syphilis attended antenatal clinics (ANCs). Among those Consequences of Pediatric HIV women, syphilis was responsible for more than 500,000 Infants who are infected with HIV through MTCT typi- adverse pregnancy outcomes, including more than cally have a rapidly progressive course: about half die 200,000 stillbirths or early fetal deaths, 92,000 neonatal within two years (Newell and others 2004; Obimbo and deaths, 65,000 preterm or low-birth-weight infants, others 2004). Infants with later MTCT of HIV through and 152,000 infected newborns (Newman and others breastfeeding may have a more indolent course than 2013). Two-thirds of these adverse outcomes occurred infants with in utero or peripartum HIV acquisition in women who had attended ANCs but were not (Becquet and others 2012; Obimbo and others 2009). screened or treated for syphilis (Newman and others Children with untreated pediatric HIV infection 2013). The WHO estimates for 2012 illustrate a decline have high risk of early mortality, severe malnutrition, from 2008: 950,000 maternal syphilis infections and growth faltering, as well as recurrent infections, Prevention of Mother-to-Child Transmission of HIV and Syphilis 117 Map 6.3 Estimated Number of Children Younger than Age 15 Years Newly Infected with HIV, by Region, 2015 IBRD 42550 | DECEMBER 2016 Eastern Europe and Central Asia North America and ... Western and Central Europe ... Middle East and North Africa 2,100 [1,400–3,200] Asia and the Pacific Western and 19,000 Central Africa [16,000–21,000] Latin America and 66,000 the Caribbean [47,000–87,000] Eastern and 2,100 Southern Africa [1,600–<2,900] 56,000 [40,000–76,000] Total: 150,000 [110,000–190,000] Source: UNAIDS 2015. Reproduced, with permission, from UNAIDS; further permission required for reuse. Note: HIV = human immunodeficiency virus. The estimated 150,000 new infections worldwide only include children younger than age 15 years. including opportunistic infections such as tuberculosis symptoms including rash; skeletal changes; hepato- (Obimbo and others 2004; Obimbo and others 2009), splenomegaly; and neurologic, renal, pulmonary, or and neurocognitive delays. If given ART early, infants ocular involvement. Moreover, congenital syphilis have significantly lower mortality but may have contin- may result in lifelong disability, particularly when ued deficiencies in growth, persistent morbidity, and undetected or detected late in infancy (Arnold and compromised neurocognitive ability (Wamalwa and Ford-Jones 2000). others 2010). A study of women who had not been screened for Relative to uninfected mothers, HIV-infected syphilis in ANCs and delivered babies in a Tanzanian mothers may have increases in other adverse out- hospital found that, among those with serological comes of pregnancy, including stillbirth, prematurity, evidence of active syphilis, 25 percent delivered a still- and low birth weight (Brocklehurst and French born baby, 20 percent a premature baby, and 33 percent 1998). Their infants, if uninfected but HIV exposed, a low-birth-weight baby. Overall, adverse events were also have increased risk of morbidity and mortality noted in 49 percent of infants born to those women, compared with infants unexposed to HIV, perhaps compared with 11 percent of women without syphilis because of increased immunologic susceptibility, (Watson-Jones, Changalucha, and others 2002). A sys- sociodemographic factors, or increased exposure to tematic review of the impact of untreated syphilis on other infectious diseases (Afran and others 2014; pregnancy outcomes found a consistently higher pro- Mofenson and Watts 2014). portion of adverse pregnancy outcomes in women with untreated syphilis than in uninfected women Consequences of Pediatric Syphilis (Gomez and others 2013). The pooled estimate of Children with congenital syphilis have a range of neonatal death was 12.3 percent in women with presentations, from asymptomatic to a variety of syphilis and 3 percent in women without syphilis. 118 Major Infectious Diseases Map 6.4 Children Younger than Age 15 Years Estimated to be Living with HIV, 2015 IBRD 42551 | DECEMBER 2016 Eastern Europe and Central Asia North America and ... Western and Central Europe ... Middle East and North Africa 11,000 [8,600–14,000] Asia and the Pacific Western and 190,000 Central Africa [140,000–170,000] Latin America and 500,000 the Caribbean [400,000–630,000] Eastern and 32,000 Southern Africa [27,000–38,000] 1 million [930,000–1.2 million] Estimated total number of pediatric HIV infections, by region, 2015 Total: 1.8 million [1.5 million–2 million] Source: UNAIDS 2015. Reproduced, with permission, from UNAIDS; further permission required for reuse. Note: HIV = human immunodeficiency virus. The estimated 1.8 million infections worldwide only include children younger than age 15 years. The pooled estimate for stillbirth or prematurity HIV testing, which can provide results during the same was 25.6 percent and for low birth weight was 12.1 visit, significantly increases the proportion of women percent among the infants of women with syphilis who receive their test results (Malonza and others (Gomez and others 2013). Treatment of pregnant 2003).6 Rapid HIV testing enables testing of women of women is estimated to avert these outcomes as out- unknown HIV status at any time they present to the lined in table 6.2. health care system, including at delivery. Testing for HIV needs to include not only diagnostic services but also careful counseling of women. Peer EFFECTIVENESS AND COVERAGE OF PMTCT counselors and mother-to-mother models have been INTERVENTIONS critically important for helping women cope with their diagnoses, make decisions about disclosure of their Effectiveness of Interventions for PMTCT of HIV results to their partners, and adhere to their medication Identification of HIV during Pregnancy and in Infants regimen (Futterman and others 2010; Shetty and others HIV testing during pregnancy initially used the enzyme- 2008; Shroufi and others 2013). linked immunosorbent assay (ELISA) test with opt-in Women may become infected with HIV during counseling, which resulted in attrition because women pregnancy or postpartum, and repeat HIV testing is elected either not to have the test or not to return for advised to detect and treat women with HIV serocon- results.5 The introduction of opt-out HIV testing—in version during this period. As PMTCT of HIV expands, which routine HIV testing is offered to all women unless women diagnosed with HIV at or before their first a woman opts out—has substantially increased the antenatal visit routinely receive HIV treatment. number of women who are tested for HIV (Creek and However, mothers who are initially seronegative but others 2007; Day and others 2004). In addition, rapid acquire HIV infection in pregnancy or postpartum are Prevention of Mother-to-Child Transmission of HIV and Syphilis 119 Map 6.5 Relative Global Concentrations of Congenital Syphilis, by Country IBRD 42552 | DECEMBER 2016 Countries with highest known burden of congenital syphilis Countries with known burden of congenital syphilis Non-focus countries for the initiative for the global elimination of congenital syphilis No data Source: Reprinted from “Advancing MDGs 4, 5 and 6: Impact of Congenital Syphilis Elimination.” Partner Brief, WHO, Geneva; http://apps.who.int/iris/bitstream/10665/70331/1/WHO_RHR _HRP_10.01_eng.pdf. 2010a. Note: The Initiative for the Global Elimination of Congenital Syphilis was launched by the WHO in 2007 to support global efforts to achieve several of the Millennium Development Goals (MDGs): MDG 4 (reduce child mortality), MDG 5 (improve maternal health), and MDG 6 (combat HIV/AIDS, malaria, and other diseases) (WHO 2010a). Table 6.2 Adverse Infant Outcomes Potentially often undiagnosed and contribute to an increasing pro- Avertable by Treatment of Maternal Syphilis portion of infant HIV infections that may not be detected until the child presents with symptomatic Outcomes averted disease (Drake and others 2014). Thus, repeat HIV test- Outcome by treatment (%) ing during pregnancy and postpartum is recommended, Stillbirth 13.3 although guidelines for frequency and timing of repeat Miscarriage 8.1 HIV tests in follow-up vary by national setting. Neonatal death (age 0–28 days) 9.3 Infants are born with maternal antibodies to HIV, Prematurity or low birth weight 5.8 making HIV serologic testing unhelpful for new- Infant with clinical evidence of syphilis 19.4 borns. HIV DNA can be diagnosed from a dried Nonneonatal infant death (age 29–365 days) 3.4 blood spot (DBS) using PCR assays in a central labo- Any adverse outcome 48.7* ratory. DBS HIV DNA virologic rather than serologic Source: LSHTM 2011. testing is recommended for infants (WHO 2010b). Note: “Treatment” refers to penicillin in pregnancy. * = “Any adverse outcome” is less than Early infant diagnosis (EID) is conducted using DBS the sum of the individual outcomes because of differences in weighting and variance that HIV DNA tests at age six weeks, followed by testing at occur when doing a meta-analysis, as well as the possibility of an infant’s having more than one outcome. age nine months and again at six weeks following 120 Major Infectious Diseases cessation of breastfeeding (Luzuriaga and Mofenson Women should be tested antenatally at their first 2016). New POC assays for HIV DNA are becoming visit in pregnancy, preferably before 16 weeks’ gesta- available that may be useful for decreasing turn- tion, and retested in the third trimester. Women who around time for EID results and expediting treatment have not been tested before delivery should be tested at of infants with HIV. delivery. Infants are born with maternal antibodies, limiting Antiretroviral Treatment for PMTCT of HIV diagnostic utility of routine treponemal or nontrepone- ART to decrease MTCT of HIV has been assessed in mal tests. Clinical diagnosis may be made in symptom- numerous randomized clinical trials (Siegfried and atic infants, with signs including mucocutaneous others 2011). Initial simple short-course regimens, lesions, bone changes evident on radiographs, syphilitic such as single-dose maternal and infant nevirapine rhinitis (“snuffles”), lymphadenopathy, and hemato- administered to the mother during labor and to logic changes. infants, decreased transmission by 50 percent. Triple- ART and infant prophylaxis regimens increase inter- Penicillin Treatment for PMTCT of Syphilis vention efficacy.7 Evidence supporting the use of ART Women with evidence of syphilis from either a trepone- for PMTCT of HIV is Grading of Recommendations mal or nontreponemal test should receive treatment Assessment, Development and Evaluation (GRADE) with penicillin (WHO 2006). In Tanzania, women with A1 (with large decreases in HIV transmission risk— syphilis who received a single dose of benzathine peni- declining from more than 30 percent to less than cillin before 28 weeks’ gestation had the same incidence 1 percent) (WHO 2012b and 2014b). of adverse pregnancy outcomes as women without syphilis (Watson-Jones, Gumodoka, and others 2002). A systematic review of the impact of penicillin treat- Effectiveness of Interventions for PMTCT of Syphilis ment on pregnancy outcomes in women with syphilis showed that treatment with at least 2.4 million units of Identification of Syphilis during Pregnancy and penicillin reduced the incidence of clinical congenital in Infants syphilis by 97 percent (relative risk 0.03) (Blencowe and Because most pregnant women with syphilis do not others 2011). The pooled estimate for reduction of still- have symptoms, a serological screening test is needed births was 82 percent, 64 percent for reduction in pre- to identify those who are infected. A nontreponemal term delivery, and 80 percent for reduction in neonatal agglutination test such as the RPR test has traditionally death (Blencowe and others 2011). The effect estimates been used. This test is cheap (less than US$0.10) and were large and consistent across studies, leading to a rapid, but the reagents need to be refrigerated, serum strong recommendation for screening and treatment needs to be separated from whole blood, and a plate according to the GRADE criteria (Blencowe and others shaker is needed, meaning that a reliable electricity 2011). Seropositive women should receive benzathine supply is required. Simple, lateral-flow POC tests have penicillin (2.4 million units), and partners should also become available in recent years; these tests are sensi- receive penicillin treatment. Infants of these women tive, specific, can be stored at ambient temperature, should receive 50,000 units/kilogram of benzathine and can be performed with a sample obtained by penicillin if asymptomatic or 10 days of crystalline or finger prick (Jafari and others 2013; Mabey and others procaine penicillin if symptomatic. The combined cost 2006). These tests detect treponemal antibodies of diagnosis and penicillin treatment of maternal syph- and cannot distinguish between active and past or ilis is less than US$1. treated infection. However, given the disease complica- tions and the simplicity and effectiveness of treatment (penicillin), these tests can provide a net public health benefit (Kuznik and others 2013). A dual POC test for PMTCT Implementation and Coverage HIV and syphilis has become available that detects PMTCT of HIV both treponemal and nontreponemal antibodies and is Concerns about stigma and loss of confidentiality both sensitive and specific, with good test performance motivated the initial intensive, opt-in voluntary coun- for the diagnosis of active syphilis (Yin and others seling and HIV testing models. Routinizing HIV test- 2013). The dual POC test is a pragmatic and attractive ing has led to much more efficient systems for PMTCT approach to promoting integration of PMTCT of HIV- of HIV (Creek and others 2007). HIV stigma may syphilis programs (Kiarie and others 2015). continue to inhibit the likelihood of HIV testing or Prevention of Mother-to-Child Transmission of HIV and Syphilis 121 ART adherence, but health-system bottlenecks appear linked to infant medical records, which would facilitate to be a comparable or greater barrier to PMTCT tracking of maternal-to-infant outcomes (Chi, Bolton- implementation (Kinuthia and others 2011). On the Moore, and Holmes 2013). EID-of-HIV programs positive side, male partner engagement may enhance involve the collection of DBS from infants at age six maternal adherence to antenatal and postnatal care weeks; the DBS are sent to a central laboratory for HIV visits and ART, while also improving infant outcomes PCR testing for detection of HIV DNA. Results from (Farquhar and others 2001; Taha and others 2007). EID programs have unacceptably long turnaround times Other implementation issues relevant to PMTCT effec- (Sutcliffe and others 2014; Woldesenbet and others tiveness include the following: 2014). Consequently, mothers often fail to receive infant ART Delays and Adherence Constraints. Previous results or remain unaware of infant HIV diagnosis, PMTCT antiretroviral regimens, such as Option A, despite testing. Children may not get a diagnosis of HIV required waiting for CD4 count results to determine until they become ill, resulting in poor long-term out- ART eligibility, which delayed ART initiation and com- comes. New POC EID assays may circumvent problems promised PMTCT programmatic effectiveness. with existing infant HIV diagnostic systems (Jani and Options B and B+ use an accelerated test-and-treat others 2014). EID detects perinatal infant HIV infec- approach, with immediate ART following HIV diagno- tions, but infants with negative early HIV tests remain at sis without waiting for a CD4 count (Taha and others risk throughout the breastfeeding period, and repeat 2007; WHO 2012b).8 Long-term adherence to ART testing is important. In addition, in settings of high HIV may decline for PMTCT regimens postpartum. In an prevalence, new maternal HIV infection acquired during early Option B+ program model in Malawi, initiation the pregnancy and breastfeeding periods contributes of Option B+ led to a rapid increase in the number of appreciably to infant HIV infections despite good women receiving ART, but only 77 percent of women PMTCT programs (Drake and others 2014). remained on ART at one year following delivery (CDC Late Postnatal Follow-Up during Breastfeeding. 2013). Option B or B+ also may decrease sexual trans- Following the six-week postnatal visit, retention mission of HIV to male HIV-uninfected partners rela- decreases, and long-term follow-up to exclude or diag- tive to Option A (Cohen and others 2011). The health nose infant HIV is erratic. With the scale-up of Option costs and benefits of Option B versus Option B+ are B+, retention beyond the early infant period needs to be not well defined (Watts and others 2009). However, improved. between pregnancies, Option B+ with continued ART will encounter fewer health-system bottlenecks than PMTCT of Syphilis Option B, in which episodic ART is administered only A 1996–97 survey of health ministries in 22 countries in during pregnancy and breastfeeding. Specifically, the Sub-Saharan Africa concluded that, although 73 percent maternal and child health (MCH) system will not need of women attended ANCs overall, and although syphilis to restart ART; however, there may be supply chain screening of ANC attendees was national policy in nearly challenges in consistent drug procurement. It is also all countries, only 38 percent of women were estimated not clear whether young asymptomatic women will to actually have been screened for syphilis (Gloyd, Chai, maintain long-term adherence to Option B+ between and Mercer 2001). Reported reasons for not performing pregnancies, or which strategies may optimize long- screening in this survey included costs of testing, treat- term adherence to ART. ment, and transport; inadequate prioritization; sociocul- Lack of Tailored Counseling Approaches. Other tural resistance; and lack of compliance or awareness by issues facing PMTCT programs require improvement or health care workers. In Tanzania, where screening and innovation. As PMTCT programs and HIV care pro- treatment of ANC attendees for syphilis is national pol- grams expand, an increasing number of women will have icy, a survey in nine districts found that only 43 percent been previously diagnosed with HIV and will have begun of 2,256 ANC attenders had been screened, and only receiving treatment before they become pregnant. 61 percent of seropositive women and 37 percent of their Stratified counseling approaches to previously diag- partners had been treated. Watson-Jones and others nosed and treated women versus newly diagnosed (2005) found that adequate training, continuity of sup- women do not exist. With Option B+, newly diagnosed plies, supervision, and quality control are critical ele- women initiate ART for life in the context of pregnancy ments for effective antenatal services but were frequently and need counseling and other interventions to motivate overlooked. The WHO noted that in 2012, 95 percent long-term ART adherence. coverage of syphilis testing was achieved in only Lack of Diagnostic System Tracking and 29 percent of 51 low- and middle-income countries Coordination. Medical records of mothers are often not (LMICs) surveyed.9 122 Major Infectious Diseases A review of syphilis screening in 13 ANCs in Nairobi, pregnancy, at which time routine HIV and syphilis test- Kenya, where blood was sent to a central laboratory for ing can occur. Downstream treatment of women with syphilis serology, found that only 291 of 540 women positive HIV results is measured systematically in coun- (54 percent) had been tested. Of 11 who were seropositive, tries targeted for PMTCT of HIV, specifically to assess only 1 had been treated. However, after these clinics intro- whether mothers were started and maintained on ART, duced same-day screening and treatment, 99.9 percent of continued in follow-up care, and had their infants tested ANC attendees were screened for syphilis, and 87.3 for HIV through EID programs (UNAIDS 2013). The percent of seropositive women and 50 percent of their MCH registers typically include information on syphilis partners received treatment (Temmerman, Mohamedalf, testing. Registers could be enhanced to improve PMTCT and Fransen 1993). The new POC tests make it possible to of syphilis by leveraging current program evaluation of offer same-day screening and treatment in any health HIV PMTCT to include variables on syphilis indicators facility, which increases the coverage of screening and (table 6.3) (WHO 2014b). Enhancements should include treatment in many settings (Hawkes and others 2011; monitoring infant receipt of antiretroviral prophylaxis Jenniskens and others 1995). and infant evaluation for clinical syphilis and treatment needs. Cross-Cutting Issues for PMTCT of HIV and Syphilis Other cross-cutting issue areas affecting PMTCT of MCH is addressed through public health systems and both HIV and syphilis include the following: was prioritized in Millennium Development Goal (MDG) 4 (to reduce child mortality) and MDG 5 Male Partner Involvement. Several studies have noted (to improve maternal health) (Chi, Bolton-Moore, and enhanced PMTCT and infant outcomes and treatment Holmes 2013); MCH is also part of Sustainable adherence among women whose male partners have Development Goal 3 (health and well-being at all ages). participated in MCH programs, either through HIV PMTCT of HIV and PMTCT of syphilis are delivered testing or ANC attendance (Aluisio and others 2011; through the same MCH system. Programs for PMTCT Kalembo and others 2013). A program offering home- of HIV have used the cascade-of-care approach to iden- based HIV testing to male partners noted high uptake of tify bottlenecks in services.10 In many settings, most male HIV testing (Osoti and others 2014). It is difficult pregnant women visit an ANC at least once during their to discern whether these benefits are the result of male Table 6.3 Cross-Cutting Health System Implementation Issues for PMTCT of HIV and Syphilis Implementation need PMTCT of syphilis PMTCT of HIV Community awareness Community has not been mobilized ¸ Required ANC attendance ¸ ¸ ANC attendance early in pregnancy (optimal for ¸ ¸ prevention) MoH recommendation of maternal testing ¸ ¸ Availability of POC diagnostic test ¸ ¸ National use of POC diagnostic test ¸ ¸ QA and QC of lab tests, good supply chain of lab ¸ ¸ tests Use of opt-out approach ¸ ¸ Availability of one-time treatment ¸ Needs lifelong treatment Availability of low-cost treatment ¸ Not a one-time cost—ART currently provided by a combination of resources including from government, PEPFAR, or GAP Usefulness of partner notification and engagement ¸ ¸ Availability of standard infant diagnostic test No laboratory test—clinical diagnosis ¸ Note: ANC = antenatal clinic; ART = antiretroviral treatment; HIV = human immunodeficiency virus; PMTCT = prevention of mother-to-child transmission; GAP = Global AIDS Program; MoH = Ministry of Health; PEPFAR = President’s Emergency Plan for AIDS Relief; POC = point-of-care; QA = quality assurance; QC = quality control. Prevention of Mother-to-Child Transmission of HIV and Syphilis 123 involvement or female self-efficacy, because few women occur at voluntary counseling and testing centers, in involve male partners in ANC attendance. Efforts to comprehensive HIV care and treatment programs, and increase this involvement include written invitations, at PMTCT programs in MCH clinics. Women with weekend openings, initiatives to provide male partners HIV may shift between PMTCT and HIV care and with incentives to attend, male-friendly clinics, and treatment programs, and Options B and B+ approaches home visits for male testing and education (Reece and ensure comparable antiretroviral regimens during this others 2010). Rwanda made a concerted effort to man- process. Women’s HIV care and treatment affect both date male attendance during antenatal care of women to sexual and mother-to-infant transmission and con- improve maternal and infant outcomes (Irakoze and tribute to community HIV prevention. Because preg- others 2012), and male HIV testing increased substan- nancy is an identifiable trigger point for accessing HIV tially in parallel with increased facility delivery. testing, it is often a sentinel event for family HIV diag- Conversely, this approach may stigmatize single moth- nosis in young couples who do not perceive themselves ers. In addition, since syphilis screening leads to male at risk for HIV but who live in settings with high HIV notification and treatment, such screening could be prevalence. integrated into male partner programs for PMTCT Syphilis is similarly preferentially detected during of HIV. pregnancy because of widespread access to care and testing. However, since no large syphilis care and treatment centers Integration of Family Planning and Primary Maternal exist, prevention and detection of syphilis typically occur at HIV and Syphilis Prevention. Postpartum visits are primary health care visits or at sexually transmitted disease opportunities to increase uptake of contraception and to (STD) treatment programs. MCH clinics have limited promote prevention of HIV and other STIs. However, experience with standard partner testing for STIs and often family planning clinics and postpartum family planning refer male partners to STD programs for syphilis testing. clinics do not routinely incorporate HIV or STI preven- An alternative approach would be to include partner syph- tion or testing. ilis testing and follow-up in routine MCH care, analogous to HIV, without referral to STD services. Community Engagement to Reduce Stigma of Testing and Treatment. Both HIV and syphilis carry a commu- nity stigma as STIs. In the case of HIV infection, the Methods for Assessing PMTCT Program combination of high community prevalence, mature Effectiveness epidemics, and increasing community knowledge and Program Evaluation Methodology for PMTCT advocacy have led to vibrant activism and pragmatic of HIV peer mentoring and counseling (Namukwaya and others Standardized national surveys using routine infant DBS 2015). In contrast, syphilis awareness lacks community- for HIV DNA testing at age six weeks can capture most level advocacy. Individuals are not familiar with the dis- mother-infant pairs, because uptake of routine six-week ease or its symptoms and typically do not know whether infant immunizations is high, regardless of HIV diagno- someone has been identified as having had syphilis. To sis or PMTCT intervention uptake (WHO 2012a). Thus, our knowledge, peer mentoring is not used, and children the evaluation of programs for PMTCT of HIV can be and parents with syphilis do not have support groups standardized because there is a hard outcome—infant comparable to HIV support groups. HIV status. However, this outcome measure is not the final infant HIV outcome, because breastmilk transmis- Alignment of the Long-Term Health Benefits of sion may occur until cessation of breastfeeding, particu- Interventions with the MDGs and General MCH larly if women do not adhere to ART. Later time Goals. Comprehensive MCH requires detection and points—nine months and thereafter—for MCH visits treatment of maternal HIV and syphilis to improve are less well attended and do not include infants who short-term and long-term outcomes for mothers and become lost to follow-up or who die before assessment. their children. Interventions may have impacts on mul- Community household-based surveys may complement tiple outcomes, including decreasing preterm birth, facility-based assessment, reach mother-infant pairs stillbirth, and other morbidity in addition to enhance- who do not attend facility services, and enable better ment of growth, cognition, and survival. estimation of HIV-free survival at later endpoints, but assessment of programmatic effectiveness may Intersection of PMTCT Interventions with General have shortcomings depending on participation and Prevention and Treatment of HIV and Syphilis. self-reporting (Conrad and others 2012; Kohler and Community-level HIV treatment and prevention others 2014; Larsson and others 2012). 124 Major Infectious Diseases UNAIDS has developed evaluation strategies for con- births. The guidelines also note the additional seque- sistent assessment of national programs for PMTCT of lae that are anticipated to decrease following implemen- HIV. In these surveys, several countries report higher tation of an effective program for PMTCT of syphilis than 90 percent PMTCT coverage, with a transmission (figure 6.2 and table 6.4). risk of less than 5 percent (UNAIDS 2015). Remarkable progress has been made in Botswana, South Africa, and Combined Validation of PMTCT of HIV and Syphilis several eastern and southern African countries. Indeed, In 2014, the WHO published a framework for validation PMTCT interventions have been responsible for a of PMTCT of HIV and syphilis for program managers substantial proportion of the declines in overall HIV and policy makers (WHO 2014a). This framework was incidence globally. However, although infant HIV infec- used to validate PMTCT of HIV and syphilis in Cuba tions have plummeted (figure 6.1), the world did not in 2015. Measures assessed include MTCT of HIV and accomplish the UNAIDS PMTCT goal of no more than congenital syphilis rate, and coverage (greater than 15,000 infections by 2015 (UNAIDS 2016b). 95 percent) of antenatal syphilis and HIV testing and treatment (WHO 2015). Program Evaluation Methodology for PMTCT of Syphilis COST-EFFECTIVENESS OF PMTCT OF HIV The Rapid Syphilis Test Toolkit is a comprehensive guide AND SYPHILIS for planning and management of syphilis prevention programs that includes guidelines on policy advocacy, Cost-effectiveness analyses allow policy makers to supply chain, cost-effectiveness, clinical training, labora- prioritize approaches for disease prevention. In this tory procedures, and monitoring and evaluation with case, they help policy makers compare PMTCT of HIV clearly defined program indicators (LSHTM 2011). with PMTCT of syphilis and potentially estimate Specifically, it states that the number of women screened, the benefits of a dual approach to PMTCT. The who tested positive, and whose male partners were cost-effectiveness studies presented are all from low- tested, as well as the number of syphilis-positive women and middle-income settings. Tables 6.5 and 6.6 present who received same-day penicillin, should be summa- cost-effectiveness estimates for PMTCT of HIV and rized from clinic register data. In addition, the number PMTCT of syphilis, respectively, extracted from sys- of cases of congenital syphilis and other complications tematic reviews of the literature (Johri and Ako-Arrey of syphilis should be summarized as a percentage of live 2011; Levin and Brouwer 2014) and using only those Figure 6.1 Number of New HIV Infections among Children in 21 Global Plan Priority Countries, 2000–15 600,000 Number of new HIV infections among children 500,000 400,000 24% decline 2000–08 300,000 60% 200,000 decline 2009–15 100,000 0 2000 2003 2006 2009 2012 2015 Source: UNAIDS 2016b. Note: HIV = human immunodeficiency virus. Prevention of Mother-to-Child Transmission of HIV and Syphilis 125 Figure 6.2 Syphilis Monitoring and Evaluation Results Pyramid Resources Staff Budget Inputs Facilities Supplies Training Training workshops and manuals Effective supply of RST and benzathine penicillin Quality of service MONITORING Outputs Quality of diagnosis Process evaluation Proficiency testing coverage Knowledge of syphilis transmission Educational materials Trained HCW Change in numbers screened and treated Change in syphilis prevalence Outcomes Change in numbers of partners tested and treated Acceptability of RST among HCW EVALUATION Integration of RST within existing PMTCT Effectiveness evaluation Proficiency of HCW on QC Increased access to syphilis testing Partner uptake Impact Change in prevalence of congenital syphilis Change in number of sites using rapid syphilis tests Economic impact Source: Adapted from LSHTM 2011, figure 1; originally published in UNAIDS/World Bank, panel 1. Note: HCW = health care workers; PMTCT = prevention of mother-to-child transmission; QC = quality control; RST = rapid syphilis test. Table 6.4 Required Indicators for Global Validation of PMTCT of articles published after 2000. In addition, a supple- HIV and Syphilis mentary search yielded more recent studies published between 2010 and 2015. We extracted 24 cost- HIV effectiveness metrics or ranges from 18 articles for Impact indicators PMTCT of HIV and 23 from 8 articles for PMTCT of Case rate of new pediatric HIV infections resulting from MTCT of HIV of ≤50 cases syphilis and integrated approaches. All estimates have per 100,000 live births; AND been converted into 2012 US dollars. MTCT rate of HIV of <5 percent in breastfeeding populations OR MTCT rate of HIV of <2 percent in nonbreastfeeding populations Cost-Effectiveness of PMTCT of HIV Process indicators The cost-effectiveness of extended combination prophy- ANC coverage (at least one visit) of ≥95 percent laxis for PMTCT of HIV is well established, especially in Coverage of pregnant women who know their HIV status of ≥95 percent high-risk areas, compared with no intervention or the use ARV coverage of HIV-positive pregnant women of ≥90 percent of single-dose nevirapine or short-course prophylaxis Congenital syphilis (Johri and Ako-Arrey 2011). In high-risk regions such as Sub-Saharan Africa, regional models estimate that Impact indicator Option A is cost-effective, ranging from US$25 to US$730 Case rate of congenital syphilis ≤50 cases per 100,000 live births per disability-adjusted life year (DALY) averted. Since Process indicators 2010, country-specific studies have focused on compar- ANC coverage (at least one visit) of ≥95 percent ing the incremental costs and benefits of new guidelines Coverage of syphilis testing of pregnant women of ≥95 percent related to treatment Options A, B, or B+. For the LMICs shown in table 6.5, these options benefit infants and Treatment of syphilis-seropositive pregnant women ≥95 percent mothers, save money over time, and are cost-effective Source: WHO 2014a. Note: ANC = antenatal care; ARV = antiretroviral; PMTCT = elimination of mother-to-child transmission; (Ciaranello and others 2013; Fasawe 2013). For resource- HIV = human immunodeficiency virus; MTCT = mother-to-child transmission. constrained countries with high risk of HIV, there are 126 Major Infectious Diseases Table 6.5 Cost-Effectiveness Analyses since 2000 of Interventions for PMTCT of HIV Country or Cost in Unit of Region Intervention Comparator region 2012 US$ outcome Study Sub-Saharan Africa Regional studies Option A No intervention SSA 25.71–43.68 DALYs averted Marseille and others 2000 Option A No intervention SSA 184.48 Infection averted Stringer and others 2000 Option A No intervention SSA 136.67–730.49 DALYs averted Sweat and others 2004 Option A No intervention AFR-E 38.94 DALYs averted Hogan and others 2005 Country studies Option A No intervention South Africa 47–99 DALYs averted Wilkinson, Floyd, and Gilks 2000 Option A and mass No intervention Chad 1,862 Infection averted Hutton, Wyss, and N’Diekhor 2003 screening Option B Current practice Malawi 40 DALYs averted Orlando and others 2010 Option B+ No intervention Tanzania 251 DALYs averted Robberstad and Evjen-Olsen 2010 Option B Option A Nigeria 171 DALYs averted Shah and others 2011 Option B Option A Uganda 65–140 DALYs averted Kuznik and others 2012 Option B+ No intervention Uganda 291–502 DALYs averted Kuznik and others 2012 Option B Option A South Africa 1,187 QALYs gained Zulliger and others 2014 a Option B+ Option B Zimbabwe 1370 YLS Ciaranello and others 2013 Option A Current practiceb Malawi 30 DALYs averted Fasawe and others 2013 b Option B Current practice Malawi 54 DALYs averted Fasawe and others 2013 Option B+ Current practiceb Malawi 50 DALYs averted Fasawe and others 2013 Option B Option A Zambia 82 QALYs gained Ishikawa and others 2014 Option B+ Option A Zambia 145 QALYs gained Ishikawa and others 2014 Latin America and the Caribbean Option A and mass No intervention Peru 3,092–7,924 DALYs averted Aldridge and others 2009 screening South Asia Universal screening No intervention India 32 YLS Kumar and others 2006 Targeted screening No intervention India 68 YLS Kumar and others 2006 East Asia and Pacific Option A No intervention SEAR-D 355 DALYs averted Hogan and others 2005 Option B Option A Thailand 2,299–3,225 DALYs averted Teerawattananon and others 2005 Option B Option A Thailand Cost saving QALYs gained Werayingyong and others 2013 Note: AFR-E = WHO member states Botswana, Burundi, the Central African Republic, the Republic of Congo, Côte d’Ivoire, the Democratic Republic of Congo, Eritrea, Ethiopia, Kenya, Lesotho, Malawi, Mozambique, Namibia, Rwanda, South Africa, Swaziland, Uganda, Tanzania, Zambia, and Zimbabwe; CEA = cost-effectiveness analysis; DALYs = disability-adjusted life years; PMTCT = prevention of mother-to-child transmission; HIV = human immunodeficiency virus; QALYs = quality-adjusted life years; SEAR-D = WHO member states Bangladesh, Bhutan, the Democratic People’s Republic of Korea, India, Maldives, Myanmar, and Nepal; SSA = Sub-Saharan Africa; YLS = years of life saved. a. This estimate is in 2013 US dollars, as in the publication, because a reliable consumer price index deflator was not available for Zimbabwe. b. Mix of interventions including single-dose nevirapine, or dual-drug regimen containing zidovudine, or triple-drug antiretroviral prophylaxis until cessation of breastfeeding. debates about the advantages and disadvantages of per DALY averted; it may even be cost saving, depending Option B+ (lifelong ART for all pregnant and HIV- on adherence to ART and relative costs of ARVs. HIV infected women), in part because Option B+ is effective treatment guidelines and PMTCT intervention options in saving lives but is associated with higher short-term are continuously evolving; country context (prevalence costs compared with Options A and B (Saeed, Kim, and rates, drug costs, health utilization rates, health system Abrams 2013). In low-prevalence settings, such as capacity) will affect the relative cost-effectiveness of and Thailand, PMTCT is cost-effective at US$2,299–US$3,225 the choice of best strategy from these and future options. Prevention of Mother-to-Child Transmission of HIV and Syphilis 127 Table 6.6 Cost-Effectiveness Analyses since 2000 of Interventions for PMTCT of Syphilis Country Cost in Unit of Region Intervention Comparator or region 2012 US$ outcome Study Sub-Saharan Global or regional Africa Rapid point-of-care ICSa No screening Sub-Saharan 11 DALYs averted Kuznik and others 2013 screen and treat with Africa penicillin Laboratory RPRa screen and No screening Global low 24–111 DALYs averted Kahn and others 2014 treat with penicillin prevalence Laboratory RPRa screen and No screening Global high Cost- DALYs averted Kahn and others 2014 treat with penicillin prevalence saving Country studies Laboratory RPRa screen and No screening Tanzania 23 DALYs averted Terris-Prestholt and others 2015 treat with penicillin Tanzania 16 DALYs averted Terris-Prestholt and others 2003 Zambia 26 DALYs averted Terris-Prestholt and others 2015 Rapid point-of-care ICSa No screening Tanzania 17 DALYs averted Terris-Prestholt and others 2015 screen and treat with Zambia 16 DALYs averted Terris-Prestholt and others 2015 penicillin Dual screening with ICSa and No screening Tanzania 18 DALYs averted Terris-Prestholt and others 2015 RPR and treat with penicillin Zambia 70 DALYs averted Terris-Prestholt and others 2015 ICS screen and treat with RPR screen and treat Tanzania 16 DALYs averted Terris-Prestholt and others 2015 penicillin Zambia 11 DALYs averted Terris-Prestholt and others 2015 ICS screen and treat with Current practice: Zambia 453 DALYs averted Larson and others 2014 penicillin 62 percent tested, 2.8 percent positive, 10.4 percent of positives treated Improved treatment: Zambia 48 DALYs averted Larson and others 2014 100 percent of positives treated Improved screening and Zambia 45 DALYs averted Larson and others 2014 treatment: 100 percent tested and treated Latin America and the Caribbean Laboratory RPRa screen and No screening Peru 139 DALYs averted Terris-Prestholt and others 2015 treat with penicillin Rapid point-of-care ICSa No screening Peru 54 DALYs averted Terris-Prestholt and others 2015 screen and treat with penicillin Dual screening with ICS and No screening Peru 76 DALYs averted Terris-Prestholt and others 2015 RPR and treat with penicillin ICS screen and treat with RPR screen and treat Peru Cost- DALYs averted Terris-Prestholt and others 2015 penicillin saving Rapid point-of-care ICS National HIV screening Haiti 21 DALYs averted Schackman and others 2007 screen added to rapid HIV program with no syphilis screen and treat screening table continues next page 128 Major Infectious Diseases Table 6.6 Cost-Effectiveness Analyses since 2000 of Interventions for PMTCT of Syphilis (continued) Country Cost in Unit of Region Intervention Comparator or region 2012 US$ outcome Study East Asia and Pacific Laboratory TRUSTa and TTPAb No screening China 369 DALYs averted Hong and others 2010 test and treat with penicillin Laboratory RPR screen and No screening China 200 DALYs averted Owusu-Edusei and others 2014 treat with penicillin Rapid ICS screen added to China 160 DALYs averted Owusu-Edusei and others 2014 rapid HIV screen and treat Note: DALYs = disability-adjusted life years; PMTCT = prevention of mother-to-child transmission; HIV = human immunodeficiency virus; ICS = point-of-care immunochromatographic strip; RPR = lab-based rapid plasma reagin test; TPPA = T. pallidum particle agglutination lab-based treponemal test; TRUST = lab-based Toluidine Red Unheated Serum Test. a. Nontreponemal tests that cannot distinguish between recent and previous infection. b. Treponemal test. At this point, Option B+ is the recommended strategy Larson and others (2014) showed that the cost per and is aligned with the WHO 2015 recommendations to DALY averted falls from US$628 to US$60 with full treat all individuals with lifelong ART. adherence to screening and treatment guidelines. Cost-Effectiveness of PMTCT of Syphilis Cost-Effectiveness of Integrated PMTCT of Syphilis and HIV Several models have demonstrated the cost-effectiveness of PMTCT in Sub-Saharan Africa, demonstrating that Although information on the integration of PMTCT of antenatal syphilis screening is cost-effective, even at syphilis and HIV is limited, two studies provide informa- low prevalence rates; it may even be cost saving in tion on the potential benefit of adding syphilis diagnosis high-prevalence settings (Blandford and others 2007; and treatment to existing PMTCT of HIV services. In Kahn and others 2014; Kuznik and others 2013; Haiti, a high-prevalence country for both HIV and Rydzak and Goldie 2008). In high-risk areas, PMTCT syphilis, adding an RPR test-and-treat model in rural of syphilis is among one of the most cost-effective areas cost US$6.83 per DALY averted, while the estimate interventions available. For example, a study of the for an urban program was US$9.95 per DALY averted cost-effectiveness of prenatal syphilis screening in (Schackman and others 2007). In a recent study in Tanzania found that the cost was US$1.44 per woman China—a low-prevalence country for both HIV and screened, US$20 per woman treated, US$16 per DALY syphilis—syphilis screening or integrated HIV and syphilis averted, or US$10.56 per DALY averted if stillbirths screening were both more cost-effective than HIV screen- averted were included in the calculation (Terris- ing alone (Owusu-Edusei and others 2014). This is a con- Prestholt and others 2003). Operational costs of testing servative finding, given the low prevalence of both HIV may influence cost-effectiveness (Levin 2007). Cost- and syphilis. Based on the results from the Haiti and China effectiveness analysis using country-level evaluation studies, it is likely that in settings with higher prevalence data has been critical for informing the choice of rates for both HIV and syphilis, integrating HIV and syph- screening strategy and for highlighting the importance ilis services would be the most cost-effective option. of achieving higher screening rates and full adherence to guidelines (Larson and others 2014; Terris-Prestholt LESSONS FROM CASE STUDIES and others 2015). Of 10 screening and treatment approaches, Terris-Prestholt and others (2015) found PMTCT of HIV: Success in South Africa that treponemal-based rapid POC syphilis test-and- As an upper-middle-income country, South Africa has treat strategies were the most cost-effective, at US$16– been able to make enormous progress in implement- US$17 per DALY averted in Zambia and Tanzania and ing PMTCT. MTCT rates decreased from more than US$54 per DALY averted in Peru. In Peru, the incre- 30 percent to approximately 4 percent in a national pro- mental cost-effectiveness of switching from a lab- grammatic evaluation of all six-week-old infants tested based test to a rapid POC test was cost saving. at 572 clinics throughout the country (Goga, Dinh, and Prevention of Mother-to-Child Transmission of HIV and Syphilis 129 Jackson 2010). Maternal HIV seroprevalence had been In Peru, 604 trained health providers implemented 32 percent at the time of the study.11 An astonishing POC testing for syphilis using the two-for-one strat- 98 percent of pregnant women received testing, and egy: offering both syphilis and HIV testing with one almost 92 percent of HIV-positive women received ART. finger prick. This approach resulted in testing and Gaps remained in CD4 testing and EID. treatment on the first visit; before POC testing was The study also illustrates that high maternal HIV introduced, a woman would typically not receive her seroprevalence contributes to the effectiveness of syphilis result until her fifth clinic visit, 27 days later PMTCT programs. When large numbers of mothers in (Garcia and others 2013). In Uganda and Zambia, the a given health center or community are HIV infected, integration of HIV and syphilis screening in ANCs, large networks of peer counselors can lead to more effec- using two POC tests, was well accepted and led to tive peer counseling, dissemination of education and increased coverage of syphilis screening (Strasser and programs, and acceptability of HIV testing and ART. others 2012). PMTCT of Syphilis: Successful Rapid POC Models PMTCT of HIV and Syphilis: An Integrated Model A study conducted in six countries showed that the Peeling and others (2004) described cases of infants born introduction of POC testing for syphilis in ANCs was to HIV-infected mothers in Haiti who successfully well accepted in a wide variety of settings and delivery escaped HIV transmission following ART and formula platforms, from the Amazon rain forest in Brazil to the feeding but who acquired congenital syphilis. These cities of China, and from primary health posts to teach- cases highlight the lost opportunity to prevent syphilis ing hospitals. After the introduction of POC testing, among women attending ANCs; despite receiving inter- more than 90 percent of those who attended ANCs were ventions for PMTCT of HIV (testing and treatment), the screened for syphilis in all six countries, and more than women failed to access simple measures to detect and 90 percent of women testing positive received treatment prevent congenital syphilis. on the same day (Mabey and others 2012). In the authors’ assessment of a Sub-Saharan African population of pregnant women (table 6.7), the costs of testing and treatment per DALY averted were estimated Table 6.7 Estimated Costs and Benefits of Screening for HIV and to be lower for syphilis (US$4–US$18.70) than for Syphilis in Theoretical Sub-Saharan African Population of 20,000 HIV (US$19.20). They noted that more efficient POC Pregnant Women Yearly diagnostic testing for syphilis and development of oral regimens could further enhance congenital syphilis Syphilis HIV* prevention efforts. Seroprevalence 8% 15% Test positive 1,600 3,000 CONCLUSIONS Result During the next few years, the concerted efforts to Stillbirth averteda 200 cases — prevent infant HIV infection will continue and per- LBW averteda 184 cases — haps accelerate (IATT 2014). This intensive focus on HIV averted* — 246 cases HIV provides an opportune time to leverage program Cost of counselling and $1.44 $18.50 improvements to similarly target elimination of con- testing per woman genital syphilis and other adverse outcomes of mater- Cost of treatment $0.5–1.00 $4.00 for single-dose nal syphilis. Programs that monitor HIV outcomes nevirapine can also track syphilis testing, treatment, and trans- mission and motivate prompt, efficient management Cost per case averted Stillbirth, $318; all adverse $506 pregnancy outcomes, $44–187b of newly diagnosed cases in pregnancy, as shown in China, Haiti, Peru, Uganda, and Zambia. In short, Cost per DALY saved $4.0–18.7b $19.2 improving PMTCT of syphilis within PMTCT of HIV Source: Peeling and others 2004. is a feasible, low-cost strategy that would yield enor- Note: DALY = disability-adjusted life year; HIV = human immunodeficiency virus; LBW = low birth weight; — = not available. mous benefits to mothers and infants. The incremen- *Data from Sweat and others 2004; Marseille and others 2000. tal cost is small and the gains—prevention of stillbirth, a. Assuming 4 percent of women have rapid plasma reagin titer of 1:8 or more, of whom 25 percent preterm birth, and long-term disability of children will have stillbirth, and 33 percent will deliver live baby weighing below 2.5 kilograms, compared with 10 percent of seronegative women (Brocklehurst and French 1998; Grimprel and others 1991). delivered by women with maternal syphilis—are b. Range taken from studies summarized in UNAIDS 2013. substantial. 130 Major Infectious Diseases NOTES REFERENCES World Bank Income Classifications as of July 2014 are as Afran, L., M. Garcia Knight, E. Nduati, B. C. Urban, follows, based on estimates of gross national income (GNI) R. S. Heyderman, and others. 2014. “HIV-Exposed per capita for 2013: Uninfected Children: A Growing Population with a Vulnerable Immune System?” Clinical and Experimental • Low-income countries (LICs) = US$1,045 or less Immunology 176 (1): 11–22. • Middle-income countries (MICs) are subdivided: Aldridge R. W., D. Iglesias, C. F. Caceres, and J. J. Miranda. (a) lower-middle-income = US$1,046–US$4,125 2009. “Determining a Cost Effective Intervention Response (b) upper-middle-income (UMICs) = US$4,126–US$12,745 to HIV/AIDS in Peru.” BMC Public Health 9: 352. • High-income countries (HICs) = US$12,746 or more. doi:10.1186/1471-2458-9-352. Aluisio, A., B. A. Richardson, R. Bosire, G. John-Stewart, 1. Increased risk of spontaneous abortion: odds ratio (OR) D. Mbori-Ngacha, and others. 2011. “Male Antenatal 4.05, 95 percent confidence interval (CI) 2.75–5.96; per- Attendance and HIV Testing Are Associated with Decreased inatal mortality: OR 1.79, 95 percent CI 1.1–2.81; infant Infant HIV Infection and Increased HIV-Free Survival.” mortality: OR 3.69, 95 percent CI 3.03–4.49; intrauterine Journal of Acquired Immune Deficiency Syndromes 56 (1): growth retardation: OR 1.7, 95 percent CI 1.43–2.02; low 76–82. birth weight: OR 2.09, 95 percent CI 1.86–2.35; preterm Arnold, S. R., and E. L. Ford-Jones. 2000. “Congenital Syphilis: birth: OR 1.83, 95 percent CI 1.63–2.06. A Guide to Diagnosis and Management.” Paediatrics and 2. In Option A, women received zidovudine as early as Child Health 5 (8): 463–69. 14 weeks into the pregnancy, a single dose of nevirapine Becquet, R., M. Marston, F. Dabis, L. H. Moulton, G. Gray, during labor, and two weeks of tenofovir and emtricita- and others. 2012. “Children Who Acquire HIV Infection bine after delivery. Perinatally Are at Higher Risk of Early Death than Those 3. In Option B, women received one of two triple anti-HIV drug Acquiring Infection through Breastmilk: A Meta-Analysis.” regimens as early as 14 weeks into the pregnancy—either (1) PLoS One 7 (2): e28510. lamivudine, zidovudine, and ritonavir-boosted lopinavir Blandford, J. M., T. L. Gift, S. Vasaikar, D. Mwesigwa-Kayongo, (called the “lamivudine combination”); or (2) tenofovir, P. Dlali, and others. 2007. “Cost-Effectiveness of On-Site emtricitabine, and ritonavir-boosted lopinavir (called the Antenatal Screening to Prevent Congenital Syphilis in Rural “tenofovir combination”). Option B+ is lifelong ART. Eastern Cape Province, Republic of South Africa.” Sexually 4. This study has not yet been published, but the press release Transmitted Diseases 34 (Suppl 7): S61–66. noted increased adverse outcomes (NIH 2014). Blencowe, H., S. Cousens, M. Kamb, S. Berman, and 5. The ELISA test detects antibodies to HIV, usually from J. E. Lawn. 2011. “Lives Saved Tool Supplement Detection either a blood sample (drawn from a vein) or saliva. A pos- and Treatment of Syphilis in Pregnancy to Reduce Syphilis itive result is often confirmed with a follow-up test such as Related Stillbirths and Neonatal Mortality.” BMC Public the Western blot, which is less likely than the ELISA test to Health 11 (Suppl 3): S9. yield a false positive result. Results from most ELISA tests Brocklehurst, P., and R. French. 1998. “The Association and confirmatory Western blot tests are usually available between Maternal HIV Infection and Perinatal Outcome: within 2 to 14 days. A Systematic Review of the Literature and Meta-Analysis.” 6. A rapid HIV test uses technology similar to the ELISA test British Journal of Obstetrics and Gynaecology 105 (8): (except blood is drawn from a finger prick instead of from a 836–48. vein) but produces results in approximately 20 minutes. As CDC (Centers for Disease Control and Prevention). 2013. with the ELISA test, however, a “preliminary positive” result “Impact of an Innovative Approach to Prevent Mother-to- must be confirmed with a second test, either a Western blot Child Transmission of HIV—Malawi, July 2011–September test or a second rapid test from another manufacturer. 2012.” Morbidity and Mortality Weekly Report 62 (8): 7. The studied triple-ART in PROMISE was either lami- 148–51. http://www.cdc.gov/mmwr/preview/mmwrhtml vudine, zidovudine, and ritonavir-boosted lopinavir, or /mm6208a3.htm. tenofovir, emtricitabine, and ritonavir-boosted lopinavir. Chi, B. H., C. Bolton-Moore, and C. B. Holmes. 2013. All infants received six weeks of nevirapine prophylaxis. “Prevention of Mother-to-Child HIV Transmission 8. Option B+ refers to lifelong ART initiated in pregnancy within the Continuum of Maternal, Newborn, and Child 9. WHO, Global Health Observatory Data Repository, Health Services.” Current Opinion in HIV and AIDS 8 (5): Antenatal care (ANC) attendees tested for syphilis at first 498–503. ANC visit, Data by country (http://apps.who.int/gho/data Ciaranello, A. L., F. Perez, B. Engelsmann, R. P. Walensky, /node.main.A1358STI). A. Mushavi, and others. 2013. “Cost-Effectiveness of World 10. The cascade of care model refers to following interventions Health Organization 2010 Guidelines for Prevention of from diagnosis to receipt of interventions to continued Mother-to-Child HIV Transmission in Zimbabwe.” Clinical retention in care. Infectious Diseases 56 (3): 430–46. 11. Maternal HIV seroprevalence was 32.0 percent (95 percent Cohen, M. S., Y. Q. Chen, M. McCauley, T. Gamble, CI 30.7–33.3). M. C. Hosseinipour, and others. 2011. “Prevention of HIV-1 Prevention of Mother-to-Child Transmission of HIV and Syphilis 131 Infection with Early Antiretroviral Therapy.” New England Chain Reaction and Rabbit Infectivity Testing to Detect Journal of Medicine 365 (6): 493–505. Treponema pallidum in Amniotic Fluid, Fetal and Conrad, P., M. De Allegri, A. Moses, E. C. Larsson, F. Neuhann, Neonatal Sera, and Cerebrospinal Fluid.” Journal of Clinical and others. 2012. “Antenatal Care Services in Rural Uganda: Microbiology 29 (8): 1711–18. Missed Opportunities for Good-Quality Care.” Qualitative Hawkes, S., N. Matin, N. Broutet, and N. Low. 2011. Health Research 22 (5): 619–29. “Effectiveness of Interventions to Improve Screening for Creek, T. L., R. Ntumy, K. Seipone, M. Smith, M. Mogodi, and Syphilis in Pregnancy: A Systematic Review and Meta- others. 2007. “Successful Introduction of Routine Opt-Out Analysis.” The Lancet Infectious Diseases 11 (9): 684–91. HIV Testing in Antenatal Care in Botswana.” Journal of Hogan, D. R., R. Baltussen, C. Hayashi, J. A. Lauer, and Acquired Immune Deficiency Syndromes 45 (1): 102–7. J. A. Salomon. 2005. “Cost Effectiveness Analysis of Day, S., D. Lakhani, M. Hankins, and C. A. Rodgers. 2004. Strategies to Combat HIV/AIDS in Developing Countries.” “Improving Uptake of HIV Testing in Patients with a British Medical Journal 331 (7530): 1431–37. Confirmed STI.” International Journal of STD and AIDS Hong, F. C., J. B. Liu, T. J. Feng, X. L. Liu, P. Pan, and 15 (9): 626–28. others. 2010. “Congenital Syphilis: An Economic Evaluation De Cock, K. M., M. G. Fowler, E. Mercier, I. de Vincenzi, of a Prevention Program in China.” Sexually Transmitted J. Saba, and others. 2000. “Prevention of Mother-to-Child Diseases 37 (1): 26–31. HIV Transmission in Resource-Poor Countries: Translating Hutton, G., K. Wyss, and Y. N’Diekhor. 2003. “Prioritization of Research into Policy and Practice.” Journal of the American Prevention Activities to Combat the Spread of HIV/AIDS Medical Association 283 (9): 1175–82. in Resource Constrained Settings: A Cost-Effectiveness Drake, A. L., A. Wagner, B. Richardson, and G. John-Stewart. Analysis from Chad, Central Africa.” International Journal 2014. “Incident HIV during Pregnancy and Postpartum and of Health Planning and Management 18 (2): 117–36. Risk of Mother-to-Child HIV Transmission: A Systematic IATT (Inter-Agency Task Team). 2014. HIV and Sexual and Review and Meta-Analysis.” PLoS Medicine 11 (2): e1001608. Reproductive Health Programming: Innovative Approaches to Farquhar, C., D. A. Mbori-Ngacha, R. K. Bosire, R. W. Nduati, Integrated Service Delivery. Compendium of Case Studies. J. K. Kreiss, and others. 2001. “Partner Notification by New York, NY: IATT. https://sustainabledevelopment.un HIV-1 Seropositive Pregnant Women: Association with .org /content/documents/1779HIV_SRH_Programming Infant Feeding Decisions.” AIDS 15 (6): 815–17. _Integrated_Service_Delivery_Case_Studies_1.pdf. Fasawe, O., C. Avila, N. Shaffer, E. Schouten, E. Chimbwandira, Irakoze, A. A., S. Nsanzimana, M. Jennifer, E. Remera, and others. 2013. “Cost-Effectiveness Analysis of Option C. Karangwa, and others. 2012. “Family Centered Approach B+ for HIV Prevention and Treatment of Mothers and in PMTCT Program, Rwanda 2005–2011.” Presentation to Children in Malawi.” PLoS One 8 (3): e57778. doi:10.1371/ the Coalition for Children Affected by AIDS Symposium, journal.pone.0057778. “Children and HIV: Closing the Gap,” Washington, DC, Futterman, D., J. Shea, M. Besser, S. Stafford, K. Desmond, July 20. and others. 2010. “Mamekhaya: A Pilot Study Combining Ishikawa, N., T. Shimbo, S. Miyano, I. Sikazwe, A. Mwango, a Cognitive-Behavioral Intervention and Mentor Mothers and others. 2014. “Health Outcomes and Cost Impact of with PMTCT Services in South Africa.” AIDS Care: the New WHO 2013 Guidelines on Prevention of Mother- Psychological and Socio-Medical Aspects of AIDS/HIV 22 (9): to-Child Transmission of HIV in Zambia.” PLoS One 9 (3): 1093–100. e90991. Garcia, P. J., C. P. Carcamo, M. Chiappe, M. Valderrama, S. Jafari, Y., R. W. Peeling, S. Shivkumar, C. Claessens, L. Joseph, La Rosa, and others. 2013. “Rapid Syphilis Tests as Catalysts and others. 2013. “Are Treponema pallidum Specific Rapid for Health Systems Strengthening: A Case Study from Peru.” and Point-of-Care Tests for Syphilis Accurate Enough for PLoS One 8 (6): e66905. Screening in Resource Limited Settings? Evidence from a Gloyd, S., S. Chai, and M. A. Mercer. 2001. “Antenatal Syphilis Meta-Analysis.” PLoS One 8 (2): e54695. in Sub-Saharan Africa: Missed Opportunities for Mortality Jani, I. V., B. Meggi, N. Mabunda, A. Vubil, N. E. Sitoe, and Reduction.” Health Policy and Planning 16 (1): 29–34. others. 2014. “Accurate Early Infant HIV Diagnosis in Goga, A., T.-H. Dinh, and D. Jackson. 2010. Evaluation of Primary Health Clinics Using a Point-of-Care Nucleic Acid the Effectiveness of the National Prevention of Mother-to- Test.” Journal of Acquired Immune Deficiency Syndromes Child Transmission (PMTCT) Programme on Infant HIV 67 (1): e1–4. Measured at Six Weeks Postpartum in South Africa. Pretoria: Jenniskens, F., E. Obwaka, S. Kirisuah, S. Moses, Department of Health, Republic of South Africa. F. M. Yusufali, and others. 1995. “Syphilis Control in Gomez, G. B., M. L. Kamb, L. M. Newman, J. Mark, N. Pregnancy: Decentralization of Screening Facilities to Broutet, and others. 2013. “Untreated Maternal Syphilis Primary Care Level, a Demonstration Project in Nairobi, and Adverse Outcomes of Pregnancy: A Systematic Kenya.” International Journal of Gynecology and Obstetrics Review and Meta-Analysis.” Bulletin of the World Health 48 (Suppl): S121–28. Organization 91 (3): 217–26. John, G. C., and J. Kreiss. 1996. “Mother-to-Child Transmission Grimprel, E., P. J. Sanchez, G. D. Wendel, J. M. Burstain, of Human Immunodeficiency Virus Type 1.” Epidemiologic G. H. McCracken Jr., and others. 1991. “Use of Polymerase Reviews 18 (2): 149–57. 132 Major Infectious Diseases John, G. C., R. W. Nduati, D. A. Mbori-Ngacha, B. A. Richardson, Lee, M. J., R. J. Hallmark, L. M. Frenkel, and G. Del Priore. D. Panteleeff, and others. 2001. “Correlates of Mother-to- 1998. “Maternal Syphilis and Vertical Perinatal Transmission Child Human Immunodeficiency Virus Type 1 (HIV-1) of Human Immunodeficiency Virus Type-1 Infection.” Transmission: Association with Maternal Plasma HIV-1 International Journal of Gynecology and Obstetrics 63 (3): 247–52. RNA Load, Genital HIV-1 DNA Shedding, and Breast Levin, C. E., and E. Brouwer. 2014. “Saving Brains: Literature Infections.” Journal of Infectious Diseases 183 (2): 206–12. Review of Reproductive, Neonatal, Child and Maternal Johri, M., and D. Ako-Arrey. 2011. “The Cost-Effectiveness of Health and Nutrition Interventions to Mitigate Basic Risk Preventing Mother-to-Child Transmission of HIV in Low- Factors to Promote Child Development.” Working Paper and Middle-Income Countries: Systematic Review.” Cost Series, GCC 14-08. Effectiveness and Resource Allocation 9: 3. Levin, C. E., M. Steele, D. Atherly, S. G. Garcia, F. Tinajeros, and Kahn, J. G., A. Jiwani, G. B. Gomez, S. J. Hawkes, H. W. Chesson, others. 2007. “Analysis of the Operational Costs of Using and others. 2014. “The Cost and Cost-Effectiveness Rapid Syphilis Tests for the Detection of Maternal Syphilis of Scaling up Screening and Treatment of Syphilis in in Bolivia and Mozambique.” Sexually Transmitted Diseases Pregnancy: A Model.” PLoS One 9 (1): e87510. 34 (Suppl 7): S47–54. Kalembo, F. W., M. Zgambo, A. N. Mulaga, D. Yukai, and LSHTM (London School of Hygiene and Tropical N. I. Ahmed. 2013. “Association between Male Partner Medicine). 2011. The Rapid Syphilis Test Toolkit: A Guide Involvement and the Uptake of Prevention of Mother-to- to Planning, Management and Implementation. London: Child Transmission of HIV (PMTCT) Interventions in LSHTM. https://www.lshtm.ac.uk/itd/crd/research Mwanza District, Malawi: A Retrospective Cohort Study.” /rapidsyphilistoolkit/rapid_syphilis_test_toolkit.pdf. PLoS One 8 (6): e66517. Luzuriaga, K., and L. M. Mofenson. 2016. “Challenges in the Kiarie, J. N., C. K. Mishra, M. Temmerman, and L. Newman. Elimination of Pediatric HIV-1 Infection.” New England 2015. “Accelerating the Dual Elimination of Mother-To- Journal of Medicine 374: 761–71. Child Transmission of Syphilis and HIV: Why Now?” Mabey, D., R. W. Peeling, R. Ballard, A. S. Benzaken, E. Galban, International Journal of Gynecology and Obstetrics and others. 2006. “Prospective, Multi-Centre Clinic-Based 130 (Suppl 1): S1–3. Evaluation of Four Rapid Diagnostic Tests for Syphilis.” Kinuthia, J., J. N. Kiarie, C. Farquhar, B. A. Richardson, Sexually Transmitted Infections 82 (Suppl 5): v13–16. R. Nduati, and others. 2011. “Uptake of Prevention of Mabey, D. C., K. A. Sollis, H. A. Kelly, A. S. Benzaken, Mother to Child Transmission Interventions in Kenya: E. Bitarakwate, and others. 2012. “Point-of-Care Tests to Health Systems Are More Influential than Stigma.” Journal Strengthen Health Systems and Save Newborn Lives: The of the International AIDS Society 14: 61. Case of Syphilis.” PLoS Medicine 9 (6): e1001233. Kohler, P. K., J. Okanda, J. Kinuthia, L. A. Mills, G. Olilo, and Malonza, I. M., B. A. Richardson, J. K. Kreiss, J. J. Bwayo, and others. 2014. “Community-Based Evaluation of PMTCT G. C. Stewart. 2003. “The Effect of Rapid HIV-1 Testing on Uptake in Nyanza Province, Kenya.” PLoS One 9 (10): Uptake of Perinatal HIV-1 Interventions: A Randomized e110110. Clinical Trial.” AIDS 17 (1): 113–18. Kumar, M., S. Birch, A. Maturana, and A. Gafni. 2006. Marseille, E., J. G. Kahn, F. Mmiro, L. Guay, P. Musoke, and “Economic Evaluation of HIV Screening in Pregnant others. 2000. “The Cost Effectiveness of a Single-Dose Women Attending Antenatal Clinics in India.” Health Policy Nevirapine Regimen to Mother and Infant to Reduce 77 (2): 233–43. Vertical HIV-1 Transmission in Sub-Saharan Africa.” Annals Kuznik, A., M. Lamorde, S. Hermans, B. Castelnuovo, B. Auerbach, of the New York Academy of Sciences 918: 53–56. and others. 2012. “Evaluating the Cost-Effectiveness of Mofenson, L. M., and D. H. Watts. 2014. “Safety of Pediatric Combination Antiretroviral Therapy for the Prevention of HIV Elimination: The Growing Population of HIV- and Mother-to-Child Transmission of HIV in Uganda.” Bulletin Antiretroviral-Exposed but Uninfected Infants.” PLoS of the World Health Organization 90 (8): 595–603. Medicine 11 (4): e1001636. Kuznik, A., M. Lamorde, A. Nyabigambo, and Y. C. Manabe. Mwapasa, V., S. J. Rogerson, J. J. Kwlek, P. E. Wilson, D. Milner, 2013. “Antenatal Syphilis Screening Using Point-of- and others. 2006. “Maternal Syphilis Infection Is Associated Care Testing in Sub-Saharan African Countries: A Cost- with Increased Risk of Mother-to-Child Transmission of Effectiveness Analysis.” PLoS Medicine 10 (11): e1001545. HIV in Malawi.” AIDS 20 (14): 1869–77. Larson, B. A., D. Lembela-Bwalya, R. Bonawitz, E. E. Hammond, Namukwaya, Z., L. Barlow-Mosha, P. Mudiope, A. Kekitinwa, D. M. Thea, and others. 2014. “Finding a Needle in the E. Matoyu, and others. 2015. “Use of Peers, Community Haystack: The Costs and Cost-Effectiveness of Syphilis Lay Persons and Village Health Team (VHT) Members Diagnosis and Treatment during Pregnancy to Prevent Improves Six-Week Postnatal Clinic (PNC) Follow-up Congenital Syphilis in Kalomo District of Zambia.” PLoS and Early Infant HIV Diagnosis (EID) in Urban and Rural One 9 (12): e113868. doi:10.1371/journal.pone.0113868. Health Units in Uganda: A One-Year Implementation Larsson, E. C., A. E. Thorson, G. Pariyo, P. Waiswa, D. Kadobera, Study.” BMC Health Services Research 15 (1). doi:10.1186/ and others. 2012. “Missed Opportunities: Barriers to HIV s12913-015-1213-5. Testing during Pregnancy from a Population Based Cohort Newell, M. L., H. Coovadia, M. Cortina-Borja, N. Rollins, Study in Rural Uganda.” PLoS One 7 (8): e37590. P. Gaillard, and others. 2004. “Mortality of Infected and Prevention of Mother-to-Child Transmission of HIV and Syphilis 133 Uninfected Infants Born to HIV-Infected Mothers in Africa: the Option B+ Approach.” Current Opinion in HIV and A Pooled Analysis.” The Lancet 364 (9441): 1236–43. AIDS 8 (5): 474–89. Newman, L., M. Kamb, S. Hawkes, G. Gomez, L. Say, and Sanchez, P. J., G. D. Wendel Jr., E. Grimprel, M. Goldberg, others. 2013. “Global Estimates of Syphilis in Pregnancy and M. Hall, and others. 1993. “Evaluation of Molecular Associated Adverse Outcomes: Analysis of Multinational Methodologies and Rabbit Infectivity Testing for the Antenatal Surveillance Data.” PLoS Medicine 10 (2): Diagnosis of Congenital Syphilis and Neonatal Central e1001396. Nervous System Invasion by Treponema pallidum.” Journal NIH (National Institutes of Health). 2014. “NIH-Sponsored of Infectious Diseases 167 (1): 148–57. Study Identifies Superior Drug Regimen for Preventing Schackman, B. R., C. P. Neukermans, S. N. Fontain, C. Nolte, Mother-to-Child HIV Transmission.” Press release, P. Joseph, and others. 2007. “Cost-Effectiveness of Rapid November 17. http://www.nih.gov/news/health/nov2014 Syphilis Screening in Prenatal HIV Testing Programs in /niaid-17.htm. Haiti.” PLoS Medicine 4 (5): e183. Obimbo, E. M., D. A. Mbori-Ngacha, J. O. Ochieng, Schmid, G. P., B. P. Stoner, S. Hawkes, and N. Broutet. 2007. B. A. Richardson, P. A. Otieno, and others. 2004. “Predictors “The Need and Plan for Global Elimination of Congenital of Early Mortality in a Cohort of Human Immunodeficiency Syphilis.” Sexually Transmitted Diseases 34 (Suppl 7): S5–10. Virus Type 1-Infected African Children.” Pediatric Infectious Schulte, J. M., S. Burkham, D. Hamaker, M. E. St. Louis, Disease Journal 23 (6): 536–43. J. M. Paffel, and others. 2001. “Syphilis among HIV-Infected Obimbo, E. M., D. Wamalwa, B. Richardson, D. Mbori-Ngacha, Mothers and Their Infants in Texas from 1988 to 1994.” J. Overbaugh, and others. 2009. “Pediatric HIV-1 in Kenya: Sexually Transmitted Diseases 28 (6): 315–20. Pattern and Correlates of Viral Load and Association Shah, M., B. Johns, A. Abimiku, and D. G. Walker. 2011. with Mortality.” Journal of Acquired Immune Deficiency “Cost-Effectiveness of New WHO Recommendations for Syndromes 51 (2): 209–15. Prevention of Mother-to-Child Transmission of HIV in a Orlando, S., M. C. Marazzi, S. Mancinelli, G. Liotta, S. Ceffa, Resource-Limited Setting.” AIDS 25 (8): 1093–102. and others. 2010. “Cost-Effectiveness of Using HAART Shetty, A. K., C. Marangwanda, L. Stranix-Chibanda, in Prevention of Mother-to-Child Transmission in the W. Chandisarewa, E. Chirapa, and others. 2008. “The DREAM-Project Malawi.” Journal of Acquired Immune Feasibility of Preventing Mother-to-Child Transmission of Deficiency Syndromes 55 (5): 631–14. HIV Using Peer Counselors in Zimbabwe.” AIDS Research Osoti, A. O., G. John-Stewart, J. Kiarie, B. Richardson, J. Kinuthia, and Therapy 5: 17. and others. 2014. “Home Visits during Pregnancy Enhance Shields, M., R. J. Guy, N. J. Jeoffreys, R. J. Finlayson, and Male Partner HIV Counselling and Testing in Kenya: A B. Donovan. 2012. “A Longitudinal Evaluation of Treponema Randomized Clinical Trial.” AIDS 28 (1): 95–103. pallidum PCR Testing in Early Syphilis.” BMC Infectious Owusu-Edusei, K. Jr., G. Tao, T. L. Gift, A. Wang, L. Wang, and Diseases 12: 353. others. 2014. “Cost-Effectiveness of Integrated Routine Shroufi, A., E. Mafara, J. F. Saint-Sauveur, F. Taziwa, and Offering of Prenatal HIV and Syphilis Screening in China.” M. C. Vinoles. 2013. “Mother to Mother (M2M) Peer Sexually Transmitted Diseases 41 (2): 103–10. Support for Women in Prevention of Mother to Child Palmer, H. M., S. P. Higgins, A. J. Herring, and M. A. Kingston. Transmission (PMTCT) Programmes: A Qualitative Study.” 2003. “Use of PCR in the Diagnosis of Early Syphilis in the PLoS One 8 (6): e64717. United Kingdom.” Sexually Transmitted Infections 79 (6): Siegfried, N., L. van der Merwe, P. Brocklehurst, and T. T. Sint. 479–83. 2011. “Antiretrovirals for Reducing the Risk of Mother-to- Peeling, R. W., D. Mabey, D. W. Fitzgerald, and D. Watson- Child Transmission of HIV Infection.” Cochrane Database Jones. 2004. “Avoiding HIV and Dying of Syphilis.” The of Systematic Reviews (7): CD003510. Lancet 364 (9445): 1561–63. Strasser, S., E. Bitarakwate, M. Gill, H. J. Hoffman, O. Musana, Reece, M., A. Hollub, M. Nangami, and K. Lane. 2010. “Assessing and others. 2012. “Introduction of Rapid Syphilis Testing Male Spousal Engagement with Prevention of Mother- within Prevention of Mother-to-Child Transmission of HIV to-Child Transmission (PMTCT) Programs in Western Programs in Uganda and Zambia: A Field Acceptability and Kenya.” AIDS Care: Psychological and Socio-Medical Aspects Feasibility Study.” Journal of Acquired Immune Deficiency of AIDS/HIV 22 (6): 743–50. Syndromes 61 (3): e40–46. Robberstad, B., and B. Evjen-Olsen. 2010. “Preventing Stringer, J. S., D. J. Rouse, S. H. Vermund, R. L. Goldenberg, Mother to Child Transmission of HIV with Highly Active M. Sinkala, and others. 2000. “Cost-Effective Use of Antiretroviral Treatment in Tanzania—A Prospective Cost- Nevirapine to Prevent Vertical HIV Transmission in Sub- Effectiveness Study.” Journal of Acquired Immune Deficiency Saharan Africa.” Journal of Acquired Immune Deficiency Syndromes 55 (3): 397–403. Syndromes 24 (4): 369–77. Rydzak, C. E., and S. J. Goldie. 2008. “Cost Effectiveness of Rapid Sutcliffe, C. G., J. H. van Dijk, F. Hamangaba, F. Mayani, and Point-of-Care Prenatal Syphilis Screening in Sub-Saharan W. J. Moss. 2014. “Turnaround Time for Early Infant HIV Africa.” Sexually Transmitted Diseases 35 (9): 775–84. Diagnosis in Rural Zambia: A Chart Review.” PLoS One Saeed, A., M. H. Kim, and E. J. Abrams. 2013. “Risks and 9 (1): e87028. Benefits of Lifelong Antiretroviral Treatment for Pregnant Sweat, M. D., K. R. O’Reilly, G. P. Schmid, J. Denison, and I. de and Breastfeeding Women: A Review of the Evidence for Zoysa. 2004. “Cost-Effectiveness of Nevirapine to Prevent 134 Major Infectious Diseases Mother-to-Child HIV Transmission in Eight African in HIV-1 Infected Children on Antiretroviral Therapy in Countries.” AIDS 18 (12): 1661–71. Kenya: A Prospective Cohort.” BMC Pediatrics 10: 33. Taha, T. E., D. R. Hoover, N. I. Kumwenda, S. A. Fiscus, Watson-Jones, D., J. Changalucha, B. Gumodoka, H. Weiss, G. Kafulafula, and others. 2007. “Late Postnatal Transmission M. Rusizoka, and others. 2002. “Syphilis in Pregnancy of HIV-1 and Associated Factors.” Journal of Infectious in Tanzania. I. Impact of Maternal Syphilis on Outcome Diseases 196 (1): 10–14. of Pregnancy.” Journal of Infectious Diseases 186 (7): Teerawattananon, Y., T. Vos, V. Tangcharoensathien, and 940–47. M. Mugford. 2005. “Cost-Effectiveness of Models for Watson-Jones, D., B. Gumodoka, H. Weiss, J. Changalucha, Prevention of Vertical HIV Transmission—Voluntary J. Todd, and others. 2002. “Syphilis in Pregnancy in Tanzania. Counseling and Testing and Choices of Drug Regimen.” II. The Effectiveness of Antenatal Syphilis Screening and Cost Effectiveness and Resource Allocation 3: 7. Single-Dose Benzathine Penicillin Treatment for the Temmerman, M., F. Mohamedalf, and L. Fransen. 1993. Prevention of Adverse Pregnancy Outcomes.” Journal of “Syphilis Prevention in Pregnancy: An Opportunity to Infectious Diseases 186 (7): 948–57. Improve Reproductive and Child Health in Kenya.” Health Watson-Jones, D., M. Oliff, F. Terris-Prestholt, J. Changalucha, Policy and Planning 8 (2): 122–27. B. Gumodoka, and others. 2005. “Antenatal Syphilis Terris-Prestholt, F., P. Vickerman, S. Torres-Rueda, N. Santesso, Screening in Sub-Saharan Africa: Lessons Learned from S. Sweeney, and others. 2015. “The Cost-Effectiveness of 10 Tanzania.” Tropical Medicine and International Health Antenatal Syphilis Screening and Treatment Approaches 10 (9): 934–43. in Peru, Tanzania, and Zambia.” International Journal of Watts, D. H., M. Lu, B. Thompson, R. E. Tuomala, W. A. Meyer Gynecology and Obstetrics 130 (Suppl 1): S73–80. III, and others. 2009. “Treatment Interruption after Terris-Prestholt, F., D. Watson-Jones, K. Mugeye, Pregnancy: Effects on Disease Progression and Laboratory L. Kumaranayake, L. Ndeki, and others. 2003. “Is Antenatal Findings.” Infectious Diseases in Obstetrics and Gynecology Syphilis Screening Still Cost Effective in Sub-Saharan 2009: 456717. Africa.” Sexually Transmitted Infections 79 (5): 375–81. Werayingyong, P., N. Phanuphak, K. Chokephaibulkit, Townsend, C. L., M. Cortina-Borja, C. S. Peckham, and S. Tantivess, N. Kullert, and others. 2013. “Economic Evaluation P. A. Tookey. 2007. “Antiretroviral Therapy and Premature of 3-Drug Antiretroviral Regimens for the Prevention of Delivery in Diagnosed HIV-Infected Women in the United Mother-to-Child HIV Transmission in Thailand.” Asia- Kingdom and Ireland.” AIDS 21 (8): 1019–26. Pacific Journal of Public Health 27 (2): NP866–76. Townsend, C. L., P. A. Tookey, M. L. Newell, and M. Cortina- WHO (World Health Organization). 2006. “Prevention of Borja. 2010. “Antiretroviral Therapy in Pregnancy: Mother-to-Child Transmission of Syphilis: Integrated Balancing the Risk of Preterm Delivery with Prevention Management of Pregnancy and Childbirth.” Standards of Mother-to-Child HIV Transmission.” Antiviral Therapy for Maternal and Neonatal Care. WHO, Geneva. http:// 15 (5): 775–83. www.who.int/reproductivehealth/publications/maternal UNAIDS (Joint United Nations Programme on HIV/AIDS). _perinatal_health/prevention_mtct_syphilis.pdf. 2011. Press Release: “World Leaders Launch Plan to ———. 2010a. “Advancing MDGs 4, 5 and 6: Impact of Eliminate New HIV Infections Among Children by 2015.” Congenital Syphilis Elimination.” Partner Brief. WHO, UNAIDS, Geneva. http://www.unaids.org/en/resources Geneva. http://apps.who.int/iris/bitstream/10665/70331/1 /presscentre/pressreleaseandstatementarchive/2011/june /WHO_RHR_HRP_10.01_eng.pdf. /20110609prglobalplanchildren. ———. 2010b. WHO Recommendations on the Diagnosis of HIV ———. 2013. Global Report: UNAIDS Report on the Global Infection in Infants and Children. Geneva: WHO. http://apps AIDS Epidemic 2013. Geneva: UNAIDS. .who.int/iris/bitstream/10665/44275/1/9789241599085 ———. 2015. 2015 Progress Report on the Global Plan towards _eng.pdf. the Elimination of New HIV Infections among Children ———. 2012a. Measuring the Impact of National PMTCT and Keeping Their Mothers Alive. Geneva: UNAIDS. Programs: Towards the Elimination of New HIV Infections http://www.unaids.org/sites/default/files/media_asset among Children by 2015 and Keeping Their Mothers Alive. /JC2774_2015ProgressReport_GlobalPlan_en.pdf. Geneva: WHO. ———. 2016a. Core Epidemiology Slides. 23 June 2016. ———. 2012b. Programmatic Update: Use of Antiretroviral Drugs http://www.unaids.org/en/resources/documents/2016 for Treating Pregnant Women and Preventing HIV Infection in /core-epidemiology-slides. Infants. HIV/AIDS Programme. Geneva: WHO. http://www ———. 2016b. On the Fast-Track to an AIDS-Free Generation. .who.int/hiv/pub/mtct/programmatic_update2012/en/. Geneva: UNAIDS. http://www.unaids.org/sites/default ———. 2013. Global Health Estimates. Online Database. /files/media_asset/GlobalPlan2016_en.pdf WHO, Geneva. http://www.who.int/healthinfo/global UNAIDS and World Bank. 2002. “National AIDS Councils: _burden_disease/en/. Monitoring and Evaluation Operations Manual.” UNAIDS, ———. 2014a. Global Guidance on Criteria and Processes for Geneva. http://www.unaids.org/sites/default/files/media Validation: Elimination of Mother-to-Child Transmission of _asset/jc808-moneval_en_1.pdf. HIV and Syphilis. Geneva: WHO. Wamalwa, D. C., E. M. Obimbo, C. Farquhar, B. A. Richardson, D. ———. 2014b. Report on Global Sexually Transmitted A. Mbori-Ngacha, and others. 2010. “Predictors of Mortality Infection Surveillance 2013. Geneva: WHO. http://apps Prevention of Mother-to-Child Transmission of HIV and Syphilis 135 .who.int/iris /bitstream/10665/112922/1/9789241507400 Yeganegh, N., H. Watts, M. Camarca, G. Soares, E. Joao, and _eng.pdf?ua=1. others. 2015. “Syphilis in HIV-Infected Mothers and Infants: ———. 2015. “WHO Validates Elimination of Mother-to- Results from the NICHD/HPTN 040 Study.” Pediatric Child Transmission of HIV and Syphilis in Cuba.” Press Infectious Disease Journal 34 (3): e52–57. release, June 30, WHO, Geneva. http://www.who.int Yin, Y. P., X. S. Chen, W. H. Wei, K. L. Gong, W. L. Cao, and /mediacentre/news/releases/2015/mtct-hiv-cuba/en/. others. 2013. “A Dual Point-of-Care Test Shows Good Wilkinson, D., K. Floyd, and C. F. Gilks. 2000. “National and Performance in Simultaneously Detecting Nontreponemal Provincial Estimated Costs and Cost Effectiveness of a and Treponemal Antibodies in Patients with Syphilis: Programme to Reduce Mother-to-Child HIV Transmission A Multisite Evaluation Study in China.” Clinical Infectious in South Africa.” South African Medical Journal 90 (8): 794–98. Diseases 56 (5): 659–65. Woldesenbet, S. A., D. Jackson, A. E. Goga, S. Crowley, Zulliger, R., S. Black, D. R. Holtgrave, A. L. Ciaranello, T. Doherty, and others. 2014. “Missed Opportunities for L. G. Bekker, and others. 2014. “Cost-Effectiveness of Early Infant HIV Diagnosis: Results of a National Study a Package of Interventions for Expedited Antiretroviral in South Africa.” Journal of Acquired Immune Deficiency Therapy Initiation during Pregnancy in Cape Town, South Syndromes 68 (3): e26–32. Africa.” AIDS and Behavior 18 (4): 697–705. 136 Major Infectious Diseases Chapter 7 Cost-Effectiveness of Interventions to Prevent HIV Acquisition Geoff P. Garnett, Shari Krishnaratne, Kate L. Harris, Timothy B. Hallett, Michael Santos, Joanne E. Enstone, Bernadette Hensen, Gina Dallabetta, Paul Revill, Simon A. J. Gregson, and James R. Hargreaves INTRODUCTION HISTORY OF THE HIV/AIDS PANDEMIC AND Because of the severe health consequences of human PREVENTION INITIATIVES immunodeficiency virus/acquired immune deficiency Clusters of fatal infectious and chronic diseases were first syndrome (HIV/AIDS) and the costs of lifelong treat- detected in 1981, leading to remarkably rapid identifica- ment, inexpensive and effective HIV prevention is tion of HIV; development of tests to identify persons bound to be cost-effective. But what constitutes HIV infected; and mapping of the routes of transmission via prevention, and can it be affordable and effective? The sex, blood products, and sharing of injection equipment use of condoms that cost a few cents and prevent a (Oppenheimer 1988). Unfortunately, it also became young adult from acquiring a chronic and fatal disease clear that over a long and variable period averaging will, over time, be cost saving. Avoiding sex with about 12 years, everyone infected would develop AIDS someone who is infected with HIV/AIDS will be even and die (Brandt 1987). This awareness lent urgency more so. What can be done to get people to use con- to identifying ways of preventing and treating HIV. doms? What can be done to facilitate the avoidance of Restrictions on who could donate blood and HIV risky sexual encounters? Additional efficacious bio- screening of blood products were found to close off medical tools have become available, but similar ques- transmission via blood products (Hoots 2001). The use tions persist: What can be done to get young women of clean needles and syringes was found to stop trans- at risk to use oral truvada effectively as preexposure mission among people who injected drugs (Fuller, Ford, prophylaxis (PrEP) and to get young men at risk to be and Rudolph 2009). Consistent and correct use of circumcised? The answers to these questions will condoms was found to stop sexual transmission of HIV determine what packages of prevention are essential, (Steiner and others 2008). Lowering the number of how much prevention programs should cost, and how sexual partners was found to reduce risks, with mutually cost-effective they can be. This chapter reviews current monogamous couples protected from sexual transmis- evidence about the efficacy, effectiveness, and costs sion (May and Anderson 1987). Although too late for of HIV/AIDS prevention products, programs, and many, this new knowledge allowed many others to avoid approaches. Corresponding author: Geoff P. Garnett, Bill & Melinda Gates Foundation, Seattle, Washington, United States; Geoff.Garnett@gatesfoundation.org. 137 acquiring HIV infection. However, it also required peo- set for treatment expansion (including prevention ple to perceive the risk and to adopt and rigorously among HIV-negative persons) (Piot and others 2015). adhere to difficult and unappealing behaviors. HIV However, logistical and social barriers mean that some continued to spread (Anderson and others 1991). delays will occur between infection and treatment, and Quantifying the impact of these interventions is many will fail HIV treatment. Even in populations difficult. It requires knowing what the incidence would in which coverage of treatment has hit the 90 percent be in their absence. Moreover, a concentrated epidemic targets for diagnosis, initiation of treatment, and sup- with heterogeneous transmission and acquisition risks pression of viral load, the disease continues to spread will become saturated (Anderson and May 1990). Thus, a (Gaolathe and others 2016). drop in incidence and leveling off of prevalence are expected, even in the absence of prevention (Hallett and CHALLENGES IN REVIEWING THE EFFICACY others 2006). Nonetheless, reported changes in risk behavior have reduced the spread of HIV in some popu- AND EFFECTIVENESS OF HIV PREVENTION lations, particularly among people who inject drugs Determining the causal impact of prevention activities (PWID) and men who have sex with men (MSM) in in affecting outcomes of interest (that is, reducing HIV high-income countries (Fuller, Ford and Rudolph 2009), infections and ultimately preserving health) has proved sex workers and their clients in Thailand (Nelson and much more challenging than for many other kinds of others 1996), and the general population in Uganda and health care intervention. We explore reasons for this Zimbabwe (Gregson and others 2007; Stoneburner and challenge related to difficulties in categorizing and defin- Low-Beer 2004). ing prevention interventions, defining and measuring In 1996, combination antiretroviral treatment was intervention endpoints, and designing studies of impact. reported to be efficacious in reducing viral replication and in reconstituting the immune system (Eron and Hirsch 2008). Effective treatment transformed the Categorizing and Defining HIV Prevention response to the epidemic, initially dramatically reducing Interventions AIDS deaths in high-income countries (Palella and oth- HIV prevention interventions have been categorized as ers 1998). Major reductions in medication costs biomedical, behavioral, or structural. These categories and increased investments from the U.S. President’s are based on whether the intervention includes use of Emergency Plan for AIDS Relief (PEPFAR); the Global a biomedical product or procedure, involves people Fund to Fight AIDS, Tuberculosis and Malaria (the changing their risk behavior, or targets changes in the Global Fund); and others led to widespread treatment in environment within which risk takes place (Merson and low- and middle-income countries (LMICs) (Ford and others 2008). Thinking of these approaches as separate others 2013). Initially, the impact of treatment on HIV and distinct ignores the requirements for interven- transmission and spread was unclear, but evidence from tions to be effective in the real world. HIV transmissi- randomized controlled trials (RCTs) demonstrated that bility needs to be reduced—either by a product used antiretroviral treatment was extremely efficacious in during exposure or by a reduction in exposure. We call preventing transmission (Cohen and others 2011). This these direct mechanisms. Getting these products to be finding and the clinical benefits of early treatment led used requires behavioral changes, for example using the World Health Organization (WHO) to recommend condoms, taking PrEP, or getting circumcised. Such that all people living with HIV should receive treatment changes are only possible when condoms are available; (WHO 2015b; Holmes and others 2017). PrEP programs and circumcision are organized; and However, the number of infected people on treat- laws do not prevent people from accessing clean needles, ment remains far below the total number infected, and condoms, circumcision, or oral PrEP. Holistic or com- further prevention is required. By the end of 2014, an bined approaches are required (Hankins and de estimated 36.9 million people were living with HIV Zalduondo 2010; Schwartländer and others 2011), and globally, and 2 million new infections were occurring the trials to test interventions need to consider all three each year (UNAIDS 2015), even with more than categories, each of which should be carefully described 15 million people globally receiving treatment. With the if they are to be replicated and scaled up (figure 7.1). global response becoming difficult to sustain, there is an Such combination approaches have been promoted to urgent need not only to scale up treatment, but also to prevent the spread of HIV (Hankins and de Zalduondo make available other affordable and effective packages of 2010; Schwärtlander and others 2011). However, prevention. Further expansion of treatment will reduce combining the use of prevention products, each with the infectiousness of infected persons; targets have been evidence of biological efficacy, can also be thought of as 138 Major Infectious Diseases Figure 7.1 Elements of Discipline-Specific and Holistic Approaches to Intervention Discipline-specific perspective Holistic perspective Types of interventions are independent, and trade-offs Holistic view recognizes the interdependencies across are made between those in which to invest. levels in a causal pathway: Interventions act to change the risk of exposure or transmission during exposure: Structures Biomedical • Development and use of condoms, interventions PrEP, VMMC, microbicides, and others Environments Behaviors • Efforts to encourage individuals Policies Sexual behaviors Direct mechanisms Behavioral to reduce behaviors that increase Demand Condoms interventions risk of HIV and to use biomedical Exposures and products Adherence VMMC transmission Laws PrEP • Efforts to improve the conditions suspected to influence behaviors Structural that lead to spread of HIV, including interventions use of incentives to reduce such behaviors Intervention programs Note: HIV = human immunodeficiency virus; PrEP = preexposure prophylaxis; VMMC = voluntary medical male circumcision. combination prevention, with a narrower perspective of intermediate risks are not reliable measures of HIV risk, which products are needed (Cremin and others 2013; so the findings of studies using other endpoints—the Vermund and others 2013). It is simpler to standardize a majority of studies—have to be treated with caution prevention product and experimentally test whether that (Garnett and others 2006). product has biological efficacy than to standardize the design of interventions to change the environment and behaviors (Hallett, White, and Garnett 2007; Lagakos and Designing Studies of Impact Gable 2008). Accordingly, with interventions requiring A third challenge pertains to the design of studies structural and behavioral components, there are major measuring the efficacy and effectiveness of HIV preven- challenges in measuring the effects and costs of preven- tion interventions. For biomedical products intended to tion because the interventions are rarely standardized and protect the individual, RCTs provide rigorous, causal units of intervention are often unclear. evidence of efficacy (Lagakos and Gable 2008), but they do not guarantee impact at scale (Hallett, White, and Garnett 2007). Furthermore, some structural and behav- Defining and Measuring Intervention Endpoints ioral elements of interventions need to be delivered to In addition to defining the interventions, it is important to communities, not individuals—for example, education define the endpoints of interest, which in studies are often and communication campaigns or changes in policies. To intermediate variables instead of HIV incidence. Ultimately, have an impact, interventions often need to reach key the goal of prevention is to reduce incidence, but measur- individuals and to scale up what protects individuals so ing incidence is difficult and expensive, especially where that the interventions protect communities. Such inter- incidence is low (Hallett, White, and Garnett 2007). If ventions can be evaluated in cluster or community ran- reducing the number of partners or increasing the use of domized trials, but conducting such trials can be condoms could be assumed to decrease HIV incidence, expensive and logistically challenging (Hallett, White, then these intermediate measures would be reasonable and Garnett 2007). When these trials find no impact, it is endpoints for trials (Laga and others 2012). Alternatively, not clear whether the intervention was ineffective or the if correlated measures such as other acute sexually trans- implementation in the trial was ineffective (Hallett, mitted infections (STIs) or pregnancy share risks with White, and Garnett 2007). RCTs are desirable if causality HIV, they can be used to indicate a change in HIV risks. is to be established, the trials need to have a valid Unfortunately, the causal pathways are often not clear and counterfactual with which to compare the effect of the Cost-Effectiveness of Interventions to Prevent HIV Acquisition 139 intervention and should be randomized to distribute disadvantage of thinking in programmatic siloes, but it unmeasured confounding variables (Gertler and others is useful for illuminating the steps needed to reduce the 2011). Evaluation using methods other than RCTs risk of acquisition (figure 7.2). A hypothetical cascade can examine the delivery of programs at scale, trends in starts with the number of individuals who would the incidence of infection and disease, qualitative data acquire infection or who are at risk over a period of time on risks and responses to interventions, and logical path- and calculates who will remain uninfected (or become ways by which interventions could have an impact. infected) because of the intervention. Steps in the Analyses from such studies create a better understanding cascade represent the potential reasons for failure of a of the results of RCTs and yield plausibility arguments prevention intervention. useful for improving implementation (Hargreaves and From the perspective of a policy maker or imple- others 2016). menter, delivering a successful intervention requires targeting the population at risk, creating demand for pre- vention in that population, having a system in place to HIV PREVENTION CASCADES supply prevention, promoting adherence, and providing a Prevention cascades could be a powerful tool for direct and biologically efficacious prevention mechanism. analyzing how a prevention product should be delivered HIV prevention trials that focus on these aspects and identifying the steps required for it to have an can be categorized. In a description of the literature, impact. To date, in studying HIV interventions, cascades Krishnaratne and others (2016) classified reviews and for treatment and prevention of mother-to-child trans- primary studies under one of the following: mission (PMTCT) have predominated (Gardner and others 2011; Mahendra and others 2007). The WHO has • Demand interventions, in which the principal aim is promoted a comprehensive approach to HIV, including to influence behavior by targeting risk perception or prevention for those who are negative (WHO 2015a). strengthening awareness of, and positive attitudes toward, A single, all-encompassing cascade is attractive. However, HIV prevention behaviors and technologies. These whether it can be populated with data and used success- interventions could include providing information, fully remains to be seen. education, and communication and aim to influence An alternative approach is to consider the different perceived norms through peer-based approaches. ways of preventing HIV and develop multiple cascades • Supply interventions, in which the principal aim is to (Garnett and others 2016). This approach has the influence the supply of HIV prevention products and messages. These interventions include mass condom Figure 7.2 HIV Prevention Cascade for a Single Intervention distribution, needle exchange initiatives, attempts to mainstream prevention within other services, and STI treatment strategies. • Use of or adherence to interventions, in which Would otherwise acquire infection during the principal aim is to support adoption or mainte- nance of prevention behaviors, including the use of prevention technologies. They include interventions period (or at risk) that provide risk counseling and target social deter- minants of behavior hypothesized to encourage or discourage access and adherence, such as conditional cash transfers or livelihood interventions. • Direct mechanisms for HIV prevention, in which the principal aim is to stop transmission. These inter- ventions include biomedical products or procedures, for example, microbicides or medical male circumci- sion (MMC). Lack of availability Lack of uptake Lack of adherence or fidelity SYSTEMATIC REVIEWS OF HIV PREVENTION Lack of efficacy Remain uninfected Limit to improvement in To understand the evidence available on prevention and the cascade address some of these challenges, Krishnaratne and Source: Garnett and others 2016. others (2016) undertook three systematic reviews of pre- Note: HIV = human immunodeficiency virus. vention interventions. They then reviewed the original 140 Major Infectious Diseases studies, reclassifying the interventions into a cascade Results from demand, supply, and use interventions framework, dissecting different endpoints, and grading could include intermediate variables, and Krishnaratne the quality of evidence. The initial search identified 666 and others (2016) included condom use and HIV testing reviews of which 88 were eligible for inclusion. From as endpoints. In addition, HIV prevalence could be these 88 reviews, 1,964 primary studies were identified, of compared between arms in a trial as a marker of past which 292 were eligible for inclusion. incidence, rather than directly assessing incidence by The division of studies within the cascade frame- following up trial participants. For this reason, HIV work is notable for several reasons (figure 7.3). First, prevalence was included as an endpoint. Table 7.1 there were many more studies of demand interven- summarizes the evidence from the review; the number tions and direct mechanisms than studies of supply of studies by type of intervention is shown with the and use interventions. Most supply interventions were number of RCTs in parentheses. mass distribution of condoms and clean needles and Of note, HIV incidence was most often an endpoint syringes. The overwhelming majority of use interven- in trials of direct mechanisms, and some of these inter- tions entailed counseling. With regard to study design, ventions were consistently found to be efficacious. use and direct interventions were highly likely to Where the endpoint was HIV incidence, the evidence for be RCTs. demand, supply, and use interventions was either mixed Two dimensions were used to summarize the or consistently ineffective; the single exception was a evidence, following the scheme that Mavedzenge, Luecke, non-RCT of couples counseling. Demand interventions and Ross (2014) used to review HIV prevention inter- were ineffective in reducing HIV prevalence, whereas ventions aimed at adolescents. The first dimension supplying condoms and clean needles and syringes was classifies the level of internal validity and replication, consistently associated with a decline in HIV prevalence emphasizing proof of causation and generalizability. It but only in non-RCT studies. The majority of studies does not include evidence of scalability, impact, or measured condom use and HIV testing rather than HIV cost-effectiveness. The second dimension describes the incidence and prevalence. Across populations, there is direction of the effect. good evidence of effectiveness in increasing condom use Figure 7.3 Mapping HIV Prevention Studies to the HIV Prevention Cascade “Supply” type interventions: 12 reviews (35 primary studies [6 RCTs]) Mass condom distribution and associated policies (20) Needle and syringe programs and associated policies (6) Health system policies: integrating family planning and HIV (6), STI control (3) “Demand” type interventions: 40 reviews (108 primary studies [24 RCTs]) IEC approaches (54) Peer-based approaches (54) “Use” interventions: 16 reviews (51 primary studies [26 RCTs]) Counseling approaches (40) Social determinants approaches: cash transfers (3), microfinance (8) “Direct mechanisms”: 29 reviews (98 primary studies [34 RCTs]) PrEP (6), condoms (4), VMMC (64), STI (7), microbicides (12), vaccines (5) Source: Based on a systematic review of HIV prevention studies in low- and middle-income countries by Krishnaratne and others 2016. Note: HIV = human immunodeficiency virus; IEC = information, education, and communication; PrEP = preexposure prophylaxis; RCT = randomized controlled trial; STI = sexually transmitted infection; VMMC = voluntary medical male circumcision. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 141 Table 7.1 Trials Assessing HIV Incidence and Prevalence, with Condom Use and Testing as Endpoints Outcome Condom Incidence; Prevalence; use; Testing; number number number number of studies of studies of studies Condom of studies Testing: (number Incidence: (number of Prevalence; (number use: QA (number QA Type of intervention of RCTs) QA rating RCTs) QA rating of RCTs) rating of RCTs) rating Demand-side interventions Impact of IEC interventions focused on youth 3 (1) B4 1 (1) B4 28 (7) A3 Impact of IEC interventions focused on men 9 (3) A2 1 (0) C1 Impact of IEC interventions focused on women 2 (2) B3 Impact of IEC interventions using mass media 1 (1) B3 9 (1) B4 Impact of IEC interventions focused on PWUD 4 (3) A1 Impact of peer-based interventions focused on youth 1 (1) B4 11 (0) C2 2 (0) C1 Impact of peer-based interventions focused on MSM 3 (1) B1 1 (0) C1 Impact of peer-based interventions focused on FSW 3 (1) B4 4 (0) C4 22 (3) A2 3 (0) C1 Impact of peer-based interventions focused on PWUD/alcohol 2 (2) B4 1 (1) B4 5 (2) B3 Impact of peer-based interventions with no population focus 10 (2) B1 1 (0) C1 Supply-side interventions Impact of interventions that integrate HIV services into routine care 1 (0) C1 5 (0) C1 Impact of clean needle and syringe programs 2 (0) C3 6 (0) C1 Impact of condom distribution interventions 3 (0) C1 20 (5) A1 Impact of community-level STI interventions 3 (3) A4 1 (1) B4 Adherence interventions Impact of couples-based counseling 1 (0) C1 9 (3) A1 4 (3) A3 Impact of HIV testing and counseling 1 (1) B4 8 (1) B2 3 (2) B1 Impact of individual-level counseling 1 (1) B3 12 (7) A1 2 (1) B3 Impact of HIV prevention counseling 7 (4) A3 Impact of microfinance interventions 1 (1) B4 8 (4) A3 1 (1) B1 Impact of cash transfer interventions 2 (2) B4 2 (2) B1 1 (1) B4 table continues next page 142 Major Infectious Diseases Table 7.1 Trials Assessing HIV Incidence and Prevalence, with Condom Use and Testing as Endpoints (continued) Outcome Condom Incidence; Prevalence; use; Testing; number number number number of studies of studies of studies Condom of studies Testing: (number Incidence: (number of Prevalence; (number use: QA (number QA Type of intervention of RCTs) QA rating RCTs) QA rating of RCTs) rating of RCTs) rating Direct mechanisms MMC heterosexual risk (female to male) 38 (3) A1 MMC heterosexual risk (male to female) 7 (1) B3 Male circumcision, MSM individual level 19 (0) C3 Condoms (heterosexual), individual level 4 (0) C1 Oral PrEP (overall), individual level 6 (6) A2 Microbicide prophylaxis, individual-level studies 12 (12) A3 STI treatment, individual-level studies 7 (7) A4 HIV vaccine, individual-level studies 5 (5) A3 Source: Krishnaratne and others 2016. Note: Level of internal validity and replication available is defined as A (3 or more RCTs), B (1–2 RCTs), and C (no RCT). The direction of effectiveness is defined as 1 (consistently effective), 2 (mainly effective), 3 (mixed results), and 4 (consistently ineffective). FSW = female sex workers; HIV = human immunodeficiency virus; IEC = information, education, and communication; MMC = medical male circumcision; MSM = men who have sex with men; STI = sexually transmitted infection; PrEP = preexposure prophylaxis; PWUD = people who use drugs; QA = quality assessment; RCT = randomized controlled trial. Blank cells = not available. and testing. Unfortunately, the impact of condom use To partly address this discrepancy, trials distinguish depends on who is using the condoms and when, that is, between participants that do and do not follow the inter- in sexual acts where they would be exposed to the virus. vention and trial protocol closely. Analysis of the effects The impact of HIV testing depends on changes in subse- according to protocol attempts to approximate the quent behavior, and there is scant evidence that this is a underlying biological effect of the product, while focus or product of HIV testing. intention-to-treat analysis attempts to approximate its Cash transfers are an area of interest. An RCT found potential effectiveness in the real world. Unfortunately, that cash transfers were associated with reduced HIV the situation is much more complicated because there is prevalence in young women in Malawi (Baird and others very likely a difference between the efficacy observed in 2012). However, the effect was reversed when the trans- the according-to-protocol analysis of a trial and the bio- fers were withdrawn. Another study found no impact of logical effect of a product in reducing transmission. A cash transfers in a less resource constrained setting, trial compares the cumulative incidence of HIV among where many girls stayed in school regardless of the inter- participants who potentially have multiple exposures. vention (HPTN 2015). Biological efficacy is the reduction in risk for one expo- sure and is the parameter that would logically be used in EFFICACY, EFFECTIVENESS, AND DURATION models of HIV transmission (Jewell and others 2015). To add to the confusion, a product could protect a OF PROTECTION fraction of individuals from all challenges (take-type The distinction between efficacy and effectiveness is efficacy) or all individuals from a fraction of chal- crucial and somewhat opaque. By definition, trials mea- lenges (degree-type efficacy) and have the same effi- sure the effect of an intervention in an artificial setting cacy (Garnett 2005). Cumulatively as the number of because the subjects have been recruited, have consented challenges increases, take-type efficacy will fare better to take part, and are observed. Accordingly, there is a than degree-type efficacy because, in the former, the well-acknowledged distinction between effectiveness in number of breakthrough infections will plateau as all a trial and in the real world, with real-world effective- those still at risk acquire infection. RCTs are not capa- ness expected to be lower than trial effectiveness. ble of distinguishing between these types of efficacy. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 143 Likewise, effectiveness in the real world will depend No trials found that the use of diaphragms affords on similarities in adherence and exposure in different significant protection. Marrazzo and Cates (2011) com- settings with those found during the trial. pared protection using a diaphragm and condom versus A further challenge is in estimating how long using a condom alone and found that using both pro- protection lasts. A trial will uncover whether protection vided no additional protection. They concluded that is short lived, but if protection wanes over the medium diaphragms should not be relied on for protection or long term, assessing the duration of protection will against STIs or HIV. be harder. The results of studies of the effectiveness of direct mechanisms of HIV prevention need to be considered, keeping these problems of interpretation Medical Male Circumcision in mind. Early in the study of heterosexually transmitted HIV infection in Sub-Saharan Africa, an association was observed at both the national (Bongaarts and others 1989) EVIDENCE OF EFFICACY OF DIRECT and individual (Cameron and others 1989) levels between circumcision status and HIV risk. Subsequent data collec- MECHANISMS OF HIV/AIDS PREVENTION tion repeatedly showed a protective effect of circumcision, Barrier Methods with a systematic review and meta-analysis of 27 studies Male and female condoms that prevent HIV from showing a 48 percent lower risk without controlling for crossing the barrier in vitro may prevent the acquisition other variables, and a meta-analysis of 15 studies showing of HIV (Steiner and others 2008). However, there is a a reduced risk of 58 percent that did account for problem in ethically and practically testing the effective- confounding (Weiss, Quigley, and Hayes 2000). ness of condoms in RCTs. Observational studies need to Despite this observational evidence, two major consider whether condom use is consistent and correct questions remained: and whether self-reported use is valid. Observed effec- tiveness will likely underestimate biological efficacy. • Could uncontrolled confounding related to the char- Good-quality studies on the effectiveness of condoms acteristics and behaviors of men from cultural groups against HIV are lacking. Estimates of effectiveness in the that circumcise explain the observed protective effect? past have been low. Weller (1993) concluded that condoms • Would an intervention providing adult male cir- were only 69 percent effective in preventing acquisition of cumcision provide the same protection as infant HIV. However, that study misaggregated some groups circumcision? on condom use and did not compare “always” users with “never” users. Other researchers attempted to address this These questions required clinical trials using ran- issue by exploring the direction of transmission. Pinkerton domization to avoid uncontrolled confounding. The and Abramson (1997) concluded that condoms were first trial in South Africa was stopped early because 90 percent to 95 percent effective when used consistently. circumcision was found to be protective, with 20 HIV However, Davis and Weller (1999) criticized their paper infections in circumcised men and 49 in uncircumcised for incorrectly categorizing “sometimes” users with never men. Calculated rates of 0.85 per 100 person years users and estimated effectiveness at 87 percent (as low as in circumcised men and 2.1 per 100 person years in 60 percent or as high as 96 percent). uncircumcised men meant that this was a 60 percent A meta-analysis by Weller and Davis (2002) concluded reduction in risk (Auvert and others 2005). In Kenya, that condoms reduced HIV seroconversion approxi- 22 circumcised men and 47 uncircumcised men mately 80 percent, comparing always users with never acquired HIV, representing a 53 percent reduction in users; their analysis used the difference between the two risk (Bailey and others 2007). In Uganda, 0.66 cases of pooled rates to estimate effectiveness. HIV per 100 person years in circumcised men and None of the reviews identified HIV effectiveness data 1.3 cases per 100 person years in uncircumcised men for female condoms. However, one systematic review of represented a 55 percent reduction in risk (Gray and the effectiveness of female-controlled barrier methods in others 2007). These three rigorous trials provided defin- preventing STI and HIV transmission concluded that itive evidence of the protective effect of adult MMC in RCTs provide evidence that female condoms confer as protecting men from heterosexual acquisition of HIV much protection from STIs as male condoms (Minnis infection over time. and Padian 2005). However, this finding was based on There is little evidence that MMC directly reduces the results for chlamydia, gonorrhea, syphilis, and tricho- risk of HIV in women through vaginal intercourse: the moniasis rather than for HIV. one RCT that included this outcome measure was 144 Major Infectious Diseases stopped early with some infections acquired by women comparing immediate PrEP to PrEP deferred for one whose partners acquired infection before the wound year found a similar 86 percent relative effectiveness from the circumcision had healed (Wawer and others (McCormack and others 2016). In both cases, infections 2009). However, the fact that male circumcision reduces occurred among those who ceased taking the drug. the incidence of HIV in men will indirectly benefit Subsequent WHO guidelines, based on a meta-analysis women by lessening their exposure to HIV. Male circum- showing 51 percent effectiveness across reviewed trials, cision has also been shown to reduce rates of genital included a strong recommendation for offering oral ulcers in men, as well as bacterial vaginosis and tricho- PrEP to persons with a high risk of acquiring HIV moniasis in female partners of circumcised men (Tobian, (WHO 2015b). Kacker, and Quinn 2014). For MSM, the evidence of protection via circumci- Vaginal or Rectal Microbicides sion is weak. A meta-analysis of observational studies A wide range of topical products to prevent HIV acquisi- concluded that there was insufficient evidence that tion has been studied. A review by Obiero and others circumcision provided protection for MSM (Millett and (2012) of 13 trials for vaginal microbicides conducted others 2008). In a subgroup analysis by sexual role in the between 1996 and 2011 found no evidence of a significant relationship, 7 of 21 studies indicated that male circum- reduction in risk of HIV in a pooled analysis, but one cision is more protective among MSM who have a proof-of-concept trial of tenofovir gel and a placebo gel mainly insertive role (Wiysonge and others 2011). conducted in South Africa found a significant reduction in the risk of acquisition (Abdool Karim and others 2010). Oral Preexposure Prophylaxis Two further phase 3 studies of tenofovir gel showed no The effectiveness of oral PrEP using either truvada significant effectiveness: one found only a 14.5 percent (tenofovir plus emtricitabine) or tenofovir alone has lower incidence of HIV infection in the tenofovir arm been studied in trials of MSM and of men and women (Marrazzo and others 2015) and one found no difference in HIV-discordant couples (figure 7.4). (Rees and others 2015). Analyses accounting for adherence have shown Work on longer-acting topical delivery of an anti- greater effectiveness with high adherence and no signifi- viral agent through a vaginal ring was tested in women cant effectiveness with poor adherence (Marrazzo and in two trials. One found 27 percent effectiveness (Baeten others 2015; Van Damme and others 2012). and others 2016), and one found 31 percent effective- A study of on-demand PrEP in MSM found ness (IPM 2016). In both cases, effectiveness was higher 86 percent effectiveness (Molina and others 2015); a trial in women older than age 21 years, with continuous Figure 7.4 Effectiveness of Oral PrEP in Randomized Controlled Trials, by Order of Increasing Effectiveness 100 50 Effectiveness (%) 0 –50 –100 –150 20 ers TD s F/ ers 20 ers TD d F/ nd 20 d 20 d 20 d 20 d 12 er 15 an rs an rs an rs an rs an TD o a h 20 oth TD th th 10 F C 15 F C 12 13 12 16 ot FT 20 zo he e he a he n he ck 12 d o do FT 15 razz ot amm ot pany ot igpe nd d rs az ot ma an 20 an an he Mar ta a or Th oo nD M en en na an cC Ch et et oli Va Gr M 20 Ba Ba M ot rs he ot Sources: Baeten and others 2012; Marazzo and others 2015. Note: FTC = emtricitabine; PrEP = preexposure prophylaxis; TDF = tenofovir. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 145 use of the ring needed to prevent HIV acquisition. MEASURING THE HIV PREVENTION Whether this effectiveness is sufficient to warrant CASCADE TO EXPLORE IMPACT launch of a product remains to be seen. From a population-based study in rural Zimbabwe, HIV prevention cascades were constructed to determine the Vaccines effectiveness of HIV prevention and gaps in preventing Systematic reviews of vaccines were included in infection (Garnett and others 2016). Figure 7.5 shows broader reviews of HIV prevention measures (Marrazzo two cascades. The cascade in panel a is for male circum- and Cates 2011; Padian and others 2010). Of four cision in 2009–11, where the cascade is applied to a trials, only one (a double-blind, placebo-controlled population of men at risk of HIV. The initial step trial conducted in more than 16,000 adults in Thailand) depends on whether there is a provider of voluntary found a protective effect. In a modified intention-to- medical male circumcision within 20 kilometers. Where treat analysis, the combination of a vaccine plus a there is, the next step includes persons who report hav- booster was 31 percent effective (Rerks-Ngarm and ing been circumcised; because adherence does not apply others 2009). to circumcision, there is no drop-off at this step. The next drop-off is where circumcision is not efficacious. Finally, persons on the right were protected by circumcision. SEXUALLY TRANSMITTED INFECTION The cascade in panel b is for HIV testing and TREATMENT AS HIV PREVENTION counseling (HTC), with a reduction in the number of HIV acquisition is correlated with the presence of other partners as the direct mechanism for protecting women STIs, and it has been hypothesized that the presence of from infection. The cascade is shown for women in two these other infections could increase the transmissibility periods, 2009–11 and 2012–13. The first step is small of HIV. Genital ulceration associated with chancroid; because most women have access to testing services, but syphilis; herpes simplex virus (HSV) type 1 or 2; or many do not use them; this situation improves over time. inflammation associated with chlamydia, gonorrhea, The greatest fall-off in protection occurs because the vast trichomoniasis, or human papillomavirus may increase majority of women tested do not reduce the number risks of HIV (Røttingen, Cameron, and Garnett 2001). of sexual partners, which suggests that HTC services Unfortunately, because of the common routes of trans- will have little impact on HIV acquisition through this mission and the impossibility of measuring the complete mechanism. sexual network, observational studies will always have unmeasured confounding. To determine the effect of controlling STIs on HIV COSTS OF HIV PREVENTION incidence, community randomized trials provided To understand the cost-effectiveness of HIV prevention enhanced STI control in intervention communities. interventions and how to budget for them, the costs of The first of these trials, conducted in Mwanza, delivering the interventions have to be known. A litera- Tanzania, using syndromic management of STIs found ture search yielded 2,151 references, of which 66 were a 40 percent reduction in HIV incidence (Grosskurth relevant. These papers varied in the interventions costed, and others 1995). This finding was not replicated in the types of costing undertaken, the analyses performed, further trials of syndromic management or mass treat- and the ability to link cost to effectiveness. Finding ment of the population (Gregson and others 2007; comparable, well-documented costing of interventions Kamali and others 2003; Wawer and others 1999). This linked to outcomes is challenging. Walker (2003) found discrepancy was explained by the importance of symp- that many interventions were not covered, costs were toms to HIV risk and stage of the HIV epidemic. inadequately described, and impact was rarely mea- Padian and others (2010) reviewed nine STI treatment sured. In another systematic review, Galárraga and trials, only one of which was effective in preventing others (2009) found that HIV prevention was extremely HIV acquisition. Three RCTs assessed the impact on cost-effective compared with treatment, but effective- HIV acquisition of suppressing HSV-2 with acyclovir. ness was rarely measured, there was a gap in examining None of the trials found a protective effect. Adherence bundles of prevention interventions, and synergies were was reportedly mixed between the trials, and good pre- not included. vention services were available to the control group, Avenir Health maintains a database of HIV interven- which may have affected behavior in all arms. tion costs from 1993 to the present (Avenir Health 2016). The strength of the HSV-2 regimen also might have UNAIDS (Joint United Nations Programme on HIV/ influenced susceptibility to HIV. AIDS) uses this database to estimate resource needs 146 Major Infectious Diseases Figure 7.5 Prevention Cascades from Rural Zimbabwe a. Cascade for voluntary medical male circumcision, data collected 2009–11 1,000 per 1,000 population of men at risk of HIV Infections prevented or not prevented 800 600 400 200 0 2009–11 Lack of availability (20 kilometers) Lack of uptake Lack of efficacy Lack of adherence Infections prevented Limit to improvement in the cascade b. Cascade for testing leading to reductions in the number of partners reported by women, 2009–11 and 2012–13 1,000 1,000 Infections prevented or not prevented per 1,000 population of women 800 800 600 600 400 400 200 200 0 0 2009–11 2012–13 Lack of HTC availability Lack of HTC uptake Lack of partner reduction Lack of efficacy Infections prevented Limit to improvement in the cascade Source: Garnett and others 2016. Note: HTC = HIV (human immunodeficiency virus) testing and counseling. (Hecht and others 2010). A large fraction of the costs in Other findings are that India has the most data on costs the database was gathered from regional experts in a of HIV prevention, costs are often given per person series of workshops. The interventions reporting costs reached, costs are generally for specific programs rather tend to be delivered in health facilities—PMTCT, than the whole system, and costs decline over time VMMC, and HTC. This last intervention, HTC, is mostly (Avenir Health 2016). geared toward diagnosing HIV-infected individuals; Two international studies costing HIV prevention however, if testing and counseling attempt to reduce nearly a decade apart had similar findings: extreme het- risky behaviors, the intervention could be considered erogeneity in unit costs across sites and possibility of preventive. The evidence suggests that persons who are economies of scale in delivery (Bautista-Arrendondo and HIV-positive do alter their behaviors, while those who others 2015; Marseille and others 2007). One study are HIV-negative do not (Hallett and others 2009). examined the costs of voluntary counseling and testing; Cost-Effectiveness of Interventions to Prevent HIV Acquisition 147 male circumcision; PMTCT; risk reduction for people Based on RCT results for adult male circumcision, who inject drugs; risk reduction for sex workers; treat- the WHO promoted the scale-up of circumcision pro- ment of STIs; information, education, and communica- grams in 14 Sub-Saharan African countries, leading to tion; condom social marketing; and school curricula in more than 9 million circumcisions (WHO 2015c). The India, Mexico, the Russian Federation, South Africa, and circumcision programs allowed costs to be estimated Uganda in 2003 and 2004 (Marseille and others 2004). across countries and across models of circumcision, The majority of costs measured were for “people reached,” including shifting tasks from doctors to nurses, using except for voluntary counseling and testing and for cir- models to improve client flow, and using circumcision cumcision. Costs for the former varied 40-fold in Uganda devices. In six countries, Bollinger and others (2014) and 2.5-fold in South Africa without adequate explana- found that the cost per circumcision varied between tion. Programs showed efficiencies of scale, but the pro- US$22 and US$70 (table 7.2). This finding is in line portion of variation differed greatly between countries. with estimated cost per circumcision used in exploring The other study examined the costs of HTC, PMTCT, the cost-effectiveness of VMMC. and VMMC in Kenya, Rwanda, South Africa, and Zambia between 2011 and 2013 (Bautista-Arrendondo and others 2015; Sosa-Rubí and others 2015). For facilities Cost-Effectiveness Analysis carrying out HTC, a 10 percent increase in scale was In the Avahan program, the number of HIV infections associated with a 5.8 percent reduction in costs. For averted was derived by comparing observed prevalence facilities carrying out VMMC, a 10 percent increase in with a modeled counterfactual representing HIV procedures was associated with a 41 percent reduction in spread without the intervention and self-reported costs, and an increase in procedures was positively increase in condom use among sex workers (Vassall correlated with quality as measured in exit interviews. and others 2014). Often the modeled effectiveness of The main focus of both studies was efficiency. However, interventions compares modeled incidence with and measuring the quality and impact of services is difficult, without the intervention; this is especially true of but necessary to determine efficiency, since lower costs interventions using products in development and could otherwise be offset by reduced effectiveness. before scale-up. For prevention interventions, effec- Extensive costing of HIV prevention service delivery tiveness is best established for VMMC and PrEP, was carried out between 2004 and 2008 as part of the Bill the direct mechanisms with the most meaningful & Melinda Gates Foundation’s Avahan Program in India. cost-effectiveness analyses. In this program, costs of prevention services depended Models were used to demonstrate that circumcision on the scale of support provided to nongovernmental would be a cost-saving intervention where circumcision organizations, extent of community involvement, and rates are low and HIV incidence is high. In a systematic organization of clinical services (Lépine and others review of circumcision cost-effectiveness, costs per 2016). A model-derived estimate of impact found a HIV infection averted varied from US$174 to US$2,808 mean incremental cost of US$785 per HIV infection (Uthman and others 2010). In a subsequent analysis, averted and US$46 per disability adjusted life-year Njeuhmeli and others (2011) found that circumcision (DALY) averted (Vassall and others 2014). would generate net savings, with predicted costs per Table 7.2 Costs of Adult Male Circumcision in Six Sub-Saharan African Countries Number of Average costs, Country Period Number of facilities circumcisions per facility 2012 US$ PPP All 99 750 (average) $49 Kenya March 2010 29 743 $38 Namibia April–May 2006 8 35 $31 South Africa April 2008–March 2009 9 3,828 $22 Tanzania 2010–11 18 1,914 $70 Uganda June–July 2009 26 286 $30 Zambia 2010 9 308 $61 Source: Bollinger and others 2014. Note: PPP = purchasing power parity. 148 Major Infectious Diseases infection averted over the period 2011–25 varying from Expenditure Analyses US$442 in Lesotho to US$4,096 in Rwanda. The unit costs assumed in models and estimated from The cost-effectiveness of oral PrEP in models is much programs are substantially different from the resources less clear because it depends on assumptions made expended on programs. about HIV incidence, costs of the program, and coverage Analyses of expenditure indicate what programs are of the PrEP. In a systematic review of models of oral costing and what interventions are being prioritized by PrEP, the cost per infection averted in a generalized HIV policy makers. PEPFAR has made expenditures available epidemic varied from cost saving to US$39,900 (Gomez online, but expenditure by one donor does not describe and others 2013). the full expenditure on a program in a country. An esti- Other cost-effectiveness analyses are of dubious mated US$4.5 billion was spent on HIV prevention in validity because they depend upon assumed effective- low- and middle-income countries in 2012, with PEPFAR ness. Topical PrEP (since found ineffective) was esti- providing US$1.6 billion of this from its total expendi- mated to cost between US$18 and US$181 per DALY ture of US$4.5 billion (UNAIDS 2015). Figure 7.6 averted and between US$1,800 and US$2,700 per life shows the distribution of PEPFAR expenditures for HIV year saved, with the major differences being due to prevention in fiscal year 2013, excluding treatment as assumptions about costs. Subsequent analyses for teno- prevention. The largest fraction of spending was on fovir gel estimated a cost of less than US$300 per DALY PMTCT and HTC. averted (Terris-Prestholt and others 2014), assuming With the addition of country spending, we can focus effectiveness. Similarly, the incremental cost-effectiveness on one country with complete expenditure data. Such for an HIV vaccine, also assuming effectiveness, was data are rarely available, but have been compiled in US$43 per DALY averted (Moodley, Gray, and Bertram Kenya (figure 7.7) and show that the majority of domes- 2016). A study of female condom program modeling tic HIV prevention resources (which excludes treatment found a low of US$107 per DALY averted in Zimbabwe as prevention) are deployed for HTC and PMTCT, and and a high of US$303 per DALY averted in Mozambique, the proportion allocated to HTC and PMTCT was even assuming that condoms would be used and would be greater than the proportion of all PEPFAR prevention effective (Mvundura and others 2015). expenditures in 19 countries on PMTCT and HTC. Figure 7.6 PEPFAR Expenditure on HIV Prevention across Selected Countries, by Category of Prevention, 2014 400 344.1 350 300 288.5 US$ (millions) 250 195.3 200 167.0 150 100 56.4 55.8 50 39.5 23.4 16.1 11.2 5.6 0 tio x, tio , lax e CS er M er PW er y ula px C C rap n- T hy ur et ula k p M HT TC x, th x, oth x, oth the atio n ns op isk op os af is W SM ID y VM kp do op ris PM ds pr texp lp rr kp d kp d ted ic y p er ris an ris an ris l an sis ed ra he oo ke th s as M al al ne ot Bl Po a er d o xu xu xu ge nd oth al an Se Se Se a al xu xu Se Se Source: PEPFAR (President’s Emergency Plan for AIDS Relief) 2014, Data Dashboard, http://www.pepfar.gov/funding/c63793.htm, accessed March 31, 2015. Note: CSW = commercial sex workers; HIV = human immunodeficiency virus; HTC = HIV testing and counseling; MSM = men who have sex with men; PEPFAR = President’s Emergency Plan for AIDS Relief; PMTCT = prevention of mother-to-child transmission; PWID = persons who inject drugs; px = prevention; VMMC = voluntary medical male circumcision. Includes 17 countries in Sub-Saharan Africa, Haiti, and Vietnam. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 149 Figure 7.7 Expenditure on HIV Prevention for Kenya, by Category of Prevention, 2012 60 53.8 50 US$ (millions) 40 35.4 30 20.9 20 15.8 14.2 8.7 8.3 8.2 10 1.9 1.9 1.7 1.7 1.3 1.1 0 n ion e e ion ID C C y SM P tin l d SW T tio lat bl lac rke cia et PE te HT M TC PW pu era at s iza af ga g kp VM M ma m so PM nic ds po uln or re bil mu W gg oo mo do rv isa om Bl n fo ity Co td sc RR un no as mm M Px Co Other international USG Domestic Sources: PEPFAR (President’s Emergency Plan for AIDS Relief) 2014, Data Dashboard, http://www.pepfar.gov/funding/c63793.htm, accessed March 31, 2015; Kenya National AIDS Spending Assessment 2014, http://files.unaids.org/en/media/unaids/contentassets/documents/data-and-analysis/tools/nasa/20141017/kenya_2011_en.pdf, last accessed October 19, 2016. Note: HIV = human immunodeficiency virus; HTC = HIV testing and counseling; MSM = men who have sex with men; PEP = postexposure prophylaxis; PMTCT = prevention of mother-to-child transmission; PWID = persons who inject drugs; Px = prevention; RR = risk reduction; SW = sex workers; USG = United States government; VMMC = voluntary medical male circumcision. Because of the limited data on effectiveness for inter- The more that prevention interventions can focus on ventions other than PMTCT, VMMC, and oral PrEP persons who would otherwise acquire or spread infec- (which has only recently been recommended), this tion, the more cost-effective they can be: targeting expenditure may be appropriate. It does illustrate that should increase cost-effectiveness. However, effective- few funds are being spent on the prevention interven- ness is not the same as impact. As interventions target tions with a weak evidence base or even on those that progressively fewer, higher-risk individuals, they may are likely efficacious in preventing transmission and become more cost-effective but have less impact. Taking that the largest proportion of prevention funding is into account the full cost of developing and implement- spent on HTC as an entry to treatment. ing programs, this lack of impact could lower the attrac- tiveness of investments in programs to develop and use new prevention products. TARGETING HIV PREVENTION This sequence can be illustrated by exploring what The relationship between expenditure and health ben- would be required to avert 20 percent of HIV infections efit is straightforward for treatment interventions. In using approximately 20 percent of the HIV prevention comparison, the cost-effectiveness of interventions for budget (figure 7.8). Assuming that there are 1.5 million prevention depends on the risk of acquiring infection new infections each year, US$4.5 billion is spent on pre- and disease. This risk can vary across individuals and vention (UNAIDS 2015), the goal explored is to reduce populations and over time, making a single measure of new infections by 300,000, and the budget is US$0.9 cost-effectiveness nonsensical. Simply viewed, preven- billion, it is possible to explore, for different effectiveness, tion interventions will be more cost-effective the higher the relationship between incidence in the target popula- incidence would otherwise be. However, changes in tion and the number of people who would have to be individual- and population-level risk as a function of covered. To achieve a given impact goal with a budget, coverage and intensity greatly complicate the relation- there is a trade-off between how well the intervention ship between costs and benefits, especially with the can be focused and the resources available per person cost-effectiveness of one intervention depending on the reached; the higher the incidence, the more could be success of other interventions. This relationship has spent per person in the program. Figure 7.8a shows how been challenging in modeling the impact of prevention, many people at a given HIV incidence would need to be where if coverage of treatment as prevention is assumed reached to avert 300,000 infections; figure 7.8b shows to be high and effective, other prevention interventions the cost per person reached with prevention allowable, have a smaller role to play (Stover and others 2014). if the cost per infection averted is to be US$3,000. 150 Major Infectious Diseases Figure 7.8 Isoclines Showing the Values Required to Achieve Target Reductions in HIV Infections and Costs of Infection Averted a. Infections averted = 300,000 per year b. Cost per infection averted = US$3,000 10 10 50% effectiveness 100% 50% effectiveness Mean background incidence rate (%) Mean background incidence rate (%) 100% 8 8 Incidence = 5% 6 Number of users = 7.5M 6 Incidence = 5% Effectiveness = 80% Cost = US$120 ppy Effectiveness = 80% 4 Incidence = 1% 4 Number of users = 38M Effectiveness = 80% 2 2 Incidence = 1% Cost = US$25 ppy Effectiveness = 80% 0 0 0 10 20 30 40 0 50 100 150 200 250 300 Number of users (M) Cost (US$ per person year) Note: HIV = human immunodeficiency virus; M = millions; ppy = per person year. In panel a, to avert 300,000 infections, the lines show effectiveness of 50, 60, 80, and 100 percent for direct mechanisms of prevention; the number of users in millions; and a given incidence of infection. In panel b, to achieve a cost per infection averted of US$3,000, the lines show effectiveness of 50, 60, 80, and 100 percent for direct mechanisms of prevention; the required cost per person year; and a given incidence of infection. Better targeting of HIV prevention to persons with a targeted, HIV prevention will, at most thresholds, be high incidence of acquisition or transmission makes the cost-effective, but that cost-effectiveness will depend on intervention more cost-effective, but this invites the the other interventions in use and the ability to target questions of how to target and what impact is possible. interventions appropriately. What resources should be used for HIV prevention and what should they be used for? Two prevention CONCLUSIONS products—VMMC and oral PrEP—conservatively Prevention has probably averted many millions of HIV should reduce HIV incidence by 50 percent; using infections, but it is impossible to be sure how many, condoms and having fewer partners could have similar given the difficulties of knowing what the scale of spread effectiveness. However, products are not interven- would have been in the absence of behavior changes tions. Interventions need to get people to use the prod- among those at risk. Despite the scale-up of effective ucts, and resources need to be directed to this effort. treatment, which can contribute to HIV prevention, An HIV infection costs either decades of lost life or other prevention interventions are needed, but which decades of expense on treatment. If an infection causes 20 ones? HIV prevention should be a cost-effective inter- DALYs and averting 1 DALY is “worth” US$500, then an vention, but 35 years into the global HIV pandemic, infection prevented is worth US$10,000. Alternatively, if large questions remain about which prevention pro- treating someone for 20 years would cost US$500 per grams are effective, how best to implement them, and year, then treating an infection would cost US$10,000. So how much should be spent on them. A fundamental if it is possible to prevent half of the infections with cur- problem is identifying those who are at risk and then rent products and HIV incidence is 10 percent, we should ensuring that they adopt preventive behavior. be spending US$500 per person per year on prevention. Large gaps are evident in the supply of prevention If HIV incidence is 1 percent, we should be spend- interventions. Furthermore, data on effectiveness are ing US$50 per person per year. If HIV incidence is available for only a few direct mechanisms. Evidence of 0.1 percent, we should be spending US$5 per person per effectiveness exists for VMMC and oral PrEP, and it is year. Alternatively, if there are 2 million HIV infections logical that using condoms and having fewer partners per year, we should be spending at least US$5 billion per will reduce risk. Costing data for HIV prevention, except year. By any account, the world is falling well short of for VMMC, are also unavailable. When appropriately providing what is needed for HIV prevention. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 151 NOTE Male Circumcision Using Primary Source Data from Sub- Saharan Africa.” PLoS One 9 (5): e84701. World Bank Income Classifications as of July 2014 are as Bongaarts, J., P. Reining, P. Way, and F. Conant. 1989. “The follows, based on estimates of gross national income (GNI) Relationship between Male Circumcision and HIV Infection per capita for 2013: in African Populations.” AIDS 3 (6): 373–77. Brandt, A. M. 1987. No Magic Bullet: A Social History of • Low-income countries (LICs) = US$1,045 or less Venereal Disease in the United States since 1880. Oxford, • Middle-income countries (MICs) are subdivided: U.K.: Oxford University Press. (a) lower-middle-income = US$1,046 to US$4,125 Cameron, W., J. N. Simonsen, L. J. D’Costa, A. R. Ronald, G. M. (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 Maitha, and others. 1989. “Female to Male Transmission • High-income countries (HICs) = US$12,746 or more. of Human Immunodeficiency Virus Type 1: Risk Factors for Seroconversion in Men.” The Lancet 334 (8660): 403–7. Choopanya, K., M. Martin, P. Suntharasamai, U. Sangkum, REFERENCES P. A. Mock, and others. 2013. “Antiretroviral Prophylaxis Abdool Karim, Q., S. S. Abdool Karim, J. A. Frohlich, for HIV Infection in Injecting Drug Users in Bangkok, A. C. Grobler, C. Baxter, and others. 2010. “Effectiveness Thailand (the Bangkok Tenofovir Study): A Randomised, and Safety of Tenofovir Gel, an Antiretroviral Microbicide, Double-Blind, Placebo-Controlled Phase 3 Trial.” The for the Prevention of HIV Infection in Women.” Science Lancet 381 (9983): 2083–90. 329 (5996): 1168–74. Cohen, M. S., Y. Q. Chen, M. McCauley, T. Gamble, M. C. Anderson, R. M., and R. M. May. 1990. Infectious Diseases of Hosseinipour, and others. 2011. “Prevention of HIV-1 Humans: Dynamics and Control. Oxford, U.K.: Oxford Infection with Early Antiretroviral Therapy.” New England University Press. Journal of Medicine 365 (6): 493–505. Anderson, R. M., R. M. May, M.-C. Boily, G. P. Garnett, and Cremin, I., R. Alsallaq, M. Dybul, P. Piot, G. Garnett, and others. J. T. Rowley. 1991. “The Spread of HIV-1 in Africa: Sexual 2013. “The New Role of Antiretrovirals in Combination Contact Patterns and the Predicted Demographic Impact of HIV Prevention: A Mathematical Modelling Analysis.” AIDS.” Nature 352 (6336): 581–89. AIDS 27 (3): 447–58. Auvert, B., D. Taljaard, E. Lagarde, J. Sobngwi-Tambekou, Davis, K. R., and S. C. Weller. 1999. “The Effectiveness of R. Sitta, and others. 2005. “Controlled Intervention Trial of Condoms in Reducing Heterosexual Transmission of HIV.” Male Circumcision for Reduction of HIV Infection Risk: Family Planning Perspectives 31 (6): 272–79. The ANRS 1265 Trial.” PLoS Medicine 2 (11): e298. Eron, J. J., Jr., and M. S. Hirsch. 2008. “Antiviral Therapy of Avenir Health. 2016. “Unit Cost Repository.” Avenir Health, Human Immunodeficiency Virus Infection.” In Sexually London. http://www.avenirhealth.org/policytools/UC/. Transmitted Diseases, 4th ed., edited by K. Holmes, Baeten, J. M., D. Donnell, P. Ndase, M. R. Mugo, J. D. Campbell, P. F. Spirling, W. E. Stamm, P. Pilot, J. N. Wasserheit, and and others. 2012. “Antiretroviral Prophylaxis for HIV others. New York: McGraw-Hill. Prevention in Heterosexual Men and Women.” New England Ford, N., M. Vitoria, G. Hirnschall, and M. Doherty. 2013. Journal of Medicine 367 (5): 399–410. “Getting to Zero HIV Deaths: Progress, Challenges, and Baeten, J. M., T. Palanee-Phillips, E. R. Brown, K. Schwartz, Ways Forward.” Journal of the International AIDS Society L. E. Soto-Torres, and others. 2016. “Use of a Vaginal Ring 16 (1): 18927. Containing Dapivirine for HIV-1 Prevention in Women.” Fuller, C. M., C. Ford, and A. Rudolph. 2009. “Injection New England Journal of Medicine February 22. doi:10.1056 Drug Use and HIV: Past and Future Considerations for /NEJMoa1506110. HIV Prevention and Interventions.” In HIV Prevention: Bailey, R. C., S. Moses, C. B. Parker, K. Agot, I. Maclean, and A Comprehensive Approach, edited by K. H. Meyer and others. 2007. “Male Circumcision for HIV Prevention in H. F. Pizer. Amsterdam: Academic Press. Young Men in Kisumu, Kenya: A Randomised Controlled Galárraga, O., M. A. Colchero, R. G. Wamai, and S. M. Bertozzi. Trial.” The Lancet 369 (9562): 643–56. 2009. “HIV Prevention Cost-Effectiveness: A Systematic Baird, S. J., R. S. Garfein, C. T. McIntosh, and B. Özler. 2012. Review.” BMC Public Health 9 (Suppl 1): S5. “Effect of a Cash Transfer Programme for Schooling on Gaolathe, T., K. Wirth, M. Pretorius Holme, J. Makhema, Prevalence of HIV and Herpes Simplex Type 2 in Malawi: S. Moyo, and others. 2016. “Botswana Is Close to Meeting A Cluster Randomised Trial.” The Lancet 379 (9823): UNAIDS 2020 Goals of 90-90-90 Coverage.” Paper pre- 1320–29. sented at the Conference on Retroviruses and Opportunistic Bautista-Arrendondo, S., S. G. Sosa-Rubí, M. Opuni, A. Kwan, Infections, Boston, MA, February 22–25. C. Chaumont, and others. 2015. “Efficiency of VMMC Gardner, E. M., M. P. McLees, J. F. Steiner, C. Del Rio, and and Its Determinants: Results from the ORPHEA Study W. J. Burman. 2011. “The Spectrum of Engagement in in Kenya, Rwanda, South Africa, and Zambia.” Paper HIV Care and Its Relevance to Test-and-Treat Strategies for presented at the International AIDS Society Conference, Prevention of HIV Infection.” Clinical Infectious Diseases Vancouver, July 19–22. 52 (6): 793–800. Bollinger, L., A. Adesina, S. Forsythe, R. Godbole, E. Reuben, Garnett, G. P. 2005. “The Role of Herd Immunity in and others. 2014. “Cost Drivers for Voluntary Medical Determining the Impact of Sexually Transmitted Disease 152 Major Infectious Diseases Vaccines.” Journal of Infectious Diseases 191 (Suppl 1): Hecht, R., J. Stover, L. Bollinger, F. Muhib, K. Case, and others. S97–106. 2010. “Financing of HIV/AIDS Programme Scale-up in Garnett, G. P., J. M. Garcia-Calleja, T. Rehle, and S. A. J. Gregson. Low-Income and Middle-Income Countries, 2009–31.” The 2006. “Behavioural Data as an Adjunct to HIV Surveillance Lancet 376 (9748): 1254–60. Data.” Sexually Transmitted Infections 82 (Suppl. 1): i57–i62. Holmes, C. B., T. B. Hallett, R. P. Walensky, T. Bärnighausen, Garnett, G. P., T. B. Hallett, A. Takaraza, J. Hargreaves, R. Rhead, Y. Pillay, and M. S. Cohen. 2017. “Effectiveness and and others. 2016. “Providing a Conceptual Framework Cost-Effectiveness of Treatment as Prevention for HIV.” for HIV Prevention Cascades and Assessing Feasibility of In Disease Control Priorities (third edition): Volume Empirical Measurement with Data from East Zimbabwe: 6, Major Infectious Diseases, edited by K. K. Holmes, A Case Study.” The Lancet HIV 3 (7): e297–e306. S. Bertozzi, B. R. Bloom, and P. Jha. Washington, DC: Gertler, P. J., S. Martinez, P. Premand, L. B. Rawlings, and World Bank. C. M. Vermeersch. 2011. Impact Evaluation in Practice. Hoots, W. K. 2001. “History of Plasma-Product Safety.” Washington, DC: World Bank. Transfusion Medical Reviews 2 (Suppl 1): 3–10. Gomez, G. B., A. Borquez, K. K. Case, A. Wheelock, A. Vassall, HPTN (HIV Prevention Trials Network). 2015. “Cash Transfers and others. 2013. “The Cost and Impact of Scaling Up Pre- Conditional on Schooling Do Not Prevent HIV Infection Exposure Prophylaxis for HIV Prevention: A Systematic among Young South African Women.” http://www.hptn Review of Cost-Effectiveness Modelling Studies.” PLoS .org/network_information/hptnnews/15/068results_IAS Medicine 10 (3): e1001401. .html. Grant, R. M., J. R. Lama, P. L. Anderson, V. McMahan, A. Y. Liu, IPM (International Partnership for Microbicides). 2016. “The and others. 2010. “Preexposure Chemoprophylaxis for HIV Ring Study 2016.” IPM, Silver Spring, MD. http://www Prevention in Men Who Have Sex with Men.” New England .ipmglobal.org/the-ring-study. Journal of Medicine 363 (27): 2587–99. Jewell, B. L., I. Cremin, M. Pickles, C. Celum, J. M. Baeten, Gray, R. H., G. Kigozi, D. Serwadda, F. Makumbi, S. Watya, and others. 2015. “Estimating the Cost-Effectiveness of and others. 2007. “Male Circumcision for HIV Prevention Pre-Exposure Prophylaxis to Reduce HIV-1 and HSV-2 in Men in Rakai, Uganda: A Randomised Trial.” The Lancet Incidence in HIV-Serodiscordant Couples in South Africa.” 369 (9562): 657–66. PLoS One 10 (1): e0115511. Gregson, S., S. Adamson, S. Papaya, T. Chimbadzwa, J. Mundondo, Kamali, A., M. Quigley, J. Nakiyingi, J. Kinsman, J. Kengeya- and others. 2007. “Impact and Process Evaluation of Integrated Kayondo, and others. 2003. “Syndromic Management of Community and Clinic-Based HIV-1 Control in Eastern Sexually-Transmitted Infections and Behaviour Change Zimbabwe.” PLoS Medicine 4 (3): e102. Interventions on Transmission of HIV-1 in Rural Uganda: Grosskurth, H., J. Todd, E. Mwijarubi, P. Mayaud, A. Nicoll, and A Community Randomised Trial.” The Lancet 361 (9358): others. 1995. “Impact of Improved Treatment of Sexually 645–52. Transmitted Diseases on HIV Infection in Rural Tanzania: Krishnaratne, S., B. Hensen, J. Cordes, J. Enstone, and Randomised Controlled Trial.” The Lancet 346 (8974): 530–36. J. R. Hargreaves. 2016. “Interventions to Strengthen the Hallett, T. B., J. Aberle-Grasse, G. Bello, L. M. Boulos, M. P. A. HIV Prevention Cascade: A Systematic Review of Reviews.” Cayemittes, and others. 2006. “Declines in HIV Prevalence The Lancet HIV 3 (7): e307–e317. Can Be Associated with Changing Sexual Behaviour in Laga, M., D. Rugg, G. Peersman, and M. Ainsworth. 2012. Uganda, Urban Kenya, Zimbabwe, and Urban Haiti.” “Evaluating HIV Prevention Effectiveness: The Perfect as Sexually Transmitted Infections 82 (Suppl 1): i1–8. the Enemy of the Good.” AIDS 26 (7): 779–83. Hallett, T. B., S. Dube, I. Cremin, B. Lopman, A. Mahomva, and Lagakos, S., and A. R. Gable, eds. 2008. Methodological Challenges others. 2009. “Testing and Counselling for HIV Prevention in Biomedical HIV Prevention Trials. Washington, DC: and Care in the Era of Scaling-up Antiretroviral Therapy.” National Academies Press. Epidemics 1 (2): 77–82. Lépine, A., S. Chandrashekar, G. Shetty, P. Vickerman, J. Bradley, Hallett, T. B., P. J. White, and G. P. Garnett. 2007. “The and others. 2016. “What Determines HIV Prevention Costs Appropriate Evaluation of HIV Prevention Interventions: at Scale? Evidence from the Avahan Programme in India.” From Experiment to Full-Scale Intervention.” Sexually Health Economics 25 (Suppl 1): 67–82. Transmitted Infections 83 (Suppl 1): i55–60. Mahendra, V. S., R. Mudoi, A. Oinam, V. Pakkela, A. Sarna, and Hankins, C. A., and B. O. de Zalduondo. 2010. “Combination others. 2007. Continuum of Care for HIV-Positive Women Prevention: A Deeper Understanding of Effective HIV Accessing Programs to Prevent Parent-to-Child Transmission: Prevention.” AIDS 24 (Suppl 4): S70–80. Findings from India. Horizons Final Report. Washington, DC: Hargreaves, J., S. Delaney-Moretlwe, T. B. Hallett, S. Johnson, Population Council. S. Kapiga, and others. 2016. “The HIV Prevention Cascade: Marrazzo, J. M., and W. Cates. 2011. “Interventions to Prevent Integrating Theories of Epidemiological, Behavioural and Sexually Transmitted Infections, Including HIV Infection.” Social Science into Programme Design and Monitoring.” Clinical Infectious Diseases 53 (Suppl 3): S64–78. The Lancet HIV 3 (7): e318–e22. Marrazzo, J. M., G. Ramjee, B. A. Richardson, K. Gomez, Harris, K., T. Guthrie, R. Omban, and G. Garnett. Forthcoming. N. Mgodi, and others. 2015. “Tenofovir-Based Preexposure “Follow the Money: Assessing the Efficiency of HIV Prophylaxis for HIV Infection among African Women.” Prevention Expenditures.” Submitted to The Lancet HIV. New England Journal of Medicine 372 (6): 509–18. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 153 Marseille, E., L. Dandona, N. Marshall, P. Gaist, S. Bautista- Sexual Acquisition of HIV Infection in Women: Systematic Arredondo, and others. 2007. “HIV Prevention Costs and Review and Meta-Analysis.” BMC Infectious Diseases Program Scale: Data from the PANCEA Project in Five Low 12: 289. and Middle-Income Countries.” BioMed Central Health Oppenheimer, G. M. 1988. “In the Eye of the Storm: The Services Research 7: 108. Epidemiological Construction of AIDS.” In AIDS, the Marseille, E., L. Dandona, J. Saba, C. McConnel, B. Rollins, and Burdens of History, edited by E. Fee and D. M. Fox. Berkeley: others. 2004. “Assessing the Efficiency of HIV Prevention University of California Press. around the World: Methods of the PANCEA Project.” Padian, N. S., S. I. McCoy, J. E. Balkus, and J. N. Wasserheit. Health Services Research 39 (6, pt 2): 1993–2012. 2010. “Weighing the Gold in the Gold Standard: Mavedzenge, S. N., E. Luecke, and D. A. Ross. 2014. “Effective Challenges in HIV Prevention Research.” AIDS 24 (5): Approaches for Programming to Reduce Adolescent 621–35. Vulnerability to HIV Infection, HIV Risk, and HIV-Related Palella, F. J., Jr., K. M. Delaney, A. C. Moorman, M. O. Loveless, Morbidity and Mortality: A Systematic Review of Systematic J. Fuhrer, and others. 1998. “Declining Morbidity and Mortality Reviews.” Journal of Acquired Immune Deficiency Syndromes among Patients with Advanced Human Immunodeficiency 66 (Suppl 2): S154–69. Virus Infection.” New England Journal of Medicine 338 (13): May, R. M., and R. M. Anderson. 1987. “Transmission Dynamics 853–60. of HIV Infection.” Nature 326 (6109): 137–42. Pinkerton, S. D., and P. R. Abramson. 1997. “Effectiveness of McCormack, S., D. T. Dunn, M. Desai, D. I. Dolling, M. Gafos, Condoms in Preventing HIV Transmission.” Social Science and others. 2016. “Pre-Exposure Prophylaxis to Prevent the and Medicine 44 (9): 1303–12. Acquisition of HIV-1 Infection (PROUD): Effectiveness Piot, P., S. S. Abdool Karim, R. Hecht, H. Legido-Quigley, Results from the Pilot Phase of a Pragmatic Open-Label K. Buse, and others. 2015. “Defeating AIDS: Advancing Randomised Trial.” The Lancet 387 (10013): 53–60. Global Health.” The Lancet 386 (9989): 171–218. Merson, M., N. Padian, T. J. Coates, G. R. Gupta, S. M. Bertozzi, Rees, H., S. A. Delany-Moretlwe, C. Lombard, D. Baron, and others. 2008. “Combination HIV Prevention.” The Lancet R. Panchia, and others. 2015. “FACTS 001 Phase III Trial of 372 (9652): 1805–6. Pericoital Tenofovir 1% Gel for HIV Prevention in Women.” Millett, G. A., S. A. Flores, G. Marks, J. B. Reed, and J. H. Herbst. Paper presented at the Conference on Retroviruses and 2008. “Circumcision Status and Risk of HIV and Sexually Opportunistic Infections (CROI), Seattle, WA, February Transmitted Infections among Men Who Have Sex with 23–26. Men: A Meta-Analysis.” Journal of the American Medical Rerks-Ngarm, S., P. Pitisuttithum, S. Nitayaphan, J. Kaewkungwal, Association 300 (14): 1674–84. J. Chiu, and others. 2009. “Vaccination with ALVAC and Minnis, A. M., and N. S. Padian. 2005. “Effectiveness of AIDSVAX to Prevent HIV-1 Infection in Thailand.” New Female Controlled Barrier Methods in Preventing Sexually England Journal of Medicine 361 (23): 2209–20. Transmitted Infections and HIV: Current Evidence and Røttingen, J. A., D. W. Cameron, and G. P. Garnett. 2001. Future Research Directions.” Sexually Transmitted Infections “A Systematic Review of the Epidemiological Interactions 81 (3): 193–200. between Classical STDs and HIV.” Sexually Transmitted Molina, J. M., C. Capitant, B. Spire, G. Pialoux, L. Cotte, and Diseases 28 (10): 579–97. others. 2015. “On-Demand Preexposure Prophylaxis in Schwartländer, B., J. Stover, T. Hallett, R. Atun, C. Avila, and Men at High Risk for HIV-1 Infection.” New England others. 2011. “Towards an Improved Investment Approach Journal of Medicine 373 (23): 2237–46. for an Effective Response to HIV/AIDS.” The Lancet 377 Moodley, N., G. Gray, and M. Bertram. 2016. “The Case (9782): 2031–41. for Adolescent HIV Vaccination in South Africa: A Cost- Sosa-Rubí, S. G., S. Bautista-Arredondo, M. Opuni, D. Contreras- Effectiveness Analysis.” Medicine 95 (4): e2528. Loya, I. Moreno, and others. 2015. “Efficiency of Facility- Mvundura, M., N. Nundy, M. Kilbourne-Brook, and P. S. Coffey. Based HTC and Its Determinants: Results from the ORPHEA 2015. “Estimating the Hypothetical Dual Health Impact and Study in Kenya, Rwanda, South Africa, and Zambia.” Cost-Effectiveness of the Woman’s Condom in Selected Sub- Paper presented at the eighth International AIDS Society Saharan African Countries.” International Journal of Women’s conference, Vancouver, BC, July 19–22. Health 7: 271–77. Steiner, M., L. Warner, K. M. Stone, and W. Cates Jr. 2008. Nelson, K. E., D. D. Celentano, S. Eiumtrakol, D. R. Hoover, “Condoms and Other Barrier Methods for Prevention C. Beyrer, and others. 1996. “Changes in Sexual Behavior and of STD/HIV Infection and Pregnancy.” In Sexually a Decline in HIV Infection among Young Men in Thailand.” Transmitted Diseases, 4th ed., edited by K. K. Holmes, New England Journal of Medicine 335 (5): 297–303. P. F. Spirling, W. E. Stamm, P. Pilot, J. N. Wasserheit, and Njeuhmeli, E., S. Forsythe, J. Reed, M. Opuni, L. Bollinger, others. New York: McGraw-Hill. and others. 2011. “Voluntary Medical Male Circumcision: Stoneburner, R. L., and D. Low-Beer. 2004. “Population-Level Modeling the Impact and Cost of Expanding Male HIV Declines and Behavioral Risk Avoidance in Uganda.” Circumcision for HIV Prevention in Eastern and Southern Science 304 (5671): 714–18. Africa.” PLoS Medicine 8 (11): e1001132. Stover, J., T. B. Hallett, Z. Wu, M. Warren, C. Gopalappa, Obiero, J., P. G. Mwthera, G. D. Hussey, and C. S. Wiysonge. and others. 2014. “How Can We Get Close to Zero? The 2012. “Vaginal Microbicides for Reducing the Risk of Potential Contribution of Biomedical Prevention and the 154 Major Infectious Diseases Investment Framework towards an Effective Response to Walker, D. 2003. “Cost and Cost-Effectiveness of HIV/AIDS HIV.” PLoS One 9 (11): e111956. Prevention Strategies in Developing Countries: Is There Terris-Prestholt, F., A. M. Foss, A. P. Cox, L. Heise, G. Meyer- an Evidence Base?” Health Policy and Planning 18 (1): Rath, and others. 2014. “Cost-Effectiveness of Tenofovir 4–17. Gel in Urban South Africa: Model Projections of HIV Wawer, M. J., F. Makumbi, G. Kigozi, D. Serwadda, S. Watya, Impact and Threshold Product Prices.” BMC Infectious and others. 2009. “Circumcision in HIV-Infected Men Diseases 14: 14. and Its Effect on HIV Transmission to Female Partners in Thigpen, M. C., P. M. Kebaabetswe, L. A. Paxton, D. K. Smith, Rakai, Uganda: A Randomised Controlled Trial.” The Lancet C. E. Rose, and others. 2012. “Antiretroviral Preexposure 374 (9685): 229–37 Prophylaxis for Heterosexual HIV Transmission in Botswana.” Wawer, M. J., N. K. Sewankambo, D. Serwadda, T. C. Quinn, New England Journal of Medicine 367 (5): 423–34. L. A. Paxton, and others. 1999. “Control of Sexually Tobian, A. A., S. Kacker, and T. C. Quinn. 2014. “Male Circumcision: Transmitted Diseases for AIDS Prevention in Uganda: A A Globally Relevant but Under-Utilized Method for the Randomised Community Trial. Rakai Project Study Group.” Prevention of HIV and Other Sexually Transmitted Infections.” The Lancet 353 (9152): 525–35. Annual Review of Medicine 65: 293–306. Weiss, H. A., M. A. Quigley, and R. J. Hayes. 2000. “Male UNAIDS (Joint United Nations Programme on HIV/AIDS). Circumcision and Risk of HIV Infection in Sub-Saharan 2015. “AIDS by the Numbers.” UNAIDS, Geneva. Africa: A Systematic Review and Meta-Analysis.” AIDS Uthman, O. A., T. A. Popoola, M. M. Uthman, and O. Aremu. 14 (15): 2361–70. 2010. “Economic Evaluations of Adult Male Circumcision Weller, S. C. 1993. “A Meta-Analysis of Condom Effectiveness for Prevention of Heterosexual Acquisition of HIV in Men in Reducing Sexually Transmitted HIV.” Social Science and in Sub-Saharan Africa: A Systematic Review.” PLoS One Medicine 36 (12): 1635–44. 5 (3): e9628. Weller, S. C., and K. Davis. 2002. “Condom Effectiveness Van Damme, L., A. Corneli, K. Ahmed, K. Agot, J. Lombaard, in Reducing Heterosexual HIV Transmission.” Cochrane and others. 2012. “Preexposure Prophylaxis for HIV Database of Systematic Reviews 1: CD003255. Infection among African Women.” New England Journal of WHO (World Health Organization). 2015a. “Consolidated Medicine 367 (5): 411–22. Strategic Information Guidelines for HIV in the Health Vassall, A., M. Pickles, S. Chandrashekar, M. C. Boily, G. Shetty, Sector.” WHO, Geneva. and others. 2014. “Cost-Effectiveness of HIV Prevention ———. 2015b. “Guideline on When to Start Antiretroviral for High-Risk Groups at Scale: An Economic Evaluation of Therapy and on Pre-Exposure Prophylaxis for HIV.” WHO, the Avahan Programme in South India.” The Lancet Global Geneva. Health 2 (9): e531–40. ———. 2015c. “Voluntary Medical Male Circumcision for HIV Vermund, S. H., S. J. Fidler, H. Ayles, N. Beyers, and R. J. Hayes. Prevention in 14 Priority Countries in East and Southern 2013. “Can Combination Prevention Strategies Reduce HIV Africa.” Progress Brief, WHO, Geneva. Transmission in Generalized Epidemic Settings in Africa? Wiysonge, C. S., E. J. Kongnyuy, M. Shey, A. S. Muula, O. B. Navti, The HPTN 071 (PopART) Study Plan in South Africa and and others. 2011. “Male Circumcision for Prevention of Zambia.” Journal of Acquired Immune Deficiency Syndrome Homosexual Acquisition of HIV in Men.” Cochrane Database 63 (Suppl 2): S221–27. of Systematic Reviews 6: CD007496. Cost-Effectiveness of Interventions to Prevent HIV Acquisition 155 Chapter 8 Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics David Wilson and Jessica Taaffe INTRODUCTION software, such as Optima, can help identify the right mix of interventions and financial allocations, resulting in In the 30 years since the global appearance of the human cost-effective programming.1 immunodeficiency virus/acquired immune deficiency This chapter reviews factors to consider when tailor- syndrome (HIV/AIDS), we have come extraordinarily ing the local response to a local epidemic. It explores far in our fight against it. We know its physical routes of transmission dynamics in concentrated, generalized, and transmission and how to prevent infection at the indi- mixed epidemics, while acknowledging the complexity vidual level, and we have proven intervention strategies and variation within these distinctions. It also addresses at the population level. Yet, despite global declines in interventions shown to work in different contexts and modeled incidence, measured incidence remains high discusses how program science can assist in reaching the in cohort studies and is growing within certain demo- right people in the right places, providing examples of graphic groups. This concerning trend underscores the successful implementation. Finally, it elaborates on what need to thoroughly examine how local and national Optima software is and how it can improve allocative HIV/AIDS epidemics are sustained. and implementation efficiency. Knowledge of epidemic and transmission dynamics will help HIV/AIDS programs invest in the right inter- ventions. Proven interventions exist for both concen- KNOW YOUR EPIDEMIC: TRANSMISSION trated and generalized epidemics. For example, simple DYNAMICS targeted interventions for concentrated epidemics have been implemented at scale with considerable durability A thorough understanding of the heterogeneity of HIV in numerous contexts. Targeting resources and interven- transmission, between and within epidemics, is the first tions to the right people and the right places is also step in a targeted response. This understanding includes essential. More than one key population may drive a appreciating transmission dynamics at national and sub- concentrated epidemic, and new infections within a national levels, across geographic regions, and within generalized or mixed epidemic come from multiple different demographics, as well as the risk factors sus- demographic groups and behaviors within the general taining transmission. population. Geographic targeting and hotspot mapping Distinguishing between concentrated, generalized, can assist in identifying both geographic areas of elevated and mixed epidemics is helpful for understanding trans- transmission and high-risk groups. Epidemic modeling mission dynamics. Corresponding author: David Wilson, Global Lead, Decision and Delivery Science; Program Director, Global HIV/AIDS Program, Health, Nutrition, and Population, World Bank; dwilson@worldbank.org. 157 • An epidemic is concentrated if “HIV has spread rap- less than 1 percent in the general population (prevalence idly in one or more defined subpopulation but is not in pregnant women as a proxy) indicated a concentrated well established in the general population” (WHO epidemic, whereas prevalence of more than 1 percent in 2013). In this context, protecting key populations, the general population indicated a generalized epidemic such as injecting drug users (IDUs), men who have (UNAIDS/WHO Working Group on Global HIV/AIDS sex with men (MSM), and sex workers, would signifi- and STI Surveillance 2000). While this approach is a cantly mitigate the epidemic. simple numerical way of defining epidemics, it does not • An epidemic is generalized if transmission at a level to address what the transmission dynamics are, how these sustain the epidemic would occur even if key popu- epidemics are sustained, where the majority of new infec- lations were protected from infection, as when “HIV tions are occurring, and where target prevention activities infection is firmly established in the general popula- are in place. For example, Wilson and Challa (2009) tion” (WHO 2013). demonstrate that three African epidemics have HIV pre- • An epidemic is mixed if transmission would be valence of more than 1 percent in the general population, sustained by both key populations and the general but they are not all considered generalized because of the population and would continue even if transmission source of new infections. in either population were stopped. Mixed epidemics Most epidemics are concentrated (map 8.1). are essentially one or more concentrated epidemics However, the global burden of HIV/AIDS is heaviest in within a generalized epidemic. Sub-Saharan Africa, where 24.7 million people live with the disease and 1.5 million new infections occur each Previously, concentrated and generalized epidemics year (more than 70 percent of the global total for exist- were distinguished from one another based on HIV pre- ing and new infections) (UNAIDS 2014b). Most HIV/ valence in key populations or the general population. AIDS epidemics in Sub-Saharan Africa are generalized Prevalence of more than 5 percent in key populations and or mixed transmission (Wilson and Halperin 2008). Map 8.1 Global HIV Transmission Dynamics IBRD 42331 | DECEMBER 2016 Global Transmission Patterns Concentrated Generalized Mixed or uncertain Source: Wilson and Halperin 2008. Note: HIV = human immunodeficiency virus. 158 Major Infectious Diseases Concentrated Epidemics Figure 8.1 HIV Risk in Key Populations Relative to the General Population Concentrated HIV/AIDS epidemics are occurring pri- marily in Asia; Australasia; Europe; the Middle East; and North, Central, and South America. They are sustained through key populations, which are at much higher risk for infection (figure 8.1). A targeted approach acknowledges and addresses the factors and behaviors that are putting these key populations at Overall risk of HIV infection, 20- to 60-fold increased risk. higher relative to general population Sex Workers Female sex workers have a risk of HIV infection that is 13.5 times higher than that of the general population (Baral and others 2012). Globally, HIV prevalence is higher among sex workers than among the general population. It is 12 times higher in 110 countries with available data and at least 50-fold higher in 4 countries 19-fold higher (UNAIDS 2014b). In 16 Sub-Saharan African countries, 13.5-fold higher female sex workers have a pooled HIV prevalence of 37 percent (Kerrigan and others 2013); these countries Female sex Men who have Injecting workers sex with men drug users include Swaziland, where as many as 70 percent of sex workers are living with HIV/AIDS (figure 8.2). Sources: Baral and others 2012; UNAIDS 2014b. Biological, behavioral, and structural factors put sex Note: HIV = human immunodeficiency virus. workers at greater risk of contracting HIV and other sexually transmitted infections (STIs). Compared with the general population, sex workers have more sexual partners and more concurrent sexual partners, are more Figure 8.2 Prevalence of HIV among Sex Workers versus likely to use injection drugs (Johnston and Corceal 2013; General Population Lau and others 2008; Needle and others 2008; UNAIDS Afghanistan Czech Republic 2006; Zohrabyan and others 2013), and are more often Philippines Tunisia subject to violence (Decker and others 2014; Deering Bulgaria Lebanon and others 2013). It is common practice for police to Romania Armenia arrest individuals because they are carrying condoms Georgia Mexico and then to confiscate them, further compromising the Bolivia Indonesia ability of sex workers to protect themselves (Ireri 2012; Madagascar Tajikistan Maseko and Ndlovu 2012; Open Society Foundations Belarus Malaysia 2012; WHO 2012). Furthermore, discrimination and Chile Vietnam stigma discourage sex workers from seeking health Honduras Dominican Republic services, including HIV testing (King and others 2013; Papua New Guinea Senegal Scorgie and others 2011). Panama Cabo Verde The spectrum of sex work is especially complex in Suriname Guyana Burundi Sub-Saharan Africa, where sexual transactions often Estonia Guinea occur in the context of poverty and inequality. Sex work Sierra Leone Jamaica involves both those who identify themselves as profes- Guinea-Bissau Chad sionals and those who do not but who do engage in Nigeria Uganda informal sex work and transactional sex. In Burkina Swaziland 0 10 20 30 40 50 60 70 80 Faso, some nonprofessional sex workers (bar waitresses Prevalence and mobile traders) have higher HIV prevalence than Prevalence among sex workers as reported by country (2012) some professional sex workers (roamers), despite fewer Prevalence among people ages 15–49 (2011) reported sexual partners (2.6 and 3.3 clients per week for Source: UNAIDS (Joint United Nations Programme on HIV/AIDS). 2012. UNAIDS Report on the waitresses and traders, respectively, versus 18.6 for pro- Global AIDS Epidemic 2012. Geneva. fessional roamers) (Nagot and others 2002). Note: HIV = human immunodeficiency virus. Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 159 Men Who Have Sex with Men Injecting Drug Users MSM is another key population in concentrated epi- An estimated 13 percent of the 12.7 million IDUs world- demics. They are 19 times more likely to be living with wide are infected with HIV; IDUs have, on average, HIV/AIDS than the general population, with prevalence 28 times greater HIV prevalence than the general of 6 percent to 15 percent (UNAIDS 2014b). Epidemics population, ranging from 1.3 to 2,000 times greater in have been spreading among MSM in many settings, even 74 countries (UNAIDS 2014b). Outside of Sub-Saharan in high-income settings with good coverage of antiretro- Africa, an estimated 30 percent of new infections come viral treatment (ART) (Sullivan and others 2009). MSM from IDUs; in many countries in Asia, Eastern Europe, account for the largest share of new infections in Latin and the Middle East, IDUs are driving the national epi- America and the Caribbean (33 percent in the Dominican demic (Mathers and others 2008; UNAIDS 2014b). In Republic, 56 percent in Peru) (figure 8.3) and for a sub- the Islamic Republic of Iran, more than two-thirds of stantial percentage of new infections in other parts of the new infections occur in IDUs and their sexual partners world (Gouws and Cuchi 2012). In Ghana, 8 percent to (figure 8.4). 18 percent of new infections occur in MSM (UNAIDS IDU-driven epidemics exhibit much heterogeneity. and World Bank 2012). In 30 countries reporting gender-specific data, HIV Biological and behavioral factors multiply the risk prevalence is higher among female than among male among MSM at both individual and population levels. IDUs (UNAIDS 2014b). This enhanced risk may be due The estimated risk of acquiring HIV is 18 times greater to high rates of sex work among female IDUs (Blouin through anal than through vaginal sex (Grulich and and others 2013). IDUs under age 25 are also at compar- Zablotska 2010). Frequent casual sex with multiple atively high risk for HIV infection, with a prevalence of partners is relatively common among MSM. In larger 5.2 percent. networks, as the number of partners increases, so does Shared needles are a main risk factor for infection, the incidence of unprotected sex (Choi, Gibson, and although sexual transmission between IDUs and their others 2004; Choi, McFarland, and others 2004; Kelly non-IDU partners can play a small to major role and others 2010). In these situations, acute infections depending on context. Modeling of epidemics for the may be more prevalent, driving higher incidence, period 2010–15 showed that HIV infection due to because acutely infected individuals have higher viral unprotected sex among and with IDUs was 5 percent of loads and are more infectious. total HIV incidence in Nairobi, Kenya, and Karachi, Sexual and other risk factors also play a role. MSM may Pakistan, but 15 percent to 45 percent in Odessa, Ukraine have sex with both men and women (and may even be in (Strathdee and others 2010). long-term heterosexual marriages or partnerships, espe- cially in the global South). Drug use is also associated with riskier sexual practices. In Bangkok, HIV prevalence among Generalized Epidemics MSM rose from 2009 to 2011 with the use of amphetamine- Generalized epidemics were previously defined as epi- type stimulants (Colfax and others 2010; Freeman and demics in which more than 1 percent of the general others 2011; van Griensven and others 2009). population was infected. Recent definitions focus on transmission dynamics, and generalized epidemics are now defined as epidemics that are sustained by hetero- Figure 8.3 Sources of New HIV Infections in Peru sexual transmission in the general population. Medical injections, 0.23% Sex workers, 0.89% The distinction between concentrated and general- Partners of injecting Clients of sex workers, 1.33% ized epidemics is important: interventions in a concen- drug users, 0.22% Injecting drug users, 1.98% No-risk blood transfusions, 0.00% trated epidemic should address specific key populations Partners of individuals who have casual heterosexual sex, 5.54% (Wilson and Halperin 2008), whereas interventions in a Female partners of men generalized epidemic should address the broader, gen- who have sex with men, 6.22% Individuals who have casual eral population, while also focusing on high risk sub- heterosexual sex, 6.30% groups within it. As heterogeneous as community Men who have Partners norms and sexual values and practices can be in the sex with men, of clients of sex 54.97% workers, 6.36% general population, so, too, are the sources of infection. For example, in KwaZulu-Natal, South Africa’s most Low-risk HIV/AIDS-afflicted region, most infections are trans- heterosexuals, 15.97% mitted through heterosexual sex (figure 8.5). However, Source: UNAIDS and OPS 2009. key populations and their sexual partners also play a Note: HIV = human immunodeficiency virus. large role. 160 Major Infectious Diseases Casual sex is highly prevalent in countries experiencing Figure 8.4 Sources of New HIV Infections in the Islamic severe generalized epidemics. In Lesotho and Swaziland, Republic of Iran where HIV prevalence was 23 percent and 27 percent, Partners of injecting drug users, 12.15% respectively, in 2013 (UNAIDS 2014b), casual sex, as Injecting drug opposed to commercial sex, is highly prevalent among users, 55.95% Sex workers, 1.27% men who work as soldiers, police, miners, drivers, guards, Clients of sex workers, 4.17% and seasonal workers (Family Health International and Partners of clients of sex workers, 2.84% others 2002; Swaziland Ministry of Health and Social Men who have Welfare 2002). sex with men, 9.94% HIV prevalence also varies across gender and age groups. In Swaziland, more than half of females ages Female partners of men who have sex with men, 4.99% 30–34 and nearly half of men ages 35–39 are living with Individuals who have casual HIV/AIDS. For people ages 18–29, prevalence is at least heterosexual sex, 0.30% Partners of individuals who have twice as high in women as in men (Swaziland Ministry No-risk medical injections, casual heterosexual sex, 0.32% of Health 2012). blood transfusions, 0.00% Low-risk heterosexuals, 8.08% Source: Nasirian 2011. Note: HIV = human immunodeficiency virus. Mixed Epidemics Mixed epidemics are characterized by both concentrated Figure 8.5 Sources of New HIV Infections in KwaZulu-Natal, and generalized transmission, often in the same area, as South Africa in much of West Africa. Transmission can also change, Injecting Drug Use (IDU), 1% Partners IDU, 0% moving from concentrated to generalized. In mixed epi- Sex workers, 6% demics that are changing from concentrated to general- Clients, 9% ized transmission, bridge populations, such as paying Partners of clients, 3% and nonpaying partners of sex workers and bisexual MSM, transmit HIV to the general population. MSM, 3% Epidemics of mixed transmission can also be geo- Low-risk Female partners of MSM, 0% graphically mixed, with concentrated epidemics occur- hetero- sexual, 61% Casual heterosexual sex, 11% ring in some areas and generalized epidemics occurring in others. In most provinces in Indonesia, the epidemic Partners CHS, 4% is concentrated and sustained in key populations of IDUs and MSM, especially transgender (waria) and Source: SACEMA 2009. male sex workers. However, in the Indonesian prov- Note: CHS = individuals who have casual heterosexual sex; HIV = human immunodeficiency virus; MSM = men who have sex with men. inces of Papua and West Papua, the epidemic is driven largely by heterosexual transmission, both by sex work- Invest in the Right Interventions ers and by the general population, and HIV prevalence is much higher than in the rest of the country, at Several biomedical and sociobehavioral interventions 2.40 percent in Papua and 0.27 percent in West Papua have been effective at preventing infection and transmis- (Indonesian National AIDS Commission 2012). sion at the individual and population levels. All are According to projections for 2000–20, the epidemic important in any comprehensive response, but specific in Jakarta began with male IDUs but quickly spread to interventions have been proven to be most effective in MSMs, who soon became the largest source of transmis- certain epidemic contexts. Prioritizing the right inter- sion. The epidemic in Papua spread steadily through ventions based on epidemic context is essential. females in the general population (figure 8.6). Interventions for Concentrated Epidemics Effective targeted interventions for concentrated epi- TAILOR THE HIV/AIDS RESPONSE demics have six core components: Tailoring the local response to a local epidemic requires • Behavior-change communication more than just knowledge of transmission dynamics. It • Condom programs and needle and syringe programs also requires the use of proven interventions that target (NSPs) the right people and that are applied effectively in the • Sexual health and harm-reduction services, including right places. opioid substitution therapy (OST) Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 161 Figure 8.6 Projected Transmission Dynamics of the Mixed HIV Epidemic in Jakarta and Papua, Indonesia, 2000–20 a. Number of new infections in Jakarta b. Sources of new infections in Jakarta, 2010 (concentrated epidemic), 2000–20 4,000 <1% 1% 3,500 1% Number of new HIV infections 3,000 8% 2,500 7% 23% 2,000 14% 1% 1,500 9% 1,000 36% 500 0 2000 2005 2010 2015 2020 c. Number of new infections in Papua d. Sources of new infections in Papua, 2010 (generalized epidemic), 2000–20 1,200 <1% 1,000 3% Number of new HIV infections 800 11% 9% 6% 600 400 72% 200 0 2000 2005 2010 2015 2020 Female injecting Male injecting Waria (transgender) Men who have sex with drug users drug users men and women MSM Clients of sex workers Indirect sex workers Direct sex workers General females General Source: Modeling by Kirby Institute, University of South Wales. Note: HIV = human immunodeficiency virus; MSM = men who have sex with men. 162 Major Infectious Diseases • HIV testing, counseling, and treatment (ART) in randomized clinical trials, behavioral interventions • Solidarity and community empowerment had a minimal effect on reducing incidence of HIV • Supportive local and national legal and policy among MSM. environments. Solid evidence exists for the use of specific interven- tions in concentrated epidemics, but coverage and imple- The World Health Organization (WHO), the United mentation are lacking in many programs. UNAIDS Nations Joint Programme on HIV/AIDS (UNAIDS), (2012) reports that 86 percent of countries have low or and the United Nations Office on Drugs and Crime pro- no coverage of NSPs, and more than 85 percent of people mote a comprehensive package of HIV interventions for who inject drugs lack access to NSP, OST, or ART services IDUs, including three priority interventions: NSPs, OST, (Stoicescu 2012). MSM and sex workers are poorly and ART. In a study of 99 cities, those with NSPs saw a represented in prevention programs, with 70 percent and 19 percent reduction in prevalence per year, while those 75 percent of countries reporting very low or no cover- without NSPs experienced an 8 percent increase over age for these populations, respectively (UNAIDS 2012). approximately one decade (MacDonald and others The inadequate coverage of sex workers is especially 2003). Among other effects, OST programs have been disconcerting, since the bulk of new infections come shown to reduce HIV incidence among IDUs (MacArthur from this group in many concentrated epidemics. and others 2012; van den Berg and others 2007). ART Finally, preexposure prophylaxis (PrEP) is a promis- has reduced transmission by reducing infectiousness ing new biomedical tool for use in key populations and (Cohen and others 2011; Quinn and others 2000; Wood is recommended by the WHO for individuals at “sub- and others 2009). stantial risk” of HIV (WHO 2015). The WHO made the Community empowerment, often leading to commu- recommendation based on evaluations of PrEP effective- nity mobilization, has been effective in programs targeting ness in a number of studies among serodiscordant sex workers. Community empowerment and mobilization couples, heterosexual men, women, MSM, IDU, and has led to increased condom use, lower risk for STIs, and transgender women, especially when adherence was increased HIV testing in India (Beattie and others 2014; high. For example, PrEP reduced HIV incidence by Blanchard and others 2013; Fonner and others 2014) and 44 percent among MSM and by 90 percent among fully Africa (Chersich and others 2013). In Asia, social pro- adherent MSM in a multicountry study (Grant and grams marketing condoms have increased condom use others 2010). PrEP has been shown to prevent infection (Lipovsek and others 2010; Ngoc and others 2011; USAID in IDUs, resulting in a 49 percent reduction of HIV inci- 2011). Along with promoting the use of condoms, other dence in PrEP groups compared with placebo groups interventions to address sexual health, including reducing (Choopanya and others 2013). Overall, across studies in the incidence of and treating STIs, have also been associ- different at-risk populations, a meta-analysis of 10 ran- ated with a reduction in HIV prevalence in sex-worker domized controlled trials found a 51 percent reduction populations (Ghys and others 2001; Ghys and others 2002; in infection risk among PrEP users versus those who Laga and others 1994; Rojanapithayakorn and Hanenberg took a placebo (Fonner and others 2016). Many studies 1996). One of the most successful interventions has been show PrEP efficacy, but success outside the context of India’s Avahan Program, which is supported by the Bill & clinical trials will not be easy. Adherence may be lower in Melinda Gates Foundation. A large-scale, cost-effective real-life implementation scenarios, challenging the effec- program promoting all of the core components of inter- tiveness of PrEP. In these settings, implementation of ventions for concentrated epidemics, the Avahan Program adherence-support programs will be very important in averted an estimated 42 percent of HIV infections in combination with PrEP. Also, achieving success in devel- 4 years and an estimated 57 percent in 10 years (Pickles oping countries will be much more difficult because and others 2013; Vassall and others 2014). fewer programs in low- and middle-income countries Condom promotion; HIV testing, counseling, and reach key populations such as MSM and IDUs. treatment; and behavior change to reduce risky sexual practices are important interventions for MSM. In the Interventions for Generalized Epidemics 1980s, behavior change on a large scale, including a Various interventions have been tested for generalized decrease in the number of sexual partners and an increase epidemics, including those that have been effective in in the use of condoms, helped to curb the epidemic in concentrated epidemics. However, randomized con- MSM in Australia and the United States (Kippax and trolled trials using HIV incidence endpoints have Race 2003; Winkelstein and others 1987). However, found only VMMC, ART-based prevention, and financial Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 163 incentives to be consistently effective (figure 8.7). Of the effectiveness of TasP at the population level. In addition, cash transfer trials, one used STIs, not HIV specifically, as randomized clinical trials are under way in several an endpoint, and another used prevalence as a proxy African field sites (Botswana, Kenya, South Africa, for incidence. Uganda, and Zambia) to evaluate changes in incidence ART-based prevention, such as TasP, has been proven as a result of TasP and other interventions. to reduce transmission. In clinical trials, TasP can reduce VMMC is extraordinarily effective at reducing HIV risk by more than 96 percent (Cohen and others 2011). incidence in generalized epidemics. In clinical trials, the Outside of clinical trials, it is still effective but less so. In protective effect of VMMC was 60 percent (Auvert and KwaZulu-Natal, South Africa, infection was 34 percent others 2005; Bailey and others 2007; Gray and others lower in areas with 30 percent to 40 percent ART cover- 2007); the long-term effect may be even higher. Longer- age than in areas with less than 10 percent coverage term follow-up suggested that incidence in persons (Tanser and others 2013). In China, infection was 26 who were circumcised, compared with those who were percent lower in serodiscordant couples in which the not, was 67 percent lower over 4.5 years in Kenya, 73 infected partner received ART than in couples in which percent lower over 2.8 years in Uganda, and 76 percent the infected partner did not (Jia and others 2013). In lower over 3.0 years in South Africa (Auvert and others Swaziland, where about 85 percent of persons with CD4 2013; Bailey and others 2010; Gray and others 2012). In cell counts of less than 350 are on ART, measured annual 13 countries, projections suggest that 80 percent imple- HIV/AIDS incidence was 2.4 percent on top of 26 mentation of VMMC (20 million circumcisions) and percent adult prevalence (Swaziland Ministry of Health sustaining this level of coverage could avert 3.4 million and Social Welfare 2002). Because of TasP’s proven pre- new infections through 2025 (Njeuhmeli and others vention benefits, the WHO now recommends a universal 2011). This target has not yet been met; at the end of TasP strategy. Under the test-and-treat strategy, all indi- 2015, only 11.6 million circumcisions had been per- viduals living with HIV/AIDS are eligible for ART, formed (UNAIDS 2016). regardless of CD4 count, and catching and treating HIV Randomized controlled trials with STI and HIV infection earlier may prevent new infections within the endpoints conducted by the World Bank and others community. As more programs adopt and implement show that financial incentives, including conditional this strategy, more information will be available on the cash transfers, can reduce HIV and other STIs (figure 8.8). The premise behind financial incentives is that the income effect on sexual behavior may mit- Figure 8.7 Effectiveness of Interventions for Preventing Infections in Generalized HIV/AIDS Epidemics igate sexually risky behavior, such as when impover- ished young girls seek out older males (who are more likely to be infected) for gifts and financial assistance Microbicides 10 1 or engage in transactional sex. In Tanzania, individu- als who received up to US$60 annually to stay STI- Behavior change 9 1 free had 25 percent lower STI prevalence (de Walque STI treatment 7 1 and others 2012). In Malawi, girls whose families received up to US$15 a month to stay in school had HIV vaccines 4 1 70 percent lower prevalence, compared with the con- trol group, members of which received no payment Preexposure prophylaxis 5 2 (Baird and others 2012). Adolescents in Lesotho who received a lottery ticket to win up to US$50 or US$100 VMMC 3 3 every four months if they stayed STI- and HIV-free Antiretroviral therapy had 25 percent lower HIV incidence, with even 9 7 –based prevention greater reductions among girls (33 percent) and in Financial incentives 5 3 the group receiving US$100 (31 percent). Although promising, these initial studies represent Total 52 19 only early efforts to determine whether financial incen- Trials completed or stopped Trials effective tives can and should be included in HIV prevention pro- grams. The Tanzania study assessed the impact on STIs Sources: Baeten and others 2012; Marrazzo and others 2015; Padian and others 2010; other than HIV, although reducing STI risk may also Peterson and others 2007; Thigpen and others 2012; Van Damme and others 2012; reduce HIV risk; the Lesotho study found only a small Weiss and others 2008. Note: HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome; statistically significant effect. The Malawi study used STI = sexually transmitted infection; VMMC = voluntary medical male circumcision. cumulative HIV prevalence among young women as a 164 Major Infectious Diseases Figure 8.8 Effect of Financial Incentives on Reducing HIV and STIs in Lesotho, Malawi, and Tanzania 1.0 1.0 1 1.0 1.0 1.0 Prevalence of HIV/AIDS, percent 0.8 a a 0.75 0.73 0.6 0.67 a 0.69 a 0.4 a 0.36 0.2 0 Control Lottery Lottery Only Control Cash Control Low High eligible females high-value transfers conditional conditional lottery cash cash transfers transfers Lesotho Malawi Tanzania Lottery Cash transfers trial RESPECT trial (HIV incidence) (HIV prevalence) (STI prevalence) Sources: Baird and others 2012; de Walque and others 2012; Nyqvist and others 2015. Note: HIV = human immunodeficiency virus; RESPECT = Rewarding STI Prevention and Control in Tanzania; STIs = sexually transmitted infections. a. Statistically significant effect. proxy for incidence, and there was no baseline or random- While an epidemic may be classified as generalized, ization. Furthermore, to the disappointment of the HIV- concentrated, or mixed at the national level, there may prevention field, the most recent evaluations of cash be much heterogeneity at the subnational level. For transfer interventions in South Africa’s generalized epi- example, Vietnam’s concentrated epidemic is sustained demic have not shown the same success as the studies in by high prevalence in MSM, male IDUs, and female Lesotho, Malawi, and Tanzania, and were challenged by sex workers, but considerable regional variation is evi- study design and context issues (HPTN 2015; Karim and dent in the number of infections and persons infected others 2015). Further studies are needed to evaluate the (map 8.2). effectiveness of financial incentives for prevention, Two approaches are useful for identifying and reach- whether they are scalable and affordable, and whether ing the right people in the right places: (1) program their outcomes are durable. science, and (2) geographic targeting and hotspot For the general population and in generalized mapping. epidemics, limited data exist on the effectiveness of pro- moting condom use and of behavior-change programs, Right People: Program Science Approach and a Cochrane review found no clear evidence that Program science is the “systematic application of theo- community interventions promoting condom use retical and empirical scientific knowledge to improve the reduce HIV or STI transmission (Moreno and others design, implementation, and evaluation of public health 2014). However, these interventions need to be sup- programmes” (Blanchard and Aral 2011, 1). Program ported as part of the overall HIV/AIDS response, as they science is useful for tailoring the response to local epi- absorb a small share of the budget. demics and has three core components: • Program intelligence: Data collection and informa- Tailor the Response for the Right People in tion gathering on modes of transmission, optimal the Right Places resource allocation, local epidemic appraisal, and It is important not only to invest in context-appropriate ethnographic-linked interventions interventions, but also to reach the right people with • Program implementation: Design, development, these interventions and to apply them in the right places. and delivery of standard intervention packages, Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 165 Map 8.2 HIV/AIDS Burden and Sources of New Infections in Vietnam, by Region, 1990 and 2014 IBRD 42332 | DECEMBER 2016 Trends in the distribution of new HIV infections among specific population groups, various provinces in Vietnam, 1990–2014 Number of people living HANOI with HIV/AIDS (2012) 100% 6,000 and above 4,000–5,999 2,000–3,999 1,000–1,999 0% <1,000 1990 2014 NORTH WEST 100% Sources of new infections (1994–2014) Low-risk men who have sex with men High-risk men who 0% have sex with men 1990 2014 Low-risk males Low-risk females Clients of sex worker Female sex workers HO CHI MINH CITY 100% People who inject drugs 0% 1990 2014 Source: UNAIDS 2014a. Note: HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome. implementation manuals, quality assurance guide- Program science has been used to assess and design the lines, support and training for coordinated program response to the epidemic in Nigeria, which is the world’s implementation, and knowledge exchange most diverse and second-largest epidemic by numbers • Program evaluation: Real-time program management infected. The transmission dynamics are mixed: 42 data and impact evaluation. percent of new infections occur in low-risk heterosexual 166 Major Infectious Diseases individuals within the general population, although key Figure 8.9 Example of Heterogeneity of Sources of New HIV populations—only 1 percent of the total population— Infections in Cross River State, Nigeria account for 23 percent of new infections (Blanchard and Partners of CHS (female), 2.3% Boyfriends, 0.4% Aral 2011; NACA 2014). In Cross River State, more than Partners of CHS (male), 0.5% Girlfriends, 2.3% 50 percent of transmission is from key populations and CHS (females), 6.3% IDUs, 3.4% their partners (Prudden and others 2013) (figure 8.9). CHS (males), 1.6% Brothel-based FSWs, 7.6% Female partners of Prevalence is equally varied, with the highest prevalence MSM, 0% Brothel-based FSW Regular female partners clients, 6.8% concentrated in four states and the lowest (well under the of clients, 7.2% Non-brothel-based 3 percent of the national average) in nine states (map 8.3). Partners of IDUs, 0.3% FSWs, 4.2% Program intelligence, gathered through biological Non-brothel-based FSW clients, 2.1% and behavioral surveillance and through local epidemic appraisals conducted initially in eight states, pro- Females engaged in transactional sex, 16.1% vided additional information on populations at Discordant couples, risk in Nigeria. HIV/AIDS prevalence is high in key 34.0% Males engaged in trasactional sex, 1.6% populations, specifically IDUs (42 percent), MSM MSM, 3.2% (17 percent), sex workers not in brothels (22 percent), MARPs: 45.0% Discordant couples: 34.0 and sex workers in brothels (27 percent) (NACA Partners of MARPs: 7.5% General population: 13.4% 2014). Local epidemic appraisals mapped out key pop- Source: Prudden and others 2013. ulation hotspots and estimated their size. Female sex Note: CHS = individuals who have casual heterosexual sex; FSW = female sex worker; HIV = human workers were the largest key population: there were immunodeficiency virus; IDU = injecting drug user; MARP = most-at-risk population; MSM = men 10,581 female sex workers, compared with 447 IDUs who have sex with men. and 495 MSM (NACA 2013). Sex workers by state were enumerated and, in some cases, geospatially mapped out (map 8.4). An assessment of the venues most fre- Right Places: Geographic Targeting and Hotspot quented by individuals seeking new sexual partners Mapping revealed that bars, clubs, and restaurants were the Hotspot mapping can help locate key populations in a primary locations in which both sexes sought out sex- concentrated epidemic and help determine the optimal ual partners (figure 8.10). allocation of national resources, based on the size of The information gathered through program intelli- the epidemic across regions. gence provided the basis for the design of a strategy that Geographic targeting can reveal locations where targeted key populations and that ranged from national- epidemics are concentrated. For example, more than to local-level planning. This strategy entailed the follow- 70 percent of India’s HIV/AIDS burden began in ing steps (NACA 2013): four southern states (Andhra Pradesh, Karnataka, Maharashtra, and Tamil Nadu) and four northeast- • Specify the target population, for example, female sex ern states (Bihar, Gujarat, Uttar Pradesh, and West workers, and develop the rationale for selection. Bengal) and spread from there (NACO 2013; World • Define the intervention package, segmented by pop- Bank 2012). ulation group. Hotspot targeting can also help locate key popula- • Set coverage targets at the macro level; for example, tions. Pakistan has a severe epidemic in its IDU popula- cover 60 percent of estimated female sex workers. tion, 69 percent of whom live in four cities (Faisalabad, • Set specific objectives for project reach at the micro Hyderabad, Karachi, and Lahore) that account for level; for example, reach 80 percent of female sex 19 percent of the total population.2 HIV is also growing workers using peer education. among female and transgender sex workers: 72 percent • Set outcome objectives; for example, obtain consis- of female sex workers are in five cities (Faisalabad, tent condom use of 80 percent. Hyderabad, Karachi, Lahore, and Multan), and 64 • Plan and implement a scaled program, including a percent of transgender sex workers are in three cities scaling-up strategy. (Karachi, Lahore, and Multan) (Blanchard 2012; Reza and others 2013) (map 8.5).3 The final component of program science is evalua- Average national HIV prevalence masks subnational tion. The Nigerian government has indicated that it variations. An analysis of 20 Sub-Saharan African coun- plans to conduct an evaluation using a combination of tries revealed spatial clustering of infections, in which dose-response behavioral and biological data and math- 14 percent of the population lives in high-prevalence ematical modeling. clusters and 16 percent lives in low-prevalence clusters, Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 167 Map 8.3 Prevalence of HIV/AIDS in Nigeria, by Region IBRD 42333 | DECEMBER 2016 SOKOTO KATSINA JIGAWA YOBE ZAMFARA BORNO KANO KEBBI KADUNA BAUCHI GOMBE NIGER AWA PLATEAU A ABUJA ABBUJA B UJA ADAM FEDERAL KWARA CAPITAL TERRITORY NASARAWA OYO EKITI TARABA OSUN KOGI BENUE HIV/AIDS ???? OGUN ONDO >8.00 ENUGU LAGOS EDO 3.01–8.00 ANAMBRA EBONYI 1.11–3.00 CROSS 0.21–1.10 DELTA IMO RIVER <0.21 ABIA RIVERS AKWA- BAYELSA IBOM Source: NACA 2014. ”Global AIDS Response Country Progress Report: Nigeria GARPR 2014.” NACA, Abuja. Note: HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome. the difference potentially due to behavioral or biological these high-burden regions while scaling back resources factors (Cuadros, Awad, and Abu-Raddad 2013). Islands for the eight lowest-burden counties, which account for of high or low prevalence, due to intense clustering, can less than 2 percent of new infections (Government of serve as prime targets for a focused response. Kenya 2013). Hotspot mapping can also guide targeted prevention Geographic targeting and hotspot mapping are efforts. In Thailand, 70 percent of new infections occur relatively new and need to be tested rigorously for in 33 of the country’s 76 provinces (National AIDS feasibility and utility. Ideally, this time- and resource- Management Center 2013); in Kenya, nine counties with intensive approach would reveal epidemic dynamics the highest burden account for 54 percent of new infec- adhering to the 80:20 rule, where 80 percent of a parame- tions but account for only 24 percent of the Kenyan ter (prevalence, new infections, risk, transmission dynam- population (Government of Kenya 2013).4 It makes ics) can be attributed to 20 percent of a larger group sense to allocate the majority of prevention resources to (population, region). This outcome may be obtained 168 Major Infectious Diseases Map 8.4 Number of Sex Workers in Nigerian Hotspots IBRD 42334 | DECEMBER 2016 GOMBE 5,772 ABUJA FEDERAL CAPITAL 24,376 TERRITORY NASARAWA 19,953 BENUE ONDO 10,034 LAGOS 9,677 ANAMBRA CROSS RIVER Number of sex workers 4,846 in hotspots 9,838 46,691 Source: NACA (National Agency for the Control of AIDS). 2013. “HIV Epidemic Appraisals in Nigeria: Evidence for Prevention Programme Planning and Implementation: Data from the First Eight States.” NACA, Abuja. Figure 8.10 Popularity of Venues Where Individuals Seek New Sexual Partners in Nigeria 70 60 58.6 59.5 Venue popularity, percent 50 40 30 20 10 14.4 15.5 14.6 14.8 4.7 0.5 6.3 4.2 0.5 6.4 0 Males seeking females Females seeking males Bars, clubs, restaurants Brothels Beauty salons, the Internet Hotels, lodges Other public places Others Source: NACA 2013. Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 169 Map 8.5 Population Density (per 1,000 Adult Males) of Key HIV/AIDS Populations in City Hotspots in Pakistan IBRD 42335 | DECEMBER 2016 Peshawar 4.2 1.8 ISLAMABAD ISLAMADAD 2.3 11.5 2.4 Faisalabad 1.7 Lahore 8.1 5.0 3.3 Quetta 1.8 10.7 6.9 12.4 Larkana 8.0 8.2 Female sex workers Injecting drug users Male sex workers 3.3 5.5 Karachi 3.5 Source: Reza and others 2013. more typically in concentrated epidemics, but even gener- mix of interventions. Developed by the Kirby Institute alized epidemics exhibit varying degrees of geographic at the University of New South Wales with input from and population-level concentration of infections. the World Bank, Optima is a mathematical model of transmission dynamics and disease progression inte- grated with an economic and financial analysis OPTIMIZE ALLOCATIVE EFFICIENCY framework. Its highly flexible structure can accom- FOR COST-EFFECTIVE PROGRAMMING: modate country-specific inputs. It can divide the population into subgroups and characterize them THE RIGHT MIX by demographics, HIV risk, disease, and clinical and In all epidemics, the response is more effective and more treatment status, including changes in health states cost-effective if it is tailored to the local transmission and population groups. The model uses a range of dynamics. To achieve maximum impact per expenditure, inputs (demographic, epidemiological, behavioral, prioritizing and optimizing the allocation of resources clinical, and financial) to generate the most appropri- are crucial; that is, the right mix of interventions must be ate ratio of interventions at different spending levels. applied. Although not all parameters have to be entered, the The use of Optima—an epidemic- and cost-modeling more data points in the model, the more reliable the software—can facilitate the effort to apply the right analysis. 170 Major Infectious Diseases The result is a set of analyses showing outcomes and gradually add counseling, testing, and treatment based on a range of programmatic allocations and services as the budget increases. Only when spending financial investments in comparison with current allo- exceeds the current budget will investments in preven- cations and investments. For example, an Optima tion be warranted for low-risk populations. analysis for Belarus suggests that current allocations Optima can also indicate ways to maximize the would have to be restructured if the number of new impact of current spending. By spending the same HIV infections is to be reduced by 2020. The optimal amount of money differently, Belarus could reduce the programmatic structure would shift funds from the number of new infections by 27 percent, and a 2014 general population to key populations, specifically analysis projected halving the number of new infections increasing NSP spending until saturation is reached, in Swaziland by 2016. This improvement in allocative and tripling and doubling spending on condom pro- and technical efficiency in Swaziland would cost only an grams targeting MSM and female sex workers, respec- additional US$8 million, as compared with 10 times that tively. To minimize new infections, the results suggest amount based on current allocations (figure 8.12). Table that Belarus should substantially increase ART cover- 8.1 lists the costs associated with scaling up specific age (figure 8.11). interventions in Swaziland, including the incremental Optima can produce optimal allocations for a range cost-effectiveness ratio for infections averted. of budget scenarios. In the tightest budget scenario, The use of Optima for Belarus is an excellent example Belarus should focus on prevention in key populations of how shifting allocations can maximize the impacts of Figure 8.11 Optima-Projected Results for Belarus: Current and Optimal Allocation of Resources for HIV/AIDS Programs No spending 10 13% 20 32% 10% 30 40 Optimal spending (% of budget) 50 17% 4% 60 <1% 3% 4% 17% 70 80 9% Current spending (%) 15% 90 10% 100 10% Current 120 26% 150 29% 200 0 5 10 15 20 Spending (US$, millions) <1% Optimal spending (%) Low-risk populations: Condom use, % Femal sex workers: Condom use, % Men who have sex with men: Condom use, % Needle and syringe program: % Opioid substitution therapy: % on treatment Antiretroviral therapy: % on treatment Voluntary counseling and testing: % testing per year Prevention of mother-to-child transmission: % per birth Source: Wilson and others 2013. Note: HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome. Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 171 Figure 8.12 Allocative and Technical Efficiency Gains and Costs Associated with Halving the Number of New HIV Infections in Swaziland by 2016 200 180 160 Spending (US$, millions) 140 Potential funding gap: 8 million 120 100 80 60 40 20 0 a Current eNSF Scale-up Optimal Optimal+cheaper Optimal+cheaper spendinga proportionate to testing testing+ current allocation management People living with HIV/AIDS Orphans and Testing Antiretroviral Men who have vulnerable children treatment sex with men Female sex workers Voluntary medical Prevention of mother- Condom use Behavior change male circumcision to-child transmission communication Source: Unpublished data from Wilson and Kerr. Note: eNSF= Extended National Strategic Framework; HIV = human immunodeficiency virus. a. The Optima model projects that incidence will decline by 13 percent by 2018 under current spending, and by 35 percent under the eNSF. Table 8.1 Optima-Projected Program Costs of Moderate Scale-up of Interventions, Number of New Infections and Deaths Averted, and Resulting Incremental Cost-Effectiveness Ratios in Swaziland by 2030 US$ unless otherwise noted Infections averted AIDS-related deaths averted Moderate scale-up of Unit costs (US$ per Discounted cost Incremental cost- Incremental cost- interventions person per year) of programa* Number effectiveness ratio Number effectiveness ratio All interventions — 74,205,074 27,355 2,713 7,413 10,010 Antiretroviral therapy 131.70 13,654,653 7,452 1,832 4,254 3,210 Voluntary medical male 130.70 (per male 2,475,488 13,291 186 1,334 1,856 circumcision age 10–49) CCTs 76.56a (per female 45,031,793 9,895 4,551 1,033 43,608 age 15–24) Prevention of mother-to-child 186.00 (per pregnant 767,735 1,164 659 221 3,480 transmission of HIV woman) TB and HIV/AIDS 247.00b 37,669,429 — — 1,220 30,883 co-treatment Source: Kelly and others 2014. Note: * = Discounted cost is today’s dollar equivalent of future costs. CCT = conditional cash transfer; TB = tuberculosis; — = not applicable. a. It was assumed that only 29 percent of the costs for the CCT program would be allocated to the HIV program, as there are multiple benefits to this intervention (see Remme and others 2014, table 3); therefore, the HIV program would not be expected to fund the full cost of the CCT program, which is US$264 per girl or young woman age 15–24 per year. b. It was assumed that the cost of the TB/HIV program would increase proportionately to the increased TB/HIV coverage over time to meet targets. As well, since it was reported that “the prevalence of HIV among TB patients is 79.6 percent, therefore 79.6 percent of total expenditure on TB was captured as TB/HIV treatment expenditure” (UNAIDS, n.d.), it follows that only 80 percent of the TB/HIV spending was allocated to the HIV program. 172 Major Infectious Diseases Figure 8.13 Optima Analysis Findings in Eastern Europe and Central Asia 100 90 Universal 80 ART Percentage of spending 70 60 KP prevention 50 LRP 40 prevention 30 Too much 20 spent on 10 overhead 0 Armenia Belarus Bulgaria Georgia Kazakhstan Kyrgyz FYR Maldova Tajikistan Ukraine Uzbekistan Republic Macedonia Other Management/HR/training Strategic info/research/M&E Infrastructure Enabling environment Blood safety/PEP/precautions Support programs STIs and OIs BCC programs Migrant/Roma programs Key populations FSW programs MSM programs IDU programs NSPs OST HCT PMTCT ART and lab monitoring Source: David Wilson, personal communication. Note: ART = antiretroviral therapy; BCC = behavioral change communication; FSW = female sex worker; HCT = HIV counseling and testing; HR = human resources; KP = key population; LRP = low-risk population; M&E = monitoring and evaluation; MSM = men who have sex with men; NSP = needle and syringe program; OI = opportunistic infection; OST = opioid substitution therapy; PEP = postexposure prophylaxis; PMTCT = prevention of mother to child transmission; PWID = people who inject drugs; STI = sexually transmitted infection. current and future spending in a concentrated epidemic. variation of epidemics is needed, together with the The opportunity to increase allocative efficiency is mod- use of proven interventions. Greater targeting of the est in generalized epidemics, such as in Swaziland, because right interventions for the right people in the right transmission is not highly concentrated and resources are places at the right times will improve the efficiency not grossly misallocated. It is difficult to determine what and effectiveness of the global response. The HIV/ programs or interventions could be assigned a lower pri- AIDS epidemic will not end without an effective vac- ority or given up. Nevertheless, Optima is an important cine or cure. In the meantime, greater understanding first step in painstaking hands-on work that focuses on of transmission dynamics and more efficient imple- intracategory analyses and the interplay between alloca- mentation and delivery of prevention, detection, and tive and implementation efficiency, keeping in mind the treatment programs can prevent a substantial propor- heterogeneity and uniqueness of each epidemic at tion of new infections. national and subnational levels. However, some general conclusions can be made from looking at a range of Optima analyses performed in Eastern Europe and NOTES Central Asia. Across the board, optimal allocations World Bank Income Classifications as of July 2014 are as include larger investments in ART for all populations and follows, based on estimates of gross national income (GNI) in HIV programs for key populations. Fewer resources per capita for 2013: could be spent on programs for low-risk populations, and the analyses reveal that far too much is spent on • Low-income countries (LICs) = US$1,045 or less overhead and program administration (figure 8.13). • Middle-income countries (MICs) are subdivided: (a) lower-middle-income = US$1,046 to US$4,125 (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 CONCLUSIONS • High-income countries (HICs) = US$12,746 or more. Financing of HIV/AIDS programs is subject to 1. For more a more detailed discussion on the use of models numerous competing priorities. To make better use to inform decision making regarding resource allocation of existing resources, a greater understanding of the for HIV program planning, please refer to chapter 9 of this heterogeneity, transmission dynamics, and geographic volume (Kahn and others 2017). Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 173 2. Calculations based on 2010 and 2011 data from City Blanchard, J. 2012. “Global Program Science Initiatives.” Paper Mayors.com (http://citymayors.com/statistics/largest-cities prepared for the National STD Prevention Conference, -mayors-mr2.html) and Knoema.com (http://knoema Minneapolis, March 12. .com/atlas/Pakistan/Population). Blanchard, J., and S. O. Aral. 2011. “Program Science: An 3. Calculations based on 2010 and 2011 data from City Initiative to Improve the Planning, Implementation, and Mayors.com (http://citymayors.com/statistics/largest Evaluation of HIV/Sexually Transmitted Infection Prevention -cities-mayors-mr2.html) and Knoema.com (http:// Programmes.” Sexually Transmitted Infections 87 (1): 2–3. knoema.com/atlas/Pakistan/Population). Blouin, K., P. Leclerc, C. Morissette, É. Roy, C. Blanchette, and 4. Calculations based on 2010 data from Kenya National others. 2013. “P3.099 Sex Work as an Emerging Risk Factor Bureau of Statistics (http://www.knbs.or.ke/index.php for HIV Seroconversion among Injection Drug Users in ?option=com_content&view=article&id=176&Itemid=645). the SurvUDI Network.” Sexually Transmitted Infections 89 (Suppl 1): A178. Chersich, M. F., S. Luchters, I. Ntaganira, A. Gerbase, Y. R. Lo, and others. 2013. “Priority Interventions to Reduce HIV REFERENCES Transmission in Sex Work Settings in Sub-Saharan Africa Auvert, B., D. Taljaard, E. Lagarde, J. Sobngwi-Tambekou, and Delivery of These Services.” Journal of the International R. Sitta, and others. 2005. “Randomized, Controlled AIDS Society 16 (1): 17980. Intervention Trial of Male Circumcision for Reduction of Choi, K. H., D. R. Gibson, L. Han, and Y. Guo. 2004. “High HIV Infection Risk: The ANRS 1265 Trial.” PLoS Medicine Levels of Unprotected Sex with Men and Women among 2 (11): e298. Men Who Have Sex with Men: A Potential Bridge of HIV Auvert, B., D. Taljaard, D. Rech, P. Lissouba, B. Singh, and Transmission in Beijing, China.” AIDS Education and others. 2013. “Association of the ANRS-12126 Male Prevention 16 (1): 19–30. Circumcision Project with HIV Levels among Men in a Choi, K. H., W. McFarland, T. B. Neilands, S. Nguyen, B. Louie, South African Township: Evaluation of Effectiveness Using and others. 2004. “An Opportunity for Prevention: Cross-Sectional Surveys.” PLoS Medicine 10 (9): e1001509. Prevalence, Incidence, and Sexual Risk for HIV among Baeten, J. M., D. Donnell, P. Ndase, N. R. Mugo, J. D. Campbell, Young Asian and Pacific Islander Men Who Have Sex with and others. 2012. “Antiretroviral Prophylaxis for HIV Men, San Francisco.” Sexually Transmitted Diseases 31 (8): Prevention in Heterosexual Men and Women.” New England 475–80. Journal of Medicine 367 (5): 399–410. Choopanya, K., M. Martin, P. Suntharasamai, U. Sangkum, Bailey, R. C., S. Moses, C. B. Parker, K. Agot, I. Maclean, and P. A. Mock, and others. 2013. “Antiretroviral Prophylaxis others. 2007. “Male Circumcision for HIV Prevention in for HIV Infection in Injecting Drug Users in Bangkok, Young Men in Kisumu, Kenya: A Randomised Controlled Thailand (The Bangkok Tenofovir Study): A Randomised, Trial.” The Lancet 369 (9562): 643–56. Double-Blind, Placebo-Controlled Phase 3 Trial.” The ———. 2010. “The Protective Effect of Adult Male Lancet 381 (9883): 2083–90. Circumcision against HIV Acquisition Is Sustained for at Cohen, M. S., Y. Q. Chen, M. McCauley, T. Gamble, Least 54 Months: Results from the Kisumu, Kenya Trial.” M. C. Hosseinipour, and others. 2011. “Prevention of XVIII International AIDS Conference, Vienna. HIV-1 Infection with Early Antiretroviral Therapy.” New Baird, S. J., R. S. Garfein, C. T. McIntosh, and B. Ozler. 2012. England Journal of Medicine 365 (6): 493–505. “Effect of a Cash Transfer Programme for Schooling on Colfax, G., G. M. Santos, P. Chu, E. Vittinghoff, A. Pluddemann, Prevalence of HIV and Herpes Simplex Type 2 in Malawi: and others. 2010. “Amphetamine-Group Substances and A Cluster Randomised Trial.” The Lancet 379 (9823): HIV.” The Lancet 376 (9739): 458–74. 1320–29. Cuadros, D. F., S. F. Awad, and L. J. Abu-Raddad. 2013. Baral, S., C. Beyrer, K. Muessig, T. Poteat, A. L. Wirtz, and oth- “Mapping HIV Clustering: A Strategy for Identifying ers. 2012. “Burden of HIV among Female Sex Workers in Populations at High Risk of HIV Infection in Sub-Saharan Low-Income and Middle-Income Countries: A Systematic Africa.” International Journal of Health Geographics 12: 28. Review and Meta-Analysis.” The Lancet Infectious Diseases Decker, M. R., A. L. Crago, S. K. Chu, S. G. Sherman, 12 (7): 538–49. M. S. Seshu, and others. 2014. “Human Rights Violations Beattie, T. S., H. L. Mohan, P. Bhattacharjee, S. Chandrashekar, against Sex Workers: Burden and Effect on HIV.” The Lancet S. Isac, and others. 2014. “Community Mobilization and 385 (9963): 186–99. Empowerment of Female Sex Workers in Karnataka Deering, K. N., T. Lyons, C. X. Feng, B. Nosyk, S. A. Strathdee, State, South India: Associations with HIV and Sexually and others. 2013. “Client Demands for Unsafe Sex: The Transmitted Infection Risk.” American Journal of Public Socioeconomic Risk Environment for HIV among Street Health 104 (8): 1516–25. and Off-Street Sex Workers.” Journal of Acquired Immune Blanchard, A. K., H. L. Mohan, M. Shahmanesh, R. Prakash, Deficiency Syndromes 63 (4): 522–31. S. Isac, and others. 2013. “Community Mobilization, de Walque, D., W. H. Dow, R. Nathan, R. Abdul, F. Abilahi, Empowerment, and HIV Prevention among Female Sex and others. 2012. “Incentivising Safe Sex: A Randomised Workers in South India.” BMC Public Health 13: 234. Trial of Conditional Cash Transfers for HIV and Sexually 174 Major Infectious Diseases Transmitted Infection Prevention in Rural Tanzania.” BMJ Indonesian National AIDS Commission. 2012. “Country Report Open 2: e000747. on the Follow-Up to the Declaration of Commitment Family Health International, the Lesotho Ministry of Health, on HIV/AIDS.” Indonesian National AIDS Commission, LAPCA (Lesotho AIDS Programme Co-ordinating Jakarta. Authority), Sechaba Consultants, and USAID (U.S. Ireri, A. W. 2012. “Police Discrimination against Commercial Agency for International Development). 2002. “HIV/AIDS Sex Workers’ Possession of Condoms.” Paper prepared for Behavioral Surveillance Survey: Lesotho 2002; Summary the 19th International AIDS Conference, Washington, DC, Technical Report—Round 1.” USAID, Washington, DC. July 22–27. Fonner, V. A., S. L. Dalglish, C. E. Kennedy, R. Baggaley, K. R. Jia, Z., Y. Mao, F. Zhang, Y. Ruan, Y. Ma, and others. 2013. O’reilly, and others. 2016. “Effectiveness and Safety of Oral “Antiretroviral Therapy to Prevent HIV Transmission in HIV Pre-Exposure Prophylaxis (PrEP) for All Populations: Serodiscordant Couples in China (2003–11): A National A Systematic Review and Meta-Analysis.” AIDS. Observational Cohort Study.” The Lancet 382 (9899): Fonner, V. A., D. Kerrigan, Z. Mnisi, S. Ketende, C. E. Kennedy, 1195–203. and others. 2014. “Social Cohesion, Social Participation, Johnston, L. G., and S. Corceal. 2013. “Unexpectedly High and HIV Related Risk among Female Sex Workers in Injection Drug Use, HIV, and Hepatitis C Prevalence Swaziland.” PLoS One 9 (1): e87527. among Female Sex Workers in the Republic of Mauritius.” Freeman, P., B. C. Walker, D. R. Harris, R. Garofalo, N. Willard, AIDS and Behavior 17 (2): 574–84. and others. 2011. “Methamphetamine Use and Risk for Kahn, J. G., L. A. Bollinger, J. Stover, and E. Marseille. 2017. HIV among Young Men Who Have Sex with Men in 8 “Improving the Efficiency of the HIV/AIDS Policy Response: U.S. Cities.” Archives of Pediatrics and Adolescent Medicine A Guide to Resource Allocation Modeling.” In Disease 165 (8): 736–40. Control Priorities (third edition): Volume 6, Major Infectious Ghys, P. D., M. O. Diallo, V. Ettiegne-Traore, K. Kale, O. Tawil, Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, and others. 2002. “Increase in Condom Use and Decline and P. Jha. World Bank: Washington, DC. in HIV and Sexually Transmitted Diseases among Female Karim, Q. A., K. Leask, A. Kharsany, H. Humphries, and others. Sex Workers in Abidjan, Côte d’Ivoire, 1991–1998.” AIDS 2015. “Impact of Conditional Cash Incentives on HSV-2 16 (2): 251–58. and HIV Prevention in Rural South African High School Ghys, P. D., M. O. Diallo, V. Ettiegne-Traore, G. A. Satten, C. K. Students: Results of the CAPRISA 007 Cluster Randomized Anoma, and others. 2001. “Effect of Interventions to Control Controlled Trial.” Presentation at the 8th International Sexually Transmitted Disease on the Incidence of HIV AIDS Society Conference on HIV Pathogenesis, Treatment, Infection in Female Sex Workers.” AIDS 15 (11): 1421–31. and Prevention, Vancouver, Canada, July 19–22. Gouws, E., and P. Cuchi. 2012. “Focusing the HIV Response Kelly, J. A., Y. A. Amirkhanian, D. W. Seal, C. M. Galletly, through Estimating the Major Modes of HIV Transmission: W. Difranceisco, and others. 2010. “Levels and Predictors of A Multi-Country Analysis.” Sexually Transmitted Infections Sexual HIV Risk in Social Networks of Men Who Have Sex 88 (Suppl 2): i76–85. with Men in the Midwest.” AIDS Education and Prevention Government of Kenya. 2013. “Kenya HIV Investment Case 22 (6): 483–95. Summary.” Government of Kenya, Nairobi. Kelly, S., A. Shattock, C. C. Kerr, T. Gama, N. Nhlabatsi, and Grant, R. M., J. R. Lama, P. L. Anderson, V. McMahan, A. Y. Liu, others. 2014. HIV Mathematical Modelling to Support and others. 2010. “Preexposure Chemoprophylaxis for HIV Swaziland’s Development of Its HIV Investment Case. Final Prevention in Men Who Have Sex with Men.” New England Report. Washington, DC: World Bank. Journal of Medicine 363 (27): 2587–99. Kerrigan, D., A. Wirtz, S. Baral, M. Decker, L. Murray, and Gray, R. H., G. Kigozi, X. Kong, V. Ssempiija, F. Makumbi, and others. 2013. The Global HIV Epidemics among Sex Workers. others. 2012. “The Effectiveness of Male Circumcision for Directions in Development. Washington, DC: World Bank. HIV Prevention and Effects on Risk Behaviors in a Posttrial King, E. J., S. Maman, J. M. Bowling, K. E. Moracco, and V. Dudina. Follow-up Study.” AIDS 26 (5): 609–15. 2013. “The Influence of Stigma and Discrimination on Gray, R. H., G. Kigozi, D. Serwadda, F. Makumbi, S. Watya, Female Sex Workers’ Access to HIV Services in St. Petersburg, and others. 2007. “Male Circumcision for HIV Prevention Russia.” AIDS and Behavior 17 (8): 2597–603. in Men in Rakai, Uganda: A Randomised Trial.” The Lancet Kippax, S., and K. Race. 2003. “Sustaining Safe Practice: Twenty 369 (9562): 657–66. Years On.” Social Science and Medicine 57 (1): 1–12. Grulich, A. E., and I. Zablotska. 2010. “Commentary: Laga, M., M. Alary, N. Nzila, A. T. Manoka, M. Tuliza, and Probability of HIV Transmission through Anal Intercourse.” others. 1994. “Condom Promotion, Sexually Transmitted International Journal of Epidemiology 39 (4): 1064–65. Diseases Treatment, and Declining Incidence of HIV-1 HPTN (HIV Prevention Trials Network). 2015. “Cash Transfers Infection in Female Zairian Sex Workers.” The Lancet Conditional on Schooling Do Not Prevent HIV Infection 344 (8917): 246–48. among Young South African Women.” Presentation at Lau, J. T. F., H. Y. Tsui, Y. Zhang, F. Cheng, L. Zhang, and others. 2008. the 8th International AIDS Society Conference on HIV “Comparing HIV-Related Syringe-Sharing Behaviors among Pathogenesis, Treatment, and Prevention, Vancouver, Female IDU Engaging versus Not Engaging in Commercial Canada, July 19–22. Sex.” Drug and Alcohol Dependence 97 (1–2): 54–63. Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 175 Lipovsek, V., A. Mukherjee, D. Navin, P. Marjara, A. Sharma, Ngoc, K. V., G. Mundy, Y. Madan, J. Ayers, D. Sherard, and others. 2010. “Increases in Self-Reported Consistent and others. 2011. “Using Formative Research to Promote Condom Use among Male Clients of Female Sex Workers Behavior Change among Male Clients of Female Sex Following Exposure to an Integrated Behaviour Change Workers in Vietnam.“ Cases in Public Health Communication Programme in Four States in Southern India.” Sexually and Marketing 5 (Proceedings): 27–47. Transmitted Infections 86 (Suppl 1): i25–32. Njeuhmeli, E., S. Forsythe, J. Reed, M. Opuni, L. Bollinger, MacArthur, G. J., S. Minozzi, N. Martin, P. Vickerman, S. Deren, and others. 2011. “Voluntary Medical Male Circumcision: and others. 2012. “Opiate Substitution Treatment and HIV Modeling the Impact and Cost of Expanding Male Transmission in People Who Inject Drugs: Systematic Review Circumcision for HIV Prevention in Eastern and Southern and Meta-Analysis.” British Medical Journal 345: e5945. Africa.” PLoS Medicine 8 (11): e1001132. MacDonald, M., M. Law, J. Kaldor, J. Hales, and G. J. Dore. Nyqvist, M. B., L. Corno, D. de Walque, and J. Svensson. 2015. 2003. “Effectiveness of Needle and Syringe Programmes “Using Lotteries to Incentivize Safer Sexual Behavior: for Preventing HIV Transmission.” International Journal of Evidence from a Randomized Controlled Trial on HIV Drug Policy 14 (5): 353–57. Prevention.” Policy Research Working Paper 7215, World Marrazzo, J. M., G. Ramjee, B. A. Richardson, K. Gomez, Bank, Washington, DC. N. Mgodi, and others. 2015. “Tenofovir-Based Preexposure Open Society Foundations. 2012. “Criminalizing Condoms: Prophylaxis for HIV Infection among African Women.” New How Policing Practices Put Sex Workers and HIV Services England Journal of Medicine 372 (6): 509–18. doi:10.1056/ at Risk in Kenya, Namibia, Russia, South Africa, the United NEJMoa1402269. States, and Zimbabwe.” Sexual Health and Rights Project, Maseko, S., and S. Ndlovu. 2012. “Condoms as Evidence: Police, Open Society Foundations, New York. Sex Workers, and Condom Confiscation in Zimbabwe.” Padian, N. S., S. I. McCoy, J. E. Balkus, and J. N. Wasserheit. Paper prepared for the 19th International AIDS Conference, 2010. “Weighing the Gold in the Gold Standard: Washington, DC, July 22–27. Challenges in HIV Prevention Research.” AIDS 24 (5): Mathers, B. M., L. Degenhardt, B. Phillips, L. Wiessing, 621–35. M. Hickman, and others. 2008. “Global Epidemiology of Peterson, L., D. Taylor, R. Roddy, G. Belai, P. Phillips, and others. Injecting Drug Use and HIV among People Who Inject 2007. “Tenofovir Disoproxil Fumarate for Prevention Drugs: A Systematic Review.” The Lancet 372 (9651): of HIV Infection in Women: A Phase 2, Double-Blind, 1733–45. Randomized, Placebo-Controlled Trial.” PLoS Clinical Moreno, R., H. Y. Nababan, E. Ota, W. M. Wariki, S. Ezoe, Trials 2 (5): e27. and others. 2014. “Structural and Community-Level Pickles, M., M. C. Boily, P. Vickerman, C. M. Lowndes, Interventions for Increasing Condom Use to Prevent the S. Moses, and others. 2013. “Assessment of the Population- Transmission of HIV and Other Sexually Transmitted Level Effectiveness of the Avahan HIV-Prevention Infections.” Cochrane Database of Systematic Reviews 7: Programme in South India: A Preplanned, Causal-Pathway- CD003363. doi:10.1002/14651858.CD003363.pub3. Based Modelling Analysis.” The Lancet Global Health 1 (5): NACA (National Agency for the Control of AIDS). 2013. “HIV e289–99. Epidemic Appraisals in Nigeria: Evidence for Prevention Prudden, H. J., C. H. Watts, P. Vickerman, N. Bobrova, Programme Planning and Implementation; Data from the L. Heise, and others. 2013. “Can the UNAIDS Modes of First Eight States.” NACA, Abuja. Transmission Model Be Improved? A Comparison of the ———. 2014. “Global AIDS Response: Country Progress Original and Revised Model Projections Using Data from a Report; Nigeria GARPR 2014.” NACA, Abuja. Setting in West Africa.” AIDS 27 (16): 2623–35. NACO (National AIDS Control Organisation). 2013. “HIV Quinn, T. C., M. J. Wawer, N. Sewankambo, D. Serwadda, Sentinel Surveillance 2012–13: A Technical Brief.” Ministry C. Li, and others. 2000. “Viral Load and Heterosexual of Health and Family Welfare, New Delhi. Transmission of Human Immunodeficiency Virus Type 1.” Nagot, N., A. Ouangre, A. Ouedraogo, M. Cartoux, P. Huygens, New England Journal of Medicine 342 (13): 921–29. and others. 2002. “Spectrum of Commercial Sex Activity in Remme, M. I., A. Vassall, B. Lutz, and C. Watts. 2014. “Financing Burkina Faso: Classification Model and Risk of Exposure Structural Interventions: Going beyond HIV-Only Value for to HIV.” Journal of Acquired Immune Deficiency Syndromes Money Assessments.” AIDS 28 (3): 425–34. 29 (5): 517–21. Reza, T., D. Y. Melesse, L. A. Shafer, M. Salim, A. Altaf, and others. Nasirian, Maryam. 2011. Modeling of New HIV Infections 2013. “Patterns and Trends in Pakistan’s Heterogeneous HIV Based on Exposure Groups in Iran: Predictions for Planning. Epidemic.” Sexually Transmitted Infections 89 (Suppl 2): Saarbrucken, Germany: Lambert Academic Publishing. ii4–10. National AIDS Management Center. 2013. Ending AIDS in Rojanapithayakorn, W., and R. Hanenberg. 1996. “The 100% Thailand. Bangkok: Ministry of Public Health. Condom Program in Thailand.” AIDS 10 (1): 1–7. Needle, R., K. Kroeger, H. Belani, A. Achrekar, C. D. Parry, and SACEMA (South African Centre for Epidemiological Modelling S. Dewing. 2008. “Sex, Drugs, and HIV: Rapid Assessment and Analysis). 2009. “The Modes of Transmission of HIV in of HIV Risk Behaviors among Street-Based Drug Using South Africa.” SACEMA, Stellenbosch. Sex Workers in Durban, South Africa.” Social Science and Scorgie, F., S. Nakato, D. O. Akoth, M. Netshivhambe, P. Chakuvinga, Medicine 67 (9): 1447–55. and others. 2011. “I Expect to Be Abused and I Have Fear”: 176 Major Infectious Diseases Sex Workers’ Experiences of Human Rights Violations and USAID (U.S. Agency for International Development). 2011. Barriers to Accessing Healthcare in Four African Countries; Final “Condom Social Marketing: Rigorous Evidence; Usable Report. Cape Town: African Sex Worker Alliance. Results.” Research to Prevention 5, USAID, New York. Stoicescu, Claudia, ed. 2012. The Global State of Harm Van Damme, L., A. Corneli, K. Ahmed, K. Agot, J. Lombaard, Reduction 2012: Towards an Integrated Response. London: and others. 2012. “Preexposure Prophylaxis for HIV Harm Reduction International. Infection among African Women.” New England Journal Strathdee, S. A., T. B. Hallett, N. Bobrova, T. Rhodes, R. Booth, of Medicine 367 (5): 411–22. doi:10.1056/NEJMoa1202614. and others. 2010. “HIV and Risk Environment for Injecting- van den Berg, C., C. Smit, G. Van Brussel, R. Coutinho, and Drug Users: The Past, Present, and Future.” The Lancet M. Prins. 2007. “Full Participation in Harm Reduction 376 (9737): 268–84. Programmes Is Associated with Decreased Risk for Human Sullivan, P. S., A. Carballo-Diéguez, T. Coates, S. M. Goodreau, Immunodeficiency Virus and Hepatitis C Virus: Evidence I. McGowan, and others. 2012. “Successes and Challenges from the Amsterdam Cohort Studies among Drug Users.” of HIV Prevention in Men Who Have Sex with Men.” The Addiction 102 (9): 1454–62. Lancet 380 (9839): 388–99. van Griensven, F., A. Varangrat, W. Wimonsate, S. Tanpradech, Sullivan, P. S., O. Hamouda, V. Delpech, J. E. Geduld, J. Prejean, K. Kladsawad, and others. 2009. “Trends in HIV Prevalence, and others. 2009. “Reemergence of the HIV Epidemic Estimated HIV Incidence, and Risk Behavior among Men Who among Men Who Have Sex with Men in North America, Have Sex with Men in Bangkok, Thailand, 2003–2007.” Journal Western Europe, and Australia, 1996–2005.” Annals of of Acquired Immune Deficiency Syndromes 53 (2): 234–39. Epidemiology 19 (6): 423–31. Vassall, A., M. Pickles, S. Chandrashekar, M.-C. Boily, G. Shetty, Swaziland Ministry of Health. 2012. Swaziland HIV Incidence and others. 2014. “Cost-Effectiveness of HIV Prevention Measurement Survey (SHIMS): First Findings Report. for High-Risk Groups at Scale: An Economic Evaluation of Mbabane: Ministry of Health. the Avahan Programme in South India.” The Lancet Global Swaziland Ministry of Health and Social Welfare. 2002. Health 2 (9): e531–40. “Swaziland Behavioral Surveillance Survey (BSS).” Ministry Weiss, H. A., J. N. Wasserheit, R. V. Barnabas, R. J. Hayes, and of Health and Social Welfare, Mbabane. L. J. Abu-Raddad. 2008. “Persisting with Prevention: The Tanser, F., T. Bärnighausen, E. Grapsa, J. Zaidi, and M. L. Newell. Importance of Adherence for HIV Prevention.” Emerging 2013. “High Coverage of ART Associated with Decline in Themes in Epidemiology 5 (8): PMC2507711. Risk of HIV Acquisition in Rural KwaZulu-Natal, South WHO (World Health Organization). 2012. “Prevention Africa.” Science 339 (6122): 966–71. and Treatment of HIV and Other Sexually Transmitted Thigpen, M. C., P. M. Kebaabetswe, L. A. Paxton, D. K. Smith, Infections for Sex Workers in Low- and Middle-Income C. E. Rose, and others. 2012. “Antiretroviral Preexposure Countries: Recommendations for a Public Health Prophylaxis for Heterosexual HIV Transmission in Approach.” Department of HIV/AIDS, WHO, Geneva. Botswana.” New England Journal of Medicine 367 (5): ———. 2013. “Consolidated ARV Guidelines.” WHO, Geneva. 423–34. doi: 10.1056/NEJMoa1110711. http://www.who.int/hiv/pub/guidelines/arv2013/intro UNAIDS (Joint United Nations Programme on HIV/AIDS). /keyterms/en/. 2006. UNAIDS Report on the Global AIDS Epidemic 2006. ———. 2014. “WHO Progress Brief: Voluntary Medical Male Geneva: UNAIDS. Circumcision for HIV Prevention in Priority Countries of ———. 2012. UNAIDS Report on the Global AIDS Epidemic East and Southern Africa.” Progress Brief, WHO, Geneva. 2012. Geneva: UNAIDS. http://www.unaids.org/sites http://www.who.int/hiv/topics/malecircumcision/male / default/files/media_asset/20121120_UNAIDS_Global -circumcision-info-2014/en/. _Report_2012_with_annexes_en_1.pdf. ———. 2015. “Guideline on When to Start Antiretroviral ———. 2014a. Local Epidemics Issue Brief. Geneva: UNAIDS. Therapy and on Pre-Exposure Prophylaxis for HIV.” WHO, ———. 2014b. UNAIDS GAP Report 2014. Geneva: UNAIDS. Geneva. ———. n.d. “National AIDS Spending Assessment 2007–2010: Wilson, D., and S. Challa. 2009. “HIV Epidemiology: Recent The Kingdom of Swaziland.” UNAIDS, Geneva. Trends and Lessons.” In The Changing HIV/AIDS Landscape, ———. 2016. Prevention Gap Report. Geneva: UNAIDS. edited by E. L. Lule, R. M. Seifman, and A. C. David, 474. http://www.unaids.org/sites/default/files/media_asset/2016 Washington, DC: World Bank. -prevention-gap-report_en.pdf. Wilson, D., and D. T. Halperin. 2008. “‘Know Your Epidemic, UNAIDS and OPS (Organización Panamericana de la Salud). Know Your Response’: A Useful Approach, If We Get It 2009. “Modos de transmisión del VIH en America Latina: Right.” The Lancet 372 (9637): 423–26. Resultados de la aplicación del modelo.” OPS, Lima. Wilson, D. P., A. Yakusik, C. Kerr, and C. Avila. 2013. “HIV UNAIDS/WHO Working Group on Global HIV/AIDS and STI Resource Needs, Efficient Allocation, and Resource Surveillance. 2000. Guidelines for Second Generation HIV Mobilization for the Republic of Belarus.” UNAIDS, Surveillance. Geneva: World Health Organization and Joint Geneva. United Nations Programme on HIV/AIDS. Winkelstein, W. Jr., M. Samuel, N. S. Padian, J. A. Wiley, UNAIDS and World Bank. 2012. “New HIV Infections by W. Lang, and others. 1987. “The San Francisco Men’s Mode of Transmission in West Africa: A Multicountry Health Study: III; Reduction in Human Immunodeficiency Analysis, 2010.” UNAIDS and World Bank, Dakar. Virus Transmission among Homosexual/Bisexual Men, Tailoring the Local HIV/AIDS Response to Local HIV/AIDS Epidemics 177 1982–86.” American Journal of Public Health 77 (6): World Bank. 2012. “HIV/AIDS in India.” World Bank, 685–89. Washington, DC. Wood, E., T. Kerr, B. D. L. Marshall, K. Li, R. Zhang, and oth- Zohrabyan, L., L. G. Johnston, O. Scutelniciuc, A. Iovita, ers. 2009. “Longitudinal Community Plasma HIV-1 RNA L. Todirascu, and others. 2013. “Determinants of HIV Concentrations and Incidence of HIV-1 among Injecting Infection among Female Sex Workers in Two Cities in the Drug Users: Prospective Cohort Study.” British Medical Republic of Moldova: The Role of Injection Drug Use and Journal 338: b1649. Sexual Risk.” AIDS and Behavior 17 (8): 2588–96. 178 Major Infectious Diseases Chapter 9 Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling James G. Kahn, Lori A. Bollinger, John Stover, and Elliot Marseille INTRODUCTION Ensuring that available resources are allocated to the Resources devoted to combating the human immuno- most-cost-effective activities is essential to pursuing the deficiency virus and acquired immune deficiency syn- aspirational “Getting to Zero” goals of the Joint United drome (HIV/AIDS) have increased dramatically since Nations Programme on HIV/AIDS (UNAIDS): zero new 2005 (Dieleman and others 2014). However, the rate infections, zero AIDS-related deaths, and zero discrimi- of increase has slowed in recent years, even though the nation. Similar challenges also face global efforts to commitment required to serve all of those in need and to control tuberculosis and malaria—resources fall short reverse the epidemic has not been reached (Schwärtlander of ambitious prevention and treatment targets. and others 2011; UNAIDS 2013, 2014; WHO 2013). In Various effectiveness, cost-effectiveness, and resource addition, new recommendations to start treatment allocation models have been developed to evaluate the earlier in the disease course mean that more resources costs and outcomes of the choices facing HIV/AIDS will be needed than previously estimated. Many of the policy makers at national and international levels. This countries with the highest prevalence of HIV/AIDS have chapter presents an overview—including features, uses, low incomes and carry a heavy burden of other diseases, and limitations—of the small subset of models that and it is particularly important to deploy resources explores the allocation of HIV/AIDS resources across judiciously. Finally, efficiency is an even greater impera- many intervention options and purposes. It does not tive in the current era of transition away from funding assess the more numerous models that analyze the dominated by international donor aid toward a funding cost-effectiveness of one or a few interventions for one model in which the national governments in affected purpose. Accordingly, it assesses the set of software tools countries bear a larger portion of the costs; this is that portray a wide range of interventions and combina- especially so since, by some metrics, national govern- tions of interventions in different settings with the goal ments are failing to increase their own contributions of providing broad guidance for improved resource rapidly enough (Resch, Ryckman, and Hecht 2015). allocation. Corresponding author: James G. Kahn, University of California, San Francisco Institute for Health Policy Studies, San Francisco, California, United States; jgkahn@ucsf.edu. 179 GENERAL OVERVIEW OF THE ROLE OF ignoring myriad contextual factors that do not have an MODELS important effect on the decision at hand. They can also portray outcomes that are not empirically measur- What Are Models? able because of technical or time constraints, such as Three types of models are relevant to determining the long-term health outcomes and costs. Finally, they cost-effectiveness of interventions. Epidemic and disease offer a more explicit and rational alternative to other models use mathematics to describe the dynamics of approaches to decision making, such as guesses, inertia, disease acquisition or progression within individuals. political expedience, or ideology. Cost-effectiveness models combine epidemic and dis- The limitations of models reflect the challenge of ease models with a quantitative description of one or analyzing a decision with imperfect information. The more intervention activities typically aimed at alter- best models are parsimonious enough to be understand- ing a specific undesirable event (such as mother-to- able and buildable, yet adequately realistic to be policy child HIV transmission), estimating each intervention’s relevant. They are technically sophisticated but easy to cost and effectiveness in reducing morbidity or use. These competing demands confront modelers with mortality. Finally, resource allocation models consider trade-offs that are sometimes difficult to navigate wisely. multiple interventions and health events simultane- This is the art of modeling. Despite best efforts, the tech- ously and in various configurations to guide the divi- nical details of models are opaque to all but the most sion of effort and funding among different strategies. sophisticated users and sometimes even to them. This Often, those disparate health events are translated into a opacity can be mitigated with clear documentation. common disease burden metric, disability-adjusted life Finally, values for the required inputs can be imprecise years (DALYs).1 or biased. For example, efficacy data may be derived This chapter focuses on the third type, resource from programs in different settings or with modified allocation models, for several reasons. implementation. To understand the importance of these uncertainties, models rely extensively on sensitivity • First, in the field of global health, the most useful analyses, that is, assessments of how results change with models for decision makers provide information different input values. Fortunately, the basic findings of that is organized and presented to help them choose models are often robust to input uncertainties. courses of action that result in better population health outcomes. Resource allocation models are designed explicitly for this purpose. COMPARISON OF HIV/AIDS RESOURCE • Second, they incorporate the two other types of models ALLOCATION MODELS or practical simplifications of them. For policy makers, it is not essential to understand the individual inter- HIV/AIDS resource allocation models include the vention models because they often examine narrow OneHealth Tool, which contains the Goals model and technical issues that do not contribute meaningfully the Resource Needs Model (RNM) by Avenir Health; to more rational resource allocation across multiple Optima HIV (part of the suite of Optima models) by interventions. the Burnet Institute and the World Bank; the AIDS • Finally, limiting this chapter to resource allocation Epidemic Model (AEM) by the East-West Center; is practical. Considering the far more numerous Epidemiological Modeling (EMOD) by the Institute for epidemic, disease, and cost-effectiveness models and Disease Modeling; and Global Health Decisions (GHD) explaining their incompatibilities would be over- by the University of California, San Francisco. Each whelming for authors and readers alike. Table 9.1 model is best used as follows: provides a brief comparison of the models reviewed in this chapter. • Goals and RNM are widely used and supported by United Nations (UN) agencies and linked with the OneHealth Tool and other disease models for broad Strengths and Weaknesses of Models health sector planning. The process is moderately Resource allocation models, if thoughtfully structured intensive, although the models can be adapted to and populated with sound, current data, are able to specific purposes in easy-to-use formulations. quantify and logically assemble diverse factors relevant • Optima HIV is widely used, supported by the to program decisions in ways that would otherwise World Bank, and consistent with the Goals model. be impossible. They highlight and integrate policy- It uses an algorithm to optimize resource alloca- relevant data and dynamics from a complex world, tion across interventions and geography for a given 180 Major Infectious Diseases Table 9.1 Comparison of Models Goals, AIDS Impact Model (AIM), and Resource Needs Model (RNM) Epidemiological Modeling in Spectrum/OneHealth Tool Optima HIV AIDS Epidemic Model (AEM) (EMOD) Global Health Decisions Institutional Avenir Health (www.avenirhealth.org Optima Consortium East-West Center (http://www Institute for Disease Modeling University of California, home /software-spectrum.php) for Decision Science .eastwestcenter.org/research (www.idmod.org) San Francisco (www (www.optimamodel.com) /research-projects/hiv-policy .globalhealthdecisions.org) -analysis-research-and-training) Disease HIV/AIDS; Spectrum also includes HIV/AIDS. Optima Consortium HIV/AIDS HIV/AIDS; other EMOD HIV/AIDS scope family planning, STIs, tuberculosis, has models for TB; malaria; models address tuberculosis, Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling malaria, NCDs, maternal and child HCV; nutrition; and others. airborne respiratory infections health, and OneHealth (health systems). (especially TB), vector-borne diseases (especially malaria), and waterborne diseases (especially polio). Main use Widely used across all epidemic Widely used across all Used primarily in Asia to model Used primarily for research; Not widely used at country-level; types, supported by United Nations epidemic types; in partnership concentrated epidemics; used generalized, concentrated- most appropriate for generalized agencies; used in partnership with with in-country stakeholders in partnership with in-country FSW-based (but not MSM or epidemics. Simplified use in-country stakeholders to support to support national strategic stakeholders to support national IDV yet), or mixed epidemics; and easy interface to explore national strategic planning; linked planning. Is one of models strategic planning. Automatic fully implemented for limited impact and costs of intervention with OneHealth Tool for broad health supported by World Bank and optimization of resources across countries. Models MTCT and combinations, implementation sector planning. Automatic optimization PEPFAR for use in operations interventions is available. sexual transmission based on scales, and delivery modalities. of resources across interventions is and technical support to partnership patterns. Simulates Manual optimization of available. More intensive process for governments. Core feature is impact and cost-effectiveness resources across interventions. full model, but simplified version is algorithm to optimize resources of scaling-up, targeting, newly available. across interventions and and varied intervention geography toward strategic implementation. Automatic objective, subect to specified user-defined optimization constraints. More intensive across interventions. process for full model, but simplified version available. table continues next page 181 182 Table 9.1 Comparison of Models (continued) Goals, AIDS Impact Model (AIM), and Resource Needs Model (RNM) Epidemiological Modeling Major Infectious Diseases in Spectrum/OneHealth Tool Optima HIV AIDS Epidemic Model (AEM) (EMOD) Global Health Decisions Interventions HIV counseling and testing, linkage to care Prevention: Condom promotion/marketing; STI treatment, PMTCT, blood safety, behavior change programs, interventions for high-risk groups (for example, FSW, IDU), voluntary medical male circumcision, PrEP, vaccines Treatment and support: antiretroviral therapy, retention in care, opportunistic infection prophylaxis, support to people living with HIV/AIDS Evolving as recommendations change. Includes interventions listed Based on user-defined local best- Targeting by age, Subset only. immediately above and also practice prevention packages gender, location, time, innovative user-specified for key populations (FSW, risk, accessibility, interventions, including MSM, PWID, FSW who inject, sociodemographics. Individual- complementary service transgendered populations) and level variation over time in modalities, targeted and cross- population-specific antiretroviral intervention participation and sectoral interventions, and therapy, which often reflects efficacy. treatment retention. standard intervention list above. Geography National, subnational, or any level for which necessary data are available. National and subnational. National, subnational, and National, subnational, or any smaller. Allows age- and level for which necessary data gender-stratified migration are available. between geographic locations and populations. Population Adults 15-49 divided into subpopulations by gender, sexual behavior (eg FSW, clients, MSMs), and injecting drug use. Also by HIV disease groups and treatment status. Perinatally infected children. Standard groups. Default standard risk and age Above plus transgender Users define groups via traits Standard groups. groups. Users may define populations. Not children. of individuals, for example, unlimited number of population risks (sex behavior, condom groups as targets of chosen use, concurrent partnerships) interventions. and health care access (use of ART). Trait intensity can vary by individual within group. Time frame 100 years; by year as desired. Specified by user, by year as 1975 to 2050, by year. Specified by user; typically 20 years, by year. desired. monthly or yearly reporting from start of epidemic until 2050. Type of model Compartmental deterministic Individual stochastic Compartmental deterministic • Divides population into groups, outcomes reflect movement between groups each time period. • Each person is portayed, • Modest flexibility, low computational requirements. outcomes reflect random chance of change each time period. • Maximum flexibility, high computational requirements. table continues next page Table 9.1 Comparison of Models (continued) Goals, AIDS Impact Model (AIM), and Resource Needs Model (RNM) Epidemiological Modeling in Spectrum/OneHealth Tool Optima HIV AIDS Epidemic Model (AEM) (EMOD) Global Health Decisions Software • Data entry and storage: Spectrum • Microsoft Excel for data • Data entry and storage: Excel • Data entry: JSON or Excel. • Data entry: prepopulated, software in Windows; data can entry with background 2013 or 2016. • Database: COMPS platform adjust in web interface. be copied from other software, for calculations in Python. • Interventions specified in Excel (http://comps.idmod.org). • Intevention specified example, Excel. • Cloud-based graphical user workbook. with JAVA graphical web • Operation: clickable run • Parameters are specified within interface. • Custom Java interface for user file or command line, interface. Spectrum. interaction; custom Java code calculation in C++. • Computation: Google Go. • Calculations within Spectrum. for computation. User inputs Demographics: (population sizes by age range, gender, major HIV risks; birth & mortality rates); Disease prevalence: (HIV, AIDS, STI); Risk behaviors: (partnerships, condom use); HIV program coverage and cost Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling Relevant variables for 7 risk categories. Flexible population groups Historical trends for prevalence Fine detail by age/gender Default values set for selected Defaults available for 100 countries. defined by user. Defaults and behaviors, program /year. Migration between countries; calibration available; available for most countries. effectiveness, and cost by key pairs of geographic locations. values modifiable by user. If resource allocation analyses populations. Risk and health care access Intevention coverage by delivery then program-related cost stratification. Health care model and risk group. functions. process flow to define health care intervention access /update/drop-out/re-initiation. Selection of output units and strata. Access Online (free); training courses are available and support is provided. Available in conjunction with Online (free). Extensive support Online (free); coordinate with trainings. /collaboration required. contact person to determine Source code (www.github.com applicability for intended use /InstituteForDiseaseModeling and request support. /EMOD), tutorials and parameters (www.idmod.org /idmdoc), installer (www.idmod .org/software), and database linked to cloud computing resources (http://comps.idmod .org). Users National, local, and international planners (including government), researchers; and monitoring and evaluation Planners propose model Planners, researchers, officers. settings and scenarios to be monitoring and evaluation implemented by IDM team or officers. research collaborators. table continues next page 183 184 Major Infectious Diseases Table 9.1 Comparison of Models (continued) Goals, AIDS Impact Model (AIM), and Resource Needs Model (RNM) Epidemiological Modeling in Spectrum/OneHealth Tool Optima HIV AIDS Epidemic Model (AEM) (EMOD) Global Health Decisions Training Workshops lasting several days. Normally 3 workshops focused Tutorials and demonstration Ten-minute video to use with on data needs and collection; files online, half-day default values; further guidance preliminary model building; and introductory trainings in required to change input values. scenario building for policy and university classes and program planning. conferences; 1-4 weeks of on-site training for detailed projects; and 4-8 weeks to calibrate a new country or setting. Technical Minor TA is free, major TA support funded via various mechanisms By arrangement (free). By arrangement (free). assistance (UN agencies, bilateral agencies, foundations, national government, and so forth). Initial set-up Default values often available. Otherwise, 3-5 days Normally done as a national Three-hour workshop or Some default values available. time for data collation and entry. process, including extensive review webinar; inputs for new setting Otherwise, 2-3 days for data of historical sources of information. 2 months; modified inputs for collation and entry. As such, generally several months existing setting minutes. for data collation and trend analysis, then projections and scenarios prepared in about 3 days, normally with vetting by national experts. Updates typically done in less than a week. User manuals Technical documentation: http:// Available at: optimamodel Available with training or email • Online at http://idmod Online intro video http://fast avenirhealth.org/software-spectrum .com/user-guide request. .org/idmdoc/—Technical .wistia.net/embed/iframe .php. documentation, topic- /h08v1rkpvf?wvideo=h08v1rkpvf. Tutorials: https://www.youtube.com specific tutorials, parameter No written manual. /user/spectrummodel definitions, and output file “data dictionary.” Analysis time Run time 1-3 minutes. Calibration several hours (if needed). Run time 1-3 minutes. Calibration Days to months, depending on Run time < 1 min. Calibration several days. complexity, from conception of hours. modeling question to results. Some standard analyses run in <1 hour. table continues next page Table 9.1 Comparison of Models (continued) Goals, AIDS Impact Model (AIM), and Resource Needs Model (RNM) Epidemiological Modeling in Spectrum/OneHealth Tool Optima HIV AIDS Epidemic Model (AEM) (EMOD) Global Health Decisions Outputs All outputs overall and by population group, by year & cumulative. By intervention scenario. Same as Goals and Optima Same as Goals and Optima Number in group, deaths, HIV/AIDS new cases and prevalence. models, plus health care use; models. Intervention participation; ART prevalence by CD4 count and stage of treatment. DALYs or QALYs. relationship and transmission Costs by interventions/care. Cost-effectiveness ratios. Optimal allocations. network over time; biomarkers such as CD4. Outputs also available as distributions. Sample Sample outputs are available online at Sample outputs are available Sample outputs are available Sample input and output files Sample outputs are outputs www.avenirhealth.org. online at www.optimamodel online at www.eastwestcenter and graphing/analysis scripts available online at www .com. .org/research/research-projects available online at www.idmod .globalhealthdecisions.org. /hiv-policy-analysis-research-and .org/software. Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling -training. Special Integrated into OneHealth Tool, Constrained optimization of Customized fit of behavioral Maximum flexibility to portray User-friendly graphical user features permitting cost and impact comparisons resource envelopes using trends to observed epidemiologic individual variation. Users can interface, direct intervention across many health sectors; Goals algorithms. Flexible group trends by adjusting transmission specify health system and care scenario comparisons. Express offers a simpler version. and intervention definition. probabilities and cofactors; flow. Modeling of multiple diseases, use of local data on program allowing cost and impact effectiveness. comparison. Optima Lite has simpler preloaded and calibrated projects. Limitations Only one adult age group (15-49) in As with all compartmental Only concentrated epidemics; Not yet implemented: viral No PrEP or vaccines; calibration Goals although outputs are available by models, individual-level data aggregate age structure based load testing and relationship to countries limited to date. five-year age group in AIM. included in aggregated form on age pattern of new male and between VL, age, and survival; No formal rollout. and homogeneity assumptions female infections. evolution and transmission of apply within specified modeled antiretroviral drug resistance; groups. transmission among MSM and IDU. Contacts John Stover (Jstover@AvenirHealth.org) David Wilson (info Tim Brown (tim@hawaii.edu) Anna Bershteyn (abershteyn Jim Kahn (jgkahn@gmail.com) @optimamodel.com) @idmod.org), Daniel Klein (dklein@idmod.org) Note: AIDS = acquired immune deficiency syndrome; ART = antiretroviral treatment; DALY = disability-adjusted life year; FSW = female sex worker; HCV = hepatitis C virus; HIV = human immunodeficiency virus; IDM = Institute for Disease Modeling; IDU = injecting drug user; MSM = men who have sex with men; MTCT = mother-to-child transmission; NCD = noncommunicable diseases; PEPFAR = U.S. President’s Emergency Plan for AIDS Relief; PMTCT = prevention of mother to child transmission; PrEP = preexposure prophylaxis; PWID = people who inject drugs; QALY = quality-adjusted life year; STI = sexually transmitted infection; TA = technical assistance; TB = tuberculosis; UN = United Nations. 185 objective, subject to budgetary, logistical, ethical, and across many areas of the health sector. Most of these mod- political constraints. els are assembled within Spectrum, a suite of integrated • AEM is used for studying concentrated epidemics, software models that provides policy makers with an especially in Asia. The task of calibrating and populat- analytical tool to support the decision-making process; ing the model is intensive, and exploring the various it is also widely known by its overlay, the OneHealth available packages encourages stakeholders to under- Tool. Many of the models also exist as Excel-based models stand local epidemics and the effectiveness of past, and web-based tools. This chapter focuses on models present, and potential future responses. useful for resource allocation for HIV/AIDS: the AIDS • EMOD is used to examine policy issues involving the Impact Model (AIM), Goals, and RNM, in particular. interplay of demographics, risks, disease progression, Spectrum is a system of policy models that support and health care. Individual-based modeling captures analysis, planning, and advocacy for health programs.2 heterogeneity and permits a nuanced portrayal— The models are used to project future needs and exam- for example, HIV/AIDS transmission is based on ine the effects of policy choices, including the impact of independent risk per sex act within partnerships that taking actions now rather than later, evaluating the costs evolve over time, and disease progression depends on and benefits of a particular policy, examining the inter- age. Health system dynamics (for example, cycles of relatedness of different policy decisions, and evaluating antiretroviral use) reflect factors such as age, gender, how a change in age and sex distribution can affect a geography, and risk. Of the models reviewed, it is the wide range of social indicators. most computationally intensive and has the most The central impact model in Spectrum is DemProj, sophisticated population portrayal and calibration. which projects the population for an entire country or • GHD is relatively simple to use, with an interface region by age and gender based on assumptions about that makes exploring the costs and effects of various fertility, mortality, and migration. A full set of demo- combinations of interventions and delivery modal- graphic indicators can be displayed for up to 100 years ities easy. It is not in widespread use, but is available into the future; urban and rural projections can also be for appropriate applications by arrangement with the prepared. Default data needed to project population are owners. GHD and Optima are exploring a collabora- provided from the estimates produced by the Population tion to incorporate key GHD features into Optima HIV. Division of the United Nations. Models not related to HIV/AIDS include FamPlan, which projects family The choice of model for a specific user depends planning requirements; Resources for the Awareness on the user’s needs and the models’ intended uses, of Population Impacts on Development, which projects strengths, and limitations. As presented in table 9.1, the social and economic consequences of high fertility some models are uniquely well suited to specific pur- and rapid population growth; Tuberculosis Impact poses, such as EMOD for detailed simulation of how Model and Estimates, which performs epidemiological individuals’ characteristics affect use of health care, and cost-effectiveness analysis of tuberculosis control and Avenir Health’s OneHealth Tool for placing HIV/ strategies; Lives Saved Tool, which estimates the cost and AIDS programming in the context of the broader health impact of scaling up child and maternal health interven- system. When models serve similar purposes, such as tions on mortality; and NonCommunicable Diseases, Avenir Health’s and Optima’s resource allocations across which calculates the impact of scaling up interventions HIV/AIDS interventions, users may want to consult the on populations affected by noncommunicable diseases. contacts for each model to discuss how comprehen- The four models related to HIV/AIDS interact with sively and efficiently each model can address users’ one another. AIM uses the Estimation and Projection needs. Annex 9A provides a list of country applications Package (EPP) module developed by the East-West for the models. Center to fit prevalence and incidence trends to surveil- A more technical comparison and assessment con- lance and survey data and then calculates the conse- ducted in early 2015 by the HIV Modeling Consortium quences of these trends for key indicators such as new (HIV Modeling Consortium 2015) was used to inform infections, deaths, need for treatment, and number of this review, although it is now slightly outdated because orphans. RNM calculates the costs associated with HIV- of ongoing model improvements. related interventions. Goals simulates HIV/AIDS inci- dence on the basis of behaviors and estimates the epidemiological effects of biomedical interventions Avenir Health Models and behavioral interventions (using an impact matrix) Over the past 40 years, Avenir Health, formerly known as to calculate infections averted and cost-effectiveness ratios. the Futures Institute, has led the development of models The Lives Saved Tool evaluates the cost and impact 186 Major Infectious Diseases of child and maternal health interventions, including The model is continuously updated to reflect the most HIV/AIDS and malaria, using inputs from AIM. recent research. AIM RNM AIM began as a relatively simple Excel-based tool RNM grew out of efforts developed in 2001 for the first developed in 1991 in collaboration with Family Health United Nations General Assembly Special Session on International under the AIDS Technical Support and HIV/AIDS to estimate the global resources required to AIDS Control and Prevention projects funded by the combat HIV/AIDS (Schwartländer and others 2001); the U.S. Agency for International Development (USAID). estimates are referred to as the Global Resource Needs The program has been revised several times since then Estimates (GRNE). Although that first Excel-based in collaboration with the UNAIDS Reference Group model was calculated at the individual country on Estimates, Models, and Projections. Since 2009, level, it was a global model and not appropriate for it has been maintained and updated with support from country-level use. After the first few rounds of the GRNE, the Bill & Melinda Gates Foundation and UNAIDS. in 2007 UNAIDS initiated a consultative process with It has evolved to become a comprehensive model countries with high burdens of HIV/AIDS to validate within Spectrum used to estimate the impact of the their country-specific portions of the GRNE, which HIV/AIDS epidemic. Several years ago, the Estimation required adapting the global model to the country level. and Projection Package (EPP) was incorporated into By 2009, the consultative process reached 60 countries, Spectrum. Both incidence and prevalence curves are and countries began to use RNM (still in Excel) for their now estimated within AIM, which then projects the own planning purposes. Because of this, RNM gradually consequences of the epidemic, including the number migrated over to Spectrum and now is used to calculate of people living with HIV/AIDS, new infections, and the funding required to expand national responses to deaths by age and gender, as well as the number of new HIV/AIDS. It estimates the costs of implementing HIV/ cases of tuberculosis and the number of orphans. Many AIDS programs, including the costs of care and treat- of these results are then used in other models in ment, prevention, and policy and program support. Spectrum. UNAIDS uses AIM to make the national RNM projects the costs of various interventions, and regional estimates it releases every two years. given assumptions about the size of various population The major inputs and outputs of AIM are as follows: groups, unit costs of interventions, and coverage targets Demographic projections are based on user inputs or (figure 9.1). Costs can be calculated from any perspective, projections prepared by the United Nations Population including provider, public, patient, and societal, depending Division. The projections start with an estimate and on the perspective of the data that are provided. projection of adult HIV/AIDS incidence, which is A significant portion of the model application process, combined with information on the age and gender described in more detail below, involves obtaining reli- distribution of incidence and progression to death to able cost data. RNM’s projections can then be used to estimate the number of new infections in adults, by age enhance knowledge of HIV/AIDS among policy makers and gender. New infections in infants are estimated and to build support for effective prevention, treatment, from prevalence among pregnant women and the rate care, and mitigation. The projection results are usually of mother-to-child transmission, which is dependent transferred to software, such as PowerPoint, for presen- on infant feeding practices and the coverage of preven- tation to leadership audiences. tion with antiretroviral agents. New infections progress RNM estimates the number of people receiving each to lower CD4 cell counts and are subject to HIV/AIDS- service by multiplying the number of people needing related mortality. Persons who receive first-line antiret- the service by the coverage rate (percentage of persons roviral therapy (ART), second-line ART, or both live needing the service who actually receive it). The resources longer than those who do not. People at any stage are needed are then estimated by multiplying the number subject to other-cause mortality at the same rates as of people receiving the service by the unit cost of people who are not infected. Adult deaths result in providing it. Before RNM can be used, both a demo- orphans. graphic and an HIV/AIDS projection must be prepared. In addition to estimating the epidemic and projecting The epidemiology section of AIM calculates the number its impacts, AIM has other features, including the ability of HIV/AIDS infections, persons needing treatment, to validate its estimates by comparing AIM outputs and orphans. This information is used in the treatment with other data sources, to perform uncertainty analyses section to calculate the costs of treatment for preventing for certain output variables, and to aggregate projections, mother-to-child transmission, HIV/AIDS, and associated for example, a series of subnational projection files. tuberculosis and opportunistic infections and can be Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 187 Figure 9.1 Structure of RNM: DemProj and AIM DemProj AIM Age- and HIV-positive population, sex-specific treatment populations, populations orphans Key Unit Resources × × Coverage populations costs required Program support Note: AIM = AIDS Impact Model; RNM = Resource Needs Model; DemProj = central model in Spectrum. used in the mitigation section to calculate the cost of incidence and prevalence and improved coverage of providing services for orphans. AIM modifies the demo- treatment, care, and support programs. It does not cal- graphic projection through HIV/AIDS deaths and the culate the optimum pattern of allocation or recommend impact of HIV/AIDS on fertility. a specific allocation of resources between prevention, care, and mitigation, although an optimization routine is Goals available. Sexual mixing is random within risk groups. The Goals model supports efforts to respond to the HIV/ Mixing between risk groups is limited to low-risk adults AIDS epidemic by showing how the amount and alloca- who can have partners from higher-risk groups. Extensive tion of funding is related to the achievement of national literature underlies both the impact matrix coefficients goals, such as the reduction of prevalence and expansion and other model parameters; these sources are well doc- of care and support. It also explores the impact of poten- umented in the manual. The Goals model underwent tial vaccines. The Goals model evolved out of an effort to an external validity check comparing 12 mathematical identify what program managers need to plan effectively. models; results were basically consistent, particularly Stover and Bollinger (2002) surveyed 14 national pro- in the short term (Eaton and others 2013). A recently gram managers and learned that their most challenging formed Models for Policy Planning Reference Group, issue was using cost-effectiveness information in their led by the HIV Modelling Consortium (http://www countries’ key priority-setting exercise, the National .hivmodelling.org), will be providing ongoing internal Strategic Plan. The model was developed to be used in and external validity checks. that process. The Goals model is a compartmentalized model, The Goals model is intended to support strategic modeling heterogeneity by dividing the adult population planning at the national level by providing a tool to link ages 15–49 years by gender and risk group: not sexu- program goals and funding. It can help answer several ally active, low-risk stable couples, medium-risk people key questions: engaging in casual sex, sex workers and clients, men who have sex with men, and people who inject drugs • How much funding is required to achieve the goals of (figure 9.2). The model calculates new infections by the strategic plan? sex and risk group as a function of behaviors and epi- • What goals can be achieved with the available resources? demiological factors such as prevalence among partners • What effect do alternate patterns of resource alloca- and stage of infection. The risk of transmission is deter- tion have on the achievement of program goals? mined by behaviors (number of partners, number of contacts per partner, and condom use) and biomedical The Goals model does not provide all the answers. factors (use of antiretroviral agents, male circumcision, It is intended to assist planners in understanding the prevalence of other sexually transmitted infections). effects of funding levels and allocation patterns on pro- Interventions can change any of these factors and affect gram impact. The model can help planners understand the future course of the epidemic. Interventions with how funding levels and patterns can lead to lower either a behavioral or biomedical effect on HIV/AIDS 188 Major Infectious Diseases Figure 9.2 Structure of the Goals Model Treatment Population group Type of transmission Low risk Sex Probability Medium risk of New HIV Behavior Needle infections High risk transmission Blood MSM IDU MTCT Coverage effectiveness Behavior change Technology Condoms Outreach STI Tx MC Education PrEP Communication Microbicides Vaccines Source: Avenir Health. http://avenirhealth.org/software-spectrum.php. Note: IDU = injecting drug user; MC = male circumcision; MSM = men who have sex with men; MTCT = mother-to-child transmission; PrEP = preexposure prophylaxis; STI Tx = sexually transmitted infection treatment. transmission are modeled, including behavior change The Goals model has been used to assess the impact through outreach; education and communication inter- of prevention and treatment at the global level (Eaton ventions; and biomedical interventions such as con- and others 2013; Schwartländer and others 2011; Stover dom distribution, voluntary medical male circumcision and others 2006) and for more than 30 applications at (VMMC), ART, preexposure prophylaxis, microbicides, the country level (see annex 9A for a list of countries). and vaccines. The effect of interventions on behaviors is modeled DMPPT through an impact matrix that summarizes the impact One of the Excel-based tools developed by Avenir Health literature to describe changes in behavior by risk group is a two-part cost and impact tool available for examining as a result of exposure to behavior change interventions the effects of VMMC; the most recent version of the (Bollinger 2008). The Goals model is then linked impact model is called Decision Makers’ Program Planning to the AIM module in Spectrum to calculate the effects Tool (DMPPT) 2.0 (http://www.malecircumcision.org). on children (ages 0–14 years) and adults older than The first DMPPT was used to estimate the costs and age 49 years. The AIM module also includes the effects impact of VMMC in many countries for adult males ages on pediatric infections of programs to prevent mother- 15–49 years. When experience showed that most VMMC to-child transmission. clients were under age 25 years, a second version of the Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 189 model was developed to evaluate the impact of targeted providing medical interventions, as well as for most coverage of VMMC services (Stover and Kripke 2014). facilities offering support functions. It also includes This tool is being developed for the web. planning functions for equipment, furniture, vehicles, and communications. OneHealth Tool • Governance. The governance module includes costing The OneHealth Tool is a series of modules overlaid templates for assessing the costs of governance on the impact models of Spectrum. It is intended for activities. medium-term strategic health planning (3–10 years) at • Logistics. The logistics module allows for the planning the national and subnational levels. The OneHealth Tool of warehouses and vehicles needed to move com- was developed by a group of UN agencies, the World modities or drugs and supplies from central ware- Bank, and the World Health Organization (WHO) in houses to the endpoints of a logistics system. response to requests made during a 2008 technical con- • Health financing. The health financing module is sultation in Senegal by countries looking for standard- used to estimate the costs of implementing health ized costing tools. The model builds on the International financing programs such as vouchers, subsidies, or Health Partnership and Joint Assessment of National cash transfers. Health Strategies and Plans framework, and experts in • Health information systems. The health information costing from all participating UN agencies contributed systems module includes templates for assessing the both fund and staff time to the technical development costs of implementing a health information system. of the model. The project also received funds from the • Budget mapping. The budget mapping module can Global Fund to Fight AIDS, Tuberculosis, and Malaria; be used to allocate intervention and health system the Global Health Workforce Alliance; and the Health costs across budget categories established by the user, Metrics Network, as well as from bilateral development to match country or international institution cost agencies. categories. The OneHealth Tool was developed because most • Financial space. The financial space module is used costing tools at the time took a disease-specific approach to analyze the financial space, including both public rather than a health systems approach (figure 9.3). In and private health expenditures, within which health addition to covering public sector health interventions at plans are expected to be executed. both national and subnational levels, it incorporates coverage of private sector interventions and includes selected nonhealth interventions that may have health Optima impacts. It is a unified tool for planning, costing, impact Optima HIV is a software package and modeling tool analysis, and financial space analysis performed jointly developed by the Optima Consortium for Decision and can be implemented at either the health system or Science in collaboration with the World Bank. It is one program level. The OneHealth Tool provides a way to of a suite of models for different disease areas that have estimate the cost and impact of interventions for HIV/ been developed by the Optima Consortium, all of which AIDS, tuberculosis, and malaria simultaneously, as well are designed to help national decision makers, program as other diseases, and to examine the resource require- managers, and funding partners achieve allocative effi- ments from the health system. Default costs from a vari- ciency and plan for financial sustainability. This is done ety of sources are available, but should be validated and by applying the Optima approach, a framework for can be subsequently modified by the user. Sources of informing public health investment choices that consists cost data include the Management Sciences for Health of the following core steps: International Drug Price Indicator Guide; the UNAIDS Global Price Reporting Mechanism; Gavi, the Vaccine • Assess the burden of disease over time, for each Alliance; and UNICEF. population group, and for each disease sequelae or The OneHealth Tool includes the following modules: state through data synthesis and epidemiological modeling. • Human resources. The human resources module • Specify the efficacy and effectiveness of interventions allows salaries, benefits, and incentives for health ser- (including different modes of delivery) that have vice providers and health management and support the potential to reduce incidence, morbidity, and personnel to be costed, along with preservice training mortality. and nonspecific in-service training. • Assess the costs required to deliver services at different • Infrastructure. The infrastructure model deals with levels of coverage, including through different service planning and costing functions for all facilities modalities and implementation or efficiency options. 190 Major Infectious Diseases Figure 9.3 Structure of the OneHealth Tool Fiscal space envelope HS4. Health HS5. Governance HS6. Financing information and leadership policies National Disease programs Service delivery planning Health system envelope Hospital Reproductive health Immunization Child health Health center Nutrition Malaria WASH Others NCDs HIV TB Outreach Community HS1. Infrastructure HS2. Human HS3. Logistics and equipment resources Source: World Health Organization (http://www.who.int/choice/onehealthtool/OneHealth_Tool_Supporting_integrated_strategic_health_planning.pdf?ua=1). Note: HIV = human immunodeficiency virus; HS = health system; NCD = noncommunicable diseases; TB = tuberculosis; WASH = Water, Sanitation, and Hygiene for All. • Define strategic objectives and national priority an optimal allocation of HIV investments. Optima HIV is targets—as well as the budgetary, logistical, ethical, the only quantitative tool currently available in the HIV and political constraints related to achieving these field that includes a formal mathematical optimization objectives—across the entire population and by routine, real-world budgetary, logistical, and political con- disease. straints, and economics of scaling up intervention • Use a formal mathematical optimization algo- programs and responses. rithm around the constructs from the previous Optima HIV is intended to address various policy steps to assess the optimal allocation of a given questions: level of resources to reduce disease burden, subject to the defined constraints. • How close is achievement of the National Strategic Plan targets under current funding? Over the strategic plan Optima HIV is a software package designed to imple- period, how close will the country get to its disease- ment the steps listed above. It consists of a mathematical related targets (a) with the current volume of funding model of disease transmission and progression, a module allocated according to current expenditure and (b) with for defining interventions and cost functions, and a math- the current volume of funding allocated optimally? ematical optimization module that integrates the epi- • How much funding is required to achieve the National demic, programmatic, and cost data in order to determine Strategic Plan targets? Over the strategic plan period Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 191 or over a longer period, according to current program Epidemiological Modeling (EMOD) implementation practices and costs, how much total The Institute for Disease Modeling3 developed the funding is required to meet the targets, and how is EMOD software primarily for use by disease model- this funding optimally allocated between programs? ers, researchers, epidemiologists, and public health • What benefits can be achieved with more efficient professionals seeking to simulate infectious disease implementation? conditions and evaluate the effectiveness of eradication • What impacts have past programs had? How would or mitigation approaches. The model is agent based, the country’s HIV/AIDS trajectory have changed if that is, portraying each individual rather than aggre- investment had not occurred in different programs, gate group behavior; in discrete time, that is, calculat- and what is the estimated cost-effectiveness of the ing transmission risk and other processes in small but past response? noncontinuous time steps; and using a Monte Carlo • What is the expected future impact of policy or pro- simulator to predict populations, that is, drawing gram implementation scenarios? What is the projected many random samples from a specified probability future trajectory of the country’s epidemic with and distribution for each input. This agent-based without investment in specific programs or with approach is computationally intensive as opposed to and without attaining program-specific targets? the fast speeds normally achieved with compartmen- tal models (whether dynamic [using differential and Optima HIV extends allocative efficiency analyses to integral equations] or in discrete time steps). The (a) include geographic prioritization and (b) integrate advantage is the ability to portray individual charac- technical efficiency within allocative efficiency, consid- teristics and transitions over time much more pre- ering the various modalities of service delivery for dif- cisely. For example, the risk of infection can reflect a ferent programs. As such, it addresses the following large set of person-specific risk factors, such as type of questions: Which service delivery modalities and mech- risk behavior and frequency, type of protective behav- anisms should be implemented in which geographic ior and consistency, geographic location, and interac- areas? How should the HIV/AIDS response prioritize tions with other individuals in the same and other investment across population groups and geographic risk groups—all of which can and do vary over time. areas, and which service delivery modalities and mech- The complex overall EMOD architecture provides anisms should be implemented and to what extent in disease transmission projections for environmental, each area, to get as close as possible to national targets sexual, vector-based, and airborne diseases and may with available resources? Additional descriptions of be adapted to support additional infectious diseases. the uses of Optima HIV for planning a national The binary software or source files are available for response are available in chapter 8 of this volume download. Data and training requirements are (Wilson and Taaffe 2017). substantial. AIDS Epidemic Model The AEM, developed in the early 1990s, is patterned after Global Health Decisions the HIV/AIDS situation in countries with concentrated The GHD model was developed by the University of epidemics, primarily in Asia. It allows countries to build California, San Francisco, to provide an HIV/AIDS locally tuned models that accurately represent their resource allocation model with a sophisticated and flex- epidemiological situations. These models can then be ible user interface prepopulated with epidemiologic and used with a set of analytic tools—the AEM workbooks programmatic data. The goal was to permit relatively (baseline, intervention, and impact analysis)—to prepare rapid but nuanced allocation of resources across popula- scenarios that analyze alternative responses to the tions and interventions. epidemic, assess the impact of these responses, and esti- A website allows users to specify a country from mate the cost of implementation. among those implemented, verify the default input These scenarios provide essential inputs into national values (for HIV/AIDS prevalence and use of ART, strategic planning processes, help countries allocate their for example), alter the values as needed, and then run resources more efficiently, and help countries identify a series of tailored intervention scale-up scenarios. weaknesses that must be addressed to strengthen their The results of each scenario are incidence, prevalence, responses. Using the AEM is an intensive process that deaths, and costs, by risk group, over time. These scenario builds stakeholder involvement in and ownership of results are stored and can be named and compared the planning process. graphically. 192 Major Infectious Diseases The back end is a deterministic compartmental model Cochrane Library, with others identified through with five risk groups (general population female, general PubMed and other sources. The evidence at this level population male, sex workers, drug users, and men is massive, diverse in form, and technically complex. who have sex with men), implemented in Google Go. • The second step is to select potentially relevant Given a set of predictions for treatment and prevalence reviews based on whether the information could affect in future years, the model uses simulated annealing—a major decisions on policy or funding, such as whether stepwise statistical sampling approach—to align model and at what scale to support a particular intervention. predictions with these benchmark projections. Important but narrower questions, such as drug dos- The model provides tiered access to functionality, ing or comparisons between very similar intervention including the use of country-specific defaults for input designs, are usually excluded, as are universally accepted values (for example, demography, epidemiology, inter- practices. All of these decisions are documented. ventions, and costs), real-time adjustment of interven- • The third step is to extract information from the tion portfolios, and manipulation of input values by selected comparisons, including context (for example, more technically informed users. Policy makers have country and type of population), research methods not used GHD. (for example, study design and outcome measures), and quantitative findings on efficacy. WHAT WORKS REVIEWS • The fourth step is to rate the strength of evidence based on the quantity and type of studies, as well A central function of policy modeling is to convey the as the precision of findings, that is, the width of the impact of interventions on health and economic outcomes. relative risk confidence interval. The result is a sum- This means that resource allocation models need to incor- mary table that presents the intervention comparisons, porate the latest evidence on intervention efficacy in chang- findings (for example, mortality and incidence), ing risks and risk behaviors. Systematic reviews of efficacy relative risk reduction, and strength of evidence for are now commonplace, but overwhelming in number and each review and study. complexity. A distilled review that conveys efficacy and • The next step, which is critical, is to combine evidence associated strength of evidence can be helpful for informing by intervention type where possible. For example, modeling and educating decision makers about the evi- if different insecticides for environmental control of dence. Thus, the GHD project initiated an activity called a disease vector (for example, a mosquito) all work What Works Reviews (WWR) in 2010 to address a per- with similar efficacy, the findings are combined into ceived gap in the availability of information about inter- a single row. All summary data are linked to orig- vention efficacy for policy discussions and models. inal extractions to allow review of the aggregation WWR translates empirical evidence on the effects decisions. of interventions into a quantitative synthesis that is • The last step is to consult with subject area experts to technically accurate while being concise and accessible review provisional findings. This step may result in to nontechnical audiences. Each estimate of efficacy the addition of new reviews or studies or adjustment is accompanied by a strength-of-evidence rating that of the interpretation of existing evidence. reflects the quantity and type of underlying studies. WWR examines both prevention and treatment for each Figure 9.4 Structure of the What Works Reviews Process health condition, with a focus on data with the most potential relevance for policy and an emphasis on health Concise, Data outcomes (for example, deaths and disease incidence) distillation consistent, accessible process Consult rather than process measures (for example, satisfaction Present key with findings experts with services or adherence). Combine evidence WWR includes nearly 50 categories of interventions by intervention type for HIV/AIDS, including some found to be ineffective.4 Rate strength of evidence (quantity and type of studies, precision) Methods Extract context, method, and quantitative findings for efficacy WWR proceeds in explicit and small steps from existing Select potentially policy-relevent systematic reviews and important new studies to key reviews and comparisons Massive, findings (figure 9.4). Search for systematic reviews and pivotal new diverse, studies for prevention and treatment for a technical health condition • The first step is to search for systematic reviews and pivotal new studies. Most reviews come from the Source: What Works Reviews (http://globalhealthdecisions.org/wwr/). Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 193 The key outcome is relative risk reduction. This is a in annex 9B. To illustrate results, this section summa- standardized metric, designed to put diverse outcome rizes the findings for biological prevention strategies. metrics (for example, odds ratio, means) onto a consis- Circumcision of adult males is 70 percent effec- tent footing (Mirzazadeh, Malekinejad, and Kahn 2015). tive in reducing transmission from females to males It equals the percentage reduction in the risk of negative based on three RCTs, other studies, and long follow-up health outcomes and can be used for mortality, morbidity, (very strong evidence). Evidence for men who have and indirect health indicators. sex with men and transgender individuals suggests Findings are presented in three parts: little if any protection from male circumcision (strong evidence). Treatment of sexually transmitted infec- • A key findings table has a row for each type of inter- tions has been examined in eight studies, with a vention, with the relative risk reduction and strength 12 percent non–statistically significant reduction in of evidence for mortality, morbidity, and other incidence and a wide confidence interval, including indicators. a negative effect (−49 percent to 48 percent), with • An overview reviews the health condition, epidemi- lower incidence of sexually transmitted infections and ology, key findings, and future directions. risk behaviors (16 percent to 23 percent, moderate • A logic model graphically represents modes of disease strength evidence). Nonoxynol-9 and microbicides acquisition and progression as well as the location of failed to reduce HIV/AIDS incidence. Data on micro- intervention opportunities. bicides containing an antiretroviral drug suggest a 37 percent reduction in HIV/AIDS incidence (weak evidence). Vaccines did not work, with exception of Strength of Evidence one trial with 30 percent efficacy, and neither did the WWR rates strength of evidence on a scale of 0–6 latex diaphragm. Preexposure prophylaxis (PreP) with (visually represented by bars). The score is based on the the antiretroviral combination tenofovir plus emtric- extent and type of studies (for example, randomized itabine reduced HIV/AIDS transmission in several controlled trials [RCTs]), quality of available systematic RCTs by 47 percent (very strong evidence). Two trials reviews, and precision (that is, narrowness of uncer- found no effect, due to low sample size and adherence. tainty bounds). The following is the typical evidence Use of antiretrovirals reduced incidence by 96 percent associated with each score: in serodiscordant couples in a large RCT in Africa, 6 = three or more RCTs, well reviewed, good precision with similar results from several earlier non-RCTs (very strong) (strong evidence). The female condom reduced the nonuse of condoms. 5 = three or more RCTs, minor problems with review or precision (strong) 4 = two RCTs, good review and precision (moderate FIELD EXPERIENCE WITH MODELS strength) INFLUENCING POLICY DECISIONS 3 = one RCT or multiple non-RCTs, good review and A typical model application associated with develop- precision (moderate strength) ing a National Strategic Plan must be integrated across 2 = one RCT or multiple non-RCTs, problems with a number of different dimensions: models used, stake- review or precision (weak) holders involved, capacity building, and dissemina- tion activities. When an application begins, the first 1 = one or more non-RCTs, serious problems with step should be to establish a stakeholder group that review or precision (very weak) will provide input throughout the process. Evaluating 0 = no evidence, because of lack of studies or extreme the current status of data available is necessary to imprecision. inform and design the process to be undertaken; some countries may have facility-based costing infor- mation for certain interventions, such as HIV/AIDS Application to HIV/AIDS counseling and testing or ART, but not for other The HIV/AIDS component of WWR was updated with interventions. new literature searches and extractions between The models to be applied then need to be ascer- December 2015 and January 2016. Figure 9.5 presents the tained; many of the models described in this chapter logic model for the broad context of HIV/AIDS interven- are integral and required in the National Strategic tion. Key findings for all intervention types are presented Plan. For the Spectrum models, a country in 194 Major Infectious Diseases Figure 9.5 Logic Model for the HIV/AIDS Component of What Works Reviews HIV Disease acquisition Disease progression Behavioral (for example HIV/AIDS counseling and testing) Biological (for example circumcision, antiretrovirals) Sexual activity Antiretroviral therapy Providing clean needles Antimicrobials (prophylaxis and treatment) Behavioral (for example, peer) Nutritional supplements Treating drug dependency HIV Needle or syringe Death infection use Antiretrovirals Disease process: Mother to child Other interventions Immune decline (falling CD4) exposure in utero, opportunistic infections perinatal, AIDS diagnosis breastfeeding Key Risk factor Intervention Outcome Source: Global Health Decisions (http://globalhealthdecisions.org/wwr/hivaids/hivaids-key-findings/). Note: HIV/AIDS = human immunodeficiency virus; acquired immune deficiency syndrome. Sub-Saharan Africa would need to apply the to allocating available resources. The steps include AIM model (to estimate the need for treatment the following: and prevention of mother-to-child-transmission ser- vices), the RNM (to estimate the resources required to • Identify and meet with national planning officials scale up from current coverage to future levels of and local consultants and perform situation analysis. desired coverage), and the Goals model (to estimate • Collect facility-based data and other data. the impact of various scale-up and resource allocation • Set up models. strategies). The resources required for universal access • Present initial results to national planning team are compared with an assessment of the resources • Revise initial analysis, as required. likely to be available based on National AIDS Spending • Present results to stakeholders and conduct prioriti- Assessments, to assess the size of the gap. zation discussions. Alternative resource allocation strategies can then • Prepare the final analysis and report. be developed that prioritize different goals— prevention, treatment, and mitigation. These alterna- Several of the models described in this chapter have tive strategies can be discussed at stakeholder been influential in policy making. However, models not meetings to reach a consensus on the best approach only can be influential in changing policy and the Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 195 policy-making process, but also the interaction of that the agenda for HIV/AIDS, including the establishment process can change the models and affect their of the Global Fund to Fight AIDS, Tuberculosis, and evolution. Malaria. As both the epidemic and the GRNE evolved, each iteration added various interventions in response to DMPPT perceived needs. For example, since the original estimates, One model that both changed policy and itself was interventions such as postexposure prophylaxis, safe changed through the policy-making process is the injection, community mobilization, and prevention for DMPPT, developed by the USAID Health Policy Initiative people living with HIV/AIDS were added. Health system in collaboration with UNAIDS, to inform decision considerations began to be included, including health makers about the potential cost and impact of options systems strengthening, training, incentives, and infra- for scaling up VMMC. When the RCT results for the structure. A separate effort to estimate the resources effect of VMMC on HIV/AIDS transmission were first needed to support orphans and vulnerable children was announced, no publicly accessible, flexible, and sup- spawned and then fed into the existing estimates (Stover ported models were available to estimate the costs and and others 2007). impact of providing VMMC services. In 2007, a large By 2010, the GRNE had expanded to contain many consultative meeting was held by UNAIDS and the WHO, interventions and the total price tag had grown com- at which consensus was reached to prioritize VMMC in mensurately, while the growth in financial resources had countries with high prevalence of HIV/AIDS and low begun to flatten out. In response to these policy issues, prevalence of male circumcision (UNAIDS, WHO, and the next round of estimates underwent an extensive SACEMA Expert Group 2009). consultative process to devise a more targeted and stra- After the model was developed, model applica- tegic approach, identifying interventions that would tions were performed for 14 Sub-Saharan African have relatively higher impact, known as the Investment countries using readily available data. Based on the Framework (Schwartländer and others 2011). Since then, results, a series of briefs were written, one for each many countries and donors have adopted this approach country and a summary brief for the region as a and developed investment cases to illustrate the validity whole (Njeuhmeli and others 2011). The U.S. President’s of the choice of strategy. Throughout this process, Emergency Plan for AIDS Relief (PEPFAR) used the models have informed policy making, and the models briefs heavily to persuade countries either to further have evolved and adapted to changes in the policy investigate the potential cost and impact of VMMC environment. based on primary source data, or simply to adopt a VMMC policy based on the initial results. The briefs AEM were extremely useful in showing the magnitude of those results so clearly. The typical process for the AEM is collaborative. Normally The original model targeted males ages 15–49 years. AEM is applied in an in-country process, organized Since then, evidence on VMMC began to show that around three initial in-country meetings: males under age 25 years were most likely to use VMMC services. Because of this finding, a new version of • The first meeting discusses data needs and inputs, the model, DMPPT 2.0, was developed to estimate the how to extract epidemiological and behavioral impact of targeting VMMC services by five-year age trends, and sources of data. This meeting is followed groups (Stover and Kripke 2014). Several applications of by an intensive period of in-country collation of rele- the new DMPPT are under way; new applications of the vant data and extraction of the required AEM inputs. costing tool are sometimes included to update previous • The second meeting reviews and uses these extracted cost estimates based on older technology. trends to build an initial model and then validate it against numerous other data sources, including male/female ratios, results of incidence studies, GRNE and early HIV/AIDS trends and more recent ART Another example of how a model can affect policy trends. The resulting model is then normally vetted is the development and use of the GRNE. The first by various in-country experts, who review both estimates were developed at the request of UNAIDS to the inputs and outputs and recommend changes establish a global price tag for the estimated funding where necessary. Based on their input, any required required for a comprehensive response to the HIV/ adjustments are made to generate a final national AIDS epidemic. Those results were influential in setting baseline model. 196 Major Infectious Diseases • In the third workshop, key stakeholders are con- • Complementary substantive areas of focus. Different vened to develop scenarios using the intervention models may vary in areas of focus. For exam- and analysis workbooks to explore the epidemic ple, one model may consider the general features impacts of different resource allocations for pre- of ART, while another may highlight differences vention and treatment programs, identify differ- among regimens or monitoring strategies. Thus, ing levels of resource availability, and determine policy makers may determine an allocation for optimal use of available resources under prevailing ART overall based on one model and allocations for epidemic conditions. specific activities within an antiretroviral program based on another. The downside is the lack of an These workshops are generally held in-country integrated assessment and the need to use an extra to maximize the engagement of all key stakeholders, model. Misalignment of two models may create ranging from behavioral scientists, epidemiologists, and confusion. If one model considers options A, B, and public health specialists, to program managers, C, but another model considers B, C, and D, users affected communities, and key decision makers. This may become frustrated. approach helps increase understanding of what the • Differing level of technical engagement by users. data are saying about the epidemic, build a common Some users prefer simpler but less flexible engage- understanding of the forces behind the epidemic, ment with a model, whereas others prefer more and inform decision makers about which choices complex and flexible engagement. Policy makers will maximize their progress in reversing the may fall in the former camp, and epidemiologists epidemic. and other academics may fall in the latter. Although Several countries, for example, Bangladesh, some models offer choice in level of engage- Indonesia, Myanmar, and the Philippines, have devel- ment, obviating this distinction, they may excel oped their own in-country AEM teams that work in either the simpler or the more detailed level of closely with national counterparts to ensure the mod- engagement. els produced meet their policy and advocacy needs. • Competition. Having multiple models may provide AEM helps countries determine where best to focus the impetus to improve model design to build a user their prevention dollars to maximize return on invest- base through quality improvement. ment. Many Asian countries have used AEM as the • Confirmation and confidence building. When dif- basis for revising their national plans, to help them in ferent models yield substantially similar results, preparing concept notes for the Global Fund, and for confidence in the validity of the findings is stronger national advocacy for more effective responses and (Hankins, Forsythe, and Njeuhmeli 2011). When expanded resources. In the Philippines, AEM scenar- results diverge, the attempt to resolve differences ios are being used to actively advocate for expanded can illuminate variations in assumptions or data HIV/AIDS resources. In Thailand, AEM was instru- values that would not otherwise have come under mental in promoting ART for all by demonstrating scrutiny. substantial downstream savings from removing • Efficiency . Perhaps the strongest argument thresholds for ART access. AEM also formed the basis for convergence is efficiency: interested parties for the analyses of the Commission on AIDS in Asia, can focus efforts on one model, building consen- emphasizing the need for responses focused on key sus on methods and inputs. A rigorous review populations and high-impact interventions given process is essential to provide the quality control existing resource constraints rather than trying to that would otherwise arise from competition and cover everybody. comparison. In 2016, there are two dominant models and other ROLE FOR MULTIPLE MODELS VERSUS less widely used models. Avenir Health’s system of mod- els is widely used in countries and global agencies for CONVERGENCE policy-making discussions. The Optima HIV model has With the availability of multiple cost-effectiveness mod- been used in dozens of countries and for global health els, often addressing similar policy territory, the issue agency decisions. Other models are used in more limited that arises is the relative merits of multiple models versus settings in specific countries and published in academic convergence on a single model. The following presents journals. They have served many of the quality control some of the advantages and disadvantages of each functions that might otherwise arise from more approach. balanced competition. Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 197 FRONTIERS OF MODELING: WHERE IS counseling, female sex workers as the target, requirement ADDED VALUE POSSIBLE? that clients return to receive test results, and indeterminate or positive HIV/AIDS test results. These results are limited Unit Cost Resources and preliminary, and the analysis needs to be repeated for All cost-effectiveness models for HIV/AIDS, tuberculosis, other interventions. Further progress in developing meth- and other diseases suffer from a significant gap in ods to measure external validity would increase users’ required input data—the unit cost of delivering inter- knowledge of the accuracy of resource allocation models ventions. Although costing studies for many interventions and their utility as an aid to decision making. are available, they have several serious limitations: many interventions or important variations in intervention delivery have not been formally costed; many geo- Implementation Approaches graphic settings are poorly represented in costing studies The bulk of massive recent spending on HIV/AIDS overall or for specific interventions; and costing methods services has been vertical: programs focused entirely are inconsistent across studies. The Global Health on prevention, treatment, or care, with no resources for Cost Consortium, funded by the Bill & Melinda Gates other diseases and largely separate operational structures. Yet Foundation, is developing a strategy for standardizing various factors highlight the need to consider horizontal existing cost data to improve comparability, extrapolate implementation: control of the disease, meaning that to new geographic settings, strategically fill gaps in exist- infected individuals live long enough to experience other ing data, and improve the efficiency and quality of col- illnesses; the ability to identify infected individuals in other lecting and analyzing cost data. These data will improve service settings, such as reproductive health; and renewed the reach of and confidence in cost-effectiveness models. interest in health system strengthening, such as high- lighted in The Lancet Global Health 2035 Commission on Investing in Health (Jamison and others 2013). Current Model Comparisons resource allocation models permit limited examination The multiple HIV/AIDS resource allocation models offer of implementation approaches, but not comprehensively important choices for potential users, with preferences (the OneHealth Tool comes closest, with explicit consider- based on the policy questions being examined and the ation of system costs); future modeling would do well to availability of detailed local data. Comparing model build in more specific options. Analysis of other imple- results is highly desirable to ensure that estimates are com- mentation issues,such as facility- versus community-based parable and valid. Comparisons have been made for gen- service delivery, with or without demand generation, eral predictions, male circumcision (Hankins, Forsythe, and geographically targeted to high-risk or high-need and Njeuhmeli 2011), and ART as prevention (Eaton and areas, would be valuable; such analysis is likely to be pos- others 2013) but not for detailed resource allocation sible in several of the reviewed models. issues, despite a comparison of model structure and fea- tures (HIV Modelling Consortium 2015). A structured output comparison would be valuable, and may be forth- Interactions coming from the HIV Modelling Consortium in late 2016. HIV/AIDS interacts with other diseases in several ways. It co-occurs in certain populations, such as with hepatitis C among persons who inject drugs. The pathophysiologies External Validity interact; for example, hepatitis C progression is sped by It has long been recognized that efficacy data collected HIV/AIDS, and CD4 decline accelerates with episodes of from research projects, often in atypically well-resourced malaria. Therapy for HIV/AIDS affects (usually reduces) situations, may not accurately portray the results that the risk of other diseases, such as tuberculosis. Capturing could be expected in typical operating programs; the these interactions and their potential implications for research findings thus have low external validity. However, intervention opportunities and health impact will more efforts to describe and enumerate the challenges to exter- accurately portray the relative merits of alternative nal validity vastly outnumber the efforts to improve or investment strategies. even measure external validity. The GHD project has taken initial steps to assess how well research results might be replicated in actual practice settings. Six external indi- Behavioral Economics cators were associated with the effect of HIV/AIDS testing Increasingly, behavioral economics—the use of cognitive on condom use: number of implementation sites, finan- psychology to influence economically relevant behaviors cial incentives, mobile mode of delivering testing and such as taking risks and seeking care—is gaining traction 198 Major Infectious Diseases in health. Cost-effectiveness models can start to incorporate Another area of controversy concerns a central question behavioral economics strategies known to be effective. The in cost-effectiveness modeling: the determination of evidence relevant for infectious and maternal-child disease whether an evaluated option is or is not cost-effective, for is in the process of being reviewed by a team at the University example, by calculating whether the incremental cost- of California, San Francisco. In addition, cost-effectiveness effectiveness ratio is above or below a threshold. The most analysis can potentially benefit from the insights of behav- widely adopted threshold was initially promoted by the ioral economics. Behavioral economics and its underlying Commission on Macroeconomics and Health and adopted prospect theory note that individuals are more averse to loss by the WHO and by WHO-CHOICE. This threshold links than attracted to equivalent gains. Perhaps users of a model per capita gross domestic product with returns on invest- will be more influenced if the presentation is framed as ments in health to define the characteristics of cost-effective missed opportunities to avert infections rather than as new and very cost-effective interventions (Hutubessy, Chisholm, opportunities to avert infections. and Edejer 2003; WHO 2002; WHO-CHOICE 2014). Many published cost-effectiveness analyses of health New Cost-Effectiveness Analysis Outcomes interventions in low-resource countries explicitly refer Cost-effectiveness analysis traditionally compares aver- to these WHO criteria as the standard for determining age incremental health impact and cost. It does not cost-effectiveness. This approach is extremely easy to consider the effects on financial solvency of high expen- apply and reflects the fact that willingness to pay for ditures, nor does it address equity. Extended cost-effec- health care depends in part on national income. tiveness analysis assesses three important considerations However, critics argue that these criteria have at least for policy makers: four major limitations: • Household out-of-pocket private expenditures 1. They have little theoretical justification. • Financial risk protection (number of cases of poverty 2. They skirt the difficult but necessary ranking of the averted) relative values of locally applicable interventions. • Distributional consequences per socioeconomic sta- 3. They omit any consideration of affordability. tus or geographic setting (Verguet, Laxminarayan, 4. Finally, the thresholds set such a low bar for cost- and Jamison 2014). effectiveness that very few interventions with evi- dence of efficacy can be ruled out. An example is provided in Disease Control Priorities, third edition, volume 2, chapter 19 on health gains and An alternative, if more labor-intensive approach, would financial risk protection (Verguet and others 2016). be to compare the cost-effectiveness of an intervention being analyzed with the cost-effectiveness of as many locally relevant interventions as possible (Marseille and CONTROVERSIES IN MODELING others 2015). The use of models to inform health policy in general and Other controversies are rooted in methodological cost-effectiveness models in particular has stimulated concerns. For example, health-state utility is difficult to debate and controversy. measure, and results vary for the same disease or condi- One of the objections is that cost-effectiveness model- tion according to which of a number of accepted meth- ing tacitly reflects ethical judgments about which ods is used to determine it. thoughtful people can disagree. For example, in any com- In addition, the related concept of disability weight parison of outcomes that uses life years, such as quality- does not vary by setting for any chosen disease or condi- -adjusted life years or DALYs, a life-saving intervention tion. The disability weights for mobility, visual, or hearing will, all else equal, favor younger rather than older peo- impairment are the same regardless of the economic status ple. Most people accept the utilitarian principle on which of the country or region in which the analyses are being this rests—as a society, we prefer to save more life years applied. Yet the practical effect on peoples’ lives of the same than fewer; others perceive it as a systematic bias against disability is likely to be greater in poorer countries where, older people. Similarly, and perhaps more controversial, for example, roads are more difficult to navigate and fewer cost-effectiveness analysis puts no greater value on iden- aids are available to assist persons with disabilities. tified lives, such as particular people who are eligible for Other concerns pertain to the fact that the data used treatment, than on anonymous, statistical lives that in models are rarely perfectly suited to the setting or might be saved through, for example, prevention activi- population being studied. Some critics believe that, in ties. Trading off identified and statistical lives challenges, view of these and other limitations, undue reliance is even offends, the ethical values of some people. placed on the results of models, they are treated as more Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 199 reliable than they actually are, and they are used to NOTES address consequential policy questions for which they World Bank Income Classifications as of July 2014 are as are unsuitably designed or parameterized. follows, based on estimates of gross national income (GNI) per capita for 2013: CONCLUSIONS • Low-income countries (LICs) = US$1,045 or less The foregoing tour of HIV/AIDS resource allocation • Middle-income countries (MICs) are subdivided: models presents a robust set of options. The models we (a) lower-middle-income = US$1,046 to US$4,125 describe are able to support the flexible examination of (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 the most critical policy questions: • High-income countries (HICs) = US$12,746 or more. • What will be the cost and health outcomes of invest- 1. https://www.youtube.com/watch?v=Exce4gy7aOk. ing in different combinations of prevention and treat- 2. Spectrum was developed with funding from the U.S. ment interventions? Agency for International Development’s Health Policy • How will those outcomes vary according to local factors Project, the Bill & Melinda Gates Foundation, Johns Hopkins University, the U.S. Fund for the United such as epidemiology, ongoing interventions, and costs? Nations Children’s Fund, UNAIDS, the World Health Organization, the Global Health Workforce Alliance, and The models do require some initial setup, although less the United Nations Population Fund. with the newer streamlined versions than has been the case 3. See the Institute for Disease Modeling, Bellevue, Washington, in the past. More nuanced questions, such as the experi- at http://idmod.org. ence of individuals with particular traits, can be examined, 4. WWR can be found at http://globalhealthdecisions.org albeit with substantially more investment of effort. /wwr/. The relative abundance of resource allocation mod- els now available, each with its own particular focus, strengths, and weaknesses, has two sides. Users can REFERENCES choose a model that fits their particular goals and pur- Bollinger, L. A. 2008. “How Can We Calculate the ‘E’ in ‘CEA’?” poses; however, assessing which model is most fit- AIDS 22 (Suppl 1): S51–57. for-purpose requires more investigation than would be Dieleman, J., C. Graves, T. Templin, E. Johnson, R. Baral, necessary in a world of fewer options. One purpose of and others. 2014. “Global Health Development Assistance this chapter is to serve as a starting point for making Remained Steady in 2013 but Did Not Align with Recipients’ such an assessment. Disease Burden.” Health Affairs 33: 878–86. A long-term challenge for models is keeping up Eaton, J. W., N. A. Menzies, J. Stover, V. Cambiano, with an ever-evolving set of prevention and treatment L. Chindelevitch, and others. 2013. “Health Benefits, Costs, approaches, and with fine-grained strategies such as and Cost-Effectiveness of Earlier Eligibility for Adult Antiretroviral Therapy and Expanded Treatment Coverage: micro-targeting of interventions to disease hotspots. The A Combined Analysis of Twelve Mathematical Models.” The models are constantly improving to reflect these innova- Lancet Global Health 2 (1): 23–34. tions as well as new analytic techniques made possible by Hankins, C., S. Forsythe, and E. Njeuhmeli. 2011. “Voluntary enhanced computing power and the advent of “big data” Medical Male Circumcision: An Introduction to the Cost, that can help inform model parameters. We believe that Impact, and Challenges of Accelerated Scaling Up.” PLoS resource allocation models will continue to provide up-to- Medicine 8 (11): e1001127. date assistance to HIV/AIDS policy makers, program HIV Modelling Consortium. 2015. “Allocative Efficiency Tools designers, and other users. Furthermore, the technology is and Methods to Support Country HIV Programme Budget adaptable to health areas outside HIV—some modeling Allocation.” Imperial College, London. techniques are already being applied to other diseases and Hutubessy, R., D. Chisholm, and T. T. Edejer. 2003. “Generalized more are anticipated. Cost-Effectiveness Analysis for National-Level Priority- Setting in the Health Sector.” Cost-Effectiveness and Resource Allocation 1 (1): 8. ANNEXES Jamison, D. T., L. H. Summers, G. Alleyne, K. J. Arrow, S. Berkley, and others. 2013. “Global Health 2035: A World Converging The annexes to this chapter are as follows. They are within a Generation.” The Lancet 382 (9908): 1898-1955. available at http://www.dcp-3.org/infectiousdiseases. Marseille, E., B. Larson, D. S. Kazi, J. G. Kahn, and S. Rosen. 2015. “Thresholds for the Cost-Effectiveness • Annex 9A. List of Countries with Model Applications of Interventions: Alternative Approaches.” Bulletin of the • Annex 9B. What Works Reviews World Health Organization 93: 118–24. 200 Major Infectious Diseases Mirzazadeh, A., M. Malekinejad, and J. G. Khan. 2015. “Relative ———. 2014. Fasttrack: Ending the AIDS Epidemic by Risk Reduction Is Useful Metric to Standardize Effect Size 2030. Geneva, Switzerland: UNAIDS. http://www for Public Health Interventions for Translational Research.” .unaids.org/sites/default / files/media_asset/JC2686 Journal of Clinical Epidemiology 68 (3): 317–23. _WAD2014report_en.pdf. Njeuhmeli, E., S. Forsythe, J. Reed, M. Opuni, L. Bollinger, UNAIDS, WHO (World Health Organization), and SACEMA and others. 2011. “Voluntary Medical Male Circumcision: (South African Centre for Epidemiological Modelling Modeling the Impact and Cost of Expanding Male and Analysis) Expert Group on Modelling the Impact Circumcision for HIV Prevention in Eastern and Southern and Cost of Male Circumcision for HIV Prevention. 2009. Africa.” PLoS 9 (11): e1001132. “Male Circumcision for HIV Prevention in High HIV Resch, S., T. Ryckman, and R. Hecht. 2015. “Funding AIDS Prevalence Settings: What Can Mathematical Modelling Programmes in the Era of Shared Responsibility: An Contribute to Informed Decision Making?” PLoS Medicine Analysis of Domestic Spending in 12 Low-Income and 6 (9): e1000109. Middle-Income Countries.” The Lancet Global Health 3 Verguet, S., R. Laxminarayan, and D. T. Jamison. 2014. (1): e52–61. “Universal Public Finance of Tuberculosis Treatment in Schwartländer, B., J. Stover, T. Hallett, R. Atun, C. Avila, and India: An Extended Cost-Effectiveness Analysis.” Health others. 2011. “Towards an Improved Investment Approach Economics 24: 318–22. for an Effective Response to HIV/AIDS.” The Lancet 377 Verguet, S., C. Pecenka, K. A. Johansson, S. T. Memirie, (9782): 2031–41. I. K. Friberg, and others. 2016. “Health Gains and Schwartländer, B., J. Stover, N. Walker, L. Bollinger, Financial Risk Protection Afforded by Treatment and J. P. Gutierrez, and others. 2001. “Resource Needs for Prevention of Diarrhea and Pneumonia in Ethiopia: HIV/AIDS.” Science 292 (5526): 2434–36. An Extended Cost-Effectiveness Analysis.” In Disease Stover, J., S. Bertozzi, J. P. Gutierrez, N. Walker, K. A. Stanecki, and Control Priorities (third edition): Volume 2, Reproductive, others. 2006. “The Global Impact of Scaling-up HIV/AIDS Maternal, Newborn, and Child Health, edited by R. Black, Prevention Programs in Low- and Middle-Income Countries.” R. Laxminarayan, M. Temmerman, and N. Walker. Science 311 (5766): 1474–76. Washington, DC: World Bank. Stover, J., and L. Bollinger. 2002. “Resource Allocation within Wilson, D., and J. Taaffe. 2017. “Tailoring the Local HIV/ HIV/AIDS Programs.” In State of the Art: AIDS and AIDS Response to the Local HIV/AIDS Epidemic.” Economics, edited by S. Forsythe. Washington, DC: POLICY In Disease Control Priorities (third edition): Volume 6, Project, Futures Group. Major Infectious Diseases, edited by K. K. Holmes, Stover, J., L. Bollinger, N. Walker, and R. Monasch. 2007. S. Bertozzi, B. R. Bloom, and P. Jha. Washington, DC: “Resource Needs to Support Orphans and Vulnerable World Bank. Children in Sub-Saharan Africa.” Health Policy and WHO (World Health Organization). 2002. World Health Planning 22 (1): 21–27. Report 2002: Reducing Risks, Promoting Healthy Life. Stover, J., and K. Kripke. 2014. “Estimating the Effects of Geneva: WHO. Targeting Voluntary Medical Male Circumcision (VMMC) ———. 2013. Consolidated Guidelines on General HIV Programs to Different Age Groups: The Decision Makers Care and the Use of Antiretroviral Drugs for Treating and Program Planning Toolkit (DMPPT 2.0).” Paper presented Preventing HIV Infection: Recommendations for a Public at the International AIDS Society biannual meeting, Health Approach. Geneva: WHO. http://www.who.int/hiv Melbourne, July 20–25. /pub/guidelines/arv2013/download/en. UNAIDS (Joint United Nations Programme on HIV/AIDS). WHO-CHOICE (Choosing Interventions That Are Cost- 2013. “Treatment 2015.” UNAIDS, Geneva. http://www Effective). 2014. “Cost-Effectiveness Results for Malaria.” unaids.org/sites/default/files/media_asset/JC2484 WHO, Geneva. http://www.who.int/choice/results/mal _treatment-2015_en_1.pdf. _afrd/en/. Improving the Efficiency of the HIV/AIDS Policy Response: A Guide to Resource Allocation Modeling 201 Chapter 10 Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention Harrell W. Chesson, Philippe Mayaud, and Sevgi O. Aral INTRODUCTION any other region. The highest estimated prevalence and Sexually transmitted infections (STIs) impose major incidence rates of chlamydia and trichomoniasis occur health and economic burdens globally. More than 35 bac- in the Americas, while the highest rates of gonorrhea and terial, viral, and parasitic pathogens have been identified syphilis are in Sub-Saharan Africa (figures 10.2 and 10.3) as sexually transmissible. An estimated 498.9 million new In general, trichomoniasis is the most prevalent STI cases of four of the curable STIs occurred among adults across regions, with the exception of Europe and the ages 15–49 years in 2008, an increase of 11.3 percent Western Pacific, where chlamydia is more prevalent. from the estimated 448.3 million new cases in 2005 A great deal of uncertainty surrounds the global and (WHO 2012a). In 2008, these cases included 105.7 mil- regional estimates of the incidence and prevalence of lion new cases of chlamydia, 106.1 million new cases of these four STIs (WHO 2012a). Relative to the size of the gonorrhea, 10.6 million new cases of syphilis, and population in each region, the Americas has the highest 276.4 million new cases of trichomoniasis (WHO 2012a). annual incidence rate of these four curable STIs (0.264), Males accounted for 266.1 million (53 percent) new followed by Africa (0.241), Western Pacific (0.130), cases. At any point in 2008, an estimated 100.4 million Europe (0.104), Eastern Mediterranean (0.085), and adults were infected with chlamydia, 36.4 million with South-East Asia (0.083). However, given heterogeneity in gonorrhea, 36.4 million with syphilis, and 187.0 million the quality of STI surveillance across regions, it is diffi- with trichomoniasis (WHO 2012a). cult to make cross-regional comparisons. The incidence and prevalence of these curable STIs The incidence of STIs can vary substantially within, varies remarkably across World Health Organization as well as across, regions according to the WHO’s Global (WHO) regions, as shown in map 10.1, figure 10.1, and Health Observatory Data Repository. In 2010, the pro- table 10.1. In general, low- and middle-income countries portion of antenatal care attendees who were positive for (LMICs) have higher estimated burdens of STIs than do syphilis was 0.2 percent in Côte d’Ivoire and 10.0 percent high-income countries (HICs) (WHO 2012a). However, in the Central African Republic; the proportion of sex comparing income and STI burden by region can be workers with active syphilis was 1.5 percent in Honduras challenging because income can vary substantially across and 17.5 percent in El Salvador; and the proportion with countries within a given region. For example, the active syphilis among men who have sex with men Americas include two relatively wealthy countries— (MSM) was 1.1 percent in Vietnam and 18.4 percent in Canada and the United States—yet the overall preva- Singapore. The incidence of STIs in a given country can lence of these four curable STIs is higher in this than in vary substantially over time. For example, the percentage Corresponding author: Harrell W. Chesson, Centers for Disease Control and Prevention, Atlanta, Georgia, United States; Hbc7@cdc.gov. 203 Map 10.1 Estimated Incidence of Four Curable STIs in Adults European Region 46.8 million Western Eastern Pacific Region Mediterranean 128.2 million Region of the Americas Region 125.7 million 26.4 million South-East Asia Region 78.5 million African Region 92.6 million Global total 498.9 million Source: WHO 2012a. Note: STI = sexually transmitted infection. Figure 10.1 Estimated Combined Incidence and Prevalence Rates of of MSM with active syphilis in Indonesia ranged from Four Curable STIs, Ages 15–49 Years, WHO Regions, 2008 4.0 percent in 2008 to 21.9 percent in 2012.1 Differences in the burden of STIs across regions, and within regions 30 over time, preclude the identification of feasible pro- grams and policies that can successfully reduce the bur- Incidence and prevalence rates, % 25 den of STIs in a cost-effective manner in all settings. 20 15 HEALTH AND ECONOMIC BURDEN OF STIs If left untreated, common STIs may cause complica- 10 tions, including pelvic inflammatory disease, ectopic pregnancy, postpartum endometriosis, infertility, and 5 chronic abdominal pain in women; adverse pregnancy outcomes, including abortion, intrauterine death, and 0 Incidence Prevalence premature delivery; neonatal and infant infections and African Region Region of the Americas blindness; urethral strictures and epididymitis in men; South-East Asia Region European Region genital malignancies; proctitis, colitis, and enteritis in Eastern Mediterranean Region Western Pacific Region MSM; arthritis secondary to gonorrhea and chlamydia; Source: WHO 2012a. liver failure and liver cancer secondary to hepatitis Note: STI = sexually transmitted infection; WHO = World Health Organization. Calculated based on B virus (HBV); myelopathy and lymphoma or leukemia the estimated number of incidence cases and prevalence cases of four curable STIs (chlamydia, due to human T-cell lymphotropic virus type 1; and gonorrhea, syphilis, and trichomoniasis) and the estimated population ages 15–49 years as estimated by WHO (2012a). The combined rates were calculated by summing the number of central nervous system disease or meningoencephalitis estimated cases and dividing by the population size. secondary to syphilis or herpes simplex virus (HSV) 204 Major Infectious Diseases Table 10.1 Estimated Annual Incidence and Prevalence of Four Curable STIs, Ages 15–49 Years, by WHO Region, 2008 WHO region (number of countries in region) South-East Eastern Western All Indicator Africa (46) Americas (35) Asia (11) Europe (53) Mediterranean (23) Pacific (37) regions Population (millions) 384.4 476.9 945.2 450.8 309.6 986.7 3,553.6 Chlamydia Incidence cases (millions) 8.3 26.4 7.2 20.6 3.2 40.0 105.7 Prevalence cases (millions) 9.1 25.2 8.0 17.3 3.0 37.8 100.4 Incidence (%) 2.2 5.5 0.8 4.6 1.0 4.1 3.0 Prevalence (%) 2.4 5.3 0.8 3.8 1.0 3.8 2.8 Gonorrhea Incidence cases (millions) 21.1 11.0 25.4 3.4 3.1 42.0 106.1 Prevalence cases (millions) 8.2 3.6 9.3 1.0 1.0 13.3 36.4 Incidence (%) 5.5 2.3 2.7 0.8 1.0 4.3 3.0 Prevalence (%) 2.1 0.8 1.0 0.2 0.3 1.3 1.0 Syphilis Incidence cases (millions) 3.4 2.8 3.0 0.2 0.6 0.5 10.6 Prevalence cases (millions) 14.3 6.7 12.3 0.3 1.6 1.2 36.4 Incidence (%) 0.9 0.6 0.3 0.0 0.2 0.1 0.3 Prevalence (%) 3.7 1.4 1.3 0.1 0.5 0.1 1.0 Trichomoniasis Incidence cases (millions) 59.7 85.4 42.9 22.6 20.2 45.7 276.4 Prevalence cases (millions) 42.8 57.8 28.7 14.3 13.2 30.1 187.0 Incidence (%) 15.5 17.9 4.5 5.0 6.5 4.6 7.8 Prevalence (%) 11.1 12.1 3.0 3.2 4.3 3.1 5.3 Four STIs combined Incidence cases (millions) 92.6 125.7 78.5 46.8 26.4 128.2 498.9 Prevalence cases (millions) 74.4 93.3 58.3 32.9 18.8 82.4 360.2 Incidence (%) 24.1 26.4 8.3 10.4 8.5 13.0 14.0 Prevalence (%) 19.4 19.6 6.2 7.3 6.1 8.4 10.1 Source: WHO 2012a. Note: STI = sexually transmitted infection; WHO = World Health Organization. infection (Aral and others 2006; Holmes and Aral 1991; are typically listed in estimates of cancer (Low and van Dam, Dallabetta, and Piot 1999). others 2006). STI sequelae disproportionately affect women and The global burden of cervical and other cancers children. STIs are one of the leading causes of mor- attributable to HPV is substantial. Of the estimated bidity and mortality, as measured by disability- 610,000 HPV-attributable cancer cases worldwide in adjusted life years (DALYs) for reproductive-age 2008, 490,000 occurred in LMICs, where 88 percent of women (Kamb and others 2007) in LMICs. Moreover, cervical cancer deaths also occurred (Forman and the health burden of STIs is often greatly underesti- others 2012). Similarly, HBV-related chronic hepatitis, mated. Although most cervical cancers are caused by liver failure, and liver cancer attributable to sexual, human papillomaviruses (HPVs), the millions of perinatal, or injection drug use transmission are sel- DALYs caused by cervical cancer are not included in dom included in estimates of morbidity attributable estimates of mortality and morbidity due to STIs; they to STIs. Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 205 Figure 10.2 Estimated Incidence Rates of Four Curable STIs, Ages health care seeking are responsible for the high levels of 15–49 Years, by WHO Region, 2008 STIs and the high rates of complications and sequelae in LMICs (Aral, Hogben, and Wasserheit 2008). STI care is 20 provided by a variety of health care providers, many of whom are poorly trained in STI case management, and 15 the quality of care is often inadequate (Mayaud and Incidence rate, % Mabey 2004). Health care seeking for STIs is often delayed and inadequate, particularly among women, as a 10 result of the asymptomatic nature of many STIs; low levels of awareness of sexual health; stigma associated 5 with genital symptoms; and tendency to seek care through traditional healers, home remedies (Mayaud and Mabey 2004; Moses and others 1994; van Dam 0 1995), and pharmacies where drugs are dispensed by ia a ilis sis workers not trained in STI treatment. e yd rrh nia ph lam no Sy mo Go Ch ho c Tri African Region Region of the Americas Factors Affecting Duration and Burden in LMICs South-East Asia Region European Region In resource-poor settings, variables that affect the dura- Eastern Mediterranean Region Western Pacific Region tion of infectiousness include adequacy of health worker training, attitudes of health workers toward marginalized Source: WHO 2012a. Note: STI = sexually transmitted infection; WHO = World Health Organization. Calculated based on groups, patient loads at health centers, availability of the estimated incidence of four curable STIs and the estimated population ages 15–49 years as drugs and clinic supplies, and cost of care (Moses and estimated by WHO (2012a). others 2002). Improvements in these factors would greatly improve STI-related services, reduce the duration Figure 10.3 Estimated Prevalence Rates of Four Curable STIs, Ages of infectiousness, and decrease the incidence of STIs 15–49 Years, by WHO Region, 2008 (Aral 2002). However, in many LMICs, worsening eco- nomic conditions; increasing burden of human immu- 15 nodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS); and occasional health crises, such as natural disasters, refugee situations, or epidemics Prevalence rate, % 10 like the recent Ebola outbreak in West Africa, can adversely affect these variables (Nam and Blanchet 2014). Sociocultural, economic, and political contexts also 5 affect sexual behaviors that contribute to the STI burden in LMICs. Changes have included rising inequality in income and other factors within countries, growing inequality among countries, increased globalization, 0 higher proportions of persons living outside of their cul- ia ea ilis sis yd rh nia tures, increased numbers of unemployed people, and ph lam or Sy mo n Go Ch larger proportions of people living in postconflict societies hoc Tri African Region Region of the Americas (Aral 2002; Aral and others 2006). All of these changes are South-East Asia Region European Region associated with increases in multipartner sexual activity. Eastern Mediterranean Region Western Pacific Region Furthermore, changes in technology, including the wide- Source: WHO 2012a. spread use of cell phones and the Internet, can facilitate Note: STI = sexually transmitted infection; WHO = World Health Organization. Calculated based on the formation of short-term sexual partnerships (Bull and the estimated prevalence of four curable STIs and the estimated population ages 15–49 years, as McFarlane 2000). These technological changes, concur- estimated by the WHO (2012a). rent with changes in norms and attitudes, have led to the expansion of transactional and commercial sex that Delayed and Inadequate Diagnosis increases sexual exposure (Aral and Ward 2014). Delayed or inadequate diagnosis and treatment of STIs Direct medical costs for eight major STIs have been in LMICs result in high rates of complications. To a large estimated at US$16.7 billion in the United States (Owusu- extent, inadequacies in health service provision and Edusei, Chesson, Gift, and others 2013). This estimate 206 Major Infectious Diseases includes costs in the United States in 2008 for gonorrhea, Moreover, population displacement in LMICs often chlamydia, syphilis, trichomoniasis, hepatitis B, diseases affects sexual networks, for example, by allowing or associated with sexually transmitted HPV, genital HSV-2 forcing sexual mixing among groups that did not mix infections, and HIV infection. The total direct medical before the displacement (Hankins and others 2002). cost for each of these STIs in 2008 was computed as the Where available, systematically collected data on estimated number of new cases in 2008 multiplied representative samples of the general population reflect by the estimated discounted lifetime cost per case, increases in a number of risky behaviors, including the adjusted to 2012 U.S. dollars. The estimated costs following: large numbers of sex partners, indiscriminate totaled US$16.7 billion (range of US$11.8 billion to choice of sex partners, short periods between the time US$22.1 billion) when including HIV. Total costs exclud- two people meet each other and the initiation of sexual ing HIV infections were US$3.2 billion. Although few activity, short time spent during the sexual encounter, estimates exist for LMICs, the direct medical costs of STIs lack or short duration of social links between sex part- are undoubtedly substantial given the high prevalence of ners, short duration of gaps between consecutive sex these and other STIs in these settings. STIs also impose partners and sexual encounters, and a tendency for both extensive productivity costs that can often exceed the partners to recruit each other for sex. These trends are direct medical costs (Owusu-Edusei, Chesson, Gift, and observed particularly among younger cohorts. Moreover, others 2013). Productivity costs are particularly burden- sexual behaviors have been changing more rapidly for some in LMICs at both the individual and the commu- women than for men (Aral and Ward 2014; Mercer and nity levels, especially for populations in which most others 2013). people are under age 40 years (World Bank 2007). The Sexual practices have also been changing. Recent data youthful age composition in these countries contributes from the United Kingdom and the United States suggest to the high prevalence and the direct costs of STIs. The trends toward initiation of sex at a younger age, greater economic burden of STIs in LMICs is so high that the frequency of same-sex and bisexual behaviors, and treatment of curable STIs is considered one of the most greater frequency of oral and anal sex (Aral and Ward cost-effective ways to improve health worldwide 2014). Although increases in oral sex began with the (World Bank 2007). generation born between 1946 and 1964, increases in STIs can impose a considerable financial burden on anal sex began with the generations born between 1965 those infected. For example, the cost of drugs is equivalent and 2000 (Aral and Ward 2014). to several days’ wages in most LMICs (Terris-Prestholt These changes may result from temporal trends in and others 2006). The direct costs associated with medical demographic and social patterns. Marriage rates have treatment of STI sequelae in LMICs have not been well declined, and divorce rates have risen in the Organisation defined, and the indirect costs associated with lost for Economic Co-operation and Development countries productivity due to STIs or STI sequelae are not known. and the United States (Aral and Ward 2014; International Futures Program 2011; Stevenson and Wolfers 2007). Globally, people who marry are doing so at older ages CHANGES IN SEXUAL BEHAVIORS than before (Aral and Ward 2014). Because of these AND PRACTICES, EPIDEMIOLOGY, AND trends, many people spend a higher percentage of their PREVENTION adult lives outside of marriage (Aral and Ward 2014), which probably increases the number of sex partners. Significant changes have occurred in sexual behaviors and Data collected in LMICs over the past two decades practices, epidemiology, and prevention. Technological have revealed the importance of sex work to the spread advances, political conflicts, the economic downturn of STIs (Baral and others 2012) and the presence (and experienced in many HICs, and natural and health crises considerable prevalence) of MSM among sex workers all have had important effects. (Baral and others 2007). These key populations have high prevalence of STIs, including HIV/AIDS, and play an important role in spreading STIs to the general Sexual Behaviors and Practices population. Most of the data on sexual behaviors and practices come from HICs. However, the increased volume of travel, sex tourism, transactional and commercial sex, Epidemiology and role of technology in establishing these connections The understanding of STI epidemiology in LMICs have expanded sexual networks beyond national has been shaped by the reemergence and escalation of boundaries (Aral and Ward 2014; Ward and Aral 2006). gonorrhea and congenital syphilis; recognition of sexual Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 207 transmission as a key factor in the spread of other Empirical and model-based studies in HICs suggest that STIs; emergence of MSM as key populations in trans- AIDS mortality contributed to declines in bacterial STIs mission; emergence and impact of HIV/AIDS mortality; through two main mechanisms: and patterns of STI spread, such as clustering and globalization. Although these issues can be difficult • Behavioral responses to the HIV/AIDS epidemic, problems globally, they are especially daunting to mon- such as increased condom use and smaller number itor, control, and prevent in LMICs. of partners • AIDS mortality among those at highest risk of acquir- Gonococcal Antimicrobial Resistance ing and transmitting STIs (Becker and Joseph 1988; Widespread, high-level gonococcal antimicrobial Boily and Brunham 1993; Boily and others 2004; resistance has been observed in Africa, South-East Chesson, Dee, and Aral 2003; Kault 1992). Asia, and the Western Pacific (Bala and others 2013; Lahra, Lo, and Whiley 2013; Ndowa and others 2013). However, a few years after effective ART became In addition to resistance to penicillin, tetracy- available in 1996, STI incidence increased in subpopula- cline, and quinolones, decreased susceptibility to tions most affected by HIV/AIDS. Syphilis outbreaks third-generation cephalosporins has been reported. among MSM have been observed in metropolitan areas Decreased susceptibility to treatment has been associ- worldwide since the late 1990s, in large part because of ated with increased gonorrhea incidence at the decreased fear of HIV/AIDS and increased survival of population level in the United States (Chesson and persons with HIV/AIDS (Chesson and Gift 2008; Stolte others 2014). These trends highlight the importance of and others 2004). sustaining and enhancing surveillance to monitor the The availability of ART has increased greatly in spread and threat of antimicrobial resistance (Lahra, LMICs. At the end of 2009, 5.25 million people in these Lo, and Whiley 2013). settings were receiving ART, compared with 4 million at the end of 2008 (WHO 2010). Peterman and Congenital Syphilis and Other Complications of Furness (2015) report notable declines in syphilis in Syphilis in Pregnancy some parts of Africa and attribute these declines in Syphilis in pregnancy can lead to a wide range of part to syndromic treatment of genital ulcers and pos- adverse outcomes, including stillbirth, fetal loss, neo- sibly the impact of AIDS mortality. Building on expe- natal death, premature and low-birthweight infants, rience acquired during the syphilis resurgence among and infection or disease in newborns (John-Stewart MSM in HICs, researchers have cautioned that an and others 2017; Kahn and others 2014; Newman and increase in STIs is possible in LMICs as a result of the others 2013; WHO 2012b). Even though these adverse increased availability of ART, particularly in areas with outcomes could be prevented through antenatal high coverage (Kenyon, Osbak, and Chico 2014; screening programs, syphilis in pregnancy imposes a Kenyon and others 2014). substantial global burden each year, resulting in 692,000–1,530,000 adverse outcomes annually (Kamb Key Populations and others 2010). As measured by DALYs, the global The role of key populations in the epidemiology of STIs burden of disease due to syphilis during pregnancy is in LMICs has become increasingly clear (Baral and oth- comparable to that of mother-to-child transmission ers 2007; Baral and others 2012). In particular, MSM are of HIV (Kahn and others 2014; Kamb and others understudied and underserved in these countries. 2010; WHO 2012b). Patterns of sexual networks linking MSM and the gen- LMICs bear a disproportionate share of the global eral population warrant future research so that appro- health and economic burden of syphilis in pregnancy priate responses can be developed. (Kamb and others 2010). In the Mwanza Region of Tanzania, from 1998 to 2000, maternal syphilis accounted Clustering, Social Determinants, and Globalization for more than 50 percent of all stillbirths and 17 percent Three additional patterns have influenced and enhanced of all adverse pregnancy outcomes among unscreened understanding of STI epidemiology in LMICs: women (Watson-Jones and others 2002). • Geographic clustering and concentration of risk AIDS Mortality behaviors and infections The probable impact of AIDS mortality, before the • Importance of context, social determinants, and advent of effective antiretroviral therapy (ART), on the structural drivers declining incidence of bacterial STIs was substantial. • Globalization. 208 Major Infectious Diseases Nonuniform distribution and clustering of risk— increasing recognition that behavioral interventions both in behaviors and infections—have been reported for have not brought sustainable decreases in incidence the epidemiologies of both HIV and other STIs (Chesson (Aral 2011; Kippax and Stephenson 2012). Concurrently, 2010a, 2010b; Leichliter and others 2010). More recent remarkable progress has been made in biomedical attention has been drawn to geographic concentration approaches to preventing HIV/AIDS, including male (Tanser and others 2009) and to the critical role of local circumcision, preexposure prophylaxis (PrEP), and context in the epidemiology of STIs. The Priorities for ART (Baeten and others 2012; Dodd, Garnett, and Local AIDS Control Efforts method can identify sites Hallett 2010; Grant and others 2010; Katz and Wright where people with high rates of partner change can 2008; Pretorius and others 2010; Thigpen and others receive STI prevention services (Weir and others 2003). 2012). Given the success of biomedical approaches to Similarly, the Situational Analysis of Sexual Health the prevention of HIV/AIDS, the field of STI preven- method can identify specific locations where vulnerable tion is drawing increasingly on biomedical interven- and at-risk people can receive STI prevention services tions, reinforced by development of effective biomedical (Benzaken and others 2012). Insight into variations in interventions for preventing STIs other than HIV. STI epidemiology has important implications for preven- More specifically, the HPV and HBV vaccines, point- tion and control, including targeting of interventions and of-care diagnostic tests for syphilis and dual tests for allocation of resources (Aral and Cates 2013). syphilis and HIV, and an understanding of the preven- In the past two decades, the importance of social tive effects of circumcision for certain STIs are begin- determinants of sexual health and structural drivers for ning to show promise in preventing specific STIs other STI epidemiology have received increasing recognition. than HIV. Examples of social determinants include low socioeco- STI prevention has also been influenced by other nomic status and poor access to quality health care insights. Prevention activities have increasingly sought (Hogben and Leichliter 2008). Underlying social, eco- to achieve impact at the population level. In addition to nomic, legal, and political structures have a notable protecting individuals, the focus has turned to decreas- influence on sexual behaviors (Hogben and Leichliter ing population incidence. This shift has brought several 2008; Parkhurst 2014). Moreover, these factors affect the other changes given that it requires system-level think- formation, evolution, and persistence of STIs in key pop- ing, planning, and evaluation. It is important to take into ulations. Finally, globalization shapes and connects sex- account how interventions may have additive, synergis- ual behaviors, practices, and networks around the world tic, or antagonistic effects (Aral 2011; Aral and Douglas (Aral, Bernstein, and Torrone 2015; Aral and Ward 2005, 2007; Parkhurst 2014). The social and epidemiological 2014; Ward and Aral 2006). context and interactions between interventions and con- Current developments in methodological approaches text have also emerged as important issues (Aral and promise to have an impact on the study and under- Cates 2013; Parkhurst 2014). standing of STI epidemiology in all settings. Two devel- More attention is being given to the elements of opments are particularly remarkable: (1) the increasing complex systems (mixing patterns, networks, cluster- use of sophisticated geographic mapping methodolo- ing, and hot spots) and to social, economic, legal, and gies (Tanser and others 2009) and (2) phylogenetic sexual structures (Blanchard and Aral 2010; Parkhurst analyses combined with social epidemiology (Avila and 2014). The need for new approaches to designing others 2014), specifically, phylogenetic and network prevention programs is now widely recognized analyses. When combined, these approaches provide (Aral and Blanchard 2012; Blanchard and Aral 2011; powerful explanations of transmission dynamics within Parkhurst 2014). and between groups; if used in conjunction with geo- With the reality of limited and declining resources, mapping, they may enhance the understanding of emphasis has been placed on accountability, resource aspects of STI prevention science, such as subgroup allocation, efficiency, prioritization, and return on targeting. investment (Over and Aral 2006). These develop- ments are changing the STI prevention field in impor- tant ways. The hope is that the next decade will bring Prevention significantly greater prevention for the money in Important changes in the approach to STI prevention LMICs, where health systems are often weak (Mills have been influenced by the HIV epidemic. During the 2014). Reforming and strengthening of health care 1980s and 1990s, behavioral prevention dominated the infrastructure may be needed before the recent advances HIV world and gained prominence in the STI domain. in STI prevention science can be successfully imple- However, since the turn of the century, there has been mented in these contexts. Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 209 EFFECTIVENESS OF STI PREVENTION Specifically, interventions were organized as primary INTERVENTIONS: LITERATURE REVIEW prevention (behavioral interventions, male circumcision, vaccines, and microbicides), STI case management, Over the past 20 years, many STI prevention interven- partner notification and management, targeted tions have been rigorously evaluated for effectiveness. In interventions and periodic presumptive treatment a review of STI prevention interventions evaluated by (PPT), mass treatment, and community-level and randomized controlled trials (RCTs) in HICs and LMICs, structural interventions. Wetmore, Manhart, and Wasserheit (2010) found that 44 of 75 interventions (59 percent) significantly reduced the risk of acquiring at least one STI. Interventions were Primary Prevention organized according to modality, including behavior Table 10.2 summarizes selected studies of the impact of change, vaginal microbicides, male circumcision, part- primary prevention interventions, categorized as behav- ner services, treatment, and vaccines. The percentage of ior change interventions, male circumcision, vaccines, trials in which a statistically significant reduction in the and microbicides. risk of a laboratory-confirmed STI was observed in the intervention arm (compared with the control arm) was Behavior Change Interventions highest for treatment, vaccines, and male circumcision, Promotion of condom use, STI and HIV education, and followed by behavioral interventions, partner services, knowledge and skill building are common behavior and vaginal microbicides. These findings are consistent change interventions. Interventions to increase condom with those of Manhart and Holmes (2005), in which 54 use are generally effective in reducing STI incidence in percent of the trials led to a significant reduction in STI high-risk populations (Celentano and others 2000; acquisition, transmission, or complications. Feldblum and others 2005; Patterson and others 2008), For this section, a literature search was conducted to although promotion of male and female condoms is identify studies of the impact of STI prevention interven- likely of modest benefit in populations already exposed tions in LMICs. The search was conducted from January to interventions promoting male condoms (Hoke and 2014 to April 2014, and the following databases were others 2007). However, Fontanet and others (1998) used: Cochrane Library, Database of Abstracts of Reviews found that female sex workers in Thailand who had the of Effects, MEDLINE, and Embase. The MEDLINE option of using female condoms in situations where search terms used to identify the relevant literature are male condoms were not used had STI incidence rates listed in annexes 10A and 10B, and these search terms that were 24 percent lower than those using male con- were amended as necessary to search the other databases. doms only. This search was supplemented with additional sources, such as the bibliographies of articles obtained in the Male Circumcision search and previous reviews of the impact of STI preven- Male circumcision has a protective effect against HSV-2, tion interventions (Manhart and Holmes 2005; Mayaud HPV, and Mycoplasma genitalium in circumcised men and Mabey 2004; Wetmore, Manhart, and Wasserheit (Auvert and others 2009; Mehta and others 2012; Tobian 2010). Although this review and that of Wetmore, and others 2009) and against trichomoniasis and bacte- Manhart, and Wasserheit (2010) overlap, there are four rial vaginosis in their female partners (Gray and others key differences. First, this review was not systematic—no 2009). Effects of male circumcision on trichomoniasis specific inclusion or exclusion criteria were applied. were mixed (Mehta and others 2009; Sobngwi-Tambekou Instead, studies were selected to highlight key aspects of and others 2009); no protective effect was observed the evidence, focusing on studies that use biological out- against gonorrhea, chlamydia, or syphilis (Mehta and comes rather than changes in attitudes or behaviors. others 2009; Sobngwi-Tambekou and others 2009; Second, the search was not limited to RCTs, but also Tobian and others 2009). Although the trials of male considered cohort and cross-sectional studies. Third, this circumcision found no significant impact on chlamydia, review focused on interventions that were evaluated in Castellsague and others (2002) and Castellsague and LMICs. Finally, it included more recent articles, pub- others (2005) present evidence that women with uncir- lished from January 2000 to July 2014, than the earlier cumcised partners have a higher prevalence of chlamy- review, which included articles published through dia than women with circumcised partners. December 2009. In this summary of the literature, interventions were HPV Vaccines: HSV, HPV, and HBV organized according to intervention modality using a An HSV-2 glycoprotein-D–adjuvant vaccine adminis- structure adapted from Mayaud and Mabey (2004). tered to persons with no serological evidence of previous 210 Major Infectious Diseases Table 10.2 Selected Evaluations of the Effectiveness of Primary STI Prevention Interventions, with a Focus on Interventions in Low- and Middle-Income Countries Type of intervention and study Description of intervention Setting Key results Behavior change interventions Ford and Peer education program Female sex workers in Bali, Indonesia Gonorrhea prevalence was lower in clusters others 2000 addressing topics such as STIs with a peer educator versus clusters without. and HIV, condom use, and condom negotiation Celentano and Institution-based, 15-month Men in Royal Thai Army Relative risk of STI infection (gonorrhea, others 2000 intervention to promote consistent syphilis, chancroid, nongonococcal urethritis) and proper condom use, reduce was 0.15 for those in intervention group versus alcohol consumption, and reduce those in control group. brothel patronage Feldblum and Male condom promotion Female sex workers in Madagascar Odds ratios for aggregate STI prevalence others 2005 intervention, comparing clinic- (chlamydia, gonorrhea, trichomoniasis) were based counseling and peer significantly lower for group with clinic plus promotion to peer promotion only peer versus peer alone. Hoke and Male and female condom Female sex workers in Madagascar STI prevalence (chlamydia, gonorrhea, others 2007 promotion intervention, comparing trichomoniasis) did not differ significantly by clinic-based counseling and peer study arm. promotion to peer promotion only among those already exposed to intensive male condom promotion Jewkes and A 50-hour program to build Men and women ages 15–26 years A 33 percent reduction in incidence of HSV-2 others 2008 knowledge and skills in Eastern Cape Province, South occurred in the intervention group. Africa Patterson and Brief condom promotion Female sex workers in Mexico A 40 percent decline in cumulative STI incidence others 2008 intervention (HIV, syphilis, gonorrhea, chlamydia) occurred in the intervention group. Chong and Online education program Adolescents attending Colombian Among those sexually active at baseline, the others 2013 addressing topics such as sexual public schools intervention group had a 5 percent reduction rights, contraception, condom use in self-reported STIs (including chlamydia, and STIs and HIV, empowerment, gonorrhea, trichomoniasis, and syphilis). and violence prevention Male circumcision Mehta and Randomized trial of circumcision to Men ages 18–24 years in Kisumu, No significant impact of circumcision on others 2009 prevent HIV and other STIs Kenya incidence of gonorrhea, chlamydia, or trichomoniasis was noted. Sobngwi- Randomized trial of circumcision to Men ages 15–29 years in Orange Male circumcision reduced trichomoniasis, but Tambekou and prevent HIV and other STIs Farm, South Africa no effect on gonorrhea and chlamydia was others 2009 noted. Auvert and Randomized trial of circumcision to Men ages 15–29 years in Orange Male circumcision reduced prevalence of high- others 2009 prevent HIV and other STIs Farm, South Africa risk HPV types in men. Tobian and Randomized trial of circumcision to Men ages 15–29 years in Rakai, Statistically significant lower rates of HSV-2 others 2009 prevent HIV and other STIs Uganda seroconversion and HPV prevalence were noted among circumcised males; no significant impact on syphilis incidence was noted. table continues next page Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 211 Table 10.2 Selected Evaluations of the Effectiveness of Primary STI Prevention Interventions, with a Focus on Interventions in Low- and Middle-Income Countries (continued) Type of intervention and study Description of intervention Setting Key results Gray and Randomized trial of circumcision to Men ages 15–29 years in Rakai, Male circumcision reduced the risk of genital others 2009 prevent HIV and other STIs Uganda, and their wives or long-term ulceration, trichomoniasis, and bacterial consensual partners vaginosis in female partners. Mehta and Randomized trial of circumcision to Men ages 18–24 years in Kisumu, Circumcision nearly halved the odds of others 2012 prevent HIV and other STIs Kenya urogenital Mycoplasma genitalium. Vaccines Stanberry and Randomized trial of safety and Men and women ages 18 years and Among women with no serological evidence of others 2002 efficacy of genital herpes vaccine older in Australia, Canada, Italy, and previous HSV-1 infection, vaccine offered partial the United States protection; no efficacy was noted for women seropositive for HSV-1 at baseline or for men regardless of their HSV serologic status. Harper and Randomized trial of safety, Women ages 15–25 years at Vaccine protected against HPV 16 and 18 others 2004 immunogenicity, and efficacy of enrollment in Brazil, Canada, and the infection and associated cervical lesions, with bivalent HPV vaccine United States evidence of cross-protection against other HPV types. Villa and Randomized trial of safety, Women ages 16–23 years in Brazil, Vaccine protected against HPV 6, 11, 16, and others 2006 immunogenicity, and efficacy of Nordic countries, and the United 18 infection and associated cervical lesions and quadrivalent HPV vaccine States genital warts. Garland and Randomized trial of safety, Women ages 16–24 years in Asia- Vaccine protected against external anogenital, others 2007 immunogenicity, and efficacy of Pacific, Europe, Latin America, and vaginal, and cervical lesions associated with quadrivalent HPV vaccine North America HPV 6, 11, 16, and 18. Future II Study Randomized trial of safety, Women ages 15–26 years in Asia- Vaccine protected against HPV 16– and 18– Group 2007 immunogenicity, and efficacy of Pacific, Europe, Latin America, and associated cervical intraepithelial neoplasia quadrivalent HPV vaccine North America grade 2 or worse. Muñoz and Randomized trial of safety, Women ages 24–45 years in Vaccine protected against HPV 6, 11, 16, and 18 others 2009 immunogenicity, and efficacy of Colombia, France, Germany, the infection and associated disease. quadrivalent HPV vaccine Philippines, Spain, Thailand, and the United States Paavonen and Randomized trial of safety, Women ages 15–25 years from 14 Vaccine protected against HPV 16– and 18– others 2009 immunogenicity, and efficacy of countries in Asia-Pacific, Europe, associated cervical intraepithelial neoplasia bivalent HPV vaccine Latin America, and North America grade 2 or worse, with some protection against HPV 31, 33, and 45. Giuliano and Randomized trial of safety, Men ages 16–26 years in Africa, Vaccine protected against HPV 6, 11, 16, and others 2011 immunogenicity, and efficacy of Asia-Pacific, Europe, Latin America, 18 infection and associated external genital quadrivalent HPV vaccine and North America lesions. Palefsky and Randomized trial of safety, MSM ages 16–26 years in Australia, Vaccine protected against anal intraepithelial others 2011 immunogenicity, and efficacy of Brazil, Canada, Croatia, Germany, neoplasia in MSM. quadrivalent HPV vaccine Spain, and the United States Joura and Randomized trial of Women ages 16–26 years in 18 Vaccine protected against high-grade cervical, others 2015 immunogenicity and efficacy of countries vulvar, and vaginal disease and persistent nonavalent HPV vaccine infection related to HPV 31, 33, 45, 52, and 58 and generated antibody response to HPV 6, 11, 16, and 18 that was noninferior to that of the quadrivalent HPV vaccine. table continues next page 212 Major Infectious Diseases Table 10.2 Selected Evaluations of the Effectiveness of Primary STI Prevention Interventions, with a Focus on Interventions in Low- and Middle-Income Countries (continued) Type of intervention and study Description of intervention Setting Key results Microbicides, suppressive therapy, and barrier methods Richardson Randomized trial of nonoxynol-9 Female sex workers in Mombasa, Intervention group had higher gonorrhea and others gel for STI prevention Kenya incidence than control group; no differences for 2001 other STIs. Roddy and Randomized trial of nonoxynol-9 Women in Cameroon Gel provided no protective effect against others 2002 gel for STI prevention urogenital gonococcal or chlamydial infection. Corey and Suppressive therapy (once-daily HSV-2–discordant couples in Suppressive therapy reduced HSV transmission others 2004 valacyclovir) to reduce HSV Australia, Canada, Europe, Latin to the susceptible partner by about 75 percent. transmission America, and the United States Ramjee and Randomized trial of diaphragm, Women in southern Africa STI incidence (chlamydia, gonorrhea) did others 2008 lubricant gel, and condoms versus not differ significantly by study arm. Among condoms alone to prevent STIs consistent users, persons in the diaphragm arm had a lower risk of acquiring gonorrhea. Karim and Randomized trial of 1 percent Women in KwaZulu-Natal, South There was a 39 percent reduction in HIV others 2010 vaginal gel formulation of tenofovir Africa acquisition (54 percent among those with high for HIV prevention adherence) and a 51 percent reduction in HSV-2 acquisition. Bukusi and Randomized trial of topical penile Heterosexual couples in Kenya The hazard ratio of diagnosis of bacterial others 2011 microbicide (62 percent alcohol in vaginosis was 1.44 in intervention arm versus gel) to prevent bacterial vaginosis control arm. in sex partners Guffey and Randomized trial of microbicides Women in Malawi, South Africa, Candidate microbicides did not protect against others 2014 BufferGel and 0.5 percent Zambia, Zimbabwe, and the United gonorrhea, chlamydia, or trichomoniasis. PRO 2000 for prevention of States nonulcerative STIs Note: HIV = human immunodeficiency virus; HPV = human papillomavirus; HSV = herpes simplex virus; MSM = men who have sex with men; STI = sexually transmitted infection. HSV-1 infection partially protected women, but not The HBV vaccine has been available for many years men, from acquiring genital herpes disease, with efficacy and is increasingly used in infants in many countries; of about 75 percent across two trials (Stanberry and vaccine programs are also now available in some coun- others 2002). In contrast, the bivalent, quadrivalent, and tries for adolescents and young adults who did not nonavalent HPV vaccines have shown remarkably high receive the vaccine as infants. However, many adults at efficacy in preventing infection and disease, and the risk today have never received the HBV vaccine. For bivalent and quadrivalent vaccines may also offer some example, in an Internet survey conducted in the United cross-protection against other types of HPV (Malagon States in 2010, 42.4 percent of HIV-negative MSM older and others 2012). These safe and effective vaccines could than age 31 years reported never having received the reduce the burden of cervical cancer and potentially other HBV vaccine (Matthews, Stephenson, and Sullivan cancers, such as vulvar, vaginal, penile, anal, and orophar- 2012). A cross-sectional survey of MSM in Beijing, yngeal cancers (Markowitz and others 2014). In HICs China, in 2012 found that only 38.9 percent had received with routine HPV vaccination programs, reductions in the HBV vaccination (Wang and others 2012). the prevalence of HPV and incidence of HPV-associated health outcomes, such as genital warts and cervical Microbicides precancers, have been observed at the population level Randomized trials have found that PrEP with antiretro- (Drolet and others 2015; Fairley and others 2009; Flagg, virals can reduce HIV acquisition among heterosexual Schwartz, and Weinstock 2013; Hariri and others 2013; men and women in serodiscordant couples (Baeten and Markowitz and others 2013; Tabrizi and others 2012). others 2012; Thigpen and others 2012) and in MSM Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 213 (Grant and others 2010). A randomized trial of couples the algorithms to detect cervical infection can be serodiscordant for HSV-2 in Australia, Canada, Europe, improved by incorporating risk scores based on factors Latin America, and the United States found that once- such as sexual history. Other studies also provide evi- daily valacyclovir for suppressive therapy reduced trans- dence that risk scores can improve the efficiency of mission of HSV-2 to the seronegative partner by about syndromic management algorithms (Cornier and oth- 75 percent (Corey and others 2004). However, an RCT in ers 2010). Pettifor and others (2000) conclude that, HIV-discordant couples in which the partner with HIV although syndromic management can be effective for was also infected with HSV-2 found that daily acyclovir managing STIs, affordable, rapid, and effective diagnos- did not reduce the risk of HIV transmission to the HIV- tic techniques to improve detection are urgently needed negative partner (Celum and others 2010). in resource-poor settings. Karim and others (2010) and Karim and Baxter Although evidence is limited, widespread implemen- (2013) report that a 1 percent vaginal gel formulation of tation of syndromic management as an approach to STI tenofovir reduced the risk of HIV and HSV-2 acquisition case management likely has reduced the burden of STIs, in a randomized trial involving women in KwaZulu- particularly in resource-poor settings (Aral and others Natal, South Africa. Specifically, it reduced HIV acquisi- 2006). A community randomized trial in Mwanza, tion by 39 percent (54 percent among those with high Tanzania, found that syndromic treatment of STIs adherence) and HSV-2 acquisition by 51 percent. resulted in a 40 percent reduction in HIV incidence and However, in a randomized, placebo-control trial of a reduction in symptomatic urethritis in men and prev- tenofovir-based PrEP in women in South Africa, Uganda, alence of syphilis seroreactivity (Grosskurth and others and Zimbabwe (the VOICE study), none of the drug 1995; Mayaud and others 1997). Prevalence rates of regimens reduced HIV-1 acquisition rates in the intent- other STIs were lower in the intervention communities to-treat analysis (Marrazzo and others 2015). Similarly, as well, although the differences were not statistically the FACTS 001 trial in more than 2,000 women in nine significant for all indicators. A community randomized sites in South Africa found that pericoital vaginal appli- trial in Masaka, Uganda, offers evidence that the syndro- cation of tenofovir 1 percent gel was not effective in mic management of STIs reduced the incidence of cur- preventing HIV acquisition (Rees and others 2015). In able STIs but not HIV (Kamali and others 2003). In a both the VOICE and the FACTS 001 studies, low rates of cluster randomized trial in Eastern Zimbabwe, an inter- adherence to the drug regimen were considered a pri- vention that included syndromic management of STI mary reason for this result. had no impact over time on the incidence of STI symp- An RCT involving sexually active women in southern toms, although male patients in the intervention com- Africa at risk for STIs found that providing condoms munities were significantly more likely than those in the alone (control) was as effective as providing a diaphragm control communities to report cessation of symptoms and lubricant gel in addition to condoms (intervention) in (Gregson and others 2007). An RCT comparing preventing chlamydia and gonorrhea (Ramjee and others enhanced syndromic management and PPT among 2008). However, consistent use of a diaphragm could be hotel-based sex workers in Bangladesh found that both protective given that the incidence of gonorrhea among strategies were effective for STI control (McCormick and women in the intervention arm was significantly lower others 2013). A randomized trial involving men in among those who reported always using a diaphragm. Malawi with urethritis found that the addition of metro- nidazole to the syndromic management of male urethri- tis can reduce trichomoniasis infection in men (Price STI Case Management and others 2003). Pettifor and others (2000) review the literature on the effectiveness of syndromic management of STIs. Their review includes 5 studies of WHO algorithms for man- Partner Notification and Management agement of urethral discharge, 5 for genital ulcers, and Alam and others (2010) conducted a systematic litera- 13 for vaginal discharge. Overall, the literature suggests ture review of the feasibility and acceptability of part- that algorithms for urethral discharge, vaginal dis- ner notification for STIs in low-resource settings and charge, and genital ulcer disease can be effective. For summarized the evidence that partner notification example, La Ruche, Lorougnon, and Digbeu (1995) interventions can yield positive outcomes. An RCT in reported therapeutic success rates of 92 percent for Harare, Zimbabwe, involving men and women with a male urethritis, 87 percent for vaginal discharge, and syndromically diagnosed STI found that a partner 100 percent for genital ulcer disease. The studies referral intervention (client-centered, private session reviewed in Pettifor and others (2000) also show that with a trained counselor) significantly increased the 214 Major Infectious Diseases likelihood that at least one partner would be reported, improve the diagnostics part of the HIV treatment compared with standard care in which the treating cli- continuum. nician discussed partner referral (Moyo and others 2002). A randomized trial in Kampala, Uganda, involv- ing men and women with a syndromically diagnosed Targeted Interventions and Periodic Presumptive STI found that a significantly higher percentage of Treatment partners were treated using patient-delivered partner Interventions commonly target groups at high risk of STI medication compared with patient-based partner refer- acquisition and transmission. These interventions can ral (Nuwaha and others 2001). include the provision of PPT, which is the systematic Although no published evaluations are available of treatment of people at high risk with a combination of the impact of partner services on STI incidence in drugs targeting the prevalent curable STIs. As shown by LMICs, evidence is available from trials conducted in four rigorous evaluations, PPT interventions can be HICs. Studies from the United States, for example, have highly effective in reducing the STI burden within tar- shown that the administration of suppressive therapy geted groups. In an RCT among female sex workers in to partners infected with HSV-2 in serodiscordant cou- Kenya, the provision of monthly prophylaxis substantially ples can reduce the incidence of HSV-2 seroconversion reduced the incidence of gonorrhea, chlamydia, and tri- in uninfected partners (Corey and others 2004) and chomoniasis, but not of HIV (Kaul and others 2004). that expedited partner treatment (including patient- Reductions of about 45 percent in the prevalence of cervi- delivered therapy to a partner) can reduce the risk of cal infection with gonorrhea and chlamydia were observed persistence or reoccurrence of gonococcal or chlamy- among commercial sex workers in the Lao People’s dial infection in the index patient (Golden and Democratic Republic after monthly PPT over a three- others 2005). Golden and others (2015) conducted a month period (O’Farrell and others 2006). Substantial community-level stepped-wedge RCT of a public reductions in STIs were also observed among hotel-based health intervention to increase the uptake of expedited sex workers in Bangladesh following the provision of partner therapy. The intervention increased the per- monthly PPT over a nine-month period (McCormick centage of persons receiving patient-delivered partner and others 2013). PPT with vaginal suppositories con- therapy and those receiving partner services. The taining metronidazole and miconazole among HIV- investigators estimated that the intervention was asso- negative women with one or more vaginal infections in ciated with reductions of about 10 percent in chlamy- Kenya and in Birmingham, Alabama, significantly reduced dia positivity and gonorrhea incidence, although these the prevalence of bacterial vaginosis among women dur- reductions were not statistically significant, perhaps as ing 12 months of follow-up, compared with women a result of inadequate statistical power and of state- receiving a placebo (McClelland and others 2015). Steen, financed uptake of parts of the intervention in control Chersich, and de Vlas (2012) noted that reductions in communities. Further trials are needed to assess the gonorrhea and chlamydia on the order of 50 percent were impact on STIs and cost-effectiveness of partner noti- common across the 15 studies included in their review of fication interventions in LMICs (Alam and others PPT of curable STIs among sex workers. 2010; Ferreira and others 2013). The WHO (2008) reviewed the effectiveness of pre- The potential benefits of partner notification strate- sumptive treatment, finding that PPT can lead to rapid, gies for STIs in LMICs are supported by encouraging short-term reductions in STI prevalence among high- results for HIV in LMICs. For example, Henley and risk groups and that ongoing STI services help sustain others (2013) found that only 3.2 index cases needed to these reductions. However, research is needed regarding be interviewed, on average, to identify one additional the use of PPT in high-risk populations and the impact person with HIV in Cameroon. Similarly, an RCT in of PPT on the emergence of antimicrobial resistance in Malawi found that 51 percent of partners returned for sexually transmitted and other pathogens. counseling and testing in the provider referral group in Reducing STI prevalence among core groups (for which health care providers notified partners, com- example, sex workers) through PPT can have notable pared with 24 percent in the passive referral group in public health effects (such as prevention of STIs in the which patients were responsible for notifying their clients of sex workers), although the evidence is limited. partners (Brown and others 2011). The health impact An intervention of PPT plus STI prevention education and cost-effectiveness of partner notification could be targeted to high-risk women in a South African mining improved substantially by integrating HIV testing into community was found to reduce the prevalence of gonor- STI clinics and providing HIV testing to partners of STI rhea and chlamydia not only in the women in the inter- clinic patients. Furthermore, this integration could vention but also in the miner population (Steen and Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 215 others 2000). In contrast, a cluster randomized trial of resulted in a temporary decrease in syphilis rates, but rates PPT conducted among female sex workers in Benin and rebounded rapidly and soon exceeded previous levels Ghana found substantial reductions in gonorrhea but not (Pourbohloul, Rekart, and Brunham 2003; Rekart and in chlamydia among sex workers themselves after nine others 2003). The intervention might have contributed to months, and no impact on the prevalence of gonorrhea or the rebound by increasing the number of people suscepti- chlamydia among their clients (Labbe and others 2012). ble to infection (Pourbohloul, Rekart, and Brunham 2003). Although PPT can be effective, interventions targeting Emergence of azithromycin-resistant Treponema pallidum high-risk groups do not have to include PPT to be effective. occurred during the intervention (Mabey 2009). The For example, Avahan, the India AIDS Initiative, offers com- impact on azithromycin resistance of other bacteria was bination interventions for high-risk groups that include not studied. For these and other reasons, researchers have activities such as peer-based education, clinical services for cautioned that mass treatment interventions should not be STIs, condom promotion and distribution, and commu- undertaken routinely (Pourbohloul, Rekart, and Brunham nity mobilization. Among female sex workers in 2003; Rekart and others 2003). Maharashtra, India, Avahan led to significant declines in the prevalence of syphilis, chlamydia, and gonorrhea (Mainkar and others 2011). It also led to reductions in syphilis among Community and Structural Interventions high-risk MSM and male-to-female transgender persons STI prevention interventions can be implemented at the (Subramanian and others 2013). Peer-mediated interven- individual, risk group, or community level. Although tions have also shown promise among female sex workers this literature review is stratified by intervention modal- in Mombasa, Kenya, where peer-based STI and HIV educa- ity and not by level of implementation, most of the tion and condom promotion among female sex workers interventions reviewed thus far were targeted to individ- increased consistent condom use with clients, but these uals or high-risk groups. interventions did not have a statistically significant impact The MEMA kwa Vijana (“good things for young on STI acquisition (Luchters and others 2008). people”) intervention, a random community interven- tion in the Mwanza Region of Tanzania, examined the impact of a multipronged intervention that included Mass Treatment school-based sexual and reproductive health education, A community RCT in Rakai, Uganda, evaluated the youth-friendly health services, peer condom promotion, efficacy of repeated mass treatment of STIs. The preva- and community activities. Although the intervention lence of syphilis seropositivity and trichomoniasis infec- increased knowledge and decreased reported risk behav- tion in women was significantly lower in intervention iors, it had no apparent effect on HIV or HSV-2 seroinci- communities than in control communities, but there dence, incidence of other STIs, or pregnancy outcomes at was no significant reduction in the prevalence of other the end of the trial (Hayes and others 2005), and no effect STIs. However, in a subanalysis of pregnant women, the on HIV after about 10 years (Doyle and others 2010). prevalence of trichomoniasis, bacterial vaginosis, gonor- Community-based interventions have also been used rhea, and chlamydia was significantly lower in commu- to improve the quality of syndromic management of nities that received mass treatment. STIs. A district RCT in Lima, Peru, examined an inter- Although rigorous evaluations of the population-level vention to improve the recognition and management of impact of mass treatment strategies in LMICs are rare, STI syndromes by pharmacy workers (Garcia and others mathematical modeling exercises suggest that mass 2003). The intervention was found to improve STI rec- treatment combined with sustained syndromic manage- ognition and management, as well as STI and HIV ment could be an effective STI control strategy and risk-reduction counseling. A subsequent trial that chose substantially reduce STI-attributable HIV incidence 20 cities throughout Peru to receive or not receive this (Korenromp and others 2000). This model suggests that intervention resulted in substantial and significant the impact of a single round of mass treatment on STI improvements in STI syndromic management at phar- incidence would be temporary without continued macies in the intervention cities but not in the control rounds of mass treatment or a sustained complementary cities. The community trial in Peru (Peru PREVEN intervention, such as syndromic management. Study) combined four intervention modalities: In general, however, mass treatment is discouraged because of its cost, adverse effects, promotion of resistance, • Provision of training, workshops, and educational and other factors (Mayaud and Mabey 2004). For example, materials to pharmacy workers and clinicians a targeted mass treatment program to provide azithromy- • STI screening and treatment for female sex workers cin to more than 4,000 at-risk persons in British Columbia by mobile outreach teams 216 Major Infectious Diseases • Provision of PPT using metronidazole to female sex Charania and others (2011) concluded that structural workers with bacterial vaginosis interventions to increase the availability of condoms do • Condom promotion among female sex workers by increase condom use, based on their review of 21 pub- mobile outreach teams and among the general pop- lished studies. However, a Cochrane Review of nine ulation by social marketing of low-cost condoms RCTs of structural and community-level interventions (Garcia and others 2012). to increase condom use found no evidence that these interventions reduced HIV or STIs (Moreno and others Adjusted for baseline prevalence, among 12,930 2014). These findings are not necessarily contradictory, young adults ages 18–29 years there was a nonsignificant given key differences in their approaches. For example, reduction in chlamydia, trichomoniasis, and gonorrhea unlike the review by Moreno and others (2014), the infection and in syphilis seroreactivity. However, signifi- review by Charania and others (2011) focused exclu- cant reductions were noted in certain subgroups, specif- sively on structural interventions, was not limited to ically young adult women and female sex workers in RCTs, and examined behavioral outcomes (condom use) intervention cities. rather than health outcomes (STI or HIV incidence). Randomized trials at the clinic level offer comparable A cash transfer program was tested in a trial of findings in Pakistan (Shah and others 2007) and South never-married women ages 13–22 years in Zomba Africa (Harrison and others 2000). A cluster randomized District of Malawi (Baird and others 2012). The provi- trial in rural Vietnam showed that educational programs sion of cash was intended to increase household income with interactive training can increase STI-related knowl- and sustain school enrollment in an attempt to offset two edge and practices of health care providers such as phar- possible risk factors for HIV and STIs: poverty and lack macists, doctors, and nurses (Lan and others 2014). of education. The cash transfer program was shown to Structural (or environmental) interventions to pre- reduce HIV and HSV-2 incidence, indicating high effec- vent STIs, including HIV, seek to change the physical and tiveness in a low-income setting (Baird and others 2012). social environments in which risky sexual behavior takes Alcohol control policies (alcohol taxation and restric- place, with a focus on making healthy options the default tions on advertising) have been proposed to reduce STIs choice (Frieden 2010; Kerrigan and others 2006). and HIV/AIDS in Sub-Saharan Africa (Chersich and Government policies and regulations are a common others 2009), given research linking alcohol consump- example of structural interventions. A government policy tion to risky sexual behaviors. In HICs, alcohol control in Puerto Plata, the Dominican Republic, requiring con- policies have been associated with substantial declines in dom use between sex workers and clients (with penalties alcohol-related health outcomes, such as motor vehicle for violations incurred by owners of sex establishments), fatalities and homicides (Cook and Durrance 2013). combined with a community-solidarity intervention, was They have also been shown to reduce risky sexual behav- associated with a 50 percent reduction in STI prevalence iors and STI incidence and to improve sexual health among female sex workers (Kerrigan and others 2006). (Chesson, Harrison, and Kassler 2000; Cohen and others This reduction was more substantial than that observed 2006; Dee 2001; Grossman, Kaestner, and Markowitz in Santo Domingo, the Dominican Republic, which 2005; Sen and Luong 2008; Staras and others 2014). received the community-solidarity intervention alone. Grossman, Kaestner, and Markowitz (2005) found that a Although not evaluated through an RCT or compar- 10 percent increase in the state excise taxes on beer was ative effectiveness design, the Thai government’s response associated with lower gonorrhea rates among males ages to prevention of HIV in the late 1980s and early 1990s 15–24 years in the United States. Dee (2001) estimated provides compelling evidence of the potential impact of that establishing a minimum legal drinking age of 21 structural interventions. The response included three years in the United States reduced childbearing by about main components: the provision of condoms to com- 6 percent among black teenagers. mercial sex venues, the imposition of sanctions on com- mercial sex venues not adhering to the 100 percent condom use policy, and a mass advertising campaign COST-EFFECTIVENESS OF STI PREVENTION advising men to use condoms with commercial sex workers (Hanenberg and others 1994). Within four INTERVENTIONS: LITERATURE REVIEW years, condom use in commercial sex acts increased to 94 The cost-effectiveness of STI prevention interventions percent from 14 percent; STIs in males declined about 80 depends in part on the degree to which reductions in percent, with notable reductions in HIV incidence as STIs other than HIV might influence the HIV epidemic. well (Hanenberg and others 1994; Punpanich, Ungchusak, In general, the estimated cost-effectiveness of STI pre- and Detels 2004). vention interventions is much higher if the potential Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 217 benefits of preventing STI-attributable HIV transmis- interventions in LMICs. The search was conducted sion or acquisition are included. Modeling exercises have through July 2014 using the same databases as those suggested that syndromic management of STIs can have listed for the literature search on effective STI preven- a substantial influence on HIV incidence in LMICs and tion interventions. Search terms used to identify the be cost saving in many scenarios (White and others relevant literature are provided in annexes 10A and 2008). However, given the scientific debate regarding the 10B. This search was supplemented with additional effects on HIV of STI treatment and prevention, some sources, such as the bibliographies of articles obtained experts have advised assessing the cost-effectiveness of in the search. Costs and cost-effectiveness ratios have STI prevention interventions without considering the been updated to 2012 U.S. dollars. The cost-effectiveness potential impacts on HIV (Galarraga and others 2009). of programs to prevent syphilis during pregnancy and This section focuses on studies that assess the cost- HIV/AIDS are not included in this review because effectiveness of STI prevention in its own right, without this topic is addressed in chapter 6 of this volume regard to the potential effects on HIV. (John-Stewart and others 2017). Table 10.3 summarizes A literature search was conducted to identify stud- selected studies of the cost-effectiveness of prevention ies of the cost-effectiveness of STI prevention interventions. Table 10.3 Selected Cost-Effectiveness Analyses of Primary STI Prevention Interventions in Low- and Middle-Income Countries Type of intervention Setting of Key cost-effectiveness results (2012 and study Description of intervention evaluated intervention U.S. dollars unless noted otherwise) Behavior change interventions Chong and others 2013 Online education program addressing Adolescents Cost per STI averted ranged from $95 to topics such as sexual rights, contraception, attending Colombian $824, depending on assumptions regarding condom use and STIs and HIV, public schools duration of intervention’s effect. empowerment, and violence prevention. HPV vaccination Goldie and others 2008; HPV vaccination of females, either alone LMICs eligible for HPV vaccination of females can be highly Levin and others 2015; or in combination with cervical cancer support from Gavi, cost-effective even in the poorest countries. Natunen and others 2013 screening. the Vaccine Alliance STI case management Sahin-Hodoglugil and Three protocols for diagnosing and treating Women in Africa Cost-effectiveness of each protocol varied others 2003 gonorrhea and chlamydia: “gold-standard” by locale. Syndromic management had two care, syndromic management, and mass key advantages: low program costs and treatment. relative ease of implementation. Adams and others 2003 Training pharmacists in syndromic Peru Intervention was cost saving from the management of urethral discharge and societal perspective. genital ulcer disease in males and vaginal discharge and pelvic inflammatory disease in females. Colvin and others 2006 Provided syndromic management packets Durban, South A savings of US$2.39 occurred per (including an information leaflet and Africa additional patient appropriately managed appropriate antibiotics) to primary care for urethral discharge in males and vaginal clinics. discharge and lower abdominal pain in females. Vickerman, Ndowa, and Modification in STI treatment guidelines LMICs Although the incorporation of HSV-2 Mayaud 2008 for syndromic management of genital ulcer treatment could increase program costs, it disease that incorporated antiviral treatment could potentially increase the proportion of for HSV-2 in certain situations. herpetic ulcers treated, while reducing the cost per ulcer appropriately treated. table continues next page 218 Major Infectious Diseases Table 10.3 Selected Cost-Effectiveness Analyses of Primary STI Prevention Interventions in Low- and Middle-Income Countries (continued) Type of intervention Setting of Key cost-effectiveness results (2012 and study Description of intervention evaluated intervention U.S. dollars unless noted otherwise) Targeted interventions and periodic presumptive treatment Borghi and others 2005 Voucher scheme to increase STI services for Managua, Average cost per STI cured was US$140.17. high-risk groups, including sex workers and Nicaragua their clients. Carrara and others 2005 Providing STI clinical services and outreach Cambodia Average cost per syndrome cured or to female sex workers and their male clients improved was about US$84.35–US$154.34 through nongovernmental organizations, for men and US$89.73–US$154.34 for with a focus on the management of genital women. discharge syndrome and genital ulcer syndrome. Marseille and others 2001 Distribution of 6,000 female condoms to Rural South Africa The intervention was estimated to have female sex workers and to women with at averted 6 HIV infections, 33 gonorrhea least one casual partner per year. infections, and 38 syphilis infections and to pay for itself in averted HIV and STI treatment costs. Lafort and others 2010 Establishing a dedicated clinic specifically Northern Cost per clinic visit was US$4.76. for high-risk populations (female sex Mozambique workers and long-distance truck drivers). Mass treatment Sahin-Hodoglugil and Mass treatment compared with syndromic Women in Africa Mass treatment offered relative advantages others 2003 management. in number and percentage of cases cured, but relative disadvantages in overall program costs and costs associated with overtreatment. Community and structural interventions Sweat and others 2004 Structural intervention was a system of Dominican Republic The cost per DALY averted was sanctions levied on sex establishment US$1,468.94 with the environmental owners for failing to follow government intervention alone. When the structural policies requiring condom use during sex intervention was included along with the work; administered in conjunction with an environmental intervention, the estimated environmental intervention that included number of HIV infections averted more than community mobilization, peer education, and doubled and the cost per DALY averted was distribution of educational materials and US$566.02. promotional items. Note: DALY = disability-adjusted life year; HIV = human immunodeficiency virus; HPV = human papillomavirus; HSV = herpes simplex virus; LMICs = low- and middle-income countries; STI = sexually transmitted infection. Primary Prevention Colombia (Chong and others 2013). The intervention Behavior Change Interventions addressed topics such as sexual rights, contraception, Although several studies have examined the cost- condom use and STIs and HIV, empowerment, and vio- effectiveness of behavioral interventions to prevent HIV lence prevention. The findings suggested a cost per STI in LMICs (McCoy, Kangwende, and Padian 2010; averted of US$95–US$824, depending on assumptions Townsend, Mathews, and Zembe 2013), the literature about the duration of the intervention’s effect (table 10.3). search yielded only one study of the cost-effectiveness of behavioral interventions to prevent other STIs. The study Male Circumcision assessed the cost-effectiveness of an online education Adult male circumcision is a cost-effective and potentially program for adolescents attending public schools in cost-saving intervention for preventing heterosexual Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 219 acquisition of HIV in men, according to a review of by locale, depending on STI prevalence, program cover- published studies (Uthman and others 2010). The cost- age, and health-seeking behavior. Syndromic manage- effectiveness of male circumcision to prevent STIs (other ment had two key advantages—low program costs and than HIV) has not been analyzed and is not of vital relative ease of implementation—which likely explains importance given that preventing HIV is the main goal of why it is often used in resource-poor settings. This find- adult male circumcision. ing is consistent with a systematic review of the costs of treating curable STIs in LMICs (Terris-Prestholt and HPV Vaccination others 2006), which found that syndromic management Although HPV vaccination is a relatively new interven- had lower costs than other management strategies. tion, a substantial body of research examines its cost- However, syndromic management had a lower estimated effectiveness in LMICs. Levin and others (2015) and impact on the percentage of chlamydia and gonorrhea Natunen and others (2013) reviewed the cost-effectiveness cases cured than the gold-standard or mass treatment of HPV vaccination in LMICs with high rates of cervical options (Sahin-Hodoglugil and others 2003). cancer. Goldie and others (2008) assessed the cost- Three studies examined the cost-effectiveness of strat- effectiveness of HPV vaccination in 72 countries eligible egies to improve the quality of syndromic management: for support from Gavi, the Vaccine Alliance. Two key themes emerge from this literature. First, HPV vaccina- • An intervention in Peru that trained pharmacists in tion of females (either alone or in combination with cer- syndromic management of urethral discharge and vical cancer screening) can be highly cost-effective even in genital ulcer disease in males and vaginal discharge the poorest countries. Second, despite favorable cost- and pelvic inflammatory disease in females was found effectiveness, HPV vaccine programs will likely not be to be cost saving from the societal perspective (Adams affordable in many countries. Gavi has helped address the and others 2003). affordability issue, making HPV vaccine available at less • An intervention in Durban, South Africa, that pro- than US$5 per dose to Gavi-eligible countries as of 2014, vided syndromic management packets (including an compared with more than US$100 per dose in HICs.2 information leaflet and appropriate antibiotics) to primary care clinics was found to cost US$2.39 per Microbicides additional patient appropriately managed for urethral A modeling study suggested that a hypothetical microbi- discharge in males and vaginal discharge and lower cide with 55 percent efficacy in preventing male-to- abdominal pain in females (Colvin and others 2006). female HIV transmission would be highly cost-effective • Vickerman, Ndowa, and Mayaud (2008) examined in LMICs with generalized epidemics, but it likely would the cost-effectiveness of a modification in the 2003 be less cost-effective in HICs (Verguet and Walsh 2010). WHO guidelines for syndromic management of geni- In another modeling study, Terris-Prestholt and others tal ulcer disease that incorporated antiviral treatment (2014) calculated per act efficacy against HIV and for HSV-2 in certain situations, such as in popula- HSV-2 consistent with the overall efficacy of the 1 tions with HSV-2 prevalence of 30 percent or more. percent vaginal gel formulation of tenofovir, as reported Although the incorporation of HSV-2 treatment could in the CAPRISA trial (Karim and others 2010). Results increase program costs, it could potentially increase indicate that the gel could be cost-effective or cost saving the proportion of herpetic ulcers treated while reduc- in LMICs, depending on its price. However, subsequent ing the cost per ulcer appropriately treated. trials have not confirmed that the gel reduces HIV/AIDS acquisition. The literature search yielded no cost- Incorporating HSV-2 treatment could be an affordable effectiveness analyses of using microbicides strictly to and cost-effective strategy in certain situations, depending prevent STIs other than HIV/AIDS. on factors such as the cost of HSV-2 therapy. Perhaps more important, the implementation of syndromic man- agement for genital ulcers that includes treatment for STI Case Management chancroid, in accordance with the WHO guidelines Sahin-Hodoglugil and others (2003) used a decision tree (WHO 2003), has been credited with major reductions in model to examine the cost-effectiveness of three proto- or even elimination of chancroid in many parts of the cols for diagnosing and treating gonorrhea and chlamy- world (Ryan, Kamb, and Holmes 2008; Spinola 2008; dia in women in Sub-Saharan Africa: gold-standard care Steen 2001). To the extent that chancroid is an important (use of the best available yet expensive diagnostic tests), risk factor for HIV transmission, syndromic management syndromic management, and mass treatment. They to reduce its incidence may be exceptionally cost-effective found that the cost-effectiveness of each strategy varied (Makasa, Buve, and Sandøy 2012). 220 Major Infectious Diseases Partner Notification and Management clinic cost of US$2,233.02 and treatment of 475 clients The literature search yielded no cost-effectiveness per month (Lafort and others 2010). Expanding the analyses of partner notification and partner manage- hours of operation, widening the geographic coverage of ment strategies for STIs other than HIV in LMICs. the clinic, and targeting additional risk groups could However, partner management strategies for HIV illus- reduce the cost per client served. trate the potential for such strategies to be cost-effective for STIs. For example, Rutstein and others (2014) Mass Treatment found that, in Sub-Saharan Africa, the incremental cost per HIV transmission averted was US$3,014.93 for Only one cost-effectiveness analysis of mass treatment “contract” notification, in which there is an agreement strategies in LMICs was found (Sahin-Hodoglugil and that the provider will attempt to notify partners if the others 2003). Their decision tree analysis suggested index patient fails to do so within one week, compared that mass treatment offered relative advantages over with passive referral in which the index patient is gold-standard care and syndromic management in encouraged to notify partners. Furthermore, partner number and percentage of cases cured, but relative notification is regarded as an efficient approach to disadvantages in overall program costs and costs asso- identifying HIV-positive individuals in need of therapy ciated with overtreatment. The decision trees used in and also identifies HIV-negative partners who may the analysis did not account for the potential for mass benefit from PrEP. treatment to promote antimicrobial resistance or for the potential adverse effects on persons treated unnecessarily. Targeted Interventions and Periodic Presumptive Treatment Borghi and others (2005) examined the cost-effectiveness Community and Structural Interventions of a voucher scheme implemented in Managua, Sweat and others (2006) examined the cost-effectiveness Nicaragua, to increase STI services for high-risk groups, of environmental and structural interventions to pre- including sex workers and their clients. The vouchers vent HIV among female sex workers in the Dominican covered free STI services from a range of providers. The Republic. The environmental intervention consisted of analysis focused on the cost of treating four STIs, and the activities such as community mobilization, peer educa- incremental cost per STI cured by the voucher interven- tion, and distribution of educational materials and tion was US$140.17. promotional items. The structural intervention was a Carrara and others (2005) examined the cost- system of sanctions levied on sex establishment owners effectiveness of providing STI clinical services and out- for failing to follow government policies requiring reach to female sex workers and their male clients in condom use during sex work. Accordingly, the Cambodia through nongovernmental organizations. structural intervention consisted of holding the estab- The analysis focused on the management of genital dis- lishment owners—not the commercial sex workers— charge syndrome and genital ulcer syndrome; the aver- responsible for ensuring that condoms were used age cost per syndrome cured or improved was about consistently in all commercial sex transactions in US$84.35 to US$154.34 for men and US$89.73 to the establishment. The cost per DALY averted was US$154.34 for women. US$1,468.94 with the environmental intervention Marseille and others (2001) examined the cost- alone. When the structural intervention was included effectiveness of an intervention to distribute female con- along with the environmental intervention, the esti- doms to female sex workers and to women with at least mated number of HIV infections averted more than one casual partner per year. The distribution of 6,000 doubled and the cost per DALY averted was reduced to female condoms was expected to avert 6 HIV infections, US$566.02. Although the cost-effectiveness ratios were 33 gonorrhea infections, and 38 syphilis infections and sensitive to various assumptions, the inclusion of the to pay for itself in averted HIV and STI treatment costs. structural intervention consistently resulted in more Increasing access to STI prevention services by estab- favorable cost-effectiveness estimates (Sweat and lishing a dedicated clinic specifically for high-risk popu- others 2006). lations could be a cost-effective strategy in LMICs. A Studies of the cost-effectiveness of structural inter- study of the costs and use of a nighttime clinic in north- ventions to prevent STIs in LMICs are rare, but struc- ern Mozambique for high-risk populations (female sex tural interventions could yield substantial and lasting workers and long-distance truck drivers) found a cost impacts at relatively low cost. For example, in a review per clinic visit of about US$4.76, based on a monthly of HIV prevention interventions in the United States, Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 221 alcohol taxation ranked as one of the most cost-effective Screening and Treatment for Syphilis in Pregnancy of all available interventions (Cohen, Wu, and Farley The prevention of mother-to-child transmission of HIV 2004). and syphilis is addressed in chapter 6 of this volume (John-Stewart and others 2017). However, screening and treatment for syphilis in pregnancy warrants special KEY ISSUES REGARDING IMPACT AND COST- mention here for several key reasons. First, the global EFFECTIVENESS burden of disease due to syphilis during pregnancy is comparable to that of mother-to-child transmission of HIV-Related Benefits of Preventing STIs Other HIV (WHO 2012b). Second, screening and treatment than HIV for syphilis in pregnancy is an inexpensive and highly Few published studies examine the cost-effectiveness of cost-effective intervention (Blandford and others 2007; interventions to control and prevent STIs other than Hawkes and others 2011; Kahn and others 2014; HIV in LMICs. The review focused primarily on studies Owusu-Edusei, Gift, and Ballard 2011; Rydzak and of the cost-effectiveness of prevention programs for spe- Goldie 2008; Schmid 2004; Terris-Prestholt and others cific STIs other than HIV that did not include costs 2003). However, despite their low cost and favorable averted and health benefits gained by preventing STI- cost-effectiveness, screening for and treatment of syphi- attributable HIV infections. The inclusion of potential lis in pregnancy are vastly underutilized in LMICs today HIV prevention benefits could substantially alter the (WHO 2012b). estimated cost-effectiveness of STI control and preven- tion programs (Chesson and Pinkerton 2000), particu- larly those targeted to high-risk populations. To the HPV and HBV Vaccination extent that prevention or control of STIs reduces the Given the scarcity of published studies on the cost- incidence of HIV, any effective STI intervention would effectiveness of interventions to prevent STIs in LMICs, be expected to be cost-effective, provided that the inter- the exceptionalism of HPV and HBV vaccination war- vention itself is not excessively costly and that its effect rants mention. HPV vaccination is unique among STI on HIV is not too small. Furthermore, STI-related inter- prevention interventions in that its effectiveness has been ventions can sometimes be cost-effective by reducing the demonstrated in RCTs, and its cost-effectiveness in progression of HIV in people infected with both HIV LMICs has been analyzed extensively, as reviewed by and another STI. For example, Vickerman and others Natunen and others (2013) and Levin and others (2015). (2011) found that suppressive therapy for HSV-2 in Similar data exist for HBV vaccination (Kane 1995). women with HSV-2 and HIV could be a cost-effective However, young girls have limited access to HPV vaccine public health intervention based on the benefits of in poorer settings because of the high cost of the vac- reducing the progression of HIV and improving the cine and other challenges associated with vaccinating (de retention of women in care, a potential benefit of HSV-2 Sanjosé and others 2012; Kane 2010). Nonetheless, Gavi’s therapy suggested by Baggaley and others (2009). support for HPV vaccines is expected to increase access in To the extent that a variety of interventions targeting LMICs and eventually reduce the disproportionate bur- curable STIs might also reduce the risk of potentially den of HPV-associated cancers in these settings. fatal, incurable, and chronic STIs other than HIV such as sexually transmitted HPV, HBV, and HSV, the cost- effectiveness of such interventions would be more favor- Income and Income Inequality able when these additional benefits are included. Aral and others (2006) examined the association Conversely, certain HIV prevention interventions between two economic measures—income and income might also reduce other STIs. However, their cost- inequality and STI burden—at the country level. For effectiveness is generally not as sensitive to the inclusion each country setting, income was measured using gross of other STI-related benefits as the reverse. For example, national income, and income inequality was measured circumcision is a highly cost-effective (and potentially using the Gini coefficient, which can range from 0 (com- cost-saving) intervention for the prevention of HIV plete equality) to 1 (complete inequality). The burden of acquisition in men (Uthman and others 2010). Because STIs was negatively associated with income and posi- it is cost-effective when considering HIV-related benefits tively associated with income inequality. Their analysis alone, there is little need to include the potential benefits suggested that these two economic measures could of preventing other STIs, at least in settings where explain almost half of the variation across countries in prevention of other STIs is not the primary goal of STI prevalence among low-risk groups (16 percent of circumcision. the variation among high-risk groups). 222 Major Infectious Diseases These findings are consistent with other analyses in CONCLUSIONS HICs. Bingham and others (2014) used the Gini coeffi- cient to examine income inequality and gonorrhea inci- STIs impose a considerable health and economic burden dence rates across 11 countries. Their analysis showed globally. Primary prevention and control of STIs in significant positive associations between income LMICs can be an efficient use of resources, although the inequality and gonorrhea rates in women. Owusu- impact and cost-effectiveness of interventions can vary Edusei, Chesson, Leichliter, and others (2013) examined substantially across settings. Furthermore, estimates of county-level data in the United States and found that the cost-effectiveness of STI control in LMICs can be racial disparities in income were associated with racial subject to considerable uncertainty and might not be disparities in STI burden. One possible explanation is generalizable across settings. The findings of this litera- that racial income disparity contributes to residential ture review should be considered in light of the limita- segregation by race, which has been identified as a social tions inherent in cost-effectiveness studies of STI control determinant of STI risk (Hogben and Leichliter 2008; in LMICs, such as incomplete cost data and imprecise Owusu-Edusei, Chesson, Leichliter, and others 2013; estimates of program impact. Thomas and Gaffield 2003). Behavioral interventions can often lead to reductions in the risk of acquiring STIs, at least in the short term. In contrast, interventions with long-lasting effects—such Research and Development Agenda as adult male circumcision and HPV and HBV Aral and others (2006) provided the following list of vaccination—can have a more pronounced impact on priorities for global STI research, and they remain prior- disease burden at the individual and population levels. ities today: Given the challenges of providing STI prevention and treatment services in LMICs, structural interventions are • Development, evaluation, and implementation of needed to make it easier and more realistic for people to STI prevention and control interventions, including choose safer behaviors. Unfortunately, establishing that a therapeutic interventions such as drug treatment and given intervention is effective and cost-effective is not therapeutic vaccines, and primary prevention inter- enough to ensure its delivery. Screening for syphilis in ventions such as prophylactic vaccines, structural pregnancy remains vastly underutilized, even though it is interventions, and behavioral interventions relatively inexpensive, effective, and cost-effective. The • Enhanced efforts to control the spread of drug- underutilization of effective and cost-effective interven- resistant strains of gonorrhea tions highlights the need for more health services research • Development and evaluation of tools and methods and stronger health systems—not only to improve the for assessing the burden of STIs and STI-related delivery of STI prevention interventions in LMICs, but sequelae and for allocating STI prevention resources also to expand access to STI prevention services, espe- efficiently to reduce this burden cially among the most vulnerable populations. • Development and evaluation of tools to promote early detection and treatment of STIs, particularly inexpensive and practical rapid diagnostic tests for ANNEXES gonorrhea and chlamydia The annexes to this chapter are as follows. They are avail- • Development and evaluation of strategies to identify able at http://www.dcp-3.org/infectiousdiseases. persons at highest risk for STIs and to offer preven- tion services to reduce their risk of acquisition and • Annex 10A. Search Terms Used to Identify Literature transmission, especially to highly stigmatized popu- on the Impact of STI Prevention Interventions lations (MSM, transgender persons, and sex workers) • Annex 10B. Search Terms Used to Identify Literature • Promotion of health services research to inform on the Cost-Effectiveness of STI Prevention the integration of practical and cost-effective pre- Interventions vention strategies or systems into the public health infrastructure • Implementation of studies in support of global elim- NOTES ination programs, such as for syphilis and possibly World Bank Income Classifications as of July 2014 are as fol- cervical cancer lows, based on estimates of gross national income (GNI) per • Continued research on the importance of social capita for 2013: determinants of STIs, with the goal of reducing racial and geographic disparities in sexual health. • Low-income countries (LICs) = US$1,045 or less Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 223 • Middle-income countries (MICs) are subdivided: ———. 2014. “Behavioral Convergence: Implications for (a) lower-middle-income = US$1,046–US$4,125 Mathematical Models of Sexually Transmitted Infection (b) upper-middle-income (UMICs) = US$4,126–US$12,745 Transmission.” Journal of Infectious Diseases 210 (Suppl 2): • High-income countries (HICs) = US$12,746 or more. S600–04. Auvert, B., J. Sobngwi-Tambekou, E. Cutler, M. Nieuwoudt, 1. These numbers represent the WHO’s best estimates, using P. Lissouba, and others. 2009. “Effect of Male Circumcision indicators that aim for comparability across countries on the Prevalence of High-Risk Human Papillomavirus and time; they are updated as more recent or revised data in Young Men: Results of a Randomized Controlled Trial become available or when changes occur in the method- Conducted in Orange Farm, South Africa.” Journal of ology used. Visit the Global Health Observatory at http:// Infectious Diseases 199 (1): 14–19. apps.who.int/gho/data/?theme=main. Avila, D., O. Keiser, M. Egger, R. Kouyos, J. Boni, and others. 2. Visit the Gavi website at http://www.gavi.org. 2014. “Social Meets Molecular: Combining Phylogenetic and Latent Class Analyses to Understand HIV-1 Transmission in Switzerland.” American Journal of Epidemiology 179 (2): 1514–25. REFERENCES Baeten, J. M., D. Donnell, P. Ndase, N. R. Mugo, J. D. Campbell, Adams, E. S., P. J. Garcia, G. P. Garnett, W. J. Edmunds, and and others. 2012. “Antiretroviral Prophylaxis for HIV K. K. Holmes. 2003. “The Cost-Effectiveness of Syndromic Prevention in Heterosexual Men and Women.” New England Management in Pharmacies in Lima, Peru.” Sexually Journal of Medicine 367 (5): 399–410. Transmitted Diseases 30 (5): 379–87. Baggaley, R. F., J. T. Griffin, R. Chapman, T. D. Hollingsworth, Alam, N., E. Chamot, S. H. Vermund, K. Streatfield, and N. Nagot, and others. 2009. “Estimating the Public Health S. Kristensen. 2010. “Partner Notification for Sexually Impact of the Effect of Herpes Simplex Virus Suppressive Transmitted Infections in Developing Countries: A Therapy on Plasma HIV-1 Viral Load.” AIDS 23 (8): Systematic Review.” BMC Public Health 10: 19. 1005–13. Aral, S. O. 2002. “Understanding Racial-Ethnic and Societal Baird, S. J., R. S. Garfein, C. T. McIntosh, and B. Ozler. 2012. Differentials in STI.” Sexually Transmitted Infections 78 (1): 2–4. “Effect of a Cash Transfer Programme for Schooling on ———. 2011. “Utility and Delivery of Behavioural Interventions Prevalence of HIV and Herpes Simplex Type 2 in Malawi: A to Prevent Sexually Transmitted Infections.” Sexually Cluster-Randomised Trial.” The Lancet 379 (9823): 1320–29. Transmitted Infections 87 (Suppl 2): ii31–ii33. Bala, M., M. Kakran, V. Singh, S. Sood, and V. Ramesh. 2013. Aral, S. O., K. Bernstein, and E. Torrone. 2015. “Geographical “Monitoring Antimicrobial Resistance in Neisseria gonor- Targeting to Improve Progression through the Sexually rhoeae in Selected Countries of the WHO South-East Asia Transmitted Infection/HIV Treatment Continua in Region between 2009 and 2012: A Retrospective Analysis.” Different Populations.” Current Opinion in HIV and AIDS Sexually Transmitted Infections 89 (Suppl 4): iv28–35. 10 (6): 477–82. Baral, S., C. Beyrer, K. Muessig, T. Poteat, A. L. Wirtz, and oth- Aral, S. O., and J. F. Blanchard. 2012. “The Program Science ers. 2012. “Burden of HIV among Female Sex Workers in Initiative: Improving the Planning, Implementation, and Low-Income and Middle-Income Countries: A Systematic Evaluation of HIV/STI Prevention Programs.” Sexually Review and Meta-Analysis.” The Lancet Infectious Diseases Transmitted Infections 88 (3): 157–59. 12 (7): 538–49. Aral, S. O., and W. Cates Jr. 2013. “Coverage, Context, and Baral, S., F. Sifakis, F. Cleghorn, and C. Beyrer. 2007. “Elevated Targeted Prevention: Optimising Our Impact.” Sexually Risk for HIV Infection among Men Who Have Sex with Transmitted Infections 89 (4): 336–40. Men in Low- and Middle-Income Countries 2000–2006: Aral, S. O., and J. M. Douglas Jr., eds. 2007. Behavioral A Systematic Review.” PLoS Medicine 4 (12): e339. Interventions for Prevention and Control of Sexually Transmitted Becker, M. H., and J. G. Joseph. 1988. “AIDS and Behavioral Diseases. New York: Springer Science+Business Media. Change to Reduce Risk: A Review.” American Journal of Aral, S. O., M. Hogben, and J. N. Wasserheit. 2008. “STD- Public Health 78 (4): 394–410. Related Health Care Seeking and Health Service Delivery.” Benzaken, A., M. Sabido, E. Galban, D. L. Rodrigues Dutra, In Sexually Transmitted Diseases, edited by K. K. Holmes, A. L. Leturiondo, and others. 2012. “HIV and Sexually P. F. Sparling, W. E. Stamm, P. Piot, J. N. Wasserheit, and Transmitted Infections at the Borderlands: Situational others. New York: McGraw-Hill. Analysis of Sexual Health in the Brazilian Amazon.” Sexually Aral, S. O., M. Over, L. Manhart, and K. K. Holmes. 2006. “Sexually Transmitted Infections 88 (4): 294–300. Transmitted Infections.” In Disease Control Priorities in Bingham, A. L., A. M. Kavanagh, C. K. Fairley, L. A. Keogh, Developing Countries, second edition, edited by D. T. Jamison, R. J. Bentley, and others. 2014. “Income Inequality and J. G. Breman, A. R. Measham, G. Alleyne, M. Claeson, D. B. Neisseria gonorrhoeae Notifications in Females: A Country- Evans, P. Jha, A. Mills, and P. Musgrove, 311–30. Washington, Level Analysis.” Sexual Health 11 (6): 556–60. DC: World Bank and Oxford University Press. Blanchard, J. F., and S. O. Aral. 2010. “Emergent Properties and Aral, S. O., and H. Ward. 2005. “Modern Day Influences on Structural Patterns in Sexually Transmitted Infection and Sexual Behavior.” Infectious Disease Clinics of North America HIV Research.” Sexually Transmitted Infections 86 (Suppl 3): 19 (2): 297–309. iii4–9. 224 Major Infectious Diseases ———. 2011. “Program Science: An Initiative to Improve the Celum, C., A. Wald, J. R. Lingappa, A. S. Magaret, R. S. Wang, Planning, Implementation, and Evaluation of HIV/Sexually and others. 2010. “Acyclovir and Transmission of HIV-1 Transmitted Infection Prevention Programmes.” Sexually from Persons Infected with HIV-1 and HSV-2.” New Transmitted Infections 87 (1): 2–3. England Journal of Medicine 362 (5): 427–39. Blandford, J. M., T. L. Gift, S. Vasaikar, D. Mwesigwa-Kayongo, Charania, M. R., N. Crepaz, C. Guenther-Gray, K. Henny, P. Dlali, and others. 2007. “Cost-Effectiveness of On-Site A. Liau, and others. 2011. “Efficacy of Structural-Level Antenatal Screening to Prevent Congenital Syphilis in Rural Condom Distribution Interventions: A Meta-Analysis of Eastern Cape Province, Republic of South Africa.” Sexually U.S. and International Studies, 1998–2007 (Structured Transmitted Diseases 34 (Suppl 7): S61–66. Abstract).” AIDS and Behavior 15 (7): 1283–97. Boily, M. C., F. I. Bastos, K. Desai, and B. Masse. 2004. “Changes Chersich, M. F., H. V. Rees, F. Scorgie, and G. Martin. 2009. in the Transmission Dynamics of the HIV Epidemic after “Enhancing Global Control of Alcohol to Reduce Unsafe the Wide-Scale Use of Antiretroviral Therapy Could Explain Sex and HIV in Sub-Saharan Africa.” Global Health 5 (16). Increases in Sexually Transmitted Infections: Results from Chesson, H. W., T. S. Dee, and S. O. Aral. 2003. “AIDS Mortality Mathematical Models.” Sexually Transmitted Diseases 31: May Have Contributed to the Decline in Syphilis Rates 100–13. in the United States in the 1990s.” Sexually Transmitted Boily, M. C., and R. C. Brunham. 1993. “The Impact of HIV Diseases 30 (5): 419–24. and Other STDs on Human Populations: Are Predictions Chesson, H. W., and T. L. Gift. 2008. “Decreases in AIDS Possible?” Infectious Disease Clinics of North America 7 (4): Mortality and Increases in Primary and Secondary Syphilis 771–92. in Men Who Have Sex with Men in the United States.” Borghi, J., A. Gorter, P. Sandiford, and Z. Segura. 2005. Journal of Acquired Immune Deficiency Syndromes 47 (2): “The Cost-Effectiveness of a Competitive Voucher Scheme 263–64. to Reduce Sexually Transmitted Infections in High-Risk Chesson, H., P. Harrison, and W. J. Kassler. 2000. “Sex under Groups in Nicaragua.” Health Policy and Planning 20 (4): the Influence: The Effect of Alcohol Policy on Sexually 222–31. Transmitted Disease Rates in the United States.” Journal of Brown, L. B., W. C. Miller, G. Kamanga, N. Nyirenda, P. Mmodzi, Law and Economics 43 (1): 215–38. and others. 2011. “HIV Partner Notification Is Effective Chesson, H. W., R. D. Kirkcaldy, T. L. Gift, K. Owusu-Edusei and Feasible in Sub-Saharan Africa: Opportunities for HIV Jr., and H. S. Weinstock. 2014. “Ciprofloxacin Resistance Treatment and Prevention.” Journal of Acquired Immune and Gonorrhea Incidence Rates in 17 Cities, United States, Deficiency Syndromes 56 (5): 437–42. 1991–2006.” Emerging Infectious Diseases 20 (4): 612–19. Bukusi, E., K. K. Thomas, R. Nguti, C. R. Cohen, N. Weiss, Chesson, H. W., and S. D. Pinkerton. 2000. “Sexually Transmitted and others. 2011. “Topical Penile Microbicide Use by Men Diseases and the Increased Risk for HIV Transmission: to Prevent Recurrent Bacterial Vaginosis in Sex Partners: Implications for Cost-Effectiveness Analyses of Sexually A Randomized Clinical Trial.” Sexually Transmitted Diseases Transmitted Disease Prevention Interventions.” Journal of 38 (6): 483–89. Acquired Immune Deficiency Syndromes 24 (1): 48–56. Bull, S. S., and M. McFarlane. 2000. “Soliciting Sex on the Chesson, H. W., M. Sternberg, J. S. Leichliter, and S. O. Aral. Internet: What Are the Risks for Sexually Transmitted 2010a. “Changes in the State-Level Distribution of Primary Diseases and HIV?” Sexually Transmitted Diseases 27 (9): and Secondary Syphilis in the USA, 1985–2007.” Sexually 545–50. Transmitted Infections 86 (Suppl 3): iii58–62. Carrara, V., F. Terris-Prestholt, L. Kumaranayake, and P. Mayaud. ———. 2010b. “The Distribution of Chlamydia, Gonorrhoea, 2005. “Operational and Economic Evaluation of an NGO-Led and Syphilis Cases across States and Counties in the USA, Sexually Transmitted Infections Intervention: North-Western 2007.” Sexually Transmitted Infections 86 (Suppl 3): iii52–57. Cambodia.” Bulletin of the World Health Organization 83 (6): Chong, A., M. Gonzalez-Navarro, D. Karlan, and M. Valdivia. 434–42. 2013. “Effectiveness and Spillovers of Online Sex Education: Castellsague, X., F. X. Bosch, N. Muñoz, C. J. Meijer, K. V. Shah, Evidence from a Randomized Evaluation in Colombian and others. 2002. “Male Circumcision, Penile Human Public Schools.” Working Paper 18776, National Bureau of Papillomavirus Infection, and Cervical Cancer in Female Economic Research, Cambridge, MA. http://www.nber.org Partners.” New England Journal of Medicine 346 (15): /papers/w18776. 1105–12. Cohen, D. A., B. Ghosh-Dastidar, R. Scribner, A. Miu, M. Scott, Castellsague, X., R. W. Peeling, S. Franceschi, S. de Sanjosé, J. S. and others. 2006. “Alcohol Outlets, Gonorrhea, and the Smith, and others. 2005. “Chlamydia trachomatis Infection Los Angeles Civil Unrest: A Longitudinal Analysis.” Social in Female Partners of Circumcised and Uncircumcised Science and Medicine 62 (12): 3062–71. Adult Men.” American Journal of Epidemiology 162 (9): Cohen, D. A., S. Y. Wu, and T. A. Farley. 2004. “Comparing 907–16. the Cost-Effectiveness of HIV Prevention Interventions.” Celentano, D. D., K. C. Bond, C. M. Lyles, S. Eiumtrakul, Journal of Acquired Immune Deficiency Syndromes 37 (3): V. F. Go, and others. 2000. “Preventive Intervention to 1404–14. Reduce Sexually Transmitted Infections: A Field Trial in Colvin, M., M. O. Bachmann, R. K. Homan, D. Nsibande, the Royal Thai Army.” Archives of Internal Medicine 160 (4): N. M. Nkwanyana, and others. 2006. “Effectiveness and 535–40. Cost-Effectiveness of Syndromic Sexually Transmitted Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 225 Infection Packages in South African Primary Care: Cluster- Transmitted Diseases by Granting Sex Workers in Thailand Randomised Trial.” Sexually Transmitted Infections 82 (4): the Choice of Using the Male or Female Condom: Results 290–94. from a Randomized Controlled Trial.” AIDS 12 (14): 1851–59. Cook, P. J., and C. P. Durrance. 2013. “The Virtuous Tax: Ford, K., D. N. Wirawan, S. S. Suastina, B. D. Reed, and P. Muliawan. Lifesaving and Crime-Prevention Effects of the 1991 Federal 2000. “Evaluation of a Peer Education Programme for Female Alcohol-Tax Increase.” Journal of Health Economics 32 (1): Sex Workers in Bali, Indonesia.” International Journal of STD 261–67. and AIDS 11 (11): 731–33. Corey, L., A. Wald, R. Patel, S. L. Sacks, S. K. Tyring, and others Forman, D., C. de Martel, C. J. Lacey, I. Soerjomataram, 2004. “Once-Daily Valacyclovir to Reduce the Risk of J. Lortet-Tieulent, and others. 2012. “Global Burden of Transmission of Genital Herpes.” New England Journal of Human Papillomavirus and Related Diseases.” Vaccine Medicine 350 (1): 11–20. 30 (Suppl 5): F12–23. Cornier, N., E. Petrova, P. Cavailler, R. Dentcheva, F. Terris- Frieden, T. R. 2010. “A Framework for Public Health Action: Prestholt, and others. 2010. “Optimising the Management of The Health Impact Pyramid.” American Journal of Public Vaginal Discharge Syndrome in Bulgaria: Cost-Effectiveness Health 100 (4): 590–95. of Four Clinical Algorithms with Risk Assessment.” Sexually Future II Study Group. 2007. “Quadrivalent Vaccine against Transmitted Infections 86 (4): 303–9. Human Papillomavirus to Prevent High-Grade Cervical Dee, T. S. 2001. “The Effects of Minimum Legal Drinking Ages Lesions.” New England Journal of Medicine 356 (19): on Teen Childbearing.” Journal of Human Resources 36 (4): 1915–27. 823–38. Galarraga, O., M. A. Colchero, R. G. Wamai, and S. M. Bertozzi. de Sanjosé, S., B. Serrano, X. Castellsague, M. Brotons, 2009. “HIV Prevention Cost-Effectiveness: A Systematic J. Muñoz, and others. 2012. “Human Papillomavirus (HPV) Review.” BMC Public Health 9 (Suppl 1): S5. and Related Cancers in the Global Alliance for Vaccines Garcia, P., J. Hughes, C. Carcamo, and K. K. Holmes. 2003. and Immunization (GAVI) Countries: A WHO/ICO HPV “Training Pharmacy Workers in Recognition, Management, Information Centre Report.” Vaccine 30 (Suppl 4): 1–83. and Prevention of STDs: District-Randomized Controlled Dodd, P. J., G. P. Garnett, and T. B. Hallett. 2010. “Examining Trial.” Bulletin of the World Health Organization 81 (11): the Promise of HIV Elimination by ‘Test and Treat’ in 806–14. Hyperendemic Settings.” AIDS 24 (5): 729–35. Garcia, P. J., C. P. Carcamo, G. P. Garnett, P. E. Campos, Doyle, A. M., D. A. Ross, K. Maganja, K. Baisley, C. Masesa, and K. K. Holmes. 2012. “Improved STD Syndrome and others. 2010. “Long-Term Biological and Behavioural Management by a Network of Clinicians and Pharmacy Impact of an Adolescent Sexual Health Intervention in Workers in Peru: The PREVEN Network.” PLoS One Tanzania: Follow-Up Survey of the Community-Based 7 (10): e47750. MEMA kwa Vijana Trial.” PLoS Medicine 7 (6): e1000287. Garland, S. M., M. Hernandez-Avila, C. M. Wheeler, G. Perez, Drolet, M., E. Benard, M. C. Boily, H. Ali, L. Baandrup, D. M. Harper, and others. 2007. “Quadrivalent Vaccine against and others. 2015. “Population-Level Impact and Herd Human Papillomavirus to Prevent Anogenital Diseases.” Effects Following Human Papillomavirus Vaccination New England Journal of Medicine 356 (19): 1928–43. Programmes: A Systematic Review and Meta-Analysis.” The Giuliano, A. R., J. M. Palefsky, S. Goldstone, E. D. Moreira, Lancet Infectious Diseases 15 (5): 565–80. M. E. Penny, and others. 2011. “Efficacy of Quadrivalent Fairley, C. K., J. S. Hocking, L. C. Gurrin, M. Y. Chen, B. Donovan, HPV Vaccine against HPV Infection and Disease in Males.” and others. 2009. “Rapid Decline in Presentations for Genital New England Journal of Medicine 364 (5): 401–11. Warts after the Implementation of a National Quadrivalent Golden, M. R., R. P. Kerani, M. Stenger, J. P. Hughes, M. Aubin, Human Papillomavirus Vaccination Program for Young and others. 2015. “Uptake and Population-Level Impact Women.” Sexually Transmitted Infections 85 (7): 499–502. of Expedited Partner Therapy (EPT) on Chlamydia tra- Feldblum, P. J., T. Hatzell, K. Van Damme, M. Nasution, chomatis and Neisseria gonorrhoeae: The Washington A. Rasamindrakotroka, and others. 2005. “Results of a State Community-Level Randomized Trial of EPT.” PLoS Randomised Trial of Male Condom Promotion among Medicine 12 (1): e1001777. Madagascar Sex Workers.” Sexually Transmitted Infections Golden, M. R., W. L. Whittington, H. H. Handsfield, J. P. Hughes, 81 (2): 166–73. W. E. Stamm, and others. 2005. “Effect of Expedited Treatment Ferreira, A., T. Young, C. Mathews, M. Zunza, and N. Low. of Sex Partners on Recurrent or Persistent Gonorrhea or 2013. “Strategies for Partner Notification for Sexually Chlamydial Infection.” New England Journal of Medicine Transmitted Infections, Including HIV.” Cochrane Database 352 (7): 676–85. of Systematic Reviews 10: CD002843. Goldie, S. J., M. O’Shea, N. G. Campos, M. Diaz, S. Sweet, and Flagg, E. W., R. Schwartz, and H. Weinstock. 2013. “Prevalence S. Y. Kim. 2008. “Health and Economic Outcomes of HPV of Anogenital Warts among Participants in Private Health 16, 18 Vaccination in 72 GAVI-Eligible Countries.” Vaccine Plans in the United States, 2003–2010: Potential Impact of 26 (32): 4080–93. HPV Vaccination.” American Journal of Public Health 103 Grant, R. M., J. R. Lama, P. L. Anderson, V. McMahan, A. Y. Liu, (8): 1428–35. and others. 2010. “Preexposure Chemoprophylaxis for HIV Fontanet, A. L., J. Saba, V. Chandelying, C. Sakondhavat, Prevention in Men Who Have Sex with Men.” New England P. Bhiraleus, and others. 1998. “Protection against Sexually Journal of Medicine 363 (27): 2587–99. 226 Major Infectious Diseases Gray, R. H., G. Kigozi, D. Serwadda, F. Makumbi, F. Nalugoda, and Innovative HIV Prevention Intervention for Developing others. 2009. “The Effects of Male Circumcision on Female Countries.” Sexually Transmitted Diseases 40 (12): 909–14. Partners’ Genital Tract Symptoms and Vaginal Infections in Hogben, M., and J. S. Leichliter. 2008. “Social Determinants a Randomized Trial in Rakai, Uganda.” American Journal of and Sexually Transmitted Disease Disparities.” Sexually Obstetrics and Gynecology 200 (1): 42–47. Transmitted Diseases 35 (12): S13–18. Gregson, S., S. Adamson, S. Papaya, J. Mundondo, C. A. Hoke, T. H., P. J. Feldblum, K. V. Damme, M. D. Nasution, Nyamukapa, and others. 2007. “Impact and Process T. W. Grey, and others. 2007. “Randomised Controlled Evaluation of Integrated Community and Clinic-Based Trial of Alternative Male and Female Condom Promotion HIV-1 Control: A Cluster-Randomised Trial in Eastern Strategies Targeting Sex Workers in Madagascar.” Sexually Zimbabwe.” PLoS Medicine 4 (3): e102. Transmitted Infections 83 (6): 448–53. Grosskurth, H., F. Mosha, J. Todd, E. Mwijarubi, A. Klokke, and Holmes, K. K., and S. O. Aral. 1991. “Behavioral Interventions others. 1995. “Impact of Improved Treatment of Sexually in Developing Countries.” In Research Issues in Human Transmitted Diseases on HIV Infection in Rural Tanzania: Behavior and STD in the AIDS Era, edited by J. Wasserheit, Randomised Controlled Trial.” The Lancet 346 (8974): S. Aral, and K. Holmes, 318–44. Washington, DC: American 530–36. Society of Microbiology Publications. Grossman, M., R. Kaestner, and S. Markowitz. 2005. “An International Futures Program. 2011. The Future of Families to Investigation of the Effects of Alcohol Policies on Youth 2030: A Synthesis Report. Paris: OECD Publications. STDs.” Advances in Health Economics and Health Services Jewkes, R., M. Nduna, J. Levin, N. Jama, K. Dunkle, and others. Research 16: 229–56. 2008. “Impact of Stepping Stones on Incidence of HIV Guffey, M. B., B. Richardson, M. Husnik, B. Makanani, and HSV-2 and Sexual Behaviour in Rural South Africa: D. Chilongozi, and others. 2014. “HPTN 035 Phase II/IIb Cluster-Randomised Controlled Trial.” British Medical Randomised Safety and Effectiveness Study of the Vaginal Journal 337: a506. Microbicides BufferGel and 0.5% PRO 2000 for the John-Stewart, G., R. Peeling, C. Levin, P. J. Garcia, D. Mabey, Prevention of Sexually Transmitted Infections in Women.” and J. Kinuthia. 2017. “Prevention of Mother-to-Child Sexually Transmitted Infections 90 (5): 363–69. Transmission of HIV and Syphilis.” In Disease Control Hanenberg, R. S., W. Rojanapithayakorn, P. Kunasol, and Priorities (third edition): Volume 6, Major Infectious D. C. Sokal. 1994. “Impact of Thailand’s HIV-Control Diseases, edited by K. K. Holmes, S. Bertozzi, B. R. Bloom, Programme as Indicated by the Decline of Sexually and P. Jha. Washington, DC: World Bank. Transmitted Diseases.” The Lancet 344 (8917): 243–45. Joura, E. A., A. R. Giuliano, O. E. Iversen, C. Bouchard, Hankins, C. A., S. R. Friedman, T. Zafar, and S. A. Strathdee. C. Mao, and others. 2015. “A 9-Valent HPV Vaccine against 2002. “Transmission and Prevention of HIV and Sexually Infection and Intraepithelial Neoplasia in Women.” New Transmitted Infections in War Settings: Implications for England Journal of Medicine 372 (8): 711–23. Current and Future Armed Conflicts.” AIDS 16: 2245–52. Kahn, J. G., A. Jiwani, G. B. Gomez, S. J. Hawkes, H. W. Chesson, Hariri, S., L. E. Markowitz, E. F. Dunne, and E. R. Unger. 2013. and others. 2014. “The Cost and Cost-Effectiveness “Population Impact of HPV Vaccines: Summary of Early of Scaling up Screening and Treatment of Syphilis in Evidence.” Journal of Adolescent Health 53 (6): 679–82. Pregnancy: A Model.” PLoS One 9: e87510. Harper, D. M., E. L. Franco, C. Wheeler, D. G. Ferris, D. Jenkins, Kamali, A., M. Quigley, J. Nakiyingi, J. Kinsman, J. Kengeya- and others. 2004. “Efficacy of a Bivalent L1 Virus-Like Kayondo, and others. 2003. “Syndromic Management Particle Vaccine in Prevention of Infection with Human of Sexually Transmitted Infections and Behaviour Change Papillomavirus Types 16 and 18 in Young Women: A Interventions on Transmission of HIV-1 in Rural Uganda: A Randomized Trial.” The Lancet 364 (9447): 1757–65. Community Randomised Trial.” The Lancet 361 (9358): 645–52. Harrison, A., S. A. Karim, K. Floyd, C. Lombard, M. Lurie, Kamb, M. L., E. Lackritz, J. Mark, D. B. Jackson, and and others. 2000. “Syndrome Packets and Health H. L. Andrews. 2007. “Sexually Transmitted Infections in Worker Training Improve Sexually Transmitted Disease Developing Countries: Current Concepts and Strategies Case Management in Rural South Africa: Randomized on Improving STI Prevention, Treatment, and Control.” Controlled Trial.” AIDS 14 (17): 2769–79. Working Paper 42797, World Bank, Washington, DC. Hawkes, S., N. Matin, N. Broutet, and N. Low. 2011. Kamb, M. L., L. M. Newman, P. L. Riley, J. Mark, S. J. Hawkes, “Effectiveness of Interventions to Improve Screening for and others. 2010. “A Road Map for the Global Elimination Syphilis in Pregnancy: A Systematic Review and Meta- of Congenital Syphilis.” Obstetrics and Gynecology Analysis.” The Lancet Infectious Diseases 11 (9): 684–91. International 2010: 312798. Hayes, R. J., J. Changalucha, D. A. Ross, A. Gavyole, J. Todd, Kane, M. A. 1995. “Global Programme for Control of Hepatitis and others. 2005. “The MEMA kwa Vijana Project: Design B Infection.” Vaccine 13 (Suppl 1): S47–49. of a Community Randomised Trial of an Innovative ———. 2010. “Global Implementation of Human Adolescent Sexual Health Intervention in Rural Tanzania.” Papillomavirus (HPV) Vaccine: Lessons from Hepatitis B Contemporary Clinical Trails 26 (4): 430–42. Vaccine.” Gynecologic Oncology 117 (Suppl 2): S32–35. Henley, C., G. Forgwei, T. Welty, M. Golden, A. Adimora, and Karim, Q. A., and C. Baxter. 2013. “Microbicides for the others. 2013. “Scale-Up and Case-Finding Effectiveness Prevention of Sexually Transmitted HIV Infection.” Expert of an HIV Partner Services Program in Cameroon: An Review of Anti-Infective Therapy 11 (1): 13–23. Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 227 Karim, Q. A., S. S. A. Karim, J. A. Frohlich, A. C. Grobler, La Ruche, G., F. Lorougnon, and N. Digbeu. 1995. “Therapeutic C. Baxter, and others. 2010. “Effectiveness and Safety Algorithms for the Management of Sexually Transmitted of Tenofovir Gel, an Antiretroviral Microbicide, for the Diseases at the Peripheral Level in Côte d’Ivoire: Prevention of HIV Infection in Women.” Science 329 Assessment of Efficacy and Cost.” Bulletin of the World (5996): 1168–74. Health Organization 73 (3): 305–13. Katz, I. T., and A. A. Wright. 2008. “Circumcision: A Surgical Leichliter, J. S., H. W. Chesson, M. Sternberg, and S. O. Aral. Strategy for HIV Prevention in Africa.” New England 2010. “The Concentration of Sexual Behaviours in the Journal of Medicine 359 (23): 2412–15. USA: A Closer Examination of Subpopulations.” Sexually Kaul, R., J. Kimani, N. J. Nagelkerke, K. Fonck, E. N. Ngugi, and Transmitted Infections 86 (Suppl 3): iii45–51. others. 2004. “Monthly Antibiotic Chemoprophylaxis and Levin, C. E., M. Sharma, Z. Olson, S. Verguet, J. F. Shi, and others. Incidence of Sexually Transmitted Infections and HIV-1 2015. “An Extended Cost-Effectiveness Analysis of Publicly Infection in Kenyan Sex Workers: A Randomized Controlled Financed HPV Vaccination to Prevent Cervical Cancer in Trial.” Journal of the American Medical Association 291 (21): China.” In Disease Control Priorities (third edition): Volume 3, 2555–62. Cancer, edited by H. Gelband, P. Jha, R. Sankaranarayanan, Kault, D. A. 1992. “Modelling the Effects of AIDS on Gonorrhea and S. Horton. Washington, DC: World Bank. Epidemiology.” Mathematical and Computer Modelling 16 Low, N., N. Broutet, Y. Adu-Sarkodie, P. Barton, M. Hossain, and (11): 3–14. S. Hawkes. 2006. “Global Control of Sexually Transmitted Kenyon, C. R., K. Osbak, J. Buyze, and R. M. Chico. 2014. Infections.” The Lancet 368 (9551): 1–16. “The Changing Relationship between Bacterial STIs and Luchters, S., M. F. Chersich, A. Rinyiru, M. S. Barasa, N. King’ola, HIV Prevalence in South Africa: An Ecological Study.” and others. 2008. “Impact of Five Years of Peer-Mediated International Journal of STD and AIDS 26 (8). Interventions on Sexual Behavior and Sexually Transmitted Kenyon, C. R., K. Osbak, and R. M. Chico. 2014. “What Infections among Female Sex Workers in Mombasa, Kenya.” Underpins the Decline in Syphilis in Southern and Eastern BMC Public Health 8: 143. Africa? An Exploratory Ecological Analysis.” International Mabey, D. 2009. “Azithromycin Resistance in Treponema palli- Journal of Infectious Diseases 29: 54–61. dum.” Sexually Transmitted Diseases 36 (12): 777–78. Kerrigan, D., L. Moreno, S. Rosario, B. Gomez, H. Jerez, and Mainkar, M. M., D. B. Pardeshi, J. Dale, S. Deshpande, others. 2006. “Environmental-Structural Interventions to S. Khazi, and others. 2011. “Targeted Interventions of the Reduce HIV/STI Risk among Female Sex Workers in the Avahan Program and Their Association with Intermediate Dominican Republic.” American Journal of Public Health Outcomes among Female Sex Workers in Maharashtra, 96 (1): 120–25. India.” BMC Public Health 11 (Suppl 6): S2. Kippax, S., and N. Stephenson. 2012. “Beyond the Distinction Makasa, M., A. Buve, and I. F. Sandøy. 2012. “Etiologic Pattern between Biomedical and Social Dimensions of HIV of Genital Ulcers in Lusaka, Zambia: Has Chancroid Been Prevention through the Lens of a Social Public Health.” Eliminated?” Sexually Transmitted Diseases 39 (10): 787–91. American Journal of Public Health 102 (5): 789–99. Malagon, T., M. Drolet, M. C. Boily, E. L. Franco, M. Jit, and Korenromp, E. L., V. C. Van, H. Grosskurth, A. Gavyole, others. 2012. “Cross-Protective Efficacy of Two Human C. P. Van der Ploeg, and others. 2000. “Model-Based Papillomavirus Vaccines: A Systematic Review and Meta- Evaluation of Single-Round Mass Treatment of Sexually Analysis.” The Lancet Infectious Diseases 12 (10): 781–89. Transmitted Diseases for HIV Control in a Rural African Manhart, L. E., and K. K. Holmes. 2005. “Randomized Population.” AIDS 14 (5): 573–93. Controlled Trials of Individual-Level, Population-Level, Labbe, A. C., J. Pepin, N. Khonde, A. Dzokoto, H. Meda, and and Multilevel Interventions for Preventing Sexually others. 2012. “Periodical Antibiotic Treatment for the Transmitted Infections: What Has Worked?” Journal of Control of Gonococcal and Chlamydial Infections among Infectious Diseases 191 (Suppl 1): S7–24. Sex Workers in Benin and Ghana: A Cluster-Randomized Markowitz, L. E., E. F. Dunne, M. Saraiya, H. W. Chesson, Placebo-Controlled Trial.” Sexually Transmitted Diseases C. R. Curtis, and others. 2014. “Human Papillomavirus 39 (4): 253–59. Vaccination: Recommendations of the Advisory Committee Lafort, Y., D. Geelhoed, L. Cumba, C. Lazaro, W. Delva, on Immunization Practices (ACIP).” Morbidity and and others. 2010. “Reproductive Health Services for Mortality Weekly Report. Recommendations and Reports 63 Populations at High Risk of HIV: Performance of a (RR-05): 1–30. Night Clinic in Tete Province, Mozambique.” BMC Health Markowitz, L. E., S. Hariri, C. Lin, E. F. Dunne, M. Steinau, Services Research 10: 144. and others. 2013. “Reduction in Human Papillomavirus Lahra, M. M., Y. R. Lo, and D. M. Whiley. 2013. “Gonococcal (HPV) Prevalence among Young Women Following HPV Antimicrobial Resistance in the Western Pacific Region.” Vaccine Introduction in the United States, National Health Sexually Transmitted Infections 89 (Suppl 4): iv19–23. and Nutrition Examination Surveys, 2003–2010.” Journal of Lan, P., H. Phuc, N. Hoa, N. Chuc, and C. Lundborg. 2014. Infectious Diseases 208 (3): 385–93. “Improved Knowledge and Reported Practice Regarding Marrazzo, J. M., G. Ramjee, B. A. Richardson, K. Gomez, Sexually Transmitted Infections among Healthcare Providers N. Mgodi, and others. 2015. “Tenofovir-Based Preexposure in Rural Vietnam: A Cluster-Randomised Controlled Prophylaxis for HIV Infection among African Women.” Educational Intervention.” BMC Infectious Diseases 14: 646. New England Journal of Medicine 372 (6): 509–18. 228 Major Infectious Diseases Marseille, E., J. G. Kahn, K. Billinghurst, and J. Saba. 2001. to the Transmission of Sexually Transmitted Diseases in “Cost-Effectiveness of the Female Condom in Preventing Kenya.” American Journal of Public Health 84 (12): 1947–51. HIV and STDs in Commercial Sex Workers in Rural South Moses, S., E. N. Ngugi, A. Costigan, C. Kariuki, I. Maclean, and Africa.” Social Science and Medicine 52 (1): 135–48. others. 2002. “Response of a Sexually Transmitted Infection Matthews, J. E., R. Stephenson, and P. S. Sullivan. 2012. “Factors Epidemic to a Treatment and Prevention Programme Associated with Self-Reported HBV Vaccination among in Nairobi, Kenya.” Sexually Transmitted Infections 78 HIV-Negative MSM Participating in an Online Sexual (Suppl 1): i114. Health Survey: A Cross-Sectional Study.” PLoS One 7 (2): Moyo, W., Z. M. Chirenje, J. Mandel, S. K. Schwarcz, J. Klausner, e30609. and others. 2002. “Impact of a Single Session of Counseling Mayaud, P., and D. Mabey. 2004. “Approaches to the Control of on Partner Referral for Sexually Transmitted Disease Sexually Transmitted Infections in Developing Countries: Treatment, Harare, Zimbabwe.” AIDS and Behavior 6 (3): Old Problems and Modern Challenges.” Sexually Transmitted 237–43. Infections 80 (3): 174–82. Muñoz, N., R. Manalastas Jr., P. Pitisuttithum, D. Tresukosol, Mayaud, P., F. Mosha, J. Todd, R. Balira, J. Mgara, and others. J. Monsonego, and others. 2009. “Safety, Immunogenicity, 1997. “Improved Treatment Services Significantly Reduce and Efficacy of Quadrivalent Human Papillomavirus the Prevalence of Sexually Transmitted Diseases in Rural (Types 6, 11, 16, 18) Recombinant Vaccine in Women Tanzania: Results of a Randomized Controlled Trial.” AIDS Aged 24–45 Years: A Randomised, Double-Blind Trial.” The 11 (15): 1873–80. Lancet 373 (9679): 1949–57. McClelland, R. S., J. E. Balkus, J. Lee, O. Anzala, J. Kimani, and Nam, S. L., and K. Blanchet. 2014. “We Mustn’t Forget Other others. 2015. “Randomized Trial of Periodic Presumptive Essential Health Services during the Ebola Crisis.” British Treatment with High-Dose Intravaginal Metronidazole and Medical Journal 349: g6837. Miconazole to Prevent Vaginal Infections in HIV-Negative Natunen, K., T. A. Lehtinen, S. Torvinen, and M. Lehtinen. Women.” Journal of Infectious Diseases 211 (12): 1875–82. 2013. “Cost-Effectiveness of HPV-Vaccination in Medium McCormick, D. F., M. Rahman, S. Zadrozny, A. Alam, L. Ashraf, or Low Income Countries with High Cervical Cancer and others. 2013. “Prevention and Control of Sexually Incidence—A Systematic Review.” Journal of Vaccines and Transmissible Infections among Hotel-Based Female Sex Vaccination 4: 172. Workers in Dhaka, Bangladesh.” Sexual Health 10 (6): Ndowa, F. J., J. M. Francis, A. Machiha, H. Faye-Kette, and 478–86. M. C. Fonkoua. 2013. “Gonococcal Antimicrobial Resistance: McCoy, S. I., R. A. Kangwende, and N. S. Padian. 2010. “Behavior Perspectives from the African Region.” Sexually Transmitted Change Interventions to Prevent HIV Infection among Infections 89 (Suppl 4): iv11–15. Women Living in Low and Middle Income Countries: Newman, L., M. Kamb, S. Hawkes, G. Gomez, L. Say, and oth- A Systematic Review.” AIDS and Behavior 14 (3): 469–82. ers. 2013. “Global Estimates of Syphilis in Pregnancy and Mehta, S. D., C. Gaydos, I. Maclean, E. Odoyo-June, S. Moses, Associated Adverse Outcomes: Analysis of Multinational and others. 2012. “The Effect of Medical Male Circumcision Antenatal Surveillance Data.” PLoS Medicine 10: e1001396. on Urogenital Mycoplasma Genitalium among Men in Nuwaha, F., F. Kambugu, P. S. Nsubuga, B. Hojer, and E. Faxelid. Kisumu, Kenya.” Sexually Transmitted Diseases 39 (4): 2001. “Efficacy of Patient-Delivered Partner Medication in 276–80. the Treatment of Sexual Partners in Uganda.” Sexually Mehta, S. D., S. Moses, K. Agot, C. Parker, J. O. Ndinya-Achola, Transmitted Diseases 28 (2): 105–10. and others. 2009. “Adult Male Circumcision Does Not Reduce O’Farrell, N., R. Oula, L. Morison, and C. T. Van. 2006. the Risk of Incident Neisseria gonorrhoeae, Chlamydia tra- “Periodic Presumptive Treatment for Cervical Infections chomatis, or Trichomonas vaginalis Infection: Results from a in Service Women in 3 Border Provinces of Laos.” Sexually Randomized, Controlled Trial in Kenya.” Journal of Infectious Transmitted Diseases 33 (9): 558–64. Diseases 200 (3): 370–78. Over, A. M., and S. O. Aral. 2006. “The Economics of Sexually Mercer, C. H., C. Tanton, P. Prah, B. Erens, P. Sonnenberg, and Transmitted Infections.” Sexually Transmitted Diseases 33 others. 2013. “Changes in Sexual Attitudes and Lifestyles in (10): S79–83. Britain through the Life Course and over Time: Findings Owusu-Edusei, K., Jr., H. W. Chesson, T. L. Gift, G. Tao, from the National Surveys of Sexual Attitudes and Lifestyles R. Mahajan, and others. 2013. “The Estimated Direct (Natsal).” The Lancet 382 (9907): 1781–94. Medical Cost of Selected Sexually Transmitted Infections Mills, A. 2014. “Health Care Systems in Low- and Middle- in the United States, 2008.” Sexually Transmitted Diseases Income Countries.” New England Journal of Medicine 370 40 (3): 197–201. (6): 552–57. Owusu-Edusei, K., Jr., H. W. Chesson, J. S. Leichliter, C. K. Kent, Moreno, R., H. Y. Nababan, E. Ota, W. M. Wariki, S. Ezoe, and and S. O. Aral. 2013. “The Association between Racial others. 2014.“Structural and Community-Level Interventions Disparity in Income and Reported Sexually Transmitted for Increasing Condom Use to Prevent the Transmission of Infections.” American Journal of Public Health 103 (5): HIV and Other Sexually Transmitted Infections.” Cochrane 910–16. Database of Systematic Reviews 7: CD003363. Owusu-Edusei, K., Jr., T. L. Gift, and R. C. Ballard. 2011. “Cost- Moses, S., E. N. Ngugi, J. E. Bradley, E. K. Njeru, G. Eldridge, Effectiveness of a Dual Non-Treponemal/Treponemal and others. 1994. “Health Care–Seeking Behavior Related Syphilis Point-of-Care Test to Prevent Adverse Pregnancy Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 229 Outcomes in Sub-Saharan Africa.” Sexually Transmitted Syphilis with Oral Azithromycin.” The Lancet 361 (9354): Diseases 38 (11): 997–1003. 313–14. Paavonen, J., P. Naud, J. Salmeron, C. M. Wheeler, S. N. Chow, Richardson, B. A., L. Lavreys, H. L. Martin Jr., C. E. Stevens, and others. 2009. “Efficacy of Human Papillomavirus E. Ngugi, and others. 2001. “Evaluation of a Low-Dose (HPV)-16/18 AS04-Adjuvanted Vaccine against Cervical Nonoxynol-9 Gel for the Prevention of Sexually Transmitted Infection and Precancer Caused by Oncogenic HPV Types Diseases: A Randomized Clinical Trial.” Sexually Transmitted (PATRICIA): Final Analysis of a Double-Blind, Randomised Diseases 28 (7): 394–400. Study in Young Women.” The Lancet 374 (9686): 301–14. Roddy, R. E., L. Zekeng, K. A. Ryan, U. Tamoufe, and K. G. Tweedy. Palefsky, J. M., A. R. Giuliano, S. Goldstone, E. D. Moreira Jr., 2002. “Effect of Nonoxynol-9 Gel on Urogenital Gonorrhea C. Aranda, and others. 2011. “HPV Vaccine against Anal and Chlamydial Infection: A Randomized Controlled HPV Infection and Anal Intraepithelial Neoplasia.” New Trial.” Journal of the American Medical Association 287 (9): England Journal of Medicine 365 (17): 1576–85. 1117–22. Parkhurst, J. O. 2014. “Structural Approaches for Prevention Rutstein, S. E., L. B. Brown, A. K. Biddle, S. B. Wheeler, of Sexually Transmitted HIV in General Populations: G. Kamanga, and others. 2014. “Cost-Effectiveness of Definitions and an Operational Approach.” Journal of the Provider-Based HIV Partner Notification in Urban Malawi.” International AIDS Society 17: 19052. Health Policy and Planning 29 (1): 115–26. Patterson, T. L., B. Mausbach, R. Lozada, H. Staines-Orozco, Ryan, C. A., M. Kamb, and K. K. Holmes. 2008. “STI Care S. J. Semple, and others. 2008. “Efficacy of a Brief Behavioral Management.” In Sexually Transmitted Diseases, edited Intervention to Promote Condom Use among Female Sex by K. K. Holmes, P. F. Sparling, W. E. Stamm, P. Piot, Workers in Tijuana and Ciudad Juarez, Mexico.” American J. N. Wasserheit, and others, 855–75. New York: McGraw-Hill. Journal of Public Health 98 (11): 2051–57. Rydzak, C. E., and S. J. Goldie. 2008. “Cost-Effectiveness Peterman, T. A., and B. W. Furness. 2015. “Public Health of Rapid Point-of-Care Prenatal Syphilis Screening in Interventions to Control Syphilis.” Sexual Health 12 (2): Sub-Saharan Africa.” Sexually Transmitted Diseases 35 (9): 126–34. 775–84. Pettifor, A., J. Walsh, V. Wilkins, and P. Raghunathan. 2000. Sahin-Hodoglugil, N. N., R. Woods, A. Pettifor, and J. Walsh. “How Effective Is Syndromic Management of STDs? A 2003. “A Comparison of Cost-Effectiveness of Three Review of Current Studies.” Sexually Transmitted Diseases Protocols for Diagnosis and Treatment of Gonococcal 27 (7): 371–85. and Chlamydial Infections in Women in Africa.” Sexually Pourbohloul, B., M. L. Rekart, and R. C. Brunham. 2003. Transmitted Diseases 30 (5): 455–69. “Impact of Mass Treatment on Syphilis Transmission: A Schmid, G. 2004. “Economic and Programmatic Aspects of Mathematical Modeling Approach.” Sexually Transmitted Congenital Syphilis Prevention.” Bulletin of the World Diseases 30 (4): 297–305. Health Organization 82 (6): 402–9. Pretorius, C., J. Stover, L. Bollinger, N. Bacaer, and B. Williams. Sen, A., and M. Luong. 2008. “Estimating the Impact of Beer 2010. “Evaluating the Cost-Effectiveness of Pre-Exposure Prices on the Incidence of Sexually Transmitted Diseases: Prophylaxis (PrEP) and Its Impact on HIV-1 Transmission Cross-Province and Time Series Evidence from Canada.” in South Africa.” PLoS One 5: e13646. Contemporary Economic Policy 26 (4): 505–17. Price, M. A., D. Zimba, I. F. Hoffman, S. C. Kaydos-Daniels, Shah, S. A., S. Kristensen, M. A. Memon, H. L. White, and W. C. Miller, and others. 2003. “Addition of Treatment for S. H. Vermund. 2007. “Syndromic Management Training for Trichomoniasis to Syndromic Management of Urethritis in Non-Formal Care Providers in Pakistan Improves Quality Malawi: A Randomized Clinical Trial.” Sexually Transmitted of Care for Sexually Transmitted Diseases (STD) Care: A Diseases 30 (6): 516–22. Randomized Clinical Trial.” South-East Asian Journal of Punpanich, W., K. Ungchusak, and R. Detels. 2004. “Thailand’s Tropical Medicine and Public Health 38 (4): 737–48. Response to the HIV Epidemic: Yesterday, Today, and Sobngwi-Tambekou, J., D. Taljaard, M. Nieuwoudt, P. Lissouba, Tomorrow.” AIDS Education and Prevention 16 (Suppl 3A): A. Puren, and others. 2009. “Male Circumcision and Neisseria 119–36. gonorrhoeae, Chlamydia trachomatis, and Trichomonas vagi- Ramjee, G., A. van der Straten, T. Chipato, G. de Bruyn, nalis: Observations after a Randomised Controlled Trial K. Blanchard, and others. 2008. “The Diaphragm and for HIV Prevention.” Sexually Transmitted Infections 85 (2): Lubricant Gel for Prevention of Cervical Sexually 116–20. Transmitted Infections: Results of a Randomized Controlled Spinola, S. M. 2008. “Chancroid and Haemophilus ducreyi.” Trial.” PLoS One 3 (10): e3488. In Sexually Transmitted Diseases, edited by K. K. Holmes, Rees, H., S. Delany-Moretlwe, D. Baron, C. Lombard, G. Gray, P. F. Sparling, W. E. Stamm, P. Piot, J. N. Wasserheit, and and others. 2015. “FACTS 001 Phase III Trial of Pericoital others, 689–99. New York: McGraw-Hill. Tenofovir 1% Gel for HIV Prevention in Women.” Stanberry, L. R., S. L. Spruance, A. L. Cunningham, D. I. Bernstein, Paper presented at the Conference on Retroviruses and A. Mindel, and others. 2002. “Glycoprotein-D–Adjuvant Opportunistic Infections (Abstract 26LB), Seattle, WA, Vaccine to Prevent Genital Herpes.” New England Journal of February 23–26. Medicine 347 (21): 1652–61. Rekart, M. L., D. M. Patrick, B. Chakraborty, J. J. Maginley, Staras, S. A., M. D. Livingston, A. M. Christou, D. H. Jernigan, and H. D. Jones, and others. 2003. “Targeted Mass Treatment for A. C. Wagenaar. 2014. “Heterogeneous Population Effects of 230 Major Infectious Diseases an Alcohol Excise Tax Increase on Sexually Transmitted Thomas, J. C., and M. E. Gaffield. 2003. “Social Structure, Race, Infections Morbidity.” Addiction 109 (6): 904–12. and Gonorrhea Rates in the Southeastern United States.” Steen, R. 2001. “Eradicating Chancroid.” Bulletin of the World Ethnicity and Disease 13 (3): 362–68. Health Organization 79 (9): 818–26. Tobian, A. A., D. Serwadda, T. C. Quinn, G. Kigozi, P. E. Gravitt, Steen, R., M. Chersich, and S. J. de Vlas. 2012. “Periodic and others. 2009. “Male Circumcision for the Prevention Presumptive Treatment of Curable Sexually Transmitted of HSV-2 and HPV Infections and Syphilis.” New England Infections among Sex Workers: Recent Experience with Journal of Medicine 360 (13): 1298–309. Implementation.” Current Opinion in Infectious Diseases Townsend, L., C. Mathews, and Y. Zembe. 2013. “A Systematic 25 (1): 100–6. Review of Behavioral Interventions to Prevent HIV Steen, R., B. Vuylsteke, T. DeCoito, S. Ralepeli, G. Fehler, and Infection and Transmission among Heterosexual, Adult others. 2000. “Evidence of Declining STD Prevalence in a Men in Low- and Middle-Income Countries.” Prevention South African Mining Community Following a Core-Group Science 14 (1): 88–105. Intervention.” Sexually Transmitted Diseases 27 (1): 1–8. Uthman, O. A., T. A. Popoola, M. M. B. Uthman, and O. Aremu. Stevenson, B., and J. Wolfers. 2007. “Marriage and Divorce: 2010. “Economic Evaluations of Adult Male Circumcision Changes and Their Driving Forces.” Working Paper 1294, for Prevention of Heterosexual Acquisition of HIV in Men National Bureau of Economic Research, Cambridge, MA. in Sub-Saharan Africa: A Systematic Review.” PLoS One http://www.nber.org/papers/w12944. 5 (3): e9628. Stolte, I. G., N. H. Dukers, R. B. Geskus, R. A. Coutinho, van Dam, C. J. 1995. “HIV, STD, and Their Current Impact and J. B. de Wit. 2004. “Homosexual Men Change to on Reproductive Health: The Need for Control of Sexually Risky Sex When Perceiving Less Threat of HIV/AIDS Transmitted Diseases.” International Journal of Gynecology since Availability of Highly Active Antiretroviral Therapy: and Obstetrics 50 (Suppl 2): S121–29. A Longitudinal Study.” AIDS 18 (2): 303–9. van Dam, C. J., G. Dallabetta, and P. Piot. 1999. “Prevention and Subramanian, T., L. Ramakrishnan, S. Aridoss, P. Goswami, Control of Sexually Transmitted Diseases in Developing B. Kanguswami, and others. 2013. “Increasing Condom Use Countries.” In Sexually Transmitted Diseases, edited by K. K. and Declining STI Prevalence in High-Risk MSM and TGs: Holmes, P. F. Sparling, P. Mardh, S. M. Lemon, W. E. Stamm, Evaluation of a Large-Scale Prevention Program in Tamil and others, 1381–90. New York: McGraw-Hill. Nadu, India.” BMC Public Health 13: 857. Verguet, S., and J. A. Walsh. 2010. “Vaginal Microbicides Save Sweat, M., D. Kerrigan, L. Moreno, S. Rosario, B. Gomez, and Money: A Model of Cost-Effectiveness in South Africa and others. 2006. “Cost-Effectiveness of Environmental-Structural the USA.” Sexually Transmitted Infections 86 (3): 212–16. Communication Interventions for HIV Prevention in the Vickerman, P., A. Devine, A. M. Foss, S. Delany-Moretlwe, Female Sex Industry in the Dominican Republic.” Journal of P. Mayaud, and others. 2011. “The Cost-Effectiveness Health Communication 11 (Suppl 2): 123–42. of Herpes Simplex Virus-2 Suppressive Therapy with Tabrizi, S. N., J. M. Brotherton, J. M. Kaldor, S. R. Skinner, Daily Aciclovir for Delaying HIV Disease Progression E. Cummins, and others. 2012. “Fall in Human Papillomavirus among HIV-1-Infected Women in South Africa.” Sexually Prevalence Following a National Vaccination Program.” Transmitted Diseases 38 (5): 401–09. Journal of Infectious Diseases 206 (11): 1645–51. Vickerman, P., F. Ndowa, and P. Mayaud. 2008. “Modelling the Tanser, F., T. Bärnighausen, G. S. Cooke, and M. L. Newell. Cost per Ulcer Treated of Incorporating Episodic Treatment 2009. “Localized Spatial Clustering of HIV Infections in for HSV-2 into the Syndromic Algorithm for Genital Ulcer a Widely Disseminated Rural South African Epidemic.” Disease.” Sexually Transmitted Infections 84 (3): 243–48. International Journal of Epidemiology 38 (4): 1008–16. Villa, L. L., K. A. Ault, A. R. Giuliano, R. L. Costa, C. A. Terris-Prestholt, F., A. M. Foss, A. P. Cox, L. Heise, G. Meyer- Petta, and others. 2006. “Immunologic Responses Rath, and others. 2014. “Cost-Effectiveness of Tenofovir Gel Following Administration of a Vaccine Targeting Human in Urban South Africa: Model Projections of HIV Impact Papillomavirus Types 6, 11, 16, and 18.” Vaccine 24 (27–28): and Threshold Product Prices.” BMC Infectious Diseases 5571–83. 14: 14. Wang, C., Y. Wang, X. Huang, X. Li, T. Zhang, and others. Terris-Prestholt, F., S. Vyas, L. Kumaranayake, P. Mayaud, and 2012. “Prevalence and Factors Associated with Hepatitis B C. Watts. 2006. “The Costs of Treating Curable Sexually Immunization and Infection among Men Who Have Sex Transmitted Infections in Low- and Middle-Income with Men in Beijing, China.” PLoS One 7 (10): e48219. Countries: A Systematic Review.” Sexually Transmitted Ward, H., and S. O. Aral. 2006. “Globalisation, the Sex Industry, Diseases 33 (10): S153–66. and Health.” Sexually Transmitted Infections 82 (5): 345–47. Terris-Prestholt, F., D. Watson-Jones, K. Mugeye, Watson-Jones, D., J. Changalucha, B. Gumodoka, H. Weiss, L. Kumaranayake, L. Ndeki, and others. 2003. “Is Antenatal M. Rusizoka, and others. 2002. “Syphilis in Pregnancy in Syphilis Screening Still Cost-Effective in Sub-Saharan Tanzania. I. Impact of Maternal Syphilis on Outcome of Africa.” Sexually Transmitted Infections 79 (5): 375–81. Pregnancy.” Journal of Infectious Diseases 186 (7): 940–47. Thigpen, M. C., P. M. Kebaabetswe, L. A. Paxton, D. K. Smith, Weir, S. S., C. Pailman, X. Mahlalela, N. Coetzee, F. Meidany, C. E. Rose, and others. 2012. “Antiretroviral Preexposure and others. 2003. “From People to Places: Focusing AIDS Prophylaxis for Heterosexual HIV Transmission in Prevention Efforts Where It Matters Most.” AIDS 17 (6): Botswana.” New England Journal of Medicine 367 (5): 423–34. 895–903. Sexually Transmitted Infections: Impact and Cost-Effectiveness of Prevention 231 Wetmore, C. M., L. E. Manhart, and J. N. Wasserheit. 2010. ———. 2010. More Developing Countries Show Universal “Randomized Controlled Trials of Interventions to Prevent Access to HIV/AIDS Services Is Possible: Sustained Sexually Transmitted Infections: Learning from the Past to Commitments Necessary to Secure Future Progress. Geneva: Plan for the Future.” Epidemiologic Reviews 32 (1): 121–36. WHO. White, R. G., K. K. Orroth, J. R. Glynn, E. E. Freeman, ———. 2012a. Global Incidence and Prevalence of Selected R. Bakker, and others. 2008. “Treating Curable Sexually Curable Sexually Transmitted Infections: 2008. Geneva: Transmitted Infections to Prevent HIV in Africa: Still an WHO. Effective Control Strategy?” Journal of Acquired Immune ———. 2012b. Investment Case for Eliminating Mother-to- Deficiency Syndromes 47 (3): 346–53. Child Transmission of Syphilis: Promoting Better Maternal WHO (World Health Organization). 2003. Guidelines for the and Child Health and Stronger Health Systems. Geneva: Management of Sexually Transmitted Infections. Geneva: WHO. WHO. World Bank. 2007. “Sexually Transmitted Infections in ———. 2008. Periodic Presumptive Treatment for Sexually Developing Countries.” Public Health at a Glance. Washington, Transmitted Infections: Experience from the Field and DC: World Bank. http://web.worldbank.org/archive Recommendations for Research. Geneva: WHO. /website01213/WEB/0__CO-54.HTM. 232 Major Infectious Diseases Chapter 11 Tuberculosis Barry R. Bloom, Rifat Atun, Ted Cohen, Christopher Dye, Hamish Fraser, Gabriela B. Gomez, Gwen Knight, Megan Murray, Edward Nardell, Eric Rubin, Joshua Salomon, Anna Vassall, Grigory Volchenkov, Richard White, Douglas Wilson, and Prashant Yadav OVERVIEW Treatment of TB disease requires multiple drugs for Despite 90 years of vaccination and 60 years of chemother- many months. These long drug regimens are challenging apy, tuberculosis (TB) remains the world’s leading cause of for both patients and health care systems, especially in death from an infectious agent, exceeding human immu- low- and middle-income countries (LMICs), where the nodeficiency virus/acquired immune deficiency syndrome disease burden often far outstrips local resources. In (HIV/AIDS) for the first time (WHO 2015b, 2016a). some areas, the incidence of drug-resistant TB, requiring The World Health Organization (WHO) estimates that even longer treatment regimens with drugs that are there are about 10.4 million new cases and 1.8 million more expensive and difficult to tolerate, is increasing. deaths from TB each year. One-third of these new cases Diagnosis in LMICs is made primarily by microscopic (about 3 million) remain unknown to the health system, examination of stained smears of sputum of suspected and many are not receiving proper treatment. patients; however, smear microscopy is capable of detect- Tuberculosis is an infectious bacterial disease caused ing only 50–60 percent of all cases (smear-positive). by Mycobacterium tuberculosis (Mtb), which is transmit- More sensitive methods of diagnosing TB and detecting ted between humans through the respiratory route and resistance to drugs have recently become available, most commonly affects the lungs, but can damage any although they are more expensive. The time between the tissue. Only about 10 percent of individuals infected onset of disease and when diagnosis is made and treat- with Mtb progress to active TB disease within their life- ment is initiated is often protracted, and such delays time; the remainder of persons infected successfully allow the transmission of disease. Although bacille contain their infection. One of the challenges of TB is Calmette–Guérin (BCG) remains the world’s most widely that the pathogen persists in many infected individuals used vaccine, its effectiveness is geographically highly in a latent state for many years and can be reactivated to variable and incomplete. Modeling suggests that more cause disease. The risk of progression to TB disease after effective vaccines will likely be needed to drive tuberculo- infection is highest soon after the initial infection and sis toward elimination in high-incidence settings. increases dramatically for persons co-infected with HIV/ The basic strategy to combat TB has been, for 40 years, AIDS or other immune-compromising conditions. to provide diagnosis and treatment to individuals Corresponding author: Barry R. Bloom, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States; bbloom@hsph.harvard.edu. 233 who are ill and who seek care at a health facility. The Moreover, the TB problem has become more pressing premise is that, if patients with active disease are cured, because of co-infection with HIV/AIDS. While globally mortality will disappear, prevalence of disease will HIV/AIDS and TB co-infection represents only decline, transmission will decline, and therefore inci- 11 percent of the total TB burden, in some areas of Sub- dence should decline. The reality in many countries is Saharan Africa with a high burden of TB, as many as more complex, and overall the decline in incidence (only three-quarters of TB patients are co-infected with HIV/ about 1.5 percent per year) has been unacceptably slow. AIDS. In those countries, efforts to control TB are over- Chemotherapy for TB is one of the most cost- whelmed by the rising number of TB cases occurring effective of all health interventions (McKee and Atun in parallel with the HIV/AIDS epidemic. And after 2006). This evidence has been central to the global pro- decades of steady decline, the incidence of TB is also motion of the WHO and Stop TB Partnership policy of increasing in some high-income countries (HICs), directly observed therapy, short course (DOTS) strategy, mainly as the result of outbreaks in vulnerable groups the package of measures combining best practices in the (WHO 2015b). diagnosis and care of patients with TB (UN General If the ultimate goal of controlling an infectious dis- Assembly 2000). The DOTS strategy to control tubercu- ease is to interrupt transmission, turning the tide on TB losis promotes standardized treatment, with supervision will require early and accurate case detection, rapid and patient support that may include, but is far broader commencement of and adherence to effective treatment than, direct observation of therapy (DOT), where a that prevents transmission, and, where possible, preven- health care worker personally observes the patient taking tive treatment of latent TB. It is universally understood the medication (WHO 2013a). that new strategies and more effective tools and inter- Thanks in part to these efforts and national and ventions will be required to reach post-2015 targets international investments, much progress has been (Bloom and Atun 2016; WHO 2015a). These interven- made in TB control over the past several decades. tions must be not only cost-effective, but also affordable Between 1990 and 2010, absolute global mortality and capable of having an impact on a very large scale. from TB declined 18.7 percent, from 1.47 million to TB control will need three new advances— 1.20 million (Lozano and others 2012) and by development of new point-of-care diagnostics, more 22 percent between 2000 and 2015 (WHO 2016a). By effective drug regimens to combat drug-susceptible 2015, an estimated 49 million lives had been saved and drug-resistant TB, and more effective vaccines. As (WHO 2016a). The internationally agreed targets for argued in this chapter, these require new strategies and TB, embraced in the United Nations (UN) Millennium tools that include moving away from the traditional Development Goals (MDGs), sought “to halt and DOTS passive case finding and toward more active case reverse the expanding incidence of tuberculosis by finding in high-burden regions; service delivery that is 2015,” and this target has been met to some extent in targeted to the most vulnerable populations and inte- all six WHO regions and in most, but not all, of the grated with other services, especially HIV/AIDS ser- world’s 22 high-burden countries (WHO 2014c). vices; and care that is based at the primary health care Despite progress, major gaps persist. Although the and community levels. Specifically, in high-burden Sustainable Development Goals (SDGs) seek to end the countries, many individuals with TB are asymptom- tuberculosis epidemic altogether (WHO 2015a, 2015c), atic, such that waiting for patients to become sick the decline in incidence has been disappointing. One of enough to seek care has not been sufficient to reduce every three TB patients remains “unknown to the health transmission and incidence markedly (Bates and others system,” many are undiagnosed and untreated, and case 2012; Mao and others 2014; Willingham and others detection and treatment success rates remain too low 2001; Wood and others 2007). A more active and in the high-burden countries. Ominously, rates of aggressive approach is needed that tackles health multidrug-resistant (MDR) TB—defined as resistance to system barriers to effective TB control. the two major TB drugs, isoniazid and rifampicin—are The strategies for controlling TB recommended by the rising globally (WHO 2011a) with the emergence of WHO have evolved significantly over time. In the early extensively drug-resistant (XDR) TB, resistant to many formulations, the central tenets of the global TB control second-line drugs, as well as strains resistant to all cur- strategy were clinical and programmatic in nature, focus- rent drugs (Dheda and others 2014; Udwadia and others ing principally on the delivery of standardized drug regi- 2012; Uplekar and others 2015). These are now primarily mens; the underlying assumption was that the problem the result of transmission rather than inadequate treat- could be solved largely by existing biomedical tools (Atun, ment (Shah and others 2017). McKee, and others 2005; Schouten and others 2011). 234 Major Infectious Diseases Yet, in many LMICs, health system weaknesses in gover- whose primary purpose is to promote, restore, or main- nance, financing, health workforce, procurement and tain health” (WHO 2000, 5). supply chain management, and information systems have In a sense, the major risk factor for acquiring TB is impeded TB control (Elzinga, Raviglione, and Maher breathing. Thus, people of all social and economic statuses 2004; Marais and others 2010; Travis and others 2004) are at risk. While TB disproportionately affects the poor, and not been adequately addressed by TB control efforts. the narrative that TB is a disease only of the poor is mis- The current global TB strategy, formulated as the End leading and counterproductive, if it leads either to further TB Strategy, is the most comprehensive ever, with three stigmatization of the disease or to the view that middle- major pillars: and high-income countries need not worry about the dis- ease. In the case of co-infection with HIV/AIDS, evidence • Integrated, patient-centered care and prevention suggests that HIV/AIDS is often more prevalent in • Social and political action to address determinants better-off populations in Africa that suffer high rates of TB. of disease The analytical framework underlying this chapter defines • Recognition of the urgent need for research to pro- key functions of the health system, ultimate goals, and con- vide new tools (WHO 2015a). textual factors that affect the health system (figure 11.1). It builds on the WHO framework (WHO 2000) as well Health systems are important and need to be strength- as health system frameworks developed by Frenk (1994), ened. As with other health interventions, the success of Hsiao and Heller (2007), and Roberts and others (2004), tuberculosis treatment and control in a country is often and national accounts (OECD, Eurostat, and WHO 2011). determined by the strength of its health system (McKee It also draws on earlier studies by Atun (2012); Atun and and Atun 2006; WHO 2003). A health system can be Coker (2008); Atun, Samyshkin, and others (2006); Samb defined in many ways, perhaps best as “all the activities and others (2009); and Swanson and others (2012). Figure 11.1 Schematic Health Systems Framework Demographic Epidemiological Political Governance and organization Equity Efficiency Health Public health Financial Legal / Ecological Financing protection regulatory Health care services User Effectiveness Responsiveness satisfaction Resource management Technological Economic Sociocultural Tuberculosis 235 The four key health system functions represented in in health systems; community-based rather than hospital- the framework are as follows: based service delivery; and greater support for research on new tools—that is, developing better diagnostics, treat- • Governance and organization. The policy and reg- ment regimens, and vaccines. Most of these approaches ulatory environment; stewardship and regulatory were included in earlier WHO policies, but were not functions of the ministry of health and its relation emphasized. They are now part of the WHO’s End TB to other levels of the health system; and structural Strategy, with which this report is fully consistent (WHO arrangements for insurers and purchasers, health care 2015a, 2015c). providers, and market regulators • Financing. The way funds are collected, funds and risks are pooled, finances are allocated, and health HISTORICAL TRENDS, CURRENT BURDEN, care providers are remunerated AND GLOBAL RESPONSE • Resource management. The way resources—physical, human, and intellectual—are generated and allo- TB has been a major killer worldwide for centuries and cated, including their geographic and needs-based has now exceeded HIV/AIDS and malaria as the world’s allocation largest cause of death from an infectious disease (Dye • Service delivery. Both population- and individual- 2015; WHO 2016a). level public health interventions and health care services provided in community, primary health care, hospitals, and other health institutions. Historical Trends TB rates have been declining in North America Each of these functions is influenced by the eco- and Western Europe since the early nineteenth century, nomic, demographic, legal, cultural, and political prior to the introduction of chemotherapy in the 1950s. context. The decline may be partly due to the natural waning of As the framework suggests, health system goals the epidemic (Blower and others 1995), but the trend include better health, financial protection, and user satis- has been too prolonged for this to provide the whole faction. Personal health services and public health inter- explanation. Researchers have recently suggested that ventions should be organized to achieve an appropriate dramatic differences between cities experiencing balance of equity (including reducing out-of-pocket marked declines prior to chemotherapy (for example, [OOP] expenditures and impoverishment of individuals New York City in the United States; London in the and families), efficiency, effectiveness (that is, the extent United Kingdom) and cities where TB remained high to which interventions are evidence based and safe), (Cape Town in South Africa) may be explained by the responsiveness, equity, and client satisfaction (as per- quality and organization of the local health system ceived by the users of services). (Hermans, Horsburgh, and Wood 2015). Other potential This chapter is organized as follows. First, we pro- explanations for the 150-year decline have been the sub- vide a detailed discussion of the global burden of ject of debate and include the following: disease and clinical context, followed by a review of approaches to diagnosis, treatment, and prevention. 1. Reduced opportunities for transmission per case, The aim throughout is to approach TB through a which may have occurred due to lower living density, health system lens and, in the latter part of the chapter, better ventilation within homes, patient isolation to provide recommendations for improving delivery within sanatoriums, and declining contacts among strategies and strengthening health systems, including elderly cases (McFarlane 1989) care, supply chain, and information systems. Because 2. Reduced susceptibility of contacts, which may have the current tools for combating TB are seriously been the result of improved nutrition and genetic inadequate, we conclude with sections on critical pressures (Lipsitch and Sousa 2002; McKeown and research and development and economic analyses of Record 1962; Shetty and others 2006) new interventions for diagnosis, treatment, and vac- 3. Reduced virulence of the pathogen, although there is cines. Throughout, emphasis is placed on data or mod- little evidence to that effect. eling of the economic costs and benefits, where available, of current or possible future interventions to While untreated TB has traditionally had a case combat this disease. fatality rate of 50 percent, there are differing opinions The chapter recommends moving toward active on the role of natural selection in resistance to pulmo- case finding in high-burden countries; greater investments nary TB in humans prior to the availability of TB drugs 236 Major Infectious Diseases (Lipsitch and Sousa 2002). It is not possible to disen- Current Burden tangle all of the factors that contributed to the decline Global Progress of TB before the widespread introduction of chemo- Although the overall burden of disease remains large, therapy or the reasons why progress has since stalled. substantial progress has been made in TB control world- What is clear is that the TB death rate in Western wide. Between 1995 and 2013, the TB case detection Europe fell 5 percent a year in the era before chemo- rate increased from 46 percent to 64 percent. In the same therapy, with declines in the United States and Western period, between 41 million and 56 million people were Europe associated with active case finding, for example, successfully treated, and by 2015 as many as 49 million X-ray screening. lives were saved (Glaziou and others 2011; WHO 2015b, By the 1990s, however, TB had emerged as a major 2016a). TB prevalence worldwide fell 47 percent by 2015, global health issue, driven largely by an increase in the and the TB-attributable mortality rate had declined number of cases in the former Soviet Union and Sub- 45 percent compared with a 1990 baseline. Saharan Africa. As the number of cases fell in other parts Since the mid-2000s, the global incidence of TB has of the world, TB incidence per capita rose in these two been declining, albeit slowly, along with the absolute regions. number of new TB cases reported each year. However, In the Russian Federation and other former Soviet incidence rates remain discouragingly high in high- countries, TB incidence and deaths rose sharply between burden countries in South-East Asia and Sub-Saharan 1990 and 2000. Understanding precisely why is nearly Africa (figure 11.2). as difficult as understanding the decline in Europe and Despite global progress, uncertainties remain in the North America. It is clear that there was a marked deteri- burden and trends in TB. For example, while estimates oration in case finding and cure rates in Russia, but this of the WHO and of the Institute for Health Metrics and likely does not explain all of the increase (Shilova and Dye Evaluation (IHME) were similar on prevalence of TB in 2001). Other factors include enhanced transmission due 2012, the trends differed: the WHO estimated a decline to the mixing of prison and civilian populations; an in cases, while the IHME estimated an increase in cases increase in susceptibility to disease following infection, over the same interval. likely linked to alcoholism and stress; poor nutrition; The progress of individual countries, organized by emphasis on hospital-based treatment and extended major international TB targets and goals in 22 high- hospitalization; poor service delivery; the spread of drug burden countries—that is, those defined by the WHO as resistance; and, more recently, the rise of HIV/AIDS infec- the 22 countries accounting for approximately 80 percent tion (Atun, Samyshkin, and others 2005; Toungoussova, of the world’s TB cases in 2015—is shown in table 11.1 Bjune, and Caugant 2006). (WHO 2015b). Figure 11.2 Estimated TB Incidence: Top-10 Countries, 2014 a. Absolute number of cases b. Annual rate per 100,000 population India Lesotho Indonesia South Africa China Swaziland Nigeria Djibouti Pakistan Namibia South Africa Mozambique Bangladesh Timor-Leste Philippines Gabon Congo, Dem. Rep. Korea, Dem. People’s Rep. Ethiopia Papua New Guinea 0 0.5 1.0 1.5 2.0 0 300 600 900 Millions Rate per 100,000 population per year Source: WHO 2015b. Note: TB = tuberculosis. The range shows the lower and upper bounds of the 95% uncertainty interval. The bullet indicates the best estimate. Tuberculosis 237 Table 11.1 2015 Target Assessment: Global, WHO Regions, and 22 High-Burden Countries TB/HIV: 2015 Global Plan Targets MDR-TB: 2015 Global Plan Targets Notified People living Treatment HIV-positive with HIV newly success rate: TB patients TB patients enrolled in HIV Estimated MDR-TB cases confirmed with known started on care who were that were detected and MDR-TB cases, Indicator HIV status (%) ART (%) started on IPT notified (%) 2011 cohort (%) Target 100% 100% 50% 100% ê75% Global Global 48 70 45 48 WHO region African (AFR) 76 69 74 47 Americas (AMR) 69 65 47 56 Eastern Mediterranean (EMR) 11 38 22 64 European (EUR) 59 54 61 46 South-East Asia (SEAR) 43 81 45 54 Western Pacific (WPR) 35 60 16 52 22 High-burden countries AFR Congo, Dem. Rep 44 48 9 59 Ethiopia 71 68 40 72 Kenya 94 84 6 70 Mozambique 91 72 17 19 31 Nigeria 88 67 3.5 18 63 South Africa 90 66 >100 >100 45 Uganda 91 65 12 77 Tanzania 83 73 0.4 10 75 Zimbabwe 92 77 3.5 53 81 AMR Brazil 65 39 56 EMR Afghanistan 26 23 6 29 Pakistan 2.8 20 70 EUR Russian Federation 33 37 SEAR Bangladesh 1.1 100 0 22 68 India 63 88 57 50 Indonesia 2.3 21 13 60 Myanmar 12 74 19 22 71 Thailand 83 59 12 WPR Cambodia 82 89 55 24 86 China 39 67 8 50 Philippines 2.1 47 41 Vietnam 70 61 24 72 table continues next page 238 Major Infectious Diseases Table 11.1 2015 Target Assessment: Global, WHO Regions, and 22 High-Burden Countries (continued) TB/HIV: 2015 Global Plan Targets MDR-TB: 2015 Global Plan Targets Notified People living Treatment HIV-positive with HIV newly success rate: TB patients TB patients enrolled in HIV Estimated MDR-TB cases confirmed with known started on care who were that were detected and MDR-TB cases, Indicator HIV status (%) ART (%) started on IPT notified (%) 2011 cohort (%) Target 100% 100% 50% 100% ê75% Classification ≥80% tested ≥80% ≥50% ≥80% detected and notified ≥75% 50–79% tested 50–79% 25–49% 50–79% detected and 50–74% notified <50% tested <50% <25% <50% detected and notified <50% Source: WHO 2015b. Note: A blank cell indicates that no data are available. TB = tuberculosis; MDR TB = multidrug-resistant TB; HIV = human immunodeficiency virus; ART = antiretroviral treatment; IPT = isoniazid preventive therapy; WHO = World Health Organization. Remaining Challenges The emergence of highly drug-resistant tuberculosis, Despite significant progress, TB remains a formidable including MDR TB (resistance to at least isoniazid and global health threat. The overall rate of decline in inci- rifampicin) and XDR TB (MDR plus resistance to at dence, by any calculation, has been frustratingly slow least one fluoroquinolone and one injectable antituber- (1.5 percent a year), and some countries and regions are cular antibiotic), has proved a serious hurdle for effective still reporting rises in TB incidence, particularly in drug- control of tuberculosis in many settings. The most recent resistant TB. Based on notification reports and surveys, estimates of the burden of MDR TB suggest that approx- there were an estimated 10.4 million new TB cases in 2015 imately 480,000 new cases of MDR TB occur each year, (WHO 2016a). Assuming lifelong latent infection, about of which only 20 percent are enrolled in treatment one-third of humanity could still be infected with Mtb (WHO 2016a). Sufficiently strong surveillance and drug (Dye and others 1999; Sudre, ten Dam, and Kochi 1992). resistance laboratory testing are not yet adequate to The estimation of TB incidence and prevalence establish whether MDR TB is rising or falling in most remains imprecise, especially in high-burden countries countries where MDR is of concern. where precision is most needed. Over the past decade, HIV/AIDS is another factor that challenges effective national surveys of the prevalence of tuberculosis dis- control of TB, especially in Southern African countries. ease have been undertaken in more than 20 countries, Of the 10.4 million new cases of TB in 2014, 1.2 million including 15 of the 22 highest-burden countries (WHO occurred in HIV-positive individuals. Among the 2015b). These prevalence surveys provide vital data in approximately 1.5 million deaths from TB in 2014, high-burden settings. Great investment is needed in 400,000 were among individuals co-infected with high-quality routine surveillance that builds on existing HIV/AIDS (WHO 2015b). Globally, around half of systems and produces robust data for assessment and patients diagnosed with TB were tested for HIV/AIDS, future planning (Dye and others 2008; WHO 2012b). although that number has increased to 79 percent in Since a quarter of TB patients are in India, the National the African region. Treating TB patients co-infected Survey planned for 2017–18 will be very important. with HIV/AIDS with antiretroviral therapy (ART) rose Many factors drive the persistence and fatality of the to 77 percent globally, which is crucial, given that treat- disease. First, case detection has been insufficient: in ing HIV-positive patients with ART reduces the risk of 2014, only about 64 percent of people who developed clinical TB by 64 percent. TB were notified as newly diagnosed cases, leaving Although dramatic improvements in TB control have approximately 3 million to 4 million cases that either been achieved over the last 25 years, the benefits have were not diagnosed or were diagnosed but not reported not been equally distributed among geographic regions. to national TB programs. Falling mortality rates were often fueled by rapid Tuberculosis 239 economic development in Asian countries. Three WHO Emerging drug resistance compelled the WHO regions—the Americas, South-East Asia, and Western to introduce DOTS-plus in 1998, with two additional Pacific—reduced prevalence 50 percent by 2013, but the requirements: (1) the capacity to perform drug-sensitivity other three regions have not yet achieved that goal. testing (DST) and (2) the ability to ensure access to In 2014, the WHO Africa region had the highest inci- second-line drugs (Stop TB Working Group on DOTS- dence rates (about 280 cases per 100,000 population) Plus for MDR-TB 2003). and 28 percent of the estimated number of cases globally. DOTS expanded further in 2014 to reflect six pro- Asia had 58 percent of total cases. China and India grammatic actions critical to an effective global TB control alone accounted for more than one in three (35 percent) strategy (box 11.1). Thus, the history of TB response of the world’s new TB cases in 2013. Approximately reflects attempts to tackle ever more fundamental causes of 78 percent of total TB deaths and 73 percent of TB TB by addressing health system drivers of the epidemic. deaths among HIV-negative people occurred in the Major organizational initiatives have sought to inten- Africa and South-East Asia regions. sify the fight against TB: the TB Alliance in 2000; the Green The spatial and temporal variation in TB incidence in Light Committee at the WHO in 2000; the Stop TB Africa is strongly correlated with the prevalence of HIV/ Initiative in 2001; the Global Fund to Fight AIDS, AIDS infection (Corbett and others 2003). Globally, an Tuberculosis, and Malaria (Global Fund), which was estimated 11 percent of all new adult TB cases were created as a major funder of country programs in 2002; infected with HIV/AIDS in 2015. In the WHO Africa UNITAID in 2006; and the Stop TB Partnership in region, the percentage of incident TB cases with HIV/AIDS 2008, which was created to coordinate, together with the varied from 8 percent in Eritrea to 77 percent in Swaziland. WHO, the many provider, advocacy, and donor groups Many of the gains in TB control globally are stalling, engaged in fighting TB. The Stop TB Partnership has 1,200 with TB incidence no longer falling in some East Asian partners in 100 countries. Further, the Global Laboratory settings, notably Hong Kong SAR, China; Japan; the Initiative was established in 2007 to strengthen laboratory Republic of Korea; and Singapore. Part of the explanation capacity in LMICs. could be that more cases are arising by reactivation from Increased attention and initiatives have been coupled a previously infected aging human population (Vynnycky with significant increases in international and domestic and others 2008; Wu and others 2010). Additionally, funding for TB. Funding for TB prevention, diagnosis, immigration from high-incidence countries is part of the and treatment reached US$6.6 billion in 2016, almost reason why the decline of TB in North America and double the level in 2006 (WHO 2015b). Overall about Western Europe has plateaued. Immigrants infected in 84 percent (US$5.5 billion) of reported TB funding their country of origin contribute, in varying degrees, to was derived from domestic sources; nevertheless, low- further transmission and outbreaks in the country where income countries (LICs) have benefited from Global they have come to live or work (Verver and others 2005). Fund financing, which accounts for about 67 percent of total TB funding for these countries. In 2016, most TB Global Response Following its declaration in 1993 that tuberculosis was a Box 11.1 global health emergency, the WHO launched the DOTS strategy in 1994 (WHO 1994). Prior to DOTS, there were WHO DOTS Strategy, 2014 at least five recommended regimens for treatment of TB, with varying effectiveness, serious adverse effects, and • Pursue high-quality DOTS expansion and varying costs (Murray, Styblo, and Rouillon 1990). DOTS enhancement aimed to create a single common best-practice strategy that • Address TB and HIV/AIDS co-infection, would be applicable to all countries and would include not MDR TB, and the needs of poor and vulner- just observing treatment but also securing political com- able populations mitments as well as adequate and sustained financing: • Contribute to health system strengthening based on primary health care • Ensure early case detection and diagnosis through • Engage all care providers quality-assured bacteriology • Empower people with TB and communities • Provide standardized treatment with supervision and through partnership. patient support Source: WHO 2014b. • Ensure effective drug supply and management • Monitor and evaluate performance and impact. 240 Major Infectious Diseases funding (US$ 4.9 billion) was spent on drug-susceptible • Halt and reverse the global growth in TB incidence TB, with MDR TB receiving about US$1.7 billion. • Decrease TB prevalence 50 percent from 1990 base- The WHO estimates that the gap in funding to achieve line levels by 2015 the SDG for TB is on the order of US$2 billion per year • Decrease TB mortality 50 percent from 1990 baseline (WHO 2015a, 2016a). levels by 2015. The current UN Sustainable Development Goals Evolving Targets seek to reduce tuberculosis deaths 95 percent and The initial DOTS targets aimed to achieve a global tuber- reduce the TB incidence rate 90 percent by 2035, effec- culosis detection rate of 70 percent and a treatment tively eliminating the disease. These are ambitious success rate of 85 percent. In 2000, the UN Millennium goals, and experts have outlined milestones in five- Declaration (UN General Assembly 2000), which estab- year increments to track progress and hold governing lished the MDGs, set in motion worldwide concerted bodies accountable. With evidence that transmission efforts to alleviate poverty and improve global health. and treatment failures in many countries continue at Astonishingly, the MDG goals did not specifically men- high levels, other metrics, including case detection tion tuberculosis, which was then causing more than and treatment success rates, are important indicators 9 million new cases and 1.5 million deaths annually. of progress. However, MDG 6, which aimed to halt and begin to Dye and others (2013) modeled what would be reverse HIV/AIDS, malaria, and “other infectious dis- required to eliminate TB as a public health problem eases,” provided sufficient language for international (less than 1 case per 100,000 population) by 2015. targets to be established for TB control. A projection of the decline in incidence at current rates The Global Plan to Stop TB 2001–15 of the Stop TB (around 1 percent annually) indicates that MDG goals Partnership established two targets in addition to the would not be met for more than 50 years. An updated general goal in MDG 6, for a total of three targets often model requires that incidence decrease 10 percent referred to in the TB literature: in 2020 and 17 percent a year thereafter (figure 11.3). Figure 11.3 Projected Acceleration in the Decline of Global Tuberculosis Incidence Rates to Target Levels, 2015–35 100 Current global trend: –1.5%/year Rate per 100,000 population/year 75 Optimize use of current and new tools –10%/year by 2025 emerging from pipeline, pursue universal health 50 coverage and social protection Introduce new tools: a vaccine, new drugs and –5%/year treatment regimens for 25 treatment of active TB disease and latent TB infection, and a point-of- –17%/year care test 10 2015 2020 2025 2030 2035 Year Source: WHO 2015c. Note: TB = tuberculosis. Tuberculosis 241 Box 11.2 WHO End TB Strategy Integrated, patient-centered care and prevention • Engagement of communities, civil society orga- • Early diagnosis of tuberculosis, including uni- nizations, and public and private care providers versal drug-susceptibility testing and systematic • Universal health coverage policy and regulatory screening of contacts and high-risk groups frameworks for case notification, vital registra- • Treatment of all people with tuberculosis, includ- tion, quality and rational use of medicines, and ing drug-resistant tuberculosis, and patient infection control support • Social protection, poverty alleviation, and actions • Collaborative tuberculosis and HIV/AIDS activi- on other determinants of tuberculosis. ties and management of comorbidities • Preventive treatment of persons at high risk and Intensified research and innovation vaccination against tuberculosis. • Discovery, development, and rapid uptake of new tools, interventions, and strategies • Research to optimize implementation and impact Bold policies and supportive systems and to promote innovations. • Political commitment with adequate resources for tuberculosis care and prevention Source: WHO 2015a, 2015c. However, such a “rate of decline has never been INFECTION AND DISEASE IN INDIVIDUALS achieved on any geographical scale for any period AND POPULATIONS of time and is not possible globally with the present suite of tools and systems for their delivery” (Dye 2013, Tuberculosis is an infectious bacterial disease caused 272–73). by Mycobacterium tuberculosis, which is transmitted by It is difficult to conceive that significant prog- aerosols and most commonly affects the lungs. Mtb is ress will be realized by 2035 without the develop- essentially found only in humans, although the related ment and application of new tools and investment pathogenic mycobacteria, M. bovis, causes disease in in health systems for TB, including reinstituting cattle and, before Pasteurization of milk, was the cause active case finding, devising more effective delivery of scrofula, TB of the lymph nodes. A related pathogen, strategies, investing in the supply chain and informa- M. leprae, is the cause of leprosy in humans. Because tion management systems, conducting research and there is no animal reservoir for Mtb, the pathogen has development into new diagnostics, and implement- evolved to persist in people for long periods of time, ing new and more effective treatment regimens and with only a portion of people developing clinical disease vaccines. with lung damage. The WHO’s current strategic plan for TB con- TB is transmitted from person to person via aerosol trol, the End TB Strategy (Uplekar and others 2015; droplets from the throat and lungs of people with WHO 2015c, 2016a) is a multifaceted program far active respiratory disease. Individuals with pulmonary more extensive than previous iterations. Recognizing or laryngeal tuberculosis produce airborne droplets both the successes of previous programs in reduc- while coughing, sneezing, or simply speaking (Lin and ing both mortality and prevalence and the failure others 2008; Loudon and Spohn 1969; Rodrigo and of present programs to reduce incidence at a rate others 1997). Inhaled infectious droplets lodge in the that will enable countries to meet the SDG targets by lung alveoli and bacilli and are taken up there by 2035, it proposes a broader and more ambitious pro- macrophages. gram based on three pillars (box 11.2). These are very ambitious and important goals that, to a large extent, will depend on investments in research Stages of TB Disease and Intervention Points and development of new tools and more effective use of TB is best understood not as a single clinical entity the available tools. but as a spectrum, which generally correlates with the 242 Major Infectious Diseases immune responses. A high, but unknown, percentage of Figure 11.4 The TB Spectrum and Possible Points of Intervention people infected with Mtb develop latent or persistent infection, but only about 10 percent develop disease in a lifetime (Chee and others 2005; Hanifa and others The spectrum of TB: Points of intervention 2009). For immune-compromised individuals, the risk is about 8 percent a year (Selwyn and others 1989). A TST Latent TB Chronic Pulmonary/ Miliary negative TST+ TB extrapulmonary TB portion of highly exposed individuals likely to be infected with Mtb fail to develop tuberculin skin tests Innate Acquired (TSTs) but remain healthy, suggesting that they are pro- tected by yet-unknown immune mechanisms. Another Miliary Prevent infection unknown proportion of individuals with tubercle bacilli Prevent or cure latency in sputum remain asymptomatic but it must be pre- Prevent transition from latent to active disease sumed capable of transmitting infection (Bates and Treat and cure disease reducing transmission others 2012; Mao and others 2014; Willingham and others 2001; Wood and others 2007). Multiple possible points for interventions exist along the care continuum, including preventing infec- Note: TB = tuberculosis; TST = tuberculin skin test. Infection with Mycobacterium tuberculosis tions, preventing establishment of latency, preventing can lead to subclinical infection, which is cured by innate or acquired immune responses; latent infection that persists; chronic or asymptomatic TB in which tubercle bacilli are found in sputum transition from latent TB to active disease (chemopro- of otherwise apparently healthy individuals; and active TB disease. In a small proportion of phylaxis), and treating persons with active disease to individuals, infection can lead directly to rapidly progressing disease, known as primary achieve cure, thereby reducing morbidity, mortality, progressive TB. Miliary TB is widely disseminated, occurring primarily in children and severely immunodeficient individuals. and transmission intensity (figure 11.4). In most healthy people, infection with M. tuberculo- sis often causes no symptoms, since the person’s immune system, through innate and acquired immunity, acts to is unknown, but the risk of reactivation persists into old kill or wall off the bacteria. Acquired cell-mediated age (Stead and Dutt 1989). immunity develops two to eight weeks after infection, The phenomenon of reactivation, often long after and granulomas—that is, infiltrating macrophages and infection or after treatment, reflects the limitations of lymphocytes—wall off the infection and limit further immune responses in assuring protection and indicates replication and spread of the organism (Aziz, Ishaq, and the challenge facing the development of effective vac- Akhwand 1985), although this is only partially protective cines that can provide long-term protection (Lin and (Andrews and others 2012; Bates and others 2007; da others 2014; Lönnroth and others 2010; Selwyn and Silva and others 2014; Lin, Ezzati, and Murray 2007; others 1989). Bacille Calmette-Guérin (BCG) is the only Slama and others 2007). licensed vaccine available today and has been used for There is a widespread belief that infection does not more than 90 years with a good safety record, except in confer significant subsequent immunity (Achkar and immunodeficient children. However, its efficacy in pre- Jenny-Avital 2011; Barry and others 2009), which is venting TB in adults has varied in different parts of the used to explain why reinfection is not uncommon world, while consistently protecting children against the (Luzze and others 2013; Verver and others 2005). most severe forms of disease: disseminated miliary TB However, recent reanalysis of many studies reveals that and TB meningitis (Mangtani and others 2014; Roy and latent TB infection does protect against active disease, others 2014). BCG vaccination is discussed further in the almost certainly by engendering protective innate or section on prevention. acquired immune responses, but that this protection is Primary infection in some individuals leads to only partial (Andrews and others 2012). Molecular fin- active TB (primary progressive tuberculosis) when the gerprinting techniques can be used to distinguish bacte- host immune response cannot effectively suppress ria obtained during relapses of prior infection from the replication of bacilli. The symptoms of active TB reinfection with new strains. The original infecting of the lung are coughing, sometimes with blood in strain can reemerge after apparent cure or ineffective sputum; chest pains; weakness; weight loss; fever; and treatment and reestablish active disease (relapse) (Khan, night sweats. Clinical tuberculosis is the sum of com- Minion, and others 2012; Luzze and others 2013). plex interactions between the pathogen and an indi- Reactivation rates vary from 1 percent to 30 percent a vidual’s immune response that facilitate mycobacterial year in different populations. Whether latent bacilli replication and cause illness, including wasting and remain viable for the full life span of all infected people granulomatous inflammation with tissue damage, Tuberculosis 243 for example, caseation, vasculitis, and fibrosis (Jagirdar case, two antagonistic subclasses of CD4+—Th1 and and Zagzag 1996; O’Garra and others 2013; Shaler and Th2—have been described, each with its own set others 2013). The most common clinical manifestation of cytokine mediators. Th1 responses, characterized of TB is pulmonary disease, and, in the most infectious by production of interferon gamma (IFN-γ), are patients, bacilli are visible microscopically on stained associated with protection, while Th2 responses, sputum smears (50–70 percent of pulmonary cases) characterized predominantly by Th2 cytokines (for (Huang, Tchetgen, Becerra, Cohen, Galea, and others example, IL-4, IL-10), are associated with antibody, 2014; Huang, Tchetgen, Becerra, Cohen, Hughes, and inflammatory responses and tissue damage. A unique others 2014). Extrapulmonary tuberculosis accounts subset of human Th1 cells has recently been described for 10–30 percent of disease, but is more common that appears strongly to correlate with protection among women and children (particularly lymphatic against mycobacterial disease. These Th1* cells consti- tuberculosis) and in people infected with HIV/AIDS tute a unique subset of human CD4+ T-cells expressing (Chadha and others 2005; Lowell, Edwards, and Palmer the chemokine receptors—CCR6, CCR4, and the RORγ 1969; MacIntyre and others 1997). nuclear transcription factor, which exclusively appears In the absence of other predisposing conditions, to produce IFN-γ to mycobacteria (Sallusto 2016). only 5–10 percent of infected people develop progres- The importance of TB among infectious diseases is sive primary disease within five years of infection attributable to the high case fatality rate among untreated (Chee and others 2005; Hanifa and others 2009; van or improperly treated cases. About two-thirds of Rie and others 2013). After five years, there is a much untreated smear-positive TB cases will die within five to lower annual risk of developing TB by the reactivation eight years; most will die within the first two years of latent infection. However, the risk in HIV-positive (Huang, Tchetgen, Becerra, Cohen, Hughes, and others individuals is on the order of 10 percent a year after 2014; Libshitz and others 1997). As illustrated in infection (Selwyn and others 1989). The risk of pro- figure 11.4, the rest will remain latent, chronically ill, or gressing to active disease is relatively high in infancy, is asymptomatic, or will self-cure. The case fatality rate for lower in older children, increases quickly during untreated microscopy smear-negative cases is lower, but adolescence (earlier in girls), and then increases more still on the order of 10–15 percent (Chadha and others slowly throughout adulthood (da Silva and others 2005; Khan, Minion, and others 2012; Libshitz and 2014; Hanifa and others 2009; Isler and others 2013; others 1997). Even among smear-positive patients Lienhardt and others 2003). The lifetime risk of devel- receiving TB drugs, the case fatality rate can exceed oping TB following infection clearly depends on the 10 percent if adherence is low or if rates of HIV/AIDS prevailing transmission rate, but is generally estimated infection and drug resistance are high. to be about 10 percent. Latent TB, which exists in an unknown percentage of people infected with M. tuberculosis, has significant M. tuberculosis Strains impact on the epidemiology and population dynamics There is striking evidence that the major strains of of tuberculosis. It represents a huge reservoir of poten- M. tuberculosis co-evolved with the major migrations tial disease and further transmission. Concomitantly, of humans from Africa to Asia, Europe, and the long-term latent infection appears to provide partial Americas (Gagneux 2012). Many strains of Mtb can be protection against developing disease (Andrews and revealed by molecular analyses, and the diversity is others 2012). thought to exacerbate drug resistance and to affect the Following Mtb infection, whether the infection effectiveness of interventions to control the disease. remains latent or develops into active disease is thought While early targeted genetic analyses suggested only to depend largely on the host’s ability to generate pro- minimal within-species diversity of M. tuberculosis tective innate and cell-mediated immunity. There is at (Keane and others 2001; Yokoyama and others 2004), present little evidence that serum antibodies provide genomic studies have revealed much more variation protection, although recent studies of serum from (Alhajri and others 2011). Subsequent examination has TST-negative highly exposed individuals indicate they provided increasing understanding of Mtb strains and may have antibodies capable of enabling macrophages how they spread globally (Talat and others 2010). to kill some bacilli in vitro (Lu and others 2016). Other investigations aim to discover whether differ- Human T-cells are highly heterogeneous. From animal ences between or within (according to Cegielski, Arab, studies, both CD4 cells and cytotoxic CD8 T-cells are and Cornoni-Huntley 2012) strains modify the ability important for protection against Mtb infection. CD4+ of the pathogen to infect hosts or are associated with cells are functionally heterogeneous. In the simplest differences in the natural history of disease (Alhajri and 244 Major Infectious Diseases others 2011; Wilkinson and others 2000). The number individuals are diagnosed and receive effective treatment. and scope of such studies is still limited, but a recent The duration of HIV/AIDS-associated TB appears to be study has associated differences in Mtb strains with the shorter than the duration of HIV-negative TB (Corbett probability of transmission of disease among house- and others 2004) or about the same (Wood and others hold exposures in Brazil (Lopez and others 2003). 2007), depending on the setting. Evidence is accumulating that strain lineages vary in Clearly, HIV/AIDS infection exerts a multifaceted strength and mechanism of host-immune stimulation suppression of the innate and acquired T-cell responses. after infection (Baker and others 2012), within-host In a sense, TB is often a sentinel for HIV/AIDS infection competitive ability (Boelaert and others 2007), rates of in high HIV/AIDS-endemic areas. Even prior to signifi- acquiring mutations (Bellamy and others 1998; Ford cant CD4+ T-cell depletion, individuals with latent TB and others 2013) and drug resistance (Borrell and can progress to active disease. Not only does HIV/AIDS Gagneux 2011), and the specific mutations acquired infection suppress immune responses to Mtb, but the (Fenner and others 2012), each of which may affect the stimulation of T-cells by antigens of Mtb may contribute within-host course of infection, disease, and response to T-cell activation, leading to the increased production to therapy (Gagneux 2012). In general, the fitness of of HIV/AIDS and acceleration of the disease process. pathogens diminishes as mutations accumulate. A vari- Clinically, the prevalence of extrapulmonary TB and ety of evidence indicates that MDR TB strains are het- disseminated TB are both increased in HIV-positive erogeneous in their transmissibility in animal models patients. Low CD4 cell counts are associated with an and human populations (Grandjean and others 2015; increased frequency of extrapulmonary TB, positive Lee, Radmonski, and others 2015). It is likely that com- mycobacterial blood cultures, and atypical chest radio- promised transmissibility can change if compensatory graphic findings, reflecting an inability of the impaired mutations arise that reestablish fitness. A particular immune response to contain infection. The rise in TB concern is the enhanced transmissibility of the Beijing incidence attributable to HIV/AIDS appears to have strain with antibiotic resistance (Ford and others 2013; peaked in most countries, as HIV/AIDS incidence has Hanekom and others 2007). Mathematical modeling declined (Dye 2015). similarly indicates that strain diversity may affect the emergence of drug resistance (Basu and Galvani 2008) and interventions (Cohen and others 2008; Colijn and Effect of TB on the Distribution of Other Diseases others 2009), but improved projections will require TB affects the presence and nature of other diseases, additional data, especially from whole-genome possibly conferring protective effects. Microbial infec- sequencing and long-term monitoring of strain types tions have the potential to influence the balance within human populations. between CD4+ T-cell functional subsets by stimulating innate immune responses and by altering cytokine pro- files, with positive or negative consequences for health TB and HIV/AIDS Co-Infection (Sallusto 2016). Mtb infection may also protect against The extent to which HIV/AIDS is fueling TB transmis- asthma, possibly by shifting the innate and acquired sion (in addition to provoking reactivation) remains Th2 response to a Th1 subset that reduces the inflam- poorly understood. One analysis suggested that 1–2 matory response. One study of Japanese children found percent of all transmission events were from HIV- that strong tuberculin responses following BCG immu- positive, smear-positive TB cases in 2000 (Corbett and nization were associated with less asthma, rhinocon- others 2003). The co-infection of TB and HIV/AIDS is junctivitis, and eczema in later childhood (Shirakawa geographically heterogeneously distributed. In countries and others 1997). A study of South African children in Eastern Europe and Southern Africa, as many as 75 found an inverse association between M. tuberculosis percent of all TB patients are HIV-positive; in others, infection and atopic rhinitis (Obihara and others such as China and India, only a small fraction of TB 2005). Comparisons among countries have found that cases are HIV-positive (Dye 2015). asthma tends to be more common where TB is less The fraction of TB infections attributable to persons common (Shirtcliffe, Weatherall, and Beasley 2002; von co-infected with HIV/AIDS depends on the prevalence Mutius and others 2000). of HIV/AIDS as well as on the infectiousness of HIV/ Interactions between other infections have also been AIDS-associated Mtb compared with that of TB cases investigated. Vigorous Th2 responses are seen in pro- not affected by HIV/AIDS. This fraction is influenced by tective immune reactions to helminth infections, and biological factors (for example, the probability of helminths can shift the balance of immune responses smear-positive pulmonary disease) and how rapidly to enhance allergic responses and thus compromise Tuberculosis 245 Th1 immune responses to BCG and M. tuberculosis risk of TB (Bates and others 2007; Lin, Ezzati, and (Hopkin 2000). Conversely, a mycobacterial-based Murray 2007; Slama and others 2007). vaccine could potentially be constructed to prevent In contrast to infection, disease progression is known allergic responses and reduce asthma. Mtb infection to be highly dependent on host risk factors, the most may protect against leprosy, as does BCG (Karonga important of which include HIV/AIDS co-infection Prevention Trial Group 1996), and natural TB trans- (Selwyn and others 1989), low body mass index mission may have contributed to the decline of leprosy (Lönnroth and others 2010), exposure to tobacco in Europe (Lietman, Porco, and Blower 1997). (WHO 2015d) and biomass fuels (indoor air pollution; There is no information at present on whether the Bates and others 2007; Lin, Ezzati, and Murray 2007; human microbiome affects responses to mycobacteria, Lin and others 2014; Slama and others 2007), diabetes but this is an area of research worthy of investigation. mellitus (Jeon and Murray 2008), and heavy alcohol use While the synergistic and antagonistic interactions (Lönnroth and others 2008; Rehm and others 2009; between bacterial, viral, and parasitic infections are WHO 2014a). Host-specific risk factors also affect TB complex and unresolved, these examples raise the gen- outcomes, including the risks of failing therapy, relaps- eral likelihood that mycobacteria influence, and are ing after treatment, and dying a TB-related death. In influenced by, the presence of other infections. addition to HIV/AIDS, smoking and diabetes are recog- nized biological risk factors for poor treatment out- comes (Kim and others 2014), and some studies have Risk Factors for TB implicated other comorbidities such as iron overload Risk factors influence the probability of infection, dis- (Yokoyama and others 2004), renal dysfunction (Keane ease, or outcome and operate on many scales (physio- and others 2001), and hematological malignancies logical, genetic, environmental, and behavioral). Once (Keane and others 2001). an individual has been exposed to a person with infec- Abundant evidence indicates that undernutrition tious pulmonary TB, his or her risk of developing sub- is associated with TB in LMICs. In national surveys clinical TB infection depends on factors that influence in India, the population-attributable risk of TB in either the ability of the person infected to transmit the undernourished adults and adolescents was two-fold disease or the susceptibility of the person exposed to or greater and greatest in rural areas (Bhargava and infection and disease. Infected persons who are acid-fast others 2014). bacillus smear- or culture-positive (Riley and Moodie Table 11.2 lists the risk factors for TB progression and 1974; Ross and Willison 1971; Tornee and others 2005), summarizes the relative risks for selected determinants who have cavitary disease (destructive lesions in the for which meta-analyses have been conducted. Although lung where the bacilli multiply to high levels; Rodrigo HIV/AIDS is a much stronger risk factor for disease pro- and others 1997) or frequent cough (Loudon and Spohn gression than other exposures, the relatively frequent 1969), or who have delayed treatment (Aziz, Ishaq, and occurrence of other determinants means that they Akhwand 1985; Lin and others 2008) are major trans- explain a higher proportion of global TB cases than does mitters of TB infection. HIV/AIDS. Table 11.3 estimates the most common Risk factors relevant to the exposed host most often reflect the social and environmental determinants of heavy exposure and include living in densely popu- Table 11.2 Key Risk Factors for Tuberculosis from lated spaces (Chadha and others 2005; Lowell, Edwards, Meta-Analyses of Randomized Controlled Trials and Palmer 1969; MacIntyre and others 1997), being incarcerated (Chadha and others 2005; Chee and oth- Key risk factor Odds ratio ers 2005), and working in occupations such as health Cigarette smoking 2.01–2.66 care that involve frequent social or direct contact with Indoor air pollution 1.4 TB patients (Hanifa and others 2009; Isler and others 2013; van Rie and others 2013). Most studies suggest Low body mass index 2.45a that, among similarly exposed contacts, the risk of TB Alcohol use (daily or alcohol use disorder) 2.94 infection does not vary much by host attributes. Diabetes mellitus 3.11 However, some recent studies report that genetic loci Source: International Institute for Population Sciences and Macro International 2007. are associated with differential risk of infection among Note: Human immunodeficiency virus/acquired immune deficiency syndrome household contacts exposed to an infectious case (da (HIV/AIDS) does not appear in the table because once it was clear that HIV/AIDS was a major risk factor for TB, it became unethical to do a prospective study that Silva and others 2014; Lienhardt and others 2003), did not offer HIV-positive patients isoniazid. while evidence indicates that smoking increases the a. Odds of tuberculosis for body mass index of 18.5 compared to 25. 246 Major Infectious Diseases Table 11.3 Attributable Fraction of Key Risk Factors for Tuberculosis Disease Progression in India, by Socioeconomic Strata Population-Attributable Fraction (%) Lowest Middle Highest Key risk factor socioeconomic stratum socioeconomic stratum socioeconomic stratum Cigarette smoking 16 10 6 Indoor air pollution 29 25 6 Low body mass index 34 27 20 Alcohol use, daily 4 2 1 HIV/AIDS seroprevalence 9 10 6 Source: Oxlade and Murray 2012. Note: HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome. Table 11.4 Global Prevalence, Relative Risk, and Attributable Fraction for Incident Tuberculosis Key risk factor Global prevalence Relative risk Attributable fraction (%) References HIV/AIDS 0.008 8.3 5.5 WHO 2009a Undernourishment 0.11 2.1 10.7 Lönnroth and others 2010 Diabetes mellitus 0.085 3.0 14.5 Jeon and Murray 2008 Heavy alcohol use 0.075 2.9 12.5 Lönnroth and others 2008 Cigarette smoking 0.21 2.6 25.1 Lin, Ezzati, and Murray 2007; Slama and others 2007 Indoor air pollution 0.41 1.5 17.0 Lin, Ezzati, and Murray 2007 Note: Relative risk estimates the magnitude of an association between exposure and disease on the basis of the incidence of disease in the exposed group relative to the unexposed group. Attributable risk is the absolute difference in incidence between an exposed and unexposed group that quantifies the risk of disease in the exposed group attributable to the exposure by removing the risk that would have occurred due to other causes. attributable risk factors in different economic strata in Risk factors also vary by socioeconomic status, as India using data from the Indian National Family Health illustrated for India. While infection with Mtb is a risk Survey (International Institute for Population Sciences for people of any economic stratum, the data from India and Macro International 2007). These data show that indicate that some known risk factors for TB are greatest multiple risk factors often converge in individuals living in the lowest socioeconomic group, a finding likely to be in poverty, further amplifying their risk of disease. true for most populations. HIV/AIDS is associated with only about 11 percent of Other less common comorbidities also modify the risk TB patients worldwide. Other risk factors, such as diabetes of disease. Persons are more likely to progress to active mellitus and smoking, occur more widely. The population- tuberculosis if they suffer from silicosis (Corbett and attributable fraction is the proportional reduction in others 2000; Snider 1978), kidney disease (Mitwalli 1991), population disease or mortality that would occur if expo- or solid (Libshitz and others 1997) and hematological sure to a risk factor were reduced to an alternative ideal (Khan and others 2005) malignancies; have undergone exposure scenario, for example, no smoking. A global gastrectomy or ileojejunal bypass surgery (Choi and estimate of population-attributable factors for TB is given others 2015; Kim and others 2014; Yokoyama and others in table 11.4. It is not possible to estimate accurately the 2004); or have received the tumor necrosis factor alpha number of people with each of the risk factors because (TNF-α) inhibitor infliximab for the treatment of rheu- the data on background risk are uncertain and risks matoid arthritis (Keane and others 2001). While these overlap. Nevertheless, the greater numbers of smokers and exposures are rare, some dramatically increase the risk of rapidly expanding number of people with diabetes melli- TB; older forms of weight-loss surgery, for example, can tus allow us to infer that the proportion of all cases due to profoundly increase risk. To date, although several case malnutrition and diabetes is five times higher and the pro- reports document TB among patients undergoing gastric portion due to smoking is eight times greater than the bypass, no systematic epidemiologic studies have been proportion due to HIV/AIDS. conducted on this risk factor (Alhajri and others 2011). Tuberculosis 247 In addition to low body mass index, several micronu- regions of the human genome, but these have not been trient deficiencies have been associated with TB consistently validated in separate populations. A locus progression. In vitro studies suggest a role for vitamin D on chromosome 11 has been associated with suscepti- in host susceptibility to disease and, in one clinical study bility in multiple populations (Chimusa and others conducted in Pakistan, 25-hydroxy-vitamin D—25(OH) 2014; Thye and others 2012). Of particular interest is a D—levels less than 9 nanograms per milliliter increased locus on chromosome 5 that encodes a component the risk of progression to active disease fivefold (Talat of interleukin 12 (IL-12) required for differentiating and others 2010). Vitamin A deficiency was found to be T-cells, which appears to confer resistance in highly associated with a 2.8 increased risk of TB in the United susceptible HIV-positive populations in East Africa States, although this finding was not statistically signifi- and for which there is evidence of positive selection cant (Cegielski, Arab, and Cornoni-Huntley 2012), and (Sobota and others 2016). reciprocal seasonal variation in vitamin D levels in South Africa correlate with TB notifications (Martineau and TB DIAGNOSIS AND SCREENING others 2011). Nutritional factors may also interact with genetic In simplest terms, the DOTS strategy for TB control polymorphisms to increase TB risk. Polymorphisms in has been to test individuals who seek care at a health the 25(OH)D receptor have been associated with TB facility for clinical symptoms of TB and to provide risk, and several studies have demonstrated a gene- appropriate drug treatment for a period of 6–24 environment interaction between 25(OH)D levels and months. With timely diagnosis and correct treatment, 25(OH)D receptor mutations (Wilkinson and others almost all people with drug-sensitive TB can be cured, 2000); there is considerable evidence that 25(OH)D3 is and even a short duration of treatment reduces the essential for human macrophages to kill Mtb in vitro bacillary load and likelihood of transmission. (Fabri and others 2011; Liu and others 2007). Vitamin Nevertheless, worldwide, the TB case detection rate D is produced in the skin by exposure to ultraviolet remains low: in 2012, about 66 percent (5.7 million) of light, and seasonal variation in vitamin D has been cor- the estimated 10.4 million people who developed TB related with the number of TB cases (Wilkinson and were newly diagnosed cases, with an estimated 3 million others 2000). This finding and the role of the skin pig- to 4 million cases remaining undiagnosed or unknown ment melanin to absorb ultraviolet light may explain to health systems. Case detection in children is of par- the increased susceptibility of dark-skinned individuals ticular concern: an estimated 1 million children devel- to TB infection and more severe disease (Martineau and oped tuberculosis in 2010, with about 32,000 children others 2011; Modlin and Bloom 2013). But correlation contracting multidrug-resistant TB disease (Seddon does not imply causation, and there is a critical need for and others 2015). well-designed clinical trials to ascertain the importance In most countries, the most common diagnostic of these factors. test is microscopic scanning of acid-fast stained bacilli Much recent work has focused on identifying the in sputum smears—a technique dating to the 1880s. genetic determinants of TB progression (Abel and oth- It is convenient but insensitive, diagnosing only ers 2014). Twin studies strongly support the hypothesis about half of all TB cases in adults (Frieden 2004) that genetic factors play a role in TB susceptibility, and and fewer in children (Detjen and others 2015) and multiple loci have been implicated through candidate HIV-positive individuals (Harries 1997). Diagnostic gene studies (Snider 1978). Some of these are rare vari- certainty is obtained when the organism is demon- ants that lead to alterations in the interferon-γ pathway strated in a laboratory after clinical evaluation of required to develop acquired immunity to mycobacteria. symptoms (usually cough) compatible with tubercu- Multiple defects in this pathway result in Mendelian losis. In the absence of diagnostic laboratory tests, susceptibility to mycobacterial diseases, which predis- clinicians need to review clinical information and poses individuals not only to disseminated infections decide whether to initiate treatment for tuberculosis, with nontuberculous mycobacteria, but also to tuber- weighing the risks of leaving possible TB untreated culosis (Bustamante and others 2014; Mitwalli 1991). against adverse drug reactions and the social and Other studies have implicated candidate genes that financial costs of committing to months of therapy affect innate immune responses (Png and others 2012). (WHO 2007). Often treatment, appropriate or not, is Several genome-wide association studies have also instituted before it is clear whether the patient has TB been reported, as reviewed in Naranbhai (2016). and whether the infection is drug-susceptible or Some have identified alleles, which occur in “gene-free” drug-resistant. 248 Major Infectious Diseases Diagnosis of MDR and XDR TB treatment regimens is often weak in health systems Diagnosis and treatment of MDR TB, in particular, has where MDR TB dominates (Keshavjee and Farmer largely faltered worldwide. Only 30 of 107 countries are 2010; Nardell and Dharmadhikari 2010). treating 75 percent or more of patients with MDR TB, Several economies—including Estonia; Hong Kong with countries experiencing high levels of loss to SAR, China; Latvia; and Singapore—that have strength- follow-up (WHO 2013a). Of the estimated 450,000 ened their health systems by improving access to diag- people who developed MDR TB in 2012, only 94,000 nosis and primary care treatment have halted the rise (20.9 percent) were detected, and just 77,000 were in MDR TB incidence (Cohen and others 2014; Cuevas started on second-line treatment. MDR and XDR TB and others 2011; Dye 2009). Strong laboratory capacity, also represent a threat to health care personnel and which has enabled rapid and definitive determination health infrastructure. Unknown numbers of nurses and of drug sensitivity, strong supply chain management physicians have acquired MDR and XDR TB, and in systems, and successful scale-up of effective treatment 2014, there were an estimated 210,000 deaths from regimens have contributed to this success (Gandhi and MDR TB (WHO 2015b). others 2010). In this context, it is technically difficult The diagnosis of drug-resistant TB and its treat- for countries to diagnose patterns of drug resistance. ment are complex, requiring laboratory capability for For this reason, the WHO created the TB Supranational drug-sensitivity testing and between 9 and 20 months Reference Laboratory Network of 24 quality-control of daily administration of drugs that are both more laboratories, which are located in every continent and toxic and less efficacious than the drugs used to treat able to carry out sophisticated testing for drug resis- drug-sensitive TB (Nathanson and others 2010). tance (WHO 2015a). Inadequate human resources, poor access to laboratory services, and low capacity to do drug-susceptibility testing and analysis partly account for low case detec- Recent Advances in TB Diagnostics tion for MDR TB (Shin and others 2008). Health sys- Tuberculosis diagnostics have advanced steadily over the tem approaches that favor hospital-based management past decade (see box 11.3). As a result, between 2007 and of MDR TB frequently have limited access to service 2012, the WHO issued 10 new policy statements on TB delivery, and scale-up of new diagnostic tools and diagnosis covering an array of approaches (Lawn 2015). Box 11.3 New Strategies for TB Diagnosis • Use of light-emitting diode (LED) microscopy as • Use of liquid culture systems as a more rapid and an alternative to conventional Ziehl-Neelsen light sensitive alternative to conventional solid culture microscopy, which has been the mainstay of TB (Palacios and others 1999). diagnosis for decades (Cuevas and others 2011) • Use of nucleic acid amplification tests (NAATs) Other avenues for developing new TB diagnostics for diagnosis of active TB, including manual hold promise: technologies such as loop-mediated isothermal • Urine-based diagnostics for detecting M. tuberculo- amplification as well as automated technologies sis antigen, for example, assays to detect lipoarabino- such as Xpert MTB/RIF (Pai, Kalantri, and Dheda mannan, especially in HIV-positive patients (Green 2006) and others 2009; Nakiyingi and others 2014) • Use of nucleic acid amplification technology • Immunochromatographic tests for rapid confirma- approaches for rapid screening for drug resis- tion of Mtb in culture (Hasegawa and others 2002) tance, for example, based on line probe assays • Exhaled air mass spectrometry for volatiles and (Pai, Kalantri, and Dheda 2006) chemical analysis (Phillips and others 2007). Tuberculosis 249 Among the new diagnostic options that have A recently developed molecular approach examin- emerged in recent years, the Xpert MTB/RIF test has ing gene expression of peripheral blood cells rather received the most attention. Xpert MTB/RIF is an than sputum has the potential to identify the subset automated deoxyribonucleic acid (DNA) amplification of healthy individuals with latent TB who are likely test that provides rapid and sensitive detection of TB to progress to active disease (Zak and others 2016). and rifampicin resistance. It uses a cartridge-based Rather than detecting components of the pathogen, system that integrates sample processing and real-time this novel method measures gene expression signa- polymerase chain reactions, accommodates use by rel- tures in peripheral white blood cells that are elevated atively unskilled health care workers, and provides in healthy individuals with latent infection prior to results in less than two hours. The system is expensive, their progression to active TB. In a panel of 16 gene costing about US$17,000 per unit, and the tests, cur- probes in three separate cohorts in different coun- rently subsidized, cost about US$10. The ability to tries, it was possible to predict persons who pro- diagnose TB and identify MDR TB from sputum in less gressed to active disease six months to one year than two hours is a major step forward in linking diag- before any symptoms could be detected clinically. At nosis to rapid initiation of treatment. However, Xpert the one-year point prior to diagnosis, the specificity MTB/RIF is currently not a technology for point-of-care of the test was around 61 percent; in HIV-positive diagnosis. In December 2010, the WHO recommended individuals, it was significantly higher, at around that the device be used for initial diagnosis in patients 80 percent. The molecular exploration of host suspected of having MDR TB or HIV/AIDS-associated responses offers new possibilities for diagnosing TB disease. Subsequently, some countries, including infection and defining the gene signatures of persons China, India, and South Africa, have purchased Xpert who do not progress to active disease, potentially equipment at reduced prices and taken advantage of enabling understanding of the genes required for volume pricing to purchase test cartridges. resistance to disease. In a similar approach, gene The Xpert system is a significant advance in acceler- expression in the whole blood of patients with either ating the diagnosis of TB, particularly MDR TB, and will latent tuberculosis or other diseases versus patients likely be a valuable new tool for major hospital and TB with active tuberculosis was compared using a vali- diagnostic laboratories, despite being dependent on a dated multicohort analytical framework. The diag- sophisticated and expensive device and relatively expen- nostic capacity of a three-gene set was found to be sive costs for each sample tested. However, a multicenter 88–90 percent in active and latent TB in samples trial in four African countries failed to demonstrate from children and adults in 10 countries (Sweeney lower TB-related morbidity (Theron and others 2014). and others 2016). Such molecular host signatures A new device model being developed, the GeneXpert could potentially serve as biomarkers for defining Omni, which is portable and battery operated, has the determinants of protection against infection or dis- potential to become a point-of-care diagnostic test in ease in future studies and vaccine trials. many more sites and is to be released later in 2017. Since shortening the time between diagnosis and ini- tiation of appropriate treatment is a major factor in TB TREATMENT reducing transmission, technologies that allow diagnosis Treatment aims to cure the disease process, rapidly and drug-sensitivity testing at the point of care are ideal. stop transmission, and prevent relapse (WHO 2006). Some innovative research is under way to achieve that Current treatment of tuberculosis requires multiple goal, but point-of-care testing remains a formidable antibiotics, guided by predicted or demonstrated anti- challenge. Considering the sheer number of patients biotic susceptibility and taken for many months. queued in busy outpatient departments, it is unlikely Context-specific treatment guidelines are usually that cough screening and sputum testing can be developed by local health authorities with guidelines effectively implemented in many of the highest-risk and oversight from the WHO. Clinical trials in the ambulatory settings—or in all resource-limited settings. twentieth century established current first-line drug Even the DNA amplification methods lack the sensitivity regimens (Fox, Ellard, and Mitchison 1999; Mitchison to detect many patients with early disease. Potentially 2004). Treatment success rates of 85 percent or more infectious TB cases will be missed, delays in diagnosis for new drug-sensitive cases are regularly reported to will occur, and patients with XDR TB will likely not the WHO from a wide variety of clinical settings respond promptly to current therapy. Traditional meth- (WHO 2012a, 2015b). ods of control will be necessary for the foreseeable future Treatment effectiveness has been eroded, however, by (WHO 2009a). the evolution and transmission of multidrug-resistant 250 Major Infectious Diseases tuberculosis. Treatment for MDR TB, which is defined as Treatment Regimens resistance to isoniazid and rifampicin (the two most Effective tuberculosis treatment needs to overcome the effective TB drugs) is longer and requires more expen- organism’s ability to persist in diverse microenviron- sive and more toxic drugs. For most patients with MDR ments under extreme conditions, including immuno- TB, the current regimens recommended by the WHO logical attack, prolonged antibiotic exposure, and last 18–24 months, and treatment success rates are much nutrient and oxygen depletion (Islam, Richards, and lower, around 60 percent. The WHO now conditionally Ojha 2012; Shaler and others 2013). Standardized treat- recommends using seven drugs to reduce the time of ment regimens and fixed-dose combination medica- treatment to nine months for uncomplicated pulmonary tions simplify good clinical care in resource-limited disease (WHO 2016b). New drug combinations, for settings. Table 11.5 presents current treatment regi- example, including bedaquiline or delaminid, which are mens, with an intensive phase followed by a continua- thought to act on new molecular targets, are being intro- tion phase (Chakraborty and others 2013; Donald and duced, but an ideal combination is likely several years McIlleron 2009; Shi and others 2011; WHO 2010, away (Villemagne and others 2012; Zumla, Nahid, and 2013b, 2016a, 2016b). Cole 2013). Patients who are effectively treated for tuberculo- First-Line Treatment of Drug-Susceptible TB sis usually show clinical response within 8–12 weeks, Rifampicin and isoniazid are the most potent drugs for both subjectively (reduced cough, fatigue, fevers, and susceptible TB and are taken throughout the course of sweats; increased appetite) and objectively (sputum first-line treatment (Donald and McIlleron 2009; smear or culture conversion from positive to nega- WHO 2010). Pyrazinamide synergistically reinforces tive; weight gain) (WHO 2010). Failure to respond to the sterilizing activity of rifampicin and, when added treatment is typically due to poor drug quality, to the first two months of treatment, reduces the dura- underdosing or malabsorption, nonadherence, drug tion of treatment to six months (Fox, Ellard, and resistance (which may broaden while on treatment), Mitchison 1999; Hong Kong Chest Service and British paradoxical reactions or immune reconstitution Medical Research Council 1979). Ethambutol is added inflammatory syndrome (IRIS), adverse drug effects, to the regimen for two months to reduce on-treatment or another disease process (bronchiectasis, malig- development of drug resistance (WHO 2010) and is nancy, pneumoconiosis, autoimmune disease, or continued for the full duration of therapy in settings organ failure). with high background prevalence of isoniazid resis- One of the embarrassing deficits in the field of TB tance. As effective as standard treatment has been, control is the ambiguous definition of “cure.” In the resistance to isoniazid, rifampicin, and pyrazinamine is twenty-first century, it should be shocking that accu- increasing in many countries, indicating that new reg- rate biomarkers for treatment response, or in fact cure, imens will need to be increasingly incorporated into are essentially nonexistent, as is the ability to predict TB treatment. relapses after treatment (Walzl and others 2008). Within clinical trials, cure is defined as no relapse after Second-Line Treatment of MDR TB one year after completing therapy. In LMICs the gen- The treatment of drug-resistant TB is evolving, and eral criterion of cure for individual patients is two recommendations are changing rapidly. Four factors negative sputum smears a month apart (WHO 2014b). make it difficult to arrive at clear, generalizable recom- Yet sputum smears are not sufficiently sensitive or pre- mendations. First, individual strains vary in their sus- cise to be certain that there is true sterilization of the ceptibility, and customized regimens might be more infection. Bacterial culture, though more sensitive, is appropriate, when possible. Second, testing susceptibil- also more time-consuming and less frequently used in ity to pyrazinamide and second- and third-line agents is resource-poor countries (Phillips and others 2016). neither widely available nor consistently reliable. Third, There are no microbiological or molecular biomarkers many agents have limited availability due to their cost or to establish whether an individual’s infection has been limited production. Finally, few comparative studies are sterilized by treatment. Recurrence can be due either available to provide data on which to make optimal to reactivation of a previously treated strain or to rein- treatment decisions. fection with a new strain. Reinfection in previously While drug-resistant disease is curable, the cure treated patients may be as common as relapse and can rate in several studies is lower than for drug-sensitive be distinguished from relapse by comparing mycobac- disease. In some studies of MDR TB, only 54 to 70 terial DNA sequences from both the original isolate percent of patients achieve treatment completion or and the recurrence (Marx and others 2014). cure (Ahuja and others 2012; Bassili and others 2013; Tuberculosis 251 Table 11.5 Tuberculosis Treatment Regimens Currently Recommended by the WHO Type of case and phase Length of regimen (months) Drug used New tuberculosis case Intensive phase 2 Rifampicin, isoniazid, pyrazinamide, ethambutol Continuation phase 4 Rifampicin, isoniazid (low risk of isoniazid resistance) or rifampicin, isoniazid, ethambutol (high risk a of isoniazid resistance) Previously treated tuberculosis case (relapse or default) b Intensive phase 2 Rifampicin, isoniazid, pyrazinamide, ethambutol, streptomycin Continuation phase 1 Rifampicin, isoniazid, pyrazinamide, ethambutol 5 Rifampicin, isoniazid, ethambutol Previously treated tuberculosis case (treatment failure) c Intensive phase 8 MDR TB regimen (see below) Continuation phase 12 MDR TB regimen (see below) MDR TB cases 2010 guideline 20 Kanamycin (or amikacin), moxifloxacind (ethionamide, cycloserine (or terizidone), pyrazinamide Intensive phase 8 Bedaquiline or delamanid, where sensitivity following the initial regimen cannot be assured (up to six months) Continuation phase 12 Moxifloxacin,d ethionamide, cycloserine, pyrazinamide 2016 short regimen (conditional recommendation) Intensive phase 4–6 Kanamycin, moxifloxacin,d prothionamide, clofazimine, pyrazinamide, high-dose isoniazid, ethambutol Continuation phase 5 Moxifloxacin,d clofazimine, ethambutol, and pyrazinamide Note: MDR TB = multidrug-resistant tuberculosis defined as resistance to isoniazid and rifampicin; WHO = World Health Organization. a. Using local epidemiological data. b. Low risk of MDR TB using local epidemiological data. The WHO recommends treatment guided by drug-susceptibility testing, especially rapid molecular tests, and suggests that standard first-line treatment be used if there is no evidence of drug resistance to isoniazid and rifampicin. c. Defined as smear positive after five months of first-line treatment, relapse or default after second or subsequent course of treatment, or active tuberculosis after contact with an MDR TB case. d. Or high-dose levofloxacin or gatifloxacin. James and others 2011; Loveday and others 2012; resource constraints and patients with advanced dis- Nathanson and others 2010). Treatment requires ease (Meressa and others 2015; WHO 2016a). new drugs, with regimens containing anywhere from Key strategies that have contributed to successful three to seven drugs that have not been previously treatment include intensive management of adverse employed (Mitnick and others 2008). In general, these effects, nutritional supplementation, adherence inter- second- and third-line agents are less potent and must ventions, and collaboration between the public health be administered for a more extended period of time, service and nongovernmental organizations (NGOs). ranging from 9 to 24 months. They are also more dif- These approaches should be routinely incorporated into ficult to administer, as most regimens contain agents programs wherever possible. such as kanamycin and amikacin that must be admin- Generally, MDR TB has substantial human, eco- istered by injection. These drugs are far more toxic nomic, and social consequences (Rouzier and others than first-line agents, causing a range of drug-specific 2010). The cost of treating MDR TB using conventional side effects. Nevertheless, it has been possible to regimens ranges from US$2,500 to US$10,000, com- achieve MDR TB cure rates of 60–80 percent irrespec- pared with US$100–US$1,000 for drug-susceptible TB tive of HIV/AIDS status in settings with severe cases (Floyd and others 2013), placing substantial costs 252 Major Infectious Diseases on high-burden countries. For example, in South Africa, Treatment should be individualized, with expert review. although MDR TB and XDR TB represent less than 3 Contraceptive advice during MDR TB treatment is percent of all TB cases detected, they consume an esti- essential. mated 35 percent of the national health budget allocated Glucocorticoids may limit the inflammatory damage to tuberculosis control (Pooran and others 2013). In associated with tuberculosis (Critchley and others some countries, the costs to treat MDR TB are estimated 2013). Evidence supports the use of glucocorticoids to exceed the total budget for TB control. for tuberculous meningitis (Prasad and Singh 2008). In May 2016, the WHO issued a conditional recom- Additionally, surgery may be necessary to improve the mendation on use of the shorter MDR TB regimen, chance of cure by removing localized disease (Marrone which would shorten the duration of treatment (to 9–12 and others 2013) and to decompress vital structures months), increase adherence and retention in care, and that are compromised by the tuberculous cavities. lower costs (about US$1,000 in drug costs per patient) A particularly devastating form of TB, tuberculous (WHO 2016a). Routine analysis of mutations conferring meningitis, has a rapid onset and is frequently fatal. resistance to isoniazid may further inform the choice of Current treatments are less effective for TB meningitis, MDR TB treatment: isolates with mutations in the pro- and higher doses of drugs may be needed to reach ther- moter region of the inhA gene are susceptible to high- apeutic levels in the central nervous system (Donald dose isoniazid but resistant to ethionamide, while those 2016). with katG mutations are resistant to high-dose isoniazid but sensitive to ethionamide (WHO 2016a). Ongoing clinical studies are beginning to form the Drug Toxicities and Interactions evidential basis for the WHO guidelines given in Prompt detection and effective management of table 11.5, and are discussed in more detail in the section adverse drug effects is essential to the integrity of a titled “Research and Development.” In this rapidly treatment program. TB antibiotics, like other medica- changing area, encouraging data suggest that higher cure tions, may interfere with drug metabolism and excre- rates are possible, perhaps with shorter courses using tion. Rifampicin potently induces expression of newer agents (see the review by Zumla, Nahid, and Cole hepatic cytochrome P450 enzymes (McIlleron and 2013). Bedaquiline and delamanid, two newly approved others 2007), substantially reducing levels of several drugs, both lead to more rapid clearance of organisms clinically important drugs including HIV/AIDS pro- and higher cure rates for MDR TB when administered tease inhibitors, warfarin, phenytoin, carbamazepine, with an optimized regimen. Similarly, the oxazolidinone and estrogen-containing contraceptives. Antiretroviral antibiotic linezolid, which is used largely to treat Gram- drugs nevirapine and efavirenz interact with rifampi- positive infections, accelerates clearance and increases cin; however, only the nevirapine interaction is clini- cure. Clofazimine, a riminophenazine dye used to treat cally significant, and current recommendations are to leprosy, is now recommended for the shortened MDR use efavirenz with rifampicin. The newer drugs, TB regimen. These new treatments may cause significant bedaquiline and delamanid, which in small studies side effects. Clofazamine may cause skin discoloration. seem to be effective against MDR- or XDR-TB, For unclear reasons, bedaquiline therapy has been asso- increase the QT interval with a risk of arythmia, and ciated with a higher death rate, while linezolid produces linezolid has serious effects on bone marrow and neu- a range of dose-limiting toxicities, including neuropathy rologic function. Clinically significant interactions and myelosuppression. should be checked regularly online.1 Treatment in Specific Situations Antiretroviral Therapy Regimens for treating tuberculosis in children are iden- In 2014, there were 10.4 million new cases of TB, of tical to those for adults. Correct dosing by weight is which 1.2 million (11 percent) were among people living essential, and the most appropriate formulation of com- with HIV/AIDS. Over the past 30 years, antiretroviral bination medications receives ongoing advocacy (WHO therapy for HIV/AIDS infection has improved to the 2013c). point where effective therapy is widely available in Tuberculosis in pregnancy can be treated with iso- LMICs, with strikingly improved mortality in patients niazid, rifampicin, pyrazinamide, and ethambutol. co-infected with HIV/AIDS and tuberculosis (Khan, Streptomycin, amikacin, and kanamycin may cause Minion, and others 2012). fetal ototoxicity and should not be used if possible Early initiation of ART reduces mortality risk in (Donald 2016). The safety of other drugs used to treat HIV-positive patients co-infected with tuberculosis and MDR TB has not been well studied in pregnancy. is therefore recommended, irrespective of CD4 count Tuberculosis 253 (WHO 2013a). To reduce mortality risk, it should be and others 2012; Smart 2010). Terminally ill patients commenced within two weeks of TB treatment.2 The may decide to improve their quality of life by discontin- goal of ART is to achieve long-term viral load suppres- uing tuberculosis treatment. Physical discomfort, psy- sion assessed with regular viral load measurements. chological distress, and unresolved end-of-life social Cotrimoxazole prophylaxis for Pneumocystis pneumo- issues can all potentially be addressed by trained com- nia, toxoplasmosis, bacterial sepsis, and malaria reduces munity health workers (CHWs) once the need has been mortality in patients co-infected with TB and HIV/ identified. These structures need ongoing local support AIDS and should be given until the CD4 count recovers and advocacy (Harding and others 2012). to above 200 cells per microliter after at least six months of ART. One significant adverse effect of combined treat- The Cascade of Care and Completion of Treatment ment is the development of IRIS, which is character- The DOTS strategy to control tuberculosis promotes ized by persisting or recurring fevers and a worsening standardized treatment, with supervision and patient of the focal tuberculous process (in profoundly support, which may include direct observation of ther- immune-suppressed patients with CD4 count below apy, where a health care worker personally observes the 50 cells per microliter) starting combination ART patient taking the medication (WHO 2013b). The scien- shortly after commencing tuberculosis treatment tific evidence on the effectiveness of DOT compared to (Meintjes and others 2008; Meintjes and others 2010). self-administered therapy is mixed. Despite the galva- IRIS can usually be controlled with steroids and non- nizing impact of the DOTS strategy in mobilizing sup- steroid anti-inflammatory agents. port and treatment activities, a systematic comparison of the effectiveness of DOT relative to self-medication in 11 random control trials failed to establish its unique Antidiabetic Treatment effectiveness in ensuring either compliance or cure Diabetes is a significant risk factor for tuberculosis (Karumbi and Garner 2015). but has received less attention than HIV/AIDS in LMICs. DOTS has been associated with reduced prevalence This will likely change given the increasing life expec- of drug resistance in the United States (Moonan and tancy and prevalence of obesity and type 2 diabetes others 2011; Pasipanodya and Gumbo 2013); other globally. All patients with tuberculosis should be screened HICs; and many LMICs, such as Cambodia, China, and for diabetes (Faurholt-Jepsen and others 2012; WHO Ethiopia (WHO 2014b, 2015b). However, in highly 2011a). Diabetic patients should have their glucose con- endemic countries, especially those burdened with HIV/ trol assessed regularly while on TB treatment as part of AIDS, even where adequate diagnosis and effective treat- integrated clinical care, and treatment should be opti- ment are provided, the strategy has not dropped inci- mized with oral antidiabetic agents and insulins. dence or transmission as much as needed. As discussed in the section on research and development, additional Surgery strategies will be needed where the forces of infection, With treatment outcomes for multidrug-resistant environment, and HIV/AIDS are driving the infection tuberculosis patients achieving only about 50 percent rate, despite effective treatment of incident cases success, surgery, once a major tool in the pre-antibiotic (Middelkoop and others 2015). era, has reemerged as an adjuvant therapeutic strategy. Some countries have experimented with involving A systematic review and meta-analysis to assess the community members to make treatment supervision evidence for the effect of surgery as an adjunct to che- more acceptable to individual patients (Datiko and motherapy found that there was little substantial data Lindtjørn 2009), but the operational issues are substan- on which to base recommendations, but there appeared tial, and a meta-analysis after DOTS implementation to be some enhancement of successful outcomes from that included community members in China found that surgery on adults treated for MDR TB (Harris and 52 percent of patients still took self-administered ther- others 2016). apy (Hou and others 2012). Nonadherent patients need to be identified early and offered practical interventions Palliative Care to assist their return into care (Toczek and others 2013; Tuberculosis remains a leading cause of death in LMICs. Yin and others 2012), including hospitalization for Suffering and the process of dying are important clinical supervised treatment and physical rehabilitation. Clinic consequences of advanced tuberculosis that should be staff who know the patient and community are in the detected and communicated by an experienced clinician best position to decide which patients need the intense who can initiate effective palliative treatment (Connor adherence support implicit in DOTS. 254 Major Infectious Diseases It is important to emphasize that early case detec- surveillance of access to immunization; and it is tion, whether by passive or active case finding, is a inexpensive. necessary but not sufficient condition for effective Considerable evidence indicates that giving BCG to control of TB. A recent analysis of the cascade of care young children is effective at preventing tuberculous of TB in India reveals the challenges of ensuring treat- meningitis and disseminated (miliary) TB (Mangtani ment completion (Subbaraman and others 2016). In and others 2014; Rodrigues, Diwan, and Wheeler 1993). this important study of about 2 million cases of con- Random control trials and case control studies have ventional TB evaluated through the Revised National shown consistently high protective efficacy of BCG Tuberculosis Control Program, the authors created a against serious childhood forms of disease (73 percent), framework and followed the cascade of care from the meningitis, and miliary TB (77 percent). The most com- number of prevalent cases to those reaching TB diag- plete analysis of the effect of BCG vaccine suggests that nostic centers, those diagnosed with TB, those regis- giving BCG to children born in 2002 prevented about tered for treatment, those who completed treatment, 29,000 cases of childhood meningitis and 11,500 cases of and finally those with recurrence-free survival at one miliary TB during the first five years of life or one case year. The results indicated for conventional TB that of meningitis for every 3,400 vaccinated children and 45 percent completed treatment and 39 percent were one case of miliary TB for every 9,300 vaccinated chil- disease free after one year. Of patients diagnosed with dren (Trunz, Fine, and Dye 2006). A recent report indi- MDR TB, only 14 percent completed treatment and cates that deferring BCG immunization to six weeks 11 percent remained disease free at one year. These after birth generates stronger and longer-lasting specific striking results clearly indicate the critical need for Th1 cellular immune responses (Kagina and others support of treatment to enable greater treatment com- 2009; Lutwama and others 2014). In some countries, pletion in India and most LMICs. children were repeatedly vaccinated over time, and there is some evidence from Taiwan, China, that multiple vac- cinations may increase protection (Chan and others TB PREVENTION 2013). A worrisome drawback, however, is the incidence of disseminated BCG infection in HIV-positive children There are three obvious strategies for preventing TB: (Hesseling and others 2007). vaccination, infection control, and chemoprophylaxis or Successful vaccines ideally prevent both infection isoniazid preventive therapy (IPT). Arguably, the most and disease among persons exposed to the pathogen. useful but perhaps least appreciated preventive interven- While it is generally believed that BCG protects against tion is simply the early diagnosis and rapid initiation of disease rather than infection, recent findings, using effective treatment of TB cases, thus reducing the infec- Interferon-Gamma Release Assays (IGRAs) that can dis- tious burden and reducing transmission. TB is unusual tinguish Mtb infection from BCG vaccination, suggest among infectious diseases, in that appropriate (and that BCG may protect to varying degrees against infec- effective) treatment of the individual patient may be the tion as well (Eisenhut and others 2009; Mangtani and most effective public health intervention to protect the others 2014; Soysal and others 2005). population. The cost-effectiveness of BCG was estimated in 2006 to be between US$40 and US$170 per disability-adjusted life year (DALY), US$8,000 and US$11,000 per life saved, Vaccination: Natural and Acquired Immunity or US$5,000 and US$8,000 per case averted (Dye 2006), BCG Vaccine making it a very cost-effective intervention. The most widely used vaccine in the world is BCG, Despite positive evidence regarding the impact of which is given to about 100 million children annually. BCG on children, BCG remains the most controversial Isolated in 1908, following attenuation through 431 pas- of all currently used vaccines, because its protective sages of a virulent M. bovis isolated from a human TB efficacy has varied widely in different parts of the case, BCG was found to be protective to some extent in world, from 77 percent protection in adolescents in the multiple animal models of TB. In its first human trial in United Kingdom (Sutherland and Springett 1987) to 1921, it was found to protect a child heavily exposed in a 0 percent protection in all age groups in South India household at high risk. (Bloom 1994; Mangtani and others 2014). Because BCG has several advantages: it can be given at birth or young children represent only a minor contributor to at any time after birth; a single inoculation produces TB transmission, BCG immunization of infants has long-lasting skin test positivity to tuberculin; it is rela- only a relatively small impact on transmission within tively stable; it produces a scar useful for epidemiological populations (Knight and others 2014). This finding is Tuberculosis 255 borne out by outcomes in parts of Europe and North there is a remarkable amount of epidemiologic evidence America that did not use BCG, where TB declined at that, in fact, Mtb latent infection does indeed provide rates that were not measurably different from those in significant protection against reinfection by engendering regions that used the vaccine (Styblo 1991). In a recent protective immune responses that likely persist (reviewed analysis using South African data, Dye (2013) found by Andrews and others 2012). Early experiments of that BCG vaccination would reduce TB in HIV-negative Heimbeck (1938) in Norwegian nurses and Sutherland, individuals by about 17 percent, to which would be Svandova, and Radhakrishna (1982) found that, among added the value of preventing transmission to HIV- healthy young individuals, being TST-positive provided positive individuals. He estimated that revaccination up to 80 percent protection against reinfection (ranging with BCG would be highly cost-effective at all combi- from 45 to 81 percent in multiple studies) compared to nations of cost (US$1–US$10 per child) and efficacy being TST-negative. In a more recent study from South (10–80 percent). Africa where it was possible to measure infections in the BCG immunization has also been shown to have apparently uninfected group and to observe cases of extremely variable protective efficacy against adult TB in tuberculosis directly, the estimated immunological pro- randomized trials and observational studies (Bloom and tection conferred by latent TB infection was 79 percent Fine 1994; Fine 1995; Mangtani and others 2014). (Andrews and others 2012). These results encourage the Explanations for the variation remain unclear. view that new vaccines with better efficacy than BCG Suggestions include the fact that BCG lacks more than could provide relatively high levels of protection, if the 100 genes of Mtb, including some putative protective protective immune responses generated are sustained antigens; the genetic make-up of different human pop- over time, as is the case of latent TB. ulations; the variable persistence of different BCG strains or preparations; and the interference by atypical Innate Immune Responses mycobacteria in the environment. It was shown many The body’s immune response is critical in protecting years ago that guinea pigs immunized with different against infection and disease, through both innate species of environmental atypical mycobacteria showed and acquired immunity. This is perhaps best exempli- different degrees of protection against Mtb. Some, such fied by the fact that, although only 10 percent of per- as M. kansasii, were as effective as BCG in animals sons infected with TB develop disease in their lifetime, (Palmer and Long 1966). It is often forgotten that M. immune-compromised individuals have a risk of microti, a murine mycobacterium, was as effective as almost 10 percent per year (Selwyn and others 1989). BCG in the British Medical Research Council human Immunodeficient individuals, such as persons infected vaccine trials (Bloom and Fine 1994; Hart and Sutherland with HIV/AIDS (Gandhi and others 2006) or receiv- 1977). This suggests that, if exposure to environmental ing anti-TNF immunotherapy for autoimmune dis- mycobacteria in a population provides some degree of eases (Wolfe and others 2004), have a markedly protection, the effects of BCG observed in that popula- increased incidence of TB. In immune-compromised tion will be comparably less than in a naïve population. HIV-positive patients in KwaZulu-Natal, the mean This could explain the large differences in BCG efficacy time from diagnosis to death from XDR or drug- in populations living in different geographic locations resistant TB was an astonishing 16 days (Gandhi and (Weir and others 2006), where children in tropical lati- others 2006). And the increased prevalence of drug- tudes show less protection (Mangtani and others 2014). resistant TB in immune-compromised individuals is For example, in the South India trial area, two-thirds of consistent with the view that the effectiveness of anti- the individuals were positive to an M. avium purified biotics depends to some extent on the antimicrobial protein derivative skin test by age 9 years and 97 percent immune response. It is intriguing that a population of were positive by ages 15–19 years (Tuberculosis healthy contacts exists in high-burden countries, who Prevention Trial 1979). Understanding the degree of almost certainly have been exposed repeatedly to environmental exposure to nontuberculous mycobacte- infection yet remain TST-negative, IGRA-negative, ria will be important in planning and evaluating any and apparently healthy. This suggests that mecha- new vaccines against TB. nisms of innate immunity may have the ability to kill the relatively small numbers of infecting tubercle A Rationale for Vaccines: Latent Infection bacilli early after respiratory infection even before Can Prevent Reinfection they can grow to numbers able to sensitize and It is widely believed that a protective vaccine is unlikely expand T-cells able to respond to TB antigens. These to be developed against TB because natural infection mechanisms are currently not understood, and greater with Mtb is ineffective at preventing reinfection. However, research is needed. 256 Major Infectious Diseases (Cavalcante and others 2010). Yet IPT is not widely Preventive Therapy used. Of the highest-burden countries, only Brazil There are two approaches to preventive therapy. For and South Africa have policies to scale up the use of HIV-positive individuals at high risk for many oppor- IPT. Even with the resources available in the United tunistic infections, cotrimoxazole is recommended States, the implementation of contact tracing and IPT routinely, and in high-burden countries between 50 has fallen short of recommendations (Lee and others and 87 percent of HIV-positive patients are receiving 2006). In trials of IPT in high-burden countries, pro- this preventive therapy (WHO 2015b). tection of TST-positive adults infected with HIV/ The major approach to prevent development of TB AIDS averaged about 60 percent, but the effects were in persons at high risk, particularly household contacts lost soon after the IPT treatment ended, and there was and HIV-positive individuals, is to screen them with a little or no impact on mortality (Churchyard and TST and, ideally, to offer persons found to be HIV- others 2012; Samandari and others 2011). By contrast, positive chemoprophylaxis, most commonly IPT, for a IPT was shown to reduce both TB incidence and mor- latent infection (Rangaka and others 2015). In perhaps tality among HIV-positive children (Zar and others the most dramatic studies, a community-based trial of 2007). A randomized controlled trial, the Temprano IPT among BCG-unvaccinated Alaska Eskimos, a com- study, compared early versus later treatment with munity with a high risk of infection, produced a ART and early versus later treatment with IPT and 60 percent decline in TB incidence that lasted more various combinations in 2,050 HIV-positive individ- than two decades in treated households (Comstock, uals with high CD4+ counts in Côte d’Ivoire. The Baum, and Snider 1979). Overall in an analysis of ran- results showed that six months of IPT resulted in a 44 domized controlled trials, preventive therapy has clear percent lower risk of severe HIV/AIDS-related illness benefits. In 2011, the WHO began recommending that and a 35 percent lower risk of death from any cause HIV-positive individuals free of symptoms suggestive of than the risks with deferred initiation of ART and no tuberculosis receive treatment with IPT for at least six IPT and that the combination reduced TB by 73 months (WHO 2011a). The risk of clinically active TB percent (Temprano ARNS Study Group 2015). Use of disease is reduced 60 percent in immunocompetent, IPT for six months reduced the incidence of TB in HIV-negative individuals (Smieja and others 2000) and Brazil not only for the duration of treatment but over 32–62 percent in HIV-positive adults who are treated a seven-year follow-up (Golub and others 2015). The with preventive therapy lasting 3 to 12 months (Akolo strong inference from this work is that the combina- and others 2010). Since these guidelines came into tion of early initiation of both ART and IPT in HIV- force, the number of HIV-positive people receiving IPT positive individuals, now adopted by the WHO, has increased sharply, rising to approximately 933,000 should become the norm in HIV/AIDS and TB in 2014 (WHO 2015b). The high risk of TB among control. persons co-infected with M. tuberculosis and HIV/AIDS Real challenges are associated with the use of IPT: motivates those encouraging wider use of preventive active disease must be excluded, where practical by therapy, especially in Africa (Stop TB Partnership 2011), radiography, before isoniazid is taken alone, and adher- but questions have been raised about the methods of ence to six or more months of daily treatment tends to screening to ensure that the persons most likely to ben- be poor among healthy people. Even in the United efit receive this treatment (Lawn and others 2012) and States, fewer than 50 percent of individuals who initi- that the persons with subclinical TB are not inadver- ated IPT completed six months of treatment (Hirsch- tently treated with isoniazid monotherapy that could Moverman and others 2015). Adverse effects include a lead to resistance. Studies among child contacts of risk of hepatitis, especially in persons co-infected with active cases have demonstrated that giving isoniazid HIV/AIDS (Ayles and Muyoyeta 2006), if IPT is admin- daily for 12 months provides 30–60 percent protection istered for long periods of time. In the United States, against active TB (Ayieko and others 2014). Recent adverse effects are on the order of 1 percent but rise to work suggests that isoniazid can be continued for lon- 4 percent if IPT is given with rifapentine (Getahun and ger than six months in HIV-positive adults with mini- others 2015). A recent cluster randomized trial of mass mal adverse effects and longer protection (den Boon screening and IPT was carried out for tuberculosis con- and others 2016). trol among gold miners in multiple mines in Thibela in At a population level, a randomized trial in high- South Africa, a community known to be at high risk for incidence urban communities in Brazil found that TB TB. Miners were given IPT for 9 months, and the effect incidence was 15 percent lower in intervention than on prevention of disease was followed for 12 months. in nonintervention communities after five years Despite the positive effect of isoniazid in preventing Tuberculosis 257 tuberculosis during the period of treatment, IPT had treatment are required to reduce the risk of transmis- no significant sustained impact on TB control in South sion substantially, regardless of sputum smear status African gold mines (Churchyard and others 2014). (Rouillon, Perdrizet, and Parrot 1976). For drug- Most discouraging, the results of a systematic review resistant TB, however, the two-week rule appeared to and meta-analysis of the cascade of care in latent TB have failed as a policy during the outbreaks in New York indicated that completion of preventive treatment was City and Miami in the 1980s and 1990s, when patients only 19 percent (Alsdurf and others 2016). with unsuspected drug resistance on conventional four- It remains unclear, other than in the case of HIV- drug therapy transmitted their infection after isolation positive individuals and child household contacts, how ended (Coronado and others 1993). Therefore, current feasible and cost-effective IPT scale-up would be in guidelines generally recommend isolating MDR TB high-burden countries. It has been a challenge for health patients until smear or culture conversion. systems in LMICs, which are finding it difficult to pro- The rate at which treatment renders tuberculosis vide treatment for diagnosed TB patients and to main- cases noninfectious was recently reexamined, employ- tain IPT for healthy contacts. A recent approach to ing the classic model of transmission from humans to shortening preventive therapy derives from studies of a guinea pigs, wherein TB transmission was established combination of long-acting rifapentine plus isoniazid, by passing exhaust air from the ward past a panel of which reduced the time of treatment from nine to three guinea pigs highly susceptible to TB (Dharmadhikari months and was better tolerated, if more expensive and others 2014). The study suggested that, like drug- (Sterling and others 2011). susceptible TB, MDR TB transmission also responds The critical question of the best preventive therapy rapidly to effective treatment, well before sputum smear with which to combat MDR TB remains unsettled. The or culture conversion (Dharmadhikari and others significant adverse effects of MDR TB regimens are a 2014). In a series of five exposure studies where mostly serious trade-off against prevention of the 10 percent of smear-positive, coughing patients with confirmed MDR cases likely to result from preventive treatment, and no TB were admitted and promptly started on therapy, optimal regimen using newer drugs has been estab- transmission to guinea pigs appears to have occurred lished (Moore 2016). The alternative is registering all predominantly from patients with unsuspected XDR such contacts, monitoring them carefully, and institut- TB who were not responsive to effective treatment. ing treatment at the earliest sign of disease. It may not be altogether fanciful to imagine, with advances in research, that more effective drug regimens Infection Control of TB Transmission in for latent infection could have a profound effect on reduc- Congregate Settings ing the global burden of TB. It is unknown what propor- Transmission and reinfection, especially of drug-resistant tion of the one-third of people on the planet exhibiting strains, is a key driver of the global TB epidemic (Wood, positive TST retain viable tubercle bacilli capable of reac- Lawn, Caldwell, and others 2011; Wood, Lawn, Johnstone- tivating and transmitting disease. But, if a practical drug Robertson, and others 2011). The benefit of isoniazid regimen could sterilize the infection in all of these indi- prophylaxis in high-risk HIV-positive populations, for viduals, this enormous reservoir of the pathogen could be example, has been rapidly reversed by ongoing transmis- eliminated in a short period of time. Rather than passively sion and reinfection soon after isoniazid is stopped detecting patients with disease, screening populations for (Samandari and others 2011). Likewise, the challenge in persons who were infected as determined by tuberculin high-transmission settings is to provide greater protec- positivity and applying the hoped-for effective mycobac- tion against reinfection than is currently provided by tericidal regimen, the great burden of latent TB could BCG immunization at birth as well as subsequent natural conceivably be reduced or eliminated. This approach exposure to Mtb and environmental mycobacteria should be considered in future research. (Tameris and others 2013). Transmission control was not specifically mentioned in the original Global Plan to Stop TB 2006–15, but the Impact of Effective Treatment on Transmission mostly hospital-based outbreak of XDR TB in 2006 dra- Of the interventions available to control transmission, it matically called attention to the problem (Gandhi and has long been taught that effective treatment ranks others 2006). Since then, control efforts have centered on highest. Treatment can be applied only to known or health care facilities, although it is widely understood suspected cases, and, to be effective, requires knowledge that transmission also occurs in homes, schools, of drug resistance. It has traditionally been thought that, churches, shelters, refugee camps, and correctional facil- for drug-susceptible TB, at least two weeks of effective ities, among other congregate settings (WHO 2009b). 258 Major Infectious Diseases Still, because they specifically bring together infectious highly effective, and Tomsk is among the few high- and vulnerable persons, health care facilities dominate burden places in the world where MDR TB rates may be the list of environments that amplify transmission at the declining (WHO 2010). population level. Principles of TB Transmission Control Hospitals as Epicenters of Transmission Since the 1985–92 resurgence of TB in the United States One key epicenter of transmission is hospitals where TB and several European settings, where institutional trans- patients reside. Because hospital exposure is better doc- mission played an important role, a three-tiered hierar- umented than many other congregate interactions, there chical approach has been adopted, based on a paradigm is a likely bias toward indicting hospitals, but it is also used in industry: administrative controls, engineering likely that hospital transmission is underestimated, for (or environmental) controls, and personal protection example, by not counting infecting strains for which (respirators). Administrative controls entail the rapid DNA fingerprints are not available. diagnosis of symptomatic, potentially infectious cases The significant reduction of the TB epidemic in New and drug resistance and the prompt initiation of effec- York City from 3,800 cases in the 1980s to 577 in 2015 tive therapy. This has recently been promoted under the indicates that multifaceted efforts are needed in large acronym, FAST (Find cases Actively by cough surveil- urban multicultural environments with large numbers lance, Separate temporarily, and Treat effectively), as a of visitors, migrants, and homeless people and increas- way to communicate the key components and facilitate ing rates of HIV/AIDS.3 Clearly DOT was helpful in adoption. Environmental controls have focused on nat- ensuring compliance in a portion of TB patients, but ural and mechanical ventilation and on the evolving perhaps more significant was the major effort to insti- technology of sustainable ultraviolet germicidal (UVGI) tute infection control in hospitals, prisons, shelters, and air disinfection. Personal respiratory protection is the congregate housing facilities (Frieden and others 1995). last tier of protection, assuming incomplete protection Using network analysis, Gandhi and others (2013) from administrative and engineering controls. Ironically, concluded that most strain-specific XDR TB although the last tier of protection, respirators are often transmission in KwaZulu-Natal occurred in hospitals the only protection available to health care workers, can- due to prolonged stay, congregate settings, and delayed not be worn continuously, and are unlikely to be worn recognition of drug resistance. Transmission patterns when treating a patient with unsuspected TB. are similar as far away as Tomsk Oblast, Siberia. A retro- spective study of the causes of drug resistance in the Measures of the Efficacy of TB Transmission Control Tomsk Oblast showed a greater than sixfold higher risk Measuring the efficacy of transmission control interven- among treatment-adherent patients hospitalized for tions has been elusive. Among process indicators are drug-susceptible TB than among patients not hospital- questions regarding whether windows are open or respi- ized (Gelmanova and others 2007). Anyone familiar rators are available, although these factors may be tied with treatment practices common to Eastern Europe too loosely to exposure to be useful. However, to the will understand why this might occur. Although Tomsk extent that undiagnosed TB patients and undiagnosed predominantly uses ambulatory treatment for new TB drug resistance are key exposure factors, process indica- cases, hospitalized patients are admitted to poorly venti- tors tied to unprotected exposure time can be measured lated, multibed rooms, tightly sealed against the cold. and reported. Institutions can document, for example, Drug susceptibility is normally only tested when patients the percentage of admissions that are screened for cough fail to respond to first-line treatment, usually following and had sputum sent to a lab; the turnaround time from months of ineffective treatment. Drug susceptibility submission until results are obtained; and the time from testing by conventional methods requires additional admission until the onset of effective treatment based on months. During this prolonged period of ineffective drug-sensitivity testing. Such measures should become treatment, there is ample opportunity for transmission routine in hospitals with access to rapid diagnostic tests. and reinfection. This scenario is not unique to Tomsk, as Few studies have been conducted not only of the effi- delays in drug-susceptibility testing occur in most TB cacy of TB infection control methods, but also of their programs where laboratory services are inadequate. cost. Apart from the great difficulty of measuring the Treatment failure is the usual indication for drug- efficacy of interventions to prevent transmission, isolat- susceptibility testing, and molecular methods are only ing the costs of infection control activities can be chal- slowly reducing the time required for results on first-line lenging, as many infection control interventions are and, much less often, second-line drugs. However, integral to hospital functions more generally. Assuming despite hospital transmission, ambulatory treatment is the presence of unsuspected, untreated patients in the Tuberculosis 259 hospital, ventilation is a key intervention, and natural regulates this small industry, and few experts are qualified ventilation ranks high among recommendations in suit- to plan installations. Still, as back-up technology for able climates. The added cost of designing and con- natural ventilation, they are a logical choice. Low-velocity structing a naturally ventilated patient waiting area is ceiling fans are recommended to assure essential room air difficult to separate from the routine capital costs of mixing. With the development of LED (light-emitting hospitals. Some insights can be gleaned from unpub- diode) UVGI, the prospect of solar-powered systems with lished data from a high-risk setting in Vladimir, Russia, a battery back-up may make upper-room germicidal UVGI training center of excellence in TB control (box 11.4). more sustainable in the near future. Cost-Effective Air Disinfection Masks and Respirators Natural ventilation alone does not provide adequate Assuming incomplete efficacy of both source control ventilation for airborne infection control in many set- and engineering or environmental control strategies, the tings. However, mechanical ventilation systems are last-tier intervention is respiratory protection—that is, often prohibitively expensive and often fail due to lack use of a device designed to exclude infectious droplet of maintenance. Room air cleaners (with filters, UVGI, nuclei from inhaled air. or both) are often sold to hospital administrators as a Masks and respirators are easily confused. Surgical simple, inexpensive fix, but they rarely move enough masks are designed to protect the environment by air to achieve the 12 or more equivalent air changes per blocking the aerosolization of some portion of exhaled hour recommended to control airborne infections. respiratory droplets and droplet nuclei, but they do Reentrapment and recirculation of the same air through not adequately protect the wearer and have a limited the device (short-circuiting) also lead to low rates of role in TB transmission control when worn short-term effective clean air delivery. by patients. MDR TB patients wearing masks were As noted in the Vladimir study, upper-room germi- 53 percent less likely to infect guinea pigs breathing cidal UVGI (with room air-mixing fans) is among the exhaust air from the ward (Dharmadhikari and others most effective and least expensive ways to achieve 2012). Recently, using the same transmission model in high-volume air disinfection. Hospital studies have South Africa, Mphaphlele and others (2015) tested the shown 70–80 percent efficacy. But like mechanical venti- efficacy of several control interventions in preventing lation and room air cleaners, caution is required. Although transmission from patients in hospital rooms to they are under development, international standards and guinea pigs. The study confirmed the previous report guidelines for safe and effective application and mainte- and showed 70–80 percent efficacy of upper-room nance are not widely available. No agency currently UVGI air disinfection. Box 11.4 Real Costs of Infection Control in Vladimir Oblast TB Dispensary, Russia Costs for high-level infection control are difficult for respirator fit testing; and US$3,000 for health to obtain. The TB Dispensary, with assistance from care worker screening. For the multistory hos- the Centers for Disease Control and Prevention, pital, with floor area of 17,000 cubic meters, the has over the past decade painstakingly introduced capital cost of a new, high-capacity ventilation and studied the impact of a variety of conven- system with negative-pressure isolation rooms tional and novel TB infection control interven- was US$345,000, and the cost of maintenance tions in Vladimir Oblast, an area with high rates was US$4,425 per year. Of three ventilation sys- of TB and drug resistance. For the entire Vladimir tems studied, the upper-room UVGI system was region (population 1.5 million), accurate esti- the least expensive intervention, at US$14 per mates of annual cost are US$350 for health care equivalent-room air change, more than nine times worker training; US$12,000 for ventilation system more cost-effective than expensive mechanical maintenance; US$10,000 for respirators; US$300 systems per equivalent-room air change. 260 Major Infectious Diseases In contrast, respirators are designed to protect the Earlier Diagnosis: Toward Active Case Finding wearer. Properly fitted, certified N-95 (or equivalent) Limitations of Passive Case Finding respirators can be 85–90 percent protective. However, Even with modern technology, effective case detection in as an intervention, respiratory protection often fails resource-poor communities with weak health systems either because the face seal leaks due to improper fit or has been difficult to introduce (Kranzer and others 2010; adjustment or, more important, because the masks are Kranzer and others 2013). The principal paradigm for not worn consistently. The cost of respirator programs diagnosing cases of TB is passive case finding, which is easily assessed, but their effectiveness is not. depends on the TB-infected individual seeking medical care. But passive case finding faces many challenges. TB Lessons for Household Transmission is most prevalent in marginalized communities that are Since TB is transmitted largely by aerosol droplets, less visible to conventional health systems. Patients are transmission is affected by the built household environ- typically poor, from disadvantaged groups, prone to ment. One study from South Africa found evidence that other diseases such as HIV/AIDS and diabetes that transmission of infection was greater, as determined by increase their vulnerability to TB, and often migrants. DNA fingerprinting of the strains, in modern-built Even when symptoms are present, in many countries, up brick housing with windows than in older shacks to a third of TB patients either fail to seek treatment or (Wood and others 2010). The reason may be that resi- do so from traditional healers before seeking medical dents in modern housing kept windows closed in an treatment, leading to more severe illness, delayed treat- effort to maintain cleanliness, whereas shacks simply ment, and increased transmission (Brouwer and others had more ventilation. Another study measured the 1998; Sreeramareddy and others 2014). effects of increasing natural ventilation in traditional TB control is premised fundamentally on the assump- housing and demonstrated that natural ventilation was tion that, if active TB cases are identified and treated, facilitated by opening doors and windows and extrapo- transmission will be diminished and ultimately inter- lated that such a change could potentially reduce the rupted. The issue of unsuspected cases is, however, a risk of household transmission by 80 percent (Lygizos serious problem that has received very little attention. and others 2013). Traditional guidelines tend to focus on known or sus- pected cases with classic symptoms and active disease. However, some forms of TB, such as asymptomatic and chronic tuberculosis, do not present with clinical symp- TURNING THE TIDE AGAINST TB toms for months or years and can transmit infection Despite the progress made in TB control over the past over extended periods of time. two decades, serious gaps persist. Although TB can be Evidence strongly indicates that the problem of treated and cured, it is still one of the deadliest infectious unsuspected or asymptomatic cases of TB and unsus- diseases in the world today. pected cases of drug resistance is significant, contribut- Three key elements are needed to achieve effective ing to the one-third of TB patients being “unknown to TB control and to meet the Sustainable Development the health system.” These patients are capable of trans- Goals: (1) early and accurate diagnosis and drug- mitting disease but not ill enough to seek care or to be sensitivity testing, (2) patient access to and completion detected by passive case finding. In a teaching hospital in of effective treatment, and (3) prevention of progression Lusaka, Zambia, for example, 900 newly admitted from latent infection to disease. Obviously, these catego- patients (70.6 percent HIV-positive) who were able to ries are not distinct; each affects and is related to the produce sputum without induction were screened. others, and all face both technical and system challenges. Testing by fluorescent microscopy and automated liquid Without greater effectiveness of these key elements, it culture detected TB in 22 percent of patients, of which will not be possible to bend the curve and dramatically 13.4 percent were unsuspected (Bates and others 2012). reduce transmission and incidence rates in all countries. This number included 18 MDR TB cases, 5 of which Turning the tide against TB therefore requires investing were unsuspected. In the same hospital, 94 patients with in new technologies—diagnostics, treatment regimens, cough, who were admitted primarily for obstetric or and vaccines—and tackling the system and strategic gynecological indications (73.4 percent HIV-positive), challenges that influence the degree to which technolog- had sputum processed in the same way; in addition, ical advances reach the people who need them and Xpert MTB/RIF was used for rapid diagnosis (Friedich translate into better heath. and others 2013). TB was diagnosed in 28 percent of the Tuberculosis 261 94 sputum specimens, of which the Xpert device detected a year) following World War II—that is, prior to the 80.8 percent compared to 50 percent by standard smear introduction of antibiotics (Dye 2015; Golub and others microscopy. Results of this kind are not new: similar 2005). In 1974, the WHO recommended discontinuing results were reported more than a decade ago in a low- active case finding with radiography, since it was no lon- HIV/AIDS setting in Lima, Peru, where 250 of 349 con- ger necessary or cost-effective in populations with low secutive new admissions to a female general medical prevalence of TB and good access to high-quality health ward were screened for TB by sputum smear, culture, care, particularly in HICs. The WHO reiterated this and radiographs. Of these, 16 percent had culture- policy in 2014, again finding that it would not be cost- proven TB, one-third of which were unsuspected, effective, and recommended that indiscriminate mass including 6 unsuspected MDR cases (Willingham and screening be avoided (WHO 2013c). However, it did others 2001). recommend systematic screening for active TB in The DOTS strategy in high-burden countries, even geographically defined subpopulations with extremely when implemented more effectively, will simply not high levels of undetected TB (1 percent prevalence or be sufficient to overcome the challenge of unsus- higher). Regrettably, this WHO recommendation has pected cases or drug resistance. In a groundbreaking not been sufficiently emphasized to stimulate countries population-based active case finding survey of HIV/ and donors to initiate and support active case finding in AIDS and TB in Sub-Saharan Africa, where the popula- high-burden countries or to have an impact on trans- tion has a 23 percent prevalence of HIV/AIDS infection, mission in those countries. Wood and others (2007) found that, despite a highly One strategy for active case finding has been to use effective DOTS TB control program with high rates of X-radiography, particularly mobile X-ray units, to detect compliance, 63 percent of adult cases with pulmonary lung lesions with computer-assisted detection in people TB were not known to the health system. Among HIV- who are relatively asymptomatic (Melendez and others negative individuals, passive case finding identified 2016; Philipsen and others 2015). This strategy is able to 67 percent of prevalent smear-positive cases, the target detect many more patients with infection than is possi- recognized for adequate DOTS implementation. But ble through passive case finding, screening for coughs, or among individuals with HIV/AIDS infection, passive self-reporting. In South Africa, for example, the only case finding identified only 33 percent of those with period in which incidence of TB cases declined occurred smear-positive TB. between 1950 and 1975, when X-ray surveillance cap- Similar findings were obtained in a large survey of tured about 10 percent of the population annually 47,000 individuals in Cambodia, in which 12 percent of (R. Wood, personal communication). While in Europe individuals were examined clinically and sputa were and North America TB control programs dramatically tested by smear and by culture. Only one-third of TB reduced the annual risk of infection in successive cohorts cases were detected by sputum smears. Importantly, 44 (Cauthen, Pio, and ten Dam 2002), such a decline has percent of the sputum-positive cases and 23 percent of not occurred in countries with high prevalence of TB the smear-negative culture-positive cases exhibited none and HIV/AIDS (Kritzinger and others 2009). of the signs of clinical tuberculosis (Mao and others In recent studies in Kenya, abnormal chest radiography 2014). A surprising demographic finding was that peo- had high sensitivity (94 percent) and reasonable specific- ple over age 50 years accounted for more than half of all ity (73 percent) for detecting tuberculosis (van’t Hoog and detected infections, a trend evident in other Asian others 2012). Radiography represents a potentially valu- countries. able population-based screen to determine which individ- Clearly, in many parts of the world where the burden uals should have their sputum tested by culture or by an of TB is low and control programs have been effective, Xpert device for definitive diagnosis. With rapid technical the need for active case finding is not great, and cost- developments, computerized reading of X-radiograms effectiveness would argue against recommending it. could allow high-throughput screening of larger num- However, in high-burden communities, passive case bers of individuals in a cost-effective way (see box 11.5). finding fails to detect early and asymptomatic cases, This is particularly true when combined with clinical leaving a third of TB patients not known to the health symptoms, where it was found that a sensitivity of system. 95 percent and negative predictive value of 98 percent could be achieved. (Melendez and others 2016). Active Case Finding: What Does the Evidence Show? Active case finding—mass screening and surveillance— Active Case Finding in Targeted Regions almost certainly contributed to the rapid decline of TB In many LMICs, patients with TB often seek care from in European countries and the Americas (5–8 percent one or more traditional providers before seeking 262 Major Infectious Diseases Box 11.5 Undetected Cases in Hospitals: The Case of FAST TB infection control needs to focus on prompt identified months after admission, and drug- screening of admitted patients with chronic cough. susceptibility testing would have taken several more This concept has been formulated into a transmis- months, during which time, other patients and staff sion control strategy called FAST (Find cases Actively would have been exposed. FAST, in other words, by cough surveillance, Separate temporarily, and was a dramatic improvement from the status quo. Treat effectively). Similar results are being obtained in a pilot study at Several pilot FAST projects have begun around the the National Institute of Diseases of the Chest world, including in Bangladesh and Russia. In a TB Hospital (NIDCH), a 680-bed facility in Dhaka, hospital in Veronesh, Russia, of almost 1,000 Bangladesh. Because respiratory symptoms are com- patients hospitalized with suspected pulmonary mon in patients admitted to NIDCH, a decision was TB, 93.5 percent were tested by Xpert MTB/RIF made to conduct universal sputum sampling. Of the within two days of admission. Of these, 407 were TB cases identified, 13 percent were unsuspected positive, and 161 were rifampin-resistant, of whom and an additional 1.3 percent were infected with 159 received MDR TB treatment within three work- MDR TB. Diagnoses were available within one to ing days of receiving the result. Under normal oper- two days of collection, and treatment was initiated ating conditions, treatment failure would have been within one day of a confirmed diagnosis. medically appropriate care (Satyanarayana and others counseling and testing. They found that 6,075 per 2011). Even symptomatic surveillance for classic signs of 100,000 of the household contacts were sputum- TB—cough, fever, and weight loss—fails to detect a sub- positive compared with 477 residents of random house- stantial number of cases, especially in persons who are holds without an index case. This finding demonstrated HIV-positive (Corbett and others 2010). the value of screening contacts of index cases in a In cluster randomized trials in Brazil, intensified case high-burden area and cautioned against random finding had a significant impact on reducing TB screening of total populations. Of the 169 previously (Cavalcante and others 2010). The WHO commissioned unidentified cases with TB detected by culture, only several studies on the effectiveness and acceptability of 6 percent were found to be sputum-positive, and active case finding or systematic screening of active TB only 11 percent were symptomatic; the remaining (WHO 2013d). The studies reported that active case 89 percent would not have been diagnosed with passive finding was highly acceptable to populations in Sub- case finding. Saharan Africa. However, the evidence is mixed. In a large, complex In a recent review of undiagnosed TB that would not community randomized trial, the Zamstar study (Ayles be found by passive case finding, Yuen and others (2015) and others 2013), 64,000 individuals in Zambia and the found that it is precisely in high-risk populations, as well Western Cape, South Africa, were surveyed for TB as in persons infected with HIV/AIDS, where intensive symptoms, sputa were taken for identifying individuals case finding and early initiation of treatment or preven- with disease, and one arm employed enhanced house- tive therapy would likely have the greatest impact for the hold active case finding with counseling. Clearly, screen- fewest resources. ing households with index cases revealed a greater In another study in South Africa (Shapiro and oth- number of TB cases. The adjusted ratio for prevalence ers 2012), a group of 2,800 individuals in households and the adjusted ratio for incidence did not differ sig- with a detected TB index case were enrolled to deter- nificantly for the enhanced case finding versus usual mine community prevalence of undetected TB and practice or for the household versus nonhousehold HIV/AIDS. Field teams screened participants for TB groups. The study identified no evidence that enhanced symptoms, collected sputum specimens for smear case finding had an effect on the burden of tuberculosis microscopy and culture, and provided HIV/AIDS at the community level. However, despite not reaching Tuberculosis 263 statistical significance, the findings suggested that the places, and at what cost need to be answered if we are to household intervention did reduce the burden of tuber- avert each additional case of drug-susceptible or MDR culosis in these communities. TB. If interrupting transmission and reducing incidence A meta-analysis of controlled studies found that more rapidly is the goal, the evidence suggests that active screening increased the number of cases found in the case finding targeted to high-burden areas can make a short term and tended to find cases earlier and with less difference. severe disease (Kranzer and others 2013). Treatment outcomes among people identified through screening were similar to outcomes among people identified Community-Based, Integrated Delivery of TB Care through passive case finding. Once again, this analysis Service delivery—from screening and diagnosing confirmed that, in many settings, more than half of the patients to administering treatment and monitoring prevalent TB cases remain undiagnosed. progress—is a key challenge in TB control, as with other A recent transmission modeling study of TB in health care services (Farmer 2013; Kim, Farmer, and South Africa suggested that the current DOTS approach Porter 2013). of passive case finding is unlikely to permit the country to reach the WHO targets for 2050. The model predicts Dominance (and Costs) of Hospital-Based Care that interventions such as active case finding, with early The costs of health service delivery are the largest cost of initiation of treatment and reduction of pretreatment tuberculosis control. While the mean costs of diagnostics loss to follow-up, could have a large impact (Knight and and tests do not vary significantly across income groups, others 2015). Thus, defining the high-burden target the costs of drugs and hospitalization do (Laurence, populations where active case finding is likely to be most Griffiths, and Vassall 2015), particularly the costs of effective is an important analysis to be undertaken. treatment (WHO 2013b, 2014b, 2016a). For example, Although most official data on TB incidence and the costs per patient of drug-sensitive TB treatment were prevalence are estimated for entire countries, a recent US$14,659 in HICs, US$840 in upper-middle-income innovative epidemiological approach asked whether countries, US$273 in lower-middle-income countries, high-burden regions within countries—TB hot spots— and US$258 in LICs, with strong positive correlation might lend themselves to targeted control efforts. Such with income (Laurence, Griffiths, and Vassall 2015). hot spots have been identified in a few countries, such as However, the mean costs of treating drug-susceptible TB Moldova, and it will be of interest to learn whether tar- were highly variable in countries at the same income geted control efforts can improve their effectiveness at level (Floyd and others 2013; Laurence, Griffiths, and lower cost than the usual countrywide programs Vassall 2015). (Dowdy and others 2012; Jenkins and others 2013; Hospitalization accounts for an average of 74 percent Manjourides and others 2012; Zelner and others 2016). of all drug-susceptible TB costs (although this varies These population experiments for TB control need to be widely between individual studies) and 64 percent of all supported and evaluated. MDR TB costs. Conversely, mean costs of outpatient We therefore recommend identifying countries and treatment were 12 times less than hospitalization costs, high-risk populations that are not responding effectively accounting for 6 percent of total costs. In one study, in to the standard DOTS strategy and designing targeted the LMIC group, India consistently had the lowest costs active case finding interventions that could have a for hospitalization and Ukraine had the highest costs for greater impact on earlier case detection and successful hospitalization and outpatient care (Laurence, Griffiths, treatment. and Vassall 2015). In Ukraine, high hospitalization costs, Initiating active case finding interventions in where patients also incurred costs, led to treatment high-burden countries will clearly require external default (Vassall and others 2009). In 2013, in countries financial assistance. Nevertheless, given the limitations with a high burden of TB and MDR TB (excluding of current tools and increasing threat of MDR TB, active China and Russia), almost 38 percent of funding (US$919 case finding may be the best strategy for reducing inci- million) was allocated to hospital inpatient and hospital dence and prevalence in the long run in high-burden outpatient care of drug-susceptible TB (WHO 2013b). areas. And although active case finding is more expen- In addition, in 2013 Russia spent US$1.6 billion on TB sive than passive case finding, added costs will be more control, most of which went toward hospital-based TB than offset by the diagnostic and treatment costs care (Atun, Samyshkin, and others 2006). However, averted—that is, TB transmitted in the hospital and in funding allocated to neglected high-transmission the community by unsuspected cases, especially MDR areas, such as prisons, remains woefully low (Lee and TB cases. The questions of how much screening, in what others 2012). 264 Major Infectious Diseases Efficient and effective delivery of health services for The cost situation in South Africa is similar to that TB is critical for improving TB outcomes globally. Yet, observed in countries where drug-susceptible and MDR the delivery of TB services is inefficient and ineffective TB are treated as inpatient care, such as Estonia due to the high reliance on hospital services and the (US$10,880) and Russia (Tomsk region, US$14,657). vertical delivery of DOTS (Atun, Samyshkin, and These high costs contrast with the lower costs recorded others 2006; Atun, Weil, and others 2010; Samb and in countries where MDR TB is managed in outpatient others 2009). For example, from 2008 to 2010 in most of clinics and at home—for example, Peru (US$2,423) and the world’s 22 high-burden countries, the average cost of the Philippines (US$3,613). In Estonia and Russia, hos- treating a patient with drug-susceptible TB was US$100– pital costs accounted for 43 and 52 percent, respectively, US$500. However, the average costs varied from US$100 of the total cost of treating an MDR TB patient to US$10,000 when Russia (a hospital-based service (Fitzpatrick and Floyd 2012). Estimates suggest that the delivery model) was included (Floyd and others 2013). global average cost per patient of treating MDR TB is In 2014, in most of the 22 high-burden countries for US$13,259 (5th–95th percentiles US$2,797–US$42,040) drug-susceptible TB, the WHO estimated that the share using an outpatient model and US$34,599 (5th–95th of hospital inpatient and outpatient costs ranged from percentiles US$6,959–US$109,154) using an inpatient 30 to 60 percent, with durations of stay ranging from 5 model (Fitzpatrick and Floyd 2012). to 56 days (WHO 2014b). In the 22 high-burden TB High hospital costs crowd out funding for cost- countries, the costs of hospital inpatient and outpatient effective interventions, such as those aimed at addressing care for managing drug-susceptible TB as a proportion the social determinants of health, as well as new diagnos- of the national TB program budget were 1–10 percent in tics and medicines that could be delivered in the com- the Democratic Republic of Congo, Ethiopia, Tanzania, munity or in primary care settings. Thailand, and Uganda; 11–20 percent in Bangladesh, Worldwide, there is substantial variation in the cost of Kenya, Myanmar, and Pakistan; and in excess of 40 treating patients using DOTS. Among the 22 high-burden percent in India (46 percent), Mozambique (51.8 countries, the cost per successfully treated patient varies percent), and Vietnam (74.2 percent) (WHO 2013b). from less than US$100 to more than US$10,000 for a While the WHO report did not have data for China and standardized treatment. Even in countries with similar Russia, earlier studies suggested that, in Russia, where per capita income levels, the cost per successfully treated drug-susceptible TB is treated predominantly in hospi- patient varies 50-fold (Floyd and others 2013). According tals, patients are kept under observation long after treat- to recent cost-effectiveness studies, the cost of treating ment completion; as a result, hospital inpatient costs one MDR TB case ranges from nearly US$15,000 in a account for around 60 percent of the total costs of treat- hospital-based program in Tomsk, Siberia, to US$2,400 ment (Atun, Samyshkin, and others 2006). It is difficult in a community-based program in Lima, Peru (Fitzpatrick to understand why China has shifted most TB care from and Floyd 2012). outpatient to hospital-based care. The failure to prioritize MDR TB adequately or to In South Africa, which has the second-highest number achieve adequate cure rates is a major problem exempli- of confirmed cases of MDR TB (WHO 2011b) and the fied by the situation in two large countries. In India, the highest number of confirmed cases of XDR TB, MDR TB MDR TB cure rate is about 11 percent (Subbaraman and accounts for only 3.5 percent of the TB disease burden, others 2016). In China, only about 5 percent of MDR but absorbs almost half of the US$218 million national patients are being treated, and the policy is formulated TB program budget (WHO 2011c). Patients with MDR entirely around hospital-based treatment, with high TB are hospitalized from the initiation of treatment until relapse rates after discharge (Zhao and others 2012). culture conversion, and the cost per patient of treating MDR TB was US$17,164, more than 40 times the cost of Efficiency and Effectiveness of Community-Based treating drug-susceptible TB (Schnippel and others TB Treatment 2013), which was estimated to be US$314–US$392 for A great deal of evidence points to the effectiveness of community-based treatment (Sinanovic and others transitioning hospital-based TB care to primary health 2003). Around 95 percent of these costs were hospitaliza- care settings (Edginton 1999) and to community-based tion costs (Schnippel and others 2013). More recent models (Ayles and others 2013; Binagwaho 2013; studies in South Africa have put the cost of inpatient Cavalcante and others 2007; Cavalcante and others treatment of MDR TB at US$6,772 (compared with 2010; Corbett and others 2010; Islam and others 2002; treatment of drug-susceptible TB at US$256), with esti- Shapiro and others 2012), even for management of mates suggesting that 45 percent of total costs are hospi- MDR TB (Fitzpatrick and Floyd 2012; Furin and others talization costs (Pooran and others 2013). 2011; Heller and others 2010; Luyirika and others 2012; Tuberculosis 265 Mitnick and others 2003; Mukherjee and others 2002; Several countries transitioning from hospital-based Nathanson and others 2006; Seung and others 2009; services to primary health care or community settings Smart 2010). In addition to clinical effectiveness and have improved quality and outcomes of TB care—for improved outcomes, community-based models appear example, Haiti (Farmer and others 1991), Latvia to be more acceptable to patients (Horter and others (Leimane and Leimans 2006), Moldova (Soltan and oth- 2014). ers 2008), Romania (Marica and others 2009), and A recent community-based model for providing TB Zambia (Miti and others 2003). Ethiopia has introduced treatment was developed in Bangladesh by BRAC, the the use of health extension workers to scale up access to world’s largest NGO (Islam and others 2002; Islam and primary health care services, including TB services, with others 2011). Individuals diagnosed with tuberculosis improved compliance and treatment outcomes (Bilal are given free treatment but asked to provide a small and others 2011; Datiko and Lindtjørn 2009, 2010), and bond, equivalent to about US$3.25, to assure that they employed village outreach programs in rural settings will complete treatment. Upon successful completion of (Shargie, Mørkve, and Lindtjørn 2006). Similarly, India treatment, the funds are returned. The treatment cost has engaged urban community volunteers to supervise per patient was US$312 in 2010. For patients unable to DOTS (Singh and others 2004). South Africa and afford the bond, the community acts as underwriter, Tanzania have used CHWs to expand access to TB ser- which establishes a social incentive to complete treat- vices (Sinanovic and others 2015; Wandwalo and others ment. Program cure rates are up to 92 percent, which is 2004), and Rwanda has achieved effective community- comparable to a government-run program, yet the costs based treatment (Binagwaho and others 2014). In addi- of the community-based program are half those of a tion to community-based care, a public-private mix with public program (Islam and others 2002). NGOs has been used in India, South Africa, and beyond to expand access to TB services and improve outcomes (Lal and others 2011; Pantoja, Floyd, and others 2009; Sinanovic and Kumaranayake 2006; Wells, Uplekar, and Figure 11.5 Cost-Effectiveness of Community-Based Interventions Compared to Health Facility–Based Interventions for Tuberculosis Pai 2015). Treatment in Select Countries In the case of drug-susceptible TB, early indications are that outpatient treatment in the community is not poorer than hospitalized care (Bassili and others 2013; Loveday and others 2012). Although current guidelines recom- mend isolating MDR TB patients until smear or culture Cost increase conversion, community-based treatment of drug-resistant TB is growing in acceptance due to its cost-effectiveness Change in cost per TB patient and a shortage of long-term hospital beds—for example, in South Africa (Brust and others 2012). The benefits of transitioning facility-based services to community-based care are substantial. In six studies, where outcomes between community-based and health facility–based TB care were similar, the costs of community-based care were 33 to 70 percent lower Cost decrease (figure 11.5). In the four countries where the treatment outcomes in community-based TB care and health facility-based TB care were better, costs were 32 to 77 percent lower. Community-based DOTS in Tanzania reduced costs 35 percent: from US$203 per patient treated at a health Less effective Equally effective More effective center to US$128 per patient treated in the community, Effect of community-based interventions compared to with almost identical treatment outcomes. This program health facility–based interventions reduced costs by lowering the number of visits to the Bangladesh Tanzania South Africa Botswana clinic (Wandwalo, Robberstad, and Morkve 2005). In Ethiopia Cambodia Pakistan Uganda Malawi and Kenya, moving to a community-based Kenya Malawi model reduced costs even more: 67 and 77 percent, respectively (Floyd and others 2003; Nganda and others Note: TB = tuberculosis. 2003) (table 11.6). 266 Major Infectious Diseases Table 11.6 Costs and Effectiveness Comparing Community-Based Tuberculosis Care to Health Facility–Based Tuberculosis Care Effectiveness of Community cost Health facility % difference in community-based care vs. Study Study country (2012 US$) cost (2012 US$) cost per patient health facility–based care Islam and others 2002 Bangladesh 172.80 259.20 33 Similar Wandwalo, Robberstad, Tanzania 216.10 331.90 35 Similar and Morkve 2005 Dick and Henchie 1998 South Africa 1,296.10 2,073.10 37 Similar Moalosi and others 2003 Botswana 5,135.90 8,543.40 40 Similar Datiko and Lindtjørn 2010 Ethiopia 138.00 332.70 59 Similar Pichenda and others 2012 Cambodia 639.30 2,131.10 70 Similar Khan, Khowaja, and Pakistan 320.83 471.16 32 Higher others 2012 Okello and others 2003 Uganda 796.52 1,405.63 43 Higher Nganda and others 2003 Kenya 752.65 2,250.51 67 Higher Floyd and others 2003 Malawi 1,040.94 4,477.99 77 Higher Human Resource Challenges to Community- resources need to provide accurate information to their Based Care patients and make it easy for patients to access and take Despite the evidence, many countries have not transi- good-quality treatment consistently until cured. tioned to community-based models of service delivery Inadequate human resource capacity remains an for TB. Barriers to adoption of innovative models of important health system barrier for TB control care delivery are often related to health system gover- (Figueroa-Munoz and others 2005; Harries and others nance, organization, and financing, in particular (exter- 2005), especially for MDR TB, which requires longer and nal and domestic) financing flows that reinforce vertical more complex interventions than drug-susceptible TB programing; provider payment systems that allocate (Keshavjee and Farmer 2010). Several factors, such as large proportions of the budget to structures and inputs weak planning, absolute shortage of health staff, limited (hospitals and hospital activities), rather than health training, inadequate skills, lack of incentives to motivate outcomes; and a health workforce that lacks suitable and retain staff, and inappropriate distribution of the skills (in particular, trained CHWs who underpin available health workforce, have contributed to the community-based service delivery). A general reluc- human resource crisis confronting global efforts to con- tance to adopt innovations may also impinge the move tain TB (Awofeso, Schelokova, and Dalhatu 2008; Buchan to community-based models, including public-private and Dal Poz 2002; Caminero 2003). partnerships, which in several countries have helped to While community-based TB treatment certainly improve outcomes while lowering costs (Atun, de Jongh, requires initial investments, including training, supervi- and others 2010; Atun, Lazarus, and others 2010; Atun, sion, and management of CHWs, the initial investment McKee, and others 2005; Atun and others 2012; Howitt is generally offset in the long run by savings accrued and others 2012; Khan, Khowaja, and others 2012). from community-based treatment rather than hospital Both active case finding and community-based TB care (Wandwalo, Robberstad, and Morkve 2005). delivery models require a competent, motivated health workforce, both at health facilities and in communities. Private Sector Challenges Following patient diagnosis, treatment must be adhered Some high-burden countries, such as India, have health to, and factors that drive patient adherence to medica- care systems in which private sector rather than public tion are complex and relate both to patients’ willingness sector care predominates. In India, about 75 percent of and ability to seek health care and their experiences health care is provided by the private sector, most often within the health system (Munro and others 2007; based on OOP payments (Pai, Daftary, and Satyanarayana Podewils and others 2013). Clinic staff with limited 2016). A high proportion of providers in many countries Tuberculosis 267 are not medically trained, and appropriate care is often In Vietnam, tuberculosis control services, which were not provided (Udwadia, Pinto, and Uplekar 2010). historically operated through a vertical network, are Patients with TB often seek care initially from private now embedded in general health services (Atun, Weil, providers and may consult three to five providers before and others 2010). Integration between care delivery receiving a correct diagnosis of TB, which delays the domains—for example, civilian and prison health care average time to initiate treatment to 60–66 days (Shin and others 2006)—has also been shown to (Satyanarayana and others 2011; Sreeramareddy and improve patient outcomes. others 2014). Vertical delivery of TB services is especially ineffi- In studies using surrogate patients presenting to cient in the context of concurrent tuberculosis and physicians with cardinal TB symptoms, correct diag- HIV/AIDS epidemics (Drobniewski and others 2004). nosis by physicians may be as low as a third (Das and As noted, the burden of TB and HIV/AIDS co-infection others 2015), and surveying physicians for how they is substantial, most acutely in Sub-Saharan Africa, and would treat a new patient diagnosed with TB revealed there is a demonstrable need to integrate TB and HIV/ that no more than a third recommended the WHO AIDS services (Corbett and others 2006; Creswell and standard treatment protocol. In this context, private others 2011; DeLuca, Chaisson, and Martinson 2009; practitioners often do not register TB patients in the Perumal, Padayatchi, and Stiefvater 2009; Sylla and health system; without notifications, it is difficult for others 2007). Emerging evidence indicates the benefits state and national TB programs to conduct planning of integrating TB control with HIV/AIDS control and to improve control of TB. Such countries need to (Gandhi and others 2009; Gasana and others 2008; improve the communication between private provid- Harris and others 2008; Huerga and others 2010; Jack ers and the health system. Innovative approaches in and others 2004; Legido-Quigley and others 2013; Miti India and Pakistan have included the creation of and others 2003; Pevzner and others 2011; Phiri and public-private mixes using interface organizations and others 2011; Uwinkindi and others 2014; Uyei and oth- information technology to improve treatment of TB ers 2011; Walton and others 2004; Zachariah and others (Khan, Minion, and others 2012; Wells, Uplekar, and 2003) and other targeted programs (Howard and Pai 2015). El-Sadr 2010; Zwarenstein and others 2011) in the primary health care setting. Integration improves From Vertical to Targeted, Integrated Delivery outcomes—for example, through concurrent screening Another challenge for TB service delivery is the domi- or through provision of cotrimoxazole during routine nance of vertically organized and financed TB programs, TB care or isoniazid during routine HIV/AIDS care and supported by external institutions (Katz and others at voluntary counseling and testing centers (Uyei and 2010), which circumvent and fail to strengthen weak others 2011). health systems in LMICs in an effort to deliver accessible Treatment for both tuberculosis and HIV/AIDS and quality TB services (Atun, Weil, and others 2010; should be integrated at the clinic as a standardized Car and others 2012; Coker, Atun, and McKee 2004; package of care, with adherence support and HIV/AIDS Coker and others 2005; Samb and others 2009; Shigayeva drug-resistance testing. It is essential that all HIV- and others 2010; Wood and others 2010). positive individuals be tested for TB. In 2004, the WHO Integrated service delivery has been shown to be introduced integrated activities to improve prevention, beneficial. By 2008, TB services were being delivered at diagnosis, and treatment of TB in people living with the primary health care level in 20 of the 22 high- HIV/AIDS (box 11.6), but the achievements have been burden countries and 83 percent of 173 countries mixed. Creating positive synergies through effective reported making progress in TB control (WHO 2013b). integration of TB control with HIV/AIDS and other However, these services were not effectively integrated targeted programs has been arduous (Ansa and others with other disease control programs as part of compre- 2012; Atun and Cooker 2008; Atun, Lazarus, and hensive primary health care. In Cambodia, hospital- others 2010; Marais and others 2010; Okot-Chono based DOTS care was the norm until stronger primary and others 2009; Uwimana, Hausler, and Zarowsky health care enabled the introduction of integrated 2012; Uwimana and Jackson 2013). Integration must community-based TB services. Similarly, the National be site- and service-specific. Rural Health Mission in India has expanded integrated While integrating diagnostic and laboratory services service delivery, including rapid scale-up of child makes great sense, integrating patients with undiag- health, TB, and combined TB and HIV/AIDS services. nosed TB in health care clinics and hospitals without Thailand extended access to primary health care–based the benefit of proper infection control measures poses TB services as part of universal health coverage. serious risks to persons with HIV/AIDS. Moreover, 268 Major Infectious Diseases Box 11.6 WHO-Recommended Collaborative TB and HIV/AIDS Activities In 2004, the WHO recommended a set of collab- 3. Provide ART and cotrimoxazole preventive ther- orative activities to improve prevention, diag- apy to TB patients living with HIV/AIDS. nosis, and treatment of TB in people living 4. Provide HIV/AIDS prevention services for TB with HIV/AIDS (WHO 2004). These recom- patients. mendations were updated in 2010 and 2011 5. Intensify TB case finding among people living (WHO 2012b). Collaborative activities include with HIV/AIDS. the following: 6. Offer isoniazid preventive treatment to people living with HIV/AIDS who do not have active TB. 7. Control the spread of TB infection in health care 1. Establish and strengthen coordination mech- and congregate settings. anisms for delivering integrated TB and HIV/ 8. Use Xpert MTB/RIF as the primary test for diag- AIDS services. nosing TB in people living with HIV/AIDS who 2. Test TB patients for HIV/AIDS. have signs and symptoms of TB. an anticipated consequence of integrating HIV/AIDS demand forecasting, ineffective drug procurement, long and TB care is the exposure of immune-compromised procurement cycles, poor-quality drugs, and delays in the HIV-positive patients to undiagnosed TB, making it delivery of diagnostics and medicines, have led to treat- essential to control infection and separate patients. ment interruptions, exacerbating drug-susceptible and Effective TB control will require health systems to MDR TB epidemics (Mathew and others 2006; van der interact with sectors that address the social determinants Werf and others 2012; Victor and others 2007). The lack of health. However, it has been argued that the program- of availability or poor quality of TB drugs in public clinics matic and biomedical focus fostered by DOTS may have, run by national TB programs or in private facilities leads in some instances, hindered multisectoral collaboration to patients missing doses, creating increased risks for and effective coverage of vulnerable communities (Ayles relapse and the emergence of drug-resistant forms of the and others 2009; den Boon and others 2007; Gler, disease. Podewils, and others 2012; van’t Hoog and others 2011; Stock-outs of TB medicines have many causes. While Wood and others 2007); geographically concentrated some stock-outs are related to poor planning, distribu- groups (de Vries and others 2014); groups at risk, such tion bottlenecks, and poor demand visibility, there are as women, children, and adolescents (Ettehad and oth- global shortages for some drugs. The problem of global ers 2012; Isaakidis and others 2013; Marais and others shortages affects many antimicrobials, especially generic 2010; Moyo and others 2014; Sheriff and others 2010); injectable agents. Very often, given the small number of and disenfranchised populations (Corbett and others manufacturers, when one manufacturer experiences 2009), for example, persons imprisoned in penitentiary problems related to quality or manufacturing, global facilities (O’Grady, Hoelscher, and others 2011; O’Grady, shortages can arise. Maeurer, and others 2011). Global Supply Chain The global supply chain of TB medicines, from the man- Stronger Supply Chains ufacturer to the patient, can be divided into two distinct A robust and well-functioning supply chain is a complex segments: the upstream supply chain (global supply but essential component of any country’s health system. chain) and the in-country supply chain (figure 11.6). An uninterrupted supply of high-quality drugs is imper- The global, or upstream, segment consists of the pro- ative to treat TB effectively and to prevent transmission of cesses related to global demand forecasting, procure- the disease or its escalation to drug-resistant strains. Weak ment, and financing. The in-country segment includes systems of supply chain management, with inadequate the quantification of needs by national TB programs, Tuberculosis 269 Figure 11.6 A Simplified Supply Chain from Manufacturer to Patient and drugs, limited demand for second-line drugs, and constrained supply, with few producers providing medi- Global supply chain In-country supply chain cines that met stringent regulatory approval. Global demand supply Forecasting In a typical healthy pharmaceutical supply chain, planning Procurement Manufacturers Pooled procurement Warehousing End patients some upstream steps such as active pharmaceutical Price negotiation Distribution to clinics ingredient (API) manufacturing, formulation, and pack- Shipment coordination Information collection aging are carried out before the buyer places an order. These forecast-driven steps lead to the finished product Source: Yadav 2010. or the API manufacturer holding some inventory, which reduces the time required to fulfill a confirmed order. In the TB medicines supply chain, however, many of procurement by ministries of health, warehousing, dis- these processes are order driven as opposed to forecast tribution, and information collection about patients driven. Manufacturers hold little, if any, finished product regarding treatment and future needs. inventory and little, if any, API inventory. All steps in the The upstream supply chain for TB medicines is upstream supply chain start only after a confirmed order fraught with market shortcomings. All commonly used is placed. And the order stream from the pooled pur- TB drugs are off-patent, holding little interest for large chaser is lumpy, meaning that it occurs in large quanti- pharmaceutical companies. As a consequence, there are ties at certain times of the year. These problems lead to few producers for most drugs. In particular, timely pro- suboptimal holding of inventory, poor planning of batch vision of life-saving second-line medicines for MDR TB size, subscale manufacturing of the finished product, has been woefully inadequate, with estimates suggesting higher costs, and excessively long lead times. These issues that less than 0.5 percent of the MDR TB cases in 2002– constrain national TB programs, which are unable to 09 were treated with drugs of known quality (Keshavjee plan much in advance due to uncertain financing and and Farmer 2012). Stock-outs of TB medicines are fre- delayed disbursement and thus face further delays in quent in LICs, and shortages exist even in HICs, where receiving supplies after they place an order. Together, standard isoniazid has not been available for more than these factors contribute to stock-outs of TB medicines at a year. the national level. Apart from the lack of manufacturer interest in these Possible mechanisms to address these problems include low-demand, limited-profit-potential medicines, the the creation of an accurate global demand forecast system, lack of procurement coordination across low- and the development of supply-contracting structures that high-demand countries further fragments the small provide limited-volume guarantees to manufacturers, and market. Lack of proper quantification at the country a buffer inventory or stockpile to smooth demand. Their level translates into a lack of robust global forecasts for applicability and cost-effectiveness depend on careful TB medicines. The upstream supply chain for the donor- analysis of the nature of demand uncertainty, supply lead funded portion of the TB market currently operates times, and ability of global program staff to operate these through a pooled procurement mechanism operated by mechanisms. the Global Drug Facility. The Global Drug Facility Some of these mechanisms are now starting to be attempts to coordinate the orders from multiple coun- used by the Global Drug Facility (Arinaminpathy and tries, especially lower-demand countries. For MDR TB others 2015), but they require adequate technical medicines, it also runs a strategic rotating stockpile to resourcing. Also, while Global Drug Facility purchasing reduce stock-outs and volatility in orders to the manu- may lead to greater coordination and pooling of orders facturer. However, the global supply chain is far from from many LICs, a significant portion of the TB burden optimal and requires significant strengthening of its is in middle-income countries, which do not pro- technical capacity to manage a small market with highly cure through the Global Drug Facility. Overcoming the uncertain demand and a fragile supply base (Institute of excessive fragmentation in the global TB drug market, Medicine 2013). especially for low-demand drugs for MDR TB, requires It is important to examine how the global supply not only robust technical solutions but also strong chain for TB medicines has evolved over time. The Green political will. Light Committee was developed in the 2000s to approve programs that deliver MDR TB treatment and provide In-Country Supply Chains access to low-cost second-line drugs. However, the Green In HICs, while the nature of health care provision and Light Committee approved a small number of MDR TB financing varies considerably, medicines are distributed treatments, creating long delays in receiving approvals primarily by private sector agencies. In LICs, and in many 270 Major Infectious Diseases of the TB-endemic countries, in particular, medicines are consumption, and future requisitioning need to flow distributed to health facilities primarily through a central through different levels of the system. Such data are medical store, regional or district stores, or a transport recorded on store ledgers, stock control cards, and requi- fleet owned by the government or a central medical store. sition forms at the district and health facility levels but Global donor–funded or national government–funded rarely get reported to higher levels of the distribution TB drugs also flow primarily through this government- system. The lack of consumption data prohibits better run distribution system (figure 11.7). overall supply planning. There is a critical need for a A multitiered distribution structure wherein TB med- simple and robust logistics management information icines are stored at multiple levels (national, regional, system to record and report these data systematically. district) before reaching TB clinics is common in most Staff need to be well trained to forecast need and countries. The distribution system maps directly to the place orders, yet staff at health facilities often have inad- administrative structure of the health system, for ease of equate capacity to estimate the quantity of medicines administration and governance, as opposed to technical needed, resulting in under- or over-ordering. One solu- or operational imperatives (Yadav 2010). tion is to deploy trained staff from district or regional Successfully managing a multitiered distribution delivery teams to visit health facilities, check the stock system for TB medicines is an information-intensive they have used, help them to estimate how much is operation. Data on stock levels at each stage, past needed for the next period, and replenish that quantity Figure 11.7 Flow of Medicines through the Public, Private, and NGO Sector in Low- and Middle-Income Countries Private sector Public sector NGO sector NGO international International Manufacturers Procurement agents warehouse Manufacturer’s NGO national national subsidiary Central medical stores warehouse National Importers or wholesalers Regional NGO regional Regional Distributors medical stores warehouse Private Sub- prescribers District District wholesalers medical stores and district hospitals City/large town Pharmacies Health center Small town/ Second-tier NGO clinics rural Community pharmacies and health worker community chemical sellers Patients Source: Yadav, Tata, and Babaley 2011. Note: NGO = nongovernmental organization. Tuberculosis 271 from the supply they carry with them. Zimbabwe has Fragmentation of the supply chain makes it difficult implemented such a system, called Delivery Team Top to achieve scale economies and to improve or verify Up (Yadav, Tata, and Babaley 2011). In a large-scale ran- quality, especially given the severe resource constraints domized pilot study in Zambia, relatively simple changes of regulatory authorities. to the information and product flow system significantly In some countries where medicine wholesaling is improved the availability of essential drugs (Vledder and highly fragmented, consolidation of wholesaling and dis- others 2015). HIV/AIDS medicine programs in multiple tribution in the supply chain is being driven by policy countries have also experimented with different variants measures such as better enforcement of distribution prac- of the distribution, requisitioning, and information flow tices and stricter reporting requirements. For instance, model. Such innovation has been lacking for the distri- when a nationwide Good Supply Practice enforcement bution of TB medicines. campaign was launched in China in 2004, the number With the recent explosion of inexpensive informa- of pharmaceutical wholesalers dropped from 16,000 to tion technology such as mobile phones, new options 7,445 (Yadav 2015). have become available for collecting and using infor- mation about clinic-level consumption. However, tech- nology will not fix all of the problems in the distribution Better Information Management and New system. Technologies Apart from lack of information, another crucial The lack of reliable and timely information impedes the cause of poor supply chain performance relates to the organization of TB control and effective discharge by lack of incentives (Yadav, Stapleton, and Van Wassenhove ministries of health of their stewardship function. In 2013): public sector supply chains often lack the ability 2009, only 4 of the 22 high-burden countries had well- to reward good performance or to remove incompetent functioning vital registration systems that appropriately workers. Better mechanisms are needed to align incen- coded causes of death (Glaziou and others 2011). tives and motivate the supply chain workforce (Spisak and others 2016), but greater accountability in the dis- Challenges tribution system for TB drugs requires richer informa- As both an infectious and chronic disease, TB presents tion about stock and consumption data. These models a range of challenges for collecting and managing have not yet been leveraged to their fullest, but they information. First, there is the challenge of case finding have the potential to be the backbone of planning in the and surveillance. Once patients are identified, diagnosis supply system. requires sputum smear and, ideally, culture and drug- Having a healthy and robust supply chain for TB sensitivity testing, a challenge in low-income settings medicines in the private sector is as important as where good laboratory facilities are rare. Sputum improving the publicly run supply chain. Significant smears can be collected in small clinics, but culture proportions of TB patients in high-burden countries and DST is a specialized process. New diagnostic tech- such as China and India seek TB treatment in private niques, particularly GeneXpert, have improved the clinics or obtain medicines in private pharmacies (Wells, situation significantly, avoiding the need for sputum Uplekar, and Pai 2015); in some countries, public- smear or culture in initial diagnostic work-up in sites private models rely on the availability and quality of that have the machines. However, many clinics have to drugs in the private sector. While countries such as send samples to other sites with machines, and any Brazil and China have created social insurance pro- patient found positive for Mtb requires follow-up grams to help patients to cover the cost of private sector testing to assess their response to treatment. Patients services, in most countries patients themselves pay out found resistant to rifampicin require follow-up drug- of pocket for private sector treatment for TB. sensitivity testing to permit individualized therapy TB medicines in the private sector are distributed (Lessem and others 2015). It is therefore critical to through a network of importers, wholesalers, sub- capture lab data from all locations and sources, includ- wholesalers, pharmacies, and drug stores. Compared to ing GeneXpert. Newer, portable models in development private sector pharmaceutical supply chains in HICs, (GeneXpert Omni) with potential for point-of-care private sector supply chains in most TB-endemic coun- diagnosis have the ability to transfer results in real time tries are excessively fragmented, with myriad small by short message service or through the Internet. Even wholesalers and distributors; intermediaries between in countries with sophisticated health care systems, like the manufacturer and the patient; and poor information South Africa, there are enormous problems getting technology and communication systems, which result the data on a sputum sample to the appropriate clinic, in poor coordination across the distribution channel. physician, and patient. 272 Major Infectious Diseases Once patients have been diagnosed successfully, they Staff used the built-in interactive questionnaire to iden- need to be tracked in a longitudinal record that captures tify symptoms and signs of TB in patients who were data on demographics, clinical condition, current and coughing for two weeks or more. In the intervention area previous medication, lab results, complications, and overall, notification of TB cases to the national TB pro- treatment response and outcomes. For first-line TB gram increased from 1,569 to 3,140 cases between 2010 treatment with DOTS, the record is typically a paper and 2011. Increasingly, CHWs in rural communities register in the clinic. This usually supplements other in LMICs are using m-health applications to find registers, including primary care visits and maternal unknown patients with TB and other diseases (Were and health. While well-designed paper registers can be effec- others 2009). tive, they are typically challenging for busy staff to keep Some groups have developed Web-based electronic up to date and accurate and lead to much duplication of medical record (EMR) systems for managing drug- data. This problem becomes much more severe with resistant TB (Fraser and others 2013). EMRs were one of large numbers of patients and mobile populations. the first e-health applications for managing TB. Partners Drug-resistant TB is a particular challenge due to the In Health developed and deployed a Web-based medical complexity of recording treatment and clinical data and record for managing MDR TB in Peru in 2001, with the long treatment times. tools to allow clinicians to view trends in lab data and E-health systems are playing an increasingly impor- changes in medications as well as to create reports for the tant role in the management of TB and are especially national TB program and funders (Fraser and others important for drug-resistant TB in LMICs. Labrique and 2002). The system was also used to collect core data for others (2013) describe 12 key functions of m-health subsequent research studies after additional data collec- (and e-health tools more generally). All of these are rele- tion and cleaning. Image data of chest X-rays captured vant to TB care: with a digital camera were included for most patients, and psychiatric records were added later. The system was 1. Client education and behavior change communica- extensively evaluated to determine its performance and tion (short message service reminders to take medi- potential impact on quality of data and delivery of care cation or attend appointments) (Fraser and others 2006). 2. Sensors and point-of-care diagnostics (attachments From 2000 onward, Peru scaled up the individualized for microscopy) treatment of drug-resistant TB, including upgrading 3. Registries and vital events tracking (community case laboratory facilities at local, district, and national levels. finding and registration) Two e-health systems were implemented as part of Peru’s 4. Data collection and reporting (research data collec- system. The first was an early m-health application to tion; Fraser and others 2012) assist staff in collecting smear and culture data from 98 5. Electronic health (medical) records small clinics in northern Lima. The system reduced the 6. Electronic decision support (information, protocols, median processing time for cultures from 23 days to algorithms, checklists) 8 days and for smears from 25 days to 12 days and signifi- 7. Provider-to-provider communication (telemedicine cantly reduced the number of errors. The intervention consultations) reduced the number of work-hours necessary to process 8. Provider work planning and scheduling (to help results by 70 percent and was preferred by all users. community health workers to manage their patients) Blaya and others (2007) evaluated a Web-based labo- 9. Provider training and education (Web-based ratory management and reporting system, eChasqui, also resources and video on mobile phone memory cards) in Lima, in a large random control trial of 1,671 patients 10. Human resource management (tracking activities, in 44 clinics, 12 of which were randomized to receive patient contacts, and location of community health initial access to the system. Error rates (mainly missing workers) results) fell 87 percent for cultures and 82 percent for 11. Supply chain management DSTs. Delays for cultures were reduced from a median of 12. Financial transactions and incentives. 8 days to 5 days and for DSTs from a median of 17 days to 11 days. In addition, the time to culture conversion fell New Technologies to Improve TB Information 20 percent (Blaya and others 2014). Results similar to Management those seen here in TB patients have been replicated for Several recent e-health applications have addressed case other diseases, particularly HIV/AIDS (Amoroso and finding. An m-health application was developed to assist others 2010; Siedner and others 2015). nonclinical CHWs to screen patients in general medical Several EMR systems have been developed to assist in clinics in Karachi, Pakistan (Theron and others 2015). the management of MDR TB, including eTB manager, Tuberculosis 273 initially developed in Brazil and now deployed in many reasons why only a minority of patients receive fully countries (Fraser and others 2013). effective treatment. Telemedicine is a potential strategy In 2008, a new EMR system was developed for MDR to support clinicians with limited knowledge of these TB care based on the open-source EMR system complex treatment protocols. This can be as simple as an OpenMRS platform (Mamlin and others 2006). The e-mail question to a specialist for advice on a specific OpenMRS-TB system provided similar functionality to patient’s drug regimen and resistance or a video confer- Peru’s EMR, but was embedded in a general-purpose ence with both clinicians and the patient. Many projects EMR platform that was also used to support a range use “store and forward telemedicine” typically involving of clinical care, including HIV/AIDS, primary care, e-mailing text and attached images (Della Mea 1999). maternal health, and oncology. It has been deployed Studies have shown that even modest-specification in Haiti, Indonesia, Pakistan, and several African digital cameras can capture images of chest X-rays good countries. A version of OpenMRS-TB was created and enough for diagnosis and management of TB (Szot and deployed in Peru for a large epidemiological study others 2004). of MDR TB (Fraser and others 2012). A new version of Telemedicine approaches often work well in small- OpenMRS-TB is currently under development for clin- scale projects but are difficult to scale up. Individual ical and research purposes in a collaboration between e-mails are not suitable for large numbers of referrals Partners In Health, Médecins Sans Frontières, and due to the difficulty of ensuring that all of the correct ThoughtWorks. OpenMRS has been developed as information is recorded accurately in the referral and the open-source software and also supports open stan- resulting assessment is recorded in the patients’ notes. dards for the coding of medical data. OpenMRS is now A more effective approach is to share data in a secure, used to support the care of patients in more than 40 Web-based EMR system like Peru’s EMR, OpenMRS-TB, LMICs (Mohammed-Rajput and others 2011; Seebregts or eTB manager (Fraser and others 2013), giving the and others 2009), with versions to support the manage- remote specialist access to the individual patient’s record. ment of HIV/AIDS, maternal-child health, and, more The remote specialist can see the full range of clinical recently, heart disease and primary care. data, lab results, and often imaging and record an assess- The design of OpenMRS offers some advantages for ment directly in the clinical record. Another challenge the development of research data management tools. with scale-up is that clinical expertise is limited and Due to the focus on safe collection, storage, and man- cannot be “spread too thin.” A more effective and scal- agement of clinical data, it includes auditing of data able strategy can be to use telemedicine as part of changes in the main database tables. This feature allows training initiatives (Geissbuhler and others 2003), along tracking of the history of changes in data items linked with better clinical guidelines and decision support for to the login of the user. OpenMRS is designed around local staff. a flexible data dictionary, called the concept dictionary, which allows new data items to be added without SMS Reminders to Improve Adherence changing the underlying structure of the database. The With the importance of achieving good adherence for dictionary simplifies the translation and maintenance managing TB and preventing the emergence of drug of items in additional languages like Spanish and has resistance, there is great interest in tools and strategies led to the development of a core dictionary mapped to to improve adherence. DOT is the best-established coding standards such as ICD-10 and SNOMED-CT approach, but questions have been raised about its scal- and shared by most users of the system. A major ability and cost. With mobile phones widely available in advantage of OpenMRS from a developer’s perspective LMICs, m-health may provide tools to support treat- is its modular software architecture, which allows ment adherence. Most work has focused on improving software modules, either from the OpenMRS module adherence to ART with text messages or interactive voice library or newly developed modules, to be plugged into prompts, with evidence of improvements in adherence the main system, adding functions without changing in some random control trials (Lester and others 2010; the core system. Pop-Eleches and others 2011), but also some negative results (Cameroon, India). Telemedicine to Support Health Workers Analysis of these studies suggests that messages cus- While clinics can generally manage the DOTS protocols tomized to each patient and the ability of patients to for treating drug-sensitive TB, management of drug- communicate with actual staff (not just automated resistant TB can be complex, especially for patients with prompts) improve adherence. To date, these tools have second-line resistance, including XDR TB. Expertise in not matched the adherence rates of effective DOT, and managing such patients is limited, which is one of the there is evidence of messaging fatigue among patients. 274 Major Infectious Diseases Further work is under way to design interventions based applications, and national reporting systems, such as the on established psychological models of behavior change District Health Information System (DHIS 2). accompanied by rigorous evaluation. Such e-health architecture approaches are being deployed in several LMICs, including Bangladesh, India, Information Technology to Manage Supply Chains Kenya, the Philippines, and Rwanda. Many countries An additional challenge for managing drug-resistant TB have now adopted the DHIS 2 to collect, manage, and has been establishing effective medication supply chains, report on health data at the district and national levels. and information systems are increasingly important for The system can take direct feeds of data from systems forecasting requirements, tracking medication shipments, such as OpenMRS and some m-health applications, and managing inventory at clinics. improving the accuracy and timeliness of data. Lack of supply of second-line TB medications is a key Embedding the specialized data collection and analysis factor in poor scale-up of treatment. While first-line tools for TB care in broader e-health systems has large drugs are low cost and generally widely available, second- benefits, such as facilitating case finding through primary line drugs are mostly used for drug-resistant TB and often care visits and lab data, identifying potential risk factors manufactured to order. Orders have to be placed months like diabetes, maintaining a complete and accurate list of in advance to ensure continuity of care, and accurate medications, and providing effective decision support forecasting is essential. Information systems can assist in based on the sharing of all key data. multiple steps in this process, including forecasting, Further evaluation is needed on the performance of ordering, tracking shipments, and managing inventory. e-health systems in LMICs as well as the clinical impact In Peru, the EMR was used to forecast medication on TB management. In addition, there are only limited requirements for treating drug-resistant TB. Combining data at present on the costs of deploying such systems data on the number of patients enrolled, their length of and maintaining good performance and usage levels in time in treatment, recruitment rate, and current regi- the long term (Blaya, Fraser, and Holt 2010). There is, mens resulted in error rates of 3 percent or less for more however, some evidence that effective EMR and labora- than 1,000 patients in both 2003 and 2004 (Fraser and tory information systems can save money in LMICs by others 2013). A related study looked at forecasting med- reducing errors and waste and speeding patient manage- ication for 68 patients in the same cohort and compared ment (Driessen and others 2013). It is also very likely to that to the usual manual methods. In one study, Peru’s be more cost-effective to embed TB management tools in EMR predicted 99 percent of one year’s needed supply existing e-health systems than to run parallel systems. of medicines, while more manual methods predicted 149 percent (Fraser and others 2013). New tools are becoming available for managing RESEARCH AND DEVELOPMENT shipments and inventory in LMICs.4 One information system for drug-resistant TB, eTB Manager, includes In addition to requiring new care delivery and health inventory management for each clinic linked to require- system strategies, turning the tide against TB will likely ments forecasting (Fraser and others 2013). These tools require technological advances that could accelerate cure, are now available for general use in drug forecasting. reduce transmission and incidence, and prevent disease. M-health tools are also being used to track inventory in Despite the effectiveness of standard drug regimens for local clinics in East Africa using text messages (SMS drug-sensitive TB, resistance is increasing, compliance for Life) and have been shown to reduce stock-outs with long treatment times is problematic, and new drugs dramatically for antimalarial drugs (Barrington and and regimens are needed. others 2010). They are also being used for TB medica- Current TB therapy has many advantages: the stan- tion. Other systems have been developed to detect coun- dard drugs are remarkably safe, with little toxicity even terfeit medication in countries like Nigeria by allowing in vulnerable populations such as pregnant women and patients to text a unique code printed on the medication children; a complete course is highly effective against container to a free number.5 drug-sensitive disease; and the drugs themselves are As noted, the OpenMRS-TB EMR platform supports affordable in the poorest parts of the world. Despite the management of a wide range of diseases and primary these advantages, first-line drugs are ineffective against care. Its use of open standards for storing and exchanging multidrug-resistant strains. And rifampin, perhaps the data supports interoperability with other e-health sys- most effective agent in the current first-line therapy, has tems, allowing systems with a range of functionality to pharmacologic interactions with many other drugs, be linked together—for example, EMR systems, labora- most notably HIV/AIDS protease inhibitors taken by tory information systems, pharmacy systems, m-health many co-infected individuals. Tuberculosis 275 The most important limitation of the current regi- patients who receive rifampin at recommended doses men is the extended time required for effective therapy. have low serum levels. Increasing the dosage of rifam- “Short course” chemotherapy is anything but short; pin or using a different rifamycin might enhance the failing to complete six months of treatment leads to clearance of infection. Several studies are looking at significant rates of relapse. Because patients feel better altered dosing regimens for rifampin or substitution in a matter of weeks, they often have little motivation of the long-acting drug rifapentine. As measured by to continue taking their medications. In addition, the a surrogate endpoint—culture conversion at two extended course means that a substantial investment months—rifapentine is no better than rifampin. In is needed to ensure a continuous drug supply. These multiple short-term controlled trials that included flu- requirements add substantially to the cost of what oroquinolones to shorten therapy, the failure rate of otherwise would be an inexpensive undertaking. This, cures at two to four months was unacceptably high. together with the training and logistics necessary None of those shortened regimens was effective enough to ensure adequate DOTS implementation, has led to to offer substantial advantages in treatment (Gillespie remarkably poor results and continued treatment and others 2014; Jindani and others 2014; Merle and failure, leading to millions of deaths each year. Thus, others 2014). the challenge is to develop not just new drugs, but The fluoroquinolone trials were phase III studies ideally new drug regimens effective in treating drug- designed to permit U.S. Federal Drug Administration sensitive and drug-resistant TB and shortening the (FDA) approval for these regimens, if successful. time of treatment. Performing such a trial is an enormous undertaking. The Better drugs and regimens could have a substantial current treatment regimen is highly successful in impact. To be most useful, they would have the following some patients, with a less than 5 percent relapse rate attributes: in most settings. Showing improved efficacy requires an enormous number of patients, which is why the fluoro- • Rapid activity, shortening the course of therapy quinolone study was designed to show noninferiority, a required for cure criterion that does not require the same level of evidence. • Safety, allowing use in a wide range of patients with- Nevertheless, these studies require thousands of patients out requiring substantial prescreening to provide confidence in the results and investments in • Easy administration, preferably oral, so that health infrastructure in the low-income settings where the trials professionals are not required are conducted, making them extremely expensive and • Limited interactions with other drugs, particularly logistically challenging. antiretroviral drugs To mitigate the costs associated with such trials, • Limited cost, making them affordable in the poorest some studies are experimenting with alternative trial parts of the world. designs. For example, one study investigated the efficacy of the oxazolidinone linezolid. Instead of recruiting There are two general paths to achieving these goals: patients with drug-sensitive disease, these investigators optimizing the use of currently available medications studied patients with MDR TB, a group where success and developing completely new drugs. Phase III clinical rates are historically lower (Gler, Skripconoka, and trials are expensive and will always be rare. It will likely others 2012; Lee, Song, and others 2015). They com- be impossible to test each new drug serially. pared regimens that had been individually designed for Given the enormous cost of developing new drugs each patient with the same regimen plus linezolid. This (the average cost of developing a new drug to licensure is study showed a faster rate of clearance of bacteria and a on the order of US$1 billion or more), the former is the lower relapse rate in the linezolid-treated patients, sug- most attractive path to making rapid changes to recom- gesting that this drug has properties that might allow mended therapy. a shorter regimen, at least in this setting. However, linezolid treatment was associated with very high levels of toxicity, far higher than could be tolerated by Optimizing the Use of Current Medications patients with drug-sensitive disease. Still, this class Is it possible to reconfigure the current drug regimen to of compounds shows promise. In another study, dela- produce better efficacy against a broader range of manid (OPC-67683), a nitro-dihydro-imidazooxazole organisms? The evidence so far is mixed. Animal derivative, was found to accelerate sputum clearance of studies and human pharmacokinetic observations have Mtb by 45 percent in two months, somewhat better than suggested some modifications of TB drug therapy standard treatment (30 percent). These studies suggest that might produce better results. For example, many that the priority given to shortening treatment time to 276 Major Infectious Diseases sputum conversion may be compromising the need to of both drugs for use in MDR infections. However, final assure cure and prevent treatment failure and relapse. approval and an expanded indication for drug-sensitive disease will require larger phase III trials. Several new regimens are currently undergoing testing Finding New Treatments in clinical trials. Like the fluoroquinolone trials, many of Producing novel agents is in some ways superior to opti- these target drug-sensitive TB with the aim of shortening mizing the use of current drugs. The length of therapy the duration of therapy without compromising the rate cannot be shortened with current drugs. And because of cure. As of this writing, trials are planned, enrolling antibiotic resistance arises from mutations rather than participants, or under way that will test whether treat- acquisition of broad determinants of resistance, virtually ment can be shortened to as little as two months using a all bacteria, including MDR TB strains, should be sensitive variety of regimens. Of these, a phase III trial testing to completely new classes of antibiotics. These advan- pretomanid/moxifloxacin/pyrazinamide (PaMZ) was tages must be balanced, however, against the substantial the farthest along. However, the trial was put on hold due costs involved in developing new drugs. Preclinical devel- to unexpected toxicity, and its future is uncertain. opment costs tens of millions of dollars, while complet- In addition to drugs that either have already been ing all of the clinical studies necessary for drug approval approved or are in late clinical trials, several preclinical can run into the hundreds of millions. And there are compounds are undergoing development. These include substantial risks along the way: only a small minority of both new members of existing classes, such as oxazolidi- compounds that enter clinical trials are approved; many nones that might have less toxicity than linezolid, and more fail to make it into clinics. completely new classes of compounds. Compounds with Nonetheless, new drug development for TB already novel mechanisms of action are particularly attractive. has achieved some notable successes (Hoagland and oth- Not only are they generally active against drug-resistant ers 2016). Two new drugs have recently received approval disease, but they may exploit pathways that could clear for human use for MDR TB under certain conditions. disease more rapidly. Both bedaquiline and delamanid have been approved in However, two issues have arisen with these new Europe; bedaquiline also has received FDA approval. compounds. First, many compounds with antibacterial These agents have interesting and novel mechanisms of activity seem to target a very limited number of bacterial action. Bedaquiline, a diarylquinoline, inhibits bacterial processes. These often induce cross-resistant mutations. adenosine triphosphate synthesis by directly blocking the Therefore, there is far less diversity than desired. Second, adenosine triphosphate synthase complex. Delamanid, a and perhaps more concerning, much remains unknown nitroimidazole, is converted by bacterial enzymes to its about the fundamental biology of infection. A principal active form, which may liberate toxic nitric oxide in the goal of new drug development is to shorten the course of process. As predicted for drugs with new molecular therapy. However, there are no in vitro correlates that mechanisms, there is little cross-resistance to existing anti- confidently predict that an early-stage compound will biotics (although altered activity of an efflux pump result in more effective therapy. This presents a consid- might decrease the efficacy of both clofazimine and erable obstacle to the drug development process. And bedaquiline). It is not clear how these agents will be used. differences in the effectiveness of drugs between labora- In rather small clinical trials, bedaquiline was effective, tory studies and patients question whether in vitro and but patients who received the drug had higher death animal models adequately predict the most effective rates, largely after treatment was completed, for unknown compounds. reasons (Cox and Laessig 2014; Diacon and others 2014; Do standard markers of rapid clearance in clinical Gupta and others 2015). Until more is known, bedaquiline trials correlate with ultimate treatment success? It is use will likely be restricted to persons with drug-resistant likely that some drugs used in TB act more slowly and TB (WHO 2013e). Less is known about delamanid and might not be seen to be effective in early bacterial clear- its optimal use at this point. ance studies and, conversely, that some drugs act rapidly The path to obtaining approval for bedaquiline and but do not sustain their effects and control infection over delamanid is both interesting and illustrative. Both were the longer term. Without biomarkers for the state of tested in patients with MDR disease in much the way that viability and magnitude of the TB bacillus in the host, linezolid was used, adding them to optimized therapies. answering these questions will likely require large trials Patients treated with the new drugs cleared infection (Wallis and Nacy 2013). significantly more rapidly and had lower rates of relapse Even in the best of circumstances, serial testing of indi- when treatment was stopped. The small numbers in vidual drugs in large trials is unlikely ever to be affordable. these trials provided the basis for “conditional” approval Moreover, these drugs will never be used alone; instead, Tuberculosis 277 they will always be used in combination with other drugs. in preclinical studies in experimental animals, and some And these combinations might be more (or, conceivably, are being tested in nonhuman primates. less) efficacious than would be predicted for individual The scientific basis for development of any effective drugs. Indeed, experiments in animal models suggest that vaccine includes (1) significant understanding of immu- some drugs can act synergistically to effect much more nological mechanisms of protection against infection rapid cure. This has led to the model of testing regimens or disease; (2) molecular correlates of mechanisms that rather than individual drugs. In this model, which would diminish the need for large, multiyear efficacy has been advanced by the TB Alliance, new drugs would trials; (3) definition of Mtb antigens that engender those be tested and approved in combinations. While the protective mechanisms; (4) means of delivering those characteristics of individual drugs might never be deter- antigens that generate or prime for protective rather mined (and, in this model, would not necessarily be than pathogenic responses; and (5) animal models that approved as individual agents), this strategy would pro- are more predictive of protection in humans than cur- vide a much more rapid and practical path to drug rent models appear to be. None of these criteria has been approval, albeit at the risk of missing information about met for any TB vaccine candidate at present. individual agents. Several special considerations are necessary in vaccine Because of the cost of developing new drugs and testing. Since vaccines, in contrast to most drugs, are regimens and the fact that the populations in greatest need given to healthy children or adults, their safety must are in LMICs, an enormous challenge remains: how can be the foremost concern. In addition, they need to be countries afford new more effective regimens, and how tested in places with a high TB burden that have labora- can the international community contribute to making tories capable of analyzing the immune correlates of them available to the populations that need them? protection; they cannot compromise the ability to test for Pharmaceutical companies have few private sector incen- infection with Mtb in IGRAs; and the preexposure to tives to invest in TB drug development or, more generally, environmental mycobacteria in places where they are in antibiotic development at present, and some sort of tested cannot compromise detection of protection. public-private mechanisms will need to be developed. From animal studies and human Mendelian genetic studies, it appears that both CD4+ and CD8+ T-cells and IFN-γ, TNF-α, and other cytokines are necessary to pro- Developing New Vaccines tect against disease (Modlin and Bloom 2013; O’Garra In the past two decades, there has been a renewed effort and others 2013). The challenge is to learn what responses to develop vaccines against TB that would provide are sufficient for protection. It is unclear how faithful greater protection than BCG. There are at least three small animal models of TB will be to the human strategies for contributing to TB control where vaccines response to vaccine candidates. There is hope that non- are being tested in clinical trials. One strategy is to deter- human primates may be the most predictive model of mine whether vaccines can prevent infection with TB human protection. Human vaccine trials showing at (prevention-of-infection trial), in which IGRAs are used least partial protection may be the only way to establish to detect infection by Mtb. A second is to test whether those conditions. The first new candidate, MVA85A, vaccines can prevent recurrence of TB or MDR TB after expressing a major Mtb antigen, 85A, in modified vac- treatment (prevention-of-recurrence trial) or possibly cinia Ankara vector, was tested in more than 3,000 South be used therapeutically with chemotherapy to accelerate Africa infants in a well-executed phase IIb trial (Tameris cure. Finally, the ultimate goal is large-scale immuniza- and others 2013). The candidate, which was reported to tion for prevention of disease (POD). About 40 vaccine induce IFN-γ and engender some protection in four candidates are at various stages of preclinical testing, and animal species, failed to protect children against either 15 vaccine candidates are currently in clinical trials infection or disease. (Jiménez-Levi 2012). BCG is known to induce Th1 T-cells and a number of Among the candidates in the pipeline (Evans and cytokines—for example, IFN-γ, TNF-α, IL-12—that, others 2013) are (1) recombinant vaccine candidates in animal models and in human Mendelian genetic expressing Mtb antigens in BCG, adenovirus, cytomeg- deficiency studies, appear to be necessary for protection alovirus (CMV), or other vectors; (2) genetically attenu- against TB. However, the immunological factors that are ated whole-cell Mtb strains, lacking either virulence sufficient to engender protection are not yet understood determinants or ability to replicate; (3) and a variety of (Modlin and Bloom 2013). It is very unlikely that subunit protein antigen candidates with adjuvants that large-scale trials of new vaccines, such as the South India would be used as boosters in children or adolescents trial, which followed 360,000 people for 15 years and primed with BCG. All of these vaccines are being tested failed to show protection in any age group, will soon be 278 Major Infectious Diseases undertaken for new candidate vaccines. Thus, there is an trials are great, several strategies are being developed to urgent scientific need to develop molecular “correlates gain insights into the critical immune mechanisms neces- of protection” that can be measured in small numbers of sary for protection in humans. Testing vaccine candidates recipients, that will predict which new vaccine candi- in TST-negative individuals and evaluating their ability dates are likely to protect against infection and disease, to prevent infection as measured by IGRAs could be and that will identify which individuals are likely to accomplished in a shorter time than disease prevention remain susceptible or relapse after treatment. In the trials. As discussed, vaccinating patients at the comple- absence of molecular markers of protection—for tion of drug treatment to prevent relapses or reinfection example, either involved in protecting against infection in high-burden areas could provide information in as or preventing latent TB from progressing to active little as one to two years. In all such trials, it will be essen- disease—serious consideration will have to be given to tial to study multiple molecular and immunological the development of a safe, attenuated, but live genetically markers to develop correlates of protection. Finally, the engineered Mtb challenge. This development could be as most promising candidates need to be tested in small valuable to TB vaccine development as live challenges groups of volunteers to understand which of the different have been in malaria and enteric vaccines. mechanisms each candidate engenders, which are likely Modeling the impact of possible new TB vaccines has to correlate with protection, and which can be used as been enormously valuable. It has shown that, because biomarkers for protection. Here innovative trial designs, children contribute little to transmission, giving a more such as matched-pair randomized trials, could provide effective vaccine than BCG only to infants and children power and information with much fewer volunteers would have little effect on the epidemic (Knight and (King and others 2009). others 2014). Modeling suggests that, in high-endemic Relevant are the older studies indicating that latent countries, vaccinating adolescents, who were the recipient TB seems to engender persisting immune responses that population found to be highly protected (about afford significant protection from disease (Andrews and 80 percent) in the original trials of BCG and M. microti others 2012), when compared with the risks of reinfec- in the United Kingdom (Hart and Sutherland 1977; tion of TB patients who have been cured (Middelkoop Sutherland and Springett 1987), would be more effec- and others 2015). This raises the concern that, if chemo- tive than vaccinating infants. Immunizing or boosting therapy kills most of the Mtb organisms, the susceptibil- adolescents just before they enter the age of highest ity of people successfully cured of TB will revert back to risk would likely have the greatest impact on the that of naïve individuals. This suggests that the immu- disease burden. nologic value of latency may lie in the persistence of Because of the need to immunize large numbers the pathogen and microbial antigens. Thus, the duration of people in a general population to obtain enough statis- of protective immune responses engendered by new tical power to establish vaccine efficacy, the approach of vaccines may be critical to protecting against reinfection the field is to test smaller numbers of individuals in high- and relapse. These findings reinforce the approach of burden countries to learn about immunological parame- vaccinating or revaccinating individuals who have been ters that may correlate with protection. For example, an “cured” by treatment to test whether vaccination will experimental trial with a small targeted population could reduce the incidence of relapse or reinfection. ascertain whether immunization could prevent reinfec- Since BCG is the most widely used vaccine in the world, tion and relapse (at the end of treatment). Targeting particularly in high-burden TB countries, the most likely vaccines to a group of treated patients would shorten the vaccine strategy will be priming with BCG or another time to learn whether there was an effect on relapse and whole-cell candidate in early childhood and boosting with reduce the costs and time required for disease prevention a live attenuated TB vaccine, a vaccine with Mtb antigens trials in large populations. It is remarkable how little expressed in a viral vector (such as an adenovirus or CMV), research has been devoted to combining chemotherapy or a subunit vaccine containing multiple epitopes plus an and immunotherapy in TB. adjuvant. Particularly exciting are new vaccine platforms that offer the possibility of generating long-enduring immune responses to Mtb antigens, including recombi- Understanding the Immune Mechanisms Necessary nant BCG vaccines designed to engender both CD4 and and Sufficient for Protection CD8 T-cells (Kaufmann and others 2014), attenuated TB Current thinking is that a single vaccine given at birth is vaccine (Spertini and others 2015), attenuated recombi- unlikely to provide sufficient protection to prevent dis- nant CMV vectors (Hansen and others 2013), and mRNA ease in adults, who have the highest risk of developing (messenger ribonucleic acid) vaccines (Chahal and others disease. Because the duration and costs of vaccine efficacy 2016; Petsch and others 2012). Clearly, vaccine trials Tuberculosis 279 comparing multiple candidate vaccines in phases II and III FINANCING FOR TB PROGRAMS are enormously expensive and time-consuming (on the order of US$50 million per trial over a period of three to The WHO estimates that funding for prevention, diag- five years). This is why there is an urgent need to define nosis, and treatment of TB reached US$6.6 billion, of molecular or immunological correlates of protection that which US$5.3 billion was for diagnosis and treatment of would enable the up-selection of the most promising can- drug-susceptible TB and US$1.8 billion was for MDR didates from small-scale human studies. TB (WHO 2014b). Of those funds, 84 percent derives An alternative being pursued is developing a safe, from domestic sources, which vary across countries genetically engineered Mtb live challenge strain that (WHO 2016a). could persist long enough to enable rapid assessment of In 2014, development assistance allocated US$1.4 the effectiveness of a vaccine to induce killing of the billion to TB (IHME 2015), a drop of 9 percent from 2013, challenge strain, but totally lacking the potential to cause two-thirds of which came from the Global Fund. The disease. With phase I human studies in a small number WHO estimates the shortfall in funds necessary to expand of individuals, the ability of a vaccine candidate to kill TB programs at about US$2.0 billion (WHO 2016a). The the challenge strain would support conducting a small Global Fund is the single largest funder of TB assistance number of phase II and phase III clinical trials to test the globally committing about 55 percent of its funds to HIV/ efficacy of a particularly effective vaccine candidate AIDS, 27 percent to malaria, and 18 percent to TB (IHME against TB infection and disease. Historically, all vaccines 2015). Yet financing from the Global Fund represents have been iterative processes with continuous learning about 50 percent of all current development assistance for and improvements. The development of biomarkers of health devoted to TB. The United States does not have a protection that would enable identification of the most separate entity for funding TB as it does for HIV/AIDS promising candidates for large clinical trials would pro- and malaria, instead channeling half of its US$500 million foundly accelerate TB vaccine development. in TB assistance through the Global Fund; the remainder Effective new vaccines are essential for TB control, yet is channeled bilaterally. In 2014, the Bill & Melinda Gates their development, testing, and regulatory approval Foundation provided about 12.6 percent of all develop- require many years and considerable investment. The ment assistance for health dedicated to TB. Other govern- slow decline in TB incidence globally, especially in ment donors were the United Kingdom (7.7 percent), high-burden countries, even as mortality and prevalence France (7.0 percent) Germany (5.1 percent), Japan (3.4 are declining, compels us to recognize the serious possi- percent), Canada (4.7 percent), and Australia (2.2 percent) bility that tuberculosis may not be controlled without a (IHME 2015). Of these funds, about 78 percent were protective vaccine. provided through the Global Fund. The WHO estimates that the level of funding required to enable a comprehensive approach to controlling TB Figure 11.8 Funding for Tuberculosis Prevention, Diagnosis, and would be on the order of US$8.8 billion, two-thirds of Treatment, by Intervention Area, 2006–14 which would be for diagnosis and treatment of drug- susceptible TB and 20 percent for drug-resistant TB 7 (WHO 2016a). This figure does not include the costs of 6 research needed to develop new drugs, vaccines, and Constant 2014 US$, billions diagnostics, which the WHO estimates would require an 5 additional US$2.0 billion. 4 Given the magnitude of the TB epidemic, the emer- gence of MDR and XDR TB, and the need to strengthen 3 health systems for TB, a greater level of funding will be 2 required to extend current efforts and enable the new 1 approaches recommended here to reduce incidence and transmission of the disease dramatically (figures 11.8 0 and 11.9). A summary of the latest 2015 recommenda- 2006 2007 2008 2009 2010 2011 2012 2013 2014 tions of the Copenhagen Consensus Center, a consor- Year tium of international economists, suggested that Total Drug-susceptible TB TB/HIV additional investment in TB would represent a good buy MDR TB Other (Economist 2015). For every US$1 invested, the return Source: WHO 2015b. was estimated to be US$43. Were US$2 billion invested to Note: TB = tuberculosis; HIV = human immunodeficiency virus; MDR = multidrug-resistant. cover the shortfall, the potential economic savings could 280 Major Infectious Diseases be on the order of US$300 billion (Lundgrun 2015). But Figure 11.9 Funding for Tuberculosis Prevention, Diagnosis, and despite increases in domestic and international financ- Treatment by Funding Source, 2006–16 ing, the WHO estimates the current funding gap to be 7 US$2.0 billion for extending current measures and an Constant 2016 US$, billions additional US$1.3 billion for research and new technolo- 6 gies. That leaves a funding gap of US$2.7 billion per year 5 to assure a full response to the TB epidemic. 4 3 ECONOMIC ANALYSES AND COST-EFFECTIVENESS 2 1 As the pipeline for new TB diagnostic technologies contin- ues to expand, health and economic evaluations are needed 0 2006 2008 2010 2012 2014 2016 to inform decisions about the most promising options to Total Domestic Inpatient and outpatient carea pursue in different settings and patient populations. Global fund Other Economic evaluation of new diagnostic approaches can seem deceptively simple, but several factors should be con- Source: WHO 2016a. sidered if these evaluations are to provide credible and a. 91 percent of funding for inpatient and outpatient care is accounted for by middle- and useful guidance for policy. These factors pertain both to high-income countries; such countries do not typically receive international donor funding for inpatient and outpatient care services. Data are estimates based on country-reported utilization. quantifying potential impact and to estimating costs. First, it is important to consider the pathway(s) by which new diagnostics are expected to lead to improved level, scaling up new technologies often involves econo- health outcomes. For instance, aspirations for new point- mies (or costs) of scale that relate to the shape of the of-care diagnostics point to the benefits of returning average cost function in relation to the quantity, which rapid test results, which can reduce loss to follow-up by reflects different mixtures of fixed and variable costs eliminating the need for a return visit. Quicker diagnosis at different scales. At a more complex level, achieving may increase the rate of initiating treatment, improve population coverage requires paying attention to health patient outcomes, and reduce transmission by decreasing system capacity, reflecting constraints not only on the the period of infectiousness. However, imperfections in budget but also on infrastructure and human and other implementation may cause real-world application to fall resources. The validity of cost-effectiveness estimates short of the maximum theoretical potential. depends on quantifying the costs of delivering an inter- Moreover, evaluating the impact and cost-effectiveness vention or strategy in a way that is consistent with the of a new approach requires comparing the new approach benefits ascribed to that intervention or strategy. Many to the status quo that it will displace. For example, a published studies fail to meet this requirement. comparative evaluation of a new diagnostic approach With regard to diagnostic approaches, important ques- should specify (1) where the diagnostic will be used (for tions should be addressed regarding patient pathways to example, only in central facilities such as district hospi- care, how a new diagnostic technology or strategy will tals or also in peripheral health centers); (2) the sequence alter the pathways, and where delays or loss to follow-up of tests and associated responses that will guide decision may occur, particularly as these might attenuate the making in different types of individuals or be based on expected benefits of a diagnostic. In HIV/AIDS, consider- particular population-level factors (for example, in indi- able attention has recently been given to the so-called viduals with HIV/AIDS or a history of treatment or in a “cascade of care,” and similar considerations are highly setting with a particular population-level HIV/AIDS salient to tuberculosis control interventions, including prevalence); and (3) the current diagnostic approach in those relating to diagnosis (Subbaraman and others 2016). these settings that constitutes the status quo comparator (that is, the extent to which bacterial culture or drug- sensitivity testing is currently being used). Cost-Effectiveness of Using Xpert MTB/RIF It is also crucial to estimate the costs associated with Xpert MTB/RIF represents a significant technological scale-up and implementation, which are often signifi- advance in accelerating diagnosis of TB and MDR disease cantly greater than the costs of the commodities per se in many settings. However, it has significant limitations. or even the costs including other health care services Several studies have examined the cost-effectiveness of that are consumed at the patient level (for example, the Xpert MTB/RIF in different LMICs. A cost-effectiveness opportunity costs of provider time). At the simplest study published in 2011 by Vassall and colleagues (2011) Tuberculosis 281 and focusing on India, South Africa, and Uganda The cost-effectiveness and affordability of first-line estimated that Xpert devices used in combination with regimens for TB treatment have been long established smear would have an incremental cost-effectiveness ratio (World Bank 1993), and attention has focused on explor- of between US$41 and US$110 compared to conventional ing the most cost-effective way to deliver treatment. use of smear plus clinical diagnosis. A 2012 modeling For countries providing TB treatment through hospitals, study (Abimbola and others 2012) estimated cost- studies have demonstrated the relative cost-effectiveness effectiveness of Xpert at the regional level in Sub-Saharan of ambulatory treatment (Floyd, Wilkinson, and Gilks Africa and found that Xpert would reduce mortality and 1997; Vassall and others 2002; Vassall and others 2009). lower overall costs. Another 2012 study focusing on Economic analysis in other settings has focused on screening with Xpert prior to initiating ART in HIV/AIDS delivering care through community structures (Floyd patients (Andrews and others 2012) found much less and others 2003; Moalosi and others 2003; Nganda and favorable results, indicating that using Xpert would cost others 2003) and ensuring effective cooperation with the US$5,100 per year of life saved compared to the next most private sector (Floyd, Arora, and others 2006; Pantoja, attractive strategy, which involved smear and culture. Floyd, and others 2009; Pantoja, Lönnroth, and others A third study from 2012 (Menzies and others 2012) exam- 2009). Although these studies demonstrate feasible and ined the cost-effectiveness of Xpert in five Southern cost-effective approaches to delivering TB treatment, the African countries and found a cost-effectiveness ratio of high costs of delivering relatively long antibiotic therapies around US$1,000 per DALY averted. Other studies have in poorly resourced health systems is a concern, as are the shown smaller-than-projected benefits of Xpert. high default rates in some settings (Kruk, Schwalbe, and Widely divergent results reflect the many challenges Aguiar 2008). These concerns are heightened by evidence of evaluating new diagnostic technologies and point to of the substantial economic and poverty impact of TB several areas for further investigation, including the treatment on patients (and households), with numerous extent to which frequent empirical treatment would studies finding that the multiple health service visits reduce the potential benefit of Xpert technology mea- required can have a severe impact on the economic sured against a counterfactual of diagnosis based strictly welfare of TB patients (Barter and others 2012). on the results from smear microscopy. As dramatic The current approach to the treatment of MDR TB scale-up of Xpert has been pursued in South Africa and presents a particular challenge from an economic per- other settings, other implementation challenges have spective. Several studies (Floyd and others 2012; Resch come to light, including the high overall budgetary and others 2006; Suarez and others 2002; Tupasi and impact of seeking high coverage of Xpert, the feasibility others 2006) have suggested that the treatment of MDR of deploying Xpert at the point of care, inability to link TB is cost-effective. A recent systematic review found that Xpert findings to health information systems, and limits the cost per DALY averted was lower than gross domestic on the potential benefit of faster diagnosis arising from product per capita in all 14 of the WHO subregions con- the failure to translate diagnostic improvements into sidered (Fitzpatrick and Floyd 2012). However, the abso- faster initiation of treatment (Lawn 2015). lute price of second-line drug regimens, even for LICs, New technologies that can be used at the point of care can run into the thousands of dollars, with a three- or are needed, and portable devices (for example, GeneXpert fourfold burden on total costs of the health system. For Omni) are under development that can be used as a example, in South Africa, treating MDR TB costs more point-of-care diagnostic test. Nevertheless, expanding than half of the total national budget for TB control their capacity for drug-sensitivity testing against the mul- (Schnippel and others 2013), with hospitalized treat- tiple drugs used in secondary and tertiary drug regimens ment costing more than US$15,000 per person treated remains a daunting challenge. (Pooran and others 2013; Schnippel and others 2013). However, these costs may be substantially reduced given the efforts to decentralize MDR TB treatment and Cost-Effectiveness of Developing New Drugs and care. Also potentially cost-effective, the cost of drug- Regimens susceptibility testing required to confirm a diagnosis of The public health case for investing in new TB drugs is MDR TB treatment can also be substantial (Acuna- clear (Zumla, Nahid, and Cole 2013), but the substantial Villaorduna and others 2008; Floyd and others 2012), cost of drug development remains a significant obstacle and, as highlighted in previous sections, culture-based to progress. Even if the efficiency of new drug develop- DST provides a substantial practical challenge in settings ment can be improved, new TB drug development is with limited laboratory capacity. Additionally, the eco- likely to remain a challenging investment decision for nomic burden of MDR TB on households may be sub- many years to come. stantially higher than the costs of first-line treatment 282 Major Infectious Diseases and is likely to be catastrophic (Ramma and others 2015; shortening treatment regimens. For example, examining survey of MDR TB patient costs in South Africa, unpub- first-line treatment, both Salomon and others (2006) lished data). and Abu-Raddad and others (2009) used a transmission In light of these challenges, increasing attention is model calibrated to the South-East Asia region and found being placed on developing low-cost models of MDR TB a substantial impact on incidence due to the introduction diagnosis and treatment for scale-up—for example, new of shortened regimens. Salomon and others (2006) found ambulatory models of care (Sinanovic and others 2015; that a noninferior two-month first-line regimen would Weiss and others 2014). In short, although the current prevent around 13–21 percent of all new TB cases approaches to the treatment of both drug-susceptible and 19–25 percent of TB deaths, depending on assump- TB and MDR TB are widely accepted to be cost-effective, tions made regarding the scale-up of current regimens the relatively high cost of treatment for both patients over an 18-year period. The study suggested that, if the and health systems, the cost of DST, the high levels of cost savings generated by treatment shortening were default, and the limited effectiveness of MDR TB treat- invested in TB case detection, two- or threefold reductions ment in low- and middle-income health systems make a in incidence might be possible. Abu-Raddad and others strong economic and public health case for investing in (2009) found that a four-month regimen with efficacy new TB drugs. similar to that of the standard of care would achieve up to In recent years, substantial investments have been a 10 percent reduction in incidence over 35 years and a made in clinical trials of new TB drugs and regimens two-month regimen with increased efficacy would achieve (Hoagland and others 2016; Zumla, Nahid, and Cole a 23 percent reduction in incidence over the same period 2013). Candidate drugs affect treatment efficacy or effec- of time. A more recent effort by Fofana and others tiveness and cost through different pathways, but the (2014) suggested a more modest, but still positive, impact, broad aim has been to improve one or more of the follow- estimating a 3 percent reduction in incidence from ing three dimensions: a four-month regimen and a 7 percent reduction from a two-month regimen over a 10-year period. • Shorten the duration of treatment; trialed examples for The analysis of the economic gains from treatment first-line regimens include four-month moxifloxacin- shortening has relied primarily on decision analytic based regimens (Gillespie and others 2014; Jindani models of patient cohorts. A study by Owens, Fofana, and others 2014) and the Bangladesh regimen and and Dowdy (2013) examined a hypothetical noninferior bedaquiline-based regimens for MDR TB (Diacon, first-line regimen and explored trade-offs between drug Donald, and others 2012). price, treatment duration, and health system treatment • Increase the efficacy of treatment, particularly for costs for a cohort of new TB patients. This study found MDR TB; trialed examples include delamanid (Gler, that a novel regimen with a four-month duration costing Skripconoka, and others 2012) and bedaquiline US$1 per day would at worst be highly cost-effective and (Diacon, Donald, and others 2012). at best be cost saving, depending on the current level of • Develop regimens that are effective in both drug- treatment costs. susceptible and MDR TB. This will require new drugs Ongoing work using an individual-based cohort for which drug resistance does not currently exist in model is exploring these trade-offs in specific country most populations. settings, using primary cost data from Bangladesh, Brazil, South Africa, and Tanzania (Zwerling and others 2016). In order to justify and support investment, various This study found that, at the cost of US$1 per day, a four- economic and modeling efforts have explored the poten- month noninferior first-line TB drug regimen would tial gains from improving these dimensions of TB treat- be cost saving in South Africa (reducing costs about ment in terms of cost-effectiveness, direct effect on 10 percent) and in Brazil (reducing costs 20 percent), treatment success, broader impact on transmission, and highly cost-effective in Tanzania (saving about US$120 patient and provider costs. In recent years, academic per DALY averted), but not cost-effective in Bangladesh. interest in both the investment in new drugs and these Even if new first-line drugs cost up to US$10 and US$58 analyses has increased. At the time of writing, consider- in South Africa and Brazil, respectively, using threshold able work is ongoing, with much new work expected on analysis, Trajman and others (2016) found that the new the horizon. TB regimen would be a cost-effective option compared to the standard regimen. This threshold price is US$0.97 in Shortening Treatment Regimens to Reduce Incidence Bangladesh and US$1.13 in Tanzania (Zwerling and To date, modeling analyses of TB treatment have focused others 2016). In all settings, the impact on health would on the population-level health gains from investing in be modest and dependent on current default rates; Tuberculosis 283 settings with higher health system and TB treatment Regimen of Anti-Tuberculosis Drugs for Patients with delivery costs would have the highest gain. Unfortunately, Multi-Drug-Resistant Tuberculosis) Trial (Nunn and the initial first-line regimens with potential regimen- others 2014). In 2016, the WHO recommended the new shortening effects coming out of trials in 2014 shortened regimen, which offers the promise of both (moxifloxacin-based regimens) failed to achieve adequate lower drug prices and lower health system costs, while levels of cure, so these gains, at the time of writing, being as effective as longer-course treatments (WHO remain hypothetical. 2016b). Economic analyses are ongoing. Zwerling and others (2016) also drew attention to the importance of patient costs, with the largest savings Developing New Drugs for Treating Drug-Susceptible achieved through reductions in the economic burden of and MDR TB as First-Line Therapies TB treatment on households. Patient cost savings ranged For the drugs discussed here, most of the work to date from US$175 (South Africa) to US$45 (Bangladesh), has explored the economic and health benefits of reduc- depending on the setting. ing the length of treatment and improving efficacy. However, new drugs or regimens that hold promise for Improving Therapeutic Efficacy in Patients with treating both drug-susceptible and rifampicin-resistant MDR TB TB also may have significant benefits. One example is Compared to drug-resistant TB treatment regimens, PaMZ (Diacon and others 2010; Diacon, Dawson, much less research has been conducted on the potential and others 2012), a trial currently on hold. This new cost-effectiveness and impact of new MDR TB regimens. treatment regimen was assumed to have a dual benefit: a In the last few years, two new MDR TB drugs (bedaqui- shortened first-line treatment for drug-susceptible line and delamanid) have come up for regulatory authority patients (four months) with an efficacy noninferior to and programmatic approval by the WHO. As part of the the current standard treatment and a shortened second- process, an exploratory cost-effectiveness analysis was line treatment for patients with rifampicin-resistant TB conducted using a decision analytic model of a cohort of (six months) with an efficacy noninferior to first-line new MDR TB cases (Vassall 2013). These analyses found treatment for drug-susceptible TB patients. In this case, both drugs to be potentially cost-effective, given their introduction of the new regimen could result in substan- impact on efficacy. However, uncertainty around bedaqui- tial cost savings (at least 35 percent reduction) from a line’s impact on mortality and, in the case of delamanid, societal perspective. The mean cost per presumptive TB the lack of randomization used when assessing long-term patient was reduced by US$23 (28 percent) for health outcomes, combined with the potential cardiotoxic effects, service treatment-related costs and US$42 (42 percent) have raised concerns. The DALYs averted varied by for patient treatment-related costs in South Africa. setting, with countries already having good outcomes When the introduction of PaMZ in a cohort of only benefiting less. However, increased efficacy, even without rifampicin-resistant patients was modeled, both the treatment shortening, also reduced costs in some cases, as reduction in costs and the gain in effect were greater. The the need for MDR TB retreatment and management of clinical safety and effectiveness of this new regimen chronic cases was reduced. The results were less certain for remains to be determined, but the economic modeling LICs; because the potential benefits of increased efficiency can be applied to any new regimen effective for treating for transmission were not included, no definitive conclu- both drug-resistant and drug-susceptible TB. sion could be reached for these settings. This model can be extended to explore the gains from In the case of bedaquiline, the impact on cost- a drug with the following optimal characteristics in all effectiveness of a shortened MDR TB regimen was also dimensions: examined, given that the trial results suggested that time to sustained sputum conversion may be reduced (Vassall 1. Short duration (two weeks maximum shortening) 2013). Examining a reduction in treatment of two 2. High efficacy (95 percent cure rate for drug-susceptible months, the cost-effectiveness analysis found cost sav- TB, 85 percent for rifampicin-resistant TB) ings at current drug prices. However, the extent of cost 3. Ability to treat all forms of TB (that is, with no circu- savings depended on the duration of hospitalization lating resistance and no need for initial DST) during treatment. The benefits in terms of DALYS 4. Drug price set at US$5 per day. averted were less clear due to the trade-off between the reduction in default and cure rates. Further trials, with Using the same model, Zwerling and others (2016) an integrated economic analysis, are ongoing that found that such a drug would be cost saving in a setting test the use of bedaquiline as part of a nine-month MDR such as South Africa. Even with a comparatively high TB regimen, the STREAM (Standardized Treatment daily cost compared to current treatment, these authors 284 Major Infectious Diseases estimated a potential reduction of 51 percent of total TB Costs to Patients of TB Treatment diagnostic and treatment costs when modeling a cohort For diagnosis and many months of treatment for TB, of 10,000 presumptive TB patients from a societal out-of-pocket costs can be catastrophic for patients and perspective. This is primarily a gain on the patient side, their families. Data on patient costs in LMICs are limited including a reduction in the direct costs of treatment and incomplete. In a systematic analysis of 11 publica- and a more rapid return to full productivity. In this high- tions on patient costs in eight countries in Africa, HIV/AIDS-prevalence setting, there is a slight increase in Ukwaja and others (2012) estimated that mean patient ART-related costs (from the health service perspective) prediagnostic costs varied between US$36 and US$196, of 4 percent and an increase in DALYs averted of corresponding to 10.4 and 35.0 percent of their annual 6.5 percent, primarily from a reduction in defaults and income, respectively. Average patient treatment costs an increase in cure rates for persons with MDR TB. ranged between US$3 and US$662, corresponding to In summary, while much of the evidence on the poten- 0.2–30.0 percent of their annual income. Prediagnostic tial economic and health benefits of investment in new TB household costs accounted for 13–18.8 percent of drugs is based on models, emerging findings suggest that patients’ annual household income, while total house- reducing the duration, improving the efficacy, and expand- hold treatment costs ranged between US$26 and US$662, ing the range of use of TB regimens may have the potential accounting for 2.9–9.3 percent of annual household for substantial economic and public health gains in most income. Consequently, 18 percent to 61 percent of settings (table 11.7). In a field with no new drugs for the patients received financial assistance from outside their last half century up until the past five years, the biomedical household to cope with the cost of TB care. Patient costs and drug development challenges cannot be underesti- in South Africa for diagnosis and treatment of MDR TB mated, but new drugs are now being trialed, and much were even more expensive: only 3 percent of patients work refining and validating these nascent predictions is were still employed, and disability grants were the expected to emerge in the coming years. primary source of income for 44 percent of patients Table 11.7 Population Impact, Patient Impact, Cost, and Cost-Effectiveness of New Tuberculosis Drugs Study and Time Impact goal Regimen Setting horizon (reduction) Cost Cost-effectiveness Shortening treatment duration Salomon and First-line, 2 month South-East Asia 2012–30: 13–21% incidence; — — others 2006 18 years 19–25% mortality Abu-Raddad First-line, 4 month, South-East Asia 2015–50: 10% incidence — — and others noninferior to 35 years 2009 standard First-line, 2 month, 23% incidence 90% efficacy in drug- resistant cases First-line, 10 days, 27% incidence 90% efficacy in drug- resistant cases Fofana and First-line, 4 month, Global, 10 years 1.9% incidence; — — others 2014 noninferior to standard nonspecific 3.5% mortality First-line, 2 month, 4.3% incidence; noninferior to standard 7.5% mortality First-line, 2 weeks, 6.7% incidence; noninferior to standard 13.1% mortality table continues next page Tuberculosis 285 Table 11.7 Population Impact, Patient Impact, Cost, and Cost-Effectiveness of New Tuberculosis Drugs (continued) Study and Time Impact goal Regimen Setting horizon (reduction) Cost Cost-effectiveness Owens, First-line, 4 month, Global, Cohort 7.9 DALYs averted Health service Highly cost-effective to cost Fofana, and noninferior to standard, nonspecific lifetime (100 cohort) cost: cost saving to saving, depending on current Dowdy 2013 US$1 per day US$5,900 (2012) treatment costs First-line, 2 month, 14.6 DALYs averted Health service Cost saving noninferior to standard, (100 cohort) cost: cost saving to US$5 per day US$20,200 (2012) Trajman and First-line, 4 month, South Africa Cohort Equivalent effect 10% societal cost Cost saving others 2016 noninferior to standard, lifetime reduction US$1 per day Brazil 20% societal cost Cost saving reduction Bangladesh 3.5% cost increase Not cost-effective (ICER: 997) Tanzania Cost neutral Cost-effective (ICER: 120) Increasing efficacy for MDR TB treatment Vassall 2013 Bedaquiline China, Estonia, Cohort Varied by setting Varied by setting from Varied from US$202 to Nepal, Russian lifetime from 0.94 to 5.27 US$823 to US$2,930, US$2,042 per DALY averted; Federation, Peru, incremental DALYs, depending on ability likely to be cost-effective in all Philippines depending on to reduce retreatment settings except low income, current treatment costs but ICER highly uncertain, success depending on assumptions regarding trial results Vassall 2013 Delamanid China, Estonia, Cohort Varied by setting Varied by setting from Varied from US$501 to Nepal, Russian lifetime from 0.94 to 1.65 US$757 to US$2,548 US$1,654 per DALY averted; Federation, Peru, incremental DALYs likely to be cost-effective in all the Philippines depending on settings except low income, current treatment but ICER highly uncertain, success depending on assumptions regarding trial results Developing regimens that are effective in both drug-susceptible and MDR TB Gomez and First- (4 month) and South Africa Cohort Equivalent effect 35% societal cost Cost saving others 2016 second-line (6 month), lifetime estimated, reduction PaMZ, noninferior to presumptive standard for first-line TB cohort and equivalent to 16% increase in 60% societal cost Cost saving first-line treatment of DALYs averted, reduction drug-sensitive TB for MDR cohort MDR patients, less than US$1 per day Future developments Gomez and Optimal: first- and South Africa Cohort 6.5% increase in 51% societal cost Cost saving others 2016 second-line, 2 weeks, lifetime DALYs averted reduction; 58.1% high efficacy, US$5 per patient costs reduction; day, no extra DST 56.5% treatment costs reduction; no change in diagnosis costs Note: DALY = disability-adjusted life year; ICER = incremental cost-effectiveness ratio; MDR = multidrug resistant; TB = tuberculosis; PaMZ = pretomanid/moxifloxacin/pyrazinamide; DST = drug-sensitivity testing; — = not available. 286 Major Infectious Diseases (Ramma and others 2015). Many of the patients reported increasingly common, and thus the costs may decrease. having no source of income before (56 percent) and Due to the pervasiveness of Mtb infection in high-burden during (47 percent) treatment. These costs were likely settings, TB vaccines targeted at adults may be used in catastrophic for many patients. mass campaigns to prevent disease. There are few prece- dents for such campaigns, and the levels of coverage that could be achievable are open to debate. Similarly, the lack Modeling the Impact of New Vaccines of acceptability of frequent mass campaigns could limit The last decade has seen a significant increase in the level the potential impact of a TB vaccine for adults. Finally, it is of investment in TB vaccine development: more than possible that vaccines could be targeted to patients being 14 TB vaccine candidates have been tested in 50 human treated with TB or MDR TB either to accelerate treatment trials, with funding of more than US$600 million times or to prevent recurrence after drug treatment. (Jiménez-Levi 2012). Despite the significant costs of Knight and others (2014) have explored a wide range of vaccine development (Brennan and Thole 2012), the vaccine profiles to understand what type of vaccine profile, complex task of selecting vaccine candidates must now targeted at what age group, would have the biggest impact occur. Therefore, there is an effort to determine globally on TB incidence in LMICs (Knight and others 2014). acceptable criteria for differentiating and subsequently Vaccine profiles were defined by both efficacy and dura- maintaining the most promising candidates in the vac- tion of protection, with a range from 40 to 80 percent and cine pipeline (Brennan and Thole 2012). from five years to lifetime, respectively. The vaccine was Mathematical modeling can be used to differentiate assumed to be introduced in 2024. A comparison was then candidates on the basis of the potential impact that made between targeting infants and targeting adolescents different vaccine characteristics targeted at different pop- or adults in mass campaigns. A new TB vaccine would ulation groups may have on the future TB burden. For likely be used as a booster to BCG, due to the broad use of example, modeling the impact of vaccines targeted at BCG and its efficacy in infants. The vaccine was assumed uninfected infants in South-East Asia has suggested that to prevent active disease in both uninfected and latently a novel vaccine introduced in 2015 could avert more than infected individuals, with 40 percent less efficacy in per- 40 percent of the TB burden by 2050 (Abu-Raddad and sons with HIV/AIDS. Vaccine “take” was modeled with an others 2009) and that, to have the most rapid impact on exact duration of protection. Each country was modeled TB burden, such a vaccine would need to be efficacious in separately, with calibration methods used to capture both both uninfected and latently infected individuals (Dye uncertainty in natural history parameters and the data on and Williams 2008). Linking such work to the economics TB burden (WHO 2013b). Vaccine coverage was taken of TB vaccines has shown that different TB vaccine pro- from similar mass campaigns in each country, such as for files can also be cost-effective (BIO Ventures for Global rubella, and were timed to occur with a frequency of every Health 2006; Ditkowsky and Schwartzman 2014; Tseng 10 years or the duration of protection (whichever was and others 2011). However, uncertainty remains about shorter) from 2024 onward. Cost-effectiveness was defined how various vaccine characteristics (efficacy, duration of as cost per DALY averted compared against gross national protection) and targeting strategies (age, HIV/AIDS income per capita from a health sector perspective, with status) may combine to maximize impact in countries tiered vaccine pricing by income group. with the largest TB burden. By estimating the burden of TB in all LMICs for The underlying economics of TB vaccines are also which data were available, the model was able to predict unclear. For example, depending on the vaccine profile, the that a vaccine targeted at adolescents or adults would likely price of both vaccine delivery and dose remains have a far greater impact on TB burden before 2050 than uncertain. TB vaccines for infants could be incorporated one targeted at infants (figure 11.10) (Knight and others into the standard infant vaccination program (DTP3 2014). This is due to the differential burden of disease [diphtheria-tetanus-pertussis]), and cost estimates are between these two age groups, with infants suffering straightforward to generate. However, the economics of from greater levels of extrapulmonary TB and therefore targeting adolescents and older populations are largely contributing less to transmission (Styblo 1991). This unknown, and only recently are these populations being conclusion remained valid over the 2024–50 period, considered as a new platform for other vaccines, such as even when considering lifetime duration of protection. human papillomavirus (Sinanovic and others 2009) or Knight and others (2014) reported that vaccines tar- measles booster. In the future, with these and other poten- geted at adolescents or adults with 10-year durations of tial vaccines (for example, measles, human papillomavirus, protection and 60 percent efficacy could be cost-effective or a future HIV/AIDS vaccine) for adolescents and adults, in LICs at US$149 (95 percent range cost saving of the targeting of adolescents via schools is likely to become US$387) per DALY averted at a cost of US$1 per dose. Tuberculosis 287 Figure 11.10 Model Calibration and Vaccine Impact in Low-Income Countries a. b. c. 1,500,000 300 90 (deaths per 100,000 population) (cases per 100,000 population) Popualation size 1,000,000 TB incidence TB mortality 200 60 500,000 100 30 0 0 0 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050 Year Year Year d. e. 100 100 (cases per 100,000 population) (cases per 100,000 population) TB incidence TB incidence 50 50 0 0 2020 2030 2040 2050 2020 2030 2040 2050 Year Year Efficacy = 40% Duration = 5 years 60% 10 years 80% Lifelong Note: Model calibration (panels a–c) and vaccine impact (panels d–e) in low-income countries. Panels a–c show median (solid dark blue line) and 95% range (blue cloud) of model fits to data. (Panel a) Human population size (per 1,000) in low-income countries, 2000–50. (Panel b) Tuberculosis (TB) incidence: cases per 100,000 population per year, 2000–50 (Panel c) TB mortality: deaths per 100,000 population per year, 2000–50. (Panels d–e) TB incidence (cases per 100,000 population per year), 2000–50, with median model output (black line) and vaccine profile impact: characteristics of efficacy (color) and duration of protection (line type) for vaccines targeted at infants (panel d) or at adolescent or adults (panel e). A vaccine targeted at infants (panel d) has a smaller impact on TB disease incidence than one targeted at adolescents or adults (panel e). The “waves” within the adolescent or adult incidence figure (panel e) are due to the impact of mass campaigns. 288 Major Infectious Diseases The same profile targeted at infants would avert only 0.89 within the poorer parts of the population: TB has a four (0.42–1.58) million TB cases while averting 17 (11–24) times greater incidence among persons in lower socio- million TB cases in adolescents or adults and would not be economic groups, and incidence is greater in rural than considered cost-effective at US$1,692 (US$634–US$4,603) urban areas (Muniyandi and others 2007). Private health per DALY averted. In threshold analysis, the price per dose expenditures also constitute a large majority of India’s at which the vaccine profile would be cost-effective was total health expenditures, and most TB patients consult determined. For example, a vaccine with 10-year duration private practitioners for their first visit, resulting in sub- of protection and 60 percent efficacy targeted at adoles- stantial OOP spending on TB (Satyanarayana and others cents or adults could be priced at up to US$20 in 2011; Uplekar and others 1998; Uplekar, Pathania, and upper-middle-income countries and still be considered Raviglione 2001). In India, as in many countries, OOP cost-effective. This reflects the large number of cases that medical costs are a leading cause of impoverishment. could be averted by targeting the age group in which most Kruk, Goldmann, and Galeo (2009) and Sengupta and cases of disease occur and in which most sources of trans- Nundy (2005) found, for example, that about 40 percent mission are found. Vaccines with short duration (five of Indian households borrowed money or sold assets to years) and low efficacy (40 percent) were also found to be pay for health care. In light of such findings, the govern- cost-effective if targeted at adolescents and adults. ment is increasingly assuming responsibility for financ- A recent review pointed out that funding for TB vac- ing TB treatment (Jha and Laxminarayan 2009). cine development significantly lacks the same support Universal public finance (UPF) is when government provided to other vaccine development efforts finances an intervention irrespective of who is receiving (Manjelievskaia and others 2016). For example, funding it. For any given health intervention, UPF entails conse- to develop an Ebola vaccine ramped up quickly follow- quences in multiple domains. First, UPF increases inter- ing the 2013–14 epidemic in Guinea, Liberia, and Sierra vention uptake. Second, it eliminates the need for private Leone, an outbreak that killed approximately 11,310 expenditures. Finally, it provides financial risk protec- persons. The global response to the outbreak was enor- tion (insurance) by covering catastrophic expenditures mous; the U.S. government alone appropriated more that would otherwise throw households into poverty. than US$5.4 billion for the Ebola emergency response in This section reports on these three consequences of UPF 2015, of which a large portion was directed to research, using extended cost-effectiveness analysis (ECEA) find- including developing Ebola vaccines. In comparison, ings on DOTs expansion in India (Verguet, Laxminarayan, funding to develop a TB vaccine received only US$85 and Jamison 2015). million to US$90 million in 2015 (based on prior years’ Because of the importance of OOP costs, interna- funding numbers), despite an estimated 29,000 people tional agencies have supported the use of health sector dying of TB per week (based on 2014 mortality esti- policies to attenuate health-related financial risks (WHO mates) and little sign of meaningful diminution in the 1999, 2010; World Bank 1993). Despite the attention global spread of Mtb and the continuing spread of MDR paid to its significant potential as part of broader social and XDR TB (Frick, Henry, and Lessem 2016; TAG 2016). insurance, UPF in practice covers few interventions in In conclusion, vaccines have an enormous potential most LMICs, with little consensus on what to cover in to reduce the incidence and prevalence of TB if they are highly resource-constrained environments. In India, targeted at adolescents or adults, suggesting that UPF has typically financed condition-specific programs increased investments in candidate vaccines targeting (for example, against leprosy, HIV/AIDS, and cataract this group are warranted. Vaccines are also likely to be blindness) or, more recently, secondary and tertiary care highly cost-effective, even if only having relatively mod- insurance such as the Rashtriya Swasthya Bima Yojana est clinical effectiveness. Given the age-dependent period and the Arogyasri (in Andhra Pradesh) programs. These of risk, special consideration should be given to develop- insurance programs are thought to provide significant ing candidate vaccines providing long-term protection. financial protection since they defray the high OOP costs associated with hospitalizations. Currently, about 70 percent of TB cases receive EXTENDED COST-EFFECTIVENESS ANALYSIS DOTS; only about half of these services are obtained for OF UNIVERSAL PUBLIC FINANCING OF TB free in the public sector. People do not always obtain TB treatment at the public hospital level for various reasons, TREATMENT including transportation cost and waiting time. Most Tuberculosis causes approximately 28 million active lower-income people prefer to see a private physician infections and 480,000 deaths in India annually (WHO after working hours than to take a day off work without 2016a). The disease burden is concentrated largely pay to visit a public hospital (Kumar and Kumar 1997). Tuberculosis 289 Table 11.8 Public Finance of Tuberculosis Treatment to Table 11.9 Borrowing to Finance Tuberculosis 90 Percent Coverage in India, by Income Quintile Treatment in India, by Income Quintile per million population per million population Income Quintile Income Quintile Outcome Total I II III IV V Outcome Total I II III IV V TB deaths averted 80 40 25 12 3 0 TB deaths averted 29 13 10 5 1 0 Private expenditures Private expenditures crowded outa 29 6 6 7 6 4 crowded outa −25 −12 −8 −4 −1 0 Insurance valuea 9 5 2 1 1 0 Insurance valuea 0 0 0 0 0 0 Note: TB = tuberculosis. Note: TB = tuberculosis. a. Figures expressed in 2011 US$, thousands. a. Figures expressed in 2011 US$, thousands. Yet, privately purchased TB treatment is often ineffective earnings can be substantial (Ananthakrishnan and others because private doctors do not necessarily provide the 2012), which would—in the absence of other forms of standard regimen with proven efficiency (Satyanarayana social insurance—increase the insurance benefits pro- and others 2011; Udwadia, Pinto, and Uplekar 2010; vided by UPF. Uplekar and Shepard 1991). A potential virtue of UPF is A limitation of focusing on the financial cost of treat- to eliminate low-quality treatment in the private sector, ment is that individuals do not get care for other reasons enabling the uptake of higher-quality treatment and as well, including lack of information about TB and increasing technical efficiency. However, this requires a its treatment. Primary health centers in India may be well-managed public sector program, something widely difficult to reach due to poor travel conditions (Muniyandi achieved only in parts of India. and others 2006), be overcrowded, and not respect the We estimated the results of expanding DOTS through dignity of patients. Health services may not always be UPF to cover 90 percent for an Indian population of 1 available, even after informational and financial barriers million (table 11.8). The total number of lives saved have been removed: increasing coverage may thus not be would be about 80 per 1 million per year, and the health feasible, and the extent of subsequent health gains from benefits would be concentrated among the bottom two UPF may be significantly reduced in the absence of income quintiles (80 percent), as TB has a higher inci- supply-side intervention. If demand for TB treatment for dence among these socioeconomic groups. The total any of these reasons does not match disease incidence, number of private expenditures averted by the program the level of coverage achievable will be constrained. would be about US$30,000 per 1 million population per Yet, using ECEA, Verguet, Laxminarayan, and Jamison year, and the bottom two quintiles would benefit from (2015) demonstrated that the potential benefits of UPF about 40 percent of the private expenditures averted. for health and financial risk protection would accrue The total financial protection provided for this popula- primarily to the poor. Reductions in OOP expenditures tion sample (measured as a money-metric value of would also benefit the poor because UPF crowds out insurance) would be about US$10,000, 80 percent of private financing of the inferior treatments frequently which would accrue to the bottom two quintiles. The purchased by persons with income constraints. Lowering total (incremental) treatment costs incurred by UPF for the costs of borrowing for the poor could potentially the sample would be about US$65,000. achieve some of the health gains of UPF, but at the cost This analysis illustrates how UPF can be used to of leaving the poor more deeply in debt (table 11.9). improve financial protection and technical efficiency by eliminating the purchase of lower-quality treatments. It is SUMMARY AND RECOMMENDATIONS only a limited take on greater possibilities. For example, a detailed assessment could provide more comprehensive While much progress has been made over the past several estimates of TB costs (for example, households’ transpor- decades, particularly in reducing mortality, TB remains a tation costs, earnings, and productivity impacts). The major infectious disease, whose severity is exacerbated by focus here is on the OOP cost of treatment and excludes the growing problem of multidrug-resistant TB, exten- the cost of earnings reduced by the disease. For example, sively drug-resistant TB, and co-infection with HIV/AIDS. the number of working days lost due to TB can range Progress in TB control requires early and accurate case from 30 to 90 in South India (Muniyandi and others 2006; detection, rapid commencement of and adherence to Muniyandi and others 2008). Indirect costs from lost effective treatment, and prevention of transmission. 290 Major Infectious Diseases The great challenge is to reduce transmission by identify- TB, represents a serious threat to health personnel in ing and treating patients with TB who are asymptomatic, high-burden countries. Even with high patient treatment unaware of their status, or unable to gain access to compliance, in these contexts, DOTS as commonly imple- treatment. mented is not reducing the transmission and incidence of This chapter offers four overarching recommenda- disease sufficiently rapidly. As this chapter discusses, sub- tions: (1) strengthen the current WHO End TB Strategy stantial numbers of individuals with TB have tubercle to emphasize active case finding in targeted high-burden bacilli in their sputa and yet are asymptomatic and not countries; (2) strengthen health systems in those coun- likely to report to a health facility and thus be detected by tries, particularly emphasizing community-based care; passive case finding. (3) strengthen information systems; and (4) invest far In this context, the WHO has revised and promulgated more in research to develop the tools necessary to con- the End TB Strategy, which emphasizes patient-centered trol TB. Fundamentally, it is necessary to revise the cur- care, treatment for all patients with drug-susceptible and rent global TB control strategies in most high-burden drug-resistant TB, increased community engagement, and, countries and to make significant new investments in for the first time, an explicit focus on research. Consistent health systems and research. with the End TB Strategy (WHO 2015a), this chapter sug- gests that a stratified approach with improved targeting in high-burden countries will be necessary in these contexts. Rethinking and Revising the DOTS Strategy Such a diversified and better-targeted strategy, should seek Since the WHO designated tuberculosis as a global emer- to accomplish the following: gency in 1993, the DOTS strategy has been the mainstay of TB control worldwide. It is based on the premise that 1. Identify high-transmission countries and hot spots patients with TB will be sufficiently ill to seek care from within countries where targeted efforts can be more the health system and, if appropriately treated, be cured effective and cost-effective. and that transmission will be reduced. Since 1994, the 2. Increase the capacity for surveillance. DOTS strategy has significantly improved the diagnosis 3. Strengthen early TB detection and diagnosis by active and treatment of individuals identified with tuberculosis case finding in certain countries and populations. in all countries and reduced mortality. The prevalence 4. Provide rapid diagnosis and enable more rapid initi- and treatment of drug-susceptible TB is a “best buy” at ation and better maintenance of treatment for both US$100–US$500 per case cured. Treatment of MDR TB drug-susceptible and multidrug-resistant TB. remains limited and more expensive. Yet the decline of 5. Expand preventive therapy of the contacts of TB TB incidence globally, particularly in the 22 highest- patients, children, and HIV-positive individuals. burden countries, has lagged the declines in HIV/AIDS 6. Combine these strategies with an approach rooted in and malaria and is far below the MDG targets set by the community-based delivery of TB services and support, Stop TB Partnership to reduce TB incidence, prevalence, wherever possible. and mortality by 50 percent compared with a baseline of 7. Improve the drug supply chain to improve access 1990. With TB exceeding HIV/AIDS and malaria as the to TB treatments that have very small markets, and largest cause of death from any single infectious disease improve information technology to enable more and drug resistance increasing, it is critical to reconsider effective control. how to control the epidemic more effectively. 8. Significantly increase resources for research on devel- DOTS has the great advantage of being a single, unify- oping new diagnostic technologies, development and ing global strategy that, in principle, can be applied to all testing of new drug regimens, and new vaccines to patients in all countries. Yet, while effective in countries prevent TB. with strong health systems, DOTS is failing to bend the incidence curve rapidly, especially in LICs with high TB All of these recommendations, while at some variance burdens and weak health systems. Management of TB has with previous DOTS protocols, are fully consistent with been particularly challenging in LMICs with high inci- the new WHO End TB strategic goals (WHO 2015a). dence and prevalence of co-infection with HIV/AIDS, The new Stop TB Partnership’s Zero TB Initiative seeks which makes treatment more difficult and relapse or to apply these recommendations to demonstrate in a reinfection more common. Of particular concern is the small number of high-burden cities that a comprehen- limited impact of DOTS on the effective control of MDR sive program of active case finding, effective treatment, TB, which requires complex diagnostic testing and is and prevention in households can be effective in limiting enormously expensive (US$5,000–US$10,000 per case transmission and reducing the incidence of TB (Stop TB cured) and only partially effective. TB, especially MDR Partnership 2016). Tuberculosis 291 Expanding Active Case Finding and others 2011; Chang and others 2012), do not require In high-burden, low-income countries, particularly those highly skilled health workers (Rachow and others 2011), with a high incidence and prevalence of TB and HIV/AIDS and have been demonstrated to be cost-effective (Vassall co-infection, earlier and more comprehensive case finding and others 2011). However, their rapid introduction has and treatment are required. Passive case finding and been hindered by weak health systems, specifically weak screening of populations for classic symptoms clearly fail supply chain management systems, weak information to detect a major proportion of existing TB cases, which systems linking diagnostic data to the site of treatment, are asymptomatic or unknown to the health system— and the need to address false-positive indications of representing up to one-third or 3 million cases—leading to rifampicin resistance (Dowdy and others 2011; Kirwan, continued transmission of TB infection. In low-burden Cárdenas, and Gilman 2012; Lawn and others 2011; Scott countries where DOTS is effective in reducing incidence, and others 2011; Trébucq and others 2011). prevalence, and mortality, the current approach should be Increased numbers of better trained and motivated expanded, and investments in active case finding are not health workers at multiple levels, better information sys- likely to be cost-effective. But, in high-burden countries, tems, and functioning logistics and supply chains are criti- targeted introduction of active case finding with modern cal if TB is to be effectively controlled. TB transmission can rapid and point-of-care diagnostic tests and mobile be prevented through investments in infection control in X-radiography for screening populations could signifi- clinics and hospitals; preventive therapy for household and cantly improve early diagnosis and early institution of HIV-positive contacts; better-designed housing for low- treatment and dramatically reduce transmission. As this income populations; and better nutrition for at-risk chapter discusses, TB treatment regimens are lengthy and populations. Efforts like these require technical and finan- not without drug-related adverse events, which, in weak cial support, not only in TB programs per se, but also in health systems, create challenges to maintaining consistent complementary activities that support the delivery of TB adherence to recommended drugs (Schaaf and others services. Without improving health systems, extending the 2009; Verma and others 2004). Health systems that can scope and targeting of TB programs, and providing addi- provide patient support, including mechanisms to incen- tional technical and financial support, it is questionable tivize treatment completion, can bolster the effectiveness how significantly and rapidly high-burden countries can of active case finding. improve TB control. Expanding Community-Based Care Strengthening Health Systems While in many countries, TB treatment has traditionally With drug-resistant TB now a global crisis, early diag- been hospital-based, in an increasing number of set- nosis and effective treatment, either through improved tings community-based treatment has been shown to access to bacterial culture and drug-sensitivity testing be as clinically effective and significantly more cost- (Dowdy and others 2008; Uys and others 2009) or effective, even when compared with clinic-based through rapid molecular diagnostics (Lin and others management of TB. Efforts should be undertaken 2012), are essential to reduce the burden of drug- to transition health systems from hospital-based to resistant TB in high-prevalence settings. Modeling community-based care in countries where doing so is studies suggest that these interventions could be highly feasible. Community-based care and treatment are cost-effective, because they provide not only potentially effective and may offset some of the increases in financ- life-saving care but also prevent further spread of ing required for active case finding. It will be important, drug-resistant disease (Menzies and others 2012). in countries with large private health care providers, to Modeling studies also suggest that infection-control develop public-private partnerships to engage the pri- interventions can be very cost-effective (Basu and others vate sector in offering the best possible diagnostic and 2007), especially when aimed at health care workers treatment protocols. Strengthening health systems to and the patients responsible for most transmission support more effective community-based TB care— (Andersson 2006; Woolhouse and others 1997). including investing in community-based health work- Although some new diagnostics and care delivery ers, information systems, and supply chains—can have models have emerged, the uptake of new tools and inno- significant impact, not only on TB, but also on other vations for TB control has been suboptimal in weak health infectious and chronic diseases, including HIV/AIDS systems (Cobelens and others 2012). New tools such as the and noncommunicable diseases such as diabetes and Xpert MTB/RIF assay enable rapid and effective diagnosis cardiovascular disease, which similarly require long- of TB and MDR TB (Boehme and others 2010; Boehme term or continuous treatment. 292 Major Infectious Diseases Strengthening Research and Development incidence is not declining at a rate required to bring this The current tools for combatting TB are woefully inade- epidemic under control. This chapter urges that the tra- quate, yet funding for TB research lags that for HIV/ ditional strategy to control TB most commonly imple- AIDS and malaria. Reducing the burden of TB and MDR mented in many countries, which has been successful in TB will require greater funding for more intensive low-burden countries of the Americas and Europe, research into new approaches to point-of-care diagnosis, needs to evolve into a more stratified and targeted shorter and more effective treatment regimens for both approach in order to meet the needs of high-burden TB and MDR TB, and better service delivery. But the populations where it has not been effective in controlling current drug pipeline is thin. There is an urgent need for transmission and reducing incidence of TB. The WHO’s new drugs and regimens that will be cost-effective and End TB Strategy now emphasizes differences in the epi- affordable. Of particular urgency is the need to develop demic in different countries and the need for research to new multidrug regimens that shorten the duration of develop new tools (WHO 2015a, 2015c). The recom- treatment, increase the number of cures of drug- mendations of this chapter are fully consistent with the susceptible TB, and, ideally, increase the number of cures WHO’s End TB Strategy. of both drug-susceptible and drug-resistant TB. A small This chapter argues that new investments are needed number of new drugs appear promising in this regard, to make health systems more responsive and effective by but require further clinical evaluation. New regimens providing greater access to improved technologies for capable of effectively treating both drug-susceptible and rapid diagnosis and to drug-sensitivity testing. It advo- drug-resistant TB would reduce the need for expensive cates for introducing new modalities, such as active case drug-sensitivity testing and could be highly cost-effective. finding and community-based care in high-burden set- Several molecular tools are being developed that could tings. Finally, it emphasizes the urgent need for greater allow prediction of risk for patients with latent infection investments in research to develop new tools for diagno- to progress to active disease and, possibly, of when treated sis, drug regimens for treatment, and vaccines for pre- cases are truly cured. This approach could enable preven- vention. Such investments will be costly—and more tive treatment of individuals at high risk of progression funding will be required to extend current efforts and to disease. The same molecular tools may be able to iden- enable the new approaches recommended here—but tify host gene signatures that would identify protective ultimately highly cost-effective for both individuals and immune responses that could become biomarkers for countries. Only with new thinking and new approaches predicting the efficacy of new vaccines. will it be possible to transform TB control to a level that Developing more effective vaccines to prevent TB can achieve the ultimate goal of eliminating TB as a infection, disease, relapse, or reinfection will be essential, global public health problem. To reduce significantly the since even a modestly successful vaccine would be highly largest cause of death from infectious disease in the cost-effective. BCG remains the most widely used vac- world and to improve the lives of people who suffer from cine in the world; while its ability to prevent severe child- it, greater financial investments will be necessary and hood consequences of TB is cost-effective, its impact on indeed justified. preventing disease in adults is questionable at best. Preventing infection and disease is the major goal of NOTES vaccine research, and, given the challenges to case detec- tion, drug resistance, and cure, it remains uncertain World Bank Income Classifications as of July 2014 are as follows, whether TB can be eliminated as a global public health based on estimates of gross national income (GNI) per capita problem without an effective preventive vaccine. for 2013: Drug, device, and vaccine companies are essential to • Low-income countries (LICs) = US$1,045 or less developing the new tools required for TB, but they have • Middle-income countries (MICs) are subdivided: few incentives to invest in a disease that occurs primarily (a) lower-middle-income = US$1,046 to US$4,125 in resource-poor countries where the returns on invest- (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 ments in long-term development and trials will be few, • High-income countries (HICs) = US$12,746 or more. if any. Public-private collaborations could make a major difference to developing new technologies for TB. 1. For example, http://reference.medscape.com/drug -interactionchecker or http://www.hiv-druginteractions.org. 2. For the WHO recommendations, see http://www.who.int The Bottom Line /hiv/topics/tb/art_hivpatients/en. TB remains the largest cause of death from an infectious 3. Described at http://www1.nyc.gov/assets/doh/downloads disease, and, in contrast to HIV/AIDS and malaria, the /pdf/tb/tb2015.pdf. Tuberculosis 293 4. For example, OpenBoxes (https://openboxes.com) and Andersson, D. I. 2006. “The Biological Cost of Mutational OpenLMIS (https://www.villagereach.org). Antibiotic Resistance: Any Practical Conclusions?” Current 5. Sproxil (https://www.sproxil.com). Opinion in Microbiology 9 (5): 461–65. Andrews, J. R., F. Noubary, R. P. Walensky, R. Cerda, E. Losina, and others. 2012. “Risk of Progression to Active Tuberculosis REFERENCES Following Reinfection with Mycobacterium tuberculosis.” Clinical Infectious Diseases 54 (6): 784–91. Abel L, J. El-Baghdadi, A. A. Bousfiha, J. L. Casanova, and Ansa, G. A., J. D. Walley, K. Siddiqi, and X. Wei. 2012. “Assessing E. Schurr. 2014. “Human Genetics of Tuberculosis: A Long the Impact of TB/HIV Services Integration on TB Treatment and Winding Road.” Archive of Philosophical Transitions of the Outcomes and Their Relevance in TB/HIV Monitoring in Royal Society of London 12: 369 (1645): 20130428. Ghana.” Infectious Diseases of Poverty 1 (December 24): 13. Abimbola, T. O., B. J. Marston, A. A. Date, J. M. Blandford, Arinaminpathy, N., T. Cordier-Lassalle, K. Lunte, and C. Dye. N. Sangrujee, and others. 2012. “Cost-Effectiveness of 2015. “The Global Drug Facility as an Intervention in Tuberculosis Diagnostic Strategies to Reduce Early Mortality the Market for Tuberculosis Drugs.” Bulletin of the World among Persons with Advanced HIV Infection Initiating Health Organization 93 (4): 237–48A. Antiretroviral Therapy.” Journal of Acquired Immune Deficiency Atun, R. 2012. “Health Systems, Systems Thinking and Syndromes 60 (1): e1–7. Innovation.” Health Policy and Planning 27 (Suppl 4): iv4–8. Abu-Raddad, L. J., L. Sabatelli, J. T. Achterberg, J. D. Sugimoto, Atun, R., and R. Coker. 2008. “Health Systems and Communicable I. M. Longini Jr., and others. 2009. “Epidemiological Disease Control: Emerging Evidence and Lessons from Central Benefits of More-Effective Tuberculosis Vaccines, Drugs, and Eastern Europe.” In Health Systems and the Challenge of and Diagnostics.” Proceedings of the National Academy of Communicable Diseases: Experiences from Europe and Latin Sciences 106 (33): 13980–85. America, edited by R. Coker, R. Atun, and M. McKee, 193–208. Achkar, J. M., and E. R. Jenny-Avital. 2011. “Incipient and Maidenhead, U.K.: Open University Press. Subclinical Tuberculosis: Defining Early Disease States Atun, R., T. de Jongh, F. Secci, K. Ohiri, and O. Adeyi. 2010. in the Context of Host Immune Response.” Journal of “Integration of Targeted Health Interventions into Health Infectious Diseases 204 (Suppl 4): S1179–86. Systems: A Conceptual Framework for Analysis.” Health Acuna-Villaorduna, C., A. Vassall, G. Henostroza, C. Seas, Policy and Planning 25 (2): 104–11. H. Guerra, and others. 2008. “Cost-Effectiveness Analysis of Introduction of Rapid, Alternative Methods to Identify Atun, R., F. M. Knaul, Y. Akachi, and J. Frenk. 2012. “Innovative Multidrug-Resistant Tuberculosis in Middle-Income Financing for Health: What Is Truly Innovative?” The Countries.” Clinical Infectious Diseases 47 (4): 487–95. Lancet 380 (9858): 2044–49. Ahuja, S. D., D. Ashkin, M. Avendano, R. Banerjee, M. Bauer, and Atun, R., J. V. Lazarus, W. Van Damme, and R. Coker. 2010. others. 2012. “Multidrug Resistant Pulmonary Tuberculosis “Interactions between Critical Health System Functions Treatment Regimens and Patient Outcomes: An Individual and HIV/AIDS, Tuberculosis, and Malaria Programmes.” Patient Data Meta-Analysis of 9,153 Patients.” PLoS Medicine Health Policy and Planning 25 (Suppl 1): 11–13. 9 (8): e1001300. Atun, R., M. McKee, F. Drobniewski, and R. Coker. 2005. Akolo, C., I. Adetifa, S. Shepperd, and J. Volmink. 2010. “Analysis of How the Health Systems Context Shapes “Treatment of Latent Tuberculosis Infection in HIV Responses to the Control of Human Immunodeficiency Infected Persons.” Cochrane Database of Systematic Reviews Virus: Case-Studies from the Russian Federation.” Bulletin 20: CD000171. of the World Health Organization 83 (10): 730–38. Alhajri, K., N. Alzerwi, K. Alsaleh, H. Bin Yousef, and M. Alzaben. Atun, R., N. Menabde, K. Saluvere, M. Jesse, and J. Habicht. 2011. “Disseminated (Miliary) Abdominal Tuberculosis after 2006. “Introducing a Complex Health Innovation— Laparoscopic Gastric Bypass Surgery.” BMJ Case Reports 2011 Primary Health Care Reforms in Estonia (Multimethods (May 12): bcr1220103591. Evaluation).” Health Policy 79 (1): 79–91. Alsdurf, H., P. C. Hill, A. Matteelli, H. Getahun, and Atun, R., Y. Samyshkin, F. Drobniewski, Y. Balabanova, I. Fedorin, D. Menzies. 2016. “The Cascade of Care in Diagnosis and and others. 2006. “Costs and Outcomes of Tuberculosis Treatment of Latent Tuberculosis Infection: A Systematic Control in the Russian Federation: Retrospective Cohort Review and Meta-Analysis.” The Lancet Infectious Diseases Analysis.” Health Policy and Planning 21 (5): 353–64. 16: 1269–78. Atun, R., Y. Samyshkin, F. Drobniewski, N. M. Skuratova, Amoroso, C., B Akimana, B. Wise, and H. S. F. Fraser. 2010. G. Gusarova, and others. 2005. “Barriers to Sustainable “Using Electronic Medical Records for HIV Care in Rural Tuberculosis Control in the Russian Federation Health Rwanda.” Studies Health Technology and Informatics 160 System.” Bulletin of the World Health Organization 83 (3): (Pt. 1): 337–41. 217–23. Ananthakrishnan, R., M. Muniyandi, A. Jeyaraj, G. Palani, Atun, R., D. E. Weil, M. Tan Eang, and D. Mwakyusa. 2010. and B. W. C. Sathiyasekaran. 2012. “Expenditure Pattern “Health-System Strengthening and Tuberculosis Control.” for TB Treatment among Patients Registered in an Urban The Lancet 375 (9732): 2169–78. Government DOTS Program in Chennai City, South India.” Awofeso, N., I. Schelokova, and A. Dalhatu. 2008. “Training Tuberculosis Research and Treatment 2012: 747924. of Front-line Health Workers for Tuberculosis Control: 294 Major Infectious Diseases Lessons from Nigeria and Kyrgyzstan.” Human Resources Bates, M. N., A. Khalakdina, M. Pai, L. Chang, F. Lessa, and for Health 6 (20): 1–9. K. R. Smith. 2007. “Risk of Tuberculosis from Exposure to Ayieko, J., L. Abuogi, B. Simchowitz, E. A. Bukusi, A. H. Tobacco Smoke: A Systematic Review and Meta-Analysis.” Smith, and others. 2014. “Efficacy of Isoniazid Prophylactic Archives of Internal Medicine 167 (4): 335–42. Therapy in Prevention of Tuberculosis in Children: A Meta- Bellamy, R., C. Ruwende, T. Corrah, K. P. McAdam, H. C. Whittle, Analysis.” BMC Infectious Diseases 14: 91. and others. 1998. “Variations in the NRAMP1 Gene and Ayles, H., and M. Muyoyeta. 2006. “Isoniazid to Prevent First Susceptibility to Tuberculosis in West Africans.” New England and Recurrent Episodes of TB.” Tropical Doctor 36 (2): 83–86. Journal of Medicine 338 (10): 640–44. Ayles, H., M. Muyoyeta, E. Du Toit, A. Schaap, S. Floyd, Bhargava, A., A. Benedetti, O. Oxlade, M. Pai, and D. Menzies. and others. 2013. “Effect of Household and Community 2014. “Undernutrition and the Incidence of Tuberculosis in Interventions on the Burden of Tuberculosis in Southern India: National and Subnational Estimates of the Population- Africa: The ZAMSTAR Community-Randomised Trial.” The Attributable Fraction Related to Undernutrition.” National Lancet 382 (9899): 1183–94. Medical Journal of India 27 (3): 128–33. Ayles, H., A. Schaap, A. Nota, C. Sismanidis, R. Tembwe, and oth- Bilal, N. K., C. H. Herbst, F. Zhao, A. Soucat, and C. Lemiere. 2011. ers. 2009. “Prevalence of Tuberculosis, HIV, and Respiratory “Health Extension Workers in Ethiopia: Improved Access and Symptoms in Two Zambian Communities: Implications for Coverage for the Rural Poor.” In Yes Africa Can: Success Stories Tuberculosis Control in the Era of HIV.” PLoS One 4 (5): e5602. from a Dynamic Continent, edited by P. Chuhan-Pole and Aziz, A., M. Ishaq, and R. Akhwand. 1985. “Infection Risk of M. Angwafo, 433–43. Washington, DC: World Bank. Sputum Positive Tuberculosis Patients to Their Family Binagwaho, A. 2013. “Resistant TB: Use the Tools Available.” Contacts with and without Chemotherapy.” Journal of Nature 494 (7436): 176. Pakistan Medical Association 35 (8): 249–52. Binagwaho, A., P. E. Farmer, S. Nsanzimana, C. Karema, M. Gasana, Baker, M. A., D. Wilson, K. Wallengren, A. Sandgren, O. Iartchouk, and others. 2014. “Rwanda 20 Years on: Investing in Life.” The and others. 2012. “Polymorphisms in the Gene That Lancet 384 (9940): 371–75. Encodes the Iron Transport Protein Ferroportin 1 Influence BIO Ventures for Global Health. 2006. “Tuberculosis Vaccines: The Susceptibility to Tuberculosis.” Journal of Infectious Diseases Case for Investment.” Washington, DC: BIO Ventures for Global 205 (7): 1043–47. Health. http://www.bvgh.org/Portals/0/Reports/2006_10_tb Barrington, J., O. Wereko-Brobby, P. Ward, W. Mwafongo, and _vaccines,_the_case_for_investment.pdf. S. Kungulwe. 2010. “SMS for Life: A Pilot Project to Improve Blaya, J. A., H. S. F. Fraser, and B. Holt. 2010. “E-Health Anti-Malarial Drug Supply Management in Rural Tanzania Technologies Show Promise in Developing Countries.” Using Standard Technology.” Malaria Journal 9 (298): 1–9. Health Affairs 29 (2): 244–51. Barry, C. E., H. I. Boshoff, V. Dartois, T. Dick, S. Ehrt, Blaya, J. A., S. S. Shin, M. Yagui, C. Contreras, P. Cegielski, and others. 2009. “The Spectrum of Latent Tuberculosis: and others. 2014. “Reducing Communication Delays and Rethinking the Biology and Intervention Strategies.” Nature Improving Quality of Care with a Tuberculosis Laboratory Reviews Microbiology 7 (12): 845–55. Information System in Resource Poor Environments: A Cluster Barter, D. M., S. O. Agboola, M. B. Murray, and T. Bärnighausen. Randomized Controlled Trial.” PLoS One 9 (4): e90110. 2012. “Tuberculosis and Poverty: The Contribution of Blaya, J. A., S. S. Shin, M. Yagui, G. Yale, C. Z. Suarez, and others. Patient Costs in Sub-Saharan Africa—A Systematic Review.” 2007.“A Web-Based Laboratory Information System to Improve BMC Public Health 12 (November 14): 980. Quality of Care of Tuberculosis Patients in Peru: Functional Bassili, A., C. Fitzpatrick, E. Qadeer, R. Fatima, K. Floyd, and Requirements, Implementation and Usage Statistics.” BMC others. 2013. “A Systematic Review of the Effectiveness Medical Informatics and Decision Making 7: 33. of Hospital- and Ambulatory-Based Management of Bloom, B. R. 1994. Tuberculosis: Pathogenesis, Protection, and Multidrug-Resistant Tuberculosis.” American Journal of Control. Washington, DC: ASM Press. Tropical Medicine and Hygiene 89 (2): 271–80. Bloom, B. R., and R. Atun. 2016. “Rethinking Global Control of Basu, S., J. R. Andrews, E. M. Poolman, N. R. Gandhi, N. S. Shah, Tuberculosis.” Science Translational Medicine 8 (329): 329. and others. 2007. “Prevention of Nosocomial Transmission Bloom, B. R., and P. E. M. Fine. 1994. “The BCG Experience: of Extensively Drug-Resistant Tuberculosis in Rural South Implications for Future Vaccines against Tuberculosis.” African District Hospitals: An Epidemiological Modelling In Tuberculosis: Pathogenesis, Protection, and Control, edited Study.” The Lancet 370 (9597): 1500–07. by B. R. Bloom, 531–57. Washington, DC: ASM Press. Basu, S., and A. P. Galvani. 2008. “The Transmission and Blower, S. M., A. R. Mclean, T. C. Porco, P. M. Small, P. C. Hopewell, Control of XDR TB in South Africa: An Operations Research and others. 1995. “The Intrinsic Transmission Dynamics of and Mathematical Modelling Approach.” Epidemiology and Tuberculosis Epidemics.” Nature Medicine 1 (8): 815–21. Infection 136 (12): 1585–98. Boehme, C. C., P. Nabeta, D. Hillemann, M. P. Nicol, S. Shenai, Bates, M., J. O’Grady, P. Mwaba, L. Chilukutu, J. Mzyece, and others. 2010. “Rapid Molecular Detection of Tuberculosis others. 2012. “Evaluation of the Burden of Unsuspected and Rifampin Resistance.” New England Journal of Medicine Pulmonary Tuberculosis and Co-Morbidity with Non- 363 (11): 1005–15. Communicable Diseases in Sputum Producing Adult Boehme, C. C., M. P. Nicol, P. Nabeta, J. S. Michael, E. Gotuzzo, Inpatients.” PLoS One 7 (7): e40774. and others. 2011. “Feasibility, Diagnostic Accuracy, and Tuberculosis 295 Effectiveness of Decentralised Use of the Xpert MTB/RIF Chadha, V., P. Kumar, P. S. Jagannatha, P. S. Vaidyanathan, Test for Diagnosis of Tuberculosis and Multidrug Resistance: and K. P. Unnikrishnan. 2005. “Average Annual Risk of A Multicentre Implementation Study.” The Lancet 377 Tuberculous Infection in India.” International Journal of (9776): 1495–505. Tuberculosis and Lung Disease 9 (1): 116–18. Boelaert, J. R., S. J. Vandecasteele, R. Appelberg, and V. R. Gordeuk. Chahal, J. S., O. F. Khan, C. L. Cooper, J. S. McPartlan, J. K. Tsosie, 2007. “The Effect of the Host’s Iron Status on Tuberculosis.” and others. 2016. “Dendrimer-RNA Nanoparticles Generate Journal of Infectious Diseases 195 (12): 1745–53. Protective Immunity against Lethal Ebola, H1N1 Influenza, and Borrell, S., and S. Gagneux. 2011. “Strain Diversity, Epistasis, and Toxoplasma gondii Challenges with a Single Dose.” Proceedings the Evolution of Drug Resistance in Mycobacterium tubercu- of the National Academy of Sciences 113 (29): E4133–42. losis.” Clinical Microbiology and Infection 17 (6): 815–20. Chakraborty, S., T. Gruber, C. E. Barry, H. I. Boshoff, and Brennan, M. J., and J. Thole. 2012. “Tuberculosis Vaccines: K. Y. Rhee. 2013. “Para-Aminosalicylic Acid Acts as an A Strategic Blueprint for the Next Decade.” Tuberculosis Alternative Substrate of Folate Metabolism in Mycobacterium 92 (Suppl 1): S6–13. tuberculosis.” Science 339 (6115): 88–91. Brouwer, J. A., M. J. Boeree, P. Kager, C. M. Varkevisser, and Chan, P. C., C. H. Yang, L. Y. Chang, K. F. Wang, Y. C. Kuo, and A. D. Harries. 1998. “Traditional Healers and Pulmonary others. 2013. “Lower Prevalence of Tuberculosis Infection Tuberculosis in Malawi.” International Journal of Tuberculosis in BCG Vaccinees: A Cross-Sectional Study in Adult Prison and Lung Disease 2 (3): 231–34. Inmates.” Thorax 68 (3): 263–68. Brust, J. C., N. S. Shah, M. Scott, K. Chaiyachati, M. Lygizos, Chang, K., W. Lu, J. Wang, K. Zhang, S. Jia, and others. and others. 2012. “Integrated, Home-Based Treatment for 2012. “Rapid and Effective Diagnosis of Tuberculosis and MDR-TB and HIV in Rural South Africa: An Alternate Rifampicin Resistance with Xpert MTB/RIF Assay: A Meta- Model of Care.” International Journal of Tuberculosis and Analysis.” Journal of Infection 64 (6): 580–88. Lung Disease 16 (8): 998–1004. Chee, C., M. Teleman, I. C. Boudville, and Y. T. Wang. 2005. Buchan, J., and M. R. Dal Poz. 2002. “Skill Mix in the Health “Contact Screening and Latent TB Infection Treatment in Care Workforce: Reviewing the Evidence.” Bulletin of the Singapore Correctional Facilities.” International Journal of World Health Organization 80 (7): 575–80. Tuberculosis and Lung Disease 9 (11): 1248–52. Bustamante, J., S. Boisson-Dupuis, L. Abel, and J. L. Casanova. Chimusa, E. R., N. Zaitlen, M. Daya, M. Moller, P. D. van 2014. “Mendelian Susceptibility to Mycobacterial Disease: Helden, and others. 2014. “Genome-Wide Association Genetic, Immunological, and Clinical Features of Inborn Study of Ancestry-Specific TB Risk in the South African Errors of IFN-y Immunity.” Seminars in Immunology Coloured Population.” Human Molecular Genetics 23 (3): 26 (6): 454–70. 796–809. Caminero, J. 2003. “Is the DOTS Strategy Sufficient to Choi, I. J., Y. W. Kim, H. S. Lee, K. W. Ryu, H. M. Yoon, and Achieve Tuberculosis Control in Low- and Middle-Income others. 2015. “Risk Factors for TB in Patients with Early Countries? 1. Need for Interventions in Universities and Gastric Cancer: Is Gastrectomy a Significant Risk Factor for Medical Schools [Unresolved Issues].” International Journal TB?” Chest 148 (3): 774–83. of Tuberculosis and Lung Disease 7 (6): 509–15. Churchyard, G., K. Fielding, J. Lewis, L. Coetzee, E. Corbett, Car, J., T. Paljärvi, M. Car, A. Kazeem, A. Majeed, and others. and others. 2012. “Community-Wide Isoniazid Preventive 2012. “Negative Health System Effects of Global Fund’s Therapy Does Not Improve TB Control among Gold Investments in AIDS, Tuberculosis. and Malaria from 2002 Miners: the Thibela TB Study, South Africa.” Paper to 2009: Systematic Review.” Journal of the Royal Society of presented at the 19th Conference on Retroviruses and Medicine Short Reports 3 (10): 70. Opportunistic Infections, Seattle, March 5–8. Cauthen, G. M., A. Pio, and H. G. ten Dam. 2002. “Annual ———. 2014. “A Trial of Mass Isoniazid Preventive Therapy Risk of Tuberculosis Infection.” Bulletin of the World Health for Tuberculosis Control.” New England Journal of Medicine Organization 80 (6): 503–11. 370 (17): 301–10. Cavalcante, S., B. Durovni, G. L. Barnes, F. B. A. Souza, R. F. Cobelens, F., S. van Kampen, E. Ochodo, R. Atun, and Silva, and others. 2010. “Community-Randomized Trial of C. Lienhardt. 2012. “Research on Implementation of Enhanced DOTS for Tuberculosis Control in Rio de Janeiro, Interventions in Tuberculosis Control in Low- and Middle- Brazil.” International Journal of Tuberculosis and Lung Disease Income Countries: A Systematic Review.” PLoS Medicine 14 (2): 203. 9 (12): e1001358. Cavalcante, S., E. Soares, A. G. F. Pacheco, R. E. Chaisson, Cohen, T., C. Colijn, A. Wright, M. Zignol, A. Pym, and others. B. Durovni, and others. 2007. “Community DOT for 2008. “Challenges in Estimating the Total Burden of Drug- Tuberculosis in a Brazilian Favela: Comparison with a Resistant Tuberculosis.” American Journal of Respiratory and Clinic Model.” International Journal of Tuberculosis and Critical Care Medicine 177 (12): 1302–06. Lung Disease 11 (5): 544–49. Cohen, T., H. E. Jenkins, C. Lu, M. McLaughlin, K. Floyd, and Cegielski, J. P., L. Arab, and J. Cornoni-Huntley. 2012. others. 2014. “On the Spread and Control of MDR-TB “Nutritional Risk Factors for Tuberculosis among Adults Epidemics: An Examination of Trends in Anti-Tuberculosis in the United States, 1971–1992.” American Journal of Drug Resistance Surveillance Data.” Drug Resistance Updates Epidemiology 176 (5): 409–22. 17 (4–6): 105–23. 296 Major Infectious Diseases Coker, R., R. A. Atun, and M. McKee. 2004. “Health-Care System Cox, E., and K. Laessig. 2014. “FDA Approval of Bedaquiline— Frailties and Public Health Control of Communicable the Benefit-Risk Balance for Drug-Resistant Tuberculosis.” Disease on the European Union’s New Eastern Border.” The New England Journal of Medicine 371 (8): 689–91. Lancet 363 (9418): 1389–92. Creswell, J., M. Raviglione, S. Ottmani, G. B. Migliori, Coker, R., B. Dimitrova, F. Drobniewski, Y. Samyshkin, M. W. Uplekar, and others. 2011. “Tuberculosis and J. Pomerleau, and others. 2005. “Health System Frailties Noncommunicable Diseases: Neglected Links and Missed in Tuberculosis Service Provision in Russia: An Analysis Opportunities.” European Respiratory Journal 37 (5): 1269–82. through the Lens of Formal Nutritional Support.” Public Critchley, J. A., F. Young, L. Orton, and P. Garner. 2013. Health 119 (9): 837–43. “Corticosteroids for Prevention of Mortality in People with Colijn, C., A. Brandes, J. Zucker, D. S. Lun, B. Weiner, and others. Tuberculosis: A Systematic Review and Meta-Analysis.” The 2009. “Interpreting Expression Data with Metabolic Flux Lancet Infectious Diseases 13 (3): 223–37. Models: Predicting Mycobacterium tuberculosis Mycolic Acid Cuevas, L. E., N. Al-Sonboli, L. Lawson, M. A. Yassin, I. Arbide, Production.” PLoS Computational Biology 5 (8): e1000489. and others. 2011. “LED Fluorescence Microscopy for the Comstock, G. W., C. Baum, and D. E. Snider Jr. 1979. “Isoniazid Diagnosis of Pulmonary Tuberculosis: A Multi-Country Prophylaxis among Alaskan Eskimos: A Final Report of the Cross-Sectional Evaluation.” PLoS Medicine 8(7): e1001057. Bethel Isoniazid Studies.” American Review of Respiratory Das, J., A. Kwan, B. Daniels, S. Satyanarayana, R. Subbaraman, Disease 119 (5): 827–30. and others. 2015. “Use of Standardised Patients to Assess Connor, S., K. Foley, R. Harding, and E. Jaramillo. 2012. Quality of Tuberculosis Care: A Pilot, Cross-Sectional “Declaration on Palliative Care and MDR/XDR-TB.” Study.” The Lancet Infectious Diseases 15 (11): 1305–13. International Journal of Tuberculosis and Lung Disease da Silva, R. C., L. Segat, H. L. da Cruz, H. C. Schindler, 16 (6): 712–13. L. M. Montenegro, and others. 2014. “Association of CD209 Corbett, E. L., T. Bandason, Y. B. Cheung, B. Makamure, and CD209L Polymorphisms with Tuberculosis Infection E. Dauya, and others. 2009. “Prevalent Infectious Tuberculosis in a Northeastern Brazilian Population.” Molecular Biology in Harare, Zimbabwe: Burden, Risk Factors and Implications Reports 41 (8): 5449–57. for Control.” International Journal of Tuberculosis and Lung Datiko, D. G., and B. Lindtjørn. 2009. “Health Extension Workers Disease 13 (10): 1231. Improve Tuberculosis Case Detection and Treatment Success Corbett, E. L., T. Bandason, T. Duong, E. Dauya, B. Makamure, in Southern Ethiopia: A Community Randomized Trial.” and others. 2010. “Comparison of Two Active Case-Finding PLoS One 4(5): e5443. Strategies for Community-Based Diagnosis of Symptomatic ———. 2010. “Cost and Cost-Effectiveness of Smear Positive Smear-Positive Tuberculosis and Control of Infectious Tuberculosis Treatment by Health Extension Workers in Tuberculosis in Harare, Zimbabwe (DETECTB): A Cluster- Southern Ethiopia: A Community Randomized Trial.” Randomised Trial.” The Lancet 376 (9748): 1244–53. PLoS One 5 (2): e9158. Corbett, E. L., S. Charalambous, V. M. Moloi, K. Fielding, Della Mea, V. 1999. “Internet Electronic Mail: A Tool for Low- A. D. Grant, and others. 2004. “Human Immunodeficiency Cost Telemedicine.” Journal of Telemedicine and Telecare Virus and the Prevalence of Undiagnosed Tuberculosis in 5 (2): 84–89. African Gold Miners.” American Journal of Respiratory and DeLuca, A., R. E. Chaisson, and N. A. Martinson. 2009. “Intensified Critical Care Medicine 170 (6): 673–79. Case Finding for Tuberculosis in Prevention of Mother to Corbett, E. L., G. J. Churchyard, T. C. Clayton, B. G. Williams, Child Transmission Programs; a Simple and Potentially Vital D. Mulder, and others. 2000. “HIV Infection and Silicosis: Addition for Maternal and Child Health.” Journal of Acquired The Impact of Two Potent Risk Factors on the Incidence Immune Deficiency Syndromes 50 (2): 196–99. of Mycobacterial Disease in South African Miners.” AIDS den Boon, S., A. Matteelli, N. Ford, and H. Getahun. 2016. 14 (17): 2759–68. “Continuous Isoniazid for the Treatment of Latent Corbett, E. L., B. Marston, G. J. Churchyard, and K. M. De Cock. Tuberculosis Infection in People Living with HIV.” AIDS 2006. “Tuberculosis in Sub-Saharan Africa: Opportunities, 30 (5): 797–801. Challenges, and Change in the Era of Antiretroviral den Boon, S., S. W. P. van Lill, M. W. Borgdorff, D. A. Enarson, Treatment.” The Lancet 367 (9514): 926–37. S. Verver, and others. 2007. “High Prevalence of Tuberculosis Corbett, E. L., C. J. Watt, N. Walker, D. Maher, B. G. Williams, in Previously Treated Patients, Cape Town, South Africa.” and others. 2003. “The Growing Burden of Tuberculosis: Emerging Infectious Diseases 13 (8): 1189. Global Trends and Interactions with the HIV Epidemic.” Detjen, A. K., A. R. DiNardo, J. Leyden, K. R. Steingart, Archives of Internal Medicine 163 (9): 1009–21. D. Menzies, and others. 2015. “Xpert MTB/RIF Assay for Coronado, V. G., C. M. Beck-Sague, M. D. Hutton, B. J. Davis, the Diagnosis of Pulmonary Tuberculosis in Children: P. Nicholas, and others. 1993. “Transmission of Multidrug- A Systematic Review and Meta-Analysis.” The Lancet Resistant Mycobacterium tuberculosis among Persons with Respiratory Medicine 3 (6): 451–61. Human Immunodeficiency Virus Infection in an Urban de Vries, G., R. Aldridge, J. A. Cayla, W. H. Haas, A. Sandgren, Hospital: Epidemiologic and Restriction Fragment Length and others. 2014. “Epidemiology of Tuberculosis in Big Polymorphism Analysis.” Journal of Infectious Diseases Cities of the European Union and European Economic 168 (4): 1052–55. Area Countries.” Eurosurveillance 19 (9): pii=20726. Tuberculosis 297 Dharmadhikari, A. S., M. Mphahlele, A. Stoltz, K. Venter, Role of Geographic Hotspots in Propagating Epidemics.” R. Mathebula, and others. 2012. “Surgical Face Masks Worn Proceedings of the National Academy of Sciences 109 (24): by Patients with Multidrug-Resistant Tuberculosis: Impact 9557–62. on Infectivity of Air on a Hospital Ward.” American Journal Driessen, J., M. Cioffi, N. Alide, Z. Landis-Lewis, G. Gamadzi, of Respiratory and Critical Care Medicine 185 (10): 1104–19. and others 2013. “Modeling Return on Investment for EMR Dharmadhikari, A. S., M. Mphahlele, K. Venter, A. Stoltz, System in Lilongwe, Malawi.” Journal of American Medical R. Mathebula, and others. 2014. “Rapid Impact of Effective Informatics Association 20 (4): 743–48. Treatment on Transmission of Multidrug-Resistant Drobniewski, F. A., R. Atun, I. Fedorin, A. Bikov, R. Coker, and Tuberculosis.” International Journal of Tuberculosis and others. 2004. “The ‘Bear Trap’: The Colliding Epidemics of Lung Disease 18 (9): 1019–25. Tuberculosis and HIV in Russia.” International Journal of Dheda, K., T. Gumbo, N. R. Gandhi, M. Murray, G. Theron, STD and AIDS 15 (10): 641–46. and others. 2014. “Global Control of Tuberculosis: From Dye, C. 2009. “Doomsday Postponed? Preventing and Reversing Extensively Drug-Resistant to Untreatable Tuberculosis.” Epidemics of Drug-Resistant Tuberculosis.” Nature Reviews The Lancet Respiratory Medicine 2 (4): 321–38. Microbiology 7 (1): 81–87. Diacon, A. H., R. Dawson, M. Hanekom, K. Narunsky, ———. 2013. “Making Wider Use of the World’s Most Widely S. J. Maritz, and others. 2010. “Early Bactericidal Activity and Used Vaccine: Bacille Calmette–Guérin Revaccination Pharmacokinetics of PA-824 in Smear-Positive Tuberculosis Reconsidered.” Journal of the Royal Society Interface 10 (87): Patients.” Antimicrobial Agents and Chemotheraphy 54 (8): 20130365. 3402–07. ———. 2015. The Population Biology of Tuberculosis. Princeton, Diacon, A. H., R. Dawson, F. von Groote-Bidlingmaier, G. Symons, NJ: Princeton University Press. A. Venter, and others. 2012. “14-Day Bactericidal Activity Dye, C., A. Bassili, A. L. Bierrenbach, J. F. Broekmans, V. K. Chadha, of PA-824, Bedaquiline, Pyrazinamide, and Moxifloxacin and others. 2008. “Measuring Tuberculosis Burden, Trends, Combinations: A Randomised Trial.” The Lancet 380 (9846): and the Impact of Control Programmes.” The Lancet Infectious 986–93. Diseases 8 (4): 233–43. Diacon, A. H., P. R. Donald, A. Pym, M. Grobusch, R. F. Patientia, Dye, C., and K. Floyd. 2006. “Tuberculosis.” In Disease Control and others. 2012. “Randomized Pilot Trial of Eight Weeks of Priorities in Developing Countries (second edition), edited Bedaquiline (TMC207) Treatment for Multidrug-Resistant by D. T. Jamison, J. G. Breman, A. R. Measham, G. Alleyne, Tuberculosis: Long-Term Outcome, Tolerability, and Effect M. Claeson, D. B. Evans, P. Jha, A. Mills, and P. Musgrove. on Emergence of Drug Resistance.” Antimicrobial Agents and Washington, DC: World Bank and Oxford University Press. Chemotheraphy 56 (6): 3271–76. Dye, C., P. Glaziou, K. Floyd, and M. Raviglione. 2013. Diacon, A. H., A. Pym, M. P. Grobusch, J. M. de los Rios, “Prospects for Tuberculosis Elimination.” Annual Review of E. Gotuzzo, and others. 2014. “Multidrug-Resistant Public Health 34 (December 14): 271–86. Tuberculosis and Culture Conversion with Bedaquiline.” Dye, C., S. Scheele, P. Dolin, V. Pathania, and M. C. Raviglione. New England Journal of Medicine 371 (8): 723–32. 1999. “Global Burden of Tuberculosis: Estimated Incidence, Dick, J., and S. Henchie. 1998. “A Cost Analysis of the Prevalence, and Mortality by Country.” Journal of the Tuberculosis Control Programme in Elsies River, Cape American Medical Association. 282 (7): 677–86. Town.” South African Medical Journal 88 (Suppl 3): 380–83. Dye, C., and B. G. Williams. 2008. “Eliminating Human Ditkowsky, J. B., and K. Schwartzman. 2014. “Potential Cost- Tuberculosis in the Twenty-First Century.” Journal of the Effectiveness of a New Infant Tuberculosis Vaccine in Royal Society Interface 5 (23): 653–62. South Africa—Implications for Clinical Trials: A Decision Economist. 2015. “The Economist Special Online Supplement.” Analysis.” PLoS One 9 (1): e83526. Economist, Copenhagen Consensus. http://www.copenhagen Donald, P. R. 2016. “Chemotherapy for Tuberculous Meningitis.” consensus.com/post-2015-consensus/economist. New England Journal of Medicine 374 (2): 179–81. Edginton, M. 1999. “Tuberculosis Patient Care Decentralised to Donald, P. R., and H. McIlleron. 2009. “Antituberculosis Drugs.” District Clinics with Community-Based Directly Observed In Tuberculosis: A Comprehensive Clinical Reference, edited by Treatment in a Rural District of South Africa.” International H. S. Schaaf and A. Zumla. Philadelphia: Saunders Elsevier. Journal of Tuberculosis and Lung Disease 3 (5): 445–50. Dowdy, D. W., A. Cattamanchi, K. R. Steingart, and M. Pai. 2011. Eisenhut, M., S. Paranjothy, I. Abubakar, S. Bracebridge, “Is Scale-Up Worth It? Challenges in Economic Analysis M. Lilley, and others. 2009. “BCG Vaccination Reduces Risk of Diagnostic Tests for Tuberculosis.” PLoS Medicine 8 (7): of Infection with Mycobacterium tuberculosis as Detected by e1001063. Gamma Interferon Release Assay.” Vaccine 27 (44): 6116–20. Dowdy, D. W., R. E. Chaisson, G. Maartens, E. L. Corbett, Elzinga, G., M. C. Raviglione, and D. Maher. 2004. “Scale and S. E. Dorman. 2008. “Impact of Enhanced Tuberculosis Up: Meeting Targets in Global Tuberculosis Control.” The Diagnosis in South Africa: A Mathematical Model of Expanded Lancet 363 (9411): 814–19. Culture and Drug Susceptibility Testing.” Proceedings of the Ettehad, D., H. S. Schaaf, J. A. Seddon, G. S. Cooke, and N. Ford. National Academy of Sciences 105 (32): 11293–98. 2012. “Treatment Outcomes for Children with Multidrug- Dowdy, D. W., J. E. Golub, R. E. Chaisson, and V. Saraceni. Resistant Tuberculosis: A Systematic Review and Meta- 2012. “Heterogeneity in Tuberculosis Transmission and the Analysis.” The Lancet Infectious Diseases 12 (6): 449–56. 298 Major Infectious Diseases Evans, T. G., M. J. Brennan, L. Barker, and J. Thole. 2013. Shorter-Course Regimens for Tuberculosis: A Model-Based “Preventive Vaccines for Tuberculosis.” Vaccine 31 (Suppl 2): Analysis.” PLoS One 9 (5): e96389. B223–26. Ford, C. B., R. R. Shah, M. K. Maeda, S. Gagneux, M. B. Murray, Fabri, M., S. Stenger, D.-M. Shin, J.-M. Yuk, P. T. Liu, and and others. 2013. “Mycobacterium tuberculosis Mutation others. 2011. “Vitamin D Is Required for IFN-γ–Mediated Rate Estimates from Different Lineages Predict Substantial Antimicrobial Activity of Human Macrophages.” Science Differences in the Emergence of Drug-Resistant Tuberculosis.” Translational Medicine 3 (104): 104. Nature Genetics 45 (7): 784–90. Farmer, P. 2013. “Chronic Infectious Disease and the Future Fox, W., G. A. Ellard, and D. A. Mitchison. 1999. “Studies on of Health Care Delivery.” New England Journal of Medicine the Treatment of Tuberculosis Undertaken by the British 369 (25): 2424–36. Medical Research Council Tuberculosis Units, 1946–1986, Farmer, P., S. Robin, S. L. Ramilus, and J. Y. Kim. 1991. with Relevant Subsequent Publications.” International Journal “Tuberculosis, Poverty, and ‘Compliance’: Lessons from of Tuberculosis and Lung Disease 3 (10, Suppl 2): S231–79. Rural Haiti.” Seminars in Respiratory Infections 6 (4): 254–60. Fraser, H. S., J. Blaya, S. S. Choi, C. Bonilla, and D. Jazayeri. 2006. Faurholt-Jepsen, D., N. Range, G. PrayGod, K. Jeremiah, “Evaluating the Impact and Costs of Deploying an Electronic M. Faurholt-Jepsen, and others. 2012. “The Role of Medical Record System to Support TB Treatment in Peru.” Anthropometric and Other Predictors for Diabetes among AMIA Annual Symposium Proceedings 2006: 265–68. Urban Tanzanians with Tuberculosis.” International Journal Fraser, H. S., A. Habib, M. Goodrich, D. Thomas, J. A. Blaya, of Tuberculosis and Lung Disease 16 (12): 1680–85. and others. 2013. “E-Health Systems for Management of Fenner, L., M. Egger, T. Bodmer, E. Altpeter, M. Zwahlen, and MDR-TB in Resource-Poor Environments: A Decade of others. 2012. “Effect of Mutation and Genetic Background on Experience and Recommendations for Future Work.” In Drug Resistance in Mycobacterium tuberculosis.” Antimicrobial MedInfo, vol. 192, 627–31. Amsterdam: IOS Press E Books. Agents and Chemotherapy 56 (6): 3047–53. Fraser, H. S., D. Jazayeri, C. D. Mitnick, J. S. Mukherjee, and Figueroa-Munoz, J., K. Palmer, M. R. Dal Poz, L. Blanc, J. Bayona. 2002. “Informatics Tools to Monitor Progress and K. Bergström, and others. 2005. “The Health Workforce Outcomes of Patients with Drug Resistant Tuberculosis in Crisis in TB Control: A Report from High-Burden Countries.” Perú.” AMIA Annual Symposium Proceedings 2002: 270–74. Human Resources for Health 3 (1): 2. Fraser, H. S., D. Thomas, J. Tomaylla, N. Garcia, L. Lecca, and Fine, P. E. 1995. “Variation in Protection by BCG: Implications others. 2012. “Adaptation of a Web-Based, Open Source of and for Heterologous Immunity.” The Lancet 346 (8986): Electronic Medical Record System Platform to Support a 1339–45. Large Study of Tuberculosis Epidemiology.” BMC Medical Fitzpatrick, M. C., and K. Floyd. 2012. “A Systematic Review of Informatics and Decision Making 12 (1): 125. the Cost and Cost Effectiveness of Treatment for Multidrug- Frenk, J. 1994. “Dimensions of Health System Reform.” Health Resistant Tuberculosis.” Pharmacoeconomics 30 (1): 63–80. Policy 27 (1): 19–34. Floyd, K., V. K. Arora, K. J. Murthy, K. Lonnroth, N. Singla, and Frick, M., I. Henry, and E. Lessem. 2016. “Falling Short of the others. 2006. “DOTS for Tuberculosis Control: Evidence Rights to Health and Scientific Progress: Inadequate TB from India.” Bulletin of the World Health Organization 84: Drug Research and Access.” Health and Human Rights 18 437–45. (1): 9–24. Floyd, K., C. Fitzpatrick, A. Pantoja, and M. Raviglione. 2013. Frieden, T. R., ed. 2004. Toman’s Tuberculosis: Case Detection, “Domestic and Donor Financing for Tuberculosis Care and Treatment, and Monitoring. Geneva: World Health Organization. Control in Low-Income and Middle-Income Countries: An Frieden, T. R., P. I. Fujiwara, R. M. Washko, and M. A. Hamburg. Analysis of Trends, 2002–11, and Requirements to Meet 1995. “Tuberculosis in New York City—Turning the Tide.” 2015 Targets.” The Lancet Global Health 1 (2): e105–15. New England Journal of Medicine 333 (4): 229–33. Floyd, K., R. Hutubessy, K. Kliiman, R. Centis, N. Khurieva, and Friedich, S. O., A. Rachow, E. Saathoff, K. Singth, C. D. Mangu, others. 2012. “Cost and Cost-Effectiveness of Multidrug- and others. 2013. “Assessment of the Sensitivity and Specificity Resistant Tuberculosis Treatment in Estonia and Russia.” of Xpert MTB/RIF Assay as an Early Sputum Biomarker of European Respiratory Journal 40 (1): 133–42. Response to Tuberculosis Treatment.” The Lancet Respiratory Floyd, K., J. Skeva, T. Nyirenda, F. Gausi, and F. Salaniponi. 2003. Medicine 1 (6): 432–70. “Cost and Cost-Effectiveness of Increased Community and Furin, J., J. Bayona, M. Becerra, P. Farmer, A. Golubkov, and Primary Care Facility Involvement in Tuberculosis Care others. 2011. “Programmatic Management of Multidrug- in Lilongwe District, Malawi.” International Journal of Resistant Tuberculosis: Models from Three Countries.” Tuberculosis and Lung Disease 7 (Suppl 1): S29–37. International Journal of Tuberculosis and Lung Disease 15 (10): Floyd, K., D. Wilkinson, and C. Gilks. 1997. “Comparison of 1294–300. Cost Effectiveness of Directly Observed Treatment (DOT) Gagneux, S. 2012. “Host–Pathogen Coevolution in Human and Conventionally Delivered Treatment for Tuberculosis: Tuberculosis.” Philosophical Transactions of the Royal Society Experience from Rural South Africa.” BMJ 315 (7120): B: Biological Sciences 367 (1590): 850–59. 1407–11. Gandhi, N. R., A. P. Moll, U. Lalloo, R. Pawinski, K. Zeller, Fofana, M. O., G. M. Knight, G. B. Gomez, R. G. White, and others. 2009. “Successful Integration of Tuberculosis and D. W. Dowdy. 2014. “Population-Level Impact of and HIV Treatment in Rural South Africa: The Sizonq’oba Tuberculosis 299 Study.” Journal of Acquired Immune Deficiency Syndromes Golub, J. E., C. I. Mohan, G. W. Comstock, and R. E. Chaisson. 50 (1): 37–43. 2005. “Active Case Finding of Tuberculosis: Historical Gandhi, N. R., A. Moll, A. W. Sturm, R. Pawinski, T. Govender, Perspective and Future Prospects.” International Journal of and others. 2006. “Extensively Drug-Resistant Tuberculosis Tuberculosis and Lung Disease 9 (11): 1183–203. as a Cause of Death in Patients Co-Infected with Gomez, G. B., D. W. Dowdy, M. L Bastos, A. Zwerling, S. Sweeney, Tuberculosis and HIV in a Rural Area of South Africa.” The and others. 2016. “Cost and Cost-Effectiveness of Tuberculosis Lancet 368 (9547): 1575–80. Treatment Shortening: A Model-based Analysis.” BMC Gandhi, N. R., P. Nunn, K. Dheda, H. S. Schaff, M. Zignol, Infectious Diseases. 16 (1): 726. and others. 2010. “Multidrug-Resistant and Extensively Grandjean, L., R. H. Gilman, L. Martin, E. Soto, and others. Drug-Resistant Tuberculosis: A Threat to Global Control of 2015. “Transmission of Multidrug-Resistant and Drug Tuberculosis.” The Lancet 375 (9728): 1830–43. Susceptible Tuberculosis within Households: A Prospective Gandhi, N. R., D. Weissman, P. Moonley, M. Ramathal, I. Elson, Cohort Study.” PLoS Medicine 12 (6): e1001843. and others. 2013. “Nosocomial Transmission of Extensively Green, C., J. F. Huggett, E. Talbot, P. Mwaba, K. Reither, and Drug-Resistant Tuberculosis in a Rural Hospital in South A. Zumla. 2009. “Rapid Diagnosis of Tuberculosis through Africa.” Journal of Infectious Diseases 207 (1): 9–17. the Detection of Mycobacterial DNA in Urine by Nucleic Gasana, M., G. Vandebriel, G. Kabanda, S. J. Tsiouris, J. Justman, Acid Amplification Methods.” The Lancet Infectious Diseases and others. 2008. “Integrating Tuberculosis and HIV Care 9 (8) 505–11. in Rural Rwanda.” International Journal of Tuberculosis and Gupta, R., M. Gao, A. Cirule, H. Xiao, L. Geiter, and others. 2015. Lung Disease 12 (Suppl 1): S39–43. “Delamanid for Extensively Drug-Resistant Tuberculosis.” Geissbuhler, A., O. Ly, C. Lovis, and J.-F. L’Haire. 2003. New England Journal of Medicine 373 (3): 291–92. “Telemedicine in Western Africa: Lessons Learned from a Hanekom, M., G. D. van der Spuy, E. Streicher, S. L. Ndabambi, Pilot Project in Mali, Perspectives and Recommendations.” C. R. McEvoy, and others. 2007. “A Recently Evolved In Annual Symposium Proceedings, vol. 2003, 249. Bethesda, Sublineage of the Mycobacterium tuberculosis Beijing Strain MD: American Medical Informatics Association. Family Is Associated with an Increased Ability to Spread Gelmanova, I. Y., S. Keshavjee, V. T. Goluchikova, V. I. Berzina, and Cause Disease.” Journal of Clinical Microbiology 45 (5): A. K. Strelis, and others. 2007. “Barriers to Successful 1483–90. Tuberculosis Treatment in Tomsk, Russian Federation: Hanifa, Y., A. Grant, J. Lewis, E. L. Corbett, K. Fielding, and Non-Adherence, Default, and the Acquisition of Multidrug others. 2009. “Prevalence of Latent Tuberculosis Infection Resistance.” Bulletin of the World Health Organization 85 among Gold Miners in South Africa.” International Journal (9): 649–732. of Tuberculosis and Lung Disease 13 (1): 39–46. Getahun, H., A. Matteelli, R. E. Chaisson, and M. Raviglione. Hansen, S. G., M. Piatak Jr., A. B. Ventura, C. M. Hughes, 2015. “Latent Mycobacterium tuberculosis Infection.” New R. M. Gilbride, and others. 2013. “Immune Clearance of England Journal of Medicine 372 (May 28): 2127–35. Highly Pathogenic SIV Infection.” Nature 502 (7469): 100–04. Gillespie, S. H., A. M. Crook, T. D. McHugh, C. M. Mendel, Harding, R., K. M. Foley, S. R. Connor, and E. Jaramillo. 2012. S. K. Meredith, and others. 2014.“Four-Month Moxifloxacin- “Palliative and End-of-Life Care in the Global Response to Based Regimens for Drug-Sensitive Tuberculosis.” New Multidrug-Resistant Tuberculosis.” The Lancet Infectious England Journal of Medicine 371 (17): 1577–87. Diseases 12 (8): 643–6. Glaziou, P., K. Floyd, E. L. Korenromp, C. Sismanidis, Harries, A. D. 1997. “Tuberculosis in Africa: Clinical Presentation A. Bierrenbach, and others. 2011. “Lives Saved by and Management.” Pharmacology and Therapeutics 73 (1): 1–50. Tuberculosis Control and Prospects for Achieving the 2015 Harries, A. D., R. Zachariah, K. Bergström, L. Blanc, F. M. Global Target for Reducing Tuberculosis Mortality.” Bulletin Salaniponi, and others. 2005. “Human Resources for of the World Health Organization 89 (8): 573–82. Control of Tuberculosis and HIV-Associated Tuberculosis Gler, M., L. Podewils, N. Munez, M. Galipot, M. I. D. Quelapio, [Unresolved Issues].” International Journal of Tuberculosis and others. 2012. “Impact of Patient and Program Factors and Lung Disease 9 (2): 128–37. on Default during Treatment of Multidrug-Resistant Harris, J. B., S. M. Hatwiinda, K. M. Randels, B. H. Chi, Tuberculosis.” International Journal of Tuberculosis and Lung N. G. Kancheya, and others. 2008. “Early Lessons from the Disease 16 (7): 955–60. Integration of Tuberculosis and HIV Services in Primary Gler, M. T., V. Skripconoka, E. Sanchez-Garavito, H. Xiao, Care Centers in Lusaka, Zambia.” International Journal of J. L. Cabrera-Rivero, and others. 2012. “Delamanid for Tuberculosis and Lung Disease 12 (7): 773–79. Multidrug-Resistant Pulmonary Tuberculosis.” New England Harris, R. C., M. S. Khan, L. J. Martin, V. Allen, D. A. Moore, Journal of Medicine 366 (23): 2151–60. and others. 2016. “The Effect of Surgery on the Outcome of Golub, J. E., S. Cohn, V. Saraceni, S. C. Cavalcante, A. G. Pacheco, Treatment for Multidrug-Resistant Tuberculosis: A Systematic and others. 2015. “Long-Term Protection from Isoniazid Review and Meta-Analysis.” BMC Infectious Diseases 16: 262. Preventive Therapy for Tuberculosis in HIV-Infected Patients Hart, P. A., and I. Sutherland. 1977. “BCG and Vole Bacillus in a Medium-Burden Tuberculosis Setting: The TB/HIV in Vaccines in the Prevention of Tuberculosis in Adolescence Rio (THRio) Study.” Clinical Infectious Diseases 60 (4): 639–45. and Early Adult Life.” BMJ 2 (6082): 293. 300 Major Infectious Diseases Hasegawa, N, T. Miura, K. Ishii, K. Yamaguchi, T. H. Lindner, Huang, C.-C., E. T. Tchetgen, M. C. Becerra, T. Cohen, J. Galea, and others. 2002. “New Simple and Rapid Test for Culture and others. 2014. “Cigarette Smoking among Tuberculosis Confirmation of Mycobacterium tuberculosis Complex: Patients Increases Risk of Transmission to Child Contacts.” A Multicenter Study.” Journal of Clinical Microbiology International Journal of Tuberculosis and Lung Disease 18 40 (3): 908–12. (11): 1285–91. Heimbeck, J. 1938. “Incidence of Tuberculosis in Young Adult Huang, C.-C., E. T. Tchetgen, M. C. Becerra, T. Cohen, Women with Special Reference to Employment.” British K. C. Hughes, and others. 2014. “The Effect of HIV- Journal of Tuberculosis 32 (3): 154–66. Related Immunosuppression on the Risk of Tuberculosis Heller, T., R. Lessells, C. G. Wallrauch, T. Bärnighausen, Transmission to Household Contacts.” Clinical Infectious C. S. Cooke, and others. 2010. “Community-Based Diseases 58 (6): 765–74. Treatment for Multidrug-Resistant Tuberculosis in Rural Huerga, H., H. Spillane, W. Guerrero, A. Odongo, and F. Varaine. KwaZulu-Natal, South Africa.” International Journal of 2010. “Impact of Introducing Human Immunodeficiency Tuberculosis and Lung Disease 14 (4): 420–26. Virus Testing, Treatment and Care in a Tuberculosis Clinic Hermans, S., C. R. Horsburgh Jr., and R. Wood. 2015. in Rural Kenya.” International Journal of Tuberculosis and “A Century of Tuberculosis Epidemiology in the Northern Lung Disease 14 (5): 611–15. and Southern Hemisphere: The Differential Impact of IHME (Institute for Health Metrics and Evaluation). 2015. Control Interventions.” PLoS One 10 (8): e0135179. “Financing Global Health 2014. Shifts in Funding as the Hesseling, A. C., B. J. Marais, R. P. Gie, H. S. Schaaf, P. E. Fine, and MDG Era Closes.” IHME, Seattle. others. 2007. “The Risk of Disseminated Bacille Calmette- Institute of Medicine. 2013. Developing and Strengthening the Guerin (BCG) Disease in HIV-Infected Children.” Vaccine Global Supply Chain for Second-Line Drugs for Multidrug- 25 (1): 14–18. Resistant Tuberculosis: Workshop Summary. Washington, Hirsch-Moverman, Y., R. Shrestha-Kuwahara, J. Bethel, DC: National Academies Press. H. M. Blumberg, T. K. Venkatappa, and others. 2015. “Latent International Institute for Population Sciences and Macro Tuberculous Infection in the United States and Canada: International. 2007. National Family Health Survey Who Completes Treatment and Why?” International Journal (NFHS-3), 2005–2006. Mumbai: International Institute for of Tuberculosis and Lung Disease 19 (1): 31–38. Population Sciences. Hoagland, D. T., J. Liu, R. B. Lee, and R. E. Lee. 2016. “New Isaakidis, P., R. Paryani, K. Samsuddin, H. Mansoor, Agents for the Treatment of Drug-Resistant Mycobacterium M. Manglani, and others. 2013. “Poor Outcomes in a Cohort tuberculosis.” Advanced Drug Delivery Reviews 102: 55–72. of HIV-Infected Adolescents Undergoing Treatment for Hong Kong Chest Service and British Medical Research Multidrug-Resistant Tuberculosis in Mumbai, India.” PLoS Council. 1979. “Controlled Trial of 6-Month and 8-Month One 8 (7): e68869. Regimens in the Treatment of Pulmonary Tuberculosis: Islam, A., M. A. May, F. Ahmed, R. A. Cash, and J. Ahmed. 2011. The Results up to 24 Months.” Tubercle 60 (4): 201–10. Making Tuberculosis History: Community-Based Solutions Hopkin, J. 2000. “Atopy, Asthma, and the Mycobacteria.” for Millions. Dhaka: University Press Limited. Thorax 55 (6): 443–45. Islam, M. D., S. Wakai, N. Ishikawa, A. M. R. Chowdhury, and Horter, S., B. Stringer, L. Reynolds, M. Shoaib, S. Kasozi, and J. P. Vaughan. 2002. “Cost-Effectiveness of Community others. 2014. “‘Home Is Where the Patient Is’: A Qualitative Health Workers in Tuberculosis Control in Bangladesh.” Analysis of a Patient-Centred Model of Care for Multi- Bulletin of the World Health Organization 80 (6): 445–50. Drug Resistant Tuberculosis.” BMC Health Services Research Islam, M. S., J. P. Richards, and A. K. Ojha. 2012. “Targeting 14 (1): 1–8. Drug Tolerance in Mycobacteria: A Perspective from Hou, W. L., F. Song, N. X. Zhang, X. X. Dong, S. Y. Cao, Mycobacterial Biofilms.” Expert Review of Anti-Infective and others. 2012. “Implementation and Community Therapy 10 (9): 1055–66. Involvement in DOTS Strategy: A Systematic Review of Isler, M., P. Rivest, J. Mason, and P. Brassard. 2013. “Screening Studies in China.” International Journal of Tuberculosis and Employees of Services for Homeless Individuals in Montréal Lung Disease 16 (11): 1433–40. for Tuberculosis Infection.” Journal of Infection and Public Howard, A. A., and W. M. El-Sadr. 2010. “Integration of Health 6 (3): 209–15. Tuberculosis and HIV Services in Sub-Saharan Africa: Jack, C., U. Lalloo, Q. A. Karim, S. A. Karim, W. El-Sadr, and Lessons Learned.” Clinical Infectious Diseases 50 (Suppl 3): others. 2004. “A Pilot Study of Once-Daily Antiretroviral S238–44. Therapy Integrated with Tuberculosis Directly Observed Howitt, P., A. Darzi, G. Z. Yang, H. Ashrafian, and R. Atun. Therapy in a Resource-Limited Setting.” Journal of Acquired 2012. “Technologies for Global Health.” The Lancet 380 Immune Deficiency Syndromes 36 (4): 929–34. (9840): 507–35. Jagirdar, J., and D. Zagzag. 1996. “Pathology and Insights into Hsiao, W. C., and P. S. Heller. 2007. “What Should Pathogenesis of Tuberculosis.” In Tuberculosis, edited by W. N. Macroeconomists Know about Health Care Policy?” Rom and S. Garay, 467–91. Boston: Little, Brown. Working Paper WP/07/13, International Monetary Fund, James, P., R. Gupta, D. J. Christopher, B. Thankagunam, and Washington, DC. B. Veeraraghavan. 2011. “MDR- and XDR-TB among Tuberculosis 301 Suspected Drug-Resistant TB Patients in a Tertiary Care Case Detection: An Impact Evaluation Study.” The Lancet Hospital in India.” Clinical Respiratory Journal 5 (1): Infectious Diseases 12 (8): 608–16. e19–25. Khan, B., P. Ahmed, K. Ullah, C. A. Hussain, I. Hussain, and Jenkins, H. E., V. Plesca, A. Ciobanu, V. Crudu, I. Galusca, S. Raza. 2005. “Frequency of Tuberculosis in Haematological and others. 2013. “Assessing Spatial Heterogeneity of Malignancies and Stem Cell Transplant Recipients.” Journal Multidrug-Resistant Tuberculosis in a High-Burden of the College of Physicians and Surgeons–Pakistan 15 (1): Country.” European Respiratory Journal 42 (5): 1291–301. 30–3. Jeon, C. Y., and M. B. Murray. 2008. “Diabetes Mellitus Increases Khan, F. A., J. Minion, A. Al-Motiri, A. Benedetti, A. D. Harries, the Risk of Active Tuberculosis: A Systematic Review of and others. 2012. “An Updated Systematic Review and 13 Observational Studies.” PLoS Medicine 5 (7): e152. Meta-Analysis on the Treatment of Active Tuberculosis in Jha, P., and R. Laxminarayan. 2009. “Choosing Health in India: Patients with HIV Infection.” Clinical Infectious Diseases An Entitlement for All Indians.” University of Toronto and 55 (8): 1154–63. Resources for the Future, New Delhi. Kim, C. H., K. H. Im, S. S. Yoo, S. Y. Lee, S. I. Cha, and Jiménez-Levi, E. 2012. Tuberculosis Research and Development: others. 2014. “Comparison of the Incidence between 2012 Report on Tuberculosis Research Funding Trends, Tuberculosis and Nontuberculous Mycobacterial Disease 2005–2011. New York: Treatment Action Group. after Gastrectomy.” Infection 42 (4): 697–704. Jindani, A., T. S. Harrison, A. J. Nunn, P. P. Phillips, G. J. Churchyard, Kim, J. Y., P. Farmer, and M. E. Porter. 2013. “Redefining Global and others. 2014. “High-Dose Rifapentine with Moxifloxacin Health-Care Delivery.” The Lancet 382 (9897): 1060–69. for Pulmonary Tuberculosis.” New England Journal of Medicine King, G., E. Gakidou, K. Imai, J. Lakin, R. T. Moore, and 371 (October 23): 1599–608. others. 2009. “Public Policy for the Poor? A Randomised Kagina, B. M., B. Abel, M. Bowmaker, T. J. Scriba, S. Gelderbloem, Assessment of the Mexican Universal Health Insurance and others. 2009. “Delaying BCG Vaccination from Birth to Programme.” The Lancet 373 (9673): 1447–54. 10 Weeks of Age May Result in an Enhanced Memory CD4 Kirwan, D. E., M. K. Cárdenas, and R. Gilman. 2012. “Rapid T Cell Response.” Vaccine 27 (40): 5488–95. Implementation of New TB Diagnostic Tests: Is it Too Karonga Prevention Trial Group. 1996. “Randomised Controlled Soon for a Global Roll-Out of Xpert MTB/RIF?” American Trial of Single BCG, Repeated BCG, or Combined BCG Journal of Tropical Medicine and Hygiene 87 (2): 197–201. and Killed Mycobacterium leprae Vaccine for Prevention Knight, G. M., P. J. Dodd, A. D. Grant, K. L. Fielding, of Leprosy and Tuberculosis in Malawi.” The Lancet 348 G. J. Churchyard, and others. 2015. “Tuberculosis Prevention (9019): 17–24. in South Africa.” PLoS One 10 (4): e0122514. Karumbi, J., and P. Garner. 2015. “Directly Observed Therapy Knight, G. M., U. K. Griffiths, T. Sumner, Y. V. Laurence, for Treating Tuberculosis.” Cochrane Database of Systematic A. Gheorghe, and others. 2014. “Impact and Cost-Effectiveness Reviews 5: CD003343. of New Tuberculosis Vaccines in Low- and Middle-Income Katz, I., M. Aziz, M. Olszak-Olszewski, R. Komatsu, D. Low- Countries.” Proceedings of the National Academy of Sciences Beer, and others. 2010. “Factors Influencing Performance of 111 (43): 15520–25. Global Fund-Supported Tuberculosis Grants.” International Kranzer, K., H. Afnan-Holmes, K. Tomlin, J. E. Golub, A. E. Shapiro, Journal of Tuberculosis and Lung Disease 14 (9): 1097–103. and others. 2013.“The Benefits to Communities and Individuals Kaufmann, S. H. E., M. F. Cotton, B. Eisele, M. Gengenbacher, of Screening for Active Tuberculosis Disease: A Systematic L. Grode, and others. 2014. “The BCG Replacement Vaccine Review.” International Journal of Tuberculosis and Lung Disease VPM1002: From Drawing Board to Clinical Trial.” Expert 17 (4): 432–46. Review of Vaccines 13 (5): 619–30. Kranzer, K., R. M. Houben, J. R. Glynn, L. G. Bekker, R. Wood, Keane, J., S. Gershon, R. P. Wise, E. Mirabile-Levens, J. Kasznica, and others. 2010. “Yield of HIV-Associated Tuberculosis and others. 2001. “Tuberculosis Associated with Infliximab, During Intensified Case Finding in Resource-Limited A Tumor Necrosis Factor A–Neutralizing Agent.” New Settings: A Systematic Review and Meta-Analysis.” The England Journal of Medicine 345 (15): 1098–104. Lancet Infectious Diseases 10 (2): 93–102. Keshavjee, S., and P. E. Farmer. 2010. “Picking Up the Pace: Kritzinger, F. E., S. den Boon, S. Verver, D. A. Enarson, Scale-Up of MDR Tuberculosis Treatment Programs.” New C. J. Lombard and others. 2009. “No Decrease in Annual England Journal of Medicine 363 (19): 1781–84. Risk of Tuberculosis Infection in Endemic Area in Cape ———. 2012. “Tuberculosis, Drug Resistance, and the History Town, South Africa.” Tropical Medicine and International of Modern Medicine.” New England Journal of Medicine Health 14 (2): 136–42. 367 (10): 931–36. Kruk, M. E., E. Goldmann, and S. Galeo. 2009. “Borrowing and Keshavjee, S., and K. Seung. 2008. “Stemming the Tide of Selling to Pay for Health Care in Low- and Middle-Income Multidrug-Resistant Tuberculosis: Major Barriers to Countries.” Health Affairs 28 (4): 1056–66. Addressing the Growing Epidemic.” Institute of Medicine Kruk, M. E., N. R. Schwalbe, and C. A. Aguiar. 2008. “Timing of report, Institute of Medicine, Washington, DC. https://www Default from Tuberculosis Treatment: A Systematic Review.” .ncbi.nlm.nih.gov/books/NBK45010. Tropical Medicine and International Health 13 (5): 703–12. Khan, A. J., S. Khowaja, F. S. Khan, F. Qazi, I. Lotia, and others. Kumar, M., and S. Kumar. 1997. “Tuberculosis Control in India: 2012. “Engaging the Private Sector to Increase Tuberculosis Role of Private Doctors.” The Lancet 350 (9087): 1329–30. 302 Major Infectious Diseases Labrique, A. B., L. Vasudevan, E. Kochi, R. Fabricant, and G. Mehl. Libshitz, H. I., H. K. Pannu, L. S. Elting, and C. D. Cooksley. 2013. “mHealth Innovations as Health System Strengthening 1997. “Tuberculosis in Cancer Patients: An Update.” Journal Tools: 12 Common Applications and a Visual Framework.” of Thoracic Imaging 12 (1): 41–46. Global Health: Science and Practice 1 (2): 160–71. Lienhardt, C., K. Fielding, J. Sillah, A. Tunkara, S. Donkor, Lal, S., M. W. Uplekar, I. Katz, K. Lonnroth, R. Komatsu, and and others. 2003. “Risk Factors for Tuberculosis Infection others. 2011. “Global Fund Financing of Public-Private Mix in Sub-Saharan Africa: A Contact Study in the Gambia.” Approaches for Delivery of Tuberculosis Care.” Tropical American Journal of Respiratory and Critical Care Medicine Medicine and International Health 16 (6): 685–92. 168 (4): 448–55. Laurence, Y. V., U. K. Griffiths, and A. Vassall. 2015. “Costs to Lietman, T., T. Porco, and S. Blower. 1997. “Leprosy and Health Services and the Patient of Treating Tuberculosis: Tuberculosis: The Epidemiological Consequences of Cross- A Systematic Literature Review.” PharmacoEconomics Immunity.” American Journal of Public Health 87 (12): 1923–27. 33 (9): 1–17. Lin, H.-H., D. Dowdy, C. Dye, M. Murray, and T. Cohen. Lawn, S. D. 2015. “Advances in Diagnostic Assays for 2012. “The Impact of New Tuberculosis Diagnostics on Tuberculosis.” Cold Spring Harbor Perspectives in Medicine Transmission: Why Context Matters.” Bulletin of the World 5 (12): pii=a017806. Health Organization 90 (10): 739–47. Lawn, S. D., S. V. Brooks, K. Kranzer, M. P. Nicol, A. Whitelaw, and Lin, H.-H., M. Ezzati, and M. Murray. 2007. “Tobacco Smoke, others. 2011. “Screening for HIV-Associated Tuberculosis Indoor Air Pollution, and Tuberculosis: A Systematic and Rifampicin Resistance before Antiretroviral Therapy Review and Meta-Analysis.” PLoS Medicine 4 (1): e20. Using the Xpert MTB/RIF Assay: A Prospective Study.” Lin, H.-H., C. Suk, H. L. Lo, R. Y. Huang, D. A. Enarson, and PLoS Medicine 8 (7): e1001067. others. 2014. “Indoor Air Pollution from Solid Fuel and Lawn, S. D., A. D. Kerkhoff, M. Vogt, and R. Wood. 2012. Tuberculosis: A Systematic Review and Meta-Analysis.” “Diagnostic Accuracy of a Low-Cost, Urine Antigen, Point- International Journal of Tuberculosis and Lung Disease of-Care Screening Assay for HIV-Associated Pulmonary 18 (5): 613–21. Tuberculosis before Antiretroviral Therapy: A Descriptive Lin, X., V. Chongsuvivatwong, L. Lin, A. Geater, and R. Lijuan. Study.” The Lancet Infectious Diseases 12 (3): 201–09. 2008. “Dose–Response Relationship between Treatment Lee, D., S. S. Lal, R. Komatsu, A. Zumla, and R. Atun. 2012. “Global Delay of Smear-Positive Tuberculosis Patients and Intra- Fund Financing of Tuberculosis Services Delivery in Prisons.” Household Transmission: A Cross-Sectional Study.” Journal of Infectious Diseases 205 (Suppl 2): S274–83. Transactions of the Royal Society of Tropical Medicine and Lee, D. K., Y.-S. Cho, S. H. Hong, W. H. Chung, and Y. C. Ahn. Hygiene 102 (8): 797–804. 2006. “Inflammatory Pseudotumor Involving the Skull Base: Lipsitch, M., and A. O. Sousa. 2002. “Historical Intensity of Response to Steroid and Radiation Therapy.” Otolaryngology— Natural Selection for Resistance to Tuberculosis.” Genetics Head and Neck Surgery 135 (1): 144–48. 161 (4): 1599–607. Lee, M., T. Song, Y. Kim, I. Jeong, S. N Cho, and C. E. Barry. Liu, P. T., S. Stenger, D. H. Tang, and R. L. Modlin. 2007. 2015. “Linezolid for XDR-TB—Final Study Outcomes.” “Cutting Edge: Vitamin D-Mediated Human Antimicrobial New England Journal of Medicine 373 (3): 290–91. http:// Activity against Mycobacterium tuberculosis Is Dependent www.nejm.org/doi/full/10.1056/NEJMc1500286#t=article. on the Induction of Cathelicidin.” Journal of Immunology Lee, R. S., N. Radmonski, J. F. Proulx, I. Levade, B. J. Shapiro 179 (4): 2060–63. and others. 2015. “Population Genomics of Mycobacterium Lönnroth, K., B. G. Williams, P. Cegielski, and C. Dye. 2010. “A tuberculosis in the Inuit.” Proceedings of the National Consistent Log-Linear Relationship between Tuberculosis Academy of Sciences 112 (44): 13609–14. Incidence and Body Mass Index.” International Journal of Legido-Quigley, H., C. M. Montgomery, P. Khan, R. Atun, Epidemiology 39 (1): 149–55. A. Fakoya, and others. 2013. “Integrating Tuberculosis Lönnroth, K., B. G. Williams, S. Stadlin, E. Jaramillo, and C. Dye. and HIV Services in Low- and Middle-Income Countries: 2008. “Alcohol Use as a Risk Factor for Tuberculosis–A A Systematic Review.” Tropical Medicine and International Systematic Review.” BMC Public Health 8 (1): 289. Health 18 (2): 199–211. Lopez, B., D. Aguilar, H. Orozco, M. Burger, C. Espitia, and Leimane, V., and J. Leimans. 2006. “Tuberculosis Control in others. 2003. “A Marked Difference in Pathogenesis and Latvia: Integrated DOTS.” Eurosurveillance 11 (3): 29–33. Immune Response Induced by Different Mycobacterium Lessem, E., H. Cox, C. Daniels, J. Furin, L. McKenna, and oth- tuberculosis Genotypes.” Clinical and Experimental ers. 2015. “Access to New Medications for the Treatment Immunology 133 (1): 30–37. of Drug-Resistant Tuberculosis: Patient, Provider and Loudon, R. G. and S. K. Spohn. 1969. “Cough Frequency Community Perspectives.” International Journal of Infectious and Infectivity in Patients with Pulmonary Tuberculosis.” Diseases 32: 56–60. American Review of Respiratory Disease 99 (1): 109–11. Lester, R. T., P. Ritvo, E. J. Mills, A. Kariri, S. Karanja, and Loveday, M., K. Wallengren, A. Voce, B. Margot, T. Reddy, others. 2010. “Effects of a Mobile Phone Short Message and others 2012. “Comparing Early Treatment Outcomes Service on Antiretroviral Treatment Adherence in Kenya of MDR-TB in a Decentralised Setting with a Centralised (WelTel Kenya1): A Randomised Trial.” The Lancet 376 Setting in KwaZulu-Natal, South Africa.” International (9755): 1838–45. Journal of Tuberculosis and Lung Disease 16 (2): 209–15. Tuberculosis 303 Lowell, A. M., L. B. Edwards, and C. E. Palmer. 1969. Mao, T. E., K. Okada, N. Yamada, S. Peou, M. Ota, and others. “Tuberculosis Morbidity and Mortality and Its Control.” 2014. “Cross-Sectional Studies of Tuberculosis Prevalence In Tuberculosis, edited by A. M. Lowell, L. B. Edwards, in Cambodia between 2002 and 2011.” Bulletin of the World and C. E. Palmer, 129–66. Cambridge, MA: Harvard Health Organization 92 (8): 573–81. University Press. Marais, B. J., M. C. Raviglione, P. R. Donald, A. D. Harries, Lozano, R., M. Naghavi, K. Foreman, S. Lim, K. Shibuya, and A. L. Kritski, and others. 2010. “Scale-Up of Services and others. 2012. “Global and Regional Mortality from 235 Research Priorities for Diagnosis, Management, and Control of Causes of Death for 20 Age Groups in 1990 and 2010: Tuberculosis: A Call to Action.” The Lancet 375 (9732): 2179–91. A Systematic Analysis for the Global Burden of Disease Marica, C., C. Didilescu, N. Galie, D. Chiotan, J. P. Zellweger, Study 2010.” The Lancet 380 (9859): 2095–128. and others. 2009. “Reversing the Tuberculosis Upwards Lu, L. L., A. W. Chung, T. R. Rosebrock, M. Ghebremichael, Trend: A Success Story in Romania.” European Respiratory W. H. Yu, and others. 2016. “A Functional Role for Journal 33 (1): 168–70. Antibodies in Tuberculosis.” Cell 167 (2): 433–43. Marrone, M. T., V. V Venkataramanan, M. Goodman, A. C. Hill, Lundgrun, B. 2015. “Ebola Kills Far Fewer Than AIDS, TB, J. A. Jereb, and others. 2013. “Surgical Interventions for and Malaria. What Should We Prioritise?” The Guardian, Drug-Resistant Tuberculosis: A Systematic Review and January 19. Meta-Analysis.” International Journal of Tuberculosis and Lutwama, F., B. M. Kagina, A. Waiswa, N. Mansoor, Lung Disease 17 (1): 6–16. S. Kirimunda, and others. 2014. “Distinct T-Cell Responses Martineau, A. R., S. Nhamoyebonde, T. Oni, M. X. Rangaka, When BCG Vaccination Is Delayed from Birth to 6 Weeks S. Marais, and others. 2011. “Reciprocal Seasonal Variation of Age in Ugandan Infants.” Journal of Infectious Diseases in Vitamin D Status and Tuberculosis Notifications in Cape 209 (6): 887–97. Town, South Africa.” Proceedings of the National Academy of Luyirika, E., H. Nsobya, R. Batamwita, P. Busingye, W. Musoke, Sciences 108 (47): 19013–17. and others. 2012. “A Home-Based Approach to Managing Marx, F. M., R. Dunbar, D. A. Enarson, B. G. Williams, Multi-Drug Resistant Tuberculosis in Uganda: A Case R. M. Warren, and others. 2014. “The Temporal Dynamics Report.” AIDS Research and Therapy 9 (1): 12. of Relapse and Reinfection Tuberculosis after Successful Luzze, H., D. F. Johnson, K. Dickman, H. Mayanja-Kizza, Treatment: A Retrospective Cohort Study.” Clinical Infectious A. Okwera, and others. 2013. “Relapse More Common Diseases 58 (12): 1676–83. Than Reinfection in Recurrent Tuberculosis 1–2 Years Mathew, T. A., T. N. Ovsyanikova, S. S. Shin, I. Y. Gelmanova, Post Treatment in Urban Uganda.” International Journal of D. A. Balbuena, and others. 2006. “Causes of Death dur- Tuberculosis and Lung Disease 17 (3): 361–67. ing Tuberculosis Treatment in Tomsk Oblast, Russia.” Lygizos, M., S. V. Shenoi, R. P. Brooks, A. Bhushan, J. C. M. International Journal of Tuberculosis and Lung Disease 10 Brust, and others. 2013. “Natural Ventilation Reduces (8): 857–63. High TB Transmission Risk in Traditional Homes in Rural McFarlane, N. 1989. “Hospitals, Housing, and Tuberculosis KwaZulu-Natal, South Africa.” BMC Infectious Diseases 13 in Glasgow, 1911–51.” Social History of Medicine 2 (1): (July 1): 300. 59–85. MacIntyre, C. R., N. Kendig, L. Kummer, S. Birago, and McIlleron, H., G. Meintjes, W. J. Burman, and G. Maartens. N. M. Graham. 1997. “Impact of Tuberculosis Control 2007. “Complications of Antiretroviral Therapy in Patients Measures and Crowding on the Incidence of Tuberculous with Tuberculosis: Drug Interactions, Toxicity, and Infection in Maryland Prisons.” Clinical Infectious Diseases Immune Reconstitution Inflammatory Syndrome.” Journal 24 (6): 1060–67. of Infectious Diseases 196 (Suppl 1): S63–75. Mamlin, B. W., P. G. Biondich, B. A. Wolfe, H. S. F. Fraser, McKee, M., and R. Atun. 2006. “Beyond Borders: Public-Health D. Jazayeri, and others. 2006. “Cooking Up an Open Source Surveillance.” The Lancet 367 (9518): 1224–26. EMR for Developing Countries: OpenMRS, a Recipe McKeown, T., and R. Record. 1962. “Reasons for the Decline for Successful Collaboration.” AMIA Annual Symposium of Mortality in England and Wales during the Nineteenth Proceedings 2006: 529–53. Century.” Population Studies 16 (2): 94–122. Mangtani, P., I. Abubakar, C. A. Ariti, and J. A. C. Sterne. Meintjes, G., S. D. Lawn, F. Scano, G. Maartens, M. A. French, 2014. “Protection by BCG Vaccine Against Tuberculosis: and others. 2008. “Tuberculosis-Associated Immune A Systematic Review of Randomized Controlled Trials.” Reconstitution Inflammatory Syndrome: Case Definitions Clinical Infectious Diseases 58 (4): 470–80. for Use in Resource-Limited Settings.” The Lancet Infectious Manjelievskaia, J., D. Erck, S. Piracha, and L. Schrager. 2016. Diseases 8 (8): 516–23. “Drug-Resistant TB: Deadly, Costly, and in Need of a Meintjes, G., R. J. Wilkinson, C. Morroni, D. J. Pepper, K. Rebe, Vaccine.” Transactions of the Royal Society of Tropical and others. 2010. “Randomized Placebo-Controlled Trial Medicine and Hygiene 110 (3): 186–91. of Prednisone for Paradoxical TB-Associated Immune Manjourides, J., H. H. Lin, S. Shin, C. Jeffery, C. Contreras, and Reconstitution Inflammatory Syndrome.” AIDS 24 (15): others. 2012. “Identifying Multidrug Resistant Tuberculosis 2381. Transmission Hotspots Using Routinely Collected Data.” Melendez, J., C. I. Sánchez, R. H. Philipsen, P. Maduskar, Tuberculosis 92 (3): 273–79. R. Dawson, and others. 2016. “An Automated Tuberculosis 304 Major Infectious Diseases Screening Strategy Combining X-Ray-Based Computer- Moore, D. A. J. 2016. “What Can We Offer to 3 Million MDRTB Aided Detection and Clinical Information.” Scientific Household Contacts in 2016?” BMC Medicine 14: 64. Reports 6: 25265. doi:10.1038/srep25265. Moyo, S., J. J. Furin, J. Hughes, J. Daniels, L. Snyman, and Menzies, N. A., T. Cohen, H.-H. Lin, M. Murray, and others. 2014. “Outcomes in Adolescents Undergoing J. A. Saloman. 2012. “Population Health Impact and Cost- Treatment for Drug-Resistant Tuberculosis in Cape Town, Effectiveness of Tuberculosis Diagnosis with Xpert MTB/ South Africa, 2008–2013.” Archives of Pediatric Infectious RIF: A Dynamic Simulation and Economic Evaluation.” Diseases 3 (3): e17934. PLoS Medicine 9 (11): e1001347. Mphaphlele, M., A. S. Dharmadhikari, P. A. Jensen, S. N. Rudnick, Meressa, D., R. M. Hurtado, J. R. Andrews, E. Diro, K. Abato, T. H. van Reenen, and others. 2015. “Institutional Tuberculosis and others. 2015. “Achieving High Treatment Success for Transmission Controlled Trial of Upper Room Ultraviolet Air Multidrug-Resistant TB in Africa: Initiation and Scale-Up Disinfection: A Basis for New Dosing Guidelines.” Amerian of MDR TB Care in Ethiopia: An Observational Cohort Journal of Respiratory and Critical Care Medicine 192 (4): 477–84. Study .” Thorax 70 (12): 1181–88. Mukherjee, J. S., S. Shin, J. Furin, M. L. Rich, F. Léandre, and Merle, C. S., K. Fielding, O. B. Sow, M. Gninafon, M. B. Lo, others. 2002. “New Challenges in the Clinical Management and others. 2014. “A Four-Month Gatifloxacin-Containing of Drug-Resistant Tuberculosis.” Infectious Diseases in Regimen for Treating Tuberculosis.” New England Journal of Clinical Practice 11 (6): 329–39. Medicine 371 (17): 1588–98. Muniyandi, M., R. Rajeswari, R. Balasubramanian, and Middelkoop, K., B. Mathema, L. Myer, E. Shashkina, A. Whitelaw, P. R. Narayanan. 2008. “A Comparison of Costs to Patients and others. 2015. “Transmission of Tuberculosis in a South with Tuberculosis Treated in a DOTS Programme with African Community with a High Prevalence of HIV Infection.” Those in a Non-DOTS Programme in South India.” Journal Journal of Infectious Diseases 211 (1): 53–61. of Health Management 10 (1): 9–24. Mitchison, D. A. 2004. “Antimicrobial Therapy of Tuberculosis: Muniyandi, M., R. Ramachandran, R. Balasubramanian, and Justification for Currently Recommended Treatment P. R. Narayanan. 2006. “Socio-Economic Dimensions of Regimens.” Seminars in Respiratory and Critical Care Tuberculosis Control: Review of Studies over Two Decades Medicine 24 (3): 307–15. from Tuberculosis Research Center.” Journal of Communicable Miti, S., V. Mfungwe, P. Reijer, and D. Maher. 2003. “Integration Diseases 38 (3): 204. of Tuberculosis Treatment in a Community-Based Home Muniyandi, M., R. Ramachandran, P. G. Gopi, V. Chandrasekaran, Care Programme for Persons Living with HIV/AIDS in R. Subramani, and others. 2007. “The Prevalence of Tuberculosis Ndola, Zambia.” International Journal of Tuberculosis and in Different Economic Strata: A Community Survey from Lung Disease 7 (9, Suppl 1): S92–98. South India [Short Communication].” International Journal of Mitnick, C. D., J. Bayona, E. Palacios, S. Shin, J. Furin, and Tuberculosis and Lung Disease 11 (9): 1042–45. others. 2003. “Community-Based Therapy for Multidrug- Munro, S. A., S. A. Lewin, H. J. Smith, M. E. Engel, A. Fretheim, Resistant Tuberculosis in Lima, Peru.” New England Journal and others. 2007. “Patient Adherence to Tuberculosis of Medicine 348 (2): 119–28. Treatment: A Systematic Review of Qualitative Research.” Mitnick, C. D., S. S. Shin, K. J. Seung, M. L. Rich, S. S. Atwood, PLoS Medicine 4 (7): e238. and others. 2008. “Extensively Drug-Resistant Tuberculosis: Murray, C. J., K. Styblo, and A. Rouillon. 1990. “Tuberculosis A Comprehensive Treatment Approach.” New England in Developing Countries: Burden, Intervention and Cost. Journal of Medicine 359 (6): 563–74. Bulletin of the International Union Against Tuberculosis and Mitwalli, A. 1991. “Tuberculosis in Patients on Maintenance Lung Disease 65 (December 8): 6–24. Dialysis.” American Journal of Kidney Diseases 18 (5): 579–82. Nakiyingi, L., V. M. Moodley, Y. C. Manabe, M. P. Nicol, Moalosi, G., K. Floyd, J. Phatshwane, T. Moeti, N. Binkin, M. Holshouse, and others. 2014. “Diagnostic Accuracy of a and T. Kenyon. 2003. “Cost-Effectiveness of Home-Based Rapid Urine Lipoarabinomannan Test for Tuberculosis in Care versus Hospital Care for Chronically Ill Tuberculosis HIV-Infected Adults.” Journal of Acquired Immune Patients, Francistown, Botswana.” International Journal of Deficiency Syndromes 66 (3): 270–79. Tuberculosis and Lung Disease 7 (Suppl 1): S80–85. Naranbhai, V. 2016. “The Role of Host Genetics (and Genomics) Modlin, R. L., and B. R. Bloom. 2013. “TB or Not TB: That Is in Tuberculosis.” Microbiology Spectrum 4 (5). doi: 10.1128 No Longer the Question.” Science Translational Medicine /microbiolspec.TBTB2-0011-2016. 5 (213): 213–16 Nardell, E., and A. Dharmadhikari. 2010. “Turning off the Mohammed-Rajput, N. A., D. C. Smith, B. Mamlin, P. Biondich, Spigot: Reducing Drug-Resistant Tuberculosis Transmission and B. N. Doebbeling. 2011. “OpenMRS, a Global Medical in Resource-Limited Settings.” International Journal of Records System Collaborative: Factors Influencing Successful Tuberculosis and Lung Disease 14 (10): 1233. Implementation.” AMIA Annual Symposium Proceedings 2011 Nathanson, E., C. Lambregts-van Weezenbeek, M. Rich, (October 22): 960–68. R. Gupta, J. Bayona, and others. 2006. “Multidrug-Resistant Moonan, P. K., T. N. Quitugua, J. M. Pogoda, G. Woo, G. Drewyer, Tuberculosis Management in Resource-Limited Settings.” and others. 2011. “Does Directly Observed Therapy (DOT) Emerging Infectious Diseases 12 (9): 1389. Reduce Drug Resistant Tuberculosis?” BMC Public Health Nathanson, E., P. Nunn, M. W. Uplekar, K. Floyd, E. Jaramillo, 11 (1): 19. and others. 2010. “MDR Tuberculosis: Critical Steps for Tuberculosis 305 Prevention and Control.” New England Journal of Medicine Palacios, J. J., J. Ferro, N. Ruiz Palma, J. M. García, H. Villar, 363 (11): 1050–58. and others. 1999. “Fully Automated Liquid Culture System Nganda, B., J. Wang’ombe, K. Floyd, and J. Kangangi. 2003. Compared with Löwenstein-Jensen Solid Medium for “Cost and Cost-Effectiveness of Increased Community Rapid Recovery of Mycobacteria from Clinical Samples.” and Primary Care Facility Involvement in Tuberculosis European Journal of Clinical Microbiology and Infectious Care in Machakos District, Kenya.” International Journal of Diseases 18 (4): 265–73. Tuberculosis and Lung Disease 7 (9, Suppl 1): S14–20. Palmer, C., and M. W. Long 1966. “Effects of Infection Nunn, A. J., I. D. Rusen, A. Van Deun, G. Torrea, P. P. Phillips, with Atypical Mycobacteria on BCG Vaccination and and others. 2014. “Evaluation of a Standardized Treatment Tuberculosis.” American Review of Respiratory Disease Regimen of Anti-Tuberculosis Drugs for Patients with Multi- 94 (4): 553–68. Drug-Resistant Tuberculosis (STREAM): Study Protocol for Pantoja, A., K. Floyd, K. P. Unnikrishnan, R. Jitendra, M. R. Padma, a Randomized Controlled Trial.” Trials 15 (September 9): 353. and others. 2009. “Economic Evaluation of Public-Private Obihara, C. C., N. Beyers, R. P. Gie, P. C. Potter, B. J. Marais, and Mix for Tuberculosis Care and Control, India. Part I. Socio- others. 2005. “Inverse Association between Mycobacterium Economic Profile and Costs among Tuberculosis Patients.” tuberculosis Infection and Atopic Rhinitis in Children.” International Journal of Tuberculosis and Lung Disease 13 (6): Allergy 60 (9): 1121–25. 698–704. OECD (Organisation for Economic Co-operation and Pantoja, A., K. Lönnroth, S. S. Lal, L. S. Chauhan, M. W. Development), Eurostat, WHO (World Health Organization). Uplekar, and others. 2009. “Economic Evaluation of Public- 2011. A System of Health Accounts. Paris: OECD Publishing. Private Mix for Tuberculosis Care and Control, India. Part http://www.who.int/health-accounts/methodology/sha2011 II. Cost and Cost-Effectiveness.” International Journal of .pdf. Tuberculosis and Lung Disease 13 (6): 705–12. O’Garra, A., P. S. Redford, F. W. McNab, C. I. Bloom, Pasipanodya, J. G., and T. Gumbo. 2013. “A Meta-Analysis of R. J. Wilkinson, and others. 2013. “The Immune Response Self-Administered versus Directly Observed Therapy Effect in Tuberculosis.” Annual Review of Immunology 31 (March): on Microbiologic Failure, Relapse, and Acquired Drug 475–527. Resistance in Tuberculosis Patients.” Clinical Infectious O’Grady, J., M. Hoelscher, R. Atun, M. Bates, P. Mwaba, Diseases 57 (1): 21–31. and others. 2011. “Tuberculosis in Prisons in Sub- Perumal, R., N. Padayatchi, and E. Stiefvater. 2009. “The Whole Saharan Africa: The Need for Improved Health Services, Is Greater Than the Sum of the Parts: Recognising Missed Surveillance, and Control.” Tuberculosis 91 (2): 173–78. Opportunities for an Optimal Response to the Rapidly O’Grady, J., M. Maeurer, R. Atun, I. Abubakar, P. Mwaba, and Maturing TB-HIV Co-Epidemic in South Africa.” BMC others. 2011. “Tuberculosis in Prisons: Anatomy of Global Public Health 9 (1): 243. Neglect.” European Respiratory Journal 38 (4): 752–54. Petsch, B., M. Schnee, A. B. Vogel, E. Lange, B. Hoffmann, and Okello, D., K. Floyd, F. Adatu, R. Odeke, and G. Gargioni. others. 2012. “Protective Efficacy of in vitro Synthesized, 2003. “Cost and Cost-Effectiveness of Community-Based Specific mRNA Vaccines against Influenza A Virus Care for Tuberculosis Patients in Rural Uganda.” The Infection.” Nature Biotechnology 30 (12): 1210–16. International Journal of Tuberculosis and Lung Disease 7 (9, Pevzner, E. S., G. Vandebriel, D. W. Lowrance, M. Gasana, Suppl 1): S72–79. and A. Finlay. 2011. “Evaluation of the Rapid Scale-Up of Okot-Chono, R., F. Mugisha, F. Adatu, E. Madraa, R. Dlodlo, Collaborative TB/HIV Activities in TB Facilities in Rwanda, and others. 2009. “Health System Barriers Affecting the 2005–2009.” BMC Public Health 11 (1): 550. Implementation of Collaborative TB-HIV Services in Philipsen, R. H., C. I. Sánchez, P. Maduskar, J. Melendez, Uganda.” International Journal of Tuberculosis and Lung L. Peters-Bax, and others. 2015. “Automated Chest- Disease 13 (8): 955–61. Radiography as a Triage for Xpert Testing in Resource- Owens, J. P., M. O. Fofana, and D. W. Dowdy. 2013. “Cost- Constrained Settings: A Prospective Study of Diagnostic Effectiveness of Novel First-Line Treatment Regimens for Accuracy and Costs.” Scientific Reports 5:12215. doi: Tuberculosis.” International Journal of Tuberculosis and 10.1038/srep12215. Lung Disease 17 (May): 590–96. Phillips, M., R. N. Cataneo, R. Condos, G. A. Ring Erickson, Oxlade, O., and M. Murray. 2012. “Tuberculosis and Poverty: J. Greenberg, and others. 2007. “Volatile Biomarkers of Why Are the Poor at Greater Risk in India?” PLoS One Pulmonary Tuberculosis in the Breath.” Tuberculosis 87 (1): 7 (11): e47533. 44–52. Pai, M., A. Daftary, and S. Satyanarayana. 2016. “TB Control: Phillips, P. P., C. M. Mendel, D. A. Burger, A. M. Crook, Challenges and Opportunities for India.” Transactions of A. J. Nunn, and others. 2016. “Limited Role of Culture the Royal Society of Tropical Medicine and Hygiene 110 (3): Conversion for Decision-Making in Individual Patient 158–60. Care and for Advancing Novel Regimens to Confirmatory Pai, M., S. Kalantri, and K. Dheda. 2006. “New Tools and Clinical Trials.” BMC Medicine 14 (February 4): 19. Emerging Technologies for the Diagnosis of Tuberculosis: Phiri, S., P. Khan, A. D. Grant, D. Gareta, H. Tweya, and Part II. Active Tuberculosis and Drug Resistance.” Expert others. 2011. “Integrated Tuberculosis and HIV Care in Review of Molecular Diagnostics 6 (3): 423–32. a Resource-Limited Setting: Experience from the Martin 306 Major Infectious Diseases Preuss Centre, Malawi.” Tropical Medicine and International Patients Who Generate Secondary Cases.” International Health 16 (11): 1397–403. Journal of Tuberculosis and Lung Disease 1 (4): 352–57. Pichenda, K., K. Nakamura, A. Morita, M. Kizuki, K. Seino, and Rodrigues, L. C., V. K. Diwan, and J. G. Wheeler. 1993. T. Takano. 2012. “Non-Hospital DOT and Early Diagnosis “Protective Effect of BCG against Tuberculous Meningitis of Tuberculosis Reduce Costs while Achieving Treatment and Miliary Tuberculosis: A Meta-Analysis.” International Success.” International Journal of Tuberculosis and Lung Journal of Epidemiology 22 (6): 1154–58. Disease 16 (6): 828–34. Ross, J. D., and J. C. Willison. 1971. “The Relationship between Png, E., B. Alisjahbana, E. Sahiratmadja, S. Marzuki, R. Nelwan, Tuberculin Reactions and the Later Development of and others. 2012. “A Genome Wide Association Study of Tuberculosis: An Investigation among Edinburgh School Pulmonary Tuberculosis Susceptibility in Indonesians.” Children in 1960–1970.” Tubercle 52 (4): 258–65. BMC Medical Genetics 13 (January 13): 5. Rouillon, A., S. Perdrizet, and R. Parrot. 1976. “Transmission Podewils, L. J., M. T. S. Gler, M. I. Quelapio, and M. P. Chen. of Tubercle Bacilli: The Effects of Chemotherapy.” Tubercle 2013. “Patterns of Treatment Interruption among Patients 57 (4): 275–99. with Multidrug-Resistant TB (MDR TB) and Association Rouzier, V. A., O. Oxlade, R. Verduga, L. Gresely, and D. Menzies. with Interim and Final Treatment Outcomes.” PLoS One 2010. “Patient and Family Costs Associated with Tuberculosis, 8 (7): e70064. Including Multidrug-Resistant Tuberculosis, in Ecuador.” Pooran, A., E. Pieterson, M. Davids, G. Theron, and K. Dheda. The International Journal of Tuberculosis and Lung Disease 2013. “What Is the Cost of Diagnosis and Management of 14 (10): 1316–22. Drug Resistant Tuberculosis in South Africa?” PLoS One Roy, A., M. Eisenhut, R. J. Harris, L. C. Rodrigues, S. Sridhar, 8 (1): e54587. and others. 2014. “Effect of BCG Vaccination Against Pop-Eleches, C., H. Thirumurthy, J. P. Habyarimana, J. G. Zivin, Mycobacterium tuberculosis Infection in Children: Systematic M. P. Goldstein, and others. 2011. “Mobile Phone Technologies Review and Meta-analysis.” BMJ 349 (August 5): g4643. Improve Adherence to Antiretroviral Treatment in a Resource- Sallusto, F. 2016. “Heterogeneity of Human CD4(+) T Cells Limited Setting: A Randomized Controlled Trial of Text against Microbes.” Annual Review of Immunology 34 (May Message Reminders.” AIDS 25 (6): 825–34. 24): 317–34. Prasad, K., and M. B. Singh. 2008. “Corticosteroids for Salomon, J. A., J. Lloyd-Smith, W. M. Getz, S. Resch, Managing Tuberculous Meningitis.” Cochrane Database of M. S. Sánchez, and others. 2006. “Prospects for Advancing Systematic Reviews 23 (1): CD002244. Tuberculosis Control Efforts through Novel Therapies.” Rachow, A., A. Zumla, N. Heinrich, G. Rojas-Ponce, B. Mtafya, PLoS Medicine 3 (8): e273. and others. 2011. “Rapid and Accurate Detection of Samandari, T., T. B. Agizew, S. Nyirenda. Z. Tedla, T. Sibanda, Mycobacterium tuberculosis in Sputum Samples by Cepheid and others. 2011. “6-Month Versus 36-Month Isoniazid Xpert MTB/RIF Assay: A Clinical Validation Study.” PLoS Preventive Treatment for Tuberculosis in Adults with HIV One 6 (6): e20458. Infection in Botswana: A Randomised, Double-Blind, Ramma, L., H. Cox, L. Wilkinson, N. Foster, L. Cunnama, and Placebo-Controlled Trial.” The Lancet 377 (9777): 1588–98. others. 2015. “Patients’ Costs Associated with Seeking and Samb, B., T. Evans, M. Dybul, R. Atun, J.-P. Moatti, and oth- Accessing Treatment for Drug-Resistant Tuberculosis in ers. 2009. “An Assessment of Interactions between Global South Africa.” International Journal of Tuberculosis and Lung Health Initiatives and Country Health Systems.” The Lancet Disease 19 (12): 1513–19. 373 (9681): 2137–69. Rangaka, M. X., S. C. Cavalcane, B. J. Marais, S. Thim, Satyanarayana, S., S. A. Nair, S. S. Chadha, R. Shivashankar, N. A. Martinson, and others. 2015. “Controlling the Seedbeds G. Sharma, and others. 2011. “From Where Are Tuberculosis of Tuberculosis: Diagnosis and Treatment of Tuberculosis Patients Accessing Treatment in India? Results from a Cross- Infection.” The Lancet 386 (10010): 2344–53. Sectional Community Based Survey of 30 Districts.” PLoS Rehm, J., A. V. Samokhvalov, M. G. Neuman, R. Room, C. Parry, One 6 (9): e24160. and others. 2009. “The Association between Alcohol Use, Schaaf, H. S., B. J. Marais, A. C. Hesseling, W. Brittle, and Alcohol Use Disorders and Tuberculosis (TB). A Systematic P. R. Ronald. 2009. “Surveillance of Antituberculosis Review.” BMC Public Health 9 (1): 450. Drug Resistance among Children from the Western Cape Resch, S. C., J. A. Salomon, M. Murray, and M. C. Weinstein. Province of South Africa—An Upward Trend.” American 2006. “Cost-Effectiveness of Treating Multidrug-Resistant Journal of Public Health 99 (8): 1486–90. Tuberculosis.” PLoS Medicine 3 (7): e241. Schnippel, K., S. Rosen, K. Shearer, N. Martinson, L. Long, Riley, R., and A. Moodie. 1974. “Infectivity of Patients with and others. 2013. “Costs of Inpatient Treatment for Multi- Pulmonary Tuberculosis in Inner City Homes.” American Drug-Resistant Tuberculosis in South Africa.” Tropical Review of Respiratory Disease 110 (6): 810–12. Medicine and International Health 18 (1): 109–16. Roberts, M., W. Hsiao, P. Berman, and M. Reich. 2004. Getting Schouten, E. J., A. Jahn, D. Midiani, S. D. Makombe, Health Reform Right: A Guide to Improving Performance and A. Mnthambala, and others. 2011. “Prevention of Mother- Equity. Oxford: Oxford University Press. to-Child Transmission of HIV and the Health-Related Rodrigo, T., J. A. Caylà, G. de Olalla P., H. Galdós-Tangüis, Millennium Development Goals: Time for a Public Health J. M. Jansà, and others. 1997. “Characteristics of Tuberculosis Approach.” The Lancet 378 (9787): 282–84. Tuberculosis 307 Scott, L. E., K. McCarthy, N. Gous, M. Nduna, A. Van Rie, Shigayeva, A., R. Atun, M. McKee, and R. Coker. 2010. “Health and others. 2011. “Comparison of Xpert MTB/RIF with Systems, Communicable Diseases and Integration.” Health Other Nucleic Acid Technologies for Diagnosing Pulmonary Policy and Planning 25 (Suppl 1): i4–20. Tuberculosis in a High HIV Prevalence Setting: A Prospective Shilova, M. V., and C. Dye. 2001. “The Resurgence of Study.” PLoS Medicine 8 (7): e1001061. Tuberculosis in Russia.” Philosophical Transactions of the Seddon, J. A., H. E. Jenkins, L. Liu, T. Cohen, R. E. Black, and Royal Society of London B: Biological Sciences 356 (1411): others. 2015. “Counting Children with Tuberculosis: Why 1069–75. Numbers Matter.” International Journal of Tuberculosis and Shin, S. S., A. D. Pasechnikov, I. Y. Gelmanova, G. G. Peremitin, Lung Disease 19 (Suppl 1): 9–16. A. K. Strelis, and others. 2006. “Treatment Outcomes in Seebregts, C. J., B. W. Mamlin, P. G. Biondich, H. S. F. Fraser, an Integrated Civilian and Prison MDR-TB Treatment B. A. Wolfe, and others. 2009. “The OpenMRS Implementers Program in Russia.” The International Journal of Tuberculosis Network.” International Journal Medical Informatics 78 (11): and Lung Disease 10 (4): 402–8. 711–20. Shin, S. S., M. Yagui, L. Ascencios, G. Yale, C. Suarez, and others. Selwyn, P. A., D. Hartel, V. A. Lewis, E. E. Schoenbaum, 2008. “Scale-Up of Multidrug-Resistant Tuberculosis S. H. Vermund, and others. 1989. “A Prospective Study of the Laboratory Services, Peru.” Emerging Infectious Diseases Risk of Tuberculosis among Intravenous Drug Users with 14 (5): 701. Human Immunodeficiency Virus Infection.” New England Shirakawa, T., T. Enomoto, S. Shimazu, and J. M. Hopkin. 1997. Journal of Medicine 320 (9): 545–50. “The Inverse Association between Tuberculin Responses Sengupta, A., and S. Nundy. 2005. “The Private Health Sector and Atopic Disorder.” Science 275 (5296): 77–79. in India.” BMJ 331 (7526): 1157–58. Shirtcliffe, P., M. Weatherall, and R. Beasley. 2002. “An Inverse Seung, K. J., D. B. Omatayo, S. Keshavjee, J. J. Furin, P. E. Farmer, Correlation between Estimated Tuberculosis Notification and others. 2009. “Early Outcomes of MDR-TB Treatment Rates and Asthma Symptoms.” Respirology 7 (2): 153–55. in a High HIV-Prevalence Setting in South Africa.” PLoS One Siedner, M. J., A. J. Lankowski, M. Kanyesigye, M. B. Bwana, 4 (9): e7186. J. E. Haberer, and others. 2015. “A Combination SMS and Shah, N. S., S. C. Auld, J. C. M. Brust, B. Mathema, N. Ismail, and Transportation Reimbursement Intervention to Improve others. 2017. “Transmission of Extensively Drug-Resistant HIV Care Following Abnormal CD4 Test Results in Rural Tuberculosis in South Africa.” New England Journal of Uganda: A Prospective Observational Cohort Study.” BMC Medicine 376: 243–53. Medicine 13 (1): 160. Shaler, C. R., C. N. Horvath, M. Jeyanathan, and Z. Xing. Sinanovic, E., K. Floyd, L. Dudley, V. Azevedo, R. Grant, and 2013. “Within the Enemy’s Camp: Contribution of others. 2003. “Cost and Cost-Effectiveness of Community- the Granuloma to the Dissemination, Persistence and Based Care for Tuberculosis in Cape Town, South Africa.” Transmission of Mycobacterium tuberculosis.” Frontiers in International Journal of Tuberculosis and Lung Disease Immunology 4 (February 14): 30. 7 (9, Suppl 1): S56–62. Shapiro, A. E., E. Variava, M. H. Rakgokong, N. Moodley, B. Luke, Sinanovic, E., and L. Kumaranayake. 2006. “Financing and and others. 2012. “Community-Based Targeted Case Finding Cost-Effectiveness Analysis of Public-Private Partnerships: for Tuberculosis and HIV in Household Contacts of Patients Provision of Tuberculosis Treatment in South Africa.” Cost with Tuberculosis in South Africa.” American Journal of Effectiveness and Resource Allocation 4: 11. Respiratory and Critical Care Medicine 185 (10): 1110–16. Sinanovic, E., J. Moodley, M. A. Barone, S. Mall., S. Cleary, and Shargie, E. B., O. Mørkve, and B. Lindtjørn. 2006. “Tuberculosis others. 2009. “The Potential Cost-Effectiveness of Adding Case-Finding through a Village Outreach Programme a Human Papillomavirus Vaccine to the Cervical Cancer in a Rural Setting in Southern Ethiopia: Community Screening Programme in South Africa.” Vaccine 27 (44): Randomized Trial.” Bulletin of the World Health Organization 6196–202. 84 (2): 112–19. Sinanovic, E., L. Ramma, A. Vassall, V. Azevedo, L. Wilkinson, Sheriff, F. G., K. P. Manji, M. M. Chagani, R. M. Mpembeni, and others. 2015. “Impact of Reduced Hospitalisation on A. M. Jusabani, and others. 2010. “Latent Tuberculosis the Cost of Treatment for Drug-Resistant Tuberculosis in among Pregnant Mothers in a Resource Poor Setting South Africa.” International Journal of Tuberculosis and Lung in Northern Tanzania: A Cross-Sectional Study.” BMC Disease 19 (2): 172–78. Infectious Diseases 10 (1): 52. Singh, A., D. Parasher, D. S. Shekhavat, S. Sahu, D. Wares, Shetty, N., M. Shemko, M. Vaz, and G. D’Souza. 2006. and others. 2004. “Effectiveness of Urban Community “An Epidemiological Evaluation of Risk Factors for Volunteers in Directly Observed Treatment of Tuberculosis Tuberculosis in South India: A Matched Case Control Patients: A Field Report from Haryana, North India [Notes Study.” International Journal of Tuberculosis and Lung from the Field].” International Journal of Tuberculosis and Disease 10 (1): 80–86. Lung Disease 8 (6): 800–802. Shi, W., X. Zhang, X. Jiang, H. Yuan, J. S. Lee, and others. 2011. Slama, K., C. Chiang, D. A. Enarson, K. Hassmiller, A. Fanning, “Pyrazinamide Inhibits Trans-Translation in Mycobacterium and others. 2007. “Tobacco and Tuberculosis: A Qualitative tuberculosis.” Science 333 (6049): 1630–32. Systematic Review and Meta-Analysis [Review Article].” 308 Major Infectious Diseases International Journal of Tuberculosis and Lung Disease 11 Styblo, K. 1991. Epidemiology of Tuberculosis: Selected Papers. (10): 1049–61. The Hague: KNCV Tuberculosis Foundation. Smart, T. 2010. “Decentralised, Patient-Centered Models of Suarez, P. G., K. Floyd, J. Portocarrero, E. Alarcon, E. Rapiti, Delivering Treatment and Palliative Care for People with and others. 2002. “Feasibility and Cost-Effectiveness of M/XDR-TB.” HIV and AIDS Treatment in Practice 166 Standardised Second-Line Drug Treatment for Chronic (October 8): 2–10. Tuberculosis Patients: A National Cohort Study in Peru.” Smieja, M. J., C. A. Marchetti, D. J. Cook, and F. M. Smaill. The Lancet 359 (9322): 1980–89. 2000. “Isoniazid for Preventing Tuberculosis in Non-HIV Subbaraman, R., R. R. Nathavitharana, S. Satyanarayana, M. Pai, Infected Persons.” Cochrane Database of Systematic Reviews B. E. Thomas, and others. 2016. “The Tuberculosis Cascade of 2: CD001363. Care in India’s Public Sector: A Systematic Review and Meta- Snider, D. Jr. 1978. “The Relationship between Tuberculosis Analysis.” PLoS Medicine 13 (10): e1002149. doi:10.1371 and Silicosis.” American Review of Respiratory Disease 118 /journal.pmed.1002149. (3): 455–60. Sudre, P., G. ten Dam, and A. Kochi. 1992. “Tuberculosis: A Sobota, R. S., C. M. Stein., N. Kodaman, L. B. Scheinfeldt, I. Maro, Global Overview of the Situation Today.” Bulletin of the and others. 2016. “A Locus at 5q33.3 Confers Resistance to World Health Organization 70 (2): 149. Tuberculosis in Highly Susceptible Individuals.” American Sutherland, I., and V. Springett. 1987. “Effectiveness of BCG Vacci- Journal of Human Genetics 98 (3): 514–24. nation in England and Wales in 1983.” Tubercle 68 (2): 81–92. Soltan, V., A. K. Henry, V. Crudu, and I. Zatusevski. 2008. Sutherland, I., E. Svandova, and S. Radhakrishna. 1982. “Increasing Tuberculosis Case Detection: Lessons from “The Development of Clinical Tuberculosis Following the Republic of Moldova.” Bulletin of the World Health Infection with Tubercle Bacilli: 1. A Theoretical Model Organization 86 (1): 71–76. for the Development of Clinical Tuberculosis Following Soysal, A., K. A. Millington, M. Bakir, D. Dosanjh, Y. Aslan, Infection, Linking from Data on the Risk of Tuberculous and others. 2005. “Effect of BCG Vaccination on Risk Infection and the Incidence of Clinical Tuberculosis in the of Mycobacterium tuberculosis Infection in Children Netherlands.” Tubercle 63 (4): 255–68. with Household Tuberculosis Contact: A Prospective Swanson, R. C., A. Cattaneo, E. Bradley, S. Chunharas, Community-Based Study.” The Lancet 366 (9495): 1443–51. R. Atun, and others. 2012. “Rethinking Health Systems Spertini, F., R. Audran, R. Chakour, O. Karoui, V. Steiner- Strengthening: Key Systems Thinking Tools and Strategies Monard, and others. 2015. “Safety of Human Immunisation for Transformational Change.” Health Policy and Planning with a Live-Attenuated Mycobacterium tuberculosis Vaccine: 27 (Suppl 4): iv54–64. A Randomised, Double-Blind, Controlled Phase I Trial.” The Sweeney, T. E., L. Braviak, C. M. Tato, and P. Khatri. 2016. Lancet Respiratory Medicine 3 (12): 953–62. “Genome-Wide Expression for Diagnosis of Pulmonary Spisak, C., L. Morgan, R. Eichler, J. Rosen, B. Serumaga, and Tuberculosis: A Multicohort Analysis.” The Lancet others. 2016. “Results-Based Financing in Mozambique’s Respiratory Medicine 4 (3): 213–24. Central Medical Store: A Review after 1 Year.” Global Health: Sylla, L., R. D. Bruce, A. Kamarulzaman, and F. L. Altice. 2007. Science and Practice 4 (1): 2–13. “Integration and Co-Location of HIV/AIDS, Tuberculosis, Sreeramareddy, C. T., Z. Z. Qin, S. Satyanarayana, R. Subbaraman, and Drug Treatment Services.” International Journal of Drug and M. Pai. 2014. “Delays in Diagnosis and Treatment of Policy 18 (4): 306–12. Pulmonary Tuberculosis in India: A Systematic Review.” Szot, A., F. Jacobson, S. Munn, D. Jazayeri, E. Nardell, and others. International Journal of Tuberculosis and Lung Disease 18 (3): 2004. “Diagnostic Accuracy of Chest X-rays Acquired Using 255–66. a Digital Camera for Low-Cost Teleradiology.” International Stead, W. W., and A. K. Dutt. 1989. “Tuberculosis in the Journal of Medicine 73 (1): 65–73. Elderly.” Seminars in Respiratory Infections 4: 189–97. TAG (Treatment Action Group). 2016. Report on Tuberculosis Sterling, T. R., M. E. Villarino, A. S. Borisov, N. Shang, F. Gordin, Research Funding Trends, 2005–2015: No Time to Lose. and others. 2011. “Three Months of Rifapentine and New York: TAG. http://www.treatmentactiongroup.org Isoniazid for Latent Tuberculosis Infection.” New England /sites/default/files/TB_FUNDING_2016. Journal of Medicine 365: 2155–66. Talat, N., S. Perry, J. Parsonnet, G. Dawood, and R. Hussain. Stop TB Partnership. 2011. “The Global Plan to Stop TB.” 2010. “Vitamin D Deficiency and Tuberculosis Progression.” World Health Organization, Geneva. Emerging Infectious Diseases 16 (5): 853–55. ———. 2016. “The ‘Zero TB Initiative’ Sparks New Action to Tameris, M. D., M. Hatherill, B. S. Landry, T. J. Scriba, End TB.” World Health Organization, Geneva, July 26. http:// M. A. Snowden, and others. 2013. “Safety and Efficacy of us3.campaign-archive2.com/?u=85207b84f0f2d8ddc9bd878de MVA85A, A New Tuberculosis Vaccine, in Infants Previously &id=6f3e04fcc9&e=abd33e01a4. Vaccinated with BCG: A Randomised, Placebo-Controlled Stop TB Working Group on DOTS-Plus for MDR-TB. Phase 2b Trial.” The Lancet 381 (9871): 1021–28. 2003. “A Prioritised Research Agenda for DOTS-Plus for Temprano ARNS Study Group. 2015. “A Trial of Early Multidrug-Resistant Tuberculosis (MDR-TB).” International Antiretrovirals and Isoniazid Preventative Therapy in Journal of Tuberculosis and Lung Disease 7 (5): 410–14. Africa.” New England Journal of Medicine 373 (9): 808–22. Tuberculosis 309 Theron, G., H. E. Jenkins, F. Cobelens, I. Abubakar, A. J. Khan, and Udwadia, Z. F., L. M. Pinto, and M. W. Uplekar. 2010. others. 2015. “Data for Action: Collection and Use of Local “Tuberculosis Management by Private Practitioners in Data to End Tuberculosis.” The Lancet 386 (10010): 2324–33. Mumbai, India: Has Anything Changed in Two Decades?” Theron, G., L. Zijenah, D. Chanda, P. Clowes, A. Rachow, and PLoS One 5 (8): e12023. others. 2014. “Feasibility, Accuracy, and Clinical Effect of Ukwaja, K. N., O. Modebe, C. Igwenyi, and I. Alobu. 2012. “The Point-of-Care Xpert MTB/RIF Testing for Tuberculosis in Economic Burden of Tuberculosis Care for Patients and Primary-Care Settings in Africa: A Multicentre, Randomised, Households in Africa: A Systematic Review.” International Controlled Trial.” The Lancet 383 (9915): 424–35. Journal of Tuberculosis and Lung Disease 16 (6): 733–39. Thye, T., E. Owusu-Dabo, F. O. Vannberg, R. van Crevel, UN (United Nations) General Assembly. 2000. “United Nations J. Curtis, and others. 2012. “Common Variants at 11p13 Millennium Declaration.” United Nations General Assembly Are Associated with Susceptibility to Tuberculosis.” Nature Resolution A/RES/55/2, UN, New York. Genetetics 44 (3): 257–59. Uplekar, M. W., S. Juvekar, S. Morankar, S. Rangan, and P. Nunn. Toczek, A., H. Cox, P. du Cros, G. Cooke, and N. Ford. 1998. “Tuberculosis Patients and Practitioners in Private 2013. “Strategies for Reducing Treatment Default in Clinics in India.” International Journal of Tuberculosis and Drug-Resistant Tuberculosis: Systematic Review and Lung Disease 2 (4): 324–29. Meta-Analysis [Review Article].” International Journal of Uplekar, M. W., V. Pathania, and M. Raviglione. 2001. “Private Tuberculosis and Lung Disease 17 (3): 299–307. Practitioners and Public Health: Weak Links in Tuberculosis Tornee, S., J. Kaewkungwal, W. Fungladda, U. Silachamroon, Control.” The Lancet 358 (9285): 912–16. P. Akarasewi, and others. 2005. “The Association between Uplekar, M. W., and D. S. Shepard. 1991. “Treatment of Environmental Factors and Tuberculosis Infection among Tuberculosis by Private General Pracitioners in India.” Household Contacts.” Southeast Asian Journal of Tropical Tubercle 72 (4): 284–90. http://www.sciencedirect.com Medicine and Public Health 36 (Suppl 4): 221–24. /science/article/pii/004138799190055W. Toungoussova, O. S., G. Bjune, and D. A. Caugant. 2006. Uplekar, M. W., D. Weil, K. Lonnroth, E. Jaramillo, C. Lienhardt, “Epidemic of Tuberculosis in the Former Soviet Union: and others. 2015. “WHO’s New End Tuberculosis Strategy.” Social and Biological Reasons.” Tuberculosis 86 (1): 1–10. The Lancet 385 (9979): 1799–801. Uwimana, J., and D. Jackson. 2013. “Integration of Tuberculosis Trajman, A., M. L. Bastos, M. Belo, J. Calaça, J. Gaspar, and and Prevention of Mother-to-Child Transmission of HIV others. 2016. “Shortened First-Line TB Treatment in Brazil: Programmes in South Africa.” International Journal of Potential Cost Savings for Patients and Health Services.” Tuberculosis and Lung Disease 17 (10): 1285–90. BMC Health Services Research 16: 27. Uwimana, J., D. Jackson, H. Hausler, and C. Zarowsky. 2012. Travis, P., S. Bennett, A. Haines, T. Pang, Z. Bhutta, and “Health System Barriers to Implementation of Collaborative others. 2004. “Overcoming Health-Systems Constraints to TB and HIV Activities Including Prevention of Mother to Achieve the Millennium Development Goals.” The Lancet Child Transmission in South Africa.” Tropical Medicine and 364 (9437): 900–906. International Health 17 (5): 658–65. Trébucq, A., D. Enarson, C. Y. Chiang, A. Van Deun, A. D. Harries, Uwinkindi, F., S. Nsanzimana, D. J. Riedel, R. Muhayimpundu, and others. 2011. “Xpert® MTB/RIF for National Tuberculosis E. Remera, and others. 2014. “Scaling Up Intensified Programmes in Low-Income Countries: When, Where and Tuberculosis Case Finding in HIV Clinics in Rwanda.” How?” International Journal of Tuberculosis and Lung Disease Journal of Acquired Immune Deficiency Syndromes 66 (2): 15 (12): 1567–72. e45–49. Trunz, B. B., P. Fine, and C. Dye. 2006. “Effect of BCG Vaccination Uyei, J., D. Coetzee, J. Macinko, and S. Guttmacher. 2011. on Childhood Tuberculous Meningitis and Miliary “Integrated Delivery of HIV and Tuberculosis Services Tuberculosis Worldwide: A Meta-Analysis and Assessment in Sub-Saharan Africa: A Systematic Review.” The Lancet of Cost-Effectiveness.” The Lancet 367 (9517): 1173–80. Infectious Diseases 11 (11): 855–67. Tseng, C.-L., O. Oxlade, D. Menzies, A. Aspler, and K. Schwartzman. Uys, P. W., R. Warren, P. D. van Helden, M. Murray, and T. C. 2011. “Cost-Effectiveness of Novel Vaccines for Tuberculosis Victor. 2009. “Potential of Rapid Diagnosis for Controlling Control: A Decision Analysis Study.” BMC Public Health Drug-Susceptible and Drug-Resistant Tuberculosis in 11 (1): 55. Communities Where Mycobacterium tuberculosis Infections Tuberculosis Prevention Trial. 1979. “Trial of BCG Vaccines in Are Highly Prevalent.” Journal of Clinical Microbiology South India for Tuberculosis Prevention: First Report.” 1979. 47 (5): 1484–90. Bulletin of the World Health Organization 57 (5): 819–27. van der Werf, M. J., M. W. Langendam, E. Huitric, and D. Manissero. Tupasi, T. E., R. Gupta, M. I. Quelapio, R. B. Orillaza, 2012. “Multidrug Resistance after Inappropriate Tuberculosis N. R. Mira, and others. 2006. “Feasibility and Cost- Treatment: A Meta-Analysis.” European Respiratory Journal Effectiveness of Treating Multidrug-Resistant Tuberculosis: 39 (6): 1511–19. A Cohort Study in the Philippines.” PLoS Medicine 3 van Rie, A., K. McCarthy, L. Scott, A. Dow, W. D. Venter, and (September 12): e352. others. 2013. “Prevalence, Risk Factors and Risk Perception Udwadia, Z. F., R. A. Amale, K. K. Ajbani, and C. Rodrigues. of Tuberculosis Infection among Medical Students and 2012. “Totally Drug-Resistant Tuberculosis in India.” Healthcare Workers in Johannesburg, South Africa.” South Clinical Infectious Diseases 54 (4): 579–81. African Medical Journal 103 (11): 853–57. 310 Major Infectious Diseases van’t Hoog, A. H., K. F. Laserson, W. A. Githui, H. K. Meme, in Hong Kong: Attributable to High Risks of Reactivation J. A. Agaya, and others. 2011. “High Prevalence of Disease.” Epidemiology and Infection 136 (7): 943–52. Pulmonary Tuberculosis and Inadequate Case Finding in Wallis, R. S., and C. Nacy. 2013. “Early Bactericidal Activity Rural Western Kenya.” American Journal of Respiratory and of New Drug Regimens for Tuberculosis.” The Lancet 381 Critical Care Medicine 183 (9): 1245–53. (9861): 111–12. van’t Hoog, A. H., H. K. Meme, K. F. Laserson, J. A. Agaya, Walton, D. A., P. E. Farmer, W. Lambert, F. Léandre, S. P. B. G. Muchiri, and others. 2012. “Screening Strategies Koenig, and others. 2004. “Integrated HIV Prevention and for Tuberculosis Prevalence Surveys: The Value of Chest Care Strengthens Primary Health Care: Lessons from Rural Radiography and Symptoms.” PLoS One 7 (7): e38691. Haiti.” Journal of Public Health Policy 25 (2): 137–58. Vassall, A. 2013. “Cost-Effectiveness of Introducing Bedaquiline Walzl, G., K. Ronacher, J. F. D. Siawaya, and H. M. Dockrell. in MDR-TB Regimens: An Exploratory Analysis.” WHO, 2008. “Biomarkers for TB Treatment Response: Challenges Geneva. and Future Strategies.” Journal of Infection 57 (2): 103–09. Vassall, A., S. Bagdadi, H. Bashour, H. Zaher, and P. V. Maaren. Wandwalo, E., N. Kapalata, S. Egwaga, and S. Morkve. 2004. 2002. “Cost-Effectiveness of Different Treatment Strategies “Effectiveness of Community-Based Directly Observed for Tuberculosis in Egypt and Syria.” International Journal Treatment for Tuberculosis in an Urban Setting in Tanzania: of Tuberculosis and Lung Disease 6 (12): 1083–90. A Randomised Controlled Trial.” International Journal of Vassall, A., Y. Chechulin, I. Raykhert, N. Osalenko, S. Svetlichnaya, Tuberculosis and Lung Disease 8 (10): 1248–54. and others. 2009. “Reforming Tuberculosis Control in Wandwalo, E., B. Robberstad, and O. Morkve. 2005. “Cost Ukraine: Results of Pilot Projects and Implications for the and Cost-Effectiveness of Community Based and Health National Scale-Up of DOTS.” Health Policy and Planning Facility Based Directly Observed Treatment of Tuberculosis 24 (1): 55–62. in Dar es Salaam, Tanzania.” Cost Effectiveness and Resource Vassall, A., S. van Kampen, H. Sohn, J. S. Michael, K. R. John, Allocation 3 (1): 6. and others. 2011. “Rapid Diagnosis of Tuberculosis with the Weir, R., G. Black, B. Nazareth, S. Floyd, S. Stenson, and Xpert MTB/RIF Assay in High Burden Countries: A Cost- others. 2006. “The Influence of Previous Exposure to Effectiveness Analysis.” PLoS Medicine 8 (11): e1001120. Environmental Mycobacteria on the Interferon—Gamma Verguet, S., R. Laxminarayan, and D. T. Jamison. 2015. Response to Bacille Calmette–Guérin Vaccination in “Universal Public Finance of Tuberculosis Treatment in Southern England and Northern Malawi.” Clinical and India: An Extended Cost-Effectiveness Analysis.” Health Experimental Immunology 146 (3): 390–99. Economics 24 (3): 318–32. Weiss, P., W. Chen, V. J. Cook, and J. C. Johnston. 2014. Verma, G., R. E. Upshur, E. Rea, and S. R. Benatar. 2004. “Treatment Outcomes from Community-Based Drug “Critical Reflections on Evidence, Ethics and Effectiveness Resistant Tuberculosis Treatment Programs: A Systematic in the Management of Tuberculosis: Public Health and Review and Meta-Analysis.” BMC Infectious Diseases 14 Global Perspectives.” BMC Medical Ethics 5 (1): 2. (June 17): 333. Verver, S., R. M. Warren, N. Beyers, M. Richardson, G. D. Wells, W. A., M. W. Uplekar, and M. Pai. 2015. “Achieving van der Spuy, and others. 2005. “Rate of Reinfection Systemic and Scalable Private Sector Engagement in Tuberculosis after Successful Treatment Is Higher than Rate Tuberculosis Care and Prevention in Asia.” PLoS Medicine of New Tuberculosis.” American Journal of Respiratory and 12 (6): e1001842. Critical Care Medicine 171 (12): 1430–35. Were, M. C., J. Kariuki, V. Chepng’eno, M. Wandabwa, S. Ndege, Victor, T., E. Streicher, C. Kewley, A. M. Jordaan, G. D. and others. 2009. “Leapfrogging Paper-Based Records Using van der Spuy, and others. 2007. “Spread of an Emerging Handheld Technology: Experience from Western Kenya.” Mycobacterium tuberculosis Drug-Resistant Strain in the Studies in Health Technology and Informatics 160 (1): 525–29. Western Cape of South Africa.” International Journal of WHO (World Health Organization). 1994. “WHO Tuberculosis Tuberculosis and Lung Disease 11 (2): 195–201. Programme: Framework for Effective Tuberculosis Control.” Villemagne, B., C. Crauste, M. Flipo, A. R. Baulard, B. Déprez, WHO, Geneva. and others. 2012. “Tuberculosis: the Drug Development ———. 1999. The World Health Report 1999: Making a Pipeline at a Glance.” European Journal of Medicinal Difference. Geneva: WHO. Chemistry 51 (May): 1–16. ———. 2000. The World Health Report 2000: Health Systems, Vledder, M., P. Yadav, J. Friedman, M. Sjoblom, amd T. Brown. Improving Performance.” Geneva: WHO. 2015. “Optimal Supply Chain Structure for Distributing ———. 2003. The World Health Report 2003: Shaping The Essential Drugs in Low-Income Countries: Results from a Future. Geneva: WHO. Randomized Experiment.” Ross School of Business Paper ———. 2004. “Interim Policy on Collaborative TB/HIV 1269, University of Michigan, Ann Arbor. Activities.” WHO, Geneva. von Mutius, E., N. Pearce, R. Beasley, S. Cheng, O. von ———. 2006. “Improving the Diagnosis and Treatment Ehrenstein, and others. 2000. “International Patterns of of Smear-Negative Pulmonary and Extrapulmonary Tuberculosis and the Prevalence of Symptoms of Asthma, Tuberculosis among Adults and Adolescents.” WHO, Stop Rhinitis, and Eczema.” Thorax 55 (6): 449–53. TB Department, Geneva. Vynnycky, E., M. Borgdorff, C. C. Leung, C. M. Tam, and ———. 2007. Global Tuberculosis Control: Surveillance, P. E. M. Fine. 2008. “Limited Impact of Tuberculosis Control Planning, Financing. Geneva: WHO. Tuberculosis 311 ———. 2009a. “Global Tuberculosis Control: A Short Update Willingham, F. F., T. L. Schmitz, M. Contreras, S. E. Kalangi, A. M. to the 2009 Report.” WHO, Geneva. Vivar, and others. 2001. “Hospital Control and Multidrug- ———. 2009b. “WHO Policy on TB Infection Control Resistant Pulmonary Tuberculosis in Female Patients, Lima, in Health-Care Facilities, Congregate Settings, and Peru.” Emerging Infectious Diseases 7 (1): 123. Households.” WHO, Geneva. Wolfe, F., K. Michaud, J. Anderson, and K. Urbansky. 2004. ———. 2010. “Multidrug and Extensively Drug-Resistant TB: “Tuberculosis Infection in Patients with Rheumatoid 2010 Global Report on Surveillance and Response.” WHO, Arthritis and the Effect of Infliximab Therapy.” Arthritis Geneva. and Rheumatism 50 (2): 372–79. ———. 2011a. “Guidelines for the Programmatic Management Wood, R., S. D. Lawn, J. Caldwell, R. Kaplan, K. Middelkoop, and of Multidrug Resistant Tuberculosis.” WHO, Geneva. others. 2011. “Burden of New and Recurrent Tuberculosis ———. 2011b. “Towards Universal Access to Diagnosis and in a Major South African City Stratified by Age and HIV- Treatment of Multidrug-Resistant and Extensively Drug- Status.” PLoS One 6 (10): e25098. Resistant Tuberculosis by 2015.” WHO Progress Report Wood, R., S. D. Lawn, S. Johnstone-Robertson, and L.-G. 2011, WHO, Geneva. Bekker. 2011. “Tuberculosis Control Has Failed in South ———. 2011c. “Tuberculosis Country Profiles: South Africa.” Africa: Time to Reappraise Strategy.” South African Medical WHO, Geneva. http://www.who.int/tb/country/data/profiles Journal 101 (2): 111–14. /en/index.html. Wood, R., S. Johnstone-Robertson, P. Uys, J. Hargrove, ———. 2012a. Global Tuberculosis Report. Geneva: WHO. K. Middelkeep, and others. 2010. “Tuberculosis Transmission ———. 2012b. “WHO Policy on Collaborative TB/HIV to Young Children in a South African Community: Modeling Activities: Guidelines for National Programmes and Other Household and Community Infection Risks.” Clincal Stakeholders.” WHO, Geneva. Infectious Diseases 51 (4): 401–8. ———. 2013a. “Consolidated Guidelines on the Use of Wood, R., K. Middelkoop, L. Myer, A. D. Grant, A. Whitelaw, and Antiretroviral Drugs for Treating and Preventing HIV others. 2007. “Undiagnosed Tuberculosis in a Community Infection.” WHO, Geneva. with High HIV Prevalence: Implications for Tuberculosis ———. 2013b. Global Tuberculosis Report 2013. Geneva: Control.” American Journal of Respiratory and Critical Care WHO. Medicine 175 (1): 87–93. ———. 2013c. Roadmap for Childhood Tuberculosis. Geneva: Woolhouse, M. E., C. Dye, J.-F. Etard, T. Smith, J. D. Charlwood, WHO. http://apps.who.int/iris/bitstream/10665/89506/1 and others. 1997. “Heterogeneities in the Transmission of /9789241506137_eng.pdf. Infectious Agents: Implications for the Design of Control ———. 2013d. “Systematic Screening for Active Tuberculosis: Programs.” Proceedings of the National Academy of Sciences Principles and Recommendations.” WHO, Geneva. http:// 94 (1): 338–42. www.who.int/tb/tbscreening. World Bank. 1993. World Development Report 1993: Investing in ———. 2013e. “The Use of Bedaquiline in the Treatment of Health. Washington, DC: World Bank. Multidrug-Resistant Tuberculosis: Interim Policy Guidance.” Wu, P., E. H. Y. Lau, B. J. Cowling, C.-C. Leung, C.-M. Tam, and WHO, Geneva. others. 2010. “The Transmission Dynamics of Tuberculosis ———. 2014a. Global Status Report on Alcohol and Health in a Recently Developed Chinese City.” PLoS One 5 (May 3): 2014. Geneva: WHO. e10468. ———. 2014b. Global Tuberculosis Report 2014. Geneva: WHO. Yadav, P. 2010. “In-Country Supply Chains: The Weakest Link ———. 2014c. “Tuberculosis.” Fact Sheet 104, WHO, Geneva. in the Health System.” Global Health Magazine 5 (Winter): http://www.who.int/mediacentre/factsheets/fs104/en/. 18–20. ———. 2015a. “The End TB Strategy.” WHO, Geneva. http:// ———. 2015. “Health Product Supply Chains in Developing www.who.int/tb/End_TB_brochure.pdf. Countries: Diagnosis of the Root Causes of Underperformance ———. 2015b. Global Tuberculosis Report. Geneva: WHO. and an Agenda for Reform.” Health Systems and Reform 1 (2): ———. 2015c. “The End TB Strategy.” Fact Sheet, WHO, Geneva. 142–54. http://www.who.int/tb/post2015_TBstrategy.pdf?ua=1. Yadav, P., O. Stapleton, and L. Van Wassenhove. 2013. “Learning ———. 2016a. Global Tuberculosis Report 2016. Geneva: WHO. from Coca-Cola.” Stanford Social Innovation Review ———. 2016b. WHO Treatment Guidelines for Drug-Resistant (Winter): n.p. Tuberculosis. 2016 Update. Geneva: WHO. http://www.who Yadav, P., H. L. Tata, and M. Babaley. 2011. The World Medicines .int/tb/MDRTBguidelines2016.pdf. Situation 2011: Storage and Supply Chain Management. ———. n.d. “The Five Elements of DOTS.” WHO, Geneva. Geneva: WHO. http://www.who.int/tb/dots/whatisdots/en/. Yin, X., X. Tu, Y. Tong, R. Yang, Y. Wang, and others. 2012. Wilkinson, R. J., M. Llewelyn, Z. Toossi, P. Patel, and G. Pasvol. “Development and Validation of a Tuberculosis Medication 2000. “Influence of Vitamin D Deficiency and Vitamin D Adherence Scale.” PLoS One 7 (12): e50328. Receptor Polymorphisms on Tuberculosis among Gujarati Yokoyama, T., R. Sato, T. Rikimaru, R. Hirai, and H. Aizawa. Asians in West London: A Case-Control Study.” The Lancet 2004. “Tuberculosis Associated with Gastrectomy.” Journal 355 (9204): 618–21. of Infection and Chemotherapy 10 (5): 299–302. 312 Major Infectious Diseases Yuen, C. M., F. Amanullah, A. Dharmadhikari, E. A. Nardell, Molecular Genetic Data.” Journal of Infectious Diseases J. A. Seddon, and others. 2015. “Turning Off the Tap: 213 (2): 287–94. Stopping Tuberculosis Transmission through Active Case- Zhao, Y., S. Xu, L. Wang, D. P. Chin, S. Wang, and others. Finding and Prompt Effective Treatment.” The Lancet 2012. “National Survey of Drug-Resistant Tuberculosis 386 (10010): 2334–43. in China.” New England Journal of Medicine 366 (23): Zachariah, R., M.-P. L. Spielmann, C. Chinji, P. Gomani, 2161–70. V. Arendt, and others. 2003. “Voluntary Counselling, HIV Zumla, A., P. Nahid, and S. T. Cole. 2013. “Advances in the Testing, and Adjunctive Cotrimoxazole Reduces Mortality Development of New Tuberculosis Drugs and Treatment in Tuberculosis Patients in Thyolo, Malawi.” AIDS 17 (7): Regimens.” Nature Reviews Drug Discovery 12 (5): 1053–61. 388–404. Zak, D., A. Penn-Nicholson, T. Scriba, E. Thompson, S. Suliman, Zwarenstein, M., L. R. Fairall, C. Lombard, P. Mayers, A. Bheekie, and others. 2016. “A Blood RNA Signature for Tuberculosis and others. 2011. “Outreach Education for Integration of Disease Risk: A Prospective Cohort Study.” The Lancet HIV/AIDS Care, Antiretroviral Treatment, and Tuberculosis 387 (10035): 2312–22. Care in Primary Care Clinics in South Africa: PALSA PLUS Zar, H. J., M. F. Cotton, S. Strauss, J. Karpakis, G. Hussey, and Pragmatic Cluster Randomised Trial.” BMJ 342 (April 21): others. 2007. “Effect of Isoniazid Prophylaxis on Mortality d2022. and Incidence of Tuberculosis in Children with HIV: Zwerling, A., G. B. Gomez, J. Pennington, F. Cobelens, A. Vassall, Randomised Controlled Trial.” BMJ 334 (7585): 136. and D. W. Dowdy. 2016. “A Simplified Cost-Effectiveness Zelner, J., M. B. Murray, M. C. Becerra, J. Galea, L. Lecca, Model to Guide Decision-Making for Shortened and others. 2016. “Identifying Hotspots of Multidrug Anti-Tuberculosis Treatment Regimens.” International Resistant Tuberculosis Transmission Using Spatial and Journal of Tuberculosis and Lung Disease 20 (2): 257–60. Tuberculosis 313 Chapter 12 Malaria Elimination and Eradication Rima Shretta, Jenny Liu, Chris Cotter, Justin Cohen, Charlotte Dolenz, Kudzai Makomva, Gretchen Newby, Didier Ménard, Allison Phillips, Allison Tatarsky, Roly Gosling, and Richard Feachem INTRODUCTION halting and beginning to reverse the global incidence of The world has made tremendous progress in the fight malaria by 2015. against malaria in the past 15 years. According to the An increasing number of countries are moving World Malaria Report, malaria case incidence was toward the elimination of malaria. Since 2000, 12 coun- reduced by 41 percent and malaria mortality rates were tries have eliminated malaria; 4 were certified as malaria reduced by 62 percent between 2000 and 2015 (WHO free by the World Health Organization (WHO) between 2016c). At the beginning of 2016, malaria was consid- 2007 and 2013 (Armenia, Morocco, Turkmenistan, and ered to be endemic in 91 countries and territories, down the United Arab Emirates); an additional 8 moved from 108 in 2000. into the WHO’s prevention-of-reintroduction phase Despite this progress, malaria continues to place a after sustaining at least three years of zero local malaria heavy toll on the world. In 2015, 212 million cases transmission (Argentina, the Arab Republic of Egypt, occurred globally, leading to 429,000 deaths, most of Iraq, Georgia, the Kyrgyz Republic, Oman, the Syrian which occurred in children under age five years in Arab Republic, and Uzbekistan); and 5 interrupted Africa. These estimates are likely to be conservative, as local transmission (Azerbaijan, Costa Rica, Paraguay, adult deaths from malaria might well be underesti- Sri Lanka,1 and Turkey). The WHO European Region mated in Africa and India (Adjuik and others 2006; reported zero indigenous cases for the first time in 2015, Bawah and Binka 2007; Dhingra and others 2010; in line with the goal of the Tashkent Declaration to Gupta and Chowdhury 2014). eliminate malaria from the region by 2015. More than 100 countries have eliminated malaria in According to the WHO (2016a), an additional 21 the past century. Of the 106 countries with ongoing countries are in a position to achieve at least one year transmission in 2000, 57 reduced malaria incidence of zero indigenous cases of malaria by 2020.2 These more than 75 percent by 2015, in line with the World dramatic declines can be attributed to the scale-up of Health Assembly target for 2015 of reducing the effective malaria control tools and technologies cou- malaria burden by 75 percent. An additional 18 coun- pled with renewed political leadership and financial tries reduced incidence by more than 50 percent commitment. (WHO 2015e), also achieving target 6C of the Bolstered by these successes, most national malaria Millennium Development Goals, which called for programs now consider elimination to be an Corresponding author: Rima Shretta, University of California, San Francisco, Global Health Sciences, San Francisco, California, United States; Rima.Shretta@ucsf.edu. 315 Table 12.1 Global Milestones and Targets for Elimination Milestones Target Goal 2020 2025 2030 Reduce malaria mortality rates globally At least 40% At least 75% At least 90% compared with 2015. Reduce malaria case incidence globally At least 40% At least 75% At least 90% compared with 2015. Eliminate malaria from countries in which At least 10 countries At least 20 countries At least 35 countries malaria was transmitted in 2015. Prevent reestablishment of malaria in all Reestablishment Reestablishment prevented Reestablishment prevented countries that are malaria free. prevented Sources: RBM Partnership 2015; WHO 2015a. attainable goal, and the idea of eradication is once Democratic Republic of Congo and Nigeria—accounting again on the global health agenda. Many countries for more than 35 percent of global malaria deaths. In have developed national elimination goals, and these areas, malaria control programs aim to maximize regional networks have been formed to facilitate col- the reduction of malaria cases and deaths; elimination laboration (Newby and others 2016). Leaders from the will likely require more potent tools and stronger health Asia Pacific Leaders Malaria Alliance and the African systems. Leaders Malaria Alliance endorsed regional goals for A few countries that have successfully reduced malaria malaria elimination by 2030 in November 2014 and transmission are struggling to maintain their gains. An January 2015, respectively, galvanizing support for increased number of cases has recently been reported elimination and eradication (APLMA 2015; United from a number of countries, including Cambodia, Nations 2015). Djibouti, Madagascar, Uganda, and República Bolivariana In this context, two new global malaria policy and advo- de Venezuela (WHO 2015e). Furthermore, as the global cacy documents supporting elimination and eradication malaria burden declines, emerging biological threats have were released in 2015: the Roll Back Malaria (RBM) the potential to critically weaken malaria responses in Partnership’s Action and Investment to Defeat Malaria several parts of the world. In 2014, 60 countries reported 2016–2030 and the WHO’s Global Technical Strategy for resistance of mosquitoes to at least one insecticide used in Malaria 2016–2030. The Global Technical Strategy (GTS), vector control strategies; resistance of parasites to artem- which the WHO ratified in May 2015, calls for at least isinin, the cornerstone of malaria chemotherapy, has another 40 percent reduction in malaria-related mortality been detected in five countries in the Greater Mekong and morbidity between 2015 and 2020. Other goals and subregion, posing a serious threat to global health targets are illustrated in table 12.1. A third document, security. launched in September 2015, From Aspiration to Action: This chapter summarizes the literature on malaria What Will It Take to End Malaria?, outlines the resources elimination; describes the progress made; and discusses and strategies needed for global eradication by 2040, call- malaria epidemiology, interventions, and challenges. In ing by 2020 commit to eradication in the next five years addition, it presents empirical information on financing (Gates and Chambers 2015). and economics, including cost information from various Despite these advances, malaria elimination and settings. It concludes with a discussion of the economic eradication face significant technical, operational, and basis for eradication and recommendations for research. financial challenges. About 3.2 billion people remain at risk of malaria; in 2015 alone, there were an estimated 214 million new cases of malaria and more than WHAT ARE ELIMINATION AND 400,000 malaria-related deaths. Global progress in malaria control and elimination is marked by vast dis- ERADICATION? parities between and within countries, with vulnerable In areas of moderate to high transmission that are imple- groups that have poor access to health services contin- menting malaria control, interventions are deployed on uing to be marginalized. The Sub-Saharan Africa region a large scale to reduce the public health burden of the shoulders the heaviest burden, with two countries—the disease. In elimination settings, targeted interventions 316 Major Infectious Diseases aim to interrupt local transmission in the specific places cases between 1957 and 2003 (Filler and others 2006). where it becomes increasingly concentrated, that is, Even countries that do not contiguously border endemic small geographic areas or special subpopulations that neighbors experience considerable importation annually: may be harder and costlier to reach. The key decisions Sri Lanka reported 49 confirmed imported malaria cases facing policy makers in low- and moderate-transmission in 2014, and in Tanzania, Zanzibar’s estimated importa- settings are when to embark on malaria elimination tion of 1.6 cases per 1,000 residents could potentially (Sabot and others 2010); which interventions to imple- produce 1,300 incident cases (Le Menach and others ment and where and when; and at what levels of inten- 2011). Transmission from imported cases may lead to sity and reach. Critical to this debate are the political and first degree introduced cases; a second degree of transmis- financial commitments that are needed long after the sion from an introduced case produces an indigenous disease stops being a public health burden. case: both are products of local transmission. Elimination Malaria elimination involves stopping indigenous accordingly requires preventing all indigenous cases, but transmission through active control measures (Cohen introduced cases may continue to occur sporadically. and others 2010; Smith and others 2009). The complete As more countries and regions eliminate malaria absence of local incidence is very unlikely to be achieved and implement measures to prevent reintroduction, in places with high intrinsic potential for transmission fewer imported infections will occur, and eradication and elevated importation of cases (Cohen and others will become increasingly feasible. See box 12.1 for the 2010). For example, even the United States, a relatively WHO definitions of control, elimination, and low transmission risk area, identified 156 locally acquired eradication. Box 12.1 Definitions of Control, Elimination, and Eradication The path to malaria-free status is characterized by four of the disease. The WHO grants this certification distinct programmatic phases: control, pre-elimination, when a country has proved, beyond reasonable elimination, and prevention of reintroduction. doubt, that the chain of local malaria transmission The terms elimination and eradication are often by Anopheles mosquitoes has been interrupted used interchangeably. For example, eradication was nationwide for at least three consecutive years. previously used to describe what is now defined as Certification of malaria elimination is managed by elimination (Feachem and others 2010). To compare the WHO Global Malaria Programme. The process programs across these phases, it is important to adhere is voluntary and can be initiated only after a country to agreed-upon terms and definitions. has submitted an official request to the WHO. The burden of proof falls on the country requesting cer- Malaria control is the reduction of disease incidence, tification. The final decision on granting a certifica- prevalence, morbidity, or mortality to a locally accept- tion of malaria elimination rests with the WHO able level as a result of deliberate efforts. Continued director-general. intervention is required to sustain control. Malaria eradication is a permanent reduction to zero Malaria elimination is the interruption of local trans- of the worldwide incidence of infection caused mission (that is, reducing the rate of malaria cases to by human malaria parasites as a result of deliberate zero) of a specified parasite in a defined geographic efforts. Once eradication has been achieved, interven- area. Continued measures are required to prevent the tion measures would no longer be needed. reestablishment of transmission. WHO certification of eliminationa is the WHO Source: WHO 2016a. a. Since the early 1960s, the WHO has maintained an official register of areas certification of a country’s malaria-free status. It where malaria elimination has been achieved. The WHO also maintains a confirms to the international community that the supplementary list to the official register, listing countries where malaria never country, at that time, has halted local transmission existed or disappeared years or decades ago and where full WHO certification of malaria elimination is not needed. The first supplementary list was published in of malaria by Anopheles mosquitoes and has created 1963 and included 23 countries. The most recent list was published in 2012 and an adequate system for preventing reestablishment included 62 countries (WHO 2012d). Malaria Elimination and Eradication 317 PROGRESS TOWARD MALARIA ELIMINATION Elimination in the Twenty-First Century Elimination in the Twentieth Century The adoption of the Global Malaria Control Strategy in 1992 (WHO 1993) and the launch of the Roll Back Until the mid-nineteenth century, malaria was endemic Malaria initiative in 1998 (Nabarro and Taylor 1998) in most countries across the globe. Countries that did stimulated increased interest and financial investment in not have malaria included the Pacific islands east of the malaria control. Increased investment in research and longitude of Vanuatu (the Buxton line) (Mendis and development resulted in highly effective malaria control others 2009), which have no Anopheles mosquitoes; or tools—notably, long-lasting insecticide-treated nets countries that were too high in elevation or too cold in (LLINs), rapid diagnostic tests, and artemisinin-based temperature (map 12.1). combination therapies (ACTs). The creation of the Global Between 1900 and 1945, only nine countries in Europe Fund to Fight AIDS, Tuberculosis, and Malaria; the eliminated malaria (Feachem, Phillips, and Targett 2009). President’s Malaria Initiative; and other financing mech- Sparked by the availability of chloroquine for treatment anisms allowed for the wide-scale deployment of these and dichloro-diphenyl-trichloroethane (DDT) for vec- new tools. The first Global Malaria Action Plan for a tor control, the WHO launched the Global Malaria malaria-free world 2008–2015 served as a valuable guide Eradication Program (GMEP) in 1955 to interrupt trans- for countries and partners to mobilize resources. Between mission in all endemic areas outside of Africa (Najera 2005 and 2014, global investment for malaria control 1999). The program relied on vector control—mainly increased from US$960 million to US$2.5 billion annu- indoor residual spraying—and systematic detection and ally, leading to dramatic declines in the global malaria treatment of cases. The campaign succeeded in eliminat- burden and rapid shrinking of the malaria map. With the ing malaria in 37 of the 143 countries or economies where end of the Millennium Development Goals in 2015 and it was endemic in 1950 (Wernsdorfer and Kouznetzov the transition to the era of the Sustainable Development 1980), including some lower-income areas with tropical Goals, the malaria community has once again committed climates such as Maldives; Mauritius; Réunion; Taiwan, to the vision of a malaria-free world. China; much of the Caribbean; Brunei Darussalam; most Table 12.2 summarizes the countries and territories of China; Hong Kong SAR, China; Singapore (Feachem that eliminated malaria between 1900 and 2015. and others 2010).3 In many other countries, the burden of Currently 35 countries are moving from low-endemic disease and deaths from malaria was greatly reduced. For malaria to elimination (Newby and others 2016). These example, in India, the number of malaria cases declined countries fit into one of two categories: (1) countries from an estimated 110 million in 1955 to fewer than that have assessed the feasibility of elimination, declared 1 million in 1968, and in Sri Lanka, the incidence of a national evidence-based goal, and launched a malaria malaria declined from an estimated 2.8 million cases in elimination strategy; or (2) those that are strongly con- 1946 to just 18 cases in 1966 (Mendis and others 2009). sidering an evidence-based national elimination goal as However, failure to sustain strong funding for the pro- determined by expert opinion, have made substantial gram, particularly in the face of increasing costs due to progress in spatially progressive elimination (by elimi- mounting drug and insecticide resistance, led to the effec- nating malaria from specific islands or geographic tive end of the GMEP in 1969 (WHO 1969) when the areas), and are greatly reducing malaria nationwide. World Health Assembly recommended that countries not These 35 countries have elimination goals ranging from yet ready for “eradication” focus on controlling malaria as 2013 to 2035, with the majority aiming for, and likely to a first step toward the ultimate goal of getting rid of achieve, elimination by 2020 (annex 12A). malaria altogether. Multilateral agencies withdrew their Five countries—Argentina, the Kyrgyz Republic, support for malaria programs in favor of general health Paraguay, Sri Lanka, and Uzbekistan—recently achieved programs. In the ensuing years, although most countries three consecutive years of zero local transmission. All but that had eliminated malaria continued to remain malaria Uzbekistan have initiated the WHO process for malaria- free, the scaling back of control efforts in malarious coun- free certification. Three other countries have achieved zero tries led to a global resurgence of the disease during the local transmission but have not yet sustained it for three 1970s and 1980s and a complete reversal of progress in consecutive years: Azerbaijan, Costa Rica, and Turkey. some countries, such as Sri Lanka and Pakistan (Abeyasinghe and others 2012; Cohen and others 2012). The experience of the GMEP provides critical lessons for contemporary elimination programs about the need Lessons Learned and Planning for Success to maintain vigilance and sustain investments during the Lessons learned from the GMEP highlight the fact that a latter stages of elimination efforts. single strategy is unlikely to be successful everywhere 318 Major Infectious Diseases Map 12.1 Malaria Transmission Worldwide, 1900, 1990, and 2015 IBRD 42556 | DECEMBER 2016 1900 Malaria transmission Malaria-free > – 3 years – 3 years Malaria-free < Eliminating malaria Controlling malaria 1900 2015 Source: Global Health Group 2016, unpublished data. Malaria Elimination and Eradication 319 given the complexities of malaria transmission systems, others 2009), and require prolonged expenditure (Sabot and given that a long-term commitment with a flexible and others 2010). Reducing transmission without suffi- strategy that includes community involvement, integra- ciently sustainable interventions to maintain those tion with health systems, and the development of agile reductions may also lead to epidemics and resurgence. surveillance systems with supporting infrastructure Out of 49 discontinued programs during GMEP, resur- is needed (Najera, Gonzalez-Silva, and Alonso 2011). gence was reported in 36 programs following cessation, A review conducted at the GMEP’s conclusion cited usually because of an inability to maintain sufficient the lack of robust assessments to determine the feasibil- financial resources (Cohen and others 2012). Countries ity of malaria eradication programs (WHO 1968; see should assess the technical, operational, and financial box 12.2), including an assessment of the technical and feasibility of achieving their goals (discussed further in operational evidence and government commitment to the section titled Prospects for Malaria Eradication) sustain funding. Attempting to eliminate malaria before before embarking on a costly restructuring of their it is feasible to do so can raise expectations, damage the programs (Moonen, Cohen, Tatem, and others 2010; credibility of the public health sector (Moonen and WHO 2014a). Table 12.2 Number of Countries and Territories That Eliminated Malaria, by Region, 1900–2015 Americas and South Asia and East Europe and Middle East and Sub-Saharan Indicator Caribbean Asia and Pacific Central Asia North Africa Africa Total Total number of countries 46 39 58 23 45 211 Malaria free 1900 2 13 3 1 1 20 1900–49 0 0 9 0 0 9 1950–78 23 5 35 4 1 68 1979–90 0 1 2 2 1 6 1991–2015 2 1 9 6 0 18 Total number of 27 20 58 13 3 121 malaria-free countries Box 12.2 Challenges to Elimination: Select Examples Despite the recent successes in eliminating malaria, when U.S. assistance ended, India was unable to challenges remain. The following discussion high- maintain its vector control activities. Resurgence lights these challenges and provides examples of over the next decade led to nearly 6 million cases. some actions taken to overcome them: A key priority identified in India’s current National Framework for Malaria Elimination 2016–2030 • Lack of sustained funding. India implemented a is funding its elimination plan with sustained widely successful program through DDT spray- domestic resources and innovative financing ing that reduced the malaria burden from an models, including cost-sharing partnerships and estimated 100 million annual cases in the early integration with other government departments 1900s to about 100,000 cases in 1965. However, (Government of India 2016).a box continues next page 320 Major Infectious Diseases Box 12.2 (continued) • Political instability and conflict. By 1975, malaria partnered to improve access to new insecticides for was eliminated throughout the former Soviet indoor residual spraying in 16 countries across Union. However, after its collapse in the early Africa. Their US$65 million Next Generation 1990s, efforts were disrupted by a lack of funding. Indoor Residual Spray Project will work with mul- Civil wars broke out in several of the former terri- tiple partners to make alternative insecticides more tories, such as Azerbaijan and Tajikistan, contrib- affordable. uting to resurgence and reintroduction. Overall • Importation. Four countries in southern Africa— strengthening of national health systems and cre- Botswana, Namibia, South Africa, and Swaziland— ation of national malaria control programs in 1998 are seeking to eliminate indigenous transmission and 1997, respectively, after gaining independence within the next five years, but many of their neigh- and achieving political stability allowed the malaria bors have much higher malaria burdens. Mobile situation to be brought under control rapidly in (moving within a country or coming back from both countries. abroad) and migrant (coming from elsewhere • Weak program vigilance. Mauritius achieved into the area) populations are primary sources malaria-free certification in 1973. However, when of imported cases, driving secondary transmis- the program was integrated into preventive health sion. As a result, the number of cases and deaths services, the malaria surveillance system was weak- between 2012 and 2013 rose in all four countries. ened. Vector control activities and screening were Cross-border initiatives are essential to addressing reduced, contributing to resurgence associated these challenges. with an influx of migrant workers. Through the In September 2015, the Global Fund approved combination of an active surveillance program US$17.8 million for the eight countries in southern that screened visitors from malarious areas, an Africa (Angola, Botswana, Mozambique, Namibia, integrated vector management strategy, and a South Africa, Swaziland, Zambia, and Zimbabwe) strong health system for detecting and responding termed the “Elimination 8” or “E8,” designed to to missed cases of imported or introduced malaria, serve as a platform for joint planning, negotiation, Mauritius has remained malaria free since 1998. and accountability toward a regionally synchronized • Drug and insecticide resistance. With few replace- malaria elimination effort. The main thrust of the ment options, drug and insecticide resistance is a E8 regional program is to expand access to early major threat to elimination. Multidrug resistance diagnosis and treatment for mobile and underserved emerged and spread rapidly within and outside populations and to enhance surveillance in the the Greater Mekong subregion (Cambodia, the border areas. Lao People’s Democratic Republic, Myanmar, • Weak health systems and program capacity. The Thailand, Vietnam, and China’s Yunnan Province), Solomon Islands and Vanuatu have had difficulty threatening effective treatment everywhere. In the maintaining robust malaria elimination programs Greater Mekong subregion, the WHO is leading as a result of weak health systems and limited an urgent, multipartner effort to eliminate P. falci- program capacity to deliver effective diagnosis and parum transmission by 2025. treatment to populations in remote areas. Both At the same time, resistance to pyrethroids, the have experienced periodic spikes in cases that have active ingredients used in insecticide-treated nets, proved challenging to bring under control. is expanding rapidly in Sub-Saharan Africa. In 2014, 27 countries had reported insecticide Sources: Cohen and others 2012; Manguin, Carnevale, and Mouchet 2008; resistance (Strode and others 2014). To combat Tatarsky and others 2011. a. Sustaining domestic and international funding as the malaria burden declines insecticide resistance, the Innovative Vector is a serious concern for most malaria-eliminating countries, 15 of which are Control Consortium and UNITAID have recently now upper-middle income and thus no longer eligible for donor funding. Malaria Elimination and Eradication 321 CHALLENGES AND THREATS TO SUCCESS groups in contact with infective vectors (Bhumiratana and others 2013; Chuquiyauri and others 2012; Hiwat In contrast to previous attempts at eradication, current and others 2012; Ngomane and de Jager 2012; Tobgay, efforts explicitly acknowledge that malaria eradication Torres, and Na-Bangchang 2011). Adult men often act requires a long-term effort incorporating multiple activi- as parasite reservoirs, with many low-density asymp- ties and embracing multiple interventions, disciplines, tomatic infections that, if left untreated and carried for approaches, and organizations. Success will be built largely long periods, contribute to seasonal transmission out- on a series of effective national and subregional elimina- breaks and epidemics (Harris and others 2010). High- tion programs, driving global eradication from the bottom risk populations, such as ethnic or political minorities up, with countries integrating malaria surveillance, trans- or mobile tribes, are also often hard to reach. These mission interruption, and treatment programs into their groups rarely seek treatment and face substantial barri- national health systems. Nevertheless, challenges exist. ers to accessing health care, including service delivery, and may be missed by disease surveillance systems Eliminating P. vivax (Hiwat and others 2012). In countries where both P. falciparum and P. vivax are As local transmission declines, the threat of secondary transmitted (mainly outside of Sub-Saharan Africa), transmission from importation becomes increasingly as P. falciparum malaria declines, the proportion of important. The greatest risk for importation is from infections due to P. vivax often rises. travel to and from neighboring or well-connected P. vivax accounts for more than 70 percent of malaria high-endemic areas (Cohen and others 2012; Tao and cases in low-transmission countries (those with fewer others 2011; Tatarsky and others 2011). Knowledge of the than 5,000 cases). Elimination is more difficult for P. vivax dynamics of population migration, both domestic and than for P. falciparum because of the presence of persis- international, and cross-border transmission is crucial for tent liver-stage infections (hypnozoites), the dormant developing appropriate surveillance and response mecha- form of the parasite responsible for relapses after months nisms. Researchers have used mobile phone data to infer or even years. In addition, gametocytes appear earlier in patterns of human movement (Tatem and others 2014; P. vivax than in P. falciparum, making onward transmis- Wesolowski and others 2012) and identify sources and sion more likely and more challenging to contain, because sinks of transmission; some programs are implementing eliminating P. vivax requires repeated blood-stage treat- spatial decision support systems (Le Menach and others ment or reliable approaches for dealing with the hypno- 2011; Marston and others 2014; Tatem and others 2014). zoite. P. vivax therefore persists as the main challenge to In some elimination settings, at a given time many malaria elimination, particularly in the late stages. malaria infections either are asymptomatic or cause Despite these difficulties, P. vivax has been eliminated only minor symptoms (Lindblade and others 2013). in many countries, including China, Mexico, Morocco, Passive surveillance misses those individuals who act as Turkey, Turkmenistan, and most recently Sri Lanka, parasite reservoirs that are infectious to mosquitoes, caus- through well-organized deployment of vector control and ing onward transmission (Sturrock and others 2013). effective treatment (El Khyari 2001; Shamuradova and A substantial proportion of infections may also be subpat- others 2012). In 2015, the WHO published a technical ent or submicroscopic, that is, the density of parasites is brief on the control and elimination of P. vivax highlight- lower than the threshold for detection by microscopy or ing the need for international donors and governments to rapid diagnostic tests. These infections account for invest in additional measures to control, eliminate, and 20 percent to 50 percent of all transmission occurrences in prevent its reestablishment (WHO 2015c). low-endemic settings (Mosha and others 2013; Okell and others 2012). Draining this asymptomatic reservoir is thus important for elimination. There is, however, growing Reaching High-Risk Populations certitude that curing all symptomatic infections will auto- In malaria-eliminating settings, parasite reservoirs are matically shrink this asymptomatic reservoir. increasingly clustered in high-risk populations or in geographically restricted foci of transmission (Sturrock and others 2013). As transmission decreases, incidence Addressing Artemisinin Resistance shifts from young children and pregnant women to all Resistance of parasites to artemisinin derivatives, the age groups, including older children and men. In Asia, mainstay of malaria treatment, is a mounting problem. this shift is exacerbated by occupational and behavioral Delayed parasite clearance times following artemisinin risk factors—such as collecting firewood, farming, monotherapy or ACT were first detected in Western hunting, or fighting in armed conflict—that put these Cambodia in 2007 and soon after along the Thai-Burmese, 322 Major Infectious Diseases the Thai-Cambodian, and the Cambodian-Vietnamese hardest to kill (Maude and others 2009). Second, borders (Carrara and others 2013; Dondorp and others artemisinin-resistant parasites are selected for concomi- 2009; Hien and others 2012; Phyo and others 2012). tant resistance to ACT partner drugs, resulting in high Plasmodium falciparum artemisinin resistance is evident late-treatment failure rates, as observed in Cambodia in five countries in the Greater Mekong subregion (WHO with dihydroartemisinin-piperaquine (Amaratunga and 2015b), most recently in Myanmar, just 25 kilometers others 2016; Duru and others 2015; Leang and others from the Indian border (map 12.2). Delayed parasite 2013; Leang and others 2015; Lon and others 2014; clearance times are correlated with some specific muta- Saunders and Lon 2016; Spring and others 2015) and tions (580C→Y, 539R→T, 543I→T, 493Y→H, and along the Thai-Myanmar border with artesunate-meflo- 446F→I) in the propeller domain of a Kelch protein gene quine (Carrara and others 2013). Although innovative located on chromosome 13 (PF3D7_1343700) (Ariey and compounds with different modes of action are in devel- others 2014; Straimer and others 2015). K13 mutant par- opment, they will not be ready for deployment before asites associated with artemisinin resistance are currently 2020 (Wells and Hooft van Huijsduijnen 2015; Wells, prevalent throughout mainland South-East Asia from Hooft van Huijsduijnen, and Van Voorhis 2015). southern Vietnam to central Myanmar (Ashley and oth- Therefore, novel strategies and regimens using available ers 2014; Ménard and others 2016; Takala-Harrison and antimalarial drugs need to be further evaluated. These others 2015). strategies may include drug rotation between different This development has major implications for malaria ACTs, extension of the three-day ACT course to five or elimination: First, parasites susceptible to artemisinin seven days, and the triple combination of artemisinin will be eliminated earliest, and the remaining parasites derivatives with two partner drugs in a three-day in low-transmission areas will be resistant and the therapy. Map 12.2 Frequency Distribution of the Wild-Type K13 Allele in Asia and Worldwide IBRD 42557 | DECEMBER 2016 Frequency of K13 non-synonymous mutant, % (sample size <30) 0% ≥4.00 to 18.00% 0.34 to 0.50% ≥18.00 to 40.00% ≥0.50 to 1.50% ≥40.00 to 66.44% ≥1.50 to 2.50% No data ≥2.50 to 4.00% Source: Ménard and others 2016. Malaria Elimination and Eradication 323 The WHO (2012a, 2015d) has labeled multidrug- efficacious treatment regimens (Ashley and others resistant malaria in the Greater Mekong subregion as a 2014; Ménard and others 2016; Takala-Harrison and regional public health disaster with the potential for others 2015). severe global consequences. In March 2015, the WHO concluded that eliminating malaria in this subregion is the only way to extend the lifespan of artemisinin deriv- MALARIA ELIMINATION INTERVENTIONS atives as an effective treatment and outlined a strategy for elimination by 2030 (WHO 2015d). AND STRATEGIES The potential spread of artemisinin resistance poses Elimination and control rely on similar interventions: a substantial risk to global health security and eco- high-quality case management, vector control, and sur- nomic development. Widespread resistance could veillance. However, while high coverage rates are desir- increase global malaria mortality by an estimated 25 able in control programs, interventions in elimination percent, with an annual economic impact of more programs must be highly targeted and tailored, and the than US$0.5 billion (Lubell and others 2014). These right tool needs to be selected according to vector and increases in mortality and in costs could undermine human behavior (table 12.3). Redistributing resources years of investments, making the case for preventing toward elimination-specific interventions, such as the spread of resistance even more compelling. strengthening surveillance systems to identify and inves- Geospatial and temporal mapping of the emergence tigate transmission foci, may produce economic effi- and spread of parasite resistance allows policy makers ciencies. However, continued investments in enhanced to mobilize resources efficiently and to adopt more program and managerial capacity are needed. Table 12.3 Key Differences between Interventions for Malaria Control and Elimination Indicator Malaria control Malaria elimination Epidemiological setting High and medium transmission Low transmission, localized, and seasonal Population at risk Entire population(s) considered to be at risk Populations living in transmission foci, high-risk groups, migrants, and mobile populations Vector control Long-lasting insecticide- Widespread coverage At-risk areas and populations; travelers to endemic areas treated nets Indoor residual spraying Widespread coverage At-risk areas and populations Larval control Larviciding Appropriate in specific circumstances where Appropriate in specific circumstances where breeding sites breeding sites can be identified and regularly can be identified and regularly targeted targeted; supplement to insecticide-treated nets and indoor residual spraying; may be better suited to urban areas Environmental management Not feasible in most high-transmission settings Feasible in targeted areas where the specific cases cannot be targeted Case management Diagnosis All suspected cases should undergo diagnostic Rapid diagnostic tests, microscopy, or both with confirmatory testing with rapid diagnostic tests or microscopy; diagnostics; quality assurance protocols implemented; highly goal is to have a confirmed diagnosis; clinical sensitive molecular diagnostic (polymerase chain reaction, diagnosis not recommended; diagnosis should loop-attenuated isothermal amplification) may be considered distinguish between parasite species; quality for quality assurance; diagnostic should distinguish between assurance protocols should be implemented parasite species Treatment P. falciparum: ACT P. falciparum: ACT plus single low dose primaquine (0.25mg/kg) table continues next page 324 Major Infectious Diseases Table 12.3 Key Differences between Interventions for Malaria Control and Elimination (continued) Indicator Malaria control Malaria elimination P. vivax: P. vivax: Blood-stage infections, chloroquine-sensitive Chloroquine-sensitive areas: Chloroquine or ACT for areas: Chloroquine or ACT blood-stage infections plus primaquine (0.25–0.5 mg/kg) Blood-stage infections, chloroquine-resistant for 14 days to ensure clearance of liver-stage infection areas: ACT or quinine during pregnancy (gametocytes) To prevent relapse: primaquine (0.25–0.5 mg/kg) Prophylaxis for travelers for 14 days G6PD deficiency: primaquine 0.75 mg/kg once a week for 8 weeks Prophylaxis for travelers Intermittent preventive treatment for pregnant n.a. women and infants Mass drug administration Seasonal malaria chemoprevention High-risk groups in geographic or demographic clusters Trials have used DHA/PIP and artemether lumefantrine accompanied by single low dose of primaquine. Surveillance Passive Monthly reporting of aggregate, confirmed cases Rapid or weekly reporting, ideally electronically, of individual to a central level cases classified by origin to a central level Active Not feasible because of high number of cases Includes case investigation, reactive case detection, proactive case detection (which may include mass screening), and foci investigation Program management Program structure Increased investment in integrated programming Vertical programming investment needed; flexibility needed in the general health system between vertical and integrated systems Human resources Large teams of dedicated staff for specific Dedicated managers; basic skills maintained among cadre of interventions; specialized skills training integrated staff High-level commitment National reduction of disease burden (morbidity, National or subnational goals of elimination; may feed into mortality) regional elimination goal; regional collaboration encouraged for controlling imported cases Source: Gosling and others 2014; RBM Partnership 2008; WHO 2012c. Note: ACT = artemisinin-based combination therapy; DHA/PIP = dihydroartemisinin-piperaquine; mg/kg = milligrams per kilogram; n.a. = not applicable. Vector Control Africa increased from 2 percent in 2000 to an estimated Vector control, a key intervention for preventing malaria 55 percent in 2015 (WHO 2015e). transmission by Anopheles mosquitoes, includes indoor However, there are threats to the sustainability of residual spraying with insecticide, use of LLINs, larvicid- these interventions. First, LLINs must be replaced at least ing, and environmental management to remove breed- every three years, and maintaining consistent use is dif- ing sites (WHO 2006). The massive gains in malaria ficult, especially when the perceived risks of malaria control in the past 15 years are attributed largely to the decline (Hsiang and others 2012). The WHO estimates scale-up of these interventions, notably LLINs (Bhatt that as many as 300 million new nets may be required and others 2015). LLINs have been most widely deployed each year to ensure that all populations at risk have in Africa, which has the highest proportion of the popu- access to LLINs in countries where LLINs are the pri- lation at risk of malaria and has malaria vectors most mary vector control strategy (WHO 2015e). Second, amenable to control with LLINs. The proportion of mosquitoes are becoming resistant to insecticides: most the population sleeping under LLINs in Sub-Saharan countries in Sub-Saharan Africa have detected resistance Malaria Elimination and Eradication 325 to pyrethroids, the main class of insecticides for indoor 2012; Helinski and others 2008; Howard and others residual spraying and LLINs (Ranson and others 2011). 2011). As an example, work to develop gene drive sys- While researchers and product development partners are tems that either suppress or replace vector populations is developing and deploying new insecticides, country proceeding (Hammond and others 2016). malaria programs must implement strategies to mitigate Ultimately, the use of an integrated approach to vec- resistance by rotating insecticides and using alternative tor control based on entomological surveillance to vector control tools (Hemingway and others 2016). understand and target unique vector behaviors and the development of new tools to target different mosquito life stages, habitats, and behaviors are essential for the Residual Transmission and New Tools for Control effective control of malaria vectors (Durnez and Despite high coverage of LLINs and indoor residual Coosemans 2013). spraying, transmission persists in many areas because of residual transmission, defined as transmission sustained Entomological Surveillance and Integrated Vector by vectors that evade contact with these two indoor Management interventions and that rest outdoors and bite humans or Robust entomological surveillance and monitoring is animals (Killeen 2014). Residual transmission poses a critical to guiding vector control interventions. particular challenge to elimination and eradication and Information on local vector species, their behaviors, and requires efficient tools to target malaria vectors. their susceptibility to insecticides as well as on coverage, Measures such as topical and spatial insect repellants usage, quality, and durability of vector control tools is (Ogoma, Moore, and Maia 2012; Wilson and others needed to inform decision making and shape local vec- 2014), insecticide-treated hammocks (Magris and others tor control strategies. Entomological expertise was the 2007), and insecticide-treated textile products (Kimani backbone of successful elimination programs in the past and others 2006; Rowland and others 1999; Thang and (Mauritius, Sri Lanka, and the United States) and should others 2009) may be more effective for protecting inform and direct future vector control strategies (Tanner individuals outdoors (Katz, Miller, and Hebert 2008). and others 2015). Innovative indoor methods such as durable wall liners Evidence-based programming and decision making and insecticidal paint could replace indoor residual and entomological intelligence are key components and spraying, and mosquito-proofed housing (using window the foundation of integrated vector management (IVM). screens) and housing modifications (closing eaves and IVM is an approach to integrated vector control that using insecticide-treated eave tubes) may be effective optimizes available resources and encourages ecological supplemental interventions (Ngufor and others 2014; soundness and sustainability. Other features of the IVM Oxborough and others 2015; Tusting and others 2015). approach include multisectoral collaboration, commu- Space spray and attract-and-kill mechanisms could tar- nity and stakeholder engagement, and integrated tools get adult vectors outdoors, and topical and systemic and structures to control disease vectors more effec- insecticide treatments for livestock can be effective for tively and efficiently (WHO 2012b). vectors that also feed on animals (Matowo and others 2013; Poché and others 2015; Pooda and others 2015; Maintenance of Low Transmission Rowland and others 2001; Shono and others 1991). The rate of progress toward elimination and the level of Researchers are also examining new approaches such interventions required to interrupt transmission depend as attractive toxic sugar baits and swarm spraying to on the strength of the health system to detect and exploit intrinsic mosquito sugar feeding and mating respond to cases; the level of investment in malaria pro- behaviors, respectively (Müller and others 2010; Qualls grams; and various other factors, including biological and others 2015). determinants, the environment, and the social, demo- More aggressive approaches to targeting immature graphic, political, and economic realities in the particu- stages of vectors, including aerial and ground larviciding lar country. Two important factors determine the risk of and breeding source reduction through environmental reestablishment of malaria: vulnerability and receptivity. management, are the mainstays of mosquito control Vulnerability is determined by the importation rate of programs in high-resource settings such as Australia and malaria into malaria-free areas; receptivity is the proba- the United States and can be considered for malaria con- bilistic risk of local mosquitos and strategies needed for trol and elimination in lower-resource settings (Floore global becoming infected with malaria parasites and 2006). Research to develop genetic and biological control subsequently transmitting the infection to humans. In of adult malaria vectors is ongoing and may be one of Canada, Europe, and the United States, vulnerability is the long-term solutions for malaria eradication (Blanford high, but receptivity is low. Thousands of imported 326 Major Infectious Diseases malaria cases arrive each year, but local mosquitoes and a few others are in the pipeline, although they are rarely become infected and transmit the infection not expected to be available in the near future. onward. In contrast, the risk of reestablishment is high in Eliminating countries also face significant threat from countries where both vulnerability and receptivity are P. vivax. Despite long being regarded as benign, acute high, such as Oman and Sri Lanka, which have previ- cases can have severe consequences. P. vivax infections ously had high rates of transmission and also receive are treated with chloroquine in areas where it remains visitors infected with malaria. In these settings, imported effective (treatment failure with chloroquine for P. vivax cases must be detected rapidly to prevent onward trans- malaria has been observed in 24 countries and con- mission to the local community. firmed in 10 countries) or with ACT where it is not. The success of achieving and sustaining elimination Primaquine, the only medicine currently available to is largely dependent on the receptivity of an area to treat hypnozoites, requires a long course of treatment malaria or “the abundant presence of anopheline vectors (7–14 days or even 8 weeks), and poor adherence can and the existence of other ecological and climatic factors lower its efficacy (John and others 2012). Furthermore, favouring malaria transmission” (WHO 2007, 84). the risk of life-threatening hemolysis in patients with Vector control is a key strategy for reducing vectorial glucose-6-phosphate dehydrogenase (G6PD) deficiency, capacity—the efficiency of the vector in transmitting a common blood disorder present in about 8 percent of malaria based on mosquito density, survival, human the population in malaria-endemic areas (Howes and biting rates, and parasite incubation period (Brady others 2012), limits its use. A reliable point-of-care test 2016). In addition, understanding the ecological and to detect G6PD deficiency is not yet widely available climatic factors that cause an increase in receptivity and (Baird 2015). Tafenoquine, a promising single-dose responding with tailored, effective vector control will be medicine against hypnozoites and relapses, is likely to be critical to elimination and eradication. available in 2018 (Eziefula and others 2012; Llanos- Cuentas and others 2014), but it has severe side effects in G6PD-deficient patients. Therefore, solving the problem Diagnosis and Treatment of G6PD diagnosis and making more sensitive, field- At present, the WHO considers quality-assured micros- deployable diagnostics more widely available have great copy the gold standard for diagnosing clinical malaria. potential for eliminating P. vivax. However, microscopy and RDTs are less sensitive at detecting low-density and subpatent infections, which Mass Drug Administration can contribute a sizable proportion of secondary cases Interest in the empiric administration of a therapeutic and onward transmission. Nucleic acid amplification antimalarial regimen to an entire population at the techniques such as polymerase chain reaction are more same time, otherwise known as mass drug administration sensitive than microscopy and RDTs and are increasingly (MDA), has recently been renewed. Proactive MDA being used in epidemiological studies; however, they are has been successfully deployed against several infec- not yet field friendly and require considerable start-up tious diseases, including lymphatic filariasis, oncho- costs and staff training. Lab-based polymerase chain cerciasis, schistosomiasis (Hotez 2009), and malaria reaction assays through pooling techniques can provide (Bruce-Chwatt 1959; Newby and others 2015; Poirot a high-throughput approach for detecting low para- and others 2013). The goal is to interrupt transmis- sitemias (Hsiang and others 2012; Imwong and others sion by treating all parasitemia in the population. 2014). However, they do not provide immediate results, MDA can potentially reduce malaria mortality and and conducting them is capital intensive. Similarly, morbidity through its direct therapeutic effect on loop-attenuated isothermal amplification can detect all individuals who receive a treatment dose of antima- species of infection at low density and high throughput, larials. It also can reduce transmission rates by reduc- is available at a relatively low marginal cost, and involves ing parasitemia prevalence and interrupting various less lab equipment, but it still requires staff capacity stages of the parasite lifecycle, and it can inhibit the (Surabattula and others 2013). The WHO recommends sporogonic cycle in the mosquito, reducing its vecto- that the use of highly sensitive diagnostic tools should rial capacity. If every member of a given population be considered only in low-transmission settings where were treated by antimalarial MDA, the prevalence of malaria diagnostic testing and treatment are already asexual parasites in the population would immedi- widely used (WHO 2014b). ately decline. ACT is the frontline therapy for uncomplicated However, knowledge gaps remain, especially regarding P. falciparum and has been widely deployed globally. The optimal size of the target population, methods to improve WHO currently recommends five ACT combinations, coverage, selection of drug-resistant parasites, and Malaria Elimination and Eradication 327 primaquine safety. Malaria elimination programs will the information on an ongoing basis to guide rapid likely use MDA in targeted ways to accelerate the impact response strategies. of vector control and ongoing diagnosis and treatment. In elimination settings, the WHO recommends inves- Current trials use a full course of dihydroartemisinin- tigation of all malaria cases to determine if they are piperaquine or artemether-lumefantrine and a single low imported or the first- (introduced) or second- (indige- dose of primaquine (Eckhoff, Gerardin, and Wenger nous) degree results of local transmission. Passive detec- 2015; White 2013). A key issue is that medicines such as tion of cases must be complemented with some form of ACTs and primaquine have been registered by drug regu- active case detection. Active case detection might take latory authorities based on a clinical indication and a the form of mass screening of high-risk individuals demonstrated risk-benefit ratio in symptomatic patients. (GHG 2013; Smith Gueye and others 2013; WHO 2013), The evidence base for its use in asymptomatic or nonin- targeted testing of specific high-risk groups, or household fected subjects will need renewed attention. In addition, visits seeking febrile or infected individuals. Active case many medicines considered for MDA are not known to detection typically costs more than passive surveillance; be safe in the first trimester of pregnancy, which presents however, the relative cost-effectiveness has not been additional problems if the medicines are deployed in assessed (Sturrock and others 2013). Less-demanding Africa, where pregnancies are rarely reported in the approaches are being explored, such as surveying chil- first trimester. dren in vaccination clinics, women in antenatal clinics, The long-term use of MDA in low-transmission set- or children attending school. tings faces several challenges. The optimum combina- Some programs proactively screen at-risk popula- tion of products and the timing, frequency, and duration tions on a periodic basis or screen the contacts of of use will depend on the endemicity, seasonality, and index cases for related infections (Moonen, Cohen, Snow, rate of importation (Newby and others 2015). For exam- and others 2010; Wickremasinghe and others 2014). ple, MDA, preferably using treatments with a long For example, migrant laborers and returning military half-life, is sensible where populations are static and the may be screened when entering a malaria-eliminating risk of importation is low (Cohen and others 2013; country, or a village may be screened before and during Gosling and others 2011). To minimize drug pressure on the malaria season to detect cases before transmission ACTs, a complete course of treatment is needed, and the begins. Focal screening and treatment of high-risk com- regimen used for MDA should differ from frontline munities and mass screening and treatment of whole treatment. At least three “rounds” of administration are populations may be used, but these approaches miss needed to affect transmission (Maude and others 2012), infected subjects who are not screened (Hoyer and requiring adequate resources and political others 2012) or persons with subpatent infections. In commitment. islands or in countries with few entry points, visitors The WHO has issued guidelines for implementation from endemic areas can be screened to prevent reintro- of MDA in different epidemiological settings (WHO duction; however, such screening is difficult to sustain. 2016a). The WHO recommends the use of MDA for the For any of these methods to be effective, diagnostic elimination of P. falciparum malaria in areas approach- tests have to be reliable and able to detect low levels of ing interruption of transmission where there is good infection, or presumptive treatment (treatment without access to treatment, effective implementation of vector a diagnostic test) can be used (WHO 2014b). control and surveillance, and minimal risk of reintroduc- Use of serology to measure past exposure could tion of infection, as well as for epidemic control and in help identify at-risk populations, especially in low- exceptional circumstances such as complex emergencies. transmission settings where infections are relatively rare Like most interventions, MDA is designed to accompany (Hsiang and others 2012). Combining serology with other interventions, including active surveillance and conventional diagnostic testing in geospatial models to vector control. produce accurate risk maps at finer scales can improve the targeting of interventions (Corran and others 2007; Epidemiological Surveillance Hsiang, Greenhouse, and Rosenthal 2014; Kelly and Robust and responsive surveillance systems that identify others 2012; Lindblade and others 2013; Sissoko and and eliminate transmission foci are critical for the suc- others 2015; Sturrock and others 2014). cess of malaria control and elimination. (Ohrt and Malaria should be made a notifiable disease (required others 2015). An ideal malaria elimination surveillance by law to be reported to government authorities) once system swiftly collects and transmits data about individ- incidence is low enough that malaria surveillance teams ual cases, classified by the origin of infection; integrates can investigate and report every individual case (Moonen, it with information on program activities; and analyzes Cohen, Snow, and others 2010). China and Swaziland 328 Major Infectious Diseases have made malaria a notifiable disease to try to increase 36 percent and 26 percent, respectively, after a booster reporting and encourage more sectors to use the surveil- dose administered 18 months after the primary series. In lance system (Cohen and others 2013; Hemingway and January 2016, the WHO released a position paper rec- others 2016). Other approaches to capturing cases that ommending further evaluation of the malaria vaccine in present outside the public sector include restricting a series of pilot implementations before considering access to antimalarials and incorporating private health wider country-level introduction (WHO 2016b). facilities into the surveillance system (Moonen, Cohen, An ideal vaccine would be more effective than RTS,S Tatem, and others 2010). at protecting individuals against infection and at stop- After elimination has been achieved, passive surveil- ping transmission of both P. falciparum and P. vivax lance at health facilities, including in the informal private (Nikolaeva, Draper, and Biswas 2015; Tran and others sector, is needed to detect and treat introduced infections. 2015). Such combinations will likely not be commer- cially available for at least another decade. Vaccines Malaria vaccines include pre-erythrocytic vaccines that aim to prevent blood-stage infection, blood-stage vac- Program Management cines that clear parasitemia and prevent clinical disease, Reorienting a program from control toward elimination and transmission-blocking vaccines that prevent infec- involves retraining staff, developing strong surveillance tion of mosquitoes and interrupt transmission (Horton capacity, building a data architecture that can monitor 2015). RTS,S, a pre-erythrocytic vaccine to prevent clin- and direct activities, instituting managerial practices that ical P. falciparum in children, is the first malaria vaccine ensure a capable and ready workforce, and changing to have completed a Phase 3 clinical trial and was program tasks from curative services to preventive com- approved by the European Medicines Agency in June munity action. These activities involve securing political 2015. Clinical trials demonstrated a vaccine efficacy and financial commitment for at least 6–10 years after for clinical malaria of 28 percent in children ages elimination has been achieved, as demonstrated by the 5–17 months, but only 18 percent in infants, the target experiences of Turkmenistan and Sri Lanka, described in population (RTSS Clinical Trials Partnership 2015) and boxes 12.3 and 12.4 (Feachem and others 2010). Box 12.3 Eliminating Malaria in Turkmenistan: Going the Last Mile Key lessons learned: its program toward eliminating transmission. The goal was reinforced by the 2005 Tashkent Declaration, • Use regional goals to drive country progress. a commitment to achieving regional elimination by • Build and sustain human resource capacity. 2015 (WHO 2005). The last local malaria case in • Maintain a dedicated budget even as priorities Turkmenistan was documented in 2004. shift. After securing high-level political and financial com- mitment, a revised elimination strategy was launched Turkmenistan eliminated malaria in the 1950s in 2007, and malaria-free status was achieved in 2010 during the Global Malaria Eradication Program. (WHO 2010). The prevention-of-reintroduction Over the next four decades, imported cases were strategy emphasized intensified surveillance at detected rapidly through a robust surveillance sys- the Afghanistan border, rapid case investigation, and tem. However, population movement after the disso- standardized reporting. Even as health priorities lution of the former Soviet Union in the 1990s led to shifted away from malaria, Turkmenistan maintained increases in local vulnerability and imported cases dedicated funding for human resources, surveillance, that escaped detection. Two P. vivax outbreaks monitoring and evaluation, and advocacy. (1998–99 and 2002–03) spurred the Turkmenistan Sources: Turkmenistan Ministry of Health and Medical Industry, WHO, and Ministry of Health and Medical Industry to reorient University of California, San Francisco, Global Health Group 2012; WHO 2005, 2010. Malaria Elimination and Eradication 329 Box 12.4 Eliminating Malaria in Sri Lanka: Flexibility under Fire Key lessons learned: and sea. Local staff conducted mobile clinics when conditions were safe and, in some areas, enlisted the cooperation of resistance fighters whose troops • Creativity and flexibility in implementation were affected by malaria. A local nongovernmental enabled practical problem solving organization with extensive presence in the • Collaboration and coordination with a range of conflict areas was enlisted to distribute long-lasting stakeholders improved program access and effi- insecticide-treated nets and provide prevention ciency even during conflict. education through volunteers. Parasitological and entomological surveillance Malaria has declined substantially in Sri Lanka in the began in 2008. By the end of the conflict in 2009, past 15 years, from 264,549 cases in 1999 to no cases in the number of cases had dropped to 558 and con- 2012. This success is particularly remarkable given the tinued to decline until 2012, when the last indige- 1983–2009 civil war, which displaced large populations nous case was recorded. Sri Lanka applied for and disrupted local health services in eight malaria- WHO certification after having achieved three endemic districts. Malaria cases peaked in 1999, with years without autochthonous transmission as of nearly 60 percent occurring in conflict districts. October 2015. The Anti-Malaria Campaign, working with the Ministry of Defense, sent essential supplies by land Sources: Sri Lanka Ministry of Health 2014; WHO and GHG 2012. Despite the need for intensified surveillance and appear to be disproportional to the marginal return. response capabilities during the elimination phase, gov- Maintaining a high level of financial support when ernments and external donors typically reduce funding transmission has been reduced to low levels remains a as incidence declines (Cohen and others 2012). Program challenge. Policy makers have to decide whether to activities are often integrated into the local health system maintain control activities indefinitely or whether to to increase efficiency (Liu and others 2013; Tatarsky and actively pursue elimination. others 2011). A review of managerial experiences with Articulating the costs of elimination and the relative disease elimination suggests that dedicated staff should benefits of investment in elimination versus control will run and oversee some tasks (vector control and rapid help inform these decisions. Three methods can be used case investigation), while local health teams could over- to assess the incremental costs and associated benefits of see others (case management, surveillance, and reporting) malaria elimination: (Gosling and others 2014). Regional collaboration can further reinforce collective • Analyzing the costs and benefits of an elimination goals and foster positive cross-border externalities and program, summarized using a benefit-cost ratio financing (Barclay, Smith, and Findeis 2012; Gosling and • Determining the financial cost savings of an elimi- others 2015; Moonen, Cohen, Snow, and others 2010). nation campaign relative to alternative scenarios (for For a description of regional initiatives, see annex 12B. example, control or resurgence costs) • Evaluating the macroeconomic impact of malaria ECONOMICS AND FINANCING OF MALARIA control and elimination against the economic burden that malaria places on society ELIMINATION One of the strongest arguments against eliminating or eradicating any disease involves the costs associated with Costs and Benefits finding and treating a decreasing number of cases Since the conclusion of the GMEP in the 1960s, several (Lines, Whitty, and Hanson 2007). These final few studies have reported the costs and consequences of cases will likely require an outlay of resources that malaria elimination and control, but few benefit-cost 330 Major Infectious Diseases Table 12.4 Benefit-Cost Ratios Associated with Malaria Elimination Programs Country or setting Study period Focus (control or elimination) Benefit-cost ratio Source Global 2010–30 Elimination 6.11 Purdy and others 2013 a Greece 1946–49 Elimination 17.09 Livadas and Athanassatos 1963 a Iraq 1958–67 Elimination 6.3 Niazi 1969 Paraguay 1965 Elimination 2.6–3.3 Ortiz 1968 India 1953–54, Controlb 9.22 Ramaiah 1980 1976–77 Control 4.14 Philippines Unspecified Controlb 2.4 Barlow and Grobar 1986 Sri Lanka 1947–55 Controlb 146.3 Barlow and Grobar 1986 2014 Prevention of reintroduction 14.3 Shretta and others 2016 b West Pakistan 1960 Control 4.9 Barlow and Grobar 1986 a. Calculated based on reported benefits and costs. b. Although the assessments considered these to be control interventions, they were conducted during the Global Malaria Eradication Program era. analyses have been conducted (table 12.4). Beyond the investment may be channeled to formerly malarious direct benefits on health, the main economic benefit areas, contributing to fiscal growth. considered in the studies is increased labor productivity Comparing the marginal benefits of control to those resulting from reductions in absenteeism. Other benefits of elimination is difficult. Elimination can improve include gains from the migration of labor into previously health equity because the last remaining foci of infection malarial areas and lower treatment costs. Most studies are often concentrated within poor or marginalized pop- assume a 10-year elimination campaign, and only two ulations (Feachem, Phillips, and Targett 2009). Prevention (Ortiz 1968; Ramaiah 1980) used empirical data. of reintroduction also protects against resurgences. All studies showed positive benefit-cost ratios, indi- Furthermore, eliminating malaria within a single country cating sizable benefits relative to costs. Benefit-cost ratios may confer substantial regional externalities and global ranged from 2.4 in the Philippines (Mills, Lubell, and public good, fostering collaboration. Elimination may Hanson 2008), 4.14 and 9.22 for control in India (Prakash also confer threshold benefits by permanently reducing and others 2003; Ramaiah 1980), 17.09 for elimination in the receptivity of an area to the reestablishment of local Greece (Livadas and Athanassatos 1963), to 146.3 and transmission (Chiyaka and others 2013; Sabot and others 14.3 for control and prevention of reintroduction, 2010; Smith Gueye and others 2013), but methods to respectively, in Sri Lanka (Barlow and Grobar 1986). Of measure the value of the diminished resurgence risk have these countries, Greece continues to report outbreaks as yet to be established. Some studies have examined the a result of imported cases, despite having eliminated relationship between elimination and tourism demand in malaria, and Sri Lanka is in the process of seeking WHO the Dominican Republic, Mauritius, and South Africa, malaria-free certification (Samaraweera 2015). but with little success because of confounding factors such as the overall increase in global travel (Maartens and Benefits others 2007; Modrek and others 2012). As benefits Many of the economic benefits associated with malaria become less tangible, they are more difficult to measure. interventions extend beyond health to include larger Gaining an understanding of the larger set of economic macroeconomic and demographic effects. Investments benefits will require better macroeconomic models that reduce private out-of-pocket expenditures on preven- quantify the links between elimination and other out- tion and treatment (Chuma, Thiede, and Molyneux comes (Mills, Lubell, and Hanson 2008). 2006; Guiguemdé and Guy 2012), increase productivity, and increase agricultural output via reclaimed land Costs and Cost Comparisons (Gallup and Sachs 2001; Mills, Lubell, and Hanson 2008; Much of the debate regarding elimination concerns Utzinger and others 2002). Lower child mortality may the government’s costs of delivering services. However, reduce fertility (Aksan and Chakraborty 2013), increase programmatic costs are only part of the picture— literacy and human capital (Lucas 2010), and eventually individuals, households, and employers also incur costs increase labor productivity. Domestic and foreign for treatment and prevention. From a programmatic Malaria Elimination and Eradication 331 perspective, costs increase as control interventions are In practice, while an abundance of literature exam- scaled up, because interventions are often provided for ines the costs of comprehensive control, studies com- free to increase coverage and to shift costs from individ- paring the costs of elimination to the costs of control to uals to programs. determine the financial cost savings of an elimination Analyses of program expenditures are limited to a few program relative to control or resurgence are scarce. studies primarily in Africa and Asia. A systematic litera- Nevertheless, once malaria is reduced to a level at which ture review identified 21 studies on the costs of malaria it is no longer a public health threat, reorienting the elimination with known data sources (Shretta and oth- program from control to elimination is likely to require ers 2016). Program expenditures were divided by the a significant one-time investment (Sabot and others cost per capita to account for differences in intended 2010). One study that projected costs to a 20- to 50-year coverage and benchmarked to the first year of data for timeline for Hainan and Jiangsu provinces in China and each country. The reported costs ranged from US$0.18 in Mauritius, Swaziland, and Zanzibar found that elim- in Mexico in 1971 (Suarez Torres 1970a) to US$27 in ination is likely to be more costly than control in the Vanuatu (Kahn and others 2009) (all in 2013 U.S. dol- short term and is likely to remain more expensive than lars). Barring a few exceptions, reported costs per capita control at substantially longer timeframes (depending were generally lowest in East Asia and Pacific and Mexico on the inputs of the post-elimination program). (Suarez Torres 1970b) and highest in African countries, Programs can also be integrated, making disease such as Mauritius (Tatarsky and others 2011), São Tomé programs more efficient as well as creating a platform and Príncipe (Kahn and others 2009), Swaziland (Kahn for mobilizing resources, even if malaria is no longer and others 2009; Sabot and others 2010), and Zanzibar considered a priority. For example, in Singapore, inte- (Sabot and others 2010). Only Mauritius seeks to pre- grating dengue and malaria surveillance facilitated vent reintroduction by screening passengers at ports of interagency collaboration and reduced transmission of entry and using targeted vector control, which may both diseases (Luckhart and others 2010). When trans- account for the high costs. mission decreases and eventually ceases, costs are likely Costs for elimination have varied but have generally to decline and eventually stabilize as efforts turn to been low. In the 1960s they were less than US$1 per preventing reintroduction primarily through surveil- person-year. Estimates from Nepal and Thailand ranged lance, vector control, and emergency response. from US$0.64 to US$1.33 per person-year in the 1980s Private out-of-pocket expenditures are also likely to (in 2006 U.S. dollars) (Mills, Lubell, and Hanson 2008). become negligible as the number of cases declines. Two A retrospective study reports elimination expenditures studies (figure 12.2) with empirical data on expendi- (including from nongovernmental funders) in Jordan, tures over multiple programmatic phases found that Lebanon, and Syria of US$0.96, US$0.73, and US$1.69 expenditures declined when moving from elimination per person-year, respectively (de Zulueta and Muir to prevention-of-reintroduction (Abeyasinghe and 1972). These estimates are lower than those from more others 2012; Smith Gueye and others 2014). A study in recent studies, and it is unclear how directly comparable Sri Lanka estimated the financial cost of prevention of they are because of variable inputs and the availability of reintroduction activities to cost US$0.37 in 2014 new and more costly tools as well as the rise of new chal- (Shretta and others 2016), less than a quarter of the lenges, such as insecticide and artemisinin resistance and expenditures in previous years (Abeyasinghe and oth- human migration (figure 12.1). ers 2012). Elimination should therefore not be justified on the basis of short-term cost savings alone. A focus only on Financial Cost Savings of Elimination Relative to relative cost savings ignores many other factors (for Alternative Scenarios example, population growth, economic development, To generate results most relevant to policy, malaria elim- reductions in malaria in neighboring countries) that ination requires a comparison of cost with a counterfac- could permanently alter the epidemiology of the area, tual scenario of malaria control, the costs of which vary reduce transmission, accelerate the elimination timeline, substantially with the level of control. Scenarios may and decrease costs (Smith and others 2013). encompass a range of alternatives, from a null state of disease without intervention to a state of controlled low-endemic malaria (Sabot and others 2010), to scenar- Macroeconomic Gains from Malaria Elimination ios illustrating the costs of doing “business as usual” with Several studies have explored the association between a relatively stable control state punctuated by spikes of malaria and economic productivity (Audibert, epidemics or resurgence when efforts are slowed. Mathonnat, and Henry 2003; Badiane and Ulimwengu 332 Major Infectious Diseases Figure 12.1 Costs of Malaria Elimination, by Country, Various Years 40 35 30 25 Cost (2012 US$) 20 15 10 5 0 2007–27 2007–27 Unspecifieda Iran, Islamic Rep. Unspecified 1964–70 1964–70 1964–70 Unspecifieda Mauritius 1855–2008a 1855–2008a 1971–76 1970 Unspecified 1984–85a Philippines 1998–2010a 2007–27a 2008 2007–27 2007–27 2007–27 Unspecified 1964–70 1952–57 Tanzania Unspecified 2008 2007–27 1991 China Iraq Jordan Lebanon Mexico Nepal São Tomé and Príncipe Solomon Islands Sri Lanka Swaziland Syrian Arab Republic Vanuatu Cost per capita Cost per person at risk Sources: China: Jackson, Sleigh, and Liu 2002; Sabot and others 2010; the Islamic Republic of Iran: Rezaei-Hemami and others 2013; Iraq, Jordan, Lebanon, the Syrian Arab Republic: de Zulueta and Muir 1972; Mauritius: Tatarsky and others 2011; Mexico: Suarez Torres 1970a, 1970b; Nepal: Kondrashin 1992; the Philippines: Liu and others 2013; São Tomé and Príncipe: Kahn and others 2009; the Solomon Islands: Beaver 2011; Sri Lanka: Abeyasinghe and others 2012; Swaziland and Tanzania (Zanzibar): Sabot and others 2010; Vanuatu (Aneityum Island): Kaneko and others 2000. a. Multiple costs per capita were reported in the original article; only the highest cost is presented in the figure. 2013; Girardin and others 2004) and can be used to 2012), and across several countries (Gallup and Sachs build the investment case. Khan (1966) estimated the 2001; McCarthy, Wolf, and Wu 2000; Okorosobo and cost of decreased efficiency attributable to malaria for others 2011) found that malaria is associated with losses Pakistan at more than US$53 million in 1960, while Dua in gross domestic product (GDP) growth. Using and others (1997) estimated more than US$347,000 in cross-country regressions, Gallup and Sachs (2001) production losses in one Indian industrial complex in demonstrated that countries with intensive malaria lost 1985. In the United States in 1914, one day lost to 1.3 percent of GDP growth per person per year between malaria was equal to US$119 in production losses (in 1965 and 1990. Similarly, McCarthy, Wolf, and Wu 2013 U.S. dollars). Many costs of malaria, such as the (2000), using WHO morbidity data, estimated that long-term effects of chronic malaria infection on lower- many high-burden countries lost at least 0.25 percent of ing educational attainment, have yet to be estimated GDP growth per year from malaria. GDP losses of (Chen and others 2016). between 0.41 percent and 8.9 percent or US$4.2 million Economic modeling using data from Ghana (Asante have been reported in Africa (Okorosobo and others and Asenso-Okyere 2003), Uganda (Orem and others 2011). The annual monetary cost of these losses was as Malaria Elimination and Eradication 333 Figure 12.2 Malaria Program Expenditures in Select Countries, by Phase 16 14 Expenditure per person at risk (2012 US$) 12 10 8 6 4 2 0 Anuradhapura Kurunegala 1948–74a 1982–2008a Apayaoa Cavite Laguna Benguet Kunena Ohangwena Omusati Aneityuma Swaziland Zanzibar Hainan Jiangau Henana Sri Lanka Mauritius Philippines Namibia China Control Controlled low endemic Elimination Prevention of reintroduction Sources: Sri Lanka: Abeyasinghe and others 2012; Mauritius: Tatarsky and others 2011; the Philippines: Liu and others 2013; Namibia: Smith Gueye and others 2014; Vanuatu (Aneityum Island): Kaneko and others 2000; Swaziland, Zanzibar, and Hainan and Jiangsu, China: Sabot and others 2010; Henan, China: Jackson, Sleigh, and Liu 2002. a. Multiple costs per capita were reported in the original article; only the highest cost is presented in the figure. high as US$13.1 million in Mali (Okorosobo and others Bleakley 2003, 2010; Hong 2011). It has also been 2011) to US$10 billion in Nigeria (Okorosobo and oth- associated with chronic diseases in later years and an ers 2011). In Thailand, the economic cost of malaria was inability to work (Hong 2013), decreased property valued at US$280 million over five years (Kühner 1971). accumulation in Côte d’Ivoire (Audibert, Mathonnat, Several studies estimated a country’s total economic and Henry 2003), and decreased spending overall loss by examining expenditures for malaria prevention, (Somi and others 2009). control, and treatment, as well as the opportunity cost of The GTS for Malaria (WHO 2015a) and Action and caregiving, debility, and premature death. For example, Investment to Defeat Malaria (RBM Partnership 2015) the loss for India was estimated to be between US$856 use transmission modeling and cost projections to esti- million and US$1.6 billion a year (Sharma 1996). Losses mate the total cost of reducing the global burden of were estimated to be US$415 million for the Philippines malaria to 90 percent of its current level by 2030. The (Barlow and Grobar 1986) and US$133.9 for Pakistan estimated cost would be about US$100 billion, resulting (Khan 1966). However, many of these historical studies in a US$208.6 billion increase in economic output. This are not population based and use secondary sources or figure is in line with a global analysis reporting that expert opinion to calculate the burden of malaria, limit- malaria reduction and elimination between 2013 and ing their contemporary use. 2035 would produce a benefit whose net present value is Exposure to malaria in childhood has been associ- US$208.6 billion and a benefit-cost ratio of 6.11 (Purdy ated with lower incomes and a greater likelihood of and others 2013). Gates and Chambers (2015) estimate poverty in adulthood in South America (Barreca 2010; that eradication could unlock US$2 trillion in economic 334 Major Infectious Diseases benefits at a cost of about US$90 billion to US$120 billion through increased government spending as well as between 2015 and 2040, yielding a return on investment innovative financing mechanisms. Box 12.5 and annex of about 17:1. 12C describe some mechanisms that are being imple- mented or considered and their applicability to malaria programs. Financing and Efficiency Development assistance for malaria quadrupled between Figure 12.3 Overseas Development Assistance Commitments for 2007 and 2013. However, the proportion of development Malaria, 2007–13 assistance directed toward malaria-eliminating countries declined more than 80 percent and continues to decline 2,500 11 (figure 12.3). Securing funding for a disease that Total disbursments (constant 2012 US$, millions) 10 occurs infrequently is challenging. Malaria-eliminating 2,000 9 countries typically have lower disease burdens and are often middle-income countries; therefore, they are a 8 lower priority for donors. The Global Fund to Fight 7 AIDS, Tuberculosis, and Malaria has historically allo- 1,500 Percent 6 cated about 7 percent of its portfolio to malaria- eliminating countries but, under its new funding model, 5 now allocates about 5 percent, representing a projected 1,000 4 decrease of 31 percent in national funding allocation—a 3 serious shortfall at a time when maintaining national gains and advancing the elimination agenda are essential 500 2 (GHG 2014; Zelman and others 2016). 1 Eliminating countries finance about 80 percent of 0 0 their malaria programs (CEPA 2013), and this spending 2007 2008 2009 2010 2011 2012 2013 has been increasing steadily since 2000. However, spend- ing still falls short of the US$8 billion per year needed to Total malaria ODA disbursements reach the 2030 targets (WHO 2015a). Percentage of malaria ODA commitments accounted for by malaria-eliminating countries Greater emphasis is being placed on building the capacity of countries to fund their own programs Source: Unpublished data from Global Health Group. Note: ODA = Overseas Development Assistance. Box 12.5 New Financing Mechanisms to Support Malaria Elimination Earmarked travel and airline taxes. Given the direct funds exist in the health sector, and ministries of link between travel and the risk of malaria transmis- health would benefit from further guidance on the sion and resurgence in Zanzibar (Le Menach and investment, finance, and legal aspects of this others 2011), the local government is considering mechanism. implementing a tax on airplane tickets. A survey finds that visitors are willing to pay a tourist airline Cash on delivery. Cash on delivery, wherein countries tax (Zanzibar Ministry of Health 2012). receive funding once they have achieved a predeter- mined target (Pertakis and Savedoff 2014), has been Endowment funds. Endowments are created as a per- included in two regional Global Fund grants to pro- manent financial asset that generates interest into vide incentives to some Sub-Saharan African coun- perpetuity or for as long as the funds are invested. tries to reduce malaria cases to zero or maintain Endowments are ideal for financing long-term activ- incidence cases below a certain threshold (CEPA ities, such as elimination and prevention of reintro- 2013). This model could also be used to encourage duction, but they require a sizable initial investment countries to achieve elimination or maintain malaria- (Adams and Victurine 2011). Few endowment free status. Malaria Elimination and Eradication 335 Technical and Operational Feasibility Box 12.6 Determining feasibility involves assessing both the tech- nical challenge—the transmission intensity and the Tools for Identifying Efficiency Gains effectiveness of the tools available to reduce it—and the operational capacity to complete the task. Other disease Receptivity risk maps. Maps of the transmission eradication campaigns suggest that eradication has only intensity that would likely occur in the absence been considered after many countries have eliminated of interventions can be generated based on pre- the disease. For example, when the goal of smallpox dictions from statistical relationships between eradication was announced, the disease had been elimi- disease occurrence and environmental or ecolog- nated in all high-income countries and was endemic ical risk factors, and they can help direct inter- in only 59 low-income countries (Barrett 2007; ventions to the places where they will have the Henderson 1987). Similarly, the poliomyelitis eradica- greatest impact—and can help withdraw inter- tion initiative was launched in 1998 only after polio had ventions in places where they are not needed. been eliminated in the Americas and all high-income countries, with indigenous transmission remaining in Elimination scenario planning tool. To guide pol- 125 countries (Aylward and others 2003; Bart, Foulds, icy and planning, elimination scenario planning and Patriarca 1996; Khan and Ehreth 2003). Malaria has applies a comprehensive framework to assessing been eliminated within many local borders, but the over- the technical, operational, and financial feasibil- all burden remains high and widespread. As burdens of ity of moving toward elimination (WHO 2014a). P. falciparum and P. vivax decrease, new strategies to Self-assessment tool. “Malaria Program Efficiency diagnose, treat, and interrupt transmission of lesser- Analysis Tool” (MPEAT) can help identify pro- studied malaria species, including P. malariae, ovale, and grammatic inefficiencies in malaria elimination knowlesi, will be needed. The true burdens of these programs and can help guide policy responses and species are largely unknown because identification by strategies to achieve better value for money (GHG microscopy or rapid diagnostic tests is not reliable 2017). (Baltzell and others 2013; Oguike and others 2011; Steenkeste and others 2010).4 Eradication of any species only succeeds if the last carrier of disease is isolated, treated, and prevented from causing further transmission. Understanding of trans- Efficiency in the portfolio and delivery of interven- mission between animal and human hosts relevant for tions will ultimately increase cost-effectiveness. More zoonotic reservoirs has only recently gained attention. efficient deployment of resources, however, requires For example, P. knowlesi, carried by the macaque mon- a robust surveillance platform in which high-quality key, is increasingly being reported in South-East Asia data can be collected and analyzed so that measures (Baird 2009; Rajahram and others 2012). of response can be adjusted in a timely manner Lessons from other campaigns suggest that for eradi- (box 12.6). cation to be feasible, a vaccine or an equivalent means is needed to convey long-term protection, as in the case of smallpox (Barrett 2007, 2013).5 However, no such mea- PROSPECTS FOR MALARIA ERADICATION sure exists for malaria. Even if other measures could be implemented to confer protection similar to a vaccine, The benefits of achieving and maintaining elimination many challenges remain. Drug resistance is on the rise, include a strong public good component—an incremen- and pyrethroid resistance has emerged after large-scale tal contribution to global malaria eradication. While distribution of LLINs (John, Ephraim, and Andrew many argue that eradication is unlikely given existing 2008; Trape and others 2011; Tulloch and others 2013). tools (Greenwood 2008; Tanner and de Savigny 2008), particularly for high-burden countries in Africa, the global pipeline for new products has never been stron- Operational Complexity ger, supporting the mounting optimism that global The smallpox and polio eradication campaigns imple- eradication is plausible. The technical and operational mented eradication-specific management systems that feasibility of eradication, the operational complexity, could be integrated into existing health systems (Aylward and the political appetite need to be considered when and others 2003), used performance indicators to assessing the prospects for eradication. measure management processes, trained adequate 336 Major Infectious Diseases numbers of staff and gave them incentives to execute The success of malaria eradication will depend on the eradication-specific tasks, developed a robust surveil- ability to mobilize collective action. At a minimum, uni- lance system, and expanded financing to support a versal political commitment to achieving an agreed-on stronger health care system (Henderson 1987). Through target is required, as are financial resources to sustain implementation of the smallpox, polio, and guinea-worm that commitment. Although countries may be willing to programs, innovative breakthroughs were made in orga- eliminate the disease within their borders, the last coun- nizing large-scale nationwide campaigns; in devising try to eliminate it has little incentive to do so on its new methods for approaching and mobilizing commu- own, given the larger interests of all other countries nities; in developing effective national surveillance net- (Barrett 2004). The smallpox eradication program nearly works and using the data to support better strategies; in failed because of lack of political commitment (Barrett fostering effective and relevant research programs to 2007), and the GMEP was cut short for the same reason. facilitate disease control; and in mobilizing support at Although global attitudes have shifted toward malaria international, national, and local levels. Lessons learned elimination and eradication, political and financial sup- from these efforts are critical for malaria eradication. port is needed to bolster the goal of global eradication, Building programs capable of proactively mitigating the should that goal be adopted for malaria. risk of transmission requires careful planning rather There are concerns that concentrating resources in than reactive emergency response measures. areas with lower burdens of disease may divert resources from lower-income countries with higher burdens of disease (Shah 2010); however, progress in low-burden Political and Financial Commitment countries is likely to drive global progress toward eradi- In 1939, Boyd summarized the prevailing public health cation (Newby and others 2016). In addition, because point of view and emphasized that ‘‘malaria control malaria-free countries stand to benefit from eradica- should not be a campaign—it should be a policy, a long- tion, they have an incentive to offer financial assistance term program. It cannot be accomplished or maintained if they are assured that the last countries will work by spasmodic effort. It requires the adoption of a practi- toward elimination (Barrett 2007; Taylor, Cutts, and cable program, the reasonable continuity of which will Taylor 1997). be sustained for a long term of years’’ (Boyd 1939, 5). See box 12.7 for future research priorities. Box 12.7 Priorities for Research While achievements made in the past 15 years give • Improved vector control strategies that target reason for optimism, some gaps and priorities for residual transmission research remain (Tanner and others 2015): • Continued research and development for a combination vaccine • Detection of emergence and spread of drug- • Appropriate models for private sector and resistant parasites using geospatial and temporal community-based surveillance and response mapping of drug resistance • Capacity building in program and health system • Epidemiologically and economically effective management and efficient mixes of interventions in different • Estimates of costs to identify and secure labora- contexts tory specimens of malaria parasites and to stock- • Serological tests to detect individual-level recent pile diagnostic, treatment, and vaccine production infections capabilities in the future • Sensitive clinical field diagnostic tests • Advocacy for engagement in the eradication • New tools for eliminating P. vivax, including the agenda identification of asymptomatic hypnozoite carriers • Sustained investments in malaria elimination • Effective approaches for mass drug administra- and eradication, including innovative financing tion in different contexts mechanisms. Malaria Elimination and Eradication 337 CONCLUSIONS • Annex 12B. Regional Initiatives to Eliminate Malaria • Annex 12C. Potential Financing Mechanisms for Despite the absence of a highly efficacious vaccine, many Malaria Elimination countries around the globe have successfully eliminated malaria and prevented its reintroduction. As malaria elimination progresses in more areas, the case for global NOTES eradication is likely to become more compelling. Promising new tools are already in the product develop- World Bank Income Classifications as of July 2014 are as ment pipeline, including radical treatments, sensitive follows, based on estimates of gross national income (GNI) rapid diagnostic tests, and next-generation vector control per capita for 2013: methods. Piloting the effective use of these innovations • Low-income countries (LICs) = US$1,045 or less will ensure that they can be scaled up safely and effec- • Middle-income countries (MICs) are subdivided: tively. The introduction of game-changing innovations— (a) lower-middle-income = US$1,046 to US$4,125 including anti-infection or transmission-blocking (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 vaccines and novel mosquito control strategies—could • High-income countries (HICs) = US$12,746 or more. substantially accelerate this next phase. As new technolo- gies and advances occur, the cost of elimination may 1. Sri Lanka obtained WHO certification as a malaria-free decline as efficiencies are realized and targeting becomes country in September 2016. increasingly focused. Elimination may become progres- 2. Algeria, Belize, Bhutan, Botswana, Cabo Verde, China, sively easier with new drug therapies, simplified treatment Comoros, Costa Rica, Ecuador, El Salvador, the Islamic regimens, and more effective vaccines. With smallpox, the Republic of Iran, Republic of Korea, Malaysia, Mexico, targeted nature of surveillance and containment and Nepal, Paraguay, Saudi Arabia, South Africa, Suriname, improved needle technology for vaccinations contributed Swaziland, Timor-Leste. 3. Despite a highly receptive environment in Taiwan, China, significantly to the success of the eradicaton campaign. intensive spraying combined with improved housing and Malaria eradication calls for a long-term investment socioeconomic conditions, better environmental manage- that will yield dividends over time. If successful, countries ment, and strong case management reduced morbidity to would no longer need to implement prevention measures, very low levels, and the WHO certified Taiwan, China, as thereby reaping an “eradication dividend” and accruing being malaria free in 1965 (Yip 2000). substantial economic benefits for all countries. However, 4. Polymerase chain reaction testing in African and Asian eliminating malaria transmission worldwide will require settings shows a higher proportion of both P. malariae and renewed focus in several areas. Strengthening the human P. ovale infections than was previously thought (Baltzell resource capacity of programs is essential. Combating the and others 2013; Barrett 2007; Oguike and others 2011). threat of importation will require collaborative regional 5. In the case of smallpox, there were no long-term carriers, surveillance efforts that reach communities and the pri- survivors gained lifetime immunity, infections were easily detected, only symptomatic persons could transmit the vate sector. In addition, as new tools become available, disease, and vaccination of only 80 percent of the popula- support will be required for their adoption and rapid tion was necessary to eliminate transmission (Barrett 2007). uptake to combat the effects of drug and insecticide resis- tance. These actions all require sustained political and financial commitment to ensure success. While increasing REFERENCES numbers of countries are moving toward financing their own programs, external assistance to the last affected Abeyasinghe, R. R., G. N. Galappaththy, C. Smith Gueye, J. G. Kahn, and R. G. Feachem. 2012. “Malaria Control countries will be essential—possibly through a dedicated and Elimination in Sri Lanka: Documenting Progress and “last-mile fund”—to ensure that the resources required to Success Factors in a Conflict Setting.” PLoS One 7 (8): complete eradication are available in the final phase. e43162. Adams, J. S., and R. Victurine. 2011. “Permanent Conservation Trusts: A Study of the Long-Term Benefits of Conservation ANNEXES Endowments.” Wildlife Conservation Society, Bronx, NY. http://www.dcnanature.org/wp-content/uploads The following annexes to this chapter are as follows. They /fundraising/Permanent-Conservation-Endowments.pdf. are available at http://www.dcp-3.org/infectiousdiseases. Adjuik, M., T. Smith, S. Clark, J. Todd, A. Garrib, and others. 2006. “Cause-Specific Mortality Rates in Sub-Saharan Africa • Annex 12A. Status and Goals of Elimination and Bangladesh.” Bulletin of the World Health Organization Countries, by Region 84 (3): 181–88. 338 Major Infectious Diseases Aksan, A.-M., and S. Chakraborty. 2013. “Childhood Disease ———. 2007. “The Smallpox Eradication Game.” Public Choice and the Precautionary Demand for Children.” Journal of 130 (1): 179–207. Population Economics 26 (3): 855–85. ———. 2013. “Economic Considerations for the Eradication Amaratunga, C., P. Lim, S. Suon, S. Sreng, S. Mao, and others. Endgame.” Philosophical Transactions of the Royal Society B: 2016. “Dihydroartemisinin-Piperaquine Resistance in Biological Sciences 368 (1623): 20120149. Plasmodium falciparum Malaria in Cambodia: A Multisite Bart, K. J., J. Foulds, and P. Patriarca. 1996. “Global Eradication Prospective Cohort Study.” The Lancet Infectious Diseases of Poliomyelitis: Benefit-Cost Analysis.” Bulletin of the 16 (3): 357–65. World Health Organization 74 (1): 35–45. APLMA (Asia Pacific Leaders Malaria Alliance). 2015. Bawah, A. A., and F. N. Binka. 2007. “How Many Years of “Asia-Pacific at the Forefront of a Global Movement to Life Could Be Saved if Malaria Were Eliminated from a Eliminate Malaria.” APLMA Blog, October 7. http://aplma Hyperendemic Area of Northern Ghana?” American Journal .org /blog/22/asia-pacific-at-the-forefront-of-a -global of Tropical Medicine & Hygiene 77 (Suppl 6): 145–52. -movement-to-eliminate-malaria/. Beaver, C. 2011. “Application of a Remoteness Index: Funding Ariey, F., B. Witkowski, C. Amaratunga, J. Beghain, A.-C. Malaria Programs.” International Journal of Geoinformatics 7 (1). Langlois, and others. 2014. “A Molecular Marker of Bhatt, S., D. J. Weiss, E. Cameron, D. Bisanzio, B. Mappin, Artemisinin-Resistant Plasmodium falciparum Malaria.” and others. 2015. “The Effect of Malaria Control on Nature 505 (7481): 50–55. Plasmodium falciparum in Africa between 2000 and 2015.” Asante, F. A., and K. Asenso-Okyere. 2003. Economic Burden of Nature 526 (7572): 207–11. Malaria in Ghana. Geneva: WHO, African Regional Office. Bhumiratana, A., P. Sorosjinda-Nunthawarasilp, W. Kaewwaen, Ashley, E. A., M. Dhorda, R. M. Fairhurst, C. Amaratunga, P. Maneekan, and S. Pimnon. 2013. “Malaria-Associated P. Lim, and others. 2014. “Spread of Artemisinin Resistance Rubber Plantations in Thailand.” Travel Medicine and in Plasmodium falciparum Malaria.” New England Journal of Infectious Disease 11 (1): 37–50. Medicine 371: 411–23. Blanford, S., N. E. Jenkins, A. F. Read, and M. B. Thomas. 2012. Audibert, M., J. Mathonnat, and M. C. Henry. 2003. “Malaria “Evaluating the Lethal and Pre-Lethal Effects of a Range and Property Accumulation in Rice Production Systems in of Fungi against Adult Anopheles stephensi Mosquitoes.” the Savannah Zone of Côte d’Ivoire.” Tropical Medicine and Malaria Journal 11: 365. International Health 8 (5): 471–83. Bleakley, H. 2003. “Disease and Development: Evidence from Aylward, R. B., A. Acharya, S. England, M. Agocs, and J. Linkins. the American South.” Journal of the European Economic 2003. “Global Health Goals: Lessons from the Worldwide Association 1 (2–3): 376–86. Effort to Eradicate Poliomyelitis.” The Lancet 362 (9387): ———. 2010. “Malaria Eradication in the Americas: A 909–14. Retrospective Analysis of Childhood Exposure.” American Badiane, O., and J. Ulimwengu. 2013. “Malaria Incidence and Economic Journal: Applied Economics 2 (2): 1–45. Agricultural Efficiency in Uganda.” Agricultural Economics Boyd, M. F. 1939. “Malaria: Retrospect and Prospect.” American 44 (1): 15–23. Journal of Tropical Medicine and Hygiene 19: 1–6. Baird, J. K. 2009. “Malaria Zoonoses.” Travel Medicine and Brady, O. 2016. “Vectorial Capacity and Vector Control: Infectious Disease 7 (5): 269–77. Reconsidering Sensitivity to Parameters for Malaria ———. 2015. “Point-of-Care G6PD Diagnostics for Elimination.” Transactions of the Royal Society of Tropical Plasmodium vivax Malaria Is a Clinical and Public Health Medicine and Hygiene 110 (2): 107–17. Urgency.” BMC Medicine 13 (December): 296. Bruce-Chwatt, L. J. 1959. “Malaria Research and Eradication Baltzell, K. A., D. Shakely, M. Hsiang, J. Kemere, A. S. Ali, and in the USSR: A Review of Soviet Achievements in the Field others. 2013. “Short Report: Prevalence of PCR Detectable of Malariology.” Bulletin of the World Health Organization Malaria Infection among Febrile Patients with a Negative 21: 737–72. Plasmodium falciparum Specific Rapid Diagnostic Test Carrara, V., K. Lwin, A. Phyo, E. Ashley, J. Wiladphaingern, and in Zanzibar.” American Journal of Tropical Medicine and others. 2013. “Malaria Burden and Artemisinin Resistance Hygiene 88 (2): 289–91. in the Mobile and Migrant Population on the Thai- Barclay, V., R. Smith, and J. Findeis. 2012. “Surveillance Myanmar Border, 1999–2011: An Observational Study.” Considerations for Malaria Elimination.” Malaria Journal PLoS One 10 (3): e1001398. 11: 304. CEPA (Cambridge Economic and Policy Associates). 2013. Barlow, R., and L. M. Grobar. 1986. “Costs and Benefits of “Financing for Malaria Elimination.” CEPA and Global Controlling Parasitic Diseases.” Technical Note PHN 8517, Health Group, University of California, San Francisco, CA. Population, Health, and Nutrition Department, World Chen, I., S. E. Clarke, R. Gosling, B. Hamainza, G. Killeen, Bank, Washington, DC. and others. 2016. “‘Asymptomatic’ Malaria: A Chronic Barreca, A. I. 2010. “The Long-Term Economic Impact of and Debilitating Infection That Should Be Treated.” PLoS In Utero and Postnatal Exposure to Malaria.” Journal of Medicine 13 (1): e1001942. Human Resources 45 (4): 865–92. Chiyaka, C., A. J. Tatem, J. M. Cohen, P. W. Gething, G. Johnston, Barrett, S. 2004. Eradication versus Control: The Economics of and others. 2013. “The Stability of Malaria Elimination.” Global Infectious Disease Policies. Geneva: WHO. Science 339 (6122): 909–10. Malaria Elimination and Eradication 339 Chuma, J. M., M. Thiede, and C. S. Molyneux. 2006. “Rethinking Feachem, R. G., A. A. Phillips, J. Hwang, C. Cotter, B. Wielgosz, the Economic Costs of Malaria at the Household Level: and others. 2010. “Shrinking the Malaria Map: Progress and Evidence from Applying a New Analytical Framework in Prospects.” The Lancet 376 (9752): 1566–78. Rural Kenya.” Malaria Journal 5: 76. Feachem, R. G., A. A. Phillips, and G. A. T. Targett, eds. Chuquiyauri, R., M. Paredes, P. Penataro, S. Torres, S. Marin, and 2009. Shrinking the Malaria Map: A Prospectus on Malaria others. 2012. “Socio-Demographics and the Development Elimination. San Francisco, CA: Global Health Group, of Malaria Elimination Strategies in the Low-Transmission University of California San Francisco. Setting.” Acta Tropica 121 (3): 292–302. Filler, S. J., J. R. MacArthur, M. Parise, R. Wirtz, M. J. Eliades, Cohen, J. M., S. Dlamini, J. M. Novotny, D. Kandula, S. Kunene, and others. 2006. Locally Acquired Mosquito-Transmitted and others. 2013. “Rapid Case-Based Mapping of Seasonal Malaria: A Guide for Investigations in the United States. Malaria Transmission Risk for Strategic Elimination Atlanta, GA: Centers for Disease Control and Prevention. Planning in Swaziland.” Malaria Journal 12: 61. Floore, T. G. 2006. “Larval Control Practices: Past and Present.” Cohen, J. M., B. Moonen, R. W. Snow, and D. L. Smith. 2010. “How Journal of the American Mosquito Control Association 22 (3): Absolute Is Zero? An Evaluation of Historical and Current 527–33. Definitions of Malaria Elimination.” Malaria Journal 9: 213. Gallup, J. L., and J. D. Sachs. 2001. “The Economic Burden Cohen, J. M., D. L. Smith, C. Cotter, A. Ward, G. Yamey, and of Malaria.” American Journal of Tropical Medicine and others. 2012. “Malaria Resurgence: A Systematic Review Hygiene 64 (Suppl 1–2): 85–96. and Assessment of Its Causes.” Malaria Journal 11: 122. Gates, B., and R. Chambers. 2015. From Aspiration to Action: Corran, P., P. Coleman, E. Riley, and C. Drakeley. 2007. What Will It Take to End Malaria? Seattle, WA: Bill and “Serology: A Robust Indicator of Malaria Transmission Melinda Gates Foundation. Intensity?” Trends in Parasitology 23 (12): 575–82. GHG (Global Health Group). 2013.“Malaria-Eliminating Country de Zulueta, J., and D. A. Muir. 1972. “Malaria Eradication in Briefings.” GHG, University of California, San Francisco, CA. the Near East.” Transactions of the Royal Society of Tropical ———. 2014. “The Impact of the Global Fund’s New Funding Medicine and Hygiene 66 (5): 679–96. Model on the 34 Malaria-Eliminating Countries.” GHG, Dhingra, N., P. Jha, V. P. Sharma, A. A. Cohen, R. M. Jotkar, and University of California, San Francisco, CA. others. 2010. “Adult and Child Malaria Mortality in India: ———. 2016. “Analysing Technical Efficiency in Malaria A Nationally Representative Mortality Survey.” The Lancet Elimination Programs: A Self-Help Toolkit.” Unpublished. 376 (9754): 1768–74. Girardin, O., D. Dao, B. G. Koudou, C. Esse, G. Cisse, and others. Dondorp, A. M., F. Nosten, P. Yi, D. Das, A. P. Phyo, and others. 2004. “Opportunities and Limiting Factors of Intensive 2009. “Artemisinin Resistance in Plasmodium falciparum Vegetable Farming in Malaria Endemic Côte d’Ivoire.” Acta Malaria.” New England Journal of Medicine 361: 455–67. Tropica 89 (2): 109–23. Dua, V., S. Sharma, A. Srivastava, and V. Sharma. 1997. Gosling, J., P. Case, J. Tulloch, D. Chandramohan, C. Smith “Bioenvironmental Control of Industrial Malaria at Bharat Gueye, and others. 2014. “Program Management Issues Heavy Electricals Ltd., Hardwar, India: Results of a Nine- in Implementation of Elimination Strategies.” GHG, Year Study (1987–95).” Journal of the American Mosquito University of California, San Francisco, CA. Control Association 13 (3): 278–85. Gosling, J., P. Case, J. Tulloch, D. Chandramohan, J. Wegbreit, Durnez, L., and M. Coosemans. 2013. “Residual Transmission and others. 2015. “Effective Program Management: A of Malaria: An Old Issue for New Approaches.” In Anopheles Cornerstone of Malaria Elimination.” American Journal of Mosquitoes: New Insights into Malaria Vectors, edited by Tropical Medicine and Hygiene 93 (1): 135–38. S. Manguin, 671–704. Rijeka: InTech. Gosling, R. D., L. Okell, J. Mosha, and D. Chandramohan. 2011. Duru, V., N. Khim, R. Leang, S. Kim, A. Domergue, and oth- “The Role of Antimalarial Treatment in the Elimination of ers. 2015. “Plasmodium falciparum Dihydroartemisinin- Malaria.” Clinical Microbiology and Infection 17 (11): 1617–23. Piperaquine Failures in Cambodia Are Associated with Government of India, Ministry of Health and Family Welfare. Mutant K13 Parasites Presenting High Survival Rates in 2016. National Framework for Malaria Elimination in India Novel Piperaquine In Vitro Assays: Retrospective and (2016–2030). New Delhi: Directorate of National Vector Prospective Investigations.” BMC Medicine 13: 305. Borne Disease Control Programme. http://www.searo.who Eckhoff, P., J. Gerardin, and E. Wenger. 2015. “Mass Campaigns .int/india/publications/national. with Antimalarial Drugs: A Modelling Comparison of Greenwood, B. M. 2008. “Control to Elimination: Implications Artemether-Lumefantrine and DHA-Piperaquine with and for Malaria Research.” Trends in Parasitology 24 (10): 449–54. without Primaquine as Tools for Malaria Control and Guiguemdé, W. A., and R. K. Guy. 2012. “An All-Purpose Elimination.” BMC Infectious Diseases 15: 144. Antimalarial Drug Target.” Cell Host and Microbe 11 (6): El Khyari, T. 2001. Malaria Elimination Strategy in Morocco: 555–57. Plan and Elements of Evaluation. Unpublished report, Gupta, I., and S. Chowdhury. 2014. “Economic Burden of Morocco Ministry of Health. Malaria in India: The Need for Effective Spending.” South- Eziefula, A. C., R. Gosling, J. Hwang, M. S. Hsiang, T. Bousema, and East Asia Journal of Public Health 3 (1): 95–102. others. 2012. “Rationale for Short Course Primaquine in Africa Hammond, A., R. Galizi, K. Kyrou, A. Simoni, C. Siniscalchi, to Interrupt Malaria Transmission.” Malaria Journal 11 (1): 360. and others. 2016. “A CRISPR-Cas9 Gene Drive System 340 Major Infectious Diseases Targeting Female Reproduction in the Malaria Mosquito Hsiang, M. S., B. Greenhouse, and P. J. Rosenthal. 2014. Vector Anopheles gambiae.” Nature Biotechnology 34 (4): “Point of Care Testing for Malaria Using LAMP, Loop 78–85. Mediated Isothermal Amplification.” Journal of Infectious Harris, I., W. W. Sharrock, L. M. Bain, K. A. Gray, A. Bobogare, Diseases 210 (8): 1167–69. and others. 2010. “A Large Proportion of Asymptomatic Hsiang, M. S., J. Hwang, S. Kunene, C. Drakeley, D. Kandula, Plasmodium Infections with Low and Sub-Microscopic and others. 2012. “Surveillance for Malaria Elimination in Parasite Densities in the Low-Transmission Setting of Swaziland: A National Cross-Sectional Study Using Pooled Temotu Province, Solomon Islands: Challenges for Malaria PCR and Serology.” PLoS One 7 (1): 329550. Diagnostics in an Elimination Setting.” Malaria Journal Imwong, M., S. Hanchana, B. Malleret, L. Renia, N. P. Day, 9: 254. and others. 2014. “High-Throughput Ultrasensitive Helinski, M., M. Hassan, W. El-Motasim, C. A. Malcolm, Molecular Techniques for Quantifying Low-Density B. G. Knols, and B. El-Sayed. 2008. “Towards a Sterile Insect Malaria Parasitemias.” Journal of Clinical Microbiology Technique Field Release of Anopheles arabiensis Mosquitoes 52 (9): 3303–9. in Sudan: Irradiation, Transportation, and Field Case Jackson, S., A. C. Sleigh, and X. L. Liu. 2002. “Cost of Malaria Experimentation.” Malaria Journal 7: 65. Control in China: Henan’s Consolidation Programme from Hemingway, J., R. Shretta, T. N. C. Wells, D. Bell, A. A. Community and Government Perspectives.” Bulletin of the Djimdé, and others. 2016. “Tools and Strategies for World Health Organization 80 (8): 653–59. Malaria Control and Elimination: What Do We Need to John, G. K., N. M. Douglas, L. von Seidlein, F. Nosten, Achieve a Grand Convergence in Malaria?” PLoS Biology J. K. Baird, and others. 2012. “Primaquine Radical Cure 12 (3): e1002380. of Plasmodium vivax: A Critical Review of the Literature.” Henderson, D. A. 1987. “Principles and Lessons from the Malaria Journal 11: 280. Smallpox Eradication Programme.” Bulletin of the World John, R., T. Ephraim, and A. Andrew. 2008. “Reduced Health Organization 65 (4): 535–46. Susceptibility to Pyrethroid Insecticide Treated Nets by the Hien, T. T., N. T. Thuy-Nhien, N. H. Phu, M. F. Boni, Malaria Vector Anopheles gambiae s. l. in Western Uganda.” N. V. Thanh, and others. 2012. “In Vivo Susceptibility of Malaria Journal 7: 92. Plasmodium falciparum to Artesunate in Binh Phuoc Kahn, J. G., S. Basu, C. Boyle, M. S. Hsiang, D. T. Jamison, and Province, Vietnam.” Malaria Journal 11: 355. others. 2009. “Financing Elimination.” In Shrinking the Hiwat, H., L. S. Hardjopawiro, W. Takken, and L. Villegas. 2012. Malaria Map: A Prospectus on Malaria Elimination, edited “Novel Strategies Lead to Pre-Elimination of Malaria in by R. G. Feachem, A. A. Phillips, and G. A. Targett, 61–80. Previously High-Risk Areas in Suriname, South America.” San Francisco, CA: GHG, University of California. Malaria Journal 11: 10. Kaneko, A., G. Taleo, M. Kalkoa, S. Yamar, T. Kobayakawa, and Hong, S. C. 2011. “Malaria and Economic Productivity: A. Bjorkman. 2000. “Malaria Eradication on Islands.” The A Longitudinal Analysis of the American Case.” Journal of Lancet 356 (9241): 1560–64. Economic History 71 (3): 654–71. Katz, T. M., J. H. Miller, and A. A. Hebert. 2008. “Insect ———. 2013. “Malaria: An Early Indicator of Later Disease and Repellents: Historical Perspectives and New Developments.” Work Level.” Journal of Health Economics 32 (3): 612–32. Journal of the American Academy of Dermatology 58 (5): Horton, R. 2015. “Vaccines: A Step Change in Malaria 865–71. Prevention?” The Lancet 385 (9978): 1591. Kelly, G. C., M. Tanner, A. Vallely, and A. Clements. 2012. Hotez, P. J. 2009. “Mass Drug Administration and Integrated “Malaria Elimination: Moving Forward with Spatial Control for the World’s High-Prevalence Neglected Tropical Decision Support Systems.” Trends in Parasitology 28 (7): Diseases.” Clinical Pharmacology and Therapeutics 85 (6): 297–304. 659–64. Khan, M. J. 1966. “Estimate of Economic Loss due to Malaria Howard, A. F., R. N’Guessan, C. J. Koenraadt, A. Asidi, in West Pakistan.” Pakistan Journal of Health 16 (3): 187–93. M. Farenhorst, and others. 2011. “First Report of the Khan, M. M., and J. Ehreth. 2003. “Costs and Benefits of Polio Infection of Insecticide-Resistant Malaria Vector Mosquitoes Eradication: A Long-Run Global Perspective.” Vaccine 21 with an Entomopathogenic Fungus under Field Conditions.” (7–8): 702–5. Malaria Journal 10: 24. Killeen, G. F. 2014. “Characterizing, Controlling and Eliminating Howes, R. E., F. B. Piel, A. P. Patil, O. A. Nyangiri, P. W. Residual Transmission.” Malaria Journal 13: 330. Gething, and others. 2012. “G6PD Deficiency Prevalence Kimani, E. W., J. M. Vulule, I. W. Kuria, and F. Mugisha. 2006. and Estimates of Affected Populations in Malaria Endemic “Use of Insecticide-Treated Clothes for Personal Protection Countries: A Geostatistical Model-Based Map.” PLoS against Malaria: A Community Trial.” Malaria Journal 5: 63. Medicine 9 (11): e1001339. Kondrashin, A. V. 1992. “Malaria in the WHO Southeast Asia Hoyer, S., S. Nguon, S. Kim, N. Habib, N. Khim, and others. Region.” Indian Journal of Malariology 29 (3): 129–60. 2012. “Focused Screening and Treatment (FSAT): A PCR- Kühner, A. 1971. “The Impact of Public Health Programs on Based Strategy to Detect Malaria Parasite Carriers and Economic Development: Report of a Study of Malaria in Contain Drug-Resistant P. Falciparum, Pailin, Cambodia.” Thailand.” International Journal of Health Services 1 (3): PLoS One 7 (1): e45797. 285–92. Malaria Elimination and Eradication 341 Leang, R., A. Barrette, D. M. Bouth, D. Ménard, R. Abdur, and Control in Yanomami Communities in the Amazon Region others. 2013. “Efficacy of Dihydroartemisinin-Piperaquine of Venezuela.” Tropical Medicine and International Health for Treatment of Uncomplicated Plasmodium falciparum 13 (3): 392–402. and Plasmodium vivax in Cambodia, 2008 to 2010.” Manguin, S., P. Carnevale, and J. Mouchet. 2008. Biodiversity of Antimicrobial Agents and Chemotherapy 57 (2): 818–26. Malaria in the World. London: John Libbey Eurotext. Leang, R., W. R. Taylor, D. M. Bouth, L. Song, J. Tarning, Marston, L., G. C. Kelly, E. Hale, A. C. Clements, A. Hodge, and others. 2015. “Evidence of Plasmodium falciparum and others. 2014. “Cost Analysis of the Development and Malaria Multidrug Resistance to Artemisinin and Implementation of a Spatial Decision Support System for Piperaquine in Western Cambodia: Dihydroartemisinin- Malaria Elimination in Solomon Islands.” Malaria Journal Piperaquine Open-Label Multicenter Clinical Assessment.” 13: 325. Antimicrobial Agents and Chemotherapy 59 (8): 4719–26. Matowo, N. S., J. Moore, S. Mapua, E. P. Madumla, I. R. Le Menach, A., A. J. Tatem, J. M. Cohen, S. I. Hay, H. Randell, and Moshi, and others. 2013. “Using a New Odour-Baited others. 2011. “Travel Risk, Malaria Importation, and Malaria Device to Explore Options for Luring and Killing Outdoor- Transmission in Zanzibar.” Scientific Reports 1 (93): 1–7. Biting Malaria Vectors: A Report on Design and Field Lindblade, K. A., L. Steinhardt, A. Samuels, S. P. Kachur, Evaluation of the Mosquito Landing Box.” Parasites and and L. Slutsker. 2013. “The Silent Threat: Asymptomatic Vectors 6 (May): 137. Parasitemia and Malaria Transmission.” Expert Review of Maude, R., W. Pontavornpinyo, S. Saralamba, R. Aguas, Anti-Infective Therapy 11 (6): 623–39. S. Yeung, and others. 2009. “The Last Man Standing Is the Lines, J., C. J. M. Whitty, and K. Hanson. 2007. Prospects for Most Resistant: Eliminating Artemisinin-Resistant Malaria Eradication and Elimination of Malaria. London: London in Cambodia.” Malaria Journal 8: 31. School of Hygiene and Tropical Medicine. Maude, R. J., D. Socheat, C. Nguon, P. Saroth, P. Dara, and Liu, J. X., G. Newby, A. Brackery, C. Smith Gueye, C. J. Candari, others. 2012. “Optimising Strategies for Plasmodium fal- and others. 2013. “Determinants of Malaria Program ciparum Malaria Elimination in Cambodia: Primaquine, Expenditures during Elimination: Case Study Evidence from Mass Drug Administration and Artemisinin Resistance.” Select Provinces in the Philippines.” PLoS One 8 (9): e73352. PLoS One 7 (5): e37166. Livadas, G., and D. Athanassatos. 1963. “The Economic Benefits McCarthy, F. D., H. Wolf, and Y. Wu. 2000. “The Growth Costs of Malaria Eradication in Greece.” Rivista de Malariologia of Malaria.” NBER Working Paper, National Bureau of 42 (December): 177–87. Economic Research, Cambridge, MA. Llanos-Cuentas, A., M. V. Lacerda, R. Rueangweerayut, Ménard, D., N. Khim, J. Beghain, A. A. Adegnika, M. Shafiul S. Krudsood, S. K. Gupta, and others. 2014. “Tafenoquine Alam, and others. 2016. “A Worldwide Map of Plasmodium Plus Chloroquine for the Treatment and Relapse Prevention falciparum K13-Propeller Polymorphisms.” New England of Plasmodium vivax Malaria (Detective): A Multicentre, Journal of Medicine 374: 2453–64. Double-Blind, Randomised, Phase 2b Dose-Selection Mendis, K., A. Rietveld, M. Warsame, A. Bosman, G. Greenwood, Study.” The Lancet 383 (9922): 1049–58. and others. 2009. “From Malaria Control to Eradication: Lon, C., J. E. Manning, P. Vanachayangkul, M. So, D. Sea, The WHO Perspective.” Tropical Medicine and International and others. 2014. “Efficacy of Two Versus Three- Health 14 (7): 802–9. Day Regimens of Dihydroartemisinin-Piperaquine Mills, A., Y. Lubell, and K. Hanson. 2008. “Malaria Eradication: for Uncomplicated Malaria in Military Personnel in The Economic, Financial, and Institutional Challenge.” Northern Cambodia: An Open-Label Randomized Trial.” Malaria Journal 7 (Suppl 1): S11. PLoS One 9 (3): e9313 Modrek, S., J. Liu, R. Gosling, and R. G. Feachem. 2012. Lubell, Y., A. Dondorp, P. Guerin, T. Drake, S. Meek, and others. “The Economic Benefits of Malaria Elimination: Do They 2014. “Artemisinin Resistance: Modelling the Potential Include Increases in Tourism?” Malaria Journal 11: 244. Human and Economic Costs.” Malaria Journal 13: 452. Moonen, B., S. Barrett, J. Tullock, and D. T. Jamison. Lucas, A. M. 2010. “The Impact of Malaria Eradication on 2009. “Making the Decision.” In Shrinking the Malaria Fertility and Education.” Department of Economics, Map: A Prospectus on Malaria Elimination, edited by Wellesley College, Wellesley, MA. R. G. Feachem, A. A. Phillips, and G. Targett. San Francisco, Luckhart, S., S. W. Lindsay, A. A. James, and T. W. Scott. CA: GHG, University of California. 2010. “Reframing Critical Needs in Vector Biology and Moonen, B., J. M. Cohen, R. W. Snow, L. Slutsker, C. Drakeley, and Management of Vector-Borne Disease.” PLoS Neglected others. 2010. “Operational Strategies to Achieve and Maintain Tropical Diseases 4 (2): e566. Malaria Elimination.” The Lancet 376 (9752): 1592–603. Maartens, F., B. Sharp, B. Curtis, J. Mthembu, and I. Hatting. Moonen, B., J. M. Cohen, A. J. Tatem, J. Cohen, S. I. Hay, and 2007. “The Impact of Malaria Control on Perceptions others. 2010. “A Framework for Assessing the Feasibility of of Tourists and Tourism Operators Concerning Malaria Malaria Elimination.” Malaria Journal 9: 322. Prevalence in Kwazulu-Natal, 1999/2000 versus 2002/2003.” Mosha, J. F., H. J. Sturrock, B. Greenhouse, B. Greenwood, Journal of Travel Medicine 14 (2): 96–104. C. J. Sutherland, and others. 2013. “Epidemiology of Magris, M., Y. Rubio-Palis, N. Alexander, B. Ruiz, N. Galván, Subpatent Plasmodium falciparum Infection: Implications and others. 2007. “Community-Randomized Trial of for Detection of Hotspots with Imperfect Diagnostics.” Lambdacyhalothrin-Treated Hammock Nets for Malaria Malaria Journal 12: 2221. 342 Major Infectious Diseases Müller, G. C., J. C. Beier, S. F. Traore, M. B. Toure, M. M. Traore, Six Countries of Africa.” European Journal of Business and and others. 2010. “Successful Field Trial of Attractive Toxic Management 3 (6): 42–62. Sugar Bait (ATSB) Plant-Spraying Methods against Malaria Orem, J. N., J. M. Kirigia, R. Azairwe, I. Kasirye, and Vectors in the Anopheles gambiae Complex in Mali, West O. Walker. 2012. “Impact of Malaria Morbidity on Gross Africa.” Malaria Journal 9: 210. Domestic Product in Uganda.” International Archives of Nabarro, D. N., and E. M. Taylor. 1998. “The Roll Back Malaria Medicine 5: 12. Campaign.” Science 280 (5372): 2062–68. Ortiz, J. R. 1968. “Estimación del costo de un programa de Najera, J. A. 1999. “Malaria Control Achievements, Problems, erradicación del paludismo.” Boletín de la Oficina Sanitaria and Strategies.” WHO and Roll Back Malaria Partnership, Panamericana 64 (2): 110–15. Geneva. Oxborough, R. M., J. Kitau, F. W. Mosha, and M. W. Rowland. Najera, J. A., M. Gonzalez-Silva, and P. L. Alonso. 2011. “Some 2015. “Modified Veranda-Trap Hut for Improved Evaluation Lessons for the Future from the Global Malaria Eradication of Vector Control Interventions.” Medical and Veterinary Programme (1955–1969).” PLoS Medicine 8 (1): e1000412. Entomology 29 (4): 371–79. doi:10.1371/journal.pmed.1000412. Pertakis, R., and W. D. Savedoff. 2014. An Introduction to Cash Newby, G., A. Bennett, E. Larson, C. Cotter, R. Shretta, and on Delivery: Aid for Funders. Washington, DC: Center for others. 2016. “The Path to Eradication: A Progress Report Global Development. on the Malaria-Eliminating Countries.” The Lancet 387 Phyo, A. P., S. Nkhoma, K. Stepniewska, E. A. Ashley, S. Nair, (10029): 177–84. and others. 2012. “Emergence of Artemisinin-Resistant Newby, G., J. Hwang, K. Koita, I. Chen, B. Greenwood, and Malaria on the Western Border of Thailand: A Longitudinal others. 2015. “Review of Mass Drug Administration for Study.” The Lancet 379 (9830): 1960–66. Malaria and Its Operational Challenges.” American Journal Poché, R. M., D. Burruss, L. Polyakova, D. M. Poché, and of Tropical Medicine and Hygiene 93 (1): 125–34. R. B. Garlapati. 2015. “Treatment of Livestock with Ngomane, L., and C. de Jager. 2012. “Changes in Malaria Systemic Insecticides for Control of Anopheles arabiensis Morbidity and Mortality in Mpumalanga Province, South in Western Kenya.” Malaria Journal 14: 351. doi:10.1186 Africa (2001–2009): A Retrospective Study.” Malaria /s12936-015-0883-0. Journal 11: 19. Poirot, E., J. Skarbinski, D. Sinclair, S. P. Kachur, L. Slutsker, Ngufor, C., M. Chouaïbou, E. Tchicaya, B. Loukou, N. Kesse, and others. 2013. “Mass Drug Administration for Malaria.” and others. 2014. “Combining Organophosphate-Treated Cochrane Database of Systematic Reviews 12: CD008846. Wall Linings and Long-Lasting Insecticidal Nets Fails to Pooda, H. S., J. B. Rayaisse, D. F. de Sale Hien, T. Lefèvre, Provide Additional Control over Long-Lasting Insecticidal S. R. Yerbanga, and others. 2015. “Administration of Nets Alone against Multiple Insecticide-Resistant Anopheles Ivermectin to Peridomestic Cattle: A Promising Approach gambiae in Côte d’Ivoire: An Experimental Hut Trial.” to Target the Residual Transmission of Human Malaria.” Malaria Journal 13: 396. Malaria Journal 14: 496. Niazi, A. D. 1969. “Approximate Estimates of the Economic Prakash, A., D. R. Bhattacharyya, P. K. Mohapatra, U. Barua, Loss Caused by Malaria with Some Estimates of the Benefits A. Phukan, and others. 2003. “Malaria Control in a Forest of M.E.P.” Bulletin of Endemic Diseases 11 (1): 28–39. Camp in an Oil Exploration Area of Upper Assam.” National Nikolaeva, D., S. J. Draper, and S. Biswas. 2015. “Toward the Medical Journal of India 16 (3): 135–38. Development of Effective Transmission-Blocking Vaccines Purdy, M., M. Robinson, K. Wei, and D. Rublin. 2013. “The for Malaria.” Expert Review of Vaccines 14 (5): 653–80. Economic Case for Combating Malaria.” American Journal Ogoma, S. B., S. H. Moore, and M. Maia. 2012. “A Systematic of Tropical Medicine and Hygiene 89 (5): 819–23. Review of Mosquito Coils and Passive Emanators: Qualls, W., G. Müller, S. Traore, M. M. Traore, K. L. Arheart, Defining Recommendations for Spatial Repellency Testing and others. 2015. “Indoor Use of Attractive Toxic Sugar Bait Methodologies.” Parasites and Vectors 5 (1): 287. (ATSB) to Effectively Control Malaria Vectors in Mali, West Oguike, M. C., M. Betson, M. Burke, D. Nolder, J. R. Stothard, Africa.” Malaria Journal 14: 301. and others. 2011. “Plasmodium ovale curtisi and Rajahram, G. S., B. E. Barber, T. William, J. Menon, N. M. Anstey, Plasmodium ovale wallikeri Circulate Simultaneously in and others. 2012. “Deaths Due to Plasmodium knowlesi African Communities.” International Journal of Parasitology Malaria in Sabah, Malaysia: Association with Reporting as 41 (6): 677–83. Plasmodium malariae and Delayed Parenteral Artesunate.” Ohrt, C., K. W. Roberts, H. J. Sturrock, J. Wegbreit, B. Y. Lee, and Malaria Journal 11: 284. others. 2015. “Surveillance Systems for Malaria Elimination.” Ramaiah, T. 1980. Cost-Benefit Analysis of Malaria Control GHG, University of California, San Francisco, CA. and Eradication Programme in India. Ahmedabad: Public Okell, L. C., T. Bousema, J. T. Griffin, A. L. Ouedraogo, Systems Group, Indian Institute of Management. A. C. Ghani, and others. 2012. “Factors Determining the Ranson, H., R. N’Guessan, J. Lines, N. Moiroux, Z. Nkuni, and Occurrence of Submicroscopic Malaria Infections and others. 2011. “Pyrethroid Resistance in African Anopheline Their Relevance for Control.” Nature Communications Mosquitoes: What Are the Implications for Malaria 3: 1237. Control?” Trends in Parasitology 27 (2): 91–98. Okorosobo, T., F. Okorosobo, G. Mwabu, J. N. Orem, and Rezaei-Hemami, M., A. Akbari-Sari, A. Raiesi, H. Vatandoost, J. M. Kirigia. 2011. “Economic Burden of Malaria in and R. Majdzadeh. 2013. “Cost Effectiveness of Malaria Malaria Elimination and Eradication 343 Interventions from Preelimination through Elimination: Smith, D. L., J. M. Cohen, C. Chiyaka, G. Johnston, P. W. Gething, A Study in Iran.” Journal of Arthropod-Borne Diseases 8 (1): 43. and others. 2013. “A Sticky Situation: The Unexpected RBM (Roll Back Malaria) Partnership. 2008. “The Global Stability of Malaria Elimination.” Philosophical Transactions Malaria Action Plan: For a Malaria-Free World.” WHO, of the Royal Society of London, Series B Biological Sciences Geneva. 368 (1623): 20120145. ———. 2015. Action and Investment to Defeat Malaria Smith, D. L., S. I. Hay, A. M. Noor, and R. W. Snow. 2009. 2016–2030: For a Malaria-Free World. Geneva: WHO on “Predicting Changing Malaria Risk after Expanded behalf of the RBM Partnership Secretariat. Insecticide-Treated Net Coverage in Africa.” Trends in Rowland, M., N. Durrani, S. Hewitt, N. Mohammed, M. Bouma, Parasitology 25 (11): 511–16. and others. 1999. “Permethrin-Treated Chaddars and Top- Smith Gueye, C., M. Gerigk, G. Newby, C. Lourenco, P. Uusiku, Sheets: Appropriate Technology for Protection against and others. 2014. “Namibia’s Path toward Malaria Malaria in Afghanistan and Other Complex Emergencies.” Elimination: A Case Study of Malaria Strategies and Costs Transactions of the Royal Society of Tropical Medicine and along the Northern Border.” BMC Public Health 14: 1190. Hygiene 93 (5): 465–72. Smith Gueye, C., K. C. Sanders, G. N. Galappaththy, C. Rundi, Rowland, M., N. Durrani, M. Kenward, N. Mohammed, T. Tobgay, and others. 2013. “Active Case Detection H. Urahman, and others. 2001. “Control of Malaria in for Malaria Elimination: A Survey among Asia Pacific Pakistan by Applying Deltamethrin Insecticide to Cattle: Countries.” Malaria Journal 12: 358. A Community-Randomised Trial.” The Lancet 357 (9271): Somi, M. F., J. R. Butler, F. Vahid, J. D. Njau, and S. Abdulla. 1837–41. 2009. “Household Responses to Health Risks and Shocks: RTSS Clinical Trials Partnership. 2015. “Efficacy and Safety A Study from Rural Tanzania Raises Some Methodological of RTS,S/AS01 Malaria Vaccine with or without a Booster Issues.” Journal of International Development 21 (2): 200–11. Dose in Infants and Children in Africa: Final Results of Spring, M. D., J. T. Lin, J. E. Manning, J. P. Vanachayangkul, a Phase 3, Individually Randomised, Controlled Trial.” S. Somethy, and others. 2015. “Dihydroartemisinin- The Lancet 386 (9988): 31–45. Piperaquine Failure Associated with a Triple Mutant Sabot, O., J. M. Cohen, M. S. Hsiang, J. G. Kahn, S. Basu, and Including Kelch13 C580Y in Cambodia: An Observational others. 2010. “Costs and Financial Feasibility of Malaria Cohort Study.” The Lancet Infectious Diseases 15 (6): 683–91 Elimination.” The Lancet 376 (9752): 1604–15. Sri Lanka Ministry of Health. 2014. “National Malaria Strategic Samaraweera, K. 2015. “Sri Lanka Eligible for WHO Malaria- Plan for Elimination and Prevention of Re-Introduction Free Certification.” The Nation, October 3. 2014–2018.” Anti-Malaria Campaign, Colombo. Saunders, D., and C. Lon. 2016. “Combination Therapies for Steenkeste, N., W. O. Rogers, L. Okell, I. Jeanne, S. Incardona, Malaria Are Failing—What Next?” The Lancet Infectious and others. 2010. “Sub-Microscopic Malaria Cases and Diseases 16 (3): 274–75. Mixed Malaria Infection in a Remote Area of High Malaria Shah, N. K. 2010. “Assessing Strategy and Equity in the Endemicity in Rattanakiri Province, Cambodia: Implication Elimination of Malaria.” PLoS Medicine 7 (8): e1000312. for Malaria Elimination.” Malaria Journal 9: 108. Shamuradova, L., S. Alieva, R. Kurdova-Mintcheva, A. Rietveld, Straimer, J., N. F. Gnadig, B. Witkowski, C. Amaratunga, R. Cibulskis, and others. 2012. “Achieving Malaria V. Duru, and others. 2015. “Drug Resistance: K13-Propeller Elimination and Certification in Turkmenistan.” Malaria Mutations Confer Artemisinin Resistance in Plasmodium Journal 11: O11. falciparum Clinical Isolates.” Science 347 (6220): 428–31. Sharma, V. P. 1996. “Malaria: Cost to India and Future Trends.” Strode, C., S. Donegan, P. Garner, A. A. Enayati, and Southeast Asian Journal of Tropical Medicine and Public J. Hemingway. 2014. “The Impact of Pyrethroid Resistance Health 27 (1): 4–14. on the Efficacy of Insecticide-Treated Bed Nets against Shono, Y., V. Jean-Francois, Y. Saint Jean, and T. Itoh. 1991. African Anopheline Mosquitoes: Systematic Review “Field Evaluation of Ultra-Low Volume Applications with and Meta-Analysis.” PLoS Medicine 11 (3): e1001619. a Mixture of D-Allethrin and D-Phenothrin for Control doi:10.1371/journal.pmed.1001619. of Anopheles albimanus in Haiti.” Journal of the American Sturrock, H. J., M. S. Hsiang, J. M. Cohen, D. L. Smith, Mosquito Control Association 7 (3): 494–95. B. Greenhouse, and others. 2013. “Targeting Asymptomatic Shretta, R., A. L. V. Avancena, and A. Hatefi. 2016. “The Malaria Infections: Active Surveillance in Control and Economics of Malaria Control and Elimination: A Elimination.” PLoS Medicine 10 (6): e1001467. Systematic Review.” Malaria Journal 15: 593. Sturrock, H. J., K. Roberts, J. Wegbreit, C. Ohrt, and R. Gosling. Shretta, R., R. Baral, A. L. Avancena, K. Fox, A. P. Dannoruwa, 2014. “Effective Responses to Malaria Importation.” GHG, and others. 2017. “An Investment Case for Preventing the University of California, San Francisco, CA. Re-Introduction of Malaria in Sri Lanka.” American Journal Suarez Torres, G. 1970a. “El programa de erradicación del of Tropical Medicine & Hygiene 96 (3): 602–15. paludismo: Plan de seis años.” Salud Pública Mexicana Sissoko, M. S., L. L. Van Den Hoogen, Y. Samake, A. Tapily, 12 (6): 751–73. A. Z. Diarra, and others. 2015. “Spatial Patterns of Plasmodium ———. 1970b. “El programa de erradicación del paludismo: falciparum Clinical Incidence, Asymptomatic Parasite Carriage Resumen del plan con incremento regional de operaciones and Anopheles Density in Two Villages in Mali.” American en parte de la vertiente del Golfo de México y en la Península Journal of Tropical Medicine and Hygiene 3 (4): 790–97. de Yucatán.” Salud Pública Mexicana 12 (6): 745–50. 344 Major Infectious Diseases Surabattula, R., M. P. Vejandla, P. C. Mallepaddi, K. Faulstich, Housing to Reduce Malaria: A Systematic Review and and R. Polavarapu. 2013. “Simple, Rapid, Inexpensive Meta-Analysis.” Malaria Journal 14 (1): 209. Platform for the Diagnosis of Malaria by Loop Mediated United Nations. 2015. “African Leaders Call for Elimination Isothermal Amplification (LAMP).” Experimental of Malaria by 2030.” Office of the UN Secretary-General’s Parasitology 134 (3): 333–40. Special Envoy for Financing the Health Millennium Tanner, M., and D. de Savigny. 2008. “Malaria Eradication Back Development Goals and for Malaria. MDG Health Envoy- on the Table.” Bulletin of the World Health Organization News, February 3. http://www.mdghealthenvoy.org/african 86 (2): 82. -leaders-call-for-elimination-of-malaria-by-2030/. Tanner, M., B. Greenwood, C. Whitty, E. K. Ansah, R. N. Price, Utzinger, J., Y. Tozan, F. Doumani, and B. H. Singer. 2002. and others. 2015. “Malaria Eradication and Elimination: “The Economic Payoffs of Integrated Malaria Control in Views on How to Translate a Vision into Reality.” BMC the Zambian Copperbelt between 1930 and 1950.” Tropical Medicine 13: 167. Medicine and International Health 7 (8): 657–77. Takala-Harrison, S., C. G. Jacob, C. Arze, M. P. Cummings, Wells, T. N., and R. Hooft van Huijsduijnen. 2015. “Ferroquine: J. C. Silva, and others. 2015. “Independent Emergence Welcome to the Next Generation of Antimalarials.” of Artemisinin Resistance Mutations among Plasmodium The Lancet Infectious Diseases 15 (12): 1365–66. falciparum in Southeast Asia.” Journal of Infectious Diseases Wells, T. N., R. Hooft van Huijsduijnen, and W. C. Van Voorhis. 211 (5): 670–79. 2015. “Malaria Medicines: A Glass Half Full?” Nature Tao, Z. Y., H. Y. Zhou, H. Xia, S. Xu, H. W. Zhu, and oth- Reviews Drug Discovery 14 (6): 424–42. ers. 2011. “Adaptation of a Visualized Loop-Mediated Wernsdorfer, W. H., and R. L. Kouznetzov. 1980. “Drug Isothermal Amplification Technique for Field Detection of Resistant Malaria: Occurrence, Control and Surveillance.” Plasmodium vivax Infection.” Parasites and Vectors 4: 115. Bulletin of the World Health Organization 58 (3): 342–352. Tatarsky, A., S. Aboobakar, J. M. Cohen, N. Gopee, A. Bheecarry, Wesolowski, A., N. Eagle, A. J. Tatem, D. L. Smith, A. M. Noor, and others. 2011. “Preventing the Reintroduction of Malaria and others. 2012. “Quantifying the Impact of Human in Mauritius: A Programmatic and Financial Assessment.” Mobility on Malaria.” Science 338 (6104): 267–70. PLoS One 6 (9): e23832. White, N. J. 2013. “Primaquine to Prevent Transmission of Tatem, A. J., Z. Huang, C. Narib, U. Kumar, D. Kandula, and falciparum Malaria.” The Lancet Infectious Diseases 13 (2): others. 2014. “Integrating Rapid Risk Mapping and Mobile 175–81. Phone Call Record Data for Strategic Malaria Elimination WHO (World Health Organization). 1968. “WHO Expert Planning.” Malaria Journal 13: 52. Committee on Malaria: Fourteenth Report.” WHO Taylor, C. E., F. Cutts, and M. E. Taylor. 1997. “Ethical Dilemmas Technical Report Series, Geneva: WHO. in Current Planning for Polio Eradication.” American ———. 1969. “Re-examination of the Global Strategy of Journal of Public Health 87 (6): 922–25. Malaria Eradication.” Twenty-Second World Health Thang, N. D., A. Erhart, N. Speybroeck, N. X. Xa, N. N. Thanh, Assembly, Part I. WHO official records number 176, annex and others. 2009. “Long-Lasting Insecticidal Hammocks for 13, 106–26. Controlling Forest Malaria: A Community-Based Trial in a ———. 1993. “A Global Malaria Control Strategy.” WHO, Rural Area of Central Vietnam.” PLoS One 4 (10): e7369. Geneva. Tobgay, T., C. E. Torres, and K. Na-Bangchang. 2011. “Malaria ———. 2005. “The Tashkent Declaration: The Move from Prevention and Control in Bhutan: Successes and Malaria Control to Elimination.” WHO Regional Office for Challenges.” Acta Tropica 117 (3): 225–28. Europe, Copenhagen. Tran, T. M., S. Portugal, S. J. Draper, and P. D. Crompton. 2015. ———. 2006. Malaria Vector Control and Personal Protection: “Malaria Vaccines: Moving forward after Encouraging First Report of a WHO Study Group. WHO Technical Report Steps.” Current Tropical Medicine Reports 2 (1): 1–3. Series 936. Geneva: WHO. Trape, J. F., A. Tall, N. Diagne, O. Ndiath, A. B. Ly, and ———. 2007. Malaria Elimination: A Field Manual for Low others. 2011. “Malaria Morbidity and Pyrethroid and Moderate Endemic Countries. Geneva: WHO. Resistance after the Introduction of Insecticide-Treated ———. 2010. “Turkmenistan Certified Malaria-Free.” WHO Bednets and Artemisinin-Based Combination Therapies: Regional Office for Europe, Copenhagen. A Longitudinal Study.” The Lancet Infectious Diseases 11 ———. 2012a. “Drug Resistance Threatens Malaria Control.” (12): 925–32. WHO, Geneva. Tulloch, J., B. David, R. D. Newman, and S. Meek. 2013. ———. 2012b. Handbook for Integrated Vector Management. “Artemisinin-Resistant Malaria in the Asia-Pacific Region.” Geneva: WHO. The Lancet 381 (9881): e16–17. ———. 2012c. “The Role of Larviciding for Malaria Control in Turkmenistan Ministry of Health and Medical Industry, Sub-Saharan Africa.” WHO, Geneva. WHO (World Health Organization), and the University ———. 2012d. World Malaria Report 2012. Geneva: WHO. of California, San Francisco, Global Health Group. 2012. ———. 2013. World Malaria Report 2013. Geneva: WHO. “Eliminating Malaria: Case-Study 1: Achieving Elimination ———. 2014a. From Malaria Control to Malaria Elimination: in Turkmenistan.” Geneva: WHO. A Manual for Elimination Scenario Planning. Geneva: WHO. Tusting, L. S., M. M. Ippolito, B. A. Willey, I. Kleinschmidt, ———. 2014b. “Policy Brief on Malaria Diagnostics in Low- G. Dorsey, and others. 2015. “The Evidence for Improving Transmission Settings.” WHO, Geneva. http://www.who Malaria Elimination and Eradication 345 .int/malaria/publications/atoz/malaria-diagnosis-low ——— and GHG (Global Health Group). 2012. Eliminating -transmission-settings-sep2014.pdf. Malaria: Case-Study 3; Progress towards Elimination in ———. 2015a. Global Technical Strategy for Malaria 2016–2030. Sri Lanka. Geneva: WHO. Geneva: WHO. Wickremasinghe, R., S. D. Fernando, J. Thillekaratne, ———. 2015b.“Q&A on Artemisinin Resistance.” WHO, Geneva. P. M. Wijeyaratne, and A. R. Wickremasinghe. 2014. http://who.int/malaria/media/artemisinin_resistance_qa/en/. “Importance of Active Case Detection in a Malaria ———. 2015c. “Recommendations on the Role of Mass Drug Elimination Programme.” Malaria Journal 13: 186. Administration, Mass Screening and Treatment, and Focal Wilson, A. L., V. Chen-Hussey, J. G. Logan, and S. W. Lindsay. Screening and Treatment for Malaria.” WHO, Geneva. 2014. “Are Topical Insect Repellents Effective against http://www.who.int/malaria/publications/atoz/role-of Malaria in Endemic Populations? A Systematic Review and -mda-for-malaria/en/index.html. Meta-Analysis.” Malaria Journal 13: 446. ———. 2015d. Strategy for Malaria Elimination in the Greater Yip, K. 2000. “Malaria Eradication: The Taiwan Experience.” Mekong Subregion (2015–2030). Geneva: WHO. Parassitologia 42 (1–2): 117–26. ———. 2015e. World Malaria Report 2015. Geneva: WHO. Zanzibar Ministry of Health. 2012. “Willingness to Pay for a ———. 2016a. “Eliminating Malaria.” WHO, Geneva. http:// Tourism Levy on International Arrivals.” Unpublished. www.who.int/malaria/publications/atoz / eliminating Zelman, B., M. Melgar, E. Larson, A. Phillips, and R. Shretta. -malaria/en/. 2016. “Global Fund Financing to the 34 Malaria-Eliminating ———. 2016b. “Malaria Vaccine.” WHO Position Paper, Countries under the New Funding Model 2014–2017: WHO, Geneva. An Analysis of National Allocations and Regional Grants.” ———. 2016c. World Malaria Report 2016. Geneva: WHO. Malaria Journal 15: 118. 346 Major Infectious Diseases Chapter 13 Malaria Control Fabrizio Tediosi, Christian Lengeler, Marcia Castro, Rima Shretta, Carol Levin, Tim Wells, and Marcel Tanner INTRODUCTION take a blood meal. These gametocytes mate and develop in the insect into motile sporozoites. This Plasmodium This chapter reviews the strategies for malaria control stage enters the host bloodstream during the next blood and empirical evidence on the costs and cost-effectiveness meal and migrates to the liver, where sporozoites develop of interventions. It then focuses on a systemic approach into liver schizonts. These rupture and produce mero- to malaria control and elimination, describing the rele- zoites, which invade red blood cells where they repro- vance of social and environmental determinants, as well duce asexually. When large numbers of parasites are as the health system factors that deliver effective coverage produced, patients experience high fever, anemia, and of malaria interventions. Finally, it reviews the tools and other symptoms. When capillaries in the brain or other technologies being developed for malaria control and vital organs (lungs) are clogged by red blood cells with their potential contribution to integrated strategies. The altered deformabilities, complications may occur (cere- chapter uses the terminology endorsed by the World bral and complicated or severe malaria), sometimes Health Organization (WHO) (see WHO 2016a). resulting in death. During the replication phase in red blood cells, gametocytes are produced as well. There are two main malaria species in humans: Natural History Plasmodium falciparum and P. vivax. P. falciparum is The ancient Romans knew that draining swamps could restricted to tropical and subtropical (wet season) prevent disease. Today we know that malaria, a disease regions. The disease was called “quotidian fever” because that has afflicted humans since the earliest records, is the fever spikes with synchronized release of parasites caused by Plasmodium spp. parasites, which, following from infected red blood cells every 24 hours. The Nobel Prize–winning studies in 1897 by Sir Ronald Ross, parasites’ escape to higher latitudes is not prevented by a are now known to be transmitted by mosquitoes (Smith shortage of mosquitoes but by the fact that their devel- and others 2012). opment in the insects is highly dependent on ambient Malaria can be transmitted—with varying degrees of temperatures; in temperate climates, this maturation efficiency—by more than 100 species of Anopheles mos- quickly exceeds the average mosquito life span of two to quitoes, a genus that is abundant worldwide. Gametocytes three weeks. P. falciparum is also the deadliest form of produced in malaria patients represent the Plasmodium malaria, because of its propensity to become severe; stage that infects mosquitoes, when female insects (which 92 percent of all malaria deaths occur in Sub-Saharan need the nutrients in vertebrate blood to produce eggs) Africa. P. vivax has found a way to avoid this climate trap Corresponding author: Fabrizio Tediosi, Department of Epidemiology and Public Health, Swiss Tropical and Public Health Institute, Basel, Switzerland; Fabrizio.tediosi@unibas.ch. 347 by remaining dormant in the liver, as hypnozoites, for In high-burden countries, the most vulnerable popu- months or even years. This form of malaria, called lations at risk for malaria tend to be women and chil- “tertian fever” for its 48-hour periodicity, was spread dren, marginalized populations, and people living in globally until the middle of the 20th century and had poverty. Young children are especially at risk, because traveled with Renaissance Europeans to the Americas. they have not yet built up the partial immune protection In Sub-Saharan Africa, P. vivax is not widespread because that adults acquire from multiple, sustained infections. most of the populations are genetically negative for Pregnant women are at risk because of placental infec- Duffy, a red blood cell receptor that P. vivax requires for tion; 30 million women living in Sub-Saharan Africa are infection; these populations are therefore resistant. at risk, leading to 10,000 maternal deaths (Marchesinig Other genetic variations, mostly affecting red blood cell and Crawley 2004) and 200,000 newborn deaths each function, attest to the enormous effect that malaria has year (WHO 2016c). Malaria and HIV (human immuno- had on human evolution. deficiency virus) co-infections occur in more than Plasmodium parasites are eminently adapted to suc- 3 million cases annually and result in 65,000 additional cessfully achieve their host-switching lifecycle. The flip deaths (Hochman and Kim 2009; WHO 2016c). side of this specialization is that the parasite species that Economically, countries with a high burden of malaria infect humans appear unable to choose other mamma- have growth rates that are 1.3 percent less per person per lian hosts. This lack of a wildlife reservoir is clearly an year than low- and malaria-free countries (Gallup and advantage in malaria eradication campaigns. Sachs 2001; WHO 2016c). Burden of Malaria Global Initiatives P. falciparum and P. vivax are by far the most prevalent Between 1955 and 1970, the WHO led a global initiative of the five species of parasites that infect humans; to eradicate malaria (Nájera, González-Silva, and Alonso P. knowlesi, P. malariae, and P. ovale are less common. 2011). Control interventions were developed to mitigate Both P. falciparum and P. vivax can be found in most the spread of the disease, starting with environmental regions: P. falciparum has the highest rates in Sub- sanitation measures and the use of dichlorodiphenyl- Saharan Africa, where P. vivax is almost absent, whereas trichloroethane (DDT) in the 1950s and 1960s. Many P. vivax is the predominant species in the Asia-Pacific countries, particularly in North America and Europe, region, accounting for 52 percent of infections (Price were successful in substantially reducing malaria trans- and others 2007). P. falciparum traditionally accounts mission and even eliminating it. However, DDT was for the majority of deaths and cases of severe malaria, abandoned because of environmental concerns, and the but the effect of P. vivax on severe morbidity is not to be world’s higher-burden countries lacked the necessary underestimated. tools, approaches, and technical assistance to eliminate Despite substantial progress, more than 1 billion the disease without the use of DDT. The goal of world- people still live in areas where malaria can be transmit- wide eradication was quietly abandoned around 1970, ted (WHO 2014). In 2015, an estimated 212 million although many countries continued to drive down the cases of malaria occurred worldwide (uncertainty inter- disease rates and some extinguished malaria. The WHO val [UI]: 148 million–304 million) (WHO 2016c). Most malaria eradication resolution of 1955 was never recalled of the cases in 2015 were in the WHO African Region and formally remains in force. (90 percent), followed by the WHO South-East Asia Decades later, the Millennium Development Goals Region (7 percent), and the WHO Eastern Mediterranean and Roll Back Malaria (RBM) Partnership’s first Global Region (2 percent). About 4 percent of estimated cases Malaria Action Plan (GMAP) (RBM Partnership 2008) globally are due to P. vivax; outside of the African led to a renewed commitment to the fight against continent, the share of P. vivax infections is 41 percent. malaria and to a substantial increase in resources. The incidence rate of malaria is estimated to have In 2007, while considering the then current state of decreased globally by 41 percent between 2000 and control as well as the potential of new tools and 2015, and by 21 percent between 2010 and 2015 (WHO approaches, the global health community at the Malaria 2016c). The massive burden in Sub-Saharan Africa is Forum of the Bill & Melinda Gates Foundation officially due mainly to P. falciparum. The Democratic Republic declared that the new goal was elimination (Roberts and of Congo and Nigeria are the most populated states Enserink 2007). Consequently, the RBM Partnership with high levels of transmission. Three Asian countries (2008) compiled the GMAP, and national control and (India, Indonesia, and Pakistan) account for more than elimination strategies were established and began being 80 percent of P. vivax cases (WHO 2014). implemented. As a result of this important shift in 348 Major Infectious Diseases paradigms and approaches and the insight gained from • Continued research and development to provide experiences from 2007 to 2015, the two guiding docu- new tools. ments for the control, elimination, and, ultimately, eradication of malaria were developed and approved by Vector Control the WHO member countries at the World Health The development and validation of ITNs was a major Assembly 2015: (1) the WHO’s (2015a) Global Technical breakthrough for vector control. Further developments Strategy for Malaria 2016–2030 (Global Technical led to the long-lasting insecticidal nets (LLINs). Current Strategy) and (2) RBM Partnership’s (2015) comple- vector control relies largely on either ITNs, especially mentary Action and Investment to Defeat Malaria LLINs, or IRS. The evidence from randomized con- 2016–2030 (AIM). Both documents were approved by trolled trials indicates that ITNs reduce cases by an esti- WHO member countries in 2015. mated 50 percent, and they reduce all-cause mortality Effective interventions, such as insecticide-treated rates in children under age five years in Sub-Saharan nets (ITNs) (effective because mosquitoes bite almost Africa by 18 percent (Lengeler 2004). exclusively between dusk and dawn) and indoor residual The WHO recommends that all persons at risk for spraying (IRS), have been massively scaled up since 2000, malaria be protected by ITNs, using roughly one net using improved insecticides. The proportion of the pop- per two people. As a result of rapidly increasing coverage, ulation at risk in Sub-Saharan Africa sleeping under an LLINs have been responsible for nearly 70 percent of the ITN for mosquitoes or being protected by IRS rose from gains made against malaria over the past 15 years, in an estimated 37 percent in 2010 (UI: 25–48 percent) to combination with IRS. This progress averted an esti- 57 percent in 2015 (UI: 44–70 percent) (WHO 2016c). mated 663 million malaria cases in Sub-Saharan Africa The proportion of the population at risk protected by alone (Bhatt and others 2015), emphasizing the central IRS declined from a peak of 5.7 percent globally in 2010 role of vector control in the control and eradication to 3.1 percent in 2015 and from 10.5 percent in 2010 agenda (malERA Consultative Group on Vector Control to 5.7 percent in 2015 in Sub-Saharan Africa (WHO 2011). 2016c). The use of insecticides—such as DDT, pyrethroids, The recent initiatives were made possible mainly by carbamates, and organophosphates in the form of the massive funding increase that began in 2002, partic- IRS—has been widely adopted around the world. ularly by the Global Fund to Fight AIDS, Tuberculosis In Sub-Saharan Africa, only countries in Southern and Malaria; the U.S. President’s Malaria Initiative; the Africa and those supported by the U.S. President’s Bill & Melinda Gates Foundation; and other donors Malaria Initiative are conducting IRS activities on a (WHO 2016b). As a direct consequence, global malaria large scale. Unfortunately, insect resistance to pyre- mortality rates were nearly halved between 2000 and throids has dramatically increased, including among 2015 (Bhatt and others 2015). Globally, 95 countries the three major malaria vectors: Anopheles gambiae ss, report ongoing transmission, and 6 are working to pre- A. arabiensis, and A. funestus (Badolo and others 2012; vent reintroduction (WHO 2016b). Mulamba and others 2014). Resistance to the other main classes of insecticides—carbamates, organochlo- rines, and organophosphates—is on the rise as well MALARIA CONTROL INTERVENTIONS: (Quinones and others 2015). The rapid spread of resis- EFFECTIVENESS, COSTS, AND tance of Anopheles mosquitoes to pyrethroids is raising COST-EFFECTIVENESS the cost of IRS substantially in many endemic areas. Two problems with surveying resistance—beyond Effectiveness and Coverage by Geographical Area weaknesses in the entomological monitoring capabili- In 2008, the first GMAP helped accelerate progress in ties in endemic settings—are the great variability in the malaria control and elimination (RBM Partnership resistance mechanisms and the lack of suitable markers 2008). The strategy included three parts, designed to be and related diagnostic tests. All of these factors seri- executed concurrently (RBM Partnership 2008): ously hamper effective monitoring. These concerns have been addressed in a five-point • Aggressive control in the malaria heartland, mainly Global Plan for Insecticide Resistance Management in Sub-Saharan Africa, to lower morbidity and mortal- malaria vector control proposed by the WHO (Mnzava ity rates and others 2015). Anecdotal evidence suggests control • Progressive elimination from the endemic margins to failure is occurring in some parts of Sub-Saharan Africa. reduce the number of countries that have to invest in This was confirmed recently by a five-year study con- fully developed malaria control programs ducted in five countries by the WHO.1 Malaria Control 349 Diagnostics when it has achieved elimination. Surveillance systems The WHO recommends testing all suspected malaria cases need to be closely interlinked with a public health by rapid diagnostic test (RDT) or microscopy. RDTs have response, namely, the availability of tailored, integrated substantially changed the individual- and community-based response packages that interrupt transmission as soon as strategies for test-and-treat campaigns, and they form the the surveillance system identifies existing, new, or backbone of the WHO-promoted test-treat-track strategy. reemerging pockets of transmission. The strategy has the following elements: The concept of surveillance and response has evolved. Today, surveillance and response more effectively link • Following up and testing every suspected malaria case the activities to detect, report, analyze, and interpret the • Treating every confirmed case public health action through integrated packages tai- • Reporting every case in a timely manner through lored to specific settings with the primary goal of stop- surveillance systems. ping transmission and treating all infected people. Surveillance response systems focus on what minimal The quality of RDTs has continuously improved, essential data are required to detect pockets of transmis- mainly because of a quality assurance program devel- sion or reintroduction. This approach differs from the oped by the Foundation for Innovative New Diagnostics classical monitoring and evaluation based on gathering and the WHO Global Malaria Programme. RDTs for all possible data, which too often leads to information P. falciparum are highly sensitive, but their sensitivity for overflow with no feedback and therefore no rapid effec- P. vivax still needs to be improved. Moreover, health care tive public health action. workers lack the means to diagnose hypnozoite carriers,2 Although surveillance was always a cornerstone of which prevents the elimination of P. vivax. RDT use, the initial GMAP, the Global Technical Strategy (WHO primarily to detect P. falciparum, has been scaled up 2015a) now builds on surveillance–response as one of its substantially in the public sector, especially in Africa; the key determinants for elimination and prevention of testing rate in suspected malaria cases increased from reintroduction. Currently, all efforts are made to opera- 40 percent to 62 percent from 2010 to 2013. However, tionalize surveillance–response approaches fully in testing before prescribing or selling treatments remains a national control and elimination programs. In 2015, challenge in the private sector throughout the world; in malaria surveillance systems detected an estimated Africa, antimalarials are often sold and used without 19 percent of cases that occur globally (UI: 16–21 percent) proper diagnosis (WHO 2014). (WHO 2016c). Treatments Costs and Cost-Effectiveness of Interventions Access to effective treatments—with WHO-recommended artemisinin-based combination therapies (ACTs) for The costs and cost-effectiveness of malaria control inter- P. falciparum and either chloroquine (where still ventions have been extensively evaluated, and a systematic efficacious) or ACTs plus primaquine for P. vivax—is review found that in most settings, malaria interventions crucial to control efforts. Between 2005 and 2013, the are among the best buys in global health based on rele- number of ACT treatment courses procured by the public vant indicators (White and others 2011). Nevertheless, and private sectors increased from 11 million to nearly the literature varies widely in the range of unit costs and 1 billion (WHO 2016b). During that timeframe, countries cost-effectiveness ratios; these variations are related to in Sub-Saharan Africa reported treating 50–100 percent of differences in the interventions evaluated, the type of malaria patients with an ACT. Using combination treat- costs included, and, most important, the methodologies ments in malaria is essential to prevent losing effective adopted. medicines to resistance, as happened repeatedly in Annexes 13A and 13B tabulate the costs and the 20th century. The Latin American and Eastern cost-effectiveness results of studies published between Mediterranean regions reported sufficient distribution of 2010 and 2015, presented in 2012 US$. The cost of medicines to treat all patients in public health facilities. malaria control interventions is relatively low in all countries, but varies widely: Surveillance Finding and detecting cases are important aspects of a • The financial cost per severe malaria case ranges national control program aimed at mitigating the spread from US$30 to US$200 in most countries; exceptions of malaria. Surveillance becomes critical when a country were observed in two studies in Myanmar and South moves from control to elimination and even more so Africa, which reported much higher costs. 350 Major Infectious Diseases • Various studies estimated only the costs of medi- SYSTEMIC APPROACHES TO MALARIA cines for uncomplicated malaria, thereby reporting CONTROL AND ELIMINATION low estimates. For most of the studies that included outpatient services, the costs varied between US$4.50 Malaria control and elimination efforts in any setting and US$30.00; the costs were higher in the few studies need to be understood in the context of prevailing eco- that included services in hospital settings. Most of logical and social systems. These highly interconnected these cost estimates do not include diagnostic tests, systems are the key drivers of control and elimination which are rather high—on average, around US$10.90 efforts. per person—again with wide variations. • The costs of preventive treatments in infants, chil- dren, and pregnant women were low (on average Environmental and Social Determinants of Malaria US$2.20, US$2.90, and US$2.60, respectively), except Environmental, health, and social system factors affect in analyses that estimated the full economic costs, the transmission intensity, seasonality, and geographi- including the noncompliant individuals in all popu- cal distribution of malaria. Social factors—such as lation strata. demographics, culture, behavior, migration patterns, socioeconomic characteristics, and politics—affect the Most of the studies available indicate rather low uptake and effectiveness of control and elimination cost-effectiveness ratios. The cost per disability-adjusted interventions. Access to health care and related behav- life year (DALY) averted for intermittent preventive ioral factors determine the vulnerability of individuals treatment for children ranged from US$13 to US$35, and and communities to infection. These factors have pos- that for preventive treatment in pregnant women was itive and negative effects, which can be modified, estimated to be less than US$2. Slightly higher, but still depending on how they interact. relatively low costs per DALYs averted were reported for In addition, many of these factors adapt to novel local case management (from less than US$2.00 to US$42.00) conditions, bringing about additional challenges to con- and for ITNs (US$4.50 to US$128.00). The costs per trol efforts. Relevant examples include the development DALYs averted by IRS were estimated to be higher at of drug and insecticide resistance, vector (Awolola and US$163 to US$183. others 2007; Chinery 1984; Sattler and others 2005) and Several studies assessed the costs and potential human behavior (Maheu-Giroux and Castro 2013), cost-effectiveness of the RTS,S/AS01 vaccine. This agent, and environmental changes (Castro and Singer 2011; which is well tolerated and partially and temporarily Gething and others 2010; Hahn and others 2014; Keiser effective, is being considered for implementation in and others 2004; Yamana and Eltahir 2013). endemic Africa. These studies showed that, conditional on assumptions of price and coverage, adding RTS,S to Environmental Determinants routine malaria control interventions could be highly Environmental determinants fall into two broad cost-effective (Galactionova and others 2017; Penny and categories: others 2016). The costs of vector control interventions were of the • Natural environment: temperature, humidity, rain- same order of magnitude as treatment costs, with wide fall, soil quality, elevation and slope, land cover, and variations, depending on the setting and the type of hydrography study. The economic costs per person protected with • Human environment: land use, land change, ITNs ranged from US$2.70 to US$9.20 in low-income deforestation, housing conditions, infrastructure countries and up to US$19.00 in upper-middle-income (water, sanitation, and waste collection), urbaniza- countries. These costs approximate those for IRS, tion, development projects (such as roads, railways, whereas the costs for insect larval source management dams, irrigation, mining, resettlement projects, are available only per intervention. and oil pipelines), and disasters abetted by human Additionally, recent estimates of the costs and poten- changes. tial returns on investments to achieve the 2030 Global Technical Strategy goals indicated a global return of up to Strategies that alter the environmental characteristics 40:1. This return is due to averting 3 billion malaria cases associated with malaria transmission were among the and 10 million malaria deaths and to increasing produc- earliest interventions tested, validated, and applied at tivity by US$4 trillion (RBM Partnership 2015; WHO larger scale (Stromquist 1920). Environmental interven- 2015a). tions (killing mosquitoes and destroying their habitats) Malaria Control 351 were crucial for the elimination of malaria in European density, migratory patterns, personal behavior, and countries and the United States, and they significantly knowledge about malaria. Behavior particularly affects reduced the burden of the disease elsewhere (Boyd 1926; the effectiveness of vector control (for example, ITN and Neiva 1940; Pomeroy 1920). Case studies documenting LLIN use). Behavior change communication (BCC) sustained success include the construction of the Panama strategies are often used to promote malaria prevention Canal (Gorgas 1915), copper mining in Zambia (Utzinger and treatment behaviors (RBM Partnership 2012) and and others 2002; Watson 1953), and rubber production can substantially increase the return on investment in in Malaysia (Watson 1921). A specific, but enlightening, malaria control (Koenker and others 2014). Although example is the story of malaria-transmitting Anopheles many malaria-endemic countries have a BCC strategy, a mosquitoes breeding in the small water bodies created in gap in the literature exists with respect to the effectiveness Bromelia plants. Bromeliads are epiphytes (plants that of BCC in promoting behavior change and ultimately grow on trees, mainly in tropical South America) that reducing malaria transmission. The design, implementa- typically provide space for small reservoirs of water in tion, and evaluation of locally adapted BCC strategies, which frogs and insects, including Anopheles species, founded in solid behavior change theory, are especially may breed. Malaria was eliminated from southern Brazil important in promoting effective and sustainable by the removal of bromeliads from urban areas and the changes, but they remain challenging for many national introduction of eucalyptus trees on which bromeliads do malaria control programs. not grow (Deane 1988; Pinotti 1951). Housing improvements, first introduced by the Italian hygienist Angelo Celli at the end of the 19th Health System Factors for Effective Coverage of century, were a crucial intervention in Europe and the Malaria Control Interventions United States; the screening of barracks during recent The scale-up of malaria interventions during the past wars was a successful intervention (Carter and Mendis decade highlights the importance of strong health sys- 2002; Lindsay, Emerson, and Charlwood 2002). The use tems (Stratton and others 2008). Effective treatment of intermittent irrigation strategies for control of provides individual benefits by curing infection and malaria around rice paddies continues to be an impor- preventing progression to severe disease stages. It also tant strategy in China (Baolin 1988; Singer and Castro provides community-level benefits by reducing the 2011). The numerous historical examples of the suc- infectious reservoir and averting the emergence and cessful use of environmental management show the spread of drug resistance (WHO 2012). crucial nature of designing integrated interventions and Ensuring effective coverage of malaria treatment is tailoring them to given socioecological settings (Keiser, particularly problematic and requires simultaneously Singer, and Utzinger 2005; Konradsen and others 2004). addressing both supply-side and demand-side chal- Environmental management often has a low priority lenges in health systems that are often weak. Efficacious in endemic areas (Lindsay, Emerson, and Charlwood therapy is available, but many patients with malaria do not 2002), although the opportunities for its adoption are have access to treatment or delay seeking treatment. excellent. Housing improvements—such as screening Providers do not always comply with treatment guide- doors, windows, and eaves; closing eaves; installing ceil- lines, so patients do not necessarily receive the correct ings; improving roofs; sealing cracks in walls; using regimen or instructions, which may lead to adherence higher-quality building materials; creating new housing problems. Even when the correct regimen is communi- designs; and installing eave tubes (Knudsen and von cated and administered, some patients will not adhere to Seidlein 2014; Lindsay, Emerson, and Charlwood 2002; it. Others may be treated with counterfeit or otherwise Ogoma and others 2009; Tusting and others 2015)—are substandard medication. All of these factors lead to applicable in many endemic areas, especially those expe- treatment failures and potentially to the development riencing rapid economic development. Construction and spread of drug resistance. and maintenance of drainage systems in expanding Recent analyses of the effectiveness of malaria service urban areas that often lack proper infrastructure will delivery have assessed supply-side determinants, includ- improve mosquito control effectiveness, as well as the ing diagnosis, staff training, and availability of antima- control of other vector-borne diseases, such as dengue larial medicines at the health facility level (Berendes and and lymphatic filariasis. others 2011; McPake and others 1999; Mikkelsen-Lopez and others 2013; Obrist and others 2007; Rao, Social Determinants Schellenberg, and Ghani 2013a, 2013b; Zurovac and Key social determinants for local populations include age, others 2008). Other studies assessed patient awareness economic activity, education, cultural beliefs, population and perception of illness, affordability of treatment, 352 Major Infectious Diseases and adherence to the treatment regimen (Littrell and achieve a similar rapid reduction in parasitemia as occurs others 2013; Mumba and others 2003; Webster and with injected artesunate (WHO 2015b). This recommen- others 2014). These studies estimated the proportion of dation is supported, particularly in children, by a large fever or malaria cases treated in the public sector accord- multisite (including Africa and Asia) randomized trial ing to the national guidelines, with estimates generated (Gomes and others 2009). Substantial potential exists to for the region as a whole (Berendes and others 2011; improve access to these treatments for the initial treat- Zurovac and Rowe 2006) and for given countries (Alonso ment of severe malaria. and others 2011; Khatib and others 2013; Littrell and New agents are needed for the treatment and preven- others 2013; Mikkelsen-Lopez and others 2013; Sumba tion of all types of malaria, mainly because of the emer- and others 2008; Webster and others 2014). Consistently, gence and spread of drug resistance. Numerous new these evaluations revealed substantial inefficiencies in chemical series and compounds have been identified malaria-related service delivery. over the past decade (Wells, van Huijsduijnen, and Van A recent comprehensive analysis of available data on Voorhis 2015). Several key characteristics are important effective coverage of malaria case management for for new molecules, described as target candidate profiles 43 countries in Sub-Saharan Africa (Galactionova and (TCPs) (Burrows and others 2017): others 2015) found considerable international variations. Effective national coverage for malaria case management • Molecules that kill the blood-stage parasites (the was found to range from 8 percent to 72 percent in the cause of clinical symptoms): In practice, most of the Sub-Saharan Africa region, for a variety of reasons. new molecules appear to have killing rates as fast as Interestingly, the correlation between effective coverage current or recent drugs and deliver an active plasma and economic development was weak, indicating that concentration from a single dose. resource constraints play only a limited role. Such pat- • New medicines that prevent the relapse of hepatic terns of intercountry variation suggest that many system hypnozoites of P. vivax and other recurrent malar- failures are amenable to change. Priority areas for malaria ias: The current standard is a 14-day course of control and eradication policies include identifying the primaquine, an 8-aminoquinoline (8-AQ). A newer reasons for poor health system performance, intervening 8-AQ, tafenoquine, was shown to be highly active to address them, and implementing the respective strate- in preventing relapse after single dosing in phase II gies in program activities. studies (Llanos-Cuentas and others 2014). However, all drugs in the class carry a risk of hemolysis in G6PD (glucose-6-phosphate dehydrogenase)–deficient NEW TOOLS AND TECHNOLOGIES FOR individuals. G6PD deficiency is frequent (around MALARIA CONTROL 10 percent) in tropical Africa (Carter and others 2011) because it is thought to afford some resistance Therapies against malaria. A roadmap for finding new chemical ACTs have become the mainstay for the case manage- entities without such a risk factor has recently been ment of uncomplicated malaria over the past decade, and published (Campo and others 2015). child-friendly versions of most of these agents have • Agents that block transmission: The other activity of become available (Bassat and others 2015). For cases of 8-AQs, including primaquine, is blocking transmis- severe and life-threatening malaria, where oral treat- sion of P. falciparum by killing gametocytes (TCP3b). ments are not an option, the key data from two pivotal In its most recent treatment guidelines, the WHO trials in Asia (Dondorp and others 2005) and Africa recommends a reduced single dose of 0.25 mg/kg (Dondorp and others 2013) demonstrated the superior- (WHO 2015b), which is assumed to be safer than ity of injected artesunate over injected quinine. However, its previous recommendation of a single 0.75 mg/ the adoption and scale-up of injectable artesunate have kg dose, following a recommendation of the WHO’s been slow. The challenge is to ensure the wide availability Malaria Policy Advisory Committee. New agents of injectable artesunate at affordable prices. The reasons without potential risks to G6PD-deficient individuals for using parenteral artesunate rather than oral ACTs are, are needed. first, that patients are often unconscious and, second, that • Chemoprevention: As the eradication agenda pro- reduction of the parasitemia quickly and profoundly ceeds, it becomes increasingly important to protect saves lives in these emergency situations. Where inject- against initial infection with agents active against able artesunate is not available, particularly where medi- hepatic schizonts. New agents of this type are needed cal facilities are lacking, rectal artesunate suppositories particularly by people entering areas of high trans- are recommended by the World Health Organization to mission from low-transmission areas. Malaria Control 353 The drug development pipeline is relatively rich in new regimens can likely be developed. However, among the molecules and targets for the rapid killing of parasites eradication goals is the availability of a simplified form of (Wells, van Huijsduijnen, and Van Voorhis 2015) therapy, ideally, a single-exposure radical cure (Burrows (figure 13.1). The recent availability of efficient controlled and others 2017). This ambitious goal makes develop- human malaria infection (CHMI) models (McCarthy and ment of effective new agents more difficult. A new medi- others 2011) has allowed the assessment of new molecules cine will be a combination of two or more active at an early stage. In these models, volunteers are infected ingredients; any new molecule that enters a combination with low (asymptomatic) densities of parasites, whose must be powerful enough by itself to kill a reasonable proliferation is monitored by PCR (polymerase chain number of parasites in the patient, preferably all of them, reaction). Following administration of experimental so that efficacy is ensured even when some parasites are drugs, a minimum inhibitory concentration (in blood) resistant to one of the partner ingredients. and a rate of parasite reduction can be calculated, both of The Global Technical Strategy (WHO 2015a) and the which are highly predictive of the effects in patients. AIM (RBM Partnership 2015) foresee a 90 percent New medicines that are fully active against emerging reduction in case mortality by 2030, underlining the resistant strains and that are administered via three-day need for new classes of medicines over the next 20 years Figure 13.1 Current Portfolio of the Global Malaria Medicines Development Effort Global Portfolio of Antimalalrial Medicines Translational Product development Access Human Patient Patient Regulatory Preclinical Postapproval * volunteers exploratory confirmatory review GSK030 DDD498 P218 Artefenomel- Tafenoquine Rectal artesunate Artemether- GlaxoSmithKline Merck KGaA (Biotec Thailand) ferroquine GSK/MMV Cipla/Strides/WHO-TDR lumefantrine Sanofi GSK/US Army Various manufacturers 1 DSM421 PA92 SJ733 KAF156/ Dihydroartemisinin Arterolne/ Artemether- (Drexel/UW/GNF) St Jude/Eisai Lumefantrine ** lumefantrine Takeda (UTSW/ piperaquine paediatric 3 piperaquine Novartis Sigma-Tau/Pierre Fabre Sun Pharma dispersible UW/Monash) Various manufacturers 2 AN13762 MMV253 ACT-451840 Cipargamin Co-trimoxazole Artesunate (Anacor) Zydus Cadila Actelion Novartis ITM Antwerp for injection Guilin 3 UCT943 JPC3210 CDRI 9778 DSM265 Artemisinin Dihydroartemisinin- UCT Jacobus Ipca Takeda (UTSW/UW/ naphthoquine piperaquine Monash) Kunming Pharma Co Sigma-Tau/Pierre Fabre 4 NPC1161B GSK607 N-ter butyl Isoquine Fosmidomycin- Artemether sub- Pyronaridine- GlaxoSmithKline LSTM/Liverpool/GSK piperaquine lingual spray artesunate Mississippi Jomaa Pharma/GmbH MRC/Suda Shin Poong 5 MK4815 Methylene Pyronaridine- Merck blue/AQ artesunate Heidelberg granules Shin Poong 5 SAR97276 Artesunate- Sanofi amodiaquine Various manufacturers 6 Artemisone Artensunate- UHKST mefloquine Cipla/DNDi/ Farmanguinhos AQ13 Sulfadoxine Immtech pyrimethamine+ amodiaquine Guilin 7 Sevuparin Dilaforette MMV048 (UCT) figure continues next page 354 Major Infectious Diseases Figure 13.1 Current Portfolio of the Global Malaria Medicines Development Effort (continued) Footnotes: Global Portfolio Target Product Profiles and Target Candidate Profiles MMV has defined Target Product Profiles and Target Candidate Profiles for medicines to support the eradication campaign. Burrows, J., R. H. van Huijsduijnen, J. J. Mohrle, C. Oeuvray, and T. N. C. Wells. 2013. “Designing the Next Generation of Medicines for Malaria Control.” Malaria Journal 12: 187, which is being updated for publication in 2017. Target Product Profiles Target Candidate Profiles indicated by bars at the bottom of Activities for each individual molecule, indicated by symbols each compound box added to each compound in the translational portfolio Burrows Burrows 3-day cure, artemisinin-based combination therapies and others 2013 and others 2017 Combinations aiming at a new single exposure radical Asexual blood stages (TCP-1,2) TCP-1 cure (TPP–1) Relapse prevention (TCP-3a) TCP-3 Severe malaria and prereferral treatment Intermittent/seasonal malaria chemoprevention Transmission reduction (TCP-3b) TCP-5 Products targeting prevention of relapse for P.vivax Chemoprevention (TCP-4) TCP-4 There are currently no products in the development portfolio meeting the single-exposure chemoprotection Additional Symbols on Global Portfolio (SEC) TPP-2 Footnotes for generic names on Global Portfolio Brought into portfolio after approval; collaborations with DNDi 1. First approval: Novartis (Brand name: Coartem®). Generics by No progress report in the past two years Ajanta, Cipla, Ipca, Strides, Macleods, Mylan. 2. First approval: Novartis (Brand name: Coartem® Dispersible). Generic by Ajanta. * Pending review or approval by WHO prequalification or by 3. Brand name: Artesun®. regulatory bodies that are ICH members or observers 4. Brand name: Eurartesim®. 5. Brand name: Pyramax® tablets and granules. ** Approved in several countries but not approved by WHO pre- 6. First approval fixed-dose combination: Sanofi/DNDi (brand qualification or regulatory bodies that are ICH members or name: ASAQ Winthrop). Generics by Ajanta, Cipla, Guilin, Ipca, observers Strides. 7. Brand name: SPAQ-COTM. Note: This figure is constructed from available information using Internet searches, Thomson Reuters Cortellis database (http://lifesciences.thomsonreuters.com /products/cortellis), and searches of the patent and published literature. It represents the portfolio of data as of December 2016. A new version is produced quarterly and is available at Medicines for Malaria Venture, http://www.mmv.org/research-development/rd-portfolio. DNDi = Drugs for Neglected Diseases initiative; ICH = International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use; MMV = Medicines for Malaria Venture; TCP = target candidate profile; TPP = target product profile; WHO = World Health Organization. if the precious gains against malaria are not to be lost in malaria episodes from a single infection. The ideal med- the future. icine for treatment would therefore have activity against the asexual and sexual blood stages of the parasite as well Transmission-Blocking Medicines as against the hypnozoites where present (P. vivax and The primary drug discovery screens have been conducted P. ovale) (Hemingway and others 2016). against blood-stage forms of the parasite. However, many of the compounds show promising activity against the Medicines for Long-Term Chemoprotection sexual stages (gametocytes) and in membrane-feeding The goal of providing chemoprotection has long been assays (feeding mosquitoes in special devices with blood one of the mainstays of malaria research and develop- from patients) (Bolscher and others 2015; Upton and ment. Historically, chemoprotection has been targeted at others 2015). The CHMI model has been modified to also tourists and nonimmune military personnel in times of allow the production and characterization of gametocytes conflict. More recent developments involve medicines to (Pasay and others 2016) and hence may provide an effec- protect children (Wilson and on behalf of the IPTc tive means of testing transmission-blocking treatments. Taskforce 2011). Studies of intermittent preventive treat- P. vivax infections are rising: P. vivax develops dor- ment of children (IPTc) show a remarkable effect when mant forms or hypnozoites, which result in multiple used throughout the rainy malaria season; the medicines Malaria Control 355 reduced the rate of infection by more than 80 percent in combination with a proprietary adjuvant, AS01 and the rate of all-cause mortality by 57 percent. The (produced by GlaxoSmithKline and the Malaria Vaccine cost of such medicine is relatively low: sulfadoxine- Initiative). The results for the phase III study involving pyrimethamine in combination with amodiaquine costs 15,460 children in 11 centers in 7 countries in Sub- less than US$0.70 for one year’s treatment. In light of Saharan Africa showed a reduction of 18–28 percent these data, several groups have proposed using ACTs, against all malaria episodes without a booster, and a such as DHA (dihydroartemisinin-piperaquine), for reduction of 26–36 percent with a booster at month 20. chemoprotection south of the equator, in regions where The protection was slightly less for severe malaria epi- sulfadoxine-pyrimethamine is ineffective. However, sodes: a 1.1–10.0 percent reduction without booster doing so would mean that the same active ingredients and 17.0–32.0 percent reduction with a booster at would be used for treatment and prevention, which is far month 20 (RTS,S/AS01 Clinical Trials Partnership from ideal. The WHO recommends against this approach 2012). The vaccine received a positive scientific opinion because of the risks of resistance and treatment failure. in July 2015 from the Committee for Medicinal Currently available agents are far from idea. Cycloguanil Products for Human Use of the European Medicines pamoate (Camolar), for example, was developed by Parke Agency. Davis in the 1960s as a long-acting form of cycloguanil. An analysis of the vaccine’s protection and its long- This low-solubility salt was developed for intramuscular term public health effect for 43 African countries reported use in primates, but it required a slow injection through a rate of initial protection of 80 percent against infection a 20-gauge needle over a 90-second period because the in children ages 5–17 months and a rate of initial protec- optimal crystal size was large and the vehicle was oleagi- tion of 65 percent in infants ages 6–12 weeks (Olotu and nous (viscous) (Schmidt and Rossan 1984). These large others 2013). Despite observed and predicted protection particle sizes were needed to achieve a drug release over of the RTS,S vaccine being short lived, when used in com- 200 days. Initial trials in humans provided a duration of bination with other malaria control strategies, such as protection from nonresistant strains of four to six months ITNs, the vaccine has the potential to avert up to 700,000 (Elslager 1969). This formulation is hardly child friendly, deaths over a 10-year period. After considering all of the requiring four 1 mL intramuscular injections through a safety data from the clinical trials, the WHO issued a 21-gauge needle. positive policy recommendation for starting to plan a New medicines for chemoprotection are urgently series of large-scale implementation programs (Penny needed because all drugs currently used in treating and others 2015; WHO Malaria Policy Advisory malaria suffer from resistance. Two compounds in phase Committee and Secretariat 2015). II studies, KAF156 (White and others 2016) (acting The WHO, working with the Malaria Vaccine Funders against the un-annotated CARL locus, a part of the Group, updated the Malaria Vaccine Technology Roadmap genome with unknown function), and DSM265 in 2013 to present the organization’s goals until 2030 (McCarthy and others 2017; Sulyok and others 2017) (an (Malaria Vaccine Funders Group 2013; Tanner and inhibitor of the enzyme dihydroorotate dehydrogenase, Alonso 2010). Specifically, by 2030, vaccines will be which is essential for the DNA synthesis of the parasites), launched that target P. falciparum and P. vivax, that have have also shown good activity against the liver schizont a protective efficacy of at least 75 percent against clinical stages and could be used in chemoprevention. Studies in malaria, that are suitable for administration to appropri- CHMI models with insect challenges are needed to vali- ate at-risk groups in malaria-endemic areas, that reduce date whether the preclinical activity can be replicated in transmission of the parasite, and that thereby substan- human subjects. They are also needed to help decide if tially reduce the incidence of human malaria infection. these medicines would require daily, weekly, or even less These new goals are in line with the Malaria frequent administration. Elimination/Eradication Roadmap (Tanner and Alonso 2010), which introduced the further concept of vaccines that interrupt malaria transmission (VIMT). These Malaria Vaccine: The Pace Quickens VIMT include classical transmission-blocking vaccines The agenda for malaria vaccines was originally pre- that target the sexual and mosquito stages, as well as pre- sented in the 2006 global malaria technology roadmap. erythrocytic and asexual stages that have an effect on By 2015, the landmark roadmap was to have registered transmission. More recently, this terminology has been a first-generation vaccine with a protective efficacy of extended to VIMT through the sexual, sporogonic, or more than 50 percent against severe disease and death mosquito stages of the parasite (Nunes and others 2014). and with a duration of protection greater than one year. The current malaria vaccine pipeline from the WHO The current frontrunner is the subunit vaccine RTS,S summary is shown in figure 13.2. 356 Major Infectious Diseases The present challenges to vaccine development are and switching between dozens of genes that encode dif- as follows: ferent cell surface proteins. Almost all successful vaccines are directed against viruses, very few act against bacteria, • A careful selection of antigens to ensure protection and almost none act against protozoans. against both P. vivax and P. falciparum is needed. Several vaccines are in human volunteer studies or • A vaccine that costs around US$1 per dose would early field trials. The probability of early success in malaria be most attractive. However, if the ideal vaccine vaccines as a whole at this stage is quite high (Pronker and may have multiple antigens to target different par- others 2013). The combination of a wealth of candidates asite stages, its production costs may increase. The and a poor success rate highlights the importance of hav- cost-benefit tradeoffs must be evaluated early to ing standardized processes for comparing candidates. The avoid producing a vaccine that might be very effective Malaria Vaccine Roadmap underscores this point and but is too expensive for widespread adoption. recommends standardized CHMI models. These same • Better and more specific adjuvants (which increase models are being used to benchmark chemoprotective antigenicity) are lacking. medicines, giving the additional advantage that chemo- • Improved understanding of why acquired immunity protective medicines and vaccines can be compared side to Plasmodium is slow to develop, incomplete, and by side. short-lived is needed; this is essential to improve the Most of the current vaccine candidates are blood-stage likelihood of finding a fully protective vaccine. vaccines and specific for one species. Again, community portfolio management is needed to ensure that vaccines The challenge is considerable, because protozoan with the potential to fulfill the aspirations of the roadmap parasites have evolved multiple strategies to evade attacks receive priority. The development of transmission-blocking from the mammalian immune system, such as encoding vaccines has received much attention, because they would Figure 13.2 Current Portfolio of the Global Malaria Vaccines Development Effort Global malaria vaccine pipeline Translational projects Vaccine candidates Phase 1a Phase 2a Phase 1b Phase 2b Phase 3 ChAd63/MVA RTS,S-AS01 ChAd63/MVA ChAd63/MVA ME-TRAP ChAd63/ Pfs 25-EPA RTS,S-AS01 + Matrix MTM MVA ME-TRAP MSP 1 ME-TRAP RTS,S-AS01 ChAd63.AMA1/ PfCelTOS FMP012 AMA1-DiCo Pf SPZ fractional dose MVA.AMA1 PfPEBS FMP2.1/ASO1B P27A ChAd63/MVA PvDBP Pfs 25-VLP Completed, reporting overdue Ad35.CS/ Polyepitope DNA GMZ2 Ad26.CS EP 1300 Phase 1a Phase 2b ChAd63/ EBA 175.R2 MSP3 [181–276] MVA (CS, TRAP, AMA) Phase 1b Phase 2b P. falciparum vaccines: Preerythrocitic Blood stage Transmission blocking P. vivax vaccines: Preerythrocitic Blood stage Transmission blocking Source: World Health Organization, http://www.who.int/immunization/research/development/Rainbow_tables/en (March 2016). Malaria Control 357 substantially affect eradication efforts. The regulatory ensure two-year follow up, and submit regulatory docu- pathway for such vaccines and drugs, which would pro- mentation was approximately six years. A review of the tect communities rather than individuals, has been regulatory lessons from this process is critical to future complicated; progress is being made in discussions efforts to shorten some of these timelines. among members of the vaccine community such as developers, clinical epidemiologists involved in trials, and regulatory agencies (Delrieu and others 2015). Vector Control There is still a paucity of antigens under study. The Success of LLINs under Threat from major exception and very promising approach is the Insecticide Resistance sporozoite vaccine from Sanaria, Inc., which is being An essential requirement for effective resistance manage- tested in early volunteer studies in Africa (Seder and ment is to speed up the development of new active insec- others 2013). Studies using more than 50,000 attenu- ticidal ingredients, as well as alternative vector control ated, aseptic, purified, cryopreserved P. falciparum spo- approaches. The Innovative Vector Control Consortium, rozoites delivered in four intravenous injections are a public-private partnership established in 2005, is man- ongoing in endemic areas and for short-term visitors aging a portfolio of novel insecticide candidates that are and travelers (Richie and others 2015). expected to deliver new public health insecticides by Finally, the timescales are important. RTS,S/AS01 2020–22. In addition to new active ingredients, the cur- started phase II (field exploratory) studies over a decade rent development pipeline of vector control products ago; the time to complete the confirmatory studies, (figure 13.3) contains repurposed and reformulated Figure 13.3 Current Portfolio of the Global Malaria Vector Control Development Effort Global portfolio of malaria vector control Product development Research (WHO pesticide evaluation scheme) Access Development Phase II Phase III (large-scale field) Registration (phase I lab trials) (small-scale [hut] trials) Repurposed crop LLIN Interceptor G2 protection LLIN Combination Chlorfenapyr/α- Actellic CS LLIRS VECTA001 Novel development BASF cypermethrin Syngenta/IVCC AI Syngenta/IVCC compounds, including (BASF/IVCC) 2016– (since 2012) fungicides 2017–2019 (2019–20) Resistance- LLIN Olyset Duo: K-Othrine LLIN Combination breaking net Pyriproxyfen/ Polyzone Novel target-based Sumitomo formulations permethrin Bayer/IVCC (since discovery, (Sumitomo/IVCC) Foundation for the 2013) 2017–2018 NIH LLIRS/repurposed active ingredients “Project Venus” Momfluorothrin from crop protection Akzo Nobel Chlorfenapyr IRS Resistance- Sumitomo (since (2020–22); Bayer, (BASF/IVCC; 2013) 2014) breaking net BASF, Sumitomoa formulations Nonpyrethroid New screening hits: insecticide-treated Syngenta, Bayer, durable wall lining Sumitomo/IVCC, (2020–22) Novel AI Bivalent carbamates Sumitomo/IVCC University of Florida, USA Species-specific, biological control of AI: active ingredient; IRS, indoor residual spray; IVCC: Innovative Vector Control; LLIRS, long- mosquitoes (>2025) lasting indoor residual spray; LLIN: long-lasting insecticidal mosquito net, LLN, long-lasting net: LSHTM, London School of Hygiene and Tropical Medicine; NIH, National Institutes of Indoxacarb + α- Health; PAMVERC, Pan-African Malaria Vector Research Consortium; aclothianidin and chlorfenapyr cypermethrin for LLINs; LSHTM, PAMVERC Dates reflect expected deployment Source: IVCC 2016. 358 Major Infectious Diseases existing insecticides, such as the microencapsulated Another potential approach is the use of attractive organophosphate insecticide pirimiphos-methyl. toxic sugar baits, which take advantage of the fact that Finally, the pipeline also includes noninsecticidal new every female mosquito needs to take one or more paradigms for vector control. As a stopgap measure to sugar meals in addition to blood meals to produce rapidly address the pyrethroid resistance issue, nets have offspring. Such baits attract mosquitoes to artificial been designed that combine an insecticide (usually a sugar sources that are toxic to the mosquito (Qualls pyrethroid) with a second chemical (usually piperonyl and others 2015). butoxide), which helps reduce enzymatic resistance in Finally, the careful and evidence-based combination Anopheles mosquitoes. These combination nets are being of multiple vector control tools is likely to provide the field tested and may prove to be effective in areas with best avenue to reduce further transmission in currently high levels of pyrethroid resistance in malaria vectors. endemic areas in which either LLINs or IRS is already As new insecticides with new modes of action become implemented at high coverage (Okumu and Moore available, nets should regain their effectiveness. Ideally, 2011). One such combination, a spatial repellent that future nets should be treated with a combination of protects houses, is associated with attractive traps located insecticides representing various modes of action, at the periphery of villages, using the “push-pull strat- thereby reducing the risk of resistance. This combination egy” (Wagman and others 2015). strategy has worked well for antimalarials and should be chosen for vector control strategies as well. CONCLUSIONS Mosquito Population Modification Strategies A promising potential intervention is the use of genet- Malaria control is one of the great success stories of ically modified mosquitoes. The recent demonstration global public health. Unprecedented success has been that CRISPR (clustered regularly interspaced short and is being achieved for a disease that, by one estimate, palindromic repeats)/Cas9 gene-editing technology may have killed half of all the people who ever lived can be used efficiently to generate genetically modified (Whitfield 2002). Even a high-transmission setting such insects (Gantz and Bier 2015) has raised hopes for as the Democratic Republic of Congo has been able to future development of malaria-resistant mosquitoes. halve the prevalence and incidence rates in the past Another encouraging finding is the recent discovery 15 years (WHO 2016b). These successes have provided that genetic modification of certain bacteria (espe- the impetus for the world to move forward to attain the cially Wolbachia sp.) from the mosquito microbiota goal of eradication. However, substantial threats to these (paratransgenesis) can lead to a dramatic reduction of achievements exist. Drug and insecticide resistance top mosquito vectorial competence (Shaw and others the list of biological and epidemiological problems, 2016). However, many challenges remain with these while the lack of political will and sustainable financing approaches, which to date have been largely confined top the list of external dangers. to the laboratory. The issue of driving new genes sus- In summary, several conclusions emerge: tainably into wild mosquito populations remains a major obstacle to widespread implementation, as is • Many new malaria control tools are being developed, testing of the efficacy and the public acceptance of thanks to product development partnerships for such approaches (WHO/TDR and Foundation for the drugs (Medicines for Malaria Venture), diagnos- National Institutes of Health 2014). tics (Foundation for Innovative New Diagnostics), and vector control tools (Innovative Vector Control Other Noninsecticidal Vector Control Approaches Consortium). New strategies are being devised that New vector control tools being developed do not rely use existing tools, with countries attempting to on insecticidal action for their effectiveness, thereby become more specific and evidence driven in their providing a welcome alternative at a time of wide- strategic plans (WHO 2015a). spread resistance. Spatial repellents have the potential • The regulatory landscape for drugs and vaccines is to significantly decrease the entry of malaria vectors not equipped to evaluate treatments meant to drive into human dwellings (Lambrechts and others 2015; low-to-zero malaria transmission. When placing this Ogoma and others 2014) and are being tested in outlook into the overall context of AIM (RBM large-scale trials around the world. Topical repellents, Partnership 2015) and the Sustainable Development in contrast, have proven largely disappointing as a Goals, one can no longer calculate the payoff of tools malaria control tool, primarily because of compliance introduced to drive elimination from low-to-zero problems (Sluydts and others 2016). transmission only in terms of present and future Malaria Control 359 cases averted; a broader economic perspective of pro- REFERENCES ductivity gained, as well as the transmission effect, is Alonso, P. L., D. Bell, K. Hanson, K. Mendis, R. D. Newman, and needed. This enhanced perspective focused on return others. 2011. “A Research Agenda for Malaria Eradication: on investments is essential to evaluate the relative Health Systems and Operational Research.” PLoS Medicine merits of the various tools and to prioritize and select 8: e1000397. their deployment in multiple settings. Awolola, T. S., A. O. Oduola, J. B. Obansa, N. J. Chukwurar, • Consequently, there is a growing need to decide how and J. P. Unyimadu. 2007. “Anopheles gambiae s.s. Breeding new control approaches are to be applied in an opti- in Polluted Water Bodies in Urban Lagos, Southwestern mal combination and an integrated way in a world Nigeria.” Journal of Vector Borne Diseases 44 (4): 241–44. with differing levels of malaria transmission and Badolo, A., A. Traore, C. M. Jones, A. Sanou, L. Flood, with contrasting malaria endemicity, even within the and others. 2012. “Three Years of Insecticide Resistance same country. Addressing this heterogeneity entails Monitoring in Anopheles gambiae in Burkina Faso: adapting the Global Technical Strategy (WHO 2015a) Resistance on the Rise?” Malaria Journal 11: 232. http:// dx.doi.org/10.1186/1475-2875-11-232. to national and even subnational levels and rigor- Baolin, L. 1988. “Environmental Management for the Control of ously implementing surveillance–response strategies Ricefield-Breeding Mosquitoes in China.” International Rice and systems. Research Institute in collaboration with the World Health Organization/Food and Agriculture Organization of the Given that all activities are based on partnership United Nations/United Nations Environment Programme approaches across the public, private, and charitable Panel of Experts on Environmental Management for Vector sectors, the roles and responsibilities of the partnerships Control, Manila. must be well defined at each level. The task of malaria Bassat, Q., B. Oqutu, A. Djimde, K. Stricker, and K. Hamed. control and elimination calls for more than joint actions. 2015. “Tailoring a Pediatric Formulation of Artemether- Rather, it requires better, well-defined, and assigned Lumefantrine for Treatment of Plasmodium falciparum Malaria.” Antimicrobial Agents and Chemotherapy 59: tasks and responsibilities, accompanied by the required 4366–74. power and authority to implement the responsibilities Berendes, S., P. Heywood, S. Oliver, and P. Garner. 2011. within national strategies and coordinated operational “Quality of Private and Public Ambulatory Health Care in plans. Low and Middle Income Countries: Systematic Review of Comparative Studies.” PLoS Medicine 8 (4): e1000433. Bhatt, S., D. J. Weiss, E. Cameron, D. Bisanzio, B. Mappin, and ANNEXES others. 2015. “The Effect of Malaria Control on Plasmodium falciparum in Africa between 2000 and 2015.” Nature 526 The annexes to this chapter are available at http://www (7572): 207–11. http://dx.doi.org/10.1038/nature15535. .dcp-3.org/infectiousdiseases. Bolscher, J. M., K. M. Koolen, G. J. Van Gemert, M. G. van de Vegte-Bolmer, T. Bousema, and others. 2015. • Annex 13A. Dataset of Costs Studies. “A Combination of New Screening Assays for Prioritization • Annex 13B. Dataset of Cost-Effectiveness Studies. of Transmission-Blocking Antimalarials Reveals Distinct Dynamics of Marketed and Experimental Drugs.” Journal of Antimicrobial Chemotherapy 70 (50): 1357–66. NOTES Boyd, M. F. 1926. “The Influence of Obstacles Unconsciously Erected against Anophelines (Housing and Screening) World Bank Income Classifications as of July 2014 are as upon the Incidence of Malaria.” American Journal of follows, based on estimates of gross national income (GNI) Tropical Medicine and Hygiene 6 (2): 157–60. per capita for 2013: Burrows, J. N., S. Duparc, W. E. Gutteridge, R. H van Huijsduijnen, W. Kaszubska, and others. 2017. “New • Low-income countries (LICs) = US$1,045 or less Developments in Anti-Malarial Target Candidate and • Middle-income countries (MICs) are subdivided: Product Profiles.” Malaria Journal 16: 26. http://www.ncbi (a) lower-middle-income = US$1,046–US$4,125 .nlm.nih.gov/pmc/articles/PMC5237200/. (b) upper-middle-income (UMICs) = US$4,126–US$12,745 Campo, B. F., O. Vandal, D. L. Wesche, and J. N. Burrows. 2015. • High-income countries (HICs) = US$12,746 or more. “Killing the Hypnozoite—Drug Discovery Approaches to Prevent Relapse in Plasmodium vivax.” Pathogens and 1. For more information on the study, visit http://www.who Global Health 109 (3): 107–22. .int/malaria/news/2016/llins-effective-tool-malaria-fight/en/. Carter, N., A. Pamba, S. Duparc, and J. N. Waitumbi. 2011. 2. P. vivax and P. ovale infections can form hypnozoites that “Frequency of Glucose-6-Phosphate Dehydrogenase can cause relapses months and even years after the initial Deficiency in Malaria Patients from Six African Countries infection. Enrolled in Two Randomized Anti-Malarial Clinical Trials.” 360 Major Infectious Diseases Malaria Journal 10: 241. http://www.ncbi.nlm.nih.gov Hahn, M. B., R. E. Gangnon, C. Barcellos, G. P. Asner, and /pmc/articles/PMC3188486/. J. A. Patz. 2014. “Influence of Deforestation, Logging, and Carter, R., and K. N. Mendis. 2002. “Evolutionary and Historical Fire on Malaria in the Brazilian Amazon.” PLoS One 9 (1). Aspects of the Burden of Malaria.” Clinical Microbiology Hemingway, J., H. Ranson, A. Magill, J. Kolaczinski, Reviews 15 (4): 564–94. C. Fornadel, and others. 2016. “Averting a Malaria Disaster: Castro, M. C., and B. H. Singer. 2011. “Malaria in the Brazilian Will Insecticide Resistance Derail Malaria Control?” The Amazon.” In Water and Sanitation-Related Diseases and Lancet 387 (10029): 1785–88. http://dx.doi.org/10.1016 the Environment: Challenges, Interventions, and Preventive /S0140-6736(15)00417-1. Measures, edited by J. M. H. Selendy, 401–20. Hoboken, NJ: Hochman, S., and K. Kim. 2009. “The Impact of HIV and Wiley-Blackwell. Malaria Coinfection: What Is Known and Suggested Venues Chinery, W. A. 1984. “Effects of Ecological Changes on the for Further Study.” Interdisciplinary Perspectives on Infectious Malaria Vectors Anopheles funestus and the Anopheles gam- Diseases 2009: 617954. biae Complex of Mosquitoes in Accra, Ghana.” Journal of IVCC. 2016. “Annual Report 2015/2016.” http://www.ivcc Tropical Medicine and Hygiene 87 (2): 75–81. .com/download/file/fid/826. Deane, L. M. 1988.“Malaria Studies and Control in Brazil.” American Keiser, J., B. H. Singer, and J. Utzinger, J. 2005. “Reducing Journal of Tropical Medicine and Hygiene 38 (2): 223–30. the Burden of Malaria in Different Eco-Epidemiological Delrieu, I., D. Leboulleux, K. Ivinson, and B. D. Gessner. 2015. Settings with Environmental Management: A Systematic “Design of a Phase III Cluster Randomized Trial to Assess Review.” The Lancet Infectious Diseases 5 (11): 695–708. the Efficacy and Safety of a Malaria Transmission Blocking Keiser, J., J. Utzinger, M. C. Castro, T. Smith, M. Tanner, and Vaccine.” Vaccine 33 (13): 1518–26. others. 2004. “Urbanization in Sub-Saharan Africa and Dondorp, A., C. I. Fanello, I. C. Hendriksen, E. Gomes, Implications for Malaria Control.” American Journal of A. Seni, and others. 2013. “Artesunate versus Quinine in Tropical Medicine and Hygiene 71 (Suppl 2): 118–27. the Treatment of Severe Falciparum Malaria in African Khatib, R. A., M. Selemani, G. A. Mrisho, I. M. Masanja, Children (AQUAMAT): An Open-Label, Randomised M. Amuri, and others. 2013. “Access to Artemisinin-Based Trial.” The Lancet 376 (9753): 1647–57. http://www.science- Anti-Malarial Treatment and Its Related Factors in Rural direct.com/science/article/pii/S0140673610619241. Tanzania.” Malaria Journal 12 (1): 155. Dondorp, A., F. F. Nosten, K. F. Stepniewska, N. F. Day, and Knudsen, J., and L. von Seidlein. 2014. Healthy Homes in N. White. 2005. “Artesunate versus Quinine for Treatment Tropical Zones: Improving Rural Housing in Asia and Africa. of Severe Falciparum Malaria: A Randomised Trial.” The Felbach, Germany: Edition Axel Menges GmbH. Lancet 366 (9487): 717–25. Koenker, H., J. Keating, M. Alilio, A. Acosta, M. Lynch, and Elslager, E. F. 1969. “Progress in Malaria Chemotherapy. 1. F. Nafo-Traore. 2014. “Strategic Roles for Behaviour Change Repository Antimalarial Drugs.” Progress in Drug Research Communication in a Changing Malaria Landscape.” Malaria 13: 170–216. Journal 13 (1). http://dx.doi.org/10.1186/1475-2875-13-1. Galactionova, K., F. Tediosi, F. Camponovo, T. A. Smith, P. W. Konradsen, F., W. van der Hoek, F. P. Amerasinghe, C. Mutero, Gething, and others. 2017. “Country-Specific Predictions and E. Boelee. 2004. “Engineering and Malaria Control: of the Cost-Effectiveness of Malaria Vaccine RTS,S/AS01 in Learning from the Past 100 Years.” Acta Tropica 89 (2): 99–108. Endemic Africa.” Vaccine 35 (1): 53–60. Lambrechts, L., N. M. Ferguson, E. Harris, E. C. Holmes, Galactionova, K., F. Tediosi, D. de Savigny, T. Smith, and E. A. McGraw, and others. 2015. “Assessing the M. Tanner. 2015. “Effective Coverage and Systems Epidemiological Effect of Wolbachia for Dengue Control.” Effectiveness for Malaria Case Management in Sub-Saharan The Lancet Infectious Diseases 15 (7): 862–66. http://dx.doi African Countries.” PLoS One 10 (5): e0127818. http:// .org/10.1016/S1473-3099(15)00091-2. dx.doi.org/10.1371/journal.pone.0127818. Lengeler, C. 2004. “Insecticide-Treated Bed Nets and Curtains Gallup, J. L., and J. D. Sachs. 2001. “The Economic Burden for Preventing Malaria.” Cochrane Database of Systematic of Malaria.” American Journal of Tropical Medicine and Reviews 2000 (2): CD000363. Hygiene 64 (Suppl 1–2): 85–96. Lindsay, S. W., P. M. Emerson, and J. D. Charlwood. 2002. Gantz, V. M., and E. Bier. 2015. “The Mutagenic Chain Reaction: “Reducing Malaria by Mosquito-Proofing Houses.” Trends A Method for Converting Heterozygous to Homozygous in Parasitology 18 (11): 510–14. Mutations.” Science 348 (6233): 442–44. http://science Littrell, M., J. Miller, M. Ndhlovu, B. Hamainza, M. Hawela, and .sciencemag.org/content/348/6233/442.abstract. others. 2013. “Documenting Malaria Case Management Gething, P. W., D. L. Smith, A. P. Patil, A. J. Tatem, R. W. Snow, Coverage in Zambia: A Systems Effectiveness Approach.” and others. 2010. “Climate Change and the Global Malaria Malaria Journal 12 (1): 371. http://www.malariajournal Recession.” Nature 465: (7296): 342–45. .com/content/12/1/371. Gomes, M. F., M. A Faiz, J. O. Gyapong, M. Warsame, Llanos-Cuentas, A., M. V. Lacerda, R. Rueangweerayut, T. Agbenyega, and others. 2009. “Pre-Referral Rectal Arte- S. Krudsood, S. K. Gupta, and others. 2014. “Tafenoquine sunate to Prevent Death and Disability in Severe Malaria: plus Chloroquine for the Treatment and Relapse Prevention a Placebo-Controlled Trial.” The Lancet 373 (9663): 14–20. of Plasmodium vivax Malaria (DETECTIVE): A Multicentre, Gorgas, G. W. C. 1915. Sanitation in Panama. New York: Double-Blind, Randomised, Phase 2b Dose-Selection D. Appleton and Co. Study.” The Lancet 383 (9922): 1049–58. Malaria Control 361 Maheu-Giroux, M., and M. C. Castro. 2013. “Do Malaria Vector Nájera, J. A., M. González-Silva, and P. L. Alonso. 2011. Control Measures Impact Disease-Related Behaviour and “Some Lessons for the Future from the Global Malaria Knowledge? Evidence from a Large-Scale Larviciding Eradication Programme (1955–1969).” PLoS Med 8 (1): Intervention in Tanzania.” Malaria Journal 12: 422. http:// e1000412. http://www.ncbi.nlm.nih.gov/pmc/articles dx.doi.org/10.1186/1475-2875-12-422. /PMC3026700. Malaria Vaccine Funders Group. 2013. Malaria Vaccine Neiva, A. 1940. “Profilaxia da malária e trabalhos de engen- Technology Roadmap. http://www.who.int/immunization haria: Notas, comentários, recordações.” Revista do Clube de /topics/malaria/vaccine_roadmap/TRM_update_nov13 Engenharia 6 (70): 60–75. .pdf?ua=1. Nunes, J. K., C. Woods, T. Carter, T. Raphael, M. J. Morin, and malERA Consultative Group on Vector Control. 2011. others. 2014. “Development of a Transmission-Blocking “A Research Agenda for Malaria Eradication: Vector Malaria Vaccine: Progress, Challenges, and the Path Control.” PLoS Medicine 8 (1): e1000401. http://www.ncbi Forward.” Vaccine 32 (343): 5531–39. .nlm.nih.gov/pmc/articles/PMC3026704/. Obrist, B., N. Iteba, C. Lengeler, A. Makemba, C. Mshana, Marchesinig, P., and J. Crawley. 2004. “Reducing the Burden of and others. 2007. “Access to Health Care in Contexts Anemia in Infants and Young Children in Malaria-Endemic of Livelihood Insecurity: A Framework for Analysis Countries of Africa: From Evidence to Action.” American and Action.” PLoS Medicine 4 (10): e308. http://dx.doi Journal of Tropical Medicine and Hygiene 71 (Suppl 2): .org/10.1371%2Fjournal.pmed.0040308. 25–34. Ogoma, S. B., K. Kannady, M. Sikulu, P. Chaki, N. Govella, McCarthy, J. S., J. Lotharius, T. Rickle, S. Chalon, J. Phillips, and and others. 2009. “Window Screening, Ceilings and Closed others. 2017. “Safety, Tolerability, Pharmacokinetics, and Eaves as Sustainable Ways to Control Malaria in Dar es Activity of the Novel Long-Acting Antimalarial DSM265: Salaam, Tanzania.” Malaria Journal 8: 221. A Two-Part First-in-Human Phase 1a/1b Randomised Ogoma, S. B., H. Ngonyani, E. T. Simfukwe, A. Mseka, J. Moore, Study.” The Lancet Infectious Diseases. 17 (6): 626–35. http:// and others. 2014. “The Mode of Action of Spatial Repellents dx.doi.org/10.1016/S1473-3099(17)30171-8. and Their Impact on Vectorial Capacity of Anopheles gam- McCarthy, J. S., S. Sekuloski, P. M. Griffin, S. Elliott, N. Douglas, biae sensu stricto.” PLoS One 9 (12): e110433. http://dx.doi and others. 2011. “A Pilot Randomised Trial of Induced .org/10.1371%2Fjournal.pone.0110433. Blood-Stage Plasmodium falciparum Infections in Healthy Okumu, F. O., and S. J. Moore. 2011. “Combining Indoor Volunteers for Testing Efficacy of New Antimalarial Drugs.” Residual Spraying and Insecticide-Treated Nets PLOS One 6 (8): e21914. http://dx.doi.org/10.1371/journal for Malaria Control in Africa: A Review of Possible .pone.0021914. Outcomes and an Outline of Suggestions for the Future.” McPake, B., D. Asiimwe, F. Mwesigye, M. Ofumbi, L. Ortenblad, Malaria Journal 10 (1): 208–22. http://dx.doi.org/10.1186 and others. 1999. “Informal Economic Activities of Public /1475-2875-10-208. Health Workers in Uganda: Implications for Quality and Olotu, A., G. Fegan, J. Wambua, G. Nyangweso, K. O. Awuondo, Accessibility of Care.” Social Science and Medicine 49 (7): and others. 2013. “Four-Year Efficacy of RTS,S/AS01E and 849–65. Its Interaction with Malaria Exposure.” New England Journal Mikkelsen-Lopez, I., F. Tediosi, G. Abdallah, M. Njozi, of Medicine 368 (12): 1111–20. http://dx.doi.org/10.1056 B. Amuri, and others. 2013. “Beyond Antimalarial Stock- /NEJMoa1207564. Outs: Implications of Health Provider Compliance on Pasay, C. J., R. Rockett, S. Sekuloski, P. Griffin, L. Marquart, Out-of-Pocket Expenditure during Care-Seeking for Fever and others. 2016. “Piperaquine Monotherapy of Drug- in South East Tanzania.” BMC Health Services Research 13: Susceptible Plasmodium falciparum Infection Results in Rapid 444. http://dx.doi.org/10.1186/1472-6963-13-444. Clearance of Parasitemia but Is Followed by the Appearance Mnzava, A. P., T. B. Knox, E. A. Temu, A. Trett, C. Fornadel, of Gametocytemia.” Journal of Infectious Diseases 214 (1): and others. 2015. “Implementation of the Global Plan for 105–13. http://dx.doi.org/10.1093/infdis/jiw128. Insecticide Resistance Management in Malaria Vectors: Penny, M. A., K. Galactionova, M. Tarantino, M. Tanner, and Progress, Challenges and the Way Forward.” Malaria T. A. Smith. 2015. “The Public Health Impact of Malaria Journal 14: 173. http://www.ncbi.nlm.nih.gov/pmc/articles Vaccine RTS,S in Malaria Endemic Africa: Country-Specific /PMC4423491/. Predictions Using 18 Month Follow-Up Phase III Data and Mulamba, C., J. M. Riveron, S. S. Ibrahim, H. Irving, Simulation Models.” BMC Medicine 13 (1): 170–90. http:// K. G. Barnes, and others. 2014. “Widespread Pyrethroid dx.doi.org/10.1186/s12916-015-0408-2. and DDT Resistance in the Major Malaria Vector Anopheles Penny, M. A., R. Verity, C. A. Bever, C. Sauboin, funestus in East Africa Is Driven by Metabolic Resistance K. Galactionova, and others. 2016. “Public Health Impact Mechanisms.” PLoS One 9 (10): e110058. http://dx.doi and Cost-Effectiveness of the RTS, S/AS01 Malaria Vaccine: .org/10.1371/journal.pone.0110058. A Systematic Comparison of Predictions from Four Mumba, M., J. Visschedijk, M. van Cleef, and B. Hausman. 2003. Mathematical Models.” The Lancet 387: 367–75. “A Piot Model to Analyse Case Management in Malaria Pinotti, M. 1951. “The Biological Basis for the Campaign against Control Programs.” Tropical Medicine and International the Malaria Vectors in Brazil.” Transactions of the Royal Society Health 8 (6): 544–51. of Tropical Medicine and Hygiene 44 (6): 663–82. 362 Major Infectious Diseases Pomeroy, A. W. J. 1920. “The Prophylaxis of Malaria in Dar 4,4’-Diacetyldiaminodiphenylsulfone, Alone and in es Salaam, East Africa.” Journal of the Royal Army Medical Combination, against Infections with Plasmodium Corps 35 (1): 44–68. Cynomolgi in Rhesus Monkeys.” Antimicrobial Agents and Price, R. N., E. Tjitra, C. A. Guerra, S. Yeung, N. J. White, and Chemotherapy 26 (5): 611–42. http://www.ncbi.nlm.nih others. 2007. “Vivax Malaria: Neglected and Not Benign.” .gov/pmc/articles/PMC179984/. American Journal of Tropical Medicine and Hygiene 77 Seder, R. A., L. J. Chang. M. E. Enama, K. L. Zephir, U. N. Sarwar, (Suppl 6): 79–87. and others. 2013. “Protection against Malaria by Intravenous Pronker, E. S., T. C. Weenen, H. Commandeur, E. H. Claassen, Immunization with a Nonreplicating Sporozoite Vaccine.” and A. D. Osterhaus. 2013. “Risk in Vaccine Research and Science 341 (6152): 1359–65. Development Quantified.” PLoS One 8 (3): e57755. http:// Shaw, W. R., P. Marcenac, L. M. Childs, C. O. Buckee, F. Baldini, dx.doi.org/10.1371%2Fjournal.pone.0057755. and others. 2016. “Wolbachia Infections in Natural Anopheles Qualls, W. A., G. C. Müller, S. F. Traore, M. M. Traore, Populations Affect Egg Laying and Negatively Correlate K. L. Arheart, and others. 2015. “Indoor Use of Attractive with Plasmodium Development.” Nature Communications Toxic Sugar Bait (ATSB) to Effectively Control Malaria 7: 11772. http://dx.doi.org/10.1038/ncomms11772. Vectors in Mali, West Africa.” Malaria Journal 14 (1): 301–8. Singer, B. H., and M. C. Castro. 2011. “Reassessing Multiple- http://dx.doi.org/10.1186/s12936-015-0819-8. Intervention Malaria Control Programs of the Past: Lessons Quinones, M. L., D. E. Norris, J. E. Conn, M. Moreno, for the Design of Contemporary Interventions.” In Water T. R. Burkot, and others. 2015. “Insecticide Resistance in and Sanitation-Related Diseases and the Environment: Areas under Investigation by the International Centers of Challenges, Interventions, and Preventive Measures, edited Excellence for Malaria Research: A Challenge for Malaria by J. M. H. Selendy, 151–66. Hoboken, NJ: Wiley-Blackwell. Control and Elimination.” Amerian Journal of Tropical Sluydts, V., L. Durnez, S. Heng, C. Gryseels, L. Canier, and Medicine and Hygiene 93 (3): 68–78. others. 2016. “Efficacy of Topical Mosquito Repellent Rao, B. V., D. Schellenberg, and A. Ghani. 2013a. “Overcoming (Picaridin) plus Long-Lasting Insecticidal Nets Versus Health Systems Barriers to Successful Malaria Treatment.” Long-Lasting Insecticidal Nets Alone for Control of Trends in Parasitology 29 (4): 164–80. Malaria: A Cluster Randomised Controlled Trial.” The ———. 2013b. “The Potential Impact of Improving Appropriate Lancet Infectious Disease 16 (10): 1169–77. http://dx.doi Treatment for Fever on Malaria and Non-Malaria Febrile .org/10.1016/S1473-3099(16)30148-7. Illness Management in Under-5s: A Decision-Tree Modeling Smith, D. L., K. E. Battle, S. I. Hay, C. M. Barker, T. W. Scott, Approach.” PLoS One 8 (7): e69654. and F. E. McKenzie. 2012. “Ross, Macdonald, and a Theory RBM (Roll Back Malaria) Partnership. 2008. Global Malaria for the Dynamics and Control of Mosquito-Transmitted Action Plan for a Malaria-Free World. Geneva: World Health Pathogens.” PLoS Pathogens 8 (4): e1002588. http://dx.doi Organization. .org/10.1371%2Fjournal.ppat.1002588. ———. 2012. The Strategic Framework for Malaria Stratton, L., M. S. O’Neill, M. E. Kruk, and M. L. Bell. Communication at the Country Level, 2012–2017. Geneva: 2008. “The Persistent Problem of Malaria: Addressing the RBM Partnership. Fundamental Causes of a Global Killer.” Social Science and ———. 2015. Action and Investment to Defeat Malaria 2016– Medicine 67 (5): 854–62. http://www.sciencedirect.com 2030. Geneva: World Health Organization, http://www /science/article/pii/S0277953608002347. .rollbackmalaria.org/about/about-rbm/aim-2016-2030. Stromquist, W. G. 1920. “Malaria Control from the Engineering Richie, T. L., P. F. Billingsley, B. K. L. Sim, E. R. James, Point of View.” American Journal of Public Health 10 (6): S. Chakravarty, and others. 2015. “Progress with Plasmodium 497–501. falciparum Sporozoite (PfSPZ)–Based Malaria Vaccines.” Sulyok, M., T. Rückle, A. Roth, R. E. Mürbeth, S. Chalon, Vaccine 33 (52): 7452–61. http://www.sciencedirect.com and others. 2017. “DSM265 for Plasmodium falciparum /science/article/pii/S0264410X15013869. Chemoprophylaxis: A Randomised, Double Blinded, Phase Roberts, L., and M. Enserink. 2007. “Did They Really Say ... 1 Trial with Controlled Human Malaria Infection.” The Eradication?” Science 318 (5856): 1544–45. http://science Lancet Infectious Diseases 17 (6): 636–44. http://dx.doi .sciencemag.org/content/318/5856/1544.abstract. .org/10.1016/S1473-3099(17)30139-1. RTS,S/AS01 Clinical Trials Partnership. 2012. “A Phase 3 Trial Sumba, P. O., L. S. Wong, H. K. Kanzaria, K. A. Johnson, and of RTS,S/AS01 Malaria Vaccine in African Infants.” New C. C. John. 2008. “Malaria Treatment-Seeking Behavior England Journal of Medicine 367 (24): 2284–95. http://dx and Recovery from Malaria in Highland Area of Kenya.” .doi.org/10.1056/NEJMoa1208394. Malaria Journal 7: 245. Sattler, M. A., D. Mtasiwa, M. Kiama, Z. Premji, M. Tanner, Tanner, M. and P. L. Alonso. 2010. “Research and Development and others. 2005. “Habitat Characterization and Spatial for the Malaria Elimination/Eradication Roadmap.” Distribution of Anopheles sp. Mosquito Larvae in Dar es Malaria Journal 9 (Suppl 2): 11. Salaam (Tanzania) during an Extended Dry Period.” Malaria Tusting, L., M. Ippolito, B. Willey, I. Kleinschmidt, G. Dorsey, Journal 4: 4. http://dx.doi.org/10.1186/1475-2875-4-4. and others. 2015. “The Evidence for Improving Housing to Schmidt, L. H., and R. N. Rossan. 1984. “Activities of Reduce Malaria: A Systematic Review and Meta-Analysis.” Respository Preparations of Cycloguanil Pamoate and Malaria Journal 14: 209. Malaria Control 363 Upton, L. M., P. M. Brock, T. S. Churcher, A. C. Ghani, ———. 2014. World Malaria Report 2014. Geneva: WHO. P. W. Gething, and others. 2015. “Lead Clinical and ———. 2015a. Global Technical Strategy for Malaria Preclinical Antimalarial Drugs Can Significantly Reduce 2016–2030. Geneva: WHO. http://apps.who.int/iris/bitstr Sporozoite Transmission to Vertebrate Populations.” eam/10665/176712/1/9789241564991_eng.pdf?ua=1. Antimicrobial Agents and Chemotherapy 59 (1): 490–97. ———. 2015b. WHO Malaria Treatment Guidelines. 3rd ed. Utzinger, J., Y. Tozan, F. Doumani, and B. H. Singer. 2002. Geneva: WHO. “The Economic Payoffs of Integrated Malaria Control in ———. 2016a. WHO Malaria Terminology. Geneva: WHO. the Zambian Copperbelt between 1930 and 1950.” Tropical http://apps.who.int/iris/bitstream/10665/208815/1/WHO Medicine and International Health 7 (8): 657–77. _HTM_GMP_2016.6_eng.pdf. Wagman, J. M., J. P. Grieco, K. Bautista, J. Polanco, I. Brice, ———. 2016b. World Malaria Report 2015. Geneva: WHO. and others. 2015. “The Field Evaluation of a Push-Pull ———. 2016c. World Malaria Report 2016. Geneva: WHO. System to Control Malaria Vectors in Northern Belize, WHO Malaria Policy Advisory Committee and Secretariat. Central America.” Malaria Journal 14: 184. http://dx.doi 2015. “Malaria Policy Advisory Committee to the WHO: .org/10.1186/s12936-015-0692-5. Conclusions and Recommendations of Seventh Biannual Watson, M. 1921. The Prevention of Malaria in the Federated Meeting (March 2015).” Malaria Journal 14: 295. http:// Malay States: A Record of Twenty Years’ Progress. London: www.ncbi.nlm.nih.gov/pmc/articles/PMC4524000/. J. Murray. WHO/TDR (Special Programme for Research and Training ———. 1953. African Highway: The Battle for Health in Central in Tropical Diseases) and Foundation for the National Africa. London: J. Murray. Institutes of Health. 2014. Guidance Framework for Webster, J., F. Baiden, J. Bawah, J. Bruce, M. Tivura, and others. Testing of Genetically Modified Mosquitoes. Geneva: 2014. “Management of Febrile Children under Five Years WHO. in Hospitals and Health Centers of Rural Ghana.” Malaria Wilson, A. L., and on behalf of the IPTc Taskforce. 2011. Journal 13: 261. “A Systematic Review and Meta-Analysis of the Efficacy Wells, T. N. C., R. H. van Huijsduijnen, and W. C. Van Voorhis. and Safety of Intermittent Preventive Treatment of Malaria 2015. “Malaria Medicines: A Glass Half Full?” Nature in Children (IPTc).” PLoS ONE 6 (2): e16976. http://www Reviews Drug Discovery 14 (6): 424–42. http://dx.doi .ncbi.nlm.nih.gov/pmc/articles/PMC3038871/. .org/10.1038/nrd4573. Yamana, T. K., and E. A. Eltahir. 2013. “Projected Impacts of White, M. T., L. Conteh, R. Cibulskis, and A. C. Ghani. 2011.“Costs Climate Change on Environmental Suitability for Malaria and Cost-Effectiveness of Malaria Control Interventions: Transmission in West Africa.” Environmental Health A Systematic Review.” Malaria Journal 10 (1): 337. Perspectives 121 (10): 1179–86. White, N. J., T. T. Duong, C. Uthaisin, F. Nosten, A. P. Phyo, and Zurovac, D., and A. K. Rowe. 2006. “Quality of Treatment for others. 2016. “Antimalarial Activity of KAF156 in Falciparum Febrile Illness among Children at Outpatient Facilities and Vivax Malaria.” New England Journal of Medicine 375 in Sub-Saharan Africa.” Annals of Tropical Medicine and (12): 1152–60. http://dx.doi.org/10.1056/NEJMoa1602250. Parasitology 100 (4): 283–96. Whitfield, J. 2002. “Portrait of a Serial Killer: A Roundup of the Zurovac, D., J. K. Tibenderana, J. Nankabirwa, J. Ssekitooleko, History and Biology of the Malaria Parasite.” Nature. http:// J. N. Njogu, and others. 2008. “Malaria Case-Management dx.doi.org/10.1038/news021001-6. under Artemether-Lumefantrine Treatment Policy WHO (World Health Organization). 2012. World Malaria in Uganda.” Malaria Journal 7: 181. http://dx.doi Report 2012. Geneva: WHO. .org/10.1186/1475-2875-7-181. 364 Major Infectious Diseases Chapter 14 Febrile Illness in Adolescents and Adults John A. Crump, Paul N. Newton, Sarah J. Baird, and Yoel Lubell INTRODUCTION overdiagnosis among febrile patients in many areas. Documented declines in malaria since 2004 (Murray and Fever is one of the most common symptoms reported by others 2014; WHO 2013) and expansion of nonmalarial patients seeking health care in low-resource areas in the infections such as dengue (Stanaway and others 2016), tropics, where it may occur either in isolation or in asso- combined with widespread use of malaria diagnostic tests ciation with other common symptoms such as cough or (WHO 2010b), mean that health care workers face a diarrhea (Feikin and others 2011; Prasad, Sharples, and growing proportion of patients with fever and a negative others 2015). Fever without localizing features presents a malaria diagnostic test. This increase is troubling because particular challenge to health care workers and health patients presenting to hospitals with fever not due to systems because it may be caused by a wide range of malaria are as likely to die as those who have malaria bacterial, fungal, parasitic, and viral infections (Crump (Reyburn and others 2004). Furthermore, vertical pro- and others 2013; Mayxay and others 2013), as well as by grams exist rarely for febrile illnesses other than malaria. noninfectious conditions. Clinical assessment has lim- This chapter identifies key challenges, issues for diagno- ited accuracy both for identifying the likely cause and for sis and treatment of relevant infections, and data gaps the early recognition of patients who will progress to for health care workers and policy makers regarding non- serious or fatal disease. Compounding the limitations of malarial fever management and its cost-effectiveness. We clinical assessment is the dearth of available epidemio- highlight the needs for increasing etiologic research, logic data on common causes of fever (Crump 2014) restructuring of burden-of-disease estimates to recognize and absence of clinical laboratory services in many areas nonmalarial fever, and development of approaches to eval- (Archibald and Reller 2001; Petti and others 2006). uating clinical interventions to improve patient outcomes. Given its prevalence and severity, malaria has been the common default diagnosis for fever without localizing features in the tropics for decades (WHO 2006). In BURDEN OF DISEASE FROM NONMALARIAL countries historically highly endemic for malaria, fever is FEBRILE ILLNESS controlled and managed with vertical programs. However, a growing number of fever etiology studies (Prasad, No estimate has been made of the global burden of dis- Murdoch, and others 2015; Reyburn and others 2004) and ability and death due to febrile illnesses without localiz- more widespread use of malaria diagnostic tests (WHO ing features. Consequently, this group of infections has 2010b) have served to unmask the problem of malaria lacked collective prominence. By extension, estimating Corresponding author: John A. Crump, Centre for International Health, University of Otago, Dunedin, New Zealand; john.crump@otago.ac.nz. 365 the effect size of interventions for fevers is challenging. undertaken in Africa (N = 16) and South and South-East Disability-adjusted life years (DALYs), including deaths, Asia (N = 11) are described below (Prasad, Murdoch, and from diarrhea and pneumonia are estimated at the others 2015). These studies illustrate both many data gaps syndrome level before being assigned to individual and sufficient geographic and seasonal heterogeneity in causes by the World Health Organization (WHO) and the etiology of fevers to require subnational empiric treat- the Institute for Health Metrics and Evaluation (IHME).1 ment guidelines (White and others 2012). The studies also However, these sources estimate the individual disease confirm the need for guidelines to respond to changes burdens of some causes of fever, such as malaria, over time in the etiology of febrile illness and in antimi- dengue, and enteric fever, but not for others, including, crobial resistance of relevant pathogens. for example, chikungunya, leptospirosis, and Q fever. Thus, the organizational structures of global, national, and academic public health institutions rarely include Geography cross-cutting expertise addressing fever across the range Africa of responsible pathogens. Similarly, fever etiology Considerable data gaps exist in our understanding research has tended to focus on one or a small number of the etiology of severe febrile illness in Africa. Few of pathogens rather than on a broad range of causes studies investigate more than one or a small group of (Prasad, Murdoch, and others 2015). pathogens, and many countries and some regions Also contributing to the difficulty in estimating the lack contemporary studies. Furthermore, standard lab- global burden of death from febrile illnesses is the inade- oratory-based case definitions are not widely used, and quacy of autopsy procedures. Verbal autopsy has limited study designs rarely include control groups or other ability to distinguish deaths caused by malaria from other approaches to estimating pathogen-specific attributable conditions with fever. Verbal autopsy may classify febrile fractions. deaths as due to malaria in malaria-endemic areas. Many Reddy, Shaw, and Crump (2010) conducted a system- causes of febrile illness are not available for assignment atic review of prospective studies of the etiology of by verbal autopsy. Even with a positive malaria diagnostic community-acquired bloodstream infection in Africa. test, deaths assigned to cerebral malaria may be found to Their findings are as follows: be due to other causes on autopsy (Mallewa and others 2007). Complete diagnostic autopsy is not widely avail- • Of 58,296 patients enrolled in 22 eligible studies able in low-resource areas, but minimally invasive in 34 locations from 1984 through 2006, 2,051 autopsy approaches are being studied as a means to (13.5 percent) of 15,166 adolescents and adults had reduce uncertainty regarding causes of death in develop- nonmalarial bloodstream infections, yielding 2,078 ing countries (Bassat and others 2013). bloodstream isolates. • Of these isolates, 1,019 (49.0 percent) were Enterobacteriaceae, including 878 (42.3 percent) NONMALARIAL FEVER IN ADOLESCENTS Salmonella enterica, of which 553 (63.0 percent) AND ADULTS were Salmonella serovar Typhi, 5 (less than Since 2010, WHO guidelines have recommended that 1.0 percent) Salmonella serovar Paratyphi A, and 291 malaria treatment decisions be based on the result of (33.1 percent) nontyphoidal Salmonella. malaria diagnostic tests (WHO 2010b). For patients with • Among the Enterobacteriaceae, Salmonella Typhi positive malaria diagnostic tests, health care workers would predominated in North Africa, whereas nontyphoidal prescribe antimalarial treatment and consider co-infections Salmonella predominated in East Africa, West and as described in detail in chapter 12 of this volume (Shretta Central Africa, and Southern Africa. and others 2017). However, for those patients with nega- • Among the 141 (6.8 percent) non–Salmonella tive malaria diagnostic tests, health care workers often lack Enterobacteriaceae, Escherichia coli accounted the epidemiological information or laboratory services for 77 (54.6 percent), Klebsiella species (spp.) for necessary to support rational diagnostic and management 24 (17.0 percent), Proteus mirabilis for 17 (12.1 percent), decisions (Crump, Gove, and Parry 2011). and Shigella spp. for 10 (7.1 percent). • Other Gram-negative organisms caused 341 blood- stream infections (16.4 percent), of which Brucella Etiology of Nonmalarial Fever spp. accounted for 275 (80.6 percent), occurring Annex 14A lists studies conducted from 1980 through predominantly in North Africa; Neisseria spp. for 2013 of the etiology of severe febrile illness among, pre- 22 (6.5 percent); Acinetobacter spp. for 16 (4.7 percent); dominantly, adolescents and adults. Key features of studies and Pseudomonas spp. for 15 (4.3 percent). 366 Major Infectious Diseases • Of the 336 (16.2 percent) Gram-positive isolates, other Gram-negative organisms. China was excluded from Streptococcus pneumoniae accounted for 198 (58.9 the review, and no reports were found from peninsular percent), Staphylococcus aureus for 111 (33.0 percent), India, representing an enormous gap in knowledge. and other streptococci for 21 (6.3 percent). In the Kathmandu Valley of Nepal, Blacksell, • Yeasts caused 39 bloodstream infections (1.9 percent Sharma, and others (2007) and Murdoch and others of the total), of which Cryptococcus spp. accounted (2004) identified the importance of typhoid, dengue, for 31 (79.5 percent) and Candida spp. for 5 leptospirosis, scrub typhus, and murine typhus. In (12.8 percent). Histoplasma capsulatum is sometimes Papua, Indonesia, Punjabi and others (2012) found isolated from blood culture (Archibald and others among 227 predominantly adult patients hospitalized 1998), but urine-antigen testing detects more cases with negative malaria diagnostic tests that the most (Lofgren and others 2012). common etiological diagnoses were typhoid, leptospiro- • Mycobacterial bloodstream infections were found in sis, rickettsioses, and dengue. 173 of the patients tested, of which 166 (96.0 percent) In a large study of patients ages 7–49 years at three were due to Mycobacterium tuberculosis complex, and health centers in rural Cambodia, Mueller and others 2 (1.2 percent) to M. avium complex. (2014) identified at least one pathogen in 73.3 percent of febrile patients. The most frequent pathogens were the The bacterial zoonoses brucellosis, leptospirosis, malaria parasites Plasmodium vivax (33.4 percent) and Q fever, and rickettsial infections are also important P. falciparum (26.5 percent). Others included patho- causes of febrile illness in Africa. Brucellosis appears to be genic Leptospira spp. (9.4 percent), influenza viruses particularly common in North Africa (Afifi and others (8.9 percent), dengue viruses (6.3 percent), and Orientia 2005; Jennings and others 2007; Reddy, Shaw, and Crump tsutsugamushi (3.9 percent). However, in the control 2010), but it also occurs in Sub-Saharan Africa (Dean group, consisting of nonfebrile persons accompanying and others 2012). Although not often sought, leptospiro- febrile patients to health centers, a potential pathogen sis is a common cause of febrile illness in Africa, identi- was identified in 40.4 percent of participants, most com- fied as the cause of fever in up to 20 percent of inpatients monly malaria parasites and Leptospira spp. in some studies (Parker and others 2007). Q fever was In a similar study, but without a control group, Mayxay responsible for 2 percent to 9 percent of febrile hospital- and others (2013) investigated the etiology of fever in izations according to a systematic review of African inpa- patients ages 5–49 years presenting at two provincial hos- tient studies (Vanderburg and others 2014). Spotted fever pitals in rural northern and southern areas of the Lao group rickettsioses and, in some locations, typhus group People’s Democratic Republic. They identified at least one rickettsioses are also common among febrile inpatients pathogen in 41 percent of patients at diagnosis, most com- (Prabhu and others 2011). Viral infections including monly dengue (8 percent), scrub typhus (7 percent), influenza (Yazdanbakhsh and Kremsner 2009) and arbo- Japanese encephalitis virus (JEV) (6 percent), leptospirosis virus infections such as chikungunya, dengue, Rift Valley (6 percent), and bacteremia (2 percent). Influenza diag- fever, and others also may occur. nostics were available for one site, where influenza B was the most frequently detected type (87 percent). However, South and South-East Asia as described in Cambodia (Kasper and others 2010), Although a relatively large number of studies have exam- 50 percent of cases of influenza B would not have been ined the epidemiology of single diseases in Asia—for identified by surveillance for influenza-like illness. In rural example, typhoid, scrub typhus, and melioidosis—they Lao PDR, the contribution of bacteremia diagnosed by cover relatively few sites that are concentrated in conventional blood cultures was relatively low (2 percent). South-East Asia (Acestor and others 2012). Vast knowl- The etiologies in children and adults were similar, but the edge gaps persist for China and India, with no studies data were not stratified by outpatients and inpatients. examining the diversity of pathogens stratified by patient With regard to patient management, Mayxay and age, outpatient or inpatient status, and disease severity. others (2013) estimated that azithromycin, doxycycline, Deen and others (2012) identified 17 studies of the ceftriaxone, and ofloxacin would have had substan- etiology of community-acquired bloodstream infection in tial efficacy for 13 percent, 12 percent, 8 percent, and South and South-East Asia. Among those, pathogenic 2 percent of patients, respectively. They suggested that organisms were isolated from 12 percent of adults. Of empiric treatment with doxycycline for patients with adults with bloodstream infections, Salmonella enterica undifferentiated fever and negative rapid diagnostic serotype Typhi was the most common bacterial pathogen tests (RDTs) for malaria and dengue could be an (30 percent). Other commonly isolated organisms in appropriate strategy for rural health workers in Lao adults were Staphylococcus aureus, Escherichia coli, and PDR. Because JEV, usually without encephalitis, was an Febrile Illness in Adolescents and Adults 367 important cause of fever, JEV vaccination is likely to uncertainty for policy makers. A major impediment to have a substantial effect on reducing the frequency of better understanding the epidemiology of diverse infec- patients presenting with fever as well as those developing tions across the continent has been the dearth of quality- encephalitis (Mayxay and others 2013). assured diagnostic facilities in rural Asia—including the Despite many data gaps and uncertainties, the evidence substantial expense and human and technical capacity highlights the importance of typhoid, dengue, scrub they would require. The situation suggests that a new typhus, leptospirosis, and influenza viruses in South and model is needed for infectious disease diagnostic facili- South-East Asia. Relative to Africa, brucellosis and Q fever ties in the rural tropics—not one copied from high- appear to be less important. Consensus is greatly needed income countries (HICs) but a model designed for the on designing fever studies that emphasize, for example, the local pathogens and environment. inclusion of control groups, especially when sampling sites that are not normally sterile, and standardized reporting. Special Groups The studies’ variation in inclusion criteria and age stratifi- Specific issues arise when considering certain subgroups cation make summarizing and comparing data between of patients with fever—particularly pregnant women, sites difficult. The lack of reports from China and India is individuals infected with the human immunodeficiency especially troubling because, presumably, most persons in virus (HIV), people with diabetes, malaria patients, and Asia developing fevers live in these two countries. people at increased risk for occupational or other The lack of an evidence base for development and types of exposure to certain nonmalarial pathogens. testing of diagnostic accuracy and cost-effectiveness of Infections and considerations affecting these special algorithms of empirical treatment creates much groups are described in table 14.1. Table 14.1 Febrile Illness Considerations for Special Groups Group Disease Comments References Pregnant women Plasmodium falciparum and Listeria Few data are available on etiology Gravett and others 2012; monocytogenes are more common. and impact of nonmalarial fevers on Kourtis, Read, and Jamieson mother, fetus, and infant. 2014; Louie and others 2010; Hepatitis E and herpes simplex virus disease, Some antimicrobials that are important Machado and others 2013; malaria, and influenza are more severe. for treatment of bacterial pathogens McGready and others 2010; are contraindicated in pregnancy. Say and others 2014 Dengue, scrub typhus, and typhoid fever may be more severe. Obstetric sepsis. HIV infection Invasive bacterial and fungal infections, Knowledge of patient’s HIV infection Huson and others 2014; including Cryptococcus neoformans, C. grubii, status assists in differential diagnosis Reddy, Shaw, and Crump nontyphoidal Salmonella enterica, and of febrile illness; provider-initiated HIV 2010; Sanders and others Mycobacterium tuberculosis are more common. testing is recommended. 2014; WHO and UNAIDS Acute HIV infection may be common among Few data are available on interaction 2007 persons seeking care for fever in areas with between HIV and several febrile concentrated or generalized epidemics. illnesses. Diabetes Tuberculosis, infections with Enterobacteriaceae, Hyperglycemia impairs antibacterial Esper, Moss, and Martin lower respiratory tract infection, urinary tract function of neutrophils and T cell– 2009; Faurholt-Jepsen and infection, skin and mucous membrane infection, mediated immune response. others 2013; Figueiredo and melioidosis are more common. and others 2010; Kapur and Harries 2013; Knapp Poorer outcomes occur with tuberculosis and Few data are available from 2013; Muller and others infections with Enterobacteriaceae. low-resource settings. 2005; Park and others 2011; Few data are available on interaction Suputtamongkol and others between diabetes and several febrile 1999; Thomsen and others illnesses such as leptospirosis and 2005; Van den Berghe and rickettsial infections. others 2006 table continues next page 368 Major Infectious Diseases Table 14.1 Febrile Illness Considerations for Special Groups (continued) Group Disease Comments References Malaria Nontyphoidal S. enterica bloodstream infection Positive malaria film may mean Abba and others 2011; and other bacteremias are more common. malaria is the cause of current illness, Feasey and others 2012; a cofactor in a concurrent infection, or Kyabayinze and others 2008; an incidental finding. Nadjm and others 2010 P. falciparum malaria rapid diagnostic tests may remain positive for a month or more after clearance of parasites. Occupational and Rural residence increases risk for leptospirosis, Knowledge of patient’s occupational Breiman and others 2012; other exposures melioidosis, scrub typhus, and spotted fever and exposure history assists in Cheng and Currie 2005; group rickettsioses. differential diagnosis of febrile illness. Shirima and others 2007; Pastoralist animal husbandry practices increase Vallée and others 2010; risk for brucellosis. Wardrop and others 2013 Q fever is more common among livestock keepers and abattoir workers. Urban residence is associated with enteric fever and murine typhus. Note: HIV = human immunodeficiency virus. Diagnosis and Treatment of Specific Infections diagnostic research and development. We will not be able The initial challenge for managing the febrile patient is to understand and provide evidence to guide health pol- deciding whether an antimicrobial agent is indicated and, icy without a surge in investment in accurate, simple, and if so, selecting the most appropriate empiric treatment. affordable diagnostic tests for these neglected diseases. Making an etiologic diagnosis allows rationalization and The success of nonstructural protein 1 (NS1)-based correction of initial treatment. Considerations for diag- dengue POCTs in assisting with dengue diagnosis in nosis and treatment of specific infections are summa- rural facilities suggests that such tests can be developed. rized in table 14.2. Other challenges concerning diagnosis and treatment Poor Quality of Diagnostics and Therapeutic Products of febrile patients involve the quality of diagnostics and The diagnostic tests, vaccines, and medicines necessary medicines for many of the pathogens considered in this to reduce the burden of fevers are often of poor quality, chapter. These issues can be divided into two groups: especially in countries with insufficient regulation inadequate diagnostic capacity and poor quality of diag- (Caudron and others 2008; Mori, Ravinetto, and nostics and therapeutic products. Jacobs 2011; Newton and others 2009). The evidence regarding antimalarials suggests severe quality issues Inadequate Diagnostic Capacity (Tabernero and others 2014), but data regarding diag- Because of the relative paucity of investment in and nostic tests, vaccines, and other classes of medicines understanding of febrile illnesses, accurate, accessible are limited. point-of-care tests (POCTs) are currently inadequate, The only long-term sustainable solution to the poor and evaluations of their diagnostic accuracy and cost- quality of diagnostics and therapeutic products would effectiveness are insufficient. Although whole-genome be the dramatic strengthening of the authorities that sequencing of pathogens is now commonly performed in regulate medicines in the 30 percent or so of countries research settings in both HICs and low- and middle- without functional capacity (Newton and others 2009). income countries (LMICs), no accurate, simple, and Health care providers should exercise caution in using affordable diagnostic tests are available for key infections diagnostic tests, in case the instructions either inflate such as typhoid, scrub typhus, and leptospirosis in the products’ claims for diagnostic accuracy or are of peripheral health care facilities (Peacock and Newton poor quality. Furthermore, they should be aware that 2008). This deficiency is in part due to intrinsic difficul- poor patient outcomes may reflect poor-quality medi- ties such as low bacterial blood load (Dittrich and others cines, both substandard and falsified, rather than the 2014) but also due to a lack of investment in targeted disease process itself. Febrile Illness in Adolescents and Adults 369 Table 14.2 Nonmalarial Febrile Diseases: Exposure, Diagnosis, Prevention, and Treatment Disease type Subgroup Exposure Diagnosis Prevention Treatment Arbovirus Dengue Arthropod • Detection of antibody and − Mosquito Supportive infections vectors NS1 antigen avoidance • Fourfold rise in antibody titer − Vector control • Demonstration of virus − Vaccine in blood by nucleic acid amplification test Japanese Arthropod • Culture or nucleic acid − Mosquito Supportive encephalitis vectors amplification test from avoidance virus serum, cerebrospinal fluid − Vector control (CSF), or tissue − Vaccine • Detection of virus-specific IgM in CSF confirmed by plaque reduction assay • Demonstration of virus in body fluid by nucleic acid amplification test Other (for Arthropod • Culture or nucleic acid − Mosquito Supportive example, vectors amplification test from avoidance chikungunya) serum or tissue − Vector control • Fourfold rise in antibody titer • Demonstration of virus in tissue by immunohistochemistry or nucleic acid amplification test Bloodstream Enteric fever Fecally • Blood culture − Improved water, Fluoroquinolones, extended- infection contaminated sanitation, and spectrum cephalosporins, or water or food food safety azithromycin, according to local − Detection and patterns of susceptibility treatment of infected persons − Vaccines Melioidosis Exposure to • Blood culture − Management • Ceftazidime contaminated • Culture of throat swab, pus, of exposure to • Carbapenems (for example, soil and water and other bodily fluids environmental meropenem) usually sources reserved for severe − Management infections or treatment of predisposing failures conditions • Amoxicillin-clavulanic acid as second-line therapy; trimethoprim- sulfamethoxazole alone for eradication phase Other Miscellaneous • Blood culture − Varies by Antimicrobials according to pathogen local patterns of susceptibility table continues next page 370 Major Infectious Diseases Table 14.2 Nonmalarial Febrile Diseases: Exposure, Diagnosis, Prevention, and Treatment (continued) Disease type Subgroup Exposure Diagnosis Prevention Treatment Brucellosis n.a. Exposure • Fourfold rise in antibody titer − Control in animal Doxycycline plus gentamicin to infected by microagglutination test husbandry or streptomycin, or doxycycline animals and • Nucleic acid amplification − Management of plus rifampin, for six weeks their products test exposure among • Blood culture occupational groups at high risk − Food safety, including pasteurization of dairy products Leptospirosis n.a. Exposure • Fourfold rise in antibody titer − Control in animal Doxycycline, penicillin, to urine or by microagglutination test husbandry and cephalosporins environments • Nucleic acid amplification rodent control contaminated test − Management of by the urine exposure among of infected • Culture of blood or urine using special media occupational animals groups at high risk − Management of exposure to environmental sources Q fever n.a. Exposure • Fourfold rise in antibody − Control in animal Tetracyclines to infected titer by immunofluorescence husbandry animals, their assay − Management of products, and • Nucleic acid amplification exposure among environments test occupational • Culture groups at high risk − Food safety Rickettsioses n.a. Arthropod • Fourfold rise in antibody − Prevention of Tetracyclines vectors, vary titer by immunofluorescence exposure to by pathogen assay vectors species • Nucleic acid amplification − Use of test prophylactic • Culture tetracyclines in very high risk groups Note: n.a. = not applicable; NS1 = nonstructural protein 1; IgM = immunoglobulin M. “Supportive” treatment refers to measures to prevent, control, or relieve complications. Integrated Management of Adolescent and Adult adolescents and adults at first-level facilities is based on Febrile Illness current WHO guidelines for the treatment of malaria Management at First-Level Health Facilities (WHO 2010b) and described in detail in chapter 12 of The WHO Integrated Management of Adolescent and this volume (Shretta and others 2017). Adult Illness (IMAI) guidelines for health workers at first- level facilities, specifically health centers and first-level Management at the District Hospital outpatient clinics, provide guidance on the management The WHO IMAI district clinician manual provides of febrile patients (WHO 2009). Management of febrile guidelines for the hospital care of adolescents and adults Febrile Illness in Adolescents and Adults 371 in low-resource areas (WHO 2011a). The manual has Diagnostic tests for many febrile illnesses other than been subjected to few evaluations to date (Rubach and malaria and dengue remain complex, expensive, and others 2015). The district clinician manual assumes limited to a few supranational reference laboratories availability of a minimum level of human resources mostly in HICs. In LMICs, their use has been restricted (medical officer, clinical officer, or senior nurse) and a to a few studies done predominantly in large cities, limited range of essential drugs, equipment, and where conditions differ from those in the vast rural areas laboratory and other investigations at the hospital level. (Acestor and others 2012). Newer diagnostics often lack Emergency management includes the use of antibacteri- standardization of both operation and interpretation als (ceftriaxone) and antimalarials (parenteral artesunate) and have not been sufficiently validated in the range of if sepsis or severe malaria is suspected. settings in which they will be used. Independent evaluations of diagnostic tests are vital Adherence to Guidelines because key information is often missing from the Considerable evidence indicates that the WHO practice details provided by the manufacturers, which may claim recommendations and diagnostic technologies are often high sensitivity and specificity without appropriate not adopted and used in low-resource areas (English and justification (Blacksell, Bell, and others 2007). For exam- others 2014). The reasons include poor dissemination, ple, high sensitivity and specificity of a dengue POCT limited training and monitoring, and limited capacity of has minimal clinical utility for a sample taken seven days human and other resources. after fever onset. A consensus statement—perhaps linked to the Standards for Reporting of Diagnostic Accuracy Diagnostic Approaches (Bossuyt and others 2003) but dedicated to evaluation Laboratory services have been a neglected component of of infectious disease diagnostic tests, especially health services in low-resource areas (Archibald and POCTs—could help improve the current situation. Reller 2001; Petti and others 2006), and patient manage- Because estimates of the sensitivity and specificity of ment has been based predominantly on syndromic diagnostic tests based on evaluation against a known approaches (WHO 2009, 2011a). Notable exceptions have but imperfect gold standard may be imprecise, been assays in support of programs for the diagnosis and Bayesian latent class models may be helpful in their management of HIV, malaria, and tuberculosis (WHO evaluation (Lim and others 2013). Use of filter paper 2010a, 2010b, 2010c). Experts have recommended clinical and POCTs as storage matrices for both serological laboratory services that should be available at various tiers and molecular diagnosis may be a practical way for- of the health service in low-resource areas (WHO 2008). ward in LMICs (Fhogartaigh and others 2015; Smit However, these services remain widely unavailable. and others 2014). Knowledge of the HIV serostatus of a febrile patient Despite these limitations, we would expect that com- is useful. Many rapid HIV antibody tests have high sen- mon pathogens, especially if sharing routes of transmis- sitivity and specificity in established HIV infection sion and risk factors, would result in common occurrence (WHO 1997). Further risk stratification for specific HIV of mixed or concurrent infections (Phommasone and co-infections may be based on the results of a CD4- others 2013). However, the laboratory diagnosis and positive T-lymphocyte count when available.2 management of mixed infections is challenging. Reports Culture of a sufficient volume of blood is useful for of mixed infections often use only serological criteria. the diagnosis of bloodstream infections (Lee and others However, the problems of antibody persistence and 2007). Once available, the results of blood cultures may interspecies cross-reaction raise uncertainty about be used to refine initial antimicrobial management. In whether these results represent true mixed infections, addition, aggregate data provide useful information sequential infections, or cross-reactions. Hence, reports about the local prevalence of pathogens causing blood- of mixed infections should include explicit discussions stream infection as well as patterns of antimicrobial resis- of the likely specificity and sensitivity of the diagnostic tance. Continuously monitored blood culture systems assays used and the likelihood that the observations rep- may shorten the time to detection and improve sensitivity resent true concurrent mixed infections (Phommasone compared with manual blood culture methods, but they and others 2013). are more expensive than manual methods. Special blood The incidence of mixed infections, including culture bottles optimize the recovery of organisms with pathogens such as Salmonella Typhi and Streptococcus particular growth requirements, including some yeasts, pneumoniae, will also be highly influenced by vaccina- mycobacteria, anaerobes, and leptospires. tion coverage. The management of mixed infections has 372 Major Infectious Diseases received little attention. With regard to bacterial mixed Tellingly, household and provider costs are lower infections, a key consideration is that although doxycy- for malaria than for other febrile illnesses among cline is likely to be efficacious for pathogens such as both ambulatory and admitted patients. This finding scrub typhus and leptospirosis, it would not be for other is evident from both indirect cost-of-illness compari- common pathogens such as Salmonella Typhi and sons and from studies comparing the costs of malaria Burkholderia pseudomallei. Combination therapy may be and nonmalarial cases in the same setting (Ansah problematic because of antagonism, such as if bacterio- and others 2013; Batwala and others 2011; Deressa, static and bactericidal antimicrobials are given in paral- Hailemariam, and Ali 2007; Kyaw and others lel, or adverse reactions. 2014; Morel and others 2008; Mustafa and Babiker 2007; Rammaert and others 2011; Yukich and others 2010). One such study compared household costs for COSTS OF FEBRILE ILLNESS AND patients diagnosed clinically as having malaria, with COST-EFFECTIVENESS OF DIAGNOSTICS and without subsequent confirmation of parasitemia, finding that patients incorrectly diagnosed with AND TREATMENTS malaria were more likely to remain symptomatic at This section briefly reviews the evidence on the costs of three weeks’ follow-up and had a higher risk of reat- febrile illness in adolescents and adults and the cost- tendance at a health facility (Hume and others 2008). effectiveness of interventions to improve its manage- Severe malaria admissions have also been found to be ment. It then summarizes three new cost-effectiveness less costly than nonmalarial diseases with similar pre- evaluations of interventions for the management of sentations on the same wards because of lower medi- fever in hospitals, first-level health facilities, and the cation costs and shorter durations of admission community. It concludes with a discussion of how future (Ayieko and English 2007; Lubell and others 2010). economic evaluations could improve on common lim- These differences in cost between malaria and non- itations in the existing literature. malaria illness could be explained by the concentra- tion of malaria-related studies in low-income Sub-Saharan African countries. However, the differ- Costs to Households and Health Care Providers ences could also indicate a genuine trend for two As a dominant reason for seeking medical care, febrile reasons. First, with extensive donor support, malaria illnesses are key drivers of health expenditures and diagnostics and treatments have become increasingly productivity losses. However, no data are available on available across the malaria-endemic world at low the economic impact of common diseases such as bru- cost. Second, the strengthening of distribution mech- cellosis and scrub typhus on either households or anisms in the public and private sectors has helped health care systems. For other diseases such as melioido- ensure access to rapid diagnosis and effective treat- sis, limited data are available from hospital settings, ment. Thus, patients suffering from malaria may fare where the visible burden of confirmed cases is likely to better than those with fever from other causes that be a small fraction of the full burden, because patho- often go undiagnosed, or for which effective treat- gens such as B. pseudomallei and Orientia tsutsuga- ment is unavailable, resulting in a longer duration of mushi are unlikely to be detected outside well-equipped illness and higher expenses for repeated seeking of hospitals. medical care (Reyburn and others 2004). Although methodological differences among cost-of- The high costs of nonmalarial febrile illnesses to illness studies impede comparison of the relative cost households and health care systems suggest consid- burden of different diseases, some general trends can be erable scope for cost-effective investment to reduce detected (figure 14.1 and annex 14B). The largest impact the impact of these illnesses through prevention, is often associated with household productivity losses diagnosis, and treatment. Households might also be rather than direct medical costs, a particular concern for willing to pay for such interventions should they adult patients. For melioidosis, the known burden is become available, as has been found regarding a concentrated among adult agricultural workers; there- typhoid vaccine in Bangladesh and leptospirosis fore, the economic impact on rural households can be prevention in the Philippines (Arbiol and others particularly hard. Among hospitalized patients with 2013; Cook and others 2009). However, there is typhoid fever in India and Indonesia, productivity losses virtually no guidance as to the cost-effectiveness of have been as high as 15 percent to 20 percent of annual diagnostics and treatments for febrile illnesses other income (Bahl and others 2004; Poulos and others 2004). than malaria. Febrile Illness in Adolescents and Adults 373 Figure 14.1 Direct and Indirect Costs of Febrile Illnesses in Adolescents and Adults, by Illness, 2012 12 10 India 8 India Vietnam 6 Cambodia Indonesia 12 4 India Sudan Ethiopia Lao PDR 2 Tanzania Timor-Leste India Ghana Myanmar China 10 India Uganda Vietnam Vietnam Average productivity losses as percentage of annual income 0 200 400 600 800 1,000 1,200 1,400 1,600 8 Brazil 6 Panama 4 Guatemala Bhutan Venezuela, RB 2 Thailand EI Salvador China Philippines Thailand Malaysia Indonesia 0 1,500 3,500 5,500 7,500 9,500 Average annual household income, US$ Chikungunya Dengue Influenza Malaria Typhoid Note: Average household income as reported in the original studies; where not reported, 2012 GDP per capita are used (http://data.worldbank.org/indicator/NY.GDP.PCAP.CD?). Circle size indicates the direct cost of illness, which includes both household and provider direct medical and nonmedical costs per febrile episode relative to household income. The largest circle on the top left, for instance, represents the direct cost of an episode of chikungunya, which in this rural Indian setting is equivalent to 18 percent of annual household income. Most malaria episodes on the bottom left consumed an equivalent of 1 percent to 3 percent of household income, as indicated by the smaller circles. These episodes also resulted in relatively fewer productivity losses, as indicated by their lower vertical positioning. As would be expected, direct costs of illness are regressive, consuming a larger proportion of household income in lower-income settings. However, as indicated by the inverse relationship between household income and productivity losses, these indirect costs also appear to be regressive, with households in lower-income settings losing higher proportions of their income per episode of febrile illness. Cost-Effectiveness Analyses of Diagnostics and artesunate over quinine for severe malaria (Lubell and Treatments of Undifferentiated Febrile Illness others 2009; Lubell and others 2011). Reviews of the cost- Major policy changes by the WHO (WHO 2010b) in the effectiveness of preventive interventions for malaria such diagnosis and treatment of malaria were well supported as insecticide-treated bednets, indoor residual spraying, by evidence of their cost-effectiveness. For example, the and intermittent preventive treatment can be found else- paradigm shift from empiric treatment of fever with anti- where (White and others 2011). malarials toward parasitological confirmation was accom- Only a few economic evaluations have been carried panied by numerous economic evaluations (Ansah and out on diagnoses and treatments of nonmalarial fevers others 2013; Batwala and others 2011; de Oliveira, Castro, in the context of LMICs (detailed in annex 14C). This and Toscano 2010; Lubell, Hopkins, and others 2008; scarcity stems from the dearth of relevant POCTs and Shillcutt and others 2008; White and others 2011; Zurovac treatments that warrant evaluation. Other than dengue and others 2006). Several economic evaluations have also and influenza POCTs, no tests have been deemed suffi- compared treatments for uncomplicated malaria, demon- ciently accurate and appropriate for use in routine care. strating the superiority of artemisinin-based combina- Few advances have been made in recent years for the tion therapies over preexisting monotherapies (Chanda treatment of nonmalarial fevers, with the exception of and others 2007; Wiseman and others 2006) and of new combination therapies for visceral leishmaniasis. 374 Major Infectious Diseases The few available evaluations include cost- POCTs for Sepsis among Patients with Febrile Illnesses effectiveness analyses of the following: in Low-Resource Settings Similarly, distinguishing patients with sepsis from those • Leptospirosis tests compared with empiric treatment with other illnesses remains a challenge. Management deci- with doxycycline (Suputtamongkol and others 2010) sions are based on clinical assessment using algorithms • Influenza POCTs for the detection of influenza such as the WHO’s IMAI guidelines (WHO 2009, 2011a). A (H1N1) in Mexico City (González-Canudas and Efforts to develop and evaluate POCTs for sepsis to others 2011) guide decisions on the use of antimicrobials are under • Eflornithine compared with melarsoprol for the treat- way. To establish the minimum performance character- ment of human African trypanosomiasis (Robays istics of such a test, Penno, Crump, and Baird (2015) and others 2008) varied the characteristics required for cost-effectiveness • Combination treatments over monotherapy for leish- of a hypothetical POCT for sepsis and applied similar maniasis (Meheus and others 2010; Olliaro and oth- methods to the bloodstream infection surveillance ers 2009) analysis of Penno, Baird, and Crump (2015). The exist- • Yet-undeveloped prereferral rectal antimicrobial sup- ing clinical assessment algorithms were compared with positories for severe febrile illness, both alone and POCT-driven management. in combination with an antimalarial suppository Based on a clinical assessment for sepsis with the (Buchanan and others 2010). established sensitivity of 83 percent and specificity of 62 percent, the authors found that a POCT for sepsis with These studies represent the scant evidence on the a specificity of 94 percent and a sensitivity of 83 percent cost-effectiveness of diagnostics and treatments for non- was cost-effective, resulting in equivalence with clinical malarial causes of fever. We next extend this limited body assessment with regard to survival but costing US$1.14 of evidence, summarizing three evaluations of diagnostic less per life saved. A POCT with sensitivity and specificity tools for the management of febrile episodes and sepsis of 100 percent, slightly superior to those of the best malaria in the hospital, first-level, and community settings. RDTs, was both cheaper and more effective than clinical assessment. Overall, this work helps establish performance Surveillance of Bloodstream Infections for Sepsis targets for POCTs for sepsis in low-resource areas. Management in Low-Resource Settings A leading cause of mortality among febrile patients in Diagnosis and Treatment Strategies for the LMICs is bacterial sepsis, but blood culture services are Management of Febrile Patients in the Community not widely available. Consequently, empiric antimicro- Lubell and others (2016) considered the implications bial management of suspected bloodstream infection is of using pathogen-specific tests for dengue and scrub based on generic guidelines that are rarely informed typhus, as compared with testing for a biomarker of host by local data on etiology and antimicrobial resistance inflammation—C-reactive protein (CRP)—to inform man- patterns. agement of fever in the community. Using data on causes of To evaluate the cost-effectiveness of surveillance of fever in outpatients in rural Lao PDR (Mayxay and others bloodstream infections to inform empiric management of 2013), the proportion of patients that would be correctly suspected sepsis in low-resource areas, Penno, Baird, and treated with an antimicrobial under each of the approaches Crump (2015) compared costs and outcomes of generic was estimated as was the cost-effectiveness of the tests. antimicrobial management with management informed After accounting for the accuracy of the tests, the by local data on etiology and patterns of antimicrobial following assumptions were made regarding how their resistance across all of Africa. Applying a decision-tree results would be used to inform treatment practices: model to a hypothetical population of febrile patients pre- senting at first-level hospitals in Africa, the authors found • Patients with positive dengue test results are not pre- that the evidence-based regimen saved an additional scribed an antimicrobial. 534 lives per 100,000 patients with fever at an increase in • Patients with positive scrub typhus test results are cost of US$25.35 per life saved, which corresponded to an prescribed an effective antimicrobial, for example, incremental cost-effectiveness ratio of US$4,749.3 tetracycline or macrolide. Although this number compares favorably to standard • Where a pathogen-specific test is negative, antimicro- cost-effectiveness thresholds, it should ultimately be com- bials are prescribed at random to 38 percent of patients, pared with other relevant policy alternatives to confirm resembling current practice, with the choice of anti- whether routine surveillance for bloodstream infections is biotic resembling current practice as observed in a cost-effective strategy in the African context. Mayxay and others (2013). Febrile Illness in Adolescents and Adults 375 • Patients with CRP levels below the threshold of 20 uncertainty surrounding the accuracy of the tests milligrams per liter do not receive an antimicrobial; (figure 14.2, panel a). those above the threshold do, with the choice of anti- The incremental costs and DALYs averted from the biotic resembling current practice. output of 500 simulations and the corresponding cost-effectiveness acceptability curves are shown in The approaches were first evaluated for their ability figure 14.3, illustrating that both the scrub typhus and to classify patients as requiring treatment. Then their the CRP tests are likely to be cost-effective, despite the ability to guide the choice of antimicrobials, with regard considerable uncertainty surrounding many model to pathogen susceptibility to the different drugs, was parameters. incorporated. Patients with positive scrub typhus tests The analysis suggests that pathogen-specific tests are therefore assumed to receive doxycycline; for other can improve the management of nonmalarial fevers, treated patients, the choice of antimicrobial is made at but their utility and cost-effectiveness are highly sensi- random based on the frequency of their (often inappro- tive to contextual factors such as heterogeneity of fever priate) use in current practice. etiologies and preexisting prescription patterns. Testing The analysis suggests that use of either pathogen- for pathogens for which there are no immediate treat- specific test offered modest improvements over cur- ment implications, such as dengue, might not offer a rent practice in the ability to classify patients as direct health benefit over current practice. However, requiring an antimicrobial (figure 14.2). The dengue this approach does not account for factors such as RDT implied a reduction in antimicrobials prescribed patient reassurance in having a confirmed diagnosis, for viral infections but a larger proportion of bacterial raising of awareness of possible danger signs, outbreak infections going untreated, while the reverse was true detection, or initiation of measures for vector control. for the scrub typhus test. Use of the CRP test implied Testing for bacterial pathogens can guide the use of both reductions in the use of antimicrobials in viral appropriate antimicrobials; therefore, testing could be infections and fewer bacterial infections going a cost-effective approach, but cost-effectiveness var- untreated. These advantages were consistent despite ies widely depending on local etiology. Testing for variation of all model parameters, including the inci- biomarkers of inflammation could offer an approach to dence of infections and baseline antimicrobial pre- targeting antimicrobials in rural settings that is cost- scription practices, and when accounting for the effective and robust to heterogeneity in causes of fevers Figure 14.2 Antimicrobial Targeting Using Lao PDR Data across a Range of Simulated Incidences and Test Accuracies a. Antimicrobials prescribed in viral b. Antimicrobials prescribed and bacterial infections correctly 100 100 80 80 Percentage of antibiotics prescribed correctly 60 60 Percent 40 40 20 20 0 0 Current Dengue ST RDT CRP RDT Current Dengue ST RDT CRP RDT practice RDT practice RDT Bacterial-untreated Bacterial-treated Viral-treated Viral-untreated Source: Lubell and others 2016; Mayxay and others 2013. Note: CRP = C-reactive protein; RDT = rapid diagnostic test; ST = scrub typhus. In panel b, the shaded boxes represent the interquartile ranges of simulated outputs, with the median marked by the mid-way horizontal line. The whiskers represent the simulation outputs closest to 1.5 times the interquartile range, with the remaining outliers shown as independent circles. 376 Major Infectious Diseases Figure 14.3 Cost-Effectiveness Plane and Cost-Effectiveness Probability Curves for the Three Strategies When Compared with a Baseline of Current Practice a. Dengue test 5 100 Probability that treatment is cost-effective (percent) Incremental cost (US$) 4 75 3 2 50 1 25 0 –1 0 –0.2 0 0.2 0.4 0 500 1,000 1,500 2,000 b. ST test 5 100 Probability that treatment is cost-effective (percent) Incremental cost (US$) 4 75 3 2 50 1 25 0 –1 0 –0.2 0 0.2 0.4 0 500 1,000 1,500 2,000 c. CRP test 5 100 Probability that treatment is cost-effective (percent) Incremental cost (US$) 4 75 3 2 50 1 25 0 –1 0 –0.2 0 0.2 0.4 0 500 1,000 1,500 2,000 DALYs averted Willingess-to-pay threshold (US$) Note: CRP = C-reactive protein; DALY = disability-adjusted life year; ST = scrub typhus. The blue solid lines indicate a willingness-to-pay threshold of US$1,400, approximating Lao PDR’s 2012 GDP per capita; the green dashed line is a much more conservative willingness-to-pay threshold of US$150. The dengue test in most instances was associated with worse health outcomes and higher costs. The scrub typhus test averted on average 0.07 DALYs as compared with current practice, and the CRP test averted on average 0.05 DALYs as compared with current practice, with median incremental cost-effectiveness ratios of US$59 and US$110, respectively. When accounting for parameter uncertainty, the scrub typhus and CRP tests are approximately 90 percent and 80 percent likely, respectively, to be cost-effective at a willingness-to-pay threshold of US$1,400. and host factors typical of the rural tropics, but realiz- range of treatable or preventable pathogens and their ing the full potential gains of this approach requires antimicrobial susceptibility patterns; uses standardized locally appropriate empirical treatment guidelines. case definitions and diagnostic assays; and uses a control group to estimate attributable fractions. The findings FUTURE RESEARCH NEEDS of such research would better apprise health care work- ers about the prior probability of specific infections Fever Etiology Research in their area and would contribute data relevant to To provide a clearer picture of the relative importance of burden-of-disease assessments for infections that cur- relevant causes of nonmalarial febrile illness, fever etiol- rently lack robust estimates of illness and death. ogy research is needed that takes into account geo- Researchers should also consider structuring graphic, seasonal, and ethnic diversity; includes a broad burden-of-disease estimates for febrile illness without Febrile Illness in Adolescents and Adults 377 localizing features similarly to estimates for diarrhea and • Extension of the analyses, where appropriate, to include pneumonia. Data from fever etiology research could be the implications for ongoing pathogen transmission applied to the syndrome-wide envelope of DALYs and • Accounting for the intervention’s impact on develop- deaths for febrile illness to develop pathogen-specific ment of antimicrobial resistance. estimates. Such an approach would provide the basis for estimating the effects of interventions, including robust Inclusion of the Intervention’s Implications for All cost-effectiveness analyses. Febrile Patients An important limitation in many evaluations of malaria diagnostics was the use of malaria-centered outcomes, Diagnostic and Clinical Management Studies such as cost per malaria case diagnosed or treated. Such POCTs for febrile illnesses can be classified into outcomes failed to account for the impact on other pathogen-specific tests and biomarker assays that might nonmalarial diseases—often affecting the majority of detect states such as the systemic inflammatory response patients. What is needed is a clear indication of how the associated with sepsis. Malaria RDTs demonstrate that intervention affects the management of febrile patients a pathogen-specific diagnostic can find an important as a whole. place in management algorithms for fever in low- One of the reasons malaria POCTs were found to be resource areas (WHO 2010b). However, current pathogen- cost-effective was this methodological shift toward inclu- specific POCTs for most other causes of fever do not sion of their potential benefit for other causes of illness, approach the performance of the best malaria RDTs such as the provision of antimicrobials to treatable bacte- (WHO 2011b). Research priorities for pathogen-specific rial infections. In contrast, there was some apprehension POCTs should be based on burden-of-disease data and that the benefits of malaria POCTs were overestimated should focus on infections for which potentially lifesav- in cost-effectiveness analyses, particularly in high- ing treatments are available. Biomarker assays that iden- transmission areas, where malaria POCTs could be detri- tify a group of patients with severe disease or those likely mental to the management of nonmalarial diseases to develop severe disease could have a role in triage and because older children and adults could be carrying in the targeting of antimicrobial agents. However, such malaria parasites to which they are clinically immune tests must be at least as sensitive as clinical evaluation to while suffering from a potentially severe nonmalarial ill- avoid excess deaths and at least as specific as clinical eval- ness (Bisoffi and others 2011). These considerations will uation to avoid antimicrobial overuse and excessive costs. be important in other pathogen-specific tests already on In addition, clinical management studies are urgently the market for diseases such as dengue and scrub typhus. needed to provide a foundation for the best empiric treatment strategies for patients with severe nonmalarial Inclusion of Factors Affecting the Intervention’s febrile illness (Crump, Gove, and Parry 2011). Studies Performance that support decisions at first-level health care facilities The performance and impact of the tests in routine about when to use and when to withhold antimicrobial settings can be affected by intermediate factors such as therapy are essential to optimizing patient outcomes interpretation of and adherence to test results, diagnostic while preserving scarce health care resources and con- test stockouts, and compromised quality due to inade- taining antimicrobial resistance. quate storage and handling. For malaria POCTs, the pre- scribing of antimalarials to patients who had tested negative for malaria was of particular concern, raising the Necessary Elements for Cost-Effectiveness Analyses question of whether POCTs were in fact cost-effective in Five elements need to be strengthened for the design of routine care as opposed to in trial settings and model- economic evaluations of diagnostics and treatments for based evaluations. febrile illnesses: An economic evaluation that accounted for compro- mised adherence demonstrated that although the • Inclusion of the intervention’s implications for all cost-effectiveness of POCTs was diminished, a substan- febrile patients rather than for a subset with specific tial degree of nonadherence could be absorbed while use pathogens of interest of the tests remained cost-effective (Lubell, Reyburn, and • Inclusion of factors affecting the intervention’s per- others 2008). Models that incorporate such factors can formance in real-life settings not only provide a more realistic assessment of the inter- • Framing of the evaluations within specified etiolog- vention’s cost-effectiveness in routine care, but also assist ical contexts and exploration of the implications of in guiding resources for training and other supportive heterogeneity interventions. 378 Major Infectious Diseases Framing of Evaluations within Specific Etiological CONCLUSIONS Contexts The diagnostic value of a test is not an inherent charac- Declines in malaria in many areas and growing recogni- teristic, but one that is strongly influenced by the prior tion of the problem of malaria overdiagnosis have probability of the condition it is designed to detect. This focused new attention on the problem of management point implies that the suitability of many pathogen- of fevers in low-resource areas. We recommend the specific tests will vary by etiological setting, both season- following in descending order of priority: ally and spatially. This circumstance is a challenge to effective policy making but also an opportunity to vastly 1. Develop accurate pathogen-specific and bacterial improve the efficient allocation of scarce resources. disease biomarker POCTs for causes of fever other than malaria. Extension of Analyses to Include Implications for 2. Improve the identification and management of Pathogen Transmission patients with bacterial and fungal sepsis and those One important concern is the degree to which the with tetracycline-responsive infections to avert mor- analysis should account for the impact of interven- bidity and mortality from febrile illness. tions on disease transmission, as opposed to a limited 3. Undertake comprehensive, standardized, and coordi- focus on direct costs and benefits to the patient. nated fever etiology research in low-resource areas to Although the latter focus can avoid many of the uncer- identify priorities for improvements in management, tainties that often pervade dynamic transmission such as selection of empiric antimicrobials, and for models, the more-conservative results of such analyses control of causes of fever other than malaria. could unduly penalize an intervention that would be 4. Gather more cost and outcome data and improve an efficient use of resources if the indirect health ben- approaches to cost-effectiveness analyses related to efits associated with reduced transmission were to be fever to strengthen resource-stratified approaches considered. to adoption and integration of interventions in the emerging field of nonmalarial febrile illness. Inclusion of Intervention’s Impact on Antimicrobial 5. Support surveillance for bloodstream infections and Resistance antimicrobial resistance in low-resource areas. Perhaps most challenging, the evaluations of diagnos- 6. Explore burden-of-disease structures that capture tics and treatments for fevers could be improved by febrile illnesses as a group to estimate the effect size including the societal costs and consequences associated of interventions for fevers. with antimicrobial resistance (Loubiere and Moatti 2010). The methodological challenges to this approach ANNEXES are considerable and likely explain this glaring omission (Coast, Smith, and Millar 1996). In the absence of such The annexes to this chapter are as follows. They are avail- measurements, presumptive treatment will almost able at http://www.dcp-3.org/infectiousdiseases. always dominate as long as the purchase cost of the drugs is less than that of the diagnostics employed to • Annex 14A. Etiology of Severe Febrile Illness: Studies determine their use, even if this analysis fails to reflect of Adolescents and Adults, by Region, 1980–2013 the broader societal costs and consequences of such a • Annex 14B. Studies of Febrile Illness Costs to strategy (Girosi and others 2006). Households and Health Care Providers Few attempts have been made to define the societal • Annex 14C. Studies Reporting the Cost-Effectiveness cost per antimicrobial consumed. One such study, in a of Diagnostics and Treatments for Nonmalarial Fevers high-income hospital setting, estimated that each defined daily dose of antimicrobial was associated with a cost of NOTES US$7–17 because of its contribution to antimicrobial resistance in that hospital (Kaier and Frank 2010). World Bank Income Classifications as of July 2014 are as fol- Despite the difficulties in deriving these values, the lows, based on estimates of gross national income (GNI) per capita for 2013: inclusion of even highly conservative estimates for these costs in the evaluation of malaria POCTs has been • Low-income countries (LICs) = US$1,045 or less shown to dramatically affect their conclusions (Lubell, • Middle-income countries (MICs) are subdivided: Reyburn, and others 2008). If not formally incorporated (a) lower-middle-income = US$1,046 to US$4,125 in the analysis, the implications of their exclusion should (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 at the very least be addressed. • High-income countries (HICs) = US$12,746 or more. Febrile Illness in Adolescents and Adults 379 1. The WHO estimates the global burden of disease through Bassat, Q., J. Ordi, J. Vila, M. R. Ismail, C. Carrilho, and others. its Global Health Estimates database, available at http:// 2013. “Development of a Post-Mortem Procedure to Reduce www.who.int/healthinfo/global_burden_disease/en/. The the Uncertainty Regarding Causes of Death in Developing Institute for Health Metrics and Evaluation (IHME) pro- Countries.” The Lancet Global Health 1 (3): e125–26. duces the Global Burden of Diseases, Injuries, and Risk Batwala, V., P. Magnussen, K. S. Hansen, and F. Nuwaha. Factors Study. However, some of the IHME’s findings 2011. “Cost-Effectiveness of Malaria Microscopy and Rapid have been controversial; for example, it estimated in 2012 Diagnostic Tests versus Presumptive Diagnosis: Implications that 1.24 million people had died from malaria worldwide for Malaria Control in Uganda.” Malaria Journal 10 (1): 372. in 2010, double the WHO estimate (The Lancet 2012). Bisoffi, Z., S. B. Sirima, F. Meheus, C. Lodesani, F. Gobbi, 2. CD4 (cluster of differentiation 4) refers to a T-lymphocyte and others. 2011. “Strict Adherence to Malaria Rapid surface antigen; CD4-positive T-lymphocyte depletion in Test Results Might Lead to a Neglect of Other Dangerous HIV infection is associated with HIV comorbidities. Diseases: A Cost Benefit Analysis from Burkina Faso.” 3. All costs are in 2012 U.S. dollars. Malaria Journal 10 (1): 226. Blacksell, S. D., D. Bell, J. Kelley, M. P. Mammen, R. V. Gibbons, and others. 2007. “Prospective Study to Determine Accuracy REFERENCES of Rapid Serological Assays for Diagnosis of Acute Dengue Abba, K., J. J. Deeks, P. Olliaro, C. M. Naing, S. M. Jackson, Virus Infection in Laos.” Clinical and Vaccine Immunology and others. 2011. “Rapid Diagnostic Tests for Diagnosing 14 (11): 1458–64. Uncomplicated P. falciparum Malaria in Endemic Blacksell, S. D., N. P. Sharma, W. Phumratanaprapin, Countries.” Cochrane Database of Systematic Reviews 7 (7): K. Jenjaroen, S. J. Peacock, and others. 2007. “Serological and CD008122. doi:10.1002/14651858.CD008122.pub2. Blood Culture Investigations of Nepalese Fever Patients.” Acestor, N., R. Cooksey, P. N. Newton, D. Ménard, P. J. Guerin, Transactions of the Royal Society of Tropical Medicine and and others. 2012. “Mapping the Aetiology of Non-Malarial Hygiene 101 (7): 686–90. Febrile Illness in Southeast Asia through a Systematic Bossuyt, P. M., J. B. Reitsma, D. E. Bruns, C. A. Gatsonis, Review—Terra Incognita Impairing Treatment Policies.” P. P. Glasziou, and others. 2003. “Towards Complete and PLoS One 7 (9): e44269. Accurate Reporting of Studies of Diagnostic Accuracy: The Afifi, S., K. Earhart, M. A. Azab, F. G. Youssef, H. El Sakka, STARD Initiative.” BMJ 326 (7379): 41–44. and others. 2005. “Hospital-Based Surveillance for Acute Breiman, R. F., L. Cosmas, H. Njuguna, A. Audi, B. Olack, and Febrile Illness in Egypt: A Focus on Community-Acquired others. 2012. “Population-Based Incidence of Typhoid Fever Bloodstream Infections.” American Journal of Tropical in an Urban Informal Settlement and a Rural Area in Kenya: Medicine and Hygiene 73 (2): 392–99. Implications for Typhoid Vaccine Use in Africa.” PLoS One Ansah, E. K., M. Epokor, C. J. Whitty, S. Yeung, and K. S. Hansen. 7: e29119. 2013. “Cost-Effectiveness Analysis of Introducing RDTs Buchanan, J., B. Mihaylova, A. Gray, and N. White. 2010. “Cost- for Malaria Diagnosis as Compared to Microscopy and Effectiveness of Pre-Referral Antimalarial, Antibacterial, Presumptive Diagnosis in Central and Peripheral Public and Combined Rectal Formulations for Severe Febrile Health Facilities in Ghana.” American Journal of Tropical Illness.” PLoS One 5 (12): e14446. Medicine and Hygiene 89 (4): 724–36. Butler, D. 2010. “Verbal Autopsy Methods Questioned.” Nature Arbiol, J., M. Borja, M. Yabe, H. Nomura, N. Gloriani, and 467 (7319): 1015. others. 2013. “Valuing Human Leptospirosis Prevention Caudron, J. M., N. Ford, M. Henkens, C. Macé, R. Kiddle- Using the Opportunity Cost of Labor.” International Monroe, and others. 2008. “Substandard Medicines in Journal of Environmental Research and Public Health 10 (5): Resource-Poor Settings: A Problem That Can No Longer 1845–60. Be Ignored.” Tropical Medicine and International Health Archibald, L. K., M. O. den Dulk, K. J. Pallangyo, and L. B. Reller. 13 (8): 1062–72. 1998. “Fatal Mycobacterium tuberculosis Bloodstream Chanda, P., F. Masiye, B. M. Chitah, N. Sipilanyambe, Infections in Febrile Hospitalized Adults in Dar es Salaam, M. Hawela, and others. 2007. “A Cost-Effectiveness Tanzania.” Clinical Infectious Diseases 26 (2): 290–96. Analysis of Artemether Lumefantrine for Treatment of Archibald, L. K., and L. B. Reller. 2001. “Clinical Microbiology Uncomplicated Malaria in Zambia.” Malaria Journal 6: 21. in Developing Countries.” Emerging Infectious Diseases 7 Cheng, A. C., and B. J. Currie. 2005. “Melioidosis: Epidemiology, (2): 302–5. Pathophysiology, and Management.” Journal of Clinical Ayieko, P., and M. English. 2007. “Case Management of Microbiology 18 (2): 383–416. Childhood Pneumonia in Developing Countries.” Pediatric Coast, J., R. D. Smith, and M. R. Millar. 1996. “Superbugs: Infectious Disease Journal 26 (5): 432–40. Should Antimicrobial Resistance Be Included as a Cost in Bahl, R., A. Sinha, C. Poulos, D. Whittington, S. Sazawal, Economic Evaluation?” Health Economics 5 (3): 217–26. and others. 2004. “Costs of Illness due to Typhoid Fever Cook, J., D. Sur, J. Clemens, and D. Whittington. 2009. in an Indian Urban Slum Community: Implications for “Evaluating Investments in Typhoid Vaccines in Two Vaccination Policy.” Journal of Health, Population and Slums in Kolkata, India.” Journal of Health, Population and Nutrition 22 (3): 304–10. Nutrition 27 (6): 711–24. 380 Major Infectious Diseases Crump, J. A. 2014. “Time for a Comprehensive Approach to the Diagnostic Tests.” Journal of Clinical Microbiology 53 (5): Syndrome of Fever in the Tropics.” Transactions of the Royal 1758–60. Society of Tropical Medicine and Hygiene 108 (2): 61–62. Figueiredo, M. A., L. C. Rodrigues, M. L. Barreto, J. W. Lima, Crump, J. A., S. Gove, and C. M. Parry. 2011. “Management M. C. Costa, and others. 2010. “Allergies and Diabetes as Risk of Adolescents and Adults with Febrile Illness in Resource Factors for Dengue Hemorrhagic Fever: Results of a Case Limited Areas.” BMJ 343: d4847. Control Study.” PLoS Neglected Tropical Diseases 4 (6): e699. Crump, J. A., A. B. Morrissey, W. L. Nicholson, R. F. Massung, Girosi, F., S. S. Olmsted, E. Keeler, D. C. Burgess, Y. W. Lim, R. A. Stoddard, and others. 2013. “Etiology of Severe Non- and others. 2006. “Developing and Interpreting Models to Malaria Febrile Illness in Northern Tanzania: A Prospective Improve Diagnostics in Developing Countries.” Nature 444 Cohort Study.” PLoS Neglected Tropical Diseases 7 (7): e2324. (Suppl 1): 3–8. Dean, A. S., L. Crump, H. Greter, E. Schelling, and J. Zinsstag. González-Canudas, J., J. M. Iglesias-Chiesa, Y. Romero-Antonio, 2012. “Global Burden of Human Brucellosis: A Systematic C. Chávez-Cortes, J. G. Gay-Molina, and others. 2011. Review of Disease Frequency.” PLoS Neglected Tropical “Cost-Effectiveness in the Detection of Influenza H1N1: Diseases 6 (10): e1865. Clinical Data versus Rapid Tests.” Revista Panamericana de Deen, J., L. von Seidlein, F. Andersen, N. Elle, N. J. White, and Salud Pública 29 (1): 1–8. others. 2012. “Community-Acquired Bacterial Bloodstream Gravett, C. A., M. G. Gravett, E. T. Martin, J. D. Bernson, Infections in Developing Countries in South and Southeast S. Khan, and others. 2012. “Serious and Life-Threatening Asia: A Systematic Review.” The Lancet Infectious Diseases Pregnancy-Related Infections: Opportunities to Reduce the 12 (6): 480–87. Global Burden.” PLoS Medicine 9 (10): e1001324. de Oliveira, M. R., G. A. de Castro, and C. M. Toscano. 2010. Hume, J. C., G. Barnish, T. Mangal, L. Armázio, E. Streat, and “Cost Effectiveness of OptiMal® Rapid Diagnostic Test for others. 2008. “Household Cost of Malaria Overdiagnosis in Malaria in Remote Areas of the Amazon Region, Brazil.” Rural Mozambique.” Malaria Journal 7: 33. Malaria Journal 9 (1): 277. Huson, M. A., S. M. Stolp, T. van der Poll, and M. P. Grobusch. Deressa, W., D. Hailemariam, and A. Ali. 2007. “Economic 2014. “Community-Acquired Bacterial Bloodstream Costs of Epidemic Malaria to Households in Rural Infections in HIV-Infected Patients: A Systematic Review.” Ethiopia.” Tropical Medicine and International Health 12 Clinical Infectious Diseases 58 (1): 79–92. (10): 1148–56. Jennings, G. J., R. A. Hajjeh, F. Y. Girgis, M. A. Fadeel, M. A. Dittrich, S., J. Castonguay-Vanier, C. E. Moore, N. Thongyoo, Maksoud, and others. 2007. “Brucellosis as a Cause of Acute P. N. Newton, and others. 2014. “Loop-Mediated Isothermal Febrile Illness in Egypt.” Transactions of the Royal Society of Amplification for Rickettsia typhi (the Causal Agent of Tropical Medicine and Hygiene 101 (7): 707–13. Murine Typhus): Problems with Diagnosis at the Limit of Kaier, K., and U. Frank. 2010. “Measuring the Externality Detection.” Journal of Clinical Microbiology 52 (3): 832–28. of Antibacterial Use from Promoting Antimicrobial English, M., D. Gathara, S. Mwinga, P. Ayieko, C. Opondo, and Resistance.” Pharmacoeconomics 28 (12): 1123–28. others. 2014. “Adoption of Recommended Practices and Kapur, A., and A. D. Harries. 2013. “The Double Burden of Basic Technologies in a Low-Income Setting.” Archives of Diabetes and Tuberculosis: Public Health Implications.” Disease in Childhood 99 (5): 452–56. Diabetes Research and Clinical Practice 101 (1): 10–19. Esper, A. M., M. Moss, and G. S. Martin. 2009. “The Effect of Kasper, M. R., T. F. Wierzba, L. Sovann, P. J. Blair, and S. D. Diabetes Mellitus on Organ Dysfunction with Sepsis: An Putnam. 2010. “Evaluation of an Influenza-Like Illness Case Epidemiological Study.” Critical Care 13 (1): R18. Definition in the Diagnosis of Influenza among Patients Faurholt-Jepsen, D., N. Range, G. PrayGod, K. Jeremiah, with Acute Febrile Illness in Cambodia.” BMC Infectious M. Faurholt-Jepsen, and others. 2013. “Diabetes Is a Strong Diseases 10 (1): 320. Predictor of Mortality during Tuberculosis Treatment: A Knapp, S. 2013. “Diabetes and Infection: Is There a Link? A Prospective Cohort Study among Tuberculosis Patients from Mini-Review.” Gerontology 59: 99–104. Mwanza, Tanzania.” Tropical Medicine and International Kourtis, A. P., J. S. Read, and D. J. Jamieson. 2014. “Pregnancy Health 18 (7): 822–29. and Infection.” New England Journal of Medicine 370 (23): Feasey, N. A., G. Dougan, R. A. Kingsley, R. S. Heyderman, and 2211–18. M. A. Gordon. 2012. “Invasive Non-Typhoidal Salmonella Kyabayinze, D. J., J. K. Tibenderana, G. W. Odong, J. B. Disease: An Emerging and Neglected Tropical Disease in Rwakimari, and H. Counihan. 2008. “Operational Accuracy Africa.” The Lancet 379 (9835): 2489–99. and Comparative Persistent Antigenicity of HRP2 Rapid Feikin, D. R., B. Olack, G. M. Bigogo, A. Audi, L. Cosmas, Diagnostic Tests for Plasmodium falciparum Malaria in a and others. 2011. “The Burden of Common Infectious Hyperendemic Region of Uganda.” Malaria Journal 7: 221. Syndromes at the Clinic and Household Level from Kyaw, S. S., T. Drake, R. Ruangveerayuth, W. Chierakul, N. J. Population-Based Surveillance in Rural and Urban Kenya.” White, and others. 2014. “Cost of Treating Inpatient fal- PLoS One 6 (1): e16085. ciparum Malaria on the Thai-Myanmar Border.” Malaria Fhogartaigh, C. N., D. A. Dance, V. Davonga, P. Tannd, Journal 13: 416. R. Phetsouvanha, and others. 2015. “A Novel Technique Lancet. 2012. “New Estimates of Malaria Deaths: Concern and for Detecting Antibiotic-Resistant Typhoid from Rapid Opportunity.” The Lancet 379 (9814): 385. Febrile Illness in Adolescents and Adults 381 Lee, A., S. Mirrett, L. B. Reller, and M. C. Weinstein. 2007. McGready, R., E. A. Ashley, V. Wuthiekanun, S. O. Tan, “Detection of Bloodstream Infections in Adults: How M. Pimanpanarak, and others. 2010. “Arthropod Borne Many Blood Cultures Are Needed?” Journal of Clinical Disease: The Leading Cause of Fever in Pregnancy on the Microbiology 45 (11): 3546–48. Thai-Burmese Border.” PLoS Neglected Tropical Diseases Lim, C., P. Wannapinij, L. White, N. P. Day, B. S. Cooper, and 4 (11): e888. others. 2013. “Using a Web-Based Application to Define the Meheus, F., M. Balasegaram, P. Olliaro, S. Sundar, S. Rijal, and Accuracy of Diagnostic Tests when the Gold Standard Is others. 2010. “Cost-Effectiveness Analysis of Combination Imperfect.” PLoS One 8 (11): e79489. Therapies for Visceral Leishmaniasis in the Indian Lofgren, S. M., E. J. Kirsch, V. P. Maro, A. B. Morrissey, L. J. Msuya, Subcontinent.” PLoS Neglected Tropical Diseases 4 (9): e818. and others. 2012. “Histoplasmosis among Hospitalized Morel, C. M., N. D. Thang, N. X. Xa, L. X. Hung, L. K. Thuan, Febrile Patients in Northern Tanzania.” Transactions of the and others. 2008. “The Economic Burden of Malaria on Royal Society of Tropical Medicine and Hygiene 106 (8): 504–7. the Household in South-Central Vietnam.” Malaria Journal Loubiere, S., and J. P. Moatti. 2010. “Economic Evaluation 7: 166. of Point-of-Care Diagnostic Technologies for Infectious Mori, M., R. Ravinetto, and J. Jacobs. 2011. “Quality of Medical Diseases.” Clinical Microbiology and Infection 16 (8): 1070–76. Devices and In Vitro Diagnostics in Resource-Limited Louie, J. K., M. Acosta, D. J. Jamieson, M. A. Honein, and Settings.” Tropical Medicine and International Health California Pandemic (H1N1) Working Group. 2010. “Severe 16 (11): 1439–49. 2009 H1N1 Influenza in Pregnant and Postpartum Women in Mueller, T. C., S. Siv, N. Khim, S. Kim, E. Fleischmann, California.” New England Journal of Medicine 362 (1): 27–35. and others. 2014. “Acute Undifferentiated Febrile Illness Lubell, Y., T. Althaus, S. D. Blacksell, D. H. Paris, M. Mayxay, and in Rural Cambodia: A 3-Year Prospective Observational others. 2016. “Modelling the Impact and Cost-Effectiveness Study.” PLoS One 9 (4): e95868. of Biomarker Tests as Compared with Pathogen-Specific Muller, L. M., K. J. Gorter, E. Hak, W. L. Goudzwaard, Diagnostics in the Management of Undifferentiated Fever F. G. Schellevis, and others. 2005. “Increased Risk of in Remote Tropical Settings.” PLoS One 11 (3): e0152420. Common Infections in Patients with Type 1 and Type 2 Lubell, Y., H. Hopkins, C. J. Whitty, S. G. Staedke, and A. Mills. Diabetes Mellitus.” Clinical Infectious Diseases 41 (3): 281–88. 2008. “An Interactive Model for the Assessment of the Murdoch, D. R., C. W. Woods, M. D. Zimmerman, P. M. Economic Costs and Benefits of Different Rapid Diagnostic Dull, R. H. Belbase, and others. 2004. “The Etiology of Tests for Malaria.” Malaria Journal 7: 21. Febrile Illness in Adults Presenting to Patan Hospital in Lubell, Y., A. J. Mills, C. J. Whitty, and S. G. Staedke. 2010. “An Kathmandu, Nepal.” American Journal of Tropical Medicine Economic Evaluation of Home Management of Malaria in and Hygiene 70 (6): 670–75. Uganda: An Interactive Markov Model.” PLoS One 5 (8): Murray, C. J., K. F. Ortblad, C. Guinovart, S. S. Lim, e12439. T. M. Wolock, and others. 2014. “Global, Regional, and Lubell, Y., H. Reyburn, H. Mbakilwa, R. Mwangi, S. Chonya, National Incidence and Mortality for HIV, Tuberculosis, and others. 2008. “The Impact of Response to the Results and Malaria during 1990–2013: A Systematic Analysis for of Diagnostic Tests for Malaria: Cost-Benefit Analysis.” the Global Burden of Disease Study 2013.” The Lancet 384 BMJ 336 (7637): 202–25. (9947): 1005–70. Lubell, Y., A. Riewpaiboon, A. M. Dondorp, L. von Seidlein, Mustafa, M. H., and M. A. Babiker. 2007. “Economic Cost of O. A. Mokuolu, and others. 2011. “Cost-Effectiveness of Malaria on Households during a Transmission Season in Parenteral Artesunate for Treating Children with Severe Khartoum State, Sudan.” Eastern Mediterranean Health Malaria in Sub-Saharan Africa.” Bulletin of the World Health Journal 13 (6): 1298–307. Organization 89 (7): 504–12. Nadjm, B., B. Amos, G. Mtove, J. Ostermann, S. Chonya, Lubell, Y., S. Yeung, A. M. Dondorp, N. P. Day, F. Nosten, and others. 2010. “WHO Guidelines for Antimicrobial and others. 2009. “Cost-Effectiveness of Artesunate for Treatment in Children Admitted to Hospital in an Area of the Treatment of Severe Malaria.” Tropical Medicine and Intense Plasmodium falciparum Transmission: Prospective International Health 14 (3): 332–27. Study.” BMJ 340: c1350. doi:10.1136/bmj.c1350. Machado, C. R., E. S. Machado, R. D. Rohloff, M. Azevedo, D. P. Newton, P. N., F. M. Fernandez, M. D. Green, J. Primo-Carpenter, Campos, and others. 2013. “Is Pregnancy Associated with and N. J. White. 2009. “Counterfeit and Substandard Anti- Severe Dengue? A Review of Data from the Rio de Janeiro Infectives in Developing Countries.” In Antimicrobial Surveillance Information System.” PLoS Neglected Tropical Resistance in Developing Countries, edited by A. de J. Sosa, D. Diseases 7 (5): e2217. K. Byarugaba, C. F. Amábile-Cuevas, P.-R. Hsueh, S. Kariuki, Mallewa, M., A. R. Fooks, D. Banda, P. Chikungwa, and I. N. Okeke, 413–46. New York: Springer. L. Mankhambo, and others. 2007. “Rabies Encephalitis in Olliaro, P., S. Darley, R. Laxminarayan, and S. Sundar. Malaria-Endemic Area, Malawi, Africa.” Emerging Infectious 2009. “Cost-Effectiveness Projections of Single and Diseases 13(1): 136–39. Combination Therapies for Visceral Leishmaniasis in Mayxay, M., J. Castonguay-Vanier, V. Chansamouth, Bihar, India.” Tropical Medicine and International Health A. Dubot-Pérès, D. H. Paris, and others. 2013. “Causes 14 (8): 918–25. of Non-Malarial Fever in Laos: A Prospective Study.” Park, S.-W., C. S. Lee, C. K. Lee, Y. G. Kwak, C. Moon, and The Lancet Global Health 1 (1): e46–54. others. 2011. “Severity Predictors in Eschar-Positive Scrub 382 Major Infectious Diseases Typhus and Role of Serum Osteopontin.” American Journal Review and Meta-Analysis.” The Lancet Infectious Diseases of Tropical Medicine and Hygiene 85 (5): 924–30. 10 (6): 417–32. Parker, T. M., C. K. Murray, A. L. Richards, A. Samir, T. Ismail, Reyburn, H., R. Mbatia, C. Drakeley, I. Carneiro, and others. 2007. “Concurrent Infections in Acute Febrile E. Mwakasungula, and others. 2004. “Overdiagnosis of Illness Patients in Egypt.” American Journal of Tropical Malaria in Patients with Severe Febrile Illness in Tanzania: Medicine and Hygiene 77 (2): 390–92. A Prospective Study.” BMJ 329 (7476): 1212–15. Peacock, S. J., and P. N. Newton. 2008. “Public Health Impact Robays, J., M. E. Raguenaud, T. Josenando, and M. Boelaert. of Establishing the Cause of Bacterial Infections in Rural 2008. “Eflornithine Is a Cost-Effective Alternative to Asia.” Transactions of the Royal Society of Tropical Medicine Melarsoprol for the Treatment of Second-Stage Human and Hygiene 102 (1): 5–6. West African Trypanosomiasis in Caxito, Angola.” Tropical Penno, E. C., S. J. Baird, and J. A. Crump. 2015. “Cost- Medicine and International Health 13 (2): 265–71. Effectiveness of Surveillance for Bloodstream Infections for Rubach, M. P., V. P. Maro, J. A. Bartlett, and J. A. Crump. 2015. Sepsis Management in Low-Resource Settings.” American “Etiologies of Illness among Patients Meeting Integrated Journal of Tropical Medicine and Hygiene 93 (4): 850–60. Management of Adolescent and Adult Illness District Penno, E. C., J. A. Crump, and S. J. Baird. 2015. Clinician Manual Criteria for Severe Infections in Northern “Performance Requirements to Achieve Cost- Tanzania: Implications for Empiric Antimicrobial Therapy.” Effectiveness of Point-of-Care Tests for Sepsis among American Journal of Tropical Medicine and Hygiene 92 (2): Patients with Febrile Illness in Low-Resource Settings.” 454–62. American Journal of Tropical Medicine and Hygiene Sanders, E. J., P. Mugo, H. A. Prins, E. Wahome, A. N. Thiong’o, 93 (4): 841–49. and others. 2014. “Acute HIV-1 Infection Is as Common as Petti, C. A., C. R. Polage, T. C. Quinn, A. R. Ronald, and M. A. Malaria in Young Febrile Adults Seeking Care in Coastal Sande. 2006. “Laboratory Medicine in Africa: A Barrier to Kenya.” AIDS 28 (9): 1357–63. Effective Health Care.” Clinical Infectious Diseases 42 (3): Say, L., D. Chou, A. Gemmill, O. Tunçalp, A. B. Moller, and others. 377–82. 2014. “Global Causes of Maternal Death: A WHO Systematic Phommasone, K., D. H. Paris, T. Anantatat, J. Castonguay- Analysis.” The Lancet Global Health 2 (6): e323–33. Vanier, S. Keomany, and others. 2013. “Concurrent Infection Shillcutt, S., C. Morel, C. Goodman, P. Coleman, D. Bell, and with Murine Typhus and Scrub Typhus in Southern Laos: others. 2008. “Cost-Effectiveness of Malaria Diagnostic The Mixed and the Unmixed.” PLoS Neglected Tropical Methods in Sub-Saharan Africa in an Era of Combination Diseases 7 (8): e2163. Therapy.” Bulletin of the World Health Organization 86 (2): Poulos, C., R. Bahl, D. Whittington, M. K. Bhan, J. D. Clemens, 101–10. and others. 2004. “A Cost-Benefit Analysis of Typhoid Fever Shirima, G. M., S. Cleveland, R. R. Kazwala, D. M. Kambarage, Immunization Programmes in an Indian Urban Slum N. French, and others. 2007. “Seroprevalence of Brucellosis Community.” Journal of Health, Population and Nutrition in Smallholder Dairy, Agropastoral, Pastoral, Beef Ranch 22 (3): 311–21. and Wildlife Animals in Tanzania.” Bulletin of Animal Prabhu, M., W. L. Nicholson, A. J. Roche, G. J. Kersh, K. A. Health and Production in Africa 55: 13–21. Fitzpatrick, and others. 2011. “Q Fever, Spotted Fever Group, Shretta, R., J. Liu, C. Cotter, J. Cohen, C. Dolenz, and others. and Typhus Group Rickettsioses among Hospitalized 2017. “Malaria Elimination and Eradication.” In Disease Febrile Patients in Northern Tanzania.” Clinical Infectious Control Priorities (third edition): Volume 6, Major Infectious Diseases 53 (4): e8–e15. Diseases, edited by K. Holmes, S. Bertozzi, B. Bloom, and P. Prasad, N., D. R. Murdoch, H. Reyburn, and J. A. Crump. 2015. Jha. Washington, DC: World Bank. “Etiology of Severe Febrile Illness in Low- and Middle- Smit, P. W., I. Elliott, R. W. Peeling, D. Mabey, and Income Countries: A Systematic Review.” PLoS One 10 (6): P. N. Newton. 2014. “An Overview of the Clinical Use e0127962. of Filter Paper in the Diagnosis of Tropical Diseases.” Prasad, N., K. J. Sharples, D. R. Murdoch, and J. A. Crump. American Journal of Tropical Medicine and Hygiene 2015. “Community Prevalence of Fever and Relationship 90 (2): 195–210. with Malaria among Infants and Children in Low-Resource Stanaway, J. D., D. S. Shepard, E. A. Undurraga, Y. A. Areas.” American Journal of Tropical Medicine and Hygiene Halasa, L. E. Coffeng, and others. 2016. “The Global 93 (1): 178–80. Burden of Dengue: An Analysis from the Global Burden Punjabi, N. H., W. R. Taylor, G. S. Murphy, S. Purwaningsih, of Disease Study 2013.” The Lancet Infectious Diseases. H. Picarima, and others. 2012. “Etiology of Acute, Non- doi:10.1016/S1473-3099(16)00026-8. Malaria, Febrile Illnesses in Jayapura, Northeastern Papua, Suputtamongkol, Y., W. Chaowagul, P. Chetchotisakd, Indonesia.” American Journal of Tropical Medicine and N. Lertpatanasuwun, S. Intaranongpai, and others. 1999. Hygiene 86 (1): 46–51. “Risk Factors for Melioidosis and Bacteremic Melioidosis.” Rammaert, B., J. Beauté, L. Borand, S. Hem, P. Buchy, and Clinical Infectious Diseases 29 (2): 408–13. others. 2011. “Pulmonary Melioidosis in Cambodia: A Suputtamongkol, Y., W. Pongtavornpinyo, Y. Lubell, C. Suttinont, Prospective Study.” BMC Infectious Diseases 11: 126. S. Hoontrakul, and others. 2010. “Strategies for Diagnosis Reddy, E. A., A. V. Shaw, and J. A. Crump. 2010. “Community- and Treatment of Suspected Leptospirosis: A Cost-Benefit Acquired Bloodstream Infections in Africa: A Systematic Analysis.” PLoS Neglected Tropical Diseases 4 (2): e610. Febrile Illness in Adolescents and Adults 383 Tabernero, P., F. M. Fernández, M. Green, P. J. Guerin, and P. N. ———. 2008. Consultation on Technical and Operational Newton. 2014. “Mind the Gaps: The Epidemiology of Poor- Recommendations for Clinical Laboratory Testing Harmonization Quality Anti-Malarials in the Malarious World: Analysis of and Standardization. Report of consensus meeting of major the WorldWide Antimalarial Resistance Network Database.” stakeholders, Maputo, Mozambique, January 22–24. Malaria Journal 13: 139. ———. 2009. Integrated Management of Adolescent and Thomsen, R. W., H. H. Hundborg, H. H. Lervang, S. P. Johnsen, Adult Illness (IMAI) Acute Care: Guidelines for First- H. C. Schønheyder, and others. 2005. “Diabetes Mellitus Level Facility Health Workers at Health Centre and as a Risk and Prognostic Factor for Community-Acquired District Outpatient Clinic. WHO/CDS/IMAI/2004.1, Bacteremia due to Enterobacteria: A 10-Year, Population- Rev. 3. Geneva: WHO. http://www.who.int/hiv/pub/imai Based Study among Adults.” Clinical Infectious Diseases /acute_care.pdf?ua=1. 40 (4): 628–31. ———. 2010a. Antiretroviral Therapy for HIV Infection in Vallée, J., T. Thaojaikong, C. E. Moore, R. Phetsouvanh, Adults and Adolescents. Recommendations for a Public A. L. Richards, and others. 2010. “Contrasting Spatial Health Approach: 2010 Revision. Geneva: WHO. Distribution and Risk Factors for Past Infection with Scrub ———. 2010b. Guidelines for the Treatment of Malaria. 2nd ed. Typhus and Murine Typhus in Vientiane City, Lao PDR.” Geneva: WHO. PLoS Neglected Tropical Diseases 12: e909. ———. 2010c. Guidelines for the Treatment of Tuberculosis. Van den Berghe, G., A. Wilmer, G. Hermans, W. Meersseman, 4th ed. Geneva: WHO. P. J. Wouters, and others. 2006. “Intensive Insulin Therapy ———. 2011a. IMAI District Clinician Manual. Hospital Care in the Medical ICU.” New England Journal of Medicine for Adolescents and Adults: Guidelines for the Management of 354 (5): 449–61. Illnesses with Limited Resources. Geneva: WHO. Vanderburg, S., M. P. Rubach, J. E. Halliday, S. Cleaveland, ———. 2011b. Malaria Rapid Diagnostic Test Performance. E. A. Reddy, and others. 2014. “Epidemiology of Coxiella Results of WHO Product Testing of Malaria RDTs: Round 3 burnetii Infection in Africa: A OneHealth Systematic (2010–2011). Geneva: WHO. Review.” PLoS Neglected Tropical Diseases 8 (4): e2787. ———. 2013. World Malaria Report 2013. Geneva: WHO. Wardrop, N. A., C. C. Kuo, H. C. Wang, A. C. Clements, WHO, and UNAIDS (Joint United Nations Programme on F. F. Lee, and others. 2013. “Bayesian Spatial Modelling and HIV/AIDS). 2007. Guidance on Provider-Initiated HIV the Significance of Agricultural Land Use to Scrub Typhus Testing and Counselling in Health Facilities. Geneva: WHO. Infection in Taiwan.” Geospatial Health 8 (1): 229–39. Wiseman, V., M. Kim, T. K. Mutabingwa, and C. J. Whitty. White, L. J., P. N. Newton, R. J. Maude, W. Pan-ngum, J. R. Fried, 2006. “Cost-Effectiveness Study of Three Antimalarial Drug and others. 2012. “Defining Disease Heterogeneity to Guide Combinations in Tanzania.” PLoS Medicine 3 (10): e373. the Empirical Treatment of Febrile Illness in Resource Poor Yazdanbakhsh, M., and P. G. Kremsner. 2009. “Influenza in Settings.” PLoS One 7 (9): e44545. Africa.” PLoS Medicine 6 (12): e1000182. White, M. T., L. Conteh, R. Cibulskis, and A. C. Ghani. Yukich, J., V. D’Acremont, J. Kahama, N. Swai, and C. Lengeler. 2011. “Costs and Cost-Effectiveness of Malaria Control 2010. “Cost Savings with Rapid Diagnostic Tests for Malaria Interventions: A Systematic Review.” Malaria Journal 10: 337. in Low-Transmission Areas: Evidence from Dar es Salaam, WHO (World Health Organization). 1997. “Joint United Nations Tanzania.” American Journal of Tropical Medicine and Programme on HIV/AIDS (UNAIDS)–WHO Revised Hygiene 83 (1): 61–68. Recommendations for the Selection and Use of HIV Antibody Zurovac, D., B. A. Larson, W. Akhwale, and R. W. Snow. 2006. Tests.” Weekly Epidemiological Record 72 (12): 81–88. “The Financial and Clinical Implications of Adult Malaria ———. 2006. Guidelines for the Treatment of Malaria. Geneva: Diagnosis Using Microscopy in Kenya.” Tropical Medicine WHO. and International Health 11 (8): 1185–94. 384 Major Infectious Diseases Chapter 15 Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children Joseph B. Babigumira, Hellen Gelband, and Louis P. Garrison Jr. INTRODUCTION laboratory facilities or technical training have become Fever is a common presenting complaint of ill children available. In light of these two developments, in 2006 all over the world. Until recently in Sub-Saharan Africa, the World Health Organization (WHO) recommended fever was synonymous with a presumed diagnosis of that parasitological confirmation precede malaria treat- malaria. However, malaria is not the only common ment except in children in high-transmission settings cause of fever or serious febrile illness (FI) in Sub- (WHO 2006), and in 2010 the WHO made the recom- Saharan Africa, and the widespread success of malaria mendation universal, even for highly exposed children control has reduced the region’s share of the FI burden. (WHO 2010). Most Sub-Saharan African countries In 2008, 16 percent of the 4.2 million deaths of children have officially adopted this policy, although few have in this region were attributed to malaria, 18 percent to been able to implement it fully. pneumonia, and 19 percent to diarrhea (Black and oth- The major advantages claimed for pretreatment ers 2010). confirmation of malaria are the following: Historically, the response to fever in children in Sub- Saharan Africa was presumptive antimalarial treat- • Prevention of unnecessary ACT use, which can save ment. Cheap, safe, and efficacious antimalarial money and reduce drug pressure that could lead to medications were widely available, and the only method resistance of diagnosis—microscopy—was scarce. The histori- • More appropriate treatment of nonmalaria fevers cally inexpensive medicines—chloroquine (CQ) and • Improved surveillance and better data for planning. sulfadoxine/pyrimethamine (SP)—succumbed to the development of drug-resistant malaria parasites; since The appropriateness of the WHO test-and-treat policy 2000, these drugs have been replaced as first-line treat- is clear in low-endemicity settings; it is not so clear in many ment by the more expensive but highly efficacious high-transmission settings, where all of the supposed artemisinin-based combination therapies (ACTs). advantages have been challenged (D’Acremont and others Rapid diagnostic tests (RDTs) that do not require 2009; English and others 2009; Graz and others 2011). Corresponding author: Joseph B. Babigumira, Global Medicines Program, Department of Global Health, University of Washington, Seattle, Washington, United States; babijo@uw.edu. 385 Many economic evaluations have compared malaria It also examines the impact of the availability of different RDTs to presumptive treatment and microscopy, using levels of diagnosis on optimal FI management strategy. cost-effectiveness methods and measuring the following: METHODOLOGY • Cost per correctly diagnosed malaria case (Bualombai and others 2003; Chanda, Castillo-Riquelme, and Analytic Overview Masiye 2009; de Oliveira, de Castro Gomes, and The reference case for this decision-analytic policy model Toscano 2010; Fernando and others 2004; Rolland and of FI management is a child under age five years present- others 2006) ing at a point of care (including pharmacies and drug • Cost per correctly diagnosed and treated malaria case sellers) with fever or history of fever. The model covers a (Batwala and others 2011; Chanda and others 2011; one-month time horizon given the short duration of Lubell and others 2007; Ly and others 2010; Rosas acute febrile illnesses such as malaria and respiratory Aguirre, Llanos Zavalaga, and Trelles de Belaunde tract infections. This chapter presents results for a run of 2009; Willcox and others 2009; Zikusooka, McIntyre, the model using parameters specific to Tanzania, which and Barnes 2008) represent the best estimates from publicly available • Cost per disability-adjusted life year (DALY) averted sources, including the published scientific literature and (Shillcutt and others 2008). various reports. The model itself can be adapted to dif- ferent regions, countries, or settings if local data are Others have used cost-benefit analysis (Bisoffi and available. A key characteristic of the analysis is the others 2011; Lubell, Hopkins, and others 2008; Lubell, planned flexibility of the ranges used for sensitivity anal- Reyburn, and others 2008). yses, which are varied to reflect current and potential Most evaluations have found that an RDT test-and- future policy goals or field realities. treat approach performs better than a microscopy test- The analysis assumes that children with fever may and-treat approach or presumptive treatment below a have malaria; treatable nonmalaria febrile illness certain level of malaria endemicity (Batwala and others (T-NMFI), which is illness that responds to appropriate 2011; Bisoffi and others 2011; Bualombai and others antibiotic treatment; or nontreatable nonmalaria febrile 2003; Chanda, Castillo-Riquelme, and Masiye 2009; illness (NT-NMFI), which is viral illness. It also assumes Hansen and others 2015; Ly and others 2010; Mosha and that severity assessments are possible at all points of care others 2010; Msellem and others 2009; Rolland and oth- (which is a simplifying assumption that is not uniformly ers 2006; Rosas Aguirre, Llanos Zavalaga, and Trelles de true) and that, although children may have malaria par- Belaunde 2009; Shillcutt and others 2008; Uzochukwu asites, the index illness may be caused by something else. and others 2009; Zikusooka, McIntyre, and Barnes 2008; (Even in cases of nonclinical malaria infection, it is Zurovac and others 2008). Microscopy performed better assumed that eliminating malaria parasites with an anti- than RDT in Brazil (de Oliveira, de Castro Gomes, and malarial drug is beneficial.) Toscano 2010), Sri Lanka for Plasmodium vivax The model assesses the costs, effectiveness, and (Fernando and others 2004), and one high-transmission cost-effectiveness of using malaria RDTs and treating setting (Willcox and others 2009) and about equivalent children with acute FI with antimalarial drugs, antibiot- to RDT in Ghana (Ansah and others 2013). ics, both, or neither. Model parameters include the fol- The most influential factors affecting the results are lowing: rural versus urban location, type of facility, malaria transmission intensity (Lubell, Hopkins, and malaria transmission intensity, etiology of FI, access to others 2008; Zurovac and others 2008), cost and accu- diagnostic technology (RDTs and microscopy), antima- racy of the RDTs (Lubell, Hopkins, and others 2008), age larial medications and antibiotics, diagnostic test perfor- (Zikusooka, McIntyre, and Barnes 2008), season (Bisoffi mance in the field, adherence to negative malaria test and others 2011), and response to negative test results results by clinicians or other prescribers, adverse drug (Bisoffi and others 2011; Lubell, Reyburn, and others events, mortality, and costs (in 2013 U.S. dollars). The 2008). analytic framework also allows for the assessment of The analysis in this chapter assesses the potential sequential treatment for FI: children who initially pre- cost-effectiveness of RDTs and their role in treatment sent with mild illness may return with severe illness. strategies for overall FI management in children under In the base case, malaria diagnosis is by RDTs if avail- age five years, taking into account transmission intensity, able, by microscopy if RDTs are not available, and by no treatment setting, and relative availability of antibiotic testing if both are unavailable. The base case assumes treatment (or no drug treatment) for nonmalaria FI. malaria treatment by ACTs if available, by another 386 Major Infectious Diseases antimalarial medicine if ACT is not available, and by a Table 15.1 Modeled Febrile Illness Management broad-spectrum antibiotic for strategies that include an Strategies antibiotic. Differential access to diagnostic tests and RDT Antimalarial Antibiotic drugs is explicitly built into the model, and the base case Comparator administered given given allows some patients to go without diagnosis or treat- ment. Variations in the impact of universal access to P-1 To none All None ACTs and antibiotics are examined using scenario and P-2 To none All If severe illness sensitivity analyses. The main sensitivity analyses assume P-3 To none All All universal access to ACTs to mirror the likely near-future RDT-1 To all If RDT None state of affairs, but this access depends critically on price. positive Key scenario and sensitivity analyses are used to answer policy questions that include the impact on opti- RDT-2 To all If RDT If severe illness positive and malaria, mal FI management strategy of increasing access to no antibiotic; diagnostic tests, antimalarial medicines, and antibiotics if severe illness and whether increasing access to these commodities and no malaria, would have the greatest impact in low- or high- treat with transmission settings, public or private settings, and antibiotic urban or rural areas. The analytic framework also allows RDT-3 To all If RDT If no malaria, an assessment to be made of the impact of prescriber positive treat with adherence to negative tests on the optimal strategy. The antibiotic model calculates the expected probability of survival and RDT-4 To all If RDT If severe illness, costs for different FI management strategies and esti- positive treat with mates the optimal strategy from the standpoint of sur- antibiotic vival, cost, and cost-effectiveness. Note: P = presumptive treatment; RDT = rapid diagnostic test. Included are the costs of diagnostic technology (RDTs and microscopy), antimalarial medicines, and broad- spectrum antibiotics; the added cost of assessing severity and administering RDTs; the cost of treating mild and • Potential future options, given improving access to severe disease; and the cost of managing adverse events. malaria diagnostic technology, ACTs, new bacterial Direct nonmedical costs incurred by patients and indirect illness diagnostics, and antibiotics. costs due to lost productivity of parents are not included. Health system costs, cost of health worker training, cost Some of the strategies are unlikely to be imple- of creating demand with behavior change communica- mented in the real world but have value as historical tion, cost of future ACT and antibiotic resistance, and comparisons or for assessment of the potential impact cost of potential RDT use in the private and informal of poor implementation. In all cases, the diagnostic sectors (such as sharps disposal) are not included. This technology is presumed to be RDT if available, and if approach constitutes a modified societal perspective for RDTs are not available, microscopy if available. The the analysis (Garrison and others 2010). antimalarial is presumed to be an ACT if available, and if an ACT is not available, an alternative antimalarial. The strategies are as follows: Comparators: Potential Strategies for Febrile Illness Management • P-1, presumptive treatment with antimalarials only Seven strategies are compared (table 15.1): three pre- for all children. Presumptive treatment of FI because sumptive strategies (P-1, P-2, and P-3) and four malaria was the historical management option in the diagnosis-based strategies (RDT-1, RDT-2, RDT-3, and vast majority of low-income, resource-constrained RDT-4). The strategies were constructed to encompass settings with high malaria endemicity in the pre-RDT the following: era. It remains the default where RDTs are not avail- able and malaria is still common. Under P-1, a child • Historical policy options presenting with fever or a history of fever is treated • Possible policy options given actual conditions in with an ACT if available, another antimalarial if an the field ACT is not available, and no treatment if ACTs or • Pragmatic policy options given system capacity and other antimalarials are not available. P-1 is the base health workforce issues comparator. Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 387 • P-2, presumptive treatment with antimalarials for are treated with an available antimalarial. All chil- all children and presumptive treatment with broad- dren testing negative for malaria are treated with a spectrum antibiotics for children with severe illness. P-2 broad-spectrum antibiotic. is modeled around the original version of Integrated • RDT-4, treatment with antimalarials for children who Management of Childhood Illness (IMCI), intro- test positive for malaria and presumptive treatment duced by the WHO in 1997 in response to increasing with broad-spectrum antibiotics for all children with under-five mortality in low-income countries (WHO severe disease. Under RDT-4, children with fever are 1999). The original version recommended that all tested for malaria using RDT or microscopy, and children with fever or a history of fever receive a first- those testing positive are treated with an available line antimalarial drug and be evaluated for signs of antimalarial. All children showing signs of severe dis- other potential causes of fever, such as rapid breath- ease are treated with a broad-spectrum antibiotic in ing for pneumonia, followed by appropriate treat- addition to the antimalarial medicine. ment. It did not explicitly recommend parasitological confirmation of malaria. • P-3, presumptive treatment with both antimalarials and antibiotics for all children. P-3 is included as a Decision-Analytic Model fallback position in recognition of the difficulty of The model consists of (1) a “front-end” decision tree clinically assessing children for pneumonia and other that classifies presenting children by their setting of serious causes of fever. Such assessment is usually not treatment, point of care, diagnostic result, and treatment carried out by caregivers making treatment decisions, received, and (2) a “back-end” Markov model that esti- even in primary care facilities that are understaffed or mates the impact of illness severity, progression, and staffed by poorly trained health care workers. mortality on costs and outcomes. • RDT-1, treatment with antimalarials for children who The front end of the model is divided into four test positive for malaria and no treatment for children parts, as shown in figure 15.1: panel a shows FI manage- who test negative. RDT-1 is included in the model to ment strategies, treatment setting, and disease etiology; demonstrate the potential consequences of untreated panel b shows malaria diagnostic test availability and T-NMFI. Children with fever are tested for malaria test results (true positive, false negative, false positive, using RDT or microscopy, and those testing positive true negative); panel c shows availability and prescrip- are treated with an available antimalarial. No antibi- tion of ACT and antibiotics; and panel d shows otics are prescribed regardless of test result or disease availability and prescription of antimalarials (CQ and severity. SP) and antibiotics. • RDT-2, treatment with antimalarials for children who In figure 15.1, panel a, children with FI may present test positive for malaria and presumptive treatment and be treated in rural or urban settings and at one of with antibiotics for children with severe illness who test five points of care: at home by a community health negative. RDT-2 mirrors the second iteration of IMCI, worker, at a general retail outlet, at a drug shop such which recommended that the assessment of children as duka la dawa baridis in Tanzania or a pharmacy, at a with fever include diagnostic testing for malaria. In private health facility, or at a public health facility the modeling framework, children with fever or a (including nongovernmental organization and faith- history of fever are tested for malaria using RDT or based facilities). Children may live and present for microscopy and those testing positive are treated with treatment in high-transmission-intensity areas (1 or an available antimalarial. Children testing negative more malaria cases per 1,000 population) or low-trans- are assessed for signs of disease severity (such as fast mission-intensity areas (0.051 cases per 1,000 popula- breathing, dehydration), and those showing signs tion). Depending on the setting, children may or may of severe disease are treated with a broad-spectrum not have parasites in their blood. Those who are para- antibiotic in addition to the antimalarial medicine. sitemic may have clinical malaria or asymptomatic • RDT-3, treatment with antimalarials for children parasitemia, in which case their illness is caused by who test positive for malaria and presumptive treat- T-NMFI, usually bacterial infection, or NT-NMFI, ment with broad-spectrum antibiotics for all children usually viral infection. Those who are not parasitemic who test negative. RDT-3 is included in the model will similarly have T-NMFI or NT-NMFI. In this anal- to demonstrate the potential consequences of pre- ysis, the combined diagnosis of malaria and T-NMFI sumptive treatment of all NMFI with antibiotics. In is modeled, but not the other combined diagnoses the model, children with fever are tested for malaria such as T-NMFI plus NT-NMFI or malaria plus using RDT or microscopy, and those testing positive NT-NMFI. 388 Major Infectious Diseases Figure 15.1 Decision-Analytic Model a. FI management strategies, treatment setting, and FI etiology b. Availability of malaria diagnostic tests and test results Malaria only RDT true positive Malaria RDT access Malaria + treatable FI RDT false negative Parasitemia Treatable FI Malaria Micro. true positive No malaria Micro. access Nontreatable FI No RDT Micro. false negative High TI Treatable FI No test No parasitemia RDT false positive Home care/CHW Nontreatable FI RDT access Low TI RDT true negative Retail shop Urban No malaria DLDP/Pharmacy Micro. false positive Private FP HF Micro. access P-1 Gov’t or NFP HF No RDT Micro. true negative P-2 Rural No test P-3 RDT-1 Febrile child RDT-2 RDT-3 RDT-4 c. Availability and prescription of ACT and antibiotics d. Availability and prescription of antimalarials and antibiotics Give Abx M Give Abx M Abx available Abx available Do not Do not Give ACT give Abx Give antimal give Abx M M Abx Abx unavailable unavailable ACT M M available Give Abx Antimal available Give Abx M M Abx available Abx available Do not Do not Do not Do not (Post-test) give ACT give Abx give antimal give Abx M M Abx Abx unavailable ACT unavailable unavailable ACT M M unavailable Give Abx M Abx available Do not Antimal give Abx unavailable M Abx unavailable M figure continues next page Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 389 Figure 15.1 Decision-Analytic Model (continued) e. Outcomes of the Markov submodel Well Mild febrile illness Severe febrile illness Dead Note: Abx = antibiotics; ACT = artemisinin-based combination therapy; Antimal = antimalarial; CHW = community health worker; DLDB = duka la dawa baridi; FI = febrile illness; FP = for profit; Gov’t = government; HF = health facility; M = transition to Markov Model; Micro. = microscopy; NFP = not for profit; RDT = rapid diagnostic test; TI = transmission intensity. In figure 15.1, panel b, children with malaria (malaria face a mortality risk. The model does not allow mortal- only and malaria plus T-NMFI) as well as children ity other than from the severe state nor complete well- without malaria are subjected to diagnostic testing ness without moving through the mild state. Both the depending on access to testing technology. At the point well and the dead states are absorbing states. The rates of care, malaria RDTs may or may not be available. If of mortality and progression depend on the medicine available, the model assumes that they are the first received and the true diagnosis as defined earlier in the choice, but if unavailable, the provider uses microscopy, model. Where diagnostic testing results in false nega- if available. If microscopy is also unavailable, children are tives or false positives, the model directs patients into assumed to be treated with an antimalarial. Depending incorrect treatment algorithms, which are included in on the performance (sensitivity and specificity) of the the model. Given that only two transitions are allowed malaria test used (RDT or microscopy), patients with from both the mild and severe states, the rates of transi- malaria are divided into two treatment pools: true posi- tion from mild illness to well and from severe illness to tive and false negative. Those without malaria are divided mild are modeled as complements of progression and into false positive and true negative pools. mortality, respectively. In figure 15.1, panels c and d, following the test and depending on the treatment pool, patients are treated with ACTs, antimalarial medicines, or antibiotics Tanzania Parameters depending on their availability and in line with one of Setting of Febrile Illness Management the seven FI management strategies in the model. At the Tanzania is 26 percent urban and 74 percent rural point of care it is assumed that if ACTs are available, they (Tanzania NBS 2011); 24 percent of children with fever are given, but if ACTS are not available, other antimalar- in urban areas and 39 percent in rural areas did not ials are given. The model also explicitly considers the seek fever treatment from health care providers in provider’s decision to give antimalarial medicines and 2010 (Tanzania NBS and Macro International 2011), antibiotics in the face of positive or negative malaria test an estimate used to represent children under home results. In the base case, when no medicines are available, care, assuming that their point of care is through a children go untreated. Adverse drug events also occur in pharmacy or drug shop or a community health worker. the model, depending on the drug given, with conse- Information from Kenya is used in the absence of quences for cost but not for mortality. information from Tanzania to estimate the distribu- The Markov model (figure 15.1, panel e) divides tion of point of care by different urban facilities acute FI into four health states: well, mild illness, severe (Molyneux and others 1999). For rural areas, a study illness, and dead. All children start in the mild or from Kilosa, Tanzania (Simba and others 2010), is severe FI health state. During every cycle, assumed to be used. An estimated 73 percent of Tanzania’s popula- one week, children in the mild state may progress to tion live in high-transmission-intensity areas, and 27 severe and children in the severe state may improve and percent live in low-transmission-intensity areas (WHO join the mild state. Children in the mild state may also 2012b). The model distribution by setting is presented move to the well state, and children in the severe state in table 15.2. 390 Major Infectious Diseases Table 15.2 Modeled Distribution of Children with Table 15.3 Estimated Average Probability of Malaria Febrile Illness in Tanzania, by Management Setting, Parasitemia and Febrile Illness Etiology in Children in 2013, Based on Best Estimates Tanzania, 2013 Setting Base case Reference Indicator Base case Reference Residence Tanzania NBS 2011 Parasitemia Urban 0.742 High transmission 0.423 Patrick Kachur and others 2006 Rural 0.258 Low transmission 0.013 Zanzibar Malaria Control Point of care Program and Karolinska Institute 2012 Urban Molyneaux and others 1999; Tanzania Malaria NBS and Macro High transmission Gosoniu and others 2012 International 2011 Clinical 0.870 Home care 0.242 Asymptomatic 0.130 General shop 0.087 Low transmission Assumption Pharmacy or drug shop 0.270 Clinical 0.999 Private health facility 0.298 Asymptomatic 0.001 Government health facility 0.103 NMFI D’Acremont 2011 Rural Simba and others Treatable 0.182 2010; Tanzania NBS and Macro Nontreatable 0.714 International 2011 Malaria and 0.343 D’Acremont 2011 Home care 0.391 treatable NMFI Note: NMFI = nonmalaria febrile illness. General shop 0.294 Pharmacy or drug shop 0.092 Availability and Performance of Diagnostic Tests Private health facility 0.166 Table 15.4 differentiates the availability of diagnostic tests Government health facility 0.057 between urban and rural settings, assuming at baseline Transmission intensity WHO 2012b that diagnosis by microscopy is not available in home High 0.730 care, general shops, and duka la dawa baridis. The sensi- tivity of microscopy in Tanzania is assumed to be 71.4 Low 0.270 percent, and specificity is assumed to be 47.3 percent. The sensitivity of RDTs is 97.0 percent, and specificity is 96.8 percent (Kahama-Maro and others 2011). Parasitemia and Etiology of Febrile Illness In low-transmission areas (Pemba and Zanzibar), the Availability of Medicines for Febrile Management parasite prevalence among children presenting with In the absence of data from Tanzania, ACT access infor- fever is 0.13 percent (Zanzibar Malaria Control Program mation from Uganda is used (ACTwatch Group, PACE, and Karolinska Institute 2012), and all patients with and IE Team 2012), combined with published estimates parasitemia are expected to have clinical malaria. In for segments of the population in Tanzania (Simba and high-transmission areas, the parasite prevalence among others 2010). As an example, in rural areas, access to children with fever is approximately 42 percent (Patrick ACTs is 71 percent in government facilities and 11 percent Kachur and others 2006), and 13 percent of patients with in private facilities in rural areas (table 15.5). parasitemia are assumed to be asymptomatic (Gosoniu and others 2012). Prescribing Practices and Prescriber Adherence to Table 15.3 presents the parasitemia and etiology of Negative Test Results malaria in Tanzania, assuming that 81 percent of acute Based on a randomized trial in Tanzania comparing respiratory infections and 75 percent of other infections RDTs with routine microscopy (Reyburn and others of unknown etiology are due to viruses. A treatable 2007), results in table 15.6 assume that all patients who co-infection occurs in 34 percent of patients with test positive or are not tested receive antimalarials malaria (D’Acremont 2011). (Lubell, Reyburn, and others 2008). Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 391 Table 15.4 Availability and Performance of Rapid Diagnostic Tests and Microscopy in Tanzania, 2013 Indicator Base case Reference RDTs Availability Urban Home care 0.000 Assumption General shop 0.001 Assumption Pharmacy or drug shop 0.056 Albertini and others 2012; CPM 2008 Private health facility 0.835 Assumption Government health facility 0.635 Masanja and others 2012; assumption Rural Home care 0.000 Assumption General shop 0.000 Assumption Pharmacy or drug shop 0.000 Assumption Private health facility 0.635 Assumption Government health facility 0.435 Masanja and others 2012 Performance Kahama-Maro and others 2011 Sensitivity 0.970 Specificity 0.968 Microscopy Availability Urban Home care 0.000 Assumption General shop 0.000 Assumption Pharmacy or drug shop 0.000 Assumption Private health facility 1.000 Assumption Government health facility 0.366 Masanja and others 2012; Tanzania NBS and Macro International 2007; assumption Rural Home care 0.000 Assumption General shop 0.000 Assumption Pharmacy or drug shop 0.000 Assumption Private health facility 0.800 Assumption Government health facility 0.190 Masanja and others 2012 Performance Kahama-Maro and others 2011; Masanja and others 2012 Sensitivity 0.714 Specificity 0.473 Note: RDTs = rapid diagnostic tests. 392 Major Infectious Diseases Table 15.5 Availability of Antimalarial Medicines and Antibiotics in Tanzania, 2013 Indicator Base case Reference ACTs Urban Home care 0.999 ACTwatch Group, PACE, and IE Team 2012 General shop 0.560 ACTwatch Group, PACE, and IE Team 2012 Pharmacy or drug shop 0.968 ACTwatch Group, PACE, and IE Team 2012 Private health facility 0.791 ACTwatch Group, PACE, and IE Team 2012 Government health facility 0.780 Chimnani and others 2010 Rural Home care 0.550 ACTwatch Group, PACE, and IE Team 2012 General shop 0.747 ACTwatch Group, PACE, and IE Team 2012 Pharmacy or drug shop 0.736 Yadav and others 2012 Private health facility 0.110 Simba and others 2010 Government health facility 0.710 Chimnani and others 2010 Other antimalarial drugs Urban Home care 0.080 ACTwatch Group, PACE, and IE Team 2012 General shop 0.001 ACTwatch Group, PACE, and IE Team 2012 Pharmacy or drug shop 0.995 ACTwatch Group, PACE, and IE Team 2012 Private health facility 0.960 ACTwatch Group, PACE, and IE Team 2012 Government health facility 0.760 Chimnani and others 2010 Rural Home care 0.111 ACTwatch Group, PACE, and IE Team 2012 General shop 0.004 ACTwatch Group, PACE, and IE Team 2012 Pharmacy or drug shop 0.999 ACTwatch Group, PACE, and IE Team 2012 Private health facility 0.943 ACTwatch Group, PACE, and IE Team 2012 Government health facility 0.670 Chimnani and others 2010 Antibiotics Urban Home care 0.000 Assumption General shop 0.200 Assumption Pharmacy or drug shop 1.000 Assumption Private health facility 1.000 Assumption Government health facility 0.770 Chimnani and others 2010 Rural Home care 0.000 Assumption General shop 0.500 Assumption Pharmacy or drug shop 1.000 Assumption Private health facility 1.000 Assumption Government health facility 0.790 Chimnani and others 2010 Note: ACTs = artemisinin-based combination therapies. Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 393 Table 15.6 Average Probability of Prescription of Table 15.7 Starting Distributions among Markov States Different Medicines in Tanzania, by Malaria Diagnostic in Tanzania, by Diagnosis, 2013 Test Result, 2013 Indicator Base case Reference Indicator Base case Reference Well 0.00 Antimalarial prescribed Mild febrile illness RDT positive 1.000 Assumption Malaria Lubell, Staedke, and RDT negative Reyburn and others 2007 others 2011 Low transmission 0.697 Low transmission intensity 0.70 High transmission 0.410 High transmission intensity 0.87 Microscopy positive 1.000 Assumption Treatable febrile illness 0.70 Lubell, Staedke, and Microscopy negative Reyburn and others 2007 others 2011 Low transmission 0.626 Nontreatable febrile illness 0.90 Assumption High transmission 0.230 Severe febrile illness No test 1.000 Assumption Malaria Lubell, Staedke, and others 2011 Antibiotic prescribed Low transmission intensity 0.30 Malaria test positive 0.140 Reyburn and others 2007 High transmission intensity 0.13 Malaria test negative 0.740 Reyburn and others 2007 Treatable febrile illness 0.30 Lubell, Staedke, and No test 0.740 Assumption others 2011 Note: RDT = rapid diagnostic test. Nontreatable febrile illness 0.10 Assumption Dead 0.00 Markov Model Parameters The starting distributions of patients in Markov states, which depend on etiology and, for malaria, on transmis- (WHO 2011a), and the cost of treating severe disease is sion intensity, are summarized in table 15.7. For malaria estimated to be US$61.07, based on a cost-effectiveness and treatable nonmalaria FIs, data from a Delphi survey analysis of intravenous artesunate for severe malaria are used (Lubell, Staedke, and others 2011). For non- (Lubell, Riewpaiboon, and others 2011). Treating treatable nonmalaria FI, a 9:1 ratio of mild to severe drug-related adverse events is assumed to be equal to the disease at baseline is assumed. cost of treating mild illness (table 15.8). Adverse Events Analyses For children receiving ACTs and artesunate-mefloquine, 11.3 percent and 63.0 percent, respectively, experience The base case and the following scenarios were analyzed: any adverse event (Mueller and others 2006); these (1) universal access to RDTs; (2) universal access to figures are used here as the probability of adverse events ACTs; (3) universal access to antibiotics; (4) universal for other antimalarials; 16 percent experience any adverse access to RDTs and ACTs; (5) universal access to RDTs event due to amoxicillin (Garbutt and others 2012). and antibiotics; (6) universal access to ACTs and anti- biotics; and (7) universal access to RDTs, ACTs, and anti- biotics. Probabilities were varied by +/− 20 percent, and Costs of Diagnosis and Treatment costs were halved and doubled for sensitivity analyses. The costs of RDTs, microscopy, ACTs, other antimalarial TreeAge Pro 2013 was used for the analyses. medicines, and antibiotics are from ACTwatch (in the absence of data for Tanzania, data from Uganda are used) and from Health Action’s International Medicines RESULTS Price Workbook for Tanzania (WHO 2012a). Personnel In both low- and high-transmission settings and overall, costs for performing RDTs and for severity assessment presumptive treatment with ACTs and antibiotics (P-3) are from Uganda. The personnel cost of treating mild leads to the fewest deaths (393 per 10,000 children), and disease is estimated for Tanzania to be US$2.66, based on treating only RDT-positive children with an antimalarial the cost of a single outpatient visit inflated to 2013 costs alone (RDT-1) leads to the most deaths (484 per 10,000) 394 Major Infectious Diseases Table 15.8 Costs of Diagnosis and Treatment of Febrile Illness in Tanzania Indicator Base case (US$) Reference RDTs Urban ACTwatch Group, PACE, and IE Team 2012 CHW and home care 1.96 Public (including PNFP HF) 1.96 Private FP HF 1.17 Drug seller (pharmacy or drug shop) 0.98 General shop or vendor 0.98 Rural ACTwatch Group, PACE, and IE Team 2012 CHW and home care 0.78 Public (including PNFP HF) 0.78 Private FP HF 1.17 Drug seller (pharmacy or drug shop) 0.90 General shop or vendor 0.90 Microscopy Urban CHW and home care — Public (including PNFP HF) 0.78 ACTwatch Group, PACE, and IE Team 2012 Private FP HF 0.78 ACTwatch Group, PACE, and IE Team 2012 Drug seller (pharmacy or drug shop) 0.93 ACTwatch Group, PACE, and IE Team 2012 General shop or vendor — Rural CHW and home care — Public (including PNFP HF) 0.39 ACTwatch Group, PACE, and IE Team 2012 Private FP HF 0.78 ACTwatch Group, PACE, and IE Team 2012 Drug seller (pharmacy or drug shop) 0.59 ACTwatch Group, PACE, and IE Team 2012 General shop or vendor — ACTs Urban CHW and home care 0.20 WHO 2011a Public (including PNFP HF) 4.01 ACTwatch Group, PACE, and IE Team 2012 Private FP HF 9.70 ACTwatch Group, PACE, and IE Team 2012 Drug seller (pharmacy or drug shop) 9.70 ACTwatch Group, PACE, and IE Team 2012 General shop or vendor 9.70 ACTwatch Group, PACE, and IE Team 2012 Rural CHW and home care 0.20 WHO 2011a Public (including PNFP HF) 4.11 ACTwatch Group, PACE, and IE Team 2012 table continues next page Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 395 Table 15.8 Costs of Diagnosis and Treatment of Febrile Illness in Tanzania (continued) Indicator Base case (US$) Reference Private FP HF 9.76 ACTwatch Group, PACE, and IE Team 2012 Drug seller (pharmacy or drug shop) 9.76 ACTwatch Group, PACE, and IE Team 2012 General shop or vendor 9.76 ACTwatch Group, PACE, and IE Team 2012 Non-ACT antimalarials Urban ACTwatch Group, PACE, and IE Team 2012 CHW and home care 0.35 Public (including PNFP HF) 4.11 Private FP HF 4.93 Drug seller (pharmacy or drug shop) 4.93 General shop and vendor 4.93 Rural ACTwatch Group, PACE, and IE Team 2012 CHW and home care 0.35 Public (including PNFP HF) 2.46 Private FP HF 4.93 Drug seller (pharmacy or drug shop) 4.93 General shop or vendor 4.93 Antibiotics Urban WHO 2011a CHW and home care 0.81 Public (including PNFP HF) 0.78 Private FP HF 1.86 Drug seller (pharmacy or drug shop) 1.86 General shop or vendor 1.86 Rural WHO 2011a CHW and home care 0.81 Public (including PNFP HF) 0.78 Private FP HF 1.86 Drug seller (pharmacy or drug shop) 1.86 General shop or vendor 1.86 RDT personnel 0.20 Babigumira and others 2009 Severity assessment 0.30 Babigumira and others 2009 Treatment Mild disease 2.66 WHO 2011a Severe disease 61.07 Lubell, Riewpaiboon, and others 2011 Adverse event 2.66 WHO 2011a Note: — = not available; ACT = artemisinin-based combination therapy; CHW = community health worker; FP = for profit; HF = health facility; PNFP = private not for profit; RDT = rapid diagnostic test. 396 Major Infectious Diseases (tables 15.9–15.11). In the base case, P-3 is also the least Presumptive treatment with ACTs and antibiotics is the costly strategy (US$251,000 per 10,000 children), but the optimal strategy and is highly cost-effective in Tanzania. costs of the strategies vary by less than US$10,000 per The ranking of the strategies varies somewhat with 10,000 children in all cases except one (RDT-4—treating endemicity levels, but the leading strategy does not change. RDT-positive children with an antimalarial and all chil- The results are robust to univariate sensitivity dren with severe disease with an antibiotic). analyses. The cost estimates are most sensitive to the Table 15.9 Survival, Mortality, Costs, and Cost-Effectiveness per 10,000 Children Presenting with Fever in Tanzania Additional Cost in 2013 Incremental Policy Survivors Deaths lives saved US$ cost in 2013 US$ Cost-effectiveness RDT-1 9,516 484 n.a. 258,100 n.a. Lower than P-1, P-3, RDT-2, and RDT-3 RDT-4 9,562 438 46 263,400 5,300 Lower than P-1, P-2, P-3, RDT-2, and RDT-3 RDT-3 9,563 437 1 255,400 −8,000 Lower than P-1 and P-3 RDT-2 9,563 437 0 256,000 600 Lower than P-1 and P-3 P-1 9,566 434 3 253,500 −2,500 Lower than P-3 P-2 9,606 394 40 258,400 4,900 Lower than P-3 P-3 9,607 393 1 251,000 −7,400 Dominant Note: n.a. = not applicable. Table 15.10 Survival, Mortality, Costs, and Cost-Effectiveness per 10,000 Children Presenting with Fever in High-Transmission Areas of Tanzania Additional Cost in 2013 Incremental Policy Survivors Deaths lives saved US$ cost in 2013 US$ Cost-effectiveness RDT-1 9,534 466 n.a. 260,200 n.a. Lower than P-1, P-3, RDT-2, and RDT-3 RDT-4 9,572 428 38 265,300 5,100 Lower than P-1, P-2, P-3, RDT-2, and RDT-3 RDT-3 9,573 427 1 257,900 −7,400 Lower than P-1 and P-3 RDT-2 9,573 427 0 258,500 600 Lower than P-1 and P-3 P-1 9,587 413 14 255,100 −3,400 Lower than P-3 P-2 9,621 379 34 259,800 4,700 Lower than P-3 P-3 9,622 378 1 252,800 −7,000 Dominant Note: n.a. = not applicable. Table 15.11 Survival, Mortality, Costs, and Cost-Effectiveness per 10,000 Children Presenting with Fever in Low-Transmission Areas in Tanzania Additional Cost in 2013 Incremental Policy Survivors Deaths lives saved US$ cost in 2013 US$ Cost-effectiveness RDT-1 9,469 531 n.a. 252,400 n.a. Lower than P-1, P-3, and RDT-3 P-1 9,508 492 39 249,300 −3,100 Lower than P-3 and RDT-3 RDT-4 9,536 464 28 258,300 9,000 Lower than P-2, P-3, RDT-2, and RDT-3 RDT-3 9,538 462 2 248,700 −9,600 Lower than P-2 and P-3 RDT-2 9,538 462 0 249,500 800 Lower than P-3 P-2 9,566 434 28 254,700 5,200 Lower than P-3 P-3 9,568 432 2 246,100 −8,600 Dominant Note: n.a. = not applicable. Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 397 starting proportion of patients in the severe Markov within the same country. Defining the criteria for delin- health state for children with T-NMFI, and mortality is eating the areas where each policy would be appropriate most sensitive to the probability of progression from is a hurdle that would be created by having two policies. mild to severe illness for children with T-NMFI. Even more difficult may be deciding when and how to move from a presumptive to a test-and-treat policy as malaria control continues to lower endemicity levels. CONCLUSIONS Until primary health care is more widely available, a Presumptive treatment of all children under age five years large proportion of fevers in high-transmission rural with fever, or only those who are severely ill, with both areas will be managed in the informal sector, where the ACTs and a broad-spectrum antibiotic can minimize analysis suggests that it is more cost-effective to focus on mortality and is projected to be highly cost-effective by treatment than on diagnostics. In low-transmission- global standards. This result is based on conditions in intensity areas and in clinical settings with well-trained Tanzania, but it is generalizable to many Sub-Saharan practitioners, diagnostics are valuable for targeting treat- African countries with similar malaria endemicity and ment. Until the burden of malaria declines more broadly, health service delivery. countries might consider a mixture of strategies tailored The WHO recommendation of definitive malaria to local conditions. diagnosis before treatment is the clinical practice ideal; physicians aim to make definitive diagnoses before pre- scribing treatment of any kind. It is a useful goal and NOTE should be adopted in clinical settings where a test— World Bank Income Classifications as of July 2014 are as microscopy or an RDT—can be conducted reliably. follows, based on estimates of gross national income (GNI) Unfortunately, the places where malaria transmission is per capita for 2013: highest also tend to be the places where the capacity for testing and reliability are most limited. Drugs are often • Low-income countries (LICs) = US$1,045 or less purchased directly from pharmacies or drug shops or • Middle-income countries (MICs) are subdivided: (a) lower-middle-income = US$1,046 to US$4,125 from poorly staffed and provisioned health facilities, (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 both public and private. In these cases, presumptive • High-income countries (HICs) = US$12,746 or more. treatment with antimalarial medicines and antibiotics for all children is the only strategy that prevents the most deaths; it is also optimal from the standpoint of survival REFERENCES and cost-effectiveness. The same conclusion was reached in an independent analysis using a net health benefit ACTwatch Group, PACE (Program for Accessible Health, approach for six Sub-Saharan African countries (Basu, Communication, and Education), and IE (Independent Modred, and Bendavid 2014). Evaluation) Team. 2012. ACTwatch Outlet Survey Report Price is a major driver of the results of these analyses. 2011 (Round 4). Endline Outlet Survey Report for the Independent Evaluation of Phase 1 of the Affordable If prevailing ACT subsidies were lost and the prices of Medicines Facility: Malaria (AMFm), Uganda. Kampala, ACTs were to rise, presumptive treatment would become Uganda: ACTwatch Group, PACE, and IE Team. less attractive from a cost standpoint, but it would not alter Albertini, A., D. Djalle, B. Faye, D. Gamboa, J. Luchavez, and the effectiveness side of the equation. Even at a low cost, others. 2012. “Preliminary Enquiry into the Availability, some individuals are unable to afford ACTs. RDTs have Price, and Quality of Malaria Rapid Diagnostic Tests in the other costs that might further reduce their cost- Private Health Sector of Six Malaria-Endemic Countries.” effectiveness, including the costs of scaling up their use, Tropical Medicine and International Health 17 (2): 147–52. distribution and storage, sharps disposal, treatment of Ansah, E. K., M. Epokor, C. J. Whitty, S. Yeung, and K. S. Hansen. potential blood-borne infections from needle stick injuries, 2013. “Cost-Effectiveness Analysis of Introducing RDTs and behavior change to encourage adherence to test results. for Malaria Diagnosis as Compared to Microscopy and Cost and cost-effectiveness from a model such as this Presumptive Diagnosis in Central and Peripheral Public Health Facilities in Ghana.” American Journal of Tropical one are but one input into the development of a global Medicine and Hygiene 89 (4): 724–36. treatment policy. This analysis does not monetize certain Babigumira, J. B., B. Castelnuovo, M. Lamorde, A. Kambugu, important externalities, such as the cost of accelerating A. Stergachis, and others. 2009. “Potential Impact of Task- the development of antibiotic resistance or antimalarial Shifting on Costs of Antiretroviral Therapy and Physician resistance (although the latter may be small). Also Supply in Uganda.” BMC Health Services Research 9 (192): 192. important, but impossible to estimate the cost of, is the Basu, S., S. Modred, and E. Bendavid. 2014. “Comparing need to maintain different policies in different areas, even Decisions for Malaria Testing and Presumptive Treatment: 398 Major Infectious Diseases A Net Health Benefit Analysis.” Medical Decision Making Immunochromatographic P.f/P.v Assay for the Diagnosis 34 (8): 996–1005. of Plasmodium vivax Malaria in a Rural Area of Sri Lanka.” Batwala, V., P. Magnussen, K. S. Hansen, and F. Nuwaha. Annals of Tropical Medicine and Parasitology 98 (1): 5–13. 2011. “Cost-Effectiveness of Malaria Microscopy and Rapid Garbutt, J. M., C. Banister, E. Spitznagel, and J. F. Piccirillo. Diagnostic Tests versus Presumptive Diagnosis: Implications 2012. “Amoxicillin for Acute Rhinosinusitis: A Randomized for Malaria Control in Uganda.” Malaria Journal 10: 372. Controlled Trial.” Journal of the American Medical Association Bisoffi, Z., S. B. Sirima, F. Meheus, C. Lodesani, F. Gobbi, 307 (7): 685–92. and others. 2011. “Strict Adherence to Malaria Rapid Garrison, L. P., Jr, E. C. Mansley, T. A. Abbott, B. W. Bresnahan, Test Results Might Lead to a Neglect of Other Dangerous J. W. Hay, and others. 2010. “Good Research Practices for Diseases: A Cost Benefit Analysis from Burkina Faso.” Measuring Drug Costs in Cost-Effectiveness Analyses: Malaria Journal 10: 226. A Societal Perspective: The ISPOR Drug Cost Task Force Black, R. E., S. Cousens, H. L. Johnson, J. E. Lawn, I. Rudan, Report—Part II.” Value in Health 13 (1): 8–13. and others. 2010. “Global, Regional, and National Causes of Gosoniu, L., A. Msengwa, C. Lengeler, and P. Vounatsou. 2012. Child Mortality in 2008: A Systematic Analysis.” The Lancet “Spatially Explicit Burden Estimates of Malaria in Tanzania: 375 (9730): 1969–87. Bayesian Geostatistical Modeling of the Malaria Indicator Bualombai, P., S. Prajakwong, N. Aussawatheerakul, Survey Data.” PLoS One 7 (5): e23966. K. Congpoung, S. Sudathip, and others. 2003. “Determining Graz, B., M. Willcox, T. Szeless, and A. Rougemont. 2011. “‘Test Cost-Effectiveness and Cost Component of Three Malaria and Treat’ or Presumptive Treatment for Malaria in High- Diagnostic Models Being Used in Remote Non-Microscope Transmission Situations? A Reflection on the Latest WHO Areas.” Southeast Asian Journal of Tropical Medicine and Guidelines.” Malaria Journal 10: 136. Public Health 34 (2): 322–33. Hansen, K. S., E. Grieve, A. Mikhail, I. Mayan, N. Mohammed, Chanda, P., M. Castillo-Riquelme, and F. Masiye. 2009. and others. 2015. “Cost-Effectiveness of Malaria Diagnosis “Cost-Effectiveness Analysis of the Available Strategies for Using Rapid Diagnostic Tests Compared to Microscopy or Diagnosing Malaria in Outpatient Clinics in Zambia.” Cost Clinical Symptoms Alone in Afghanistan.” Malaria Journal Effectiveness and Resource Allocation 7: 5. 14: 217. Chanda, P., B. Hamainza, H. B. Moonga, V. Chalwe, P. Banda, Kahama-Maro, J., V. D’Acremont, D. Mtasiwa, B. Genton, and and others. 2011. “Relative Costs and Effectiveness of C. Lengeler. 2011. “Low Quality of Routine Microscopy for Treating Uncomplicated Malaria in Two Rural Districts in Malaria at Different Levels of the Health System in Dar es Zambia: Implications for Nationwide Scale-up of Home- Salaam.” Malaria Journal 10: 332. Based Management.” Malaria Journal 10: 159. Lubell, Y., H. Hopkins, C. J. Whitty, S. G. Staedke, and A. Mills. Chimnani, J., J. Kamunyori, E. Bancroft, N. Nazarewicz, 2008. “An Interactive Model for the Assessment of the N. Kisoka, and others. 2010. Tanzania: Review of the Health Economic Costs and Benefits of Different Rapid Diagnostic Facility Report and Request Forms at MSD Zonal Stores. Tests for Malaria.” Malaria Journal 7: 21. Arlington, VA: USAID Deliver Project. Lubell, Y., H. Reyburn, H. Mbakilwa, R. Mwangi, K. Chonya, CPM (Center for Pharmaceutical Management). 2008. and others. 2007. “The Cost-Effectiveness of Parasitologic Accredited Drug Dispensing Outlets in Tanzania: Strategies Diagnosis for Malaria-Suspected Patients in an Era of for Enhancing Access to Medicines Program. Report pre- Combination Therapy.” American Journal of Tropical pared for the Strategies for Enhancing Access to Medicines Medicine and Hygiene 77 (Suppl 6): 128–32. Program. Arlington, VA: Management Sciences for Health. ———. 2008. “The Impact of Response to the Results of D’Acremont, V. 2011. “Understanding and Improving Malaria Diagnostic Tests for Malaria: Cost-Benefit Analysis.” British Diagnosis in Health Facilities in Dar es Salaam, Tanzania.” Medical Journal 336 (7637): 202–5. PhD dissertation, University of Basel. Lubell, Y., A. Riewpaiboon, A. M. Dondorp, L. von Seidlein, D’Acremont, V., C. Lengeler, H. Mshinda, D. Mtasiwa, O. A. Mokuolu, and others. 2011. “Cost-Effectiveness of M. Tanner, and others. 2009. “Time to Move from Parenteral Artesunate for Treating Children with Severe Presumptive Malaria Treatment to Laboratory-Confirmed Malaria in Sub-Saharan Africa.” Bulletin of the World Health Diagnosis and Treatment in African Children with Fever.” Organization 89 (7): 504–12. PLoS Medicine 6 (1): e252. Lubell, Y., S. G. Staedke, B. M. Greenwood, M. R. Kamya, de Oliveira, M. R., A. de Castro Gomes, and C. M. Toscano. M. Molyneux, and others. 2011. “Likely Health Outcomes 2010. “Cost-Effectiveness of OptiMal® Rapid Diagnostic for Untreated Acute Febrile Illness in the Tropics in Test for Malaria in Remote Areas of the Amazon Region, Decision and Economic Models; A Delphi Survey.” PLoS Brazil.” Malaria Journal 9: 277. One 6 (2): e17439. English, M., H. Reyburn, C. Goodman, and R. W. Snow. 2009. Ly, A. B., A. Tall, R. Perry, L. Baril, A. Badiane, and others. “Abandoning Presumptive Antimalarial Treatment for Febrile 2010. “Use of HRP-2-Based Rapid Diagnostic Test for Children Aged Less than Five Years—A Case of Running Plasmodium falciparum Malaria: Assessing Accuracy and before We Can Walk?” PLoS Medicine 6 (1): e1000015. Cost-Effectiveness in the Villages of Dielmo and Ndiop, Fernando, S. D., N. D. Karunaweera, W. P. Fernando, Senegal.” Malaria Journal 9: 153. N. Attanayake, and A. R. Wickremasinghe. 2004. Masanja, I., M. Selemani, B. Amuri, D. Kajungu, R. Khatib, and “A Cost Analysis of the Use of the Rapid, Whole-Blood, others. 2012. “Increased Use of Malaria Rapid Diagnostic Cost-Effectiveness of Strategies for the Diagnosis and Treatment of Febrile Illness in Children 399 Tests Improves Targeting of Anti-Malarial Treatment in http://www.nbs.go.tz/nbs/takwimu/references/Tanzania_in Rural Tanzania: Implications for Nationwide Rollout of _Figures2010.pdf. Malaria Rapid Diagnostic Tests.” Malaria Journal 11 (1): 221. Tanzania NBS (National Bureau of Statistics) and Macro Molyneux, C. S., V. Mung’Ala-Odera, T. Harpham, and International. 2007. Tanzania Service Provision Assessment R. W. Snow. 1999. “Maternal Responses to Childhood Survey 2006. Dar es Salaam: NBS and Macro International. Fevers: A Comparison of Rural and Urban Residents in ———. 2011. Tanzania Demographic and Health Survey Coastal Kenya.” Tropical Medicine and International Health 2010. Dar es Salaam: NBS and Macro International. 4 (12): 836–45. http://www.measuredhs.com/pubs/pdf/FR243/FR243 Mosha, J. F., L. Conteh, F. Tediosi, S. Gesase, J. Bruce, and [24June2011].pdf. others. 2010. “Cost Implications of Improving Malaria Uzochukwu, B. S., E. N. Obikeze, O. E. Onwujekwe, C. A. Onoka, Diagnosis: Findings from North-Eastern Tanzania.” PLoS and U. K. Griffiths. 2009. “Cost-Effectiveness Analysis of One 5 (1): e8707. Rapid Diagnostic Test, Microscopy, and Syndromic Approach Msellem, M. I., A. Martensson, G. Rotllant, A. Bhattarai, in the Diagnosis of Malaria in Nigeria: Implications for J. Stromberg, and others. 2009. “Influence of Rapid Malaria Scaling-up Deployment of ACT.” Malaria Journal 8: 265. Diagnostic Tests on Treatment and Health Outcome in WHO (World Health Organization). 1999. Improving Child Fever Patients, Zanzibar: A Crossover Validation Study.” Health: Integrated Management of Childhood Illnesses; The PLoS Medicine 6 (4): e1000070. Integrated Approach. Geneva: WHO. http://www.who.int Mueller, E. A., M. van Vugt, W. Kirch, K. Andriano, P. Hunt, /child_adolescent_health/documents /chd_97_12_Rev_2 and others. 2006. “Efficacy and Safety of the Six-Dose /en/index.html. Regimen of Artemether-Lumefantrine for Treatment ———. 2006. Guidelines for the Treatment of Malaria. Geneva: of Uncomplicated Plasmodium falciparum Malaria in WHO. Adolescents and Adults: A Pooled Analysis of Individual ———. 2010. Guidelines for the Treatment of Malaria, second Patient Data from Randomized Clinical Trials.” Acta Tropica edition. Geneva: WHO. 100 (1–2): 41–53. ———. 2011. “Estimates of Unit Costs for Patient Services for Patrick Kachur, S., J. Schulden, C. A. Goodman, H. Kassala, B. F. United Republic of Tanzania.” WHO, Geneva. http://www Elling, and others. 2006. “Prevalence of Malaria Parasitemia .who.int/choice/country/tza/cost/en/. among Clients Seeking Treatment for Fever or Malaria at ———. 2012a. International Medicines Price Workbook: Part I. Drug Stores in Rural Tanzania 2004.” Tropical Medicine and Geneva: WHO and Health Action International/Tanzania. International Health 11 (4): 441–51. ———. 2012b. World Malaria Report 2012. Geneva: WHO. Reyburn, H., H. Mbakilwa, R. Mwangi, O. Mwerinde, R. Olomi, http://www.who.int/malaria/publications/world_malaria and others. 2007. “Rapid Diagnostic Tests Compared with _report_2012/wmr2012_full_report.pdf. Malaria Microscopy for Guiding Outpatient Treatment Willcox, M. L., F. Sanogo, B. Graz, M. Forster, F. Dakouo, of Febrile Illness in Tanzania: Randomised Trial.” British and others. 2009. “Rapid Diagnostic Tests for the Home- Medical Journal 334 (7590): 403. Based Management of Malaria, in a High-Transmission Rolland, E., F. Checchi, L. Pinoges, S. Balkan, J. P. Guthmann, Area.” Annals of Tropical Medicine and Parasitology 103 and others. 2006. “Operational Response to Malaria (1): 3–16. Epidemics: Are Rapid Diagnostic Tests Cost-Effective?” Yadav, P., J. L. Cohen, S. Alphs, J. Arkedis, P. S. Larson, and others. Tropical Medicine and International Health 11 (4): 398–408. 2012. “Trends in Availability and Prices of Subsidized ACT Rosas Aguirre, A. M., L. F. Llanos Zavalaga, and M. Trelles de over the First Year of the AMFm: Evidence from Remote Belaunde. 2009. “Cost-Effectiveness Ratio of Using Rapid Regions of Tanzania.” Malaria Journal 11: 1–11. Tests for Malaria Diagnosis in the Peruvian Amazon.” Zanzibar Malaria Control Program and Karolinska Institute. Revista Panamericana de Salud Pública 25 (5): 377–88. 2012. “Malaria in Zanzibar 1999–2011.” Presentation by Shillcutt, S., C. Morel, C. Goodman, P. Coleman, D. Bell, and Anders Björkman, May 22. http://www.infektion.net/sites others. 2008. “Cost-Effectiveness of Malaria Diagnostic /default/files/pdf/malaria_zanzibar_anders_bjorkman.pdf. Methods in Sub-Saharan Africa in an Era of Combination Zikusooka, C. M., D. McIntyre, and K. I. Barnes. 2008. Therapy.” Bulletin of the World Health Organization 86 (2): “Should Countries Implementing an Artemisinin-Based 101–10. Combination Malaria Treatment Policy also Introduce Simba, D. O., M. Warsame, D. Kakoko, Z. Mrango, G. Tomson, Rapid Diagnostic Tests?” Malaria Journal 7: 176. and others. 2010. “Who Gets Prompt Access to Artemisinin- Zurovac, D., B. A. Larson, J. Skarbinski, L. Slutsker, R. W. Snow, Based Combination Therapy? A Prospective Community- and M. J. Hamel. 2008. “Modeling the Financial and Based Study in Children from Rural Kilosa, Tanzania.” PLoS Clinical Implications of Malaria Rapid Diagnostic Tests One 5 (8): e12104. in the Case-Management of Older Children and Adults in Tanzania NBS (National Bureau of Statistics). 2011. “Tanzania in Kenya.” American Journal of Tropical Medicine and Hygiene Figures 2010.” NBS, Ministry of Finance, Dar es Salaam, June. 78 (6): 884–91. 400 Major Infectious Diseases Chapter 16 Viral Hepatitis Stefan Z. Wiktor INTRODUCTION protection from hepatitis B infection by immunization also prevents hepatitis D. Viral hepatitis is caused by five distinct viruses (hepati- Hepatitis B and C chronic infections can be treated tis A, B, C, D, and E), which have different routes of effectively. The new direct acting antiviral (DAA) medi- transmission and varying courses of disease (table 16.1). cines for hepatitis C can cure more than 90 percent of those According to the Global Health Estimates, deaths from with chronic infection with a two to three month course of acute and chronic hepatitis in 2012 were the tenth treatment. Hepatitis C treatment could also reduce hepati- leading cause of death and the sixteenth leading cause tis C transmission because people who have been cured do of disability. In 2013, an estimated 1.45 million per- not transmit the infection. There is no cure for chronic sons (95 percent uncertainty interval 1.38 million to hepatitis B, but effective antiviral treatments can suppress 1.54 million) died from viral hepatitis; this estimate viral replication and prevent disease progression. includes deaths due to acute hepatitis, as well as hepatitis-related liver cancer and cirrhosis (Stanaway and others 2016). Furthermore, while deaths from INCIDENCE, PREVALENCE, AND infectious diseases such as HIV/AIDS, malaria, and DISTRIBUTION tuberculosis are decreasing, deaths from hepatitis increased by 63 percent between 1990 and 2013. Most Because infection with the hepatitis viruses is often (96 percent) hepatitis deaths are caused by hepatitis B asymptomatic, there are no reliable estimates of the virus (HBV) and hepatitis C virus (HCV)—these two incidence of acute and chronic viral hepatitis. An esti- viruses cause chronic, lifelong infection resulting in mated 250 million people live with chronic hepatitis B progressive liver damage leading to cirrhosis and hepa- infection; 80 million have chronic hepatitis C infection tocellular carcinoma (figure 16.1). (Gower and others 2014; Schweitzer and others 2015). The burden of hepatitis infection is not equally The prevalence of hepatitis B and C varies consider- distributed globally. Mortality rates from hepatitis are ably in different regions. The areas of highest preva- highest in West Africa and parts of Asia; in absolute lence for hepatitis B are West Africa, where in some numbers, East Asia and South Asia account for the great- countries more than 8 percent of the population is est number of people dying from hepatitis—51 percent infected, and East and Central Asia (map 16.1). For of the total number of deaths. hepatitis C infection, the regions with the highest prev- Effective interventions exist to prevent transmission alence are West and Central Africa, Eastern Europe, of viral hepatitis (table 16.2). Safe and effective vaccines and Central Asia. The prevalence of hepatitis C infection have been developed to prevent hepatitis A, B, and E, and is extremely high in a few countries, most notably, Corresponding author: Stefan Z. Wiktor, School of Public Health, University of Washington, Seattle, Washington, United States; wiktors@uw.edu. 401 Table 16.1 Characteristics of Main Types of Viral Hepatitis Infections Hepatitis B with or Hepatitis A without hepatitis D Hepatitis C Hepatitis E Mode of transmission Contaminated food or water Blood, sex, mother-to-child Contaminated food or water Undercooked pork and pork liver Number of chronic 0 Approximately 250 million Approximately 80 Very few and mainly in immuno- infections million suppressed persons Disease outcomes Fulminant hepatitis Fulminant hepatitis; cirrhosis and hepatocellular Fulminant hepatitis; maternal carcinoma mortality Figure 16.1 Number of Deaths per Year Due to Hepatitis, HIV/AIDS, the Arab Republic of Egypt and Pakistan, where high Tuberculosis, and Malaria, 2000–15 incidence continues, largely the result of weak preven- tion measures, such as reuse of syringes and needles in 2.5 health care settings (Ahmed and others 2013; Mostafa and others 2010). 2.0 Deaths (millions) 1.5 NATURAL HISTORY AND MORTALITY FROM 1.0 VIRAL HEPATITIS The natural history of the hepatitis viruses can be cate- 0.5 gorized based on whether they cause chronic infection. All hepatitis viruses cause acute hepatitis, which can 0 result in fulminant hepatitis in rare cases and may be 2000 2005 2010 2015 fatal. Hepatitis B and hepatitis C also can cause chronic Hepatitis HIV Malaria Tuberculosis infection. Hepatitis D is an incomplete virus that can Sources: Hepatitis: GBD 2013 Mortality and Causes of Death Collaborators 2015; HIV: UNAIDS only replicate and cause infection in the presence of 2015; Malaria: WHO and UNICEF 2015; Tuberculosis: WHO 2015a. hepatitis B. Note: HIV/AIDS = human immunodeficiency virus/acquired immune deficiency syndrome. The risk of developing chronic hepatitis B infection depends on the age at infection; it declines from more than 90 percent among children infected during the first Table 16.2 Elements of a Comprehensive Hepatitis Prevention year to 20 percent to 30 percent for children infected Program between the ages of 1 and 5 years, and 6 percent for chil- dren ages 5 to 15 years (Edmunds and others 1993). For Hepatitis virus hepatitis C, approximately one-third of individuals will Prevention intervention types spontaneously clear infection, while the remaining two- Safe water and food A, E thirds will develop chronic infection. For both viruses, Hepatitis A vaccination according to the country’s A chronic infection is marked by continued replication of epidemiological situation the virus in the liver, which can lead to cirrhosis, hepato- Hepatitis B vaccination for all children and B, D cellular carcinoma, or both (Chen, Iloeje, and Yang 2007; administration of birth dose; vaccination of Thein and others 2008). These two conditions account health care workers and at-risk adults for more than 90 percent of all deaths due to viral hepa- Access to safe blood (universal screening of all blood B, C, D titis (figure 16.2) (Stanaway and others 2016). Chronic donations in a quality-assured manner) hepatitis B or hepatitis C infections cause an estimated Access to sterile injections and other invasive medical B, C, D 78 percent of all liver cancer and 57 percent of all liver equipment in formal and informal health settings cirrhosis (Perz and others 2006). Because of the higher prevalence of hepatitis B and hepatitis C in Asia and Access to sterile injection equipment and other harm- B, C, D Sub-Saharan Africa, countries in these regions, which reduction measures for people who inject drugs are least able to deal with these diseases, experience the Promotion of safe sex practices A, B, C highest rates of death due to viral hepatitis. 402 Major Infectious Diseases Map 16.1 Hepatitis Mortality Rates and Virus Distribution, by Global Burden of Disease Region, 2013 IBRD 42553 | DECEMBER 2016 Mortality rate Proportion attibutable (per 100,000 per year) to each virus, 2013 >10.00 Hepatitis A virus 10.00–14.99 Hepatitis B virus 15.00–22.49 Hepatitis C virus 22.50–33.49 Hepatitis E virus ≥33.50 Source: Stanaway and others 2016. Figure 16.2 Number of Hepatitis-Related Deaths, by Virus TRANSMISSION OF HEPATITIS VIRUSES Type, 2013 Hepatitis A and hepatitis E are transmitted through the fecal-oral route by contact with contaminated food 800 or water. Hepatitis E is also a zoonotic infection that Number of deaths (thousands) can be spread by eating undercooked or uncooked 600 pork or deer meat (Kamar and others 2012). Sexual transmission of hepatitis A by frequent oral-anal con- 400 tact is common among men who have sex with men (Jin and others 2007). Hepatitis B, C, and D are trans- 200 mitted through blood and bodily fluids. Globally, most hepatitis B infections occur through mother-to-child 0 and early-life horizontal transmission between family HAV HBV HCV HEV members. Among adults, transmission occurs through Hepatocellular carcinoma Cirrhosis Acute sexual intercourse, as well as through unsafe injection practices and transfusion of unscreened blood and Source: Stanaway and others 2016. Note: HAV = hepatitis A virus; HBV = hepatitis B virus; HCV = hepatitis C virus; sexual transmission (Goldstein and others 2005). Most HEV = hepatitis E virus. infections with HCV occur through unsafe injections, Viral Hepatitis 403 either in medical settings from reuse of medical equip- depends on the socioeconomic status of a country. In ment and substandard application of infection control countries where sanitation practices are improving as measures, or through unsafe practices among people a result of improved socioeconomic conditions, many who inject drugs. Sexual transmission of hepatitis C is children escape hepatitis A infection, thus leaving rare in heterosexual couples but more common among them susceptible when they become adults. In these HIV/AIDS-infected men who have sex with men countries, wide-scale hepatitis A vaccination is likely to (Tohme and Holmberg 2010). be cost-effective and is encouraged (Jacobsen and Wiersma 2010). INTERVENTIONS Hepatitis B Table 16.2 summarizes the effective interventions for The most notable achievement in hepatitis prevention is preventing the transmission of viral hepatitis. the reduction in incidence of acute and chronic hepatitis B infection as a result of universal childhood hepatitis B vaccination. At the end of 2013, 183 out of 194 countries Sanitation had introduced universal childhood vaccination; global Hepatitis A and E infections can be prevented through coverage with three doses of hepatitis B vaccine is esti- improved sanitation. Although no reliable estimates are mated to be 81 percent (figure 16.3) (WHO 2015b). available, it is likely that the incidence of hepatitis A Universal infant vaccination with high coverage levels and E has declined as part of the overall reduction in has led to major reductions in the prevalence of chronic the number of deaths due to diarrhea. Between 1990 hepatitis B infection among children. In China, the and 2012, all regions experienced major declines in the prevalence of chronic hepatitis B infection declined annual number of diarrheal diseases attributable to from approximately 8 percent in 1992 to 1 percent in inadequate water, sanitation, and hygiene, from 2006 among children ages one to four years (Liang and 1.8 million to 842,000 (WHO 2014). others 2009). The World Health Organization (WHO) Western Pacific Region, which includes China and sev- eral other high-prevalence countries, is on track to reach Vaccination its goal of reducing the prevalence of chronic hepatitis B Hepatitis A infection to less than 1 percent by 2017 (Wiesen, An effective hepatitis A vaccine exists, and 18 countries Diorditsa, and Li 2016). Countries where hepatitis B have introduced universal childhood hepatitis A vacci- vaccine coverage has been high for several decades have nation. Whether universal vaccination is appropriate noted reductions in death rates from hepatocellular car- cinoma (Hung and others 2015). Challenges remain to further reductions in incidence. Figure 16.3 Global Coverage of Childhood Hepatitis Immunization and Full protection for children requires that they receive Birth Dose, 2000–13 the first vaccine dose within 24 hours of birth; this dose is 100 termed a birth dose. Of 194 countries, 94 have introduced 90 2013 Global coverage the birth dose in the vaccine schedule, and an estimated HB3 = 80% 38 percent of children globally receive this early dose 80 Birth dose = 38% (WHO 2015c). There are many logistical challenges to the 70 Percent coverage delivery of the birth dose, including that most births occur 60 at home in some countries. Treating pregnant women who 50 have a high viral load of hepatitis B with antiviral medica- 40 tions during pregnancy can prevent mother-to-child 30 transmission of hepatitis B (Pan and others 2016). 20 10 Hepatitis E 0 Outbreaks of hepatitis E occur primarily in Africa and South-East Asia through contaminated drinking water. 01 00 02 03 04 06 09 10 11 13 05 07 08 12 20 20 20 20 20 20 20 20 20 20 20 20 20 20 An effective vaccine has been developed against one of HB3 Birth dose the genotypes of hepatitis E. However, its effectiveness in Source: WHO 2015b. outbreak settings and among children has not been Note: HB3 = three doses of hepatitis B vaccine. assessed. For these reasons, the WHO has not made a 404 Major Infectious Diseases recommendation for its use in national vaccination pro- drug use have been most successful in addressing the grams (WHO 2015c). epidemics of injecting drug use, hepatitis, and HIV/AIDS (Palmateer and others 2014). Despite solid public health evidence demonstrating the effectiveness of harm Safety of Health Care Injections reduction interventions, many decision makers remain In 2000, unsafe health care injections accounted for an reluctant to implement or scale up such programs estimated 32 percent of hepatitis B infections, 40 because of their controversial nature. Ongoing dedicated percent of hepatitis C infections, and 5 percent of HIV advocacy efforts are still needed to initiate and sustain infections acquired in low- and middle-income coun- harm reduction programs across the globe. tries (LMICs) (Hauri, Armstrong, and Hutin 2004). An updated analysis estimated that between 2000 and 2010, the number of infections due to unsafe injections Access to Treatment declined by 91 percent for hepatitis B and 83 percent for Even a comprehensive hepatitis prevention program can- hepatitis C, primarily attributable to a reduction in the not prevent deaths occurring among persons already reuse of injection equipment (Pepin and others 2014). infected. Fortunately, effective treatment exists to treat Addressing health care–associated infections, partic- persons with chronic hepatitis B and C infections. HBV ularly unsafe injections, is difficult because in many replication can be suppressed, and therapy is lifelong in countries the health system is fragmented and poorly most cases (Yapali, Talaat, and Lok 2014). By effectively regulated. In some areas, the health care system overuses suppressing the replication of HBV, hepatitis B treatment injections because of the perception that injected med- can result in a reversal of cirrhosis (Marcellin and others icines are more effective than oral medicines. The 2013). Hepatitis C can be cured in as little as eight weeks approach to reducing this risk needs to combine raising by providing treatment with DAA medications. Persons awareness about the effectiveness of oral medicines with cured of hepatitis C have improved quality of life and the introduction of needles and syringes with reuse- lower rates of liver-related and overall mortality (Smith- prevention features (WHO 2015d). Palmer, Cerri, and Valentine 2015). Despite the availability of these medicines, very few people, particularly in LMICs, are treated. The high price Safety of Blood Supply of the medicines is an important factor. Prices for DAAs in The risk of transmission of hepatitis B and C viruses high-income countries (HICs) generally exceed US$50,000 through the transfusion of unsafe blood can be dra- per person. Through price negotiations, middle-income matically lowered by appropriate selection of donors countries such as Brazil are able to obtain the same medi- and universal, quality-assured testing of blood dona- cines for approximately US$7,000 (Andrieux-Meyer and tions. Many countries, however, still collect a signifi- others 2015). In low-income countries, through the intro- cant proportion of the blood for transfusion from duction of generic medications, the price to treat one donors who have a high prevalence of hepatitis infec- person is less than US$500. Hepatitis B medicines tion, for example, paid donors, and not all countries are available in generic formulations and prices can be as screen 100 percent of their blood collections for hepa- low as US$4/month. Other important challenges to scal- titis B or C. Accordingly, the risk of transfusion-related ing up are the delay in drug registration of the DAAs and transmission of these viruses remains unacceptably the lack of a workforce trained in hepatitis treatment. An high in these settings (WHO 2011). additional important barrier is that most persons with chronic hepatitis infection are still undiagnosed and lack the opportunities to be tested because of the low availabil- Harm Reduction Programs ity, high complexity, and high cost of diagnostics. People who inject drugs are the group most highly affected by viral hepatitis. Globally, an estimated 67 percent of people who inject drugs have evidence of COSTS, COST-EFFECTIVENESS, AND hepatitis C infection and 8 percent have hepatitis B EXTENDED COST-EFFECTIVENESS ANALYSES (Nelson and others 2011). An effective package of inter- vention, termed harm reduction, has been developed and RESULTS includes the provision of sterile injecting equipment and The quality and availability of data on the cost- opioid substitution therapy (WHO, UNODC, and effectiveness of the interventions to prevent and treat UNAIDS 2009). Countries that have provided these hepatitis infection vary by intervention and are primar- services as part of a public health response to injecting ily limited to HICs (table 16.3). Viral Hepatitis 405 Table 16.3 Cost-Effectiveness of Interventions for Hepatitis Intervention Cost-effectiveness and affordability Sources Vaccination, hepatitis A • Universal childhood vaccination is very cost-effective in UMICs and Anonychuk and others 2008 more cost-effective than targeted approaches. • Depending on incidence of hepatitis A infection and vaccination approach, vaccination is sometimes cost-effective or not cost-effective. Vaccination, hepatitis B • Universal childhood vaccination is very cost-effective in LMICs and Miller and McCann 2000 HICs and can be cost saving depending on level of incidence and cost of vaccine administration. Interventions to reduce • Combination of patient and health care worker education and provision Dziekan and others 2003 unnecessary and unsafe injections of single-use injection equipment is very cost-effective. Needle- and syringe-exchange • Limited data are available and only for HICs, with a range of cost- Kwon and others 2012; programs among people who effectiveness results from not cost-effective to highly cost-effective. Pollack 2001 inject drugs Screening blood transfusion for • No data are available on cost-effectiveness related to hepatitis B or C Creese and others 2002 hepatitis B, C infection. • Screening is highly cost-effective in preventing HIV/AIDS infections in Sub-Saharan Africa so could be for hepatitis, too. Hepatitis B treatment • Treatment is cost-effective in HICs and MICs. Buti and others 2013; Toy, • Treatment can be cost saving assuming access to low-cost, generic Hutton, and So 2015 medicines. Hepatitis C treatment • Treatment is cost-effective in HICs, despite very high cost of treatment. Younossi and Henry 2014 • Few studies have been conducted in LMICs. • Treatment is likely to be cost-effective in LICs and some MICs that have negotiated favorable medicine prices or have access to inexpensive generic medicines. Note: HICs = high-income countries; LICs = low-income countries; LMICs = low- and middle-income countries; MICs = middle-income countries; UMICs = upper-middle-income countries. Definition of very cost-effective and cost-effective follows: If cost per disability-adjusted life year averted is less than the per capita income of the country, the intervention is very cost-effective; if less than three times the per capita income, it is cost-effective. Immunization with no vaccination. The only vaccine with greater Hepatitis B childhood immunization is the most estimated impact is that for measles (WHO 2013). cost-effective of the interventions to prevent infec- Assessing the cost-effectiveness of hepatitis A vac- tion, given the high morbidity associated with chronic cination is complicated because the risk of symptom- hepatitis B infection and the low cost of vaccination. atic disease is linked with increasing age at time of The cost-effectiveness of vaccination is directly linked infection. In countries where nearly the entire popula- to the population prevalence of infection; according tion is infected in childhood, when most infections are to one analysis, the cost per year of life saved ranged asymptomatic, there is very little morbidity, and from US$6 to US$51 in LMICs and upper-middle- thereby cost, associated with hepatitis A infection. In income countries but was much higher (US$8,712– countries where the incidence of infection is lower, US$12,197) in HICs because of higher administration analyses show that universal childhood vaccination is costs and lower prevalence of hepatitis B infection. more cost-effective than targeted approaches. Hepatitis In certain country-specific analyses, for example, in A vaccination of adults is less cost-effective than for China, universal hepatitis B vaccination is cost saving children. In a systematic review, 64 percent of studies as is catch-up vaccination of children and adolescents assessing universal vaccination in children had cost- (Hutton, So, and Brandeau 2010; Lu and others 2013). effectiveness ratios of less than US$28,606 per year of An estimated 5.3 million to 6.0 million future hepati- life saved or quality-adjusted life year (QALY) gained, tis B–related deaths could be averted if 90 percent of compared with only 29 percent of studies in adults children in 94 LMICs were vaccinated, compared (Anonychuk and others 2008). 406 Major Infectious Diseases Injection-Associated Infections of liver-related deaths and 26 percent to 31 percent of Only limited data are available on the economic aspects of hepatocellular carcinoma deaths; treatment would be the prevention of health care–associated hepatitis cost saving at a drug price of less than US$32–US$75 infection. An analysis of the cost-effectiveness of measures a month (Toy, Hutton, and So 2015). to reduce unnecessary and unsafe injections resulted in a cost-effectiveness ratio between US$20.02 and US$3,280 CONCLUSIONS per disability-adjusted life year (DALY) averted, depend- ing on the region (Dziekan and others 2003). The arrival of highly effective treatments for hepatitis B Needle- and syringe-exchange programs for people and C and an improved understanding of the attendant who inject drugs are shown to be effective in preventing burden of disease have resulted in a call for increased HIV/AIDS, but with regard to the effectiveness of these action to eliminate hepatitis. In view of the differences in programs in the prevention of hepatitis C infection, the the distribution of hepatitis viruses and the related burden data are more limited and only from HICs. One analysis of disease, policy makers will need to develop national concluded that the cost to prevent one hepatitis C infec- plans that are tailored to their epidemiologic patterns tion was between US$357,586 and US$1.4 million; a and response capacity. The goals of the United Nations study from Australia assessing the cost-effectiveness of 2030 Agenda for Sustainable Development include a harm reduction to prevent hepatitis C and HIV/AIDS goal to combat the epidemic of hepatitis (UN 2015). The infections found that a needle- and syringe-exchange WHO has developed a global hepatitis strategy to elimi- program resulted in US$801–US$16,840 per QALY nate hepatitis B and C as major public health threats by gained (Kwon and others 2012; Pollack 2001). 2030 (WHO 2016). The means to achieve these goals exist. What is needed is increased advocacy and country-level investment and action. Treatment The high cost of the new hepatitis C medicines has engendered considerable controversy. Economic analy- NOTE ses concerning treatment have yielded widely varying World Bank Income Classifications as of July 2014 are as fol- results, partly because of differences in the assessed treat- lows, based on estimates of gross national income (GNI) per ment regimens; virus genotypes; and types of patient capita for 2013: populations, for example, those undergoing treatment for the first time versus those who failed previous treat- • Low-income countries (LICs) = US$1,045 or less ment. Most of the analyses show that treatment with • Middle-income countries (MICs) are subdivided: DAAs for hepatitis C genotype 1 regardless of the stage (a) lower-middle-income = US$1,046 to US$4,125 of liver disease is highly cost-effective in the United (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 States, with an incremental cost-effectiveness ratio of less • High-income countries (HICs) = US$12,746 or more. than US$71,517 (Tice and others 2015). Few studies have assessed the cost-effectiveness of the treatments in LMICs. These analyses are complicated by the varying REFERENCES and rapidly changing prices of DAAs in these countries Ahmed, B., T. Ali, H. Qureshi, and S. Hamid. 2013. (Andrieux-Meyer and others 2015). “Population-Attributable Estimates for Risk Factors The economics of treatment for hepatitis B differ Associated with Hepatitis B and C: Policy Implications for from those for hepatitis C. Although the price of medi- Pakistan and Other South Asian Countries.” Hepatology cines is much lower on a per-day basis, treatment is Internatonal 7 (2): 500–507. lifelong for most patients with HBV infection; however, Andrieux-Meyer, I., J. Cohn, E. S. de Araujo, and S. S. Hamid. analyses show hepatitis B treatment to be cost-effective. 2015. “Disparity in Market Prices for Hepatitis C Virus In a systematic review, most of the studies conducted in Direct-Acting Drugs.” The Lancet Global Health 3 (11): HICs showed that treating compared to not treating e676–77. doi:10.1016/S2214-109X(15)00156-4. Anonychuk, A. M., A. C. Tricco, C. T. Bauch, B. Pham, V. Gilca, resulted in an incremental cost-effectiveness ratio of less and others. 2008. “Cost-Effectiveness Analyses of Hepatitis than US$70,000 per QALY gained (Buti and others A Vaccine: A Systematic Review to Explore the Effect of 2013). Several studies have assessed the economics of Methodological Quality on the Economic Attractiveness of treatment in China, which has the largest number of Vaccination Strategies.” Pharmacoeconomics 26 (1): 17–32. persons with chronic hepatitis B infection. In one study, Buti, M., I. Oyaguez, V. Lozano, and M. A. Casado. 2013. entecavir or tenofovir monotherapy treatment in non- “Cost Effectiveness of First-Line Oral Antiviral cirrhotic patients would prevent 49 percent to 69 percent Therapies for Chronic Hepatitis B: A Systematic Review.” Viral Hepatitis 407 Pharmacoeconomics 31 (1): 63–75. doi:10.1007/s40273-012 Kwon, J. A., J. Anderson, C. C. Kerr, H. H. Thein, L. Zhang, and -0009-2. others. 2012. “Estimating the Cost-Effectiveness of Needle- Chen, C. J., U. H. Iloeje, and H. I. Yang. 2007. “Long-Term Syringe Programs in Australia.” AIDS 26 (17): 2201–10. Outcomes in Hepatitis B: The REVEAL-HBV Study.” Clinics doi:10.1097/QAD.0b013e3283578b5d. in Liver Disease 11 (4): 797–816, viii. Liang, X., S. Bi, W. Yang, L. Wang, G. Cui, and others. 2009. Creese, A., A. Floyd, A. Alban, and L. Guinness. 2002. “Cost- “Epidemiological Serosurvey of Hepatitis B in China: Effectiveness of HIV/AIDS Interventions in Africa: A Declining HBV Prevalence Due to Hepatitis B Vaccination.” Systematic Review of the Evidence.” The Lancet 359 (9318): Vaccine 27 (47): 6550–57. doi:10.1016/j.vaccine.2009.08.048. 1635–43. Lu, S. Q., S. M. McGhee, X. Xie, J. Cheng, and R. Fielding. 2013. Dziekan, G., D. Chisholm, B. Johns, J. Rovira, and Y. J. Hutin. “Economic Evaluation of Universal Newborn Hepatitis B 2003. “The Cost-Effectiveness of Policies for the Safe and Vaccination in China.” Vaccine 31 (14): 1864–69. doi:10 Appropriate Use of Injection in Healthcare Settings.” .1016/j.vaccine.2013.01.020. Bulletin of the World Health Organization 81 (4): 277–85. Marcellin, P., E. Gane, M. Buti, N. Afdhal, W. Sievert, and Edmunds, W. J., G. F. Medley, D. J. Nokes, A. J. Hall, and H. C. others. 2013. “Regression of Cirrhosis during Treatment Whittle. 1993. “The Influence of Age on the Development with Tenofovir Disoproxil Fumarate for Chronic Hepatitis of the Hepatitis B Carrier State.” Proceedings of the B: A 5-Year Open-Label Follow-Up Study.” The Lancet 381 Royal Society: Biological Sciences 253 (1337): 197–201. (9865): 468–75. doi:10.1016/S0140-6736(12)61425-1. doi: 10.1098/rspb.1993.0102. Miller, M. A., and L. McCann. 2000. “Policy Analysis of GBD 2013 Mortality and Causes of Death Collaborators. 2015. the Use of Hepatitis B, Haemophilus influenzae Type b-, “Global, Regional, and National Age-Sex Specific All-Cause Streptococcus pneumoniae-Conjugate and Rotavirus and Cause-Specific Mortality for 240 Causes of Death, Vaccines in National Immunization Schedules.” Health 1990–2013: A Systematic Analysis for the Global Burden Economics 9 (1): 19–35. of Disease Study 2013.” The Lancet 385 (9963): 117–71. Mostafa, A., S. M. Taylor, M. el-Daly, M. el-Hoseiny, I. Bakr, doi:10.1016/S0140-6736(14)61682-2. and others. 2010. “Is the Hepatitis C Virus Epidemic Over Goldstein, S. T., F. Zhou, S. C. Hadler, B. P. Bell, E. E. Mast, and in Egypt? Incidence and Risk Factors of New Hepatitis others. 2005. “A Mathematical Model to Estimate Global C Virus Infections.” Liver International 30 (4): 560–66. Hepatitis B Disease Burden and Vaccination Impact.” doi:10.1111/j.1478-3231.2009.02204.x. International Journal of Epidemiology 34 (6): 1329–39. Nelson, P. K., B. M. Mathers, B. Cowie, H. Hagan, D. Des doi:10.1093/ije/dyi206. Jarlais, and others. 2011. “Global Epidemiology of Hepatitis Gower, E., C. Estes, S. Blach, K. Razavi-Shearer, and H. Razavi. B and Hepatitis C in People Who Inject Drugs: Results 2014. “Global Epidemiology and Genotype Distribution of of Systematic Reviews.” The Lancet 378 (9791): 571–83. the Hepatitis C Virus Infection.” Journal of Hepatology 61 doi:10.1016/S0140-6736(11)61097-0. (Suppl 1): S45–57. doi:10.1016/j.jhep.2014.07.027. Palmateer, N. E., A. Taylor, D. J. Goldberg, A. Munro, Hauri, A. M., G. L. Armstrong, and Y. J. Hutin. 2004. “The Global C. Aitken, and others. 2014. “Rapid Decline in HCV Burden of Disease Attributable to Contaminated Injections Incidence among People Who Inject Drugs Associated Given in Health Care Settings.” International Journal of STD with National Scale-Up in Coverage of a Combination of and AIDS 15 (1): 7–16. doi:10.1258/095646204322637182. Harm Reduction Interventions.” PLoS One 9 (8): e104515. Hung, G. Y., J. L. Horng, H. J. Yen, C. Y. Lee, and L. Y. Lin. 2015. doi:10.1371/journal.pone.0104515. “Changing Incidence Patterns of Hepatocellular Carcinoma Pan, C. Q., Z. Duan, E. Dai, S. Zhang, G. Han, and others. 2016. among Age Groups in Taiwan.” Journal of Hepatology 63 (6): “Tenofovir to Prevent Hepatitis B Transmission in Mothers 1390–96. doi:10.1016/j.jhep.2015.07.032. with High Viral Load.” New England Journal of Medicine Hutton, D. W., S. K. So, and M. L. Brandeau. 2010. “Cost- 374 (24): 2324–34. doi:10.1056/NEJMoa1508660. Effectiveness of Nationwide Hepatitis B Catch-Up Pepin, J., C. N. Abou Chakra, E. Pepin, V. Nault, and L. Valiquette. Vaccination among Children and Adolescents in China.” 2014. “Evolution of the Global Burden of Viral Infections Hepatology 51 (2): 405–14. doi:10.1002/hep.23310. from Unsafe Medical Injections, 2000–2010.” PLoS One Jacobsen, K. H., and S. T. Wiersma. 2010. “Hepatitis A 9 (6): e99677. doi:10.1371/journal.pone.0099677. Virus Seroprevalence by Age and World Region, 1990 Perz, J. F., G. L. Armstrong, L. A. Farrington, Y. J. Hutin, and and 2005.” Vaccine 28 (41): 6653–57. doi:10.1016/j B. P. Bell. 2006. “The Contributions of Hepatitis B Virus .vaccine.2010.08.037. and Hepatitis C Virus Infections to Cirrhosis and Primary Jin, F., G. P. Prestage, I. Zablostka, R. Rawstone, S. C. Kippax, Liver Cancer Worldwide.” Journal of Hepatology 45 (4): and others. 2007. “High Rates of Sexually Transmitted 529–38. doi:10.1016/j.jhep.2006.05.013. Infections in HIV Positive Homosexual Men: Data from Pollack, H. A. 2001. “Cost-Effectiveness of Harm Reduction Two Community Based Cohorts.” Sexually Transmitted in Preventing Hepatitis C among Injection Drug Users.” Infections 83: 397–99. Medical Decision Making 21 (5): 357–67. Kamar, N., R. Bendall, F. Legrand-Abravanel, N. S. Xia, S. Ijaz, Schweitzer, A., J. Horn, R. T. Mikolajczyk, G. Krause, and and others. 2012. “Hepatitis E.” The Lancet 379 (9835): J. J. Ott. 2015. “Estimations of Worldwide Prevalence of 2477–88. doi:10.1016/S0140-6736(11)61849-7. Chronic Hepatitis B Virus Infection: A Systematic Review 408 Major Infectious Diseases of Data Published between 1965 and 2013.” The Lancet 386 ———. 2013. Global Vaccine Action Plan 2011–2020. Geneva: (10003): 1546–55. doi:10.1016/S0140-6736(15)61412-X. WHO. Smith-Palmer, J., K. Cerri, and W. Valentine. 2015. “Achieving ———. 2014. Preventing Diarrhoea through Better Water, Sustained Virologic Response in Hepatitis C: A Systematic Sanitation and Hygiene Exposures and Impacts in Low- and Review of the Clinical, Economic and Quality of Life Middle-Income Countries. Geneva: WHO. http://www.who Benefits.” BMC Infectious Diseases 15: 19. doi:10.1186 .int/water_sanitation_health/gbd_poor_water/en/. /s12879-015-0748-8. ———. 2015a. Global Tuberculosis Report 2015. Geneva: WHO. Stanaway, J. D., A. D. Flaxman, M. Naghavi, C. Fitzmaurice, http://www.who.int/tb/publications/global_report/en/. T. Vos, and others. 2016. “The Global Burden of Viral ———. 2015b. “WHO–UNICEF Estimate of HepB3 Coverage.” Hepatitis from 1990 to 2013: Findings from the Global Web page. http://apps.who.int/immunization_monitoring Burden of Disease Study 2013.” The Lancet. doi:10.1016 /globalsummary/timeseries/tswucoveragehepb3.html. /S0140-6736(16)30579-7. ———. 2015c. “Hepatitis E Vaccine: WHO Position Paper. Thein, H. H., Q. Yi, G. J. Dore, and M. D. Krahn. 2008. May 2015.” Weekly Epidemiological Record 90 (18): “Estimation of Stage-Specific Fibrosis Progression Rates in 185–200. Chronic Hepatitis C Virus Infection: A Meta-Analysis and ———. 2015d. “Who Guideline on the Use of Safety-Engineered Meta-Regression.” Hepatology 48 (2): 418–31. doi:10.1002 Syringes for Intramuscular, Intradermal and Subcutaneous /hep.22375. Injections in Health-Care Settings.” WHO, Geneva. http:// Tice, J. A., D. A. Ollendorf, H. S. Chahal, J. G. Kahn, E. Marseille, www.who.int/injection_safety/global-campaign/injection and others. 2015. The Comparative Clinical Effectiveness and -safety_guidline.pdf. Value of Novel Combination Therapies for the Treatment ———. 2016. WHO Global Health Sector Strategy on Viral of Patients with Genotype 1 Chronic Hepatitis C Infection. Hepatitis: Towards Ending Viral Hepatitis. Geneva: WHO. A Technology Assessment Final Report. Boston, MA: http://apps.who.int/iris/bitstream/10665/246177/1/WHO Institute for Clinical and Economic Review. -HIV-2016.06-eng.pdf?ua=1. Tohme, R. A., and S. D. Holmberg. 2010. “Is Sexual Contact WHO and UNICEF (United Nations Children’s Fund). a Major Mode of Hepatitis C Virus Transmission?” 2015. Achieving the Malaria MDG Target: Reversing the Hepatology 52 (4): 1497–505. doi:10.1002/hep.23808. Incidence of Malaria 2000–2015. Geneva: WHO and Toy, M., D. W. Hutton, and S. K. So. 2015. “Cost-Effectiveness and UNICEF. Cost Thresholds of Generic and Brand Drugs in a National WHO, UNODC (United Nations Office on Drugs and Crime), Chronic Hepatitis B Treatment Program in China.” PLoS One and UNAIDS. 2009. Technical Guide for Countries to Set 10 (11): e0139876. doi:10.1371/journal.pone.0139876. Targets for Universal Access to HIV Prevention, Treatment UN (United Nations). 2015. “United Nations General and Care for Injecting Drug Users. Geneva: WHO, Regional Assembly Resolution A/RES/70/1 – Transforming Our Office for Europe, UNAIDS, UNODC. World: The 2030 Agenda for Sustainable Development.” Wiesen, E., S. Diorditsa, and X. Li. 2016. “Progress towards United Nations, New York. http://www.un.org/ga/search Hepatitis B Prevention through Vaccination in the Western /view_doc.asp?symbol=A/RES/70/1&Lang=E. Pacific, 1990–2014.” Vaccine 34 (25): 2855–62. doi: 10.1016 UNAIDS (Joint United Nations Programme on HIV/AIDS). /j.vaccine.2016.03.060. 2015. How AIDS Changed Everything. MDG 6: 15 Years, Yapali, S., N. Talaat, and A. S. Lok. 2014. “Management of Hepatitis 15 Lessons of Hope from the AIDS Response. Geneva: B: Our Practice and How It Relates to the Guidelines.” Clinical UNAIDS. http://www.unaids.org/en/resources/campaigns Gastroenterology and Hepatology 12 (1): 16–26. /HowAIDSchangedeverything. Younossi, Z., and L. L. Henry. 2014. “The Impact of the New WHO (World Health Organization). 2011. Global Database Antiviral Regimens on Patient Reported Outcomes and on Blood Safety. Blood Transfusion Safety [web page]. Health Economics of Patients with Chronic Hepatitis C.” WHO, Geneva. Digestive and Liver Disease 46 (Suppl 5): S186–96. Viral Hepatitis 409 Chapter 17 An Investment Case for Ending Neglected Tropical Diseases Christopher Fitzpatrick, Uzoma Nwankwo, Edeltraud Lenk, Sake J. de Vlas, and Donald A. P. Bundy Box 17.1 Key Messages • Neglected tropical diseases (NTDs) together treatment and vector control combined require account for a significant and inequitably distrib- less than 0.1 percent of domestic health spending. uted global disease burden, similar in order of Domestic value for money is enhanced by the magnitude to those of tuberculosis or malaria at unprecedented scale of the London Declaration approximately 22 million disability-adjusted life- donation of medicines for nine of the most prev- years (DALYs) in 2012. alent NTDs. • Cost-effective interventions to end NTDs are • Reaching those targets could avert an estimated available for as little as US$3 per DALY averted; 519 million DALYs from 2015 to 2030, compared these interventions reach the poorest and most to 1990 and the beginning of concerted efforts to marginalized populations and provide an inte- control NTDs. grated approach to treat multiple diseases. • The benefit to affected individuals in terms of • Ambitious eradication, elimination, and control averted out-of-pocket health expenditures and targets for individual diseases emerged with the lost productivity exceeds US$342 billion over the launch of the World Health Organization’s NTD same period. roadmap in 2012; the Sustainable Development • The net benefit to affected individuals is about Goals target “the end of NTDs” by 2030. US$25 for every dollar to be invested by public • Interventions to end NTDs are affordable globally; and philanthropic funders between 1990 and estimated treatment costs are US$750 million per 2030—a 30 percent annualized rate of return. year for 2015 to 2020 and US$300 million per • The end of NTDs represents a fair and effi- year for 2020 to 2030. cient transfer toward universal health coverage • Interventions to end NTDs are affordable for and social protection for those who are least the governments of most endemic countries; well-off. Corresponding author: Christopher Fitzpatrick, Department of Control of Neglected Tropical Diseases, World Health Organization, Geneva, Switzerland; fitzpatrickc@who.int. 411 INTRODUCTION is now firmly embedded within the Sustainable Development Goals (SDGs) for 2030, under target 3.3, The neglected tropical diseases (NTDs) affect more than reflecting the promise to “leave no one behind.” 1 billion of the poorest and most marginalized people of This chapter reaffirms the case that NTDs account for the world. These infections are a consequence of the envi- a significant and unfairly distributed global disease bur- ronmental and socioeconomic conditions in which the den, cost-effective interventions exist to reduce that poor live, and the ill health and disability they cause are a burden, these interventions are affordable, and they are primary factor locking the poor into poverty. They are good investments in universal health coverage and social diseases of the most neglected people who live in coun- protection. It builds on the second edition of the Disease tries that lack the basic resources to control them. Yet this Control Priorities (DCP2) project (Hotez and others chapter demonstrates that the tools to end this neglect 2006) with new data and analysis. It also takes into already exist, and that there are compelling economic account new strategies and tools that have been intro- arguments that ending these diseases would be one of the duced since 2006 and the increasingly ambitious elimi- most cost-effective of global public health programs. nation and eradication targets for individual diseases The NTD concept was developed to draw attention to that have emerged since 2012, including the unprece- this opportunity that was overlooked by the Millennium dented donation by the pharmaceutical industry under Development Goals (MDGs). At least 18 diseases are the London Declaration of more than a billion medi- recognized as NTDs by World Health Assembly resolu- cines annually to treat nine of the most important tions; the latest addition is mycetoma (WHO 2013, NTDs. Finally, it helps provide a longer-term perspective 2016). The World Health Organization (WHO) has set on SDG target 3.3 and the 2030 goals. specific targets for control, elimination, and eradication This chapter is structured around three key NTD of a subset of these diseases (table 17.1). These are the interventions, rather than individual NTDs, in recogni- NTDs that we focus on in this chapter. The end of NTDs tion of the increasingly integrated delivery of interven- tions to the poorest, most remote, and otherwise most Table 17.1 Global Targets for Control, Elimination, and Eradication marginalized communities of the world. These interven- toward “the End of NTDs” tions are as follows: Indicator Target • Preventive chemotherapy by mass drug administration Incidence/ • Eradication of Guinea worm disease (2015b) and yaws (2020) • Innovative and intensified disease management prevalencea • Global elimination of leprosy (2020), lymphatic filariasis • Vector ecology and management. (2020), trachoma (2020), onchocerciasis (2025), and human African trypanosomiasis (2020, with zero incidence in 2030) For simplicity of analysis, we focus on a subset of the • Regional elimination of schistosomiasis (2020), rabies (2020), NTDs recognized by the WHO. We do not provide a full and visceral leishmaniasis (2020) analysis of veterinary public health interventions against • Regional interruption of intradomiciliary transmission of zoonotic NTDs or of water, sanitation, and hygiene Chagas disease (2020) (WASH). These conditions are beyond the scope of this • 25 percent reduction in the number of cases of dengue (2020, chapter, but WASH is addressed in chapter 9 in volume 7 compared with 2010) of the third edition of Disease Control Priorities (Hutton Mortality • 50 percent reduction in number of deaths attributable to and Chase 2017). Chapter 13 in volume 8 (Bundy and dengue (2020, compared with 2010) others 2017) discusses mass deworming programs, and Coveragea • 75 percent coverage with preventive chemotherapy for food- chapter 29 (Ahuja and others 2017) in volume 8 ana- borne trematode infections and soil-transmitted helminthiasis lyzes the economics of such programs. (2020) • 70 percent of all cases of Buruli ulcer detected and treated (2020) • Universal coverage against NTDsc (2030) BURDEN OF NTDs Source: WHO 2012. The MDGs evolved around HIV/AIDS malaria, and Note: NTDs = neglected tropical diseases. tuberculosis (TB); other diseases were overlooked and not a. Reaching the incidence and coverage targets should result in at least a 90 percent reduction in the number of people requiring interventions against NTDs between 2015 and 2030; this is the combined prioritized for funding. The focus was on mortality. These NTD indicator that will be monitored under Sustainable Development Goal target 3.3. three diseases accounted for about 3.9 million deaths in b. Target year for Guinea worm eradication has not been updated; only 22 cases were reported in 2015. 2000, including about 875,000 for malaria; NTDs c. 80 percent service coverage and 100 percent financial protection of people requiring at least one of five key interventions against NTDs: preventive chemotherapy; innovative and intensified disease accounted for about 242,000 deaths (Horton 2012; Hotez management; vector ecology and management; veterinary public health: water, sanitation, and hygiene. and others 2014). Several of the NTDs do not kill; they do, 412 Major Infectious Diseases however, disable, disfigure, and even impair the cognitive tropical diseases. Dengue has reemerged in high-income development of children. Today, the focus of the SDGs countries that had not seen cases in decades. Chagas has broadened to include healthy lives and well-being for disease now affects migrant populations across all at all ages. North America and Europe. Today, a majority of the poor The disability-adjusted life year (DALY) is meant to lives in countries assessed as middle income or above. account for years of life lost because of premature death, With national income and other secular trends being as well as years of life lived with disability. In practice, it generally upward in the LMICs where NTDs are most tends to underestimate the burden of NTDs in part prevalent, the overall burden of disease has been coming because of gaps in the data from low- and middle-income down since at least 2000. However, the persistence of countries (LMICs). Even so, in 2012, the NTDs accounted NTDs in middle- and even high-income countries indi- for approximately 22 million DALYs globally, which cates that some communities have been left behind by amounts to about 40 percent of the DALYs for malaria the macroeconomic development of the past decades. and about 1 percent of the global total. The contribution A review found that more than 60 percent of stud- of individual NTDs to the total is shown in table 17.2. ies reveal inequalities in the prevalence of NTDs across This burden is heavy, especially for regions and coun- socioeconomic groups (Houweling and others 2016). tries where NTDs are most endemic. In several countries in For example, in rural Nigeria, the prevalence of ascariasis Sub-Saharan Africa, NTDs make up more than 6 percent among children ranges from 10 percent when both of the total burden of disease. parents have at least primary education, to 31 percent However, the NTDs are not only a concern of low- when only the mother does, 53 percent when only the income countries (LICs) in Sub-Saharan Africa. A sig- father does, and 96 percent when neither parent does nificant burden is shouldered by the poorest and most (Ugbomoiko and others 2009). marginalized communities of middle-income countries, In Ethiopia, trichiasis cases (a consequence of as evidenced by figure 17.1. Indeed, environmental trachoma) are more likely to occur in poorer house- change and population movement have redefined holds, whether measured by asset ownership, Table 17.2 Disease Burden (Mortality and Morbidity) of Malaria and NTDs, 2012 DALYs (thousands) % YLD (thousands) % YLL (thousands) % Parasitic and vector diseases 72,006 2.62 11,697 1.58 60,309 3.01 Malaria 55,111 2.01 4,301 0.58 50,810 2.54 Trypanosomiasis 1,264 0.05 9 0.00 1,256 0.06 Chagas disease 528 0.02 326 0.04 202 0.01 Schistosomiasis 4,026 0.15 3,179 0.43 848 0.04 Leishmaniasis 3,374 0.12 128 0.02 3,245 0.16 Lymphatic filariasis 2,839 0.10 2,839 0.38 0 0.00 Onchocerciasis 598 0.02 598 0.08 0 0.00 Leprosya 257 0.01 6 0.00 251 0.01 Dengue 1,445 0.05 12 0.00 1,432 0.07 Trachoma 299 0.01 299 0.04 0 0.00 Rabies 2,265 0.08 0 0.00 2,265 0.11 Intestinal nematode infections 5,266 0.19 5,057 0.68 209 0.01 Ascariasis 1,355 0.05 1,146 0.15 209 0.01 Trichuriasis 666 0.02 666 0.09 0 0.00 Hookworm disease 3,246 0.12 3,246 0.44 0 0.00 Total excluding malaria 22,161 0.81 12,453 1.68 9,708 0.48 Source: World Health Organization, http://www.who.int/healthinfo/global_burden_disease/estimates/en/index2.html. Note: NTDs = neglected tropical diseases. Cause-specific disability-adjusted life year (DALYs), years of life lost (YLLs), and years lived with disability (YLDs). Percentages are expressed relative to the global total. a. Leprosy is formally not a parasitic disease, it is caused by a mycobacterium. Furthermore, we suspect that the YLD and YLL numbers for leprosy may have been inverted; we nonetheless report them here as in the original source. An Investment Case for Ending Neglected Tropical Diseases 413 Figure 17.1 Disease Burden of NTDs, by Country Income Group, 2012 self-rated wealth, or peer-rated wealth (the odds range from 2.8 to 8.2) (Habtamu and others 2015). Compared 12 with controls, those with trichiasis are also signifi- cantly less likely to participate in economically pro- ductive activities, more likely to report difficulty in 10 performing activities, and more likely to receive assis- tance in performing productive activities (Habtamu and others 2015). 8 In addition to the disease burden is the heavy eco- DALYs (millions) nomic burden that NTDs impose on patients and their families. Most of the economic cost comes in lost pro- 6 ductivity, usually working time (and wages), but also agricultural land, lost to morbidity and disability. The extent of loss of productive inputs depends on the type and severity of the NTD as well as where it occurs 4 (table 17.3). The particularly high economic cost of blindness motivated the World Bank’s first investments in health, with the creation of the Onchocerciasis Control 2 Programme in West Africa in 1975. Added to the productivity losses are the direct medical costs of diagnosis and treatment and, even if tests and 0 medicines are offered free of charge, direct nonmedical High income Low income Lower-middle Upper-middle costs associated with accessing or adhering to treatment. income income The latter include transportation, accommodation, and Hookworm disease Dengue Schistosomiasis food. Altogether, costs can easily exceed 20 percent of Trichuriasis Leprosy Chagas disease annual household income, a threshold for so-called cata- Ascariasis Onchocerciasis Trypanosomiasis strophic cost that can propel a previously stable household Rabies Lymphatic filariasis into penury and unsupportable debt (Ruan and others Trachoma Leishmaniasis 2016). Protection against this risk requires further public Source: WHO 2014a. sector investment in finding cases early, treating patients Note: DALYs = disability-adjusted life years; NTDs = neglected tropical diseases. free of charge, and, as required, other social protection to cover transport and other nonmedical and indirect costs. Table 17.3 Economic Burden (Lost Productivity) of Selected NTDs These are some well-documented examples: Annualized loss in Buruli ulcer. In Cameroon, the cost of hospitalization NTD Sequela productive input (%)a attibutable to Buruli ulcer (caused by Mycobacterium Chagas disease General (excluding severe 2–5 ulcerans) has been estimated to be 25 percent of house- heart failure) hold annual earnings, despite treatment being available Leprosy Disfigurement 28 free of charge. In Ghana, medical costs made up less than Lymphatic filariasis Hydrocele 14–19 4 percent of total direct costs; the largest cost (81 percent of direct costs) is transportation to treatment (WHO 2015b). Lymphedema 2–23 Onchocerciasis Visual impairment 14–38 Chagas disease. The cost of Chagas disease (caused by Blindness 79–100 Trypanosoma cruzi) was estimated in 2013 to be about Schistosomiasis General 1–23 US$7 billion per year, including lost productivity (Lee Soil-transmitted Anemia 0.1–6.0 and others 2013). Health care costs accounted for slightly helminthiasis less than 9 percent of this total. The cost of treatment Trachoma Visual impairment 25 ranges from less than US$200 to more than US$30,000 Blindness 60–100 per person per year in endemic countries, and exceeds US$40,000 in the United States (WHO 2015b). Visceral leishmaniasis General (treated) 6–30 Sources: Ibe and others 2015; Lenk and others 2016. Note: NTD = neglected tropical disease. Dengue. In Cambodia and Vietnam, “between half a. Minimum and maximum from the available literature. and two-thirds of affected households have incurred 414 Major Infectious Diseases debt as a result of treatment for dengue” (WHO 2015b, PROOF OF CONCEPT FOR ENDING NTDs 82). The economic burden of the disease is measured in the billions of dollars annually; urbanization and Despite being sidelined in the MDGs, an integrated climate change are conspiring to raise the cost even approach to the prevention and control of NTDs began higher (Constenla, Garcia, and Lefcourt 2015; Martelli to take shape during the MDG era, and by the end, NTD and others 2015; Shepard, Undurraga, and Halasa 2013; interventions had delivered a number of successes. These Shepard and others 2011; Shepard and others 2014; successes include a reduction of 80 percent in new Undurraga and others 2015). human African trypanosomiasis (HAT) cases between 2000 and 2014, to an estimated less than 4,000 cases; and a reduction of 75 percent in the number of cases of Human African trypanosomiasis. In the Democratic visceral leishmaniasis (kala-azar) in Bangladesh, India, Republic of Congo, the cost to affected households and Nepal between 2005 (when a regional program in a typical rural community represents more than was launched) and 2014, to a reported 10,209 cases. In 40 percent of annual household income. New and more 2000, there were more than 130,000 cases of dracuncu- effective melarsoprol-free treatment has increased the liasis (Guinea worm); in 2015, there were only 22 average cost to treat one patient with second-stage gam- reported cases, reflecting near-eradication (figure 17.2). biense sleeping sickness from US$30 in 2001 to US$440 Map 17.1 shows the reported numbers of cases of these in 2010 (WHO 2015b). three NTDs targeted for elimination or eradication. For other NTDs, especially those for which cases are Leprosy. Erythema nodosum leprosum is a common not routinely reported to the WHO, country-level prog- immune-mediated complication of leprosy. In a dis- ress has been made toward the interruption of transmis- trict of West Bengal, India, the total household cost of sion. For example, by 2014, 18 countries reported having erythema nodosum leprosum was about 28 percent been able to stop preventive chemotherapy for lym- of monthly household income (Chandler and others phatic filariasis, and 8 countries have stopped mass anti- 2015). Direct costs accounted for 35 percent of this total. biotic treatment for trachoma, because the set targets Total household costs exceeded 40 percent of household had been reached. income for more than one-third of cases. Other NTDs that have been eliminated in certain countries or that are under surveillance for verification Visceral leishmaniasis. In Bihar, India, 83 percent of affected households belong to the two lowest Figure 17.2 Reported Number of Cases of Three Neglected Tropical wealth quintiles (the poorest 40 percent) (Boelaert Diseases Targeted for Elimination or Eradication, 2000–15 and others 2009). In Bangladesh, India, Nepal, and Sudan, 25 percent to 75 percent of affected house- 125 holds experience some type of financial catastrophe in obtaining a diagnosis and treatment, even when tests Number of cases reported (thousands) and medicines are provided free of charge (Anoopa 100 and others 2006; Meheus and others 2013; Ozaki and others 2011; Sundar and others 2010; Uranw and 75 others 2013). In addition to the health (death and disability) and economic burden, there is also the social and psycholog- 50 ical (mental health) burden of NTDs because of stigma. Reasons given for stigmatization include appear- ance, fear of contagion, burden on family, hereditary 25 etiology, promiscuity, and performance impediment. This burden is harder to quantify, but there is evidence that no less than 10 NTDs are associated with stigma, 0 with especially strong evidence related to leprosy, lym- 2000 2005 2010 2015 phatic filariasis, Buruli ulcer, onchocerciasis, and leish- Year maniasis (Hofstraat and van Brakel 2016). The visible Dracunculiasis Human African trypanomiasis impact of NTDs has been shown to be an important Visceral leishmaniasis (Indian subcontinent) determinant of stigma; disease management should therefore have a positive effect on stigma. Source: WHO Global Health Observatory, (http://apps.who.int/gho/data/node.main.A1629?lang=en). An Investment Case for Ending Neglected Tropical Diseases 415 Map 17.1 Countries in Which One or More NTDs Have Been Eliminated or Are under Surveillance for Verification of Elimination, since 2000 IBRD 42555 | OCTOBER 2016 NTDs that have been eliminated since 2000 Dracunculiasis Onchocerciasis Trachoma Yaws Lymphatic filariasis Rabies Visceral leishmaniasis Source: WHO Global Health Observatory, (http://apps.who.int/gho/data/node.main.A1629?lang=en). Note: NTD = neglected tropical disease. of elimination are illustrated in figure 17.3. Those coun- INTERVENTIONS TO END NTDs tries are already reaping the economic and financial rewards that come with having eliminated a disease and The WHO recommends five interventions for the stopping treatment, thereby freeing up resources for control, elimination, and eradication of the NTDs: other public health priorities. preventive chemotherapy by mass drug administra- Progress on NTD-related mortality includes a reduction tion; innovative and intensified disease management; in deaths caused by visceral leishmaniasis, rabies, schistoso- vector ecology and management; veterinary public miasis, HAT, Chagas disease, and soil-transmitted helmin- health services; and the provision of safe water, sanita- thiases (that is, ascariasis, collectively estimated to be tion, and hygiene (WHO 2010; see the discussion in 142,000 deaths in 2012, down from about 220,000 in 2000 volume 7 of this series [Hutton and Chase 2017]). We (WHO 2014a). review the evidence for all but the last two of these Much of the burden of NTDs occurs in morbidity interventions. rather than mortality, and here, too, the progress has been good, albeit somewhat less dramatic, with a decrease of 19 percent in the total number of DALYs between Delivering Large-Scale Preventive Treatment to Entire 2000 and 2012, from 1 percent of the global burden of Communities disease to 0.8 percent (WHO 2014a). There have been Preventive chemotherapy involves the large-scale delivery logistical challenges, of course, that have differed greatly of medicines to eligible populations at regular intervals. between diseases and between countries. However, elim- Medicines donated to and distributed through the WHO ination of dracunculiasis, for example, has been achieved are quality assured and safe for administration by non- in some of the most difficult settings in the world. health workers. Table 17.4 provides the disease-specific 416 Major Infectious Diseases details of how preventive chemotherapy is delivered. Figure 17.3 Update on the Global Status of Implementation of In many areas, these diseases do not occur exclusive of each Preventive Chemotherapy, 2008–20 other, but are co-endemic. A combination of medicines is recommended in this scenario. Integrated delivery of treat- Required trajectory 75 ments for more than one disease is now the norm in several countries, with resulting cost savings (WHO 2015b). Preventive chemotherapy is effective toward elimina- 60 tion only if the threshold coverage is sustained annually 50 Current trajectory Coverage (%) for at least three years or longer, depending on the disease. The WHO has set clear thresholds for effective 40 Trajectory before 2012 program coverage, by disease (table 17.4), meaning delivery of medicines to a minimum percentage of eligi- 30 ble individuals during approximately the same time period, with 100 percent geographic coverage of endemic areas. If the threshold coverage is not met, the disease burden is reduced but will return when preventive che- motherapy is stopped. If threshold coverage is met, countries can stop mass treatment, or at least reduce its 2008 2012 2016 2020 Year frequency, and shift resources to integrated disease sur- veillance and other public health priorities. Source: WHO (http://www.who.int/neglected_diseases/preventive_chemotherapy/PC_Update.pdf?ua=1). The global population in need of preventive chemo- Note: Analysis includes current and projected percentage of people receiving preventive therapy is reported to be 1.7 billion as of 2014, of which chemotherapy for at least one of these diseases (including lymphatic filariasis, onchocerciasis, schistosomiasis, and soil-transmitted helminthiases) out of the estimated number of people 851 million people actually received treatment, leaving a requiring preventive chemotherapy. The required trajectory assumes a linear scale-up to 75 percent, coverage gap of approximately 50 percent (WHO 2015c). the minimum target coverage indicated in table 17.4. Table 17.4 Selected NTDs Targeted by Preventive Chemotherapy Target population (minimum Frequency and duration of Disease Causative organism and transmission Medicine, single dose effective coverage) intervention Lymphatic Parasites (Wuchereria bancrofti, Brugia Albendazole 400 mg with Ivermectin: ≥ age 5 years Annually for at least 5 years filariasis malayi, Brugia timori ) transmitted by ivermectin (150–200 μg/kg) Diethlycarbamazine: ≥ age 2 mosquito or with diethylcarbamazine years (65%) 6 mg/kg Onchocerciasis Parasites (Onchocerca volvulus) Ivermectin 150 μg/kg or > age 5 years (80%)c Annually for at least 10–15 transmitted by blackfly mcg/kg years Schistosomiasis People are infected by parasites Praziquantel 40 mg/kg SAC (ages 5–14 years) and Once a year, or every two (S. haematobium, S. mansoni, adults at risk (75%) to three years, depending S. japonicum) when exposed to on community prevalence, freshwater infested by cercariae released for variable or unknown by intermediate host snails. durationa Soil-transmitted The main parasites that infect people Albendazole 400 mg Pre-SAC (< age 5 years) and Once or twice a year helminthiases are the roundworm (Ascaris lumbricoides), Mebendazole 500 mg SAC (age 5–14 years) (75%) depending on community the whipworm (Trichuris trichiura) and prevalence, for variable or the hookworms (Necator americanus unknown durationa and Ancylostoma duodenale ) Trachoma Bacterial infection (Chlamydia Azithromycin 20 mg/kg > age 6 months (80%) Annually, with the number of trachomatis) through contact with to a maximum of 1g rounds given before review infected people or spread by flies (This is given as part of dependent on the prevalence a SAFEb strategy) of disease at last estimate Source: WHO (http://apps.who.int/iris/bitstream/10665/43545/1/9241547103_eng.pdf). Note: g = gram; kg = kilograms; mg = milligrams; SAC = school-age children; μg or mcg = micrograms. a. Treatment is geared toward reducing the intensity of infection in individuals. The frequency of treatment may be reduced over time, but ultimately, the duration will depend on improved water, sanitation, and hygiene. b. SAFE strategy comprises: S-eyelid surgery for trichiasis, A-antibiotics, F- facial cleanliness, and E-environmental improvement. c. 80% of the eligible population is roughly equivalent to 65% of the total population. An Investment Case for Ending Neglected Tropical Diseases 417 Still, it is important to acknowledge the tremendous network has been heavily subsidized by communities. progress that has been made in shifting the trajectory in Members are selected as community drug distributors and coverage up from the levels of 2000–11 (figure 17.3). In paid in kind or with cash incentives by their communities. 2012, the WHO NTD Roadmap (WHO 2012) and sub- This approach has worked well in small-scale projects, but sequent London Declaration drove the pivot from the many communities are overstretched. Bridging the cover- trajectory in 2011 (green line) to the current trajectory age gap will require investment in delivery chains within (dark blue line). It also remains clear that the current endemic countries; how much is needed and what the trajectory is not sufficient to meet the required level of return will be are described later in the chapter. coverage of 75 percent (light blue line) early enough (2016) to ensure that treatment can be stopped or its frequency reduced by 2020 (WHO 2012). Preventing the Transmission of NTDs by Vectors Morbidity management and disability prevention is a Vectors are living organisms that can transmit infec- related intervention for those for whom preventive che- tious diseases between humans or from animals to motherapy has arrived too late, particularly for those with humans (and vice versa). Many of these vectors are a long history of infection with lymphatic filariasis or bloodsucking insects that ingest disease-producing trachoma, now suffering the chronic consequences. Adult microorganisms during a blood meal from an infected filarial worms lodge in the lymphatic system and disrupt host (human or animal) and later inject it into a new the immune system, resulting in swelling of the scrotum host during their subsequent blood meal. Mosquitoes and lower limbs. Repeated reinfection with Chlamydia are the best known vector, transmitting malaria as well trachomatis gives rise to trichiasis, in which eyelashes rub as lymphatic filariasis. Others include blackflies (oncho- on the eyeball, leading to corneal opacification and cerciasis), sandflies (leishmaniasis), and triatomine bugs blindness. Hydrocele and trichiasis surgery, as well as (Chagas disease). lymphedema management, are complementary to mea- Vector ecology and management aims to control the sures to reduce infection prevalence—the benefits being transmission of the causative pathogens of insect-borne more visible than those from the distribution of drugs. NTDs with proven interventions that are applied in an The pharmaceutical industry has expressed its com- ecologically friendly manner. The WHO’s integrated mitment to meet medicine requirements. A full list of vector management strategy has found use in areas with medicines donated by the pharmaceutical industry is multiple vector-borne diseases or where preventive che- available in annex 5 of the third global report on NTDs motherapy is contraindicated because of the risk of (WHO 2015b). These are highly effective medicines— severe adverse events (for example, in areas of oncho- most of which are on the WHO’s list of essential cerciasis and Loa co-endemicity) and where there is no medicines—being provided for free. Logistic constraints other intervention to control infection (for example, have, however, hindered the ability of NTD programs to dengue, chikungunya). NTDs targeted primarily by ensure universal access to these medicines. The delivery vector management are listed in table 17.5. Table 17.5 Selected NTDs Targeted by Vector Ecology and Management Interventions Disease (vector) Description of vector management intervention Chagas disease (hematophagous Spray homes and surroundings with residual insecticides triatomine or “kissing bug”) Improve walls and roofs of dwellings Use bednets and other personal control measures Dengue, chikungunya (female mosquitos Individual and household protection Aedes aegypti and Aedes albopictus) Vector surveillance Biological, chemical, and environmental control (including insecticides) Outbreak preparedness and response Visceral leishmaniasis or kala-azar In areas of the Indian subcontinent where vector control is not already being undertaken by malaria (female sandflies) programs: Vector surveillance Indoor residual spraying and use of bednets Note: NTDs = neglected tropical diseases. Vector control interventions also exist for onchocerciasis, human African trypanosomiasis, and lymphatic filariasis; however, these are not discussed here. There is little recent evidence on their cost-effectiveness or, in the case of lymphatic filariasis, evidence is limited to areas that are co-endemic with malaria. 418 Major Infectious Diseases Coverage with vector management for the prevention and Nepal launched a comprehensive strategy including of Chagas disease and dengue remains uneven. More integrated vector management. In Nepal elimination has than 100 million people still require an attack phase of been achieved in 12 districts. In Bangladesh, the number vector control for the interruption of intradomiciliary of hyperendemic subdistricts (upazilas) decreased from transmission of Chagas disease in Latin America. This eight in 2012 to two in 2014, with elimination achieved in intensive phase involves residual insecticide spraying by about 90 percent of endemic upazilas (WHO 2015d). specialized mobile teams two times with a six-month intermediate period (to kill all insects, including the ones coming from eggs) and dwelling or house hygiene and Providing Treatment and Care to Individuals improvement (plastering) to avoid reinfestation. Not all NTD cases can be prevented by preventive chemo- Although 2 billion to 4 billion people are at risk of therapy and vector management, for example, because the dengue or chikungunya, only a handful of counties offer drugs are too toxic. A complementary approach focuses coverage with sustained vector management involving on the innovative and intensified clinical management biological, environmental, and chemical measures of diseases. Innovation and intensification refer to a shift adapted to the needs of the communities. Most countries from passive management to active surveillance, early only respond to dengue outbreaks when it is too late to diagnosis, and treatment, with the aim to eliminate or make much of a difference (Stahl and others 2013). control, not just to manage. Treatment of Buruli ulcer, for However, sustained vector control interventions could example, has evolved from late-stage surgical removal of make a very large difference, not only for dengue and infected or dead tissue and correction of deformity to the chikungunya, but also for other diseases transmitted by early-stage use of antibiotics (a combination of rifampicin Aedes aegypti and Aedes albopictus. and streptomycin or amikacin). The gains go beyond Progress toward elimination of visceral leishmaniasis health benefits to include reductions in the cost of hospi- as a public health problem on the Indian subcontinent talization to health systems and to individuals. The NTDs provides evidence of the impact that vector management for which the primary intervention is disease manage- can have. In 2005, the governments of Bangladesh, India, ment are listed in table 17.6. Table 17.6 Selected NTDs Targeted by Disease Management Interventions Disease (agent) Disease management intervention Buruli ulcer Early case detection and antibiotic treatment, including rifampicin and streptomycin, or rifampicin and clarithromycin (Mycobacterium ulcerans) Surgical removal of dead skin and grafting Rehabilitation for deformities Chagas disease Early case detection and treatment with nifurtimox and benznidazole (Trypanosoma cruzi) Adequate screening of blood for transfusion Screening (testing) of organ, tissue, or cell donors and receivers Screening of newborns and other children of infected mothers to provide early diagnosis and treatment Other morbidity-specific treatment Human African trypanosomiasis Early case detection and treatment with pentamidine or suramin or nifurtimox-eflornithine combination (Trypanosoma) treatment (NECT), depending on the stage of the disease Leishmaniases (Leishmania spp) Early case detection For visceral leishmaniasis, treatment options include: sodium stibogluconate, meglumine antimonite, paromomycin, liposomal amphotericin B or miltefosine, depending on the parasite species and the endemic region For cutaneous leishmaniasis, management options include local or systemic treatments with antileishmanial drugs or local procedures with cryotherapy, thermotherapy Leprosy Early case detection and treatment with multidrug regimen (combination of rifampicin, dapsone, and (Mycobacterium leprae) clofazimine) and management of morbidity and prevention of disability Yaws Formerly, identification of the population at risk of infection by case finding (active and passive) and treatment (Treponema pallidum pertenue) with injectable penicillin Currently, Total Community Treatment followed by Total Targeted Treatment of confirmed cases and their contacts with single dose of azithromycin Note: NTDs = neglected tropical diseases. An Investment Case for Ending Neglected Tropical Diseases 419 The inclusion here of yaws warrants an explanation, There is now a rich literature—34 studies of 23 countries given the recent shift in strategy from individual treat- and at least 91 sites over 19 years—documenting the cost ment with injectable penicillin to mass treatment with per person treated in diverse settings (Fitzpatrick and single-dose azithromycin. While the risk is thought to be others, forthcoming). The average unit cost is US$0.40 low in populations with little previous antibiotic expo- (in 2015 U.S. dollars) in financial terms, but the average sure, surveillance is undertaken to guard against antimi- unit cost increases to US$0.70 in studies that also con- crobial resistance. This mass treatment for yaws is sider the economic cost of ministry of health staff time similar to preventive chemotherapy but is known as and assets. About half of the available estimates of the Total Community Treatment. In keeping with the con- economic unit cost fall between US$0.30 and US$1.00, vention within the NTD community, this chapter con- but they range from a low of US$0.02 in large-scale pro- siders Total Community Treatment as separate from grams to US$2.9 in smaller projects. Benchmarking tools preventive chemotherapy. can help assess the value for money in mass treatment campaigns (WHO 2015b). COST AND COST-EFFECTIVENESS OF Cost-Effectiveness INTERVENTIONS TO END NTDS Hotez and others (2006) described preventive chemo- This section is a synthesis and update of the review of the therapy as one of the most cost-effective interventions cost and cost-effectiveness conducted for the WHO’s NTD available in public health. The large and unprecedented report (WHO 2015b) and subsequent systematic reviews. donation of NTD medicines in the London Declaration in 2012 strengthens that case from the perspective of national health systems. Indeed, reviews continue to show Preventive Chemotherapy that preventive chemotherapy remains cost-effective, Unit Cost of Delivery even with an expansion beyond the traditional zones of Advocacy around preventive chemotherapy has typically focus, or with an increase in treatment frequency to accel- put the cost per person treated at less than US$0.50. erate progress (Keating and others 2014). While useful for advocacy, the focus on a single number Some of the more recent cost-effectiveness analyses misrepresents the complexity of delivering “free” donated are presented in table 17.7, with results standardized for medicines to more than 1 billion people across the world. prices in 2012. Table 17.7 Recent Cost-Effectiveness Analyses of Preventive Chemotherapy 2012 US$ per DALY averted, relative to Disease Study Intervention Setting Target population doing nothing Lymphatic filariasis Turner and others 2016 Albendazole + ivermectin Global All 28a Onchocerciasis Turner and others 2014 Ivermectin, annual Africa Mesoendemic (microfilarial 15 prevalence: 40%) Hyperendemic (60%) 6 Highly hyperendemic (80%) 3 Schistosomiasis and STH Turner and others 2015 Albendazole + praziquantel Global School-age children 5-80 Lo and others 2015 Côte d’Ivoire School-age children 114 Trachoma Baltussen and Smith Mass treatment with Sub-Saharan 95% coverage 22–83b 2012 azithromycin + trichiasis Africa surgery Note: DALY = disability-adjusted life year; STH = soil-transmitted helminthiasis. a. Stone and others (2016) do not report the number of DALYs averted relative to a null (do nothing) scenario; they report the incremental costs and effects of a hypothetical eradication program over the baseline elimination program. b. This estimate is based on Baltussen and Smith (2012) using median purchasing-power-parity exchange rates for Sub-Saharan Africa in 2012 to convert from international dollars of 2000. Baltussen and Smith (2012) used 2012 market prices for azithromycin; an assumption of zero cost would be closer to the reality of the current situation. See text for further comment. 420 Major Infectious Diseases Annual mass treatment with ivermectin is estimated All of the cost-effectiveness ratios described above to cost 2012 US$3–US$15 per DALY averted, depending are well below the threshold of one times gross domes- on the degree of onchocerciasis endemicity; biannual tic product (GDP) per capita, implying that they are mass treatment would cost an additional 2012 very cost-effective (WHO-CHOICE 2012). Even so, US$12–US$36 per DALY averted in hyperendemic areas they may be overstated. Integrated delivery of more (Turner and others 2014). This does not take into than one medicine at a time is safe and there is evi- account the substantial collateral reduction in DALYs dence that it will reduce cost (Evans and others 2011; attributable to the impact of ivermectin on lymphatic Leslie and others 2013). filariasis, soil-transmitted helminthiasis (STH), and scabies (Krotneva and others 2015). With regard to STH, there has been a recent contro- Vector Ecology and Management versy about the extent to which it is possible to detect Chagas Disease population-level benefit from mass treatment. However, The cost-effectiveness of vector control for Chagas there is no doubt that infected persons are at risk of dis- disease in the Argentinean Chaco region has been ease and require treatment; the WHO recommendation estimated to be 2004 US$45–US$132 per human case is for mass treatment in communities where prevalence averted, depending on the strategy chosen (Vazquez- exceeds 20 percent (see Bundy and others 2017 for a Prokopec and others 2009). A mixed strategy—vertical discussion of these issues). A recent review finds that (centralized) attack phase followed by a horizontal most studies present results within the range of being (community-led) surveillance phase—is thought to be highly cost-effective according to World Bank thresholds the most cost-effective option. A comparison of vector (Turner and others 2015). Mass treatment of school-age control policies for Chagas disease in Colombia to a children in Côte d’Ivoire for STH and schistosomiasis do-nothing policy revealed net benefits for all consid- together costs 2014 US$118 (2012 US$114) per DALY ered villages at a willingness to pay of 2004 US$631 averted relative to doing nothing (Lo and others 2015). (2012 US$940) per DALY averted (Castillo-Riquelme Mass treatment of the entire community would also be and others 2008). cost-effective, at 2014 US$167 (2012 US$161) per DALY averted relative to school-age children only (Lo and oth- ers 2015). Combination with other interventions is also Dengue possible. Mass treatment for STH costs 2012 US$13 per For dengue, DCP2 put the cost per DALY averted by vector control at US$1,992–US$3,139 (Cattand and DALY averted when added to a vitamin A supplementa- others 2006). Since then, the dengue economics litera- tion campaign for children ages 6 months to 14 years in ture suggests lower cost-effectiveness ratios that range Uganda (Fiedler and Semakula 2014). from 2005 US$227 (2012 US$334) per DALY averted Of the medicines delivered as preventive chemo- by larval control in Cambodia to 2009 US$615– therapy, azithromycin, a broad-spectrum antibiotic, US$1,267 (2012 US$779–US$1,604) per DALY averted has the greatest market value. The cost-effectiveness of by adult mosquito control in Brazil (Suaya and others preventive chemotherapy for trachoma (relative to 2007; Luz and others 2011). Environmental change, other interventions for the prevention of blindness) including urbanization and climate change, strengthen depends crucially on assumptions about the cost of the investment case for sustained vector control, which azithromycin. There is no market price for azithromy- is cost-effective even in the era of a low-cost, medium- cin for use in global trachoma elimination. Applying efficacy vaccine. If benefits for the control of chikun- the market price of azithromycin for use in smaller- gunya and Zika virus (transmitted by the same vector) scale programs, mass treatment combined with trichia- were taken into account, the cost per DALY averted sis surgery costs about 2012 US$83 per DALY averted would be even lower. in Sub-Saharan Africa, relative to doing nothing (Baltussen and Smith 2012). In practice, azithromycin is available as a free donation to trachoma-elimination programs worldwide, and the cost per DALY averted is Disease Management lower than the cost using market pricing. An earlier Cutaneous Leishmaniasis study suggested a 73 percent decrease in the cost per The cost-effectiveness of interventions for cutaneous DALY averted with donated azithromycin (Baltussen leishmaniasis was not specifically discussed in DCP2. The and others 2005). Therefore, the cost per DALY averted difficulty with assessing their cost-effectiveness is that this is probably closer to 2012 US$22. form of the disease is not fatal and disability weights An Investment Case for Ending Neglected Tropical Diseases 421 may not fully reflect the social stigma associated with deliver those treatments early enough to prevent disabil- disfigurement. Nonetheless, recent economic evaluations ity and further transmission. suggest that interventions for early diagnosis and treat- ment can be cost-effective, ranging from 2010 US$156 Yaws (2012 US$218) per DALY averted in Argentina to 2003 A global yaws eradication campaign could be established US$1,200 (2012 US$3,000) per DALY averted by treat- with a relatively modest investment in the period to ment in a complex emergency setting in Afghanistan 2020—about 2012 US$100 million to US$500 million in (Orellano, Vazquez, and Salomon 2013; Reithinger and 12 endemic countries. Yaws eradication would cost Coleman 2007). about 2012 US$26 per year lived without disability or 2012 US$324 per DALY averted. Global financial sup- Visceral Leishmaniasis port is not yet available in the same proportions as for Liposomal amphotericin B (AmBisome) has been found other NTDs. The cost to the public sector would be sig- to be the most effective treatment option available for nificantly reduced by donations of azithromycin for the Indian subcontinent for visceral leishmaniasis yaws, as is done for trachoma (Fitzpatrick, Asiedu, and (Meheus and Boelaert 2010). Recent donations of Jannin 2014; WHO 2015b). AmBisome have also made it the most cost-effective More evidence is needed on the cost-effectiveness of treatment option from the health system perspective. In interventions for Buruli ulcer and mycetoma. Such evidence Bangladesh, a comprehensive elimination program will likely come from evaluations of integrated approaches involved active case detection, single-dose treatment to screening, diagnosing, and treating these and other with donated AmBisome, indoor residual spraying, skin-related NTDs, especially cutaneous leishmaniasis, lep- long-lasting insecticide treated nets, and environmental rosy, and yaws. vector management. It was the most cost-effective option available at thresholds above 50 percent of GDP per capita, and cost far less than 50 percent of GDP per FAIRNESS OF INTERVENTIONS TO END NTDS capita per DALY averted relative to doing nothing Mass treatment is an intervention that favors women in (Federici and others, forthcoming). most countries. NTDs could have a disproportionate impact on the health and well-being of girls and women Human African Trypanosomiasis (including pregnant women) because they negatively The latest economic evaluations identified in a recent affect female reproductive health; exacerbate anemia in review focused on the Democratic Republic of Congo women of reproductive age; and increase susceptibility and Angola, of which the Democratic Republic of to sexually transmitted infections, including HIV/AIDS. Congo makes up most of the remaining burden of HAT Mass treatment turns out to be quite favorable to in the world (Sutherland and others 2015). Case detec- women. In coverage surveys from 37 countries from tion, diagnosis, and treatment were considered cost- which data were available, the gender ratio (female-to- effective at 2002 US$17 (2012 US$79) per DALY averted male) was between 0.96 and 1.17 (Worrell and Mathieu in the Democratic Republic of Congo. However, the 2012). Data from Uganda also suggest that coverage current treatment, nifurtimox-eflornithine combina- tends to be higher among females than males (Rilkoff tion therapy (NECT), has not yet been evaluated for 2013). Men tend to be away from home more often cost-effectiveness. Given that NECT too is donated and than the women within a household, whether for work has efficacy in excess of 90 percent, we would anticipate or travel. at least similar ratios. Inequity persists along other dimensions of socioeco- nomic status. Disaggregation of NTD intervention cover- Leprosy age is not yet routinely done. The disaggregated data we do For leprosy, DCP2 put the cost per DALY averted by case have are from household surveys that ask whether children detection and treatment at less than US$50 (Remme and ages 6–59 months had been given deworming medication others 2006). Since then, the pharmaceutical industry in the past six months. This is only a subset of the popula- has committed an unlimited number of treatments to tion requiring treatment for one NTD—STH—but it overcome the disease. From the health system perspec- points to both a challenge and an opportunity. In most tive, therefore, treatment is more cost-effective than ever. countries, deworming coverage is similar in both rural and The economic evaluation of leprosy elimination pro- urban areas, but higher among educated and wealthy grams focuses primarily on the cost-effectiveness of households who need it least (figure 17.4). A dozen or so interventions to detect more cases earlier (Ezenduka and countries have demonstrated that higher rates of coverage others 2012; Idema and others 2010). The challenge is to can be achieved among those who need it most. 422 Major Infectious Diseases Figure 17.4 Deworming Coverage among Children Ages 6–59 Months in 55 Countries, Ratio Wealth Education Residence 1 2 3 4 5 6 7 Ratio of deworming coverage, high relative to low socioeconomic status Source: Demographic and Health Surveys, most recent available (2006–14) from 55 countries. Note: Ratio of deworming coverage: Residence shows the ratio of coverage in urban relative to rural areas; Education shows the ratio of coverage among households headed by those with higher relative to no education; and Wealth gives the ratio of coverage among the wealthiest 20 percent relative to the poorest 20 percent of households. TARGETS FOR THE SCALE-UP OF eradication or elimination targets for individual NTDs INTERVENTIONS TO END NTDs are being brought together under a single indicator for 2030: reduction in the number of people requiring The current improved support for control and elimina- interventions against NTDs. tion of NTDs reflects a well-structured strategy, avail- This indicator will capture but is not limited to erad- ability of cost-effective interventions, and a clear road ication of yaws (2020); global elimination of leprosy map against which progress can be measured. Endemic (2020), lymphatic filariasis (2020), trachoma (2020), countries have adopted global targets and milestones in onchocerciasis (2025), and HAT (2020, with zero inci- national NTD master plans. Within these plans the dence in 2030); and regional elimination of schistoso- national NTD programs are responsible for ensuring miasis (2020) and visceral leishmaniasis (2020). These that all donated essential medicines are delivered to all remain critical milestones on the path toward the end of the citizens requiring them. the NTD epidemic by 2030. If these milestones are met, the total number of people requiring treatment for NTDs may begin to decrease as soon as 2017, as diseases Treatment Targets are eradicated, eliminated, and controlled. The WHO’s NTD Roadmap (WHO 2012) set clear tar- Between 2015 and 2030, we should see a 90 percent gets for the eradication or elimination of 11 of the 17 reduction in the number of people in need of mass and NTDs considered by 2020. Eradication is the “perma- individual treatment globally. The projected 90 percent nent reduction to zero of a specific pathogen, as a result reduction in the number of people requiring treatment of deliberate efforts, with no more risk of reintroduc- will be associated with a projected 75 percent reduction tion,” while elimination (of transmission) is the “reduc- in DALYs, from 12 million in 2015 to 3 million in 2030, tion to zero of the incidence of infection caused by a expected from the achievement of NTD Roadmap tar- specific pathogen in a defined geographical area, with gets for nine NTDs (figure 17.5). The decrease in the minimal risk of reintroduction, as a result of deliberate total number of people requiring treatment against efforts; continued actions to prevent re-establishment of NTDs from about 1.6 billion in 2015 to less than 300 transmission may be required” (WHO 2015a, 1). million means far less death, disability, and disfigure- Elimination thresholds are defined differently by dis- ment; but it also means far less cost to households and to ease, but in general involve the reduction of disease the health system. This is why we speak of spending on impact to below levels of public health importance. SDG these NTDs as an investment. target 3.3 is to “end the epidemic” of NTDs. For global At the same time, achievement of the 2020 targets and monitoring purposes, the existing coverage and even the end of treatment is not exactly the same as the An Investment Case for Ending Neglected Tropical Diseases 423 Figure 17.5 Health Impact of Achieving the WHO’s NTD Roadmap Targets for 2020 and Sustaining Them until the “End of NTDs” by 2030, in Relation to Having Done Nothing, Expressed as a Continuation of the 1990 Situation, Corrected for Demographic Trends 50 40 30 DALYs (millions) 20 10 0 1990 2000 2010 2020 2030 Year Averted DALYs Remaining DALYs Sources: de Vlas and others 2016; https://erasmusmcmgz.shinyapps.io/dissemination/. Note: DALYs = disability-adjusted life years; NTDs = neglected tropical diseases. These trends are for the nine so-called London Declaration NTDs only, which explains why the total number of DALYs is less than the 22 million DALYs reported in table 17.2 for 2012. end of NTDs. Rehabilitation and disability inclusion will the investment target for treatment (preventive chemother- need to be sustained well beyond 2030 for those people apy and individual treatment) is about US$750 million per for whom prevention arrived too late and for whom the year during 2015–20, and about US$300 million per year consequences are irreversible. Vector control for Chagas during 2020–30. These amounts exclude the value of disease and dengue are interventions that are also likely donated medicines, estimated to be about $4.5 billion to extend beyond 2030. To be sustainable, these longer- when using the lowest prices negotiated on the market term services and interventions will need to be included (MSH 2014). An estimated US$2.8 billion is required to within benefit packages under universal health coverage. deliver donated medicines to end users either in the form of mass treatment or facility-based care. Where will this investment come from? Between 2012 Investment Targets and 2014, about $200 million to $300 million was dis- The third WHO global NTD report set investment targets bursed or committed by foreign donors (Uniting to for reaching the 2020 Roadmap and 2030 SDG targets Combat Neglected Tropical Diseases 2014). This is about (WHO 2015b). Including vector control for Chagas disease one-tenth of the investment target in endemic countries. and dengue, a total of $18 billion is targeted for the period At most, it would cover the investment required in LICs. 2015–30, or about US$2.1 billion per year. Most of the As reflected in the discourse around the SDGs, the time investment in vector control is required in upper-middle- has come for a shift in focus from foreign donors to income countries (UMICs). Excluding vector control, domestic investment by governments and stakeholders in 424 Major Infectious Diseases endemic countries. The third WHO global NTD report NTDs, 96 percent of the health gains would be attributed argued that targets for domestic investment should be set to averted disability, and within the intensified-disease- such that the realization of the end of NTDs will not management NTDs, 95 percent of the impact would be depend disproportionately on foreign aid (WHO 2015b). realized from averted deaths. In 2011, the domestic share of total expenditure on The impact compares favorably to the total invest- health was 71 percent in LICs, 98 percent in lower- ment of US$27 billion thought to be required in the middle-income countries, and more than 99 percent period 2011–30 for achievement of global targets for in UMICs (WHO 2014b). Allowing for an upward those nine NTDs (our calculations based on abovemen- trend toward 2030 in line with recent trajectories in eco- tioned targets for 2015–30). That investment implies nomic growth, the domestic share in LICs could rise to US$45 per DALY averted or US$178 per irreversible 93 percent by 2030 (WHO 2015b). Of course, this aver- disease manifestation averted over the same period. age conceals considerable variation between countries. In addition to their impact as measured by DALYs, Nonetheless, if recent trajectories in economic growth NTDs are known to cause financial hardship among are maintained, by 2030 the domestic share could exceed affected individuals, which can exacerbate the cycle of pov- 80 percent in all of them (WHO 2015b). We apply these erty. A conservative estimate suggests that the end of domestic shares to the total investment target for NTDs NTDs would avert a total of international dollars (I$) 35 to obtain domestic investment targets for NTDs. billion in out-of-pocket (OOP) health expenditure by For all income groups, domestic investment targets affected individuals between 2011 and 2030 (Lenk and for NTDs decrease after 2020 in absolute (dollar) terms others, forthcoming; Redekop and others, forthcoming). as coverage targets are achieved and NTDs are con- This averted cost includes medicines, tests, travel, and food trolled, eliminated, or eradicated. These targets for not covered by public providers or by health insurance. It domestic investment are affordable. The domestic does not capture the additional averted costs of household investment target for NTDs represents less than one- coping mechanisms, such as indebtedness, or the irrevers- tenth of 1 percent of domestic expenditure on health ible consequences of catastrophic health expenditure. expected within the group of lower-middle-income Progress toward the end of NTDs would avert a further countries for 2015–30. The percentage is highest for the I$622 billion in wages lost by affected individuals between group of LICs, where the domestic investment target 2011 and 2030. This number does not include the signifi- for NTDs is nevertheless still well below 1 percent of cant long-term benefits of school attendance for employ- domestic expenditure on health. ment (Ahuja and others 2015). The choice by the Global In many endemic countries, the costs of community- Burden of Disease (2010) project (on which the analysis level interventions are borne by endemic communities was based) not to include so-called subtle morbidities, such themselves, who provide the volunteers and incentives. as impaired cognitive development, poor mental health In addition to financing delivery, all levels of govern- from stigma, and discrimination because of disfigurement, ment have a critical role in ensuring that nonfinancial is controversial. The benefit to affected individuals of barriers to access are also addressed. They can do this by averted OOP health expenditures and lost wages is there- providing waivers and supporting drug logistics through fore conservatively estimated to be I$657 billion between all administrative levels, especially at the very critical 2011 and 2030, or an average of I$33 billion per year. level of the community, where delivery occurs. We convert the benefits reported by Redekop and The end of NTDs is an achievable and affordable others (forthcoming) and Lenk and others (forthcom- SDG target for which endemic countries could take ing) from 2010 international dollars to 2015 U.S. dollars political and financial control. The returns to their poor- for direct comparison to the investment targets described est citizens will be substantial. in the previous section. Of course, some of this benefit is attributable to investments made before 2011. We conservatively assume RETURN ON INVESTMENTS TO END NTDs that investments begun in 1990 were at the level of those The health impact of meeting the WHO 2020 targets in 2011 (in real terms); 1990 is assumed to mark the and the end of NTDs by 2030 has been calculated for beginning of concerted global efforts to control most nine NTDs (de Vlas and others 2016). Between 2011 and NTDs and 2011 is assumed to mark the beginning of the 2030, 600 million DALYs would be averted: 30 million recent scale-up in investment to eliminate them. In real- DALYs per year on average (figure 17.5). These health ity, investments before 2011 were probably lower than gains include about 150 million irreversible disease this in most countries. We do not consider the invest- manifestations averted (such as blindness) and 5 million ments in improved housing and water and sanitation averted deaths. Among the preventive-chemotherapy that occurred over the same period; these investments An Investment Case for Ending Neglected Tropical Diseases 425 were not targeted at the NTDs but nonetheless contrib- Lower net benefits and rates of return in LICs are uted to their control. due in large part to the way in which averted productiv- We then calculate a rough estimate of the net benefit ity losses have been valued, that is, using GDP per capita to affected individuals from 1990 to 2020 (NTD Roadmap of the bottom 20 percent of the population of each targets) and to 2030 (the SDG target). Net benefit per country. This approach assigns a lower benefit to dollar invested is the present value of the benefit to lower-income and more unequal countries. Good phys- affected individuals minus the present value of the cost to ical health without disability is arguably more impor- public and philanthropic funders, divided by the present tant in LICs with large informal sectors that revolve value of the cost to funders. We apply a discount rate of around subsistence. As a result, the numbers are partic- 3 percent per year for both costs and benefits. ularly conservative estimates of the net benefit for The net benefit to affected individuals is US$17 for affected individuals in LICs. every dollar invested by funders during the period What is clear from even this preliminary analysis is 1990–2020 and US$28 for every dollar invested in the that investment in interventions against NTDs is a period 1990–2030 (table 17.8). It ranges from US$8 per fair and efficient investment in social justice. The bene- dollar invested in Africa to US$398 per dollar invested in fit to affected individuals—the poorest and most the Western Pacific (including China), and US$4 per marginalized—greatly exceeds the cost to public and dollar invested in LICs to US$273 per dollar invested in philanthropic funders of providing it. If the new social UMICs. compact articulated at the Addis Ababa Conference on Taking into account the period during which the Financing for Development is to involve transfers to the investments and returns are to be made, we also calculate poor (as SDGs 1 and 10 on ending poverty and reducing an annualized compounded rate of return. The end of inequalities suggest that it should), then ending NTDs is NTDs offers a 31 percent rate of return overall. It ranges an efficient way of making those transfers. from 19 percent per year in the Eastern Mediterranean This benefit can be measured by OOP health expen- Region to 59 percent in the Western Pacific (including diture and productivity losses averted. It thereby sup- China), and 14 percent in LICs to 54 percent in UMICs. ports two additional targets of the SDGs: universal Table 17.8 Benefits, Costs, Net Benefits, and Rates of Return on the End of Selected NTDs, Best Estimates Benefit to Affected Net Benefit to Affected Individuals, Cost to Funders, Individuals per Dollar Annualized Compounded 2015 US$ (billions) 2015 US$ (billions) Invested by Funders Rate of Return (%) 2015–2020 2015–2030 2015–2020 2015–2030 1990–2020 1990–2030 1990–2020 1990–2030 Preventive-chemotherapy NTDsa 119.7 399.0 2.8 6.2 27.4 42.8 35 35 b Disease-management NTDs 5.4 20.9 1.1 2.2 0.9 2.8 11 13 Total 125.1 419.9 3.9 8.4 16.9 28.4 30 31 African Region 9.2 40.6 1.5 3.0 3.0 8.0 18 20 Region of the Americas 5.6 21.9 0.1 0.3 10.2 26.1 24 25 Eastern Mediterranean Region 1.9 8.5 0.2 0.3 3.9 13.0 16 19 European Region 0.1 0.5 <0.1 <0.1 5.1 14.1 19 21 South-East Asian Region 27.1 98.9 1.9 4.5 4.9 10.1 19 21 Western Pacific Region 80.5 246.4 0.2 0.4 260.6 397.7 59 59 Low-income countries 7.5 29.3 1.3 2.5 1.3 4.0 11 14 Lower-middle-income countries 29.2 113.2 2.4 5.3 6.1 12.3 22 23 Upper-middle-income countries 87.6 274.3 0.3 0.5 165.4 272.7 54 54 Sources: Based on Lenk and others 2016; Redekop and others, forthcoming; WHO 2015b. Note: NTDs = neglected tropical diseases. a. Integrated delivery of preventive chemotherapy medicines for lymphatic filariasis, onchocerciasis, schistosomiasis, soil-transmitted helminthiases, and trachoma; also includes post-preventive chemotherapy surveillance and morbidity management and disability prevention. b. Individual management of human African trypanosomiasis (HAT), leprosy, and visceral leishmaniasis (VL); also includes active case finding for HAT, leprosy, and VL, and vector control for VL; includes the cost of integrated disease surveillance in HAT-endemic areas. 426 Major Infectious Diseases health coverage and social protection. Universal health Convergence in NTDs also stands out in the grand coverage means, among others, financial risk protection convergence narrative in its sustainability. Whereas the against OOP health expenditure. Social protection incremental costs of convergence increase from US$62 includes benefits for people of working age in case of billion in 2015 to US$86 billion in 2030, the costs of disability. As countries struggle with how to finance uni- convergence in NTDs decrease from US$750 million per versal health coverage and social protection, prioritizing year during 2015–20 to about US$300 million per year interventions to end NTDs can guide countries’ first during 2020–30. In this sense, there is a special role for decisive steps on the long path toward those goals. NTDs in leading the way toward the grand convergence Investment in interventions against NTDs puts progress through the elimination of infectious diseases targeted on universal health coverage within reach of even the by the SDGs. weakest health systems. CONCLUSIONS PLACE OF NTDS IN THE GRAND The elimination of the NTDs was a late and ad hoc CONVERGENCE addition to the MDG era, leaving a legacy of 22 million In 2013, the Lancet Commission on Investing in Health DALYs in 2012, a burden not far behind those of addressed the question of whether the world could malaria and TB. As we enter the SDG era the world is achieve a grand convergence, in which poorer countries seeking to rethink this opportunity and look toward the would see their infectious, maternal, and child health end of NTDs in 2030. The ambitious WHO NTD outcomes converge with the levels of wealthier nations— Roadmap (WHO 2012) and the massive donation of through increased investments in health interventions treatments for nine NTDs have built on successes in and systems to combat common causes of mortality and integrated treatment of multiple diseases in the poorest morbidity. There are now estimates of what the grand and most marginalized populations, and contribute to convergence might achieve and what investment would the potential of ending NTDs for as little as US$3 per be required by 2030 (Boyle and others 2015). DALY averted. Those estimates focus on lower-middle-income The evidence is clearly in favor of including the fol- countries. They consider the costs of scaling up interven- lowing within the package of essential interventions for tions for reproductive, maternal, and child health; and all low-income endemic countries (based on a cost per HIV/AIDS, TB, and malaria interventions; as well as the DALY averted of 2012 US$250 or less): preventive cost of strengthening health systems. They suggest that chemotherapy for at least five NTDs; comprehensive convergence would avert more than 130 million deaths control (including vector control) for visceral leishma- between 2015 and 2030. The incremental costs of con- niasis; and early detection and treatment of cutaneous vergence would be about US$62 billion in 2015, rising to leishmaniasis, HAT, and leprosy. Other interventions about US$86 billion in 2030. The end of NTDs is a against NTDs should also be included on a country- high-impact and low-cost contribution to the grand by-country basis. Populations requiring vector control convergence. for Chagas disease and dengue and mass treatment for Convergence in the burden of NTDs would avert yaws need to be mapped out; current evidence indicates about 519 million DALYs in the period 2015–30, includ- that these can be highly effective interventions in ing about 5 million deaths or 4 percent of the conver- lower-middle-income and upper-middle-income gence total of 130 million deaths (de Vlas and others endemic countries. 2016). As table 17.8 shows, the cost in lower-middle- Our estimates suggest that the costs of ending NTDs income countries is US$7.8 billion in 2015–30; this are affordable globally (for example, US$750 million per amount is US$0.5 billion per year or less than 1 percent year in 2015–20 and US$300 million per year in 2020– of the convergence total. The NTDs compare favorably 30, for preventive chemotherapy against five NTDs) and with other major diseases as measured by deaths that affordable for the governments of most endemic coun- could be averted by 2030 and the investment needed. tries at less than 0.1 percent of domestic health spending. This comparison is favorable even though up to The estimated benefits to affected individuals in averted 96 percent of the health gains from convergence in OOP health expenditures and lost productivity exceed NTDs would be in averted disability rather than death US$342 billion over the same period. The end of NTDs (de Vlas and others 2016). A more inclusive metric of offers a net benefit to affected individuals of about grand convergence would reveal the true contribution of US$25 for every dollar invested by funders—a 30 percent convergence in NTDs. annualized rate of return. It is a fair and efficient An Investment Case for Ending Neglected Tropical Diseases 427 investment in universal health coverage and social pro- Anoopa, S. D., C. Bern, B. Varghese, R. Chowdhury, R. Haque, tection for the least well-off. and others. 2006. “The Economic Impact of Visceral We compared NTDs with other public health Leishmaniasis on Households in Bangladesh.” Tropical programs by revisiting the Grand Convergence in Medicine and International Health 11 (5): 757–64. Baltussen, R., and A. Smith. 2012. “Cost-Effectiveness of Health by 2030, as proposed by the Lancet Commission Strategies to Combat Vision and Hearing Loss in Sub- on Investing in Health. Here the comparison is striking: Saharan Africa and South East Asia: Mathematical ending NTDs would avert about 4 percent of the con- Modelling Study.” BMJ 344 (1): e615. vergence total of 130 million deaths, and it would do so Baltussen, R., M. Sylla, K. D. Frick, and S. P. Mariotti. 2005. for less than 1 percent of the total convergence cost. “Cost-Effectiveness of Trachoma Control in Seven World This makes excellent health and economic sense on its Regions.” Ophthalmic Epidemiology 12 (2): 91–101. own, but it is an underestimate of the true scale of ben- Boelaert, M., F. Meheus, A. Sanchez, S. P. Singh, V. Vanlerberghe, efit given that it ignores the 96 percent of health gains and others. 2009. “The Poorest of the Poor: A Poverty attributable to averted disability, rather than death, Appraisal of Households Affected by Visceral Leishmaniasis which is equivalent to about 519 million DALYs averted in Bihar, India.” Tropical Medicine and International Health from 2015 to 2030. 14 (6): 639–44. Boyle, C. F., C. Levin, A. Hatefi, S. Madriz, and N. Santos. Coordinated efforts to end the NTDs are emerging 2015. “Achieving a ‘Grand Convergence’ in Global Health: as among the largest public health programs in the Modeling the Technical Inputs, Costs, and Impacts from world, and the most cost-effective and affordable. 2016 to 2030.” PLoS One 10 (10): e0140092. Indeed, the global program costs are on the scale of a Bundy, D. A. P., L. J. Appleby, M. Bradley, K. Croke, T. D. rounding error at less than 1 percent of the grand con- Hollingsworth, and others. 2017. “Mass Deworming vergence investment, yet offer a substantial return on Programs in Middle Childhood and Adolescence.” investment. Furthermore, because these are diseases of In Disease Control Priorities (third edition): Volume 8, poverty, the NTD agenda is specifically pro-poor; Child and Adolescent Health and Development, edited by because the programs target morbidity, they are also D. A. P. Bundy, N. de Silva, S. Horton, D. T. Jamison, and specifically pro-development. All in all, this suggests G. C. Patton. Washington, DC: World Bank. that NTD programs have a special role in leading the Castillo-Riquelme, M., Z. Chalabi, J. Lord, F. Guhl, D. Campbell-Lendrum, and others. 2008. “Modelling world’s efforts toward more fairness in health and the Geographic Variation in the Cost-Effectiveness of attainment of the SDGs. Control Policies for Infectious Vector Diseases: The Example of Chagas Disease.” Journal of Health Economics 27 (2): 405–26. NOTE Cattand, P., P. Desjeux, M. G. Guzman, J. Jannin, A. Kroeger, World Bank Income Classifications as of July 2014 are as and others. 2006. “Tropical Diseases Lacking Adequate follows, based on estimates of gross national income (GNI) Control Measures: Dengue, Leishmaniasis, and African per capita for 2013: Trypanosomiasis.” In Disease Control Priorities in Developing Countries (second edition), edited by D. T. Jamison, • Low-income countries (LICs) = US$1,045 or less J. G. Breman, A. R. Measham, G. Alleyne, M. Claeson, • Middle-income countries (MICs) are subdivided: D. B. Evans, P. Jha, A. Mills, and P. Musgrove. Washington, (a) lower-middle-income = US$1,046 to US$4,125 DC: World Bank and Oxford University Press. (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 Chandler, D. J., K. S. Hansen, B. Mahato, J. Darlong, A. John, • High-income countries (HICs) = US$12,746 or more. and others. 2015. “Household Costs of Leprosy Reactions (ENL) in Rural India.” PLoS Neglected Tropical Diseases 9 (1): e0003431. Constenla, D., C. Garcia, and N. Lefcourt. 2015. “Assessing the REFERENCES Economics of Dengue: Results from a Systematic Review Ahuja, A., S. Baird, J. H. Hicks, M. Kremer, E. Miguel, and others. of the Literature and Expert Survey.” Pharmacoeconomics 2015. “When Should Governments Subsidize Health? The 33 (11): 1107–35. Case of Mass Deworming.” World Bank Economic Review de Vlas, S. J., W. A. Stolk, E. A. le Rutte, J. A. C. Hontelez, 29 (Suppl 1): S9–24. R. Bakker, and others. 2016. “Concerted Efforts to Control Ahuja, A., S. Baird, J. H. Hicks, M. Kremer, and E. Miguel. or Eliminate Neglected Tropical Diseases: How Much 2017. “The Economics of Mass Deworming Programs.” Health Will Be Gained?” PLoS Neglected Tropical Diseases In Disease Control Priorities (third edition): Volume 8, 10 (2): e0004386. Child and Adolescent Health and Development, edited by Evans, D., D. McFarland, W. Adamani, A. Eigege, E. Miri, D. A. P. Bundy, N. de Silva, S. Horton, D. T. Jamison, and and others. 2011. “Cost-Effectiveness of Triple Drug G. C. Patton. Washington, DC: World Bank. Administration (TDA) with Praziquantel, Ivermectin, and 428 Major Infectious Diseases Albendazole for the Prevention of Neglected Tropical Economic Burden of Onchocerciasis on Households in Diseases in Nigeria.” Annals of Tropical Medicine and Enugu State, South-East Nigeria.” PLoS Neglected Tropical Parasitology 105 (8): 537–47. Diseases 9 (11): e0004231. Ezenduka, C., E. Post, S. John, A. Suraj, A. Namadi, and others. Idema, W. J., I. M. Majer, D. Pahan, L. Oskam, S. Polinder, and 2012. “Cost-Effectiveness Analysis of Three Leprosy Case others. 2010. “Cost-Effectiveness of a Chemoprophylactic Detection Methods in Northern Nigeria.” PLoS Neglected Intervention with Single Dose Rifampicin in Contacts of Tropical Diseases 6 (9): e1818. New Leprosy Patients.” PLoS Neglected Tropical Diseases Federici, C., C. Fitzpatrick, A. Be-nazir, F. Meheus, and D. Dagne. 4 (11): e874. Forthcoming. “The Cost-Effectiveness of a Comprehensive Keating, J., J. O. Yukich, S. Mollenkopf, and F. Tediosi. 2014. Programme to Eliminate Visceral Leishmaniasis in Bangladesh.” “Lymphatic Filariasis and Onchocerciasis Prevention, Fiedler, J. L., and R. Semakula. 2014. “An Analysis of the Treatment, and Control Costs across Diverse Settings: A Costs of Uganda’s Child Days Plus: Do Low Costs Reveal Systematic Review.” Acta Tropica 135 (1): 86–95. an Efficient Program or an Underfinanced One?” Food and Krotneva, S. P., L. E. Coffeng, M. Noma, H. G. Zouré, L. Bakoné, Nutrition Bulletin 35 (1): 92–104. and others. 2015. “African Program for Onchocerciasis Fitzpatrick, C., K. Asiedu, and J. Jannin. 2014. “Where the Road Control 1995–2010: Impact of Annual Ivermectin Mass Ends, Yaws Begins? The Cost-Effectiveness of Eradication Treatment on Off-Target Infectious Diseases.” PLoS versus More Roads.” PLoS Neglected Tropical Diseases Neglected Tropical Diseases 9 (9): 1–14. 8 (9): e3165. Lee, B. Y., K. M. Bacon, M. E. Bottazzi, and P. J. Hotez. Fitzpatrick, C., M. Madin-Warburton, F. M. Fleming, 2013. “Global Economic Burden of Chagas Disease: A T. Scheider, F. Mehus, K. Asiedu, A. Solomon, A. Montresor, Computational Simulation Model.” The Lancet Infectious G. Biswas. Forthcoming. “Benchmarking the Cost per Diseases 13 (4): 342–48. Person of Mass Treatment for Selected Neglected Tropical Lenk, E. J., W. K. Redekop, M. Luyendijk, C. Fitzpatrick, Diseases: An Approach Based on Literature Review and L. Niessen, and others. Forthcoming. “The Socioeconomic Meta-Regression with Web-Based Software Application.” Benefit to Individuals of Achieving the 2020 Targets for PLoS Neglected Tropical Diseases (under review). Neglected Tropical Diseases Controlled or Eliminated By Habtamu, E., T. Wondie, S. Aweke, Z. Tadesse, M. Zerihun, Innovative and Intensified Disease Management.” and others. 2015. “Trachoma and Relative Poverty: A Lenk, E. J., W. K. Redekop, M. Luyendijk, A. J. Rijnsburger, and Case-Control Study.” PLoS Neglected Tropical Diseases J. L. Severens. 2016. “Productivity Loss Related to Neglected 9 (11): e0004228. Tropical Diseases Eligible for Preventive Chemotherapy: Hofstraat, K., and W. H. van Brakel. 2016. “Social Stigma A Systematic Literature Review.” PLoS Neglected Tropical towards Neglected Tropical Diseases: A Systematic Review.” Diseases 10 (2): e0004397. International Health 8 (Suppl 1): i53–70. Leslie, J., A. Garba, K. Boubacar, Y. Yayé, H. Sebongou, and others. Horton, R. 2012. “GBD 2010: Understanding Disease, Injury, 2013. “Neglected Tropical Diseases: Comparison of the and Risk.” The Lancet 380 (9859): 2053–54. Costs of Integrated and Vertical Preventive Chemotherapy Hotez, P. J., M. Alvarado, M.-G. Basáñez, I. Bolliger, R. Bourne, Treatment in Niger.” International Health 5 (1): 78–84. and others. 2014. “The Global Burden of Disease Study 2010: Lo, N. C., I. I. Bogoch, B. G. Blackburn, G. Raso, E. K. N’Goran, Interpretation and Implications for the Neglected Tropical and others. 2015. “Comparison of Community-Wide, Diseases.” PLoS Neglected Tropical Diseases 8 (7): e2865. Integrated Mass Drug Administration Strategies for Hotez, P. J., D. A. P. Bundy, K. Beegle, S. Brooker, L. Drake, Schistosomiasis and Soil-Transmitted Helminthiasis: A and others. 2006. “Helminth Infections: Soil-Transmitted Cost-Effectiveness Modelling Study.” The Lancet Global Helminth Infections and Schistosomiasis.” In Disease Health 3 (10): e629–38. Control Priorities in Developing Countries (second edition), Luz, P. M., T. Vanni, J. Medlock, A. D. Paltiel, and A. P. Galvani. edited by D. T. Jamison, J. G. Breman, A. R. Measham, 2011. “Dengue Vector Control Strategies in an Urban G. Alleyne, M. Claeson, D. B. Evans, P. Jha, A. Mills, and Setting: An Economic Modelling Assessment.” The Lancet P. Musgrove, 1245–59. Washington, DC: World Bank and 377 (9778): 1673–80. Oxford University Press. Martelli, C. M. T., J. B. Siqueira, M. P. P. D. Parente, A. L. Zara, Houweling, T. A. J., H. E. Karim-Kos, M. C. Kulik, W. A. Stolk, C. S. Oliveira, and others. 2015. “Economic Impact of J. A. Haagsma, and others. 2016. “Socioeconomic Inequalities Dengue: Multicenter Study across Four Brazilian Regions.” in Neglected Tropical Diseases: A Systematic Review.” PLoS PLoS Neglected Tropical Diseases 9 (9): e0004042. Neglected Tropical Diseases 10 (5): e0004546. Meheus, F., A. A. Abuzaid, R. Baltussen, B. M. Younis, Hutton, G., and C. Chase. 2017. “Water Supply, Sanitation, M. Balasegaram, and others. 2013. “The Economic Burden and Hygiene.” In Disease Control Priorities (third edition): of Visceral Leishmaniasis in Sudan: An Assessment of Volume 7, Injury Prevention and Environmental Health, Provider and Household Costs.” American Journal of edited by C. N. Mock, O. Kobusingye, R. Nugent, and Tropical Medicine and Hygiene 89 (6): 1146–53. K. Smith. Washington, DC: World Bank. Meheus, F., and M. Boelaert. 2010. “The Burden of Visceral Ibe, O., O. Onwujekwe, B. Uzochukwu, M. Ajuba, and Leishmaniasis in South Asia.” Tropical Medicine and P. Okonkwo. 2015. “Exploring Consumer Perceptions and International Health 15 (Suppl 2): 1–3. An Investment Case for Ending Neglected Tropical Diseases 429 MSH (Management Sciences for Health). 2014. International Suaya, J. A., D. S. Shepard, M.-S. Chang, M. Caram, S. Hoyer, Drug Price Indicator Guide 2014. Medford, MA: Management and others. 2007. “Cost-Effectiveness of Annual Targeted Sciences for Health. http://erc.msh.org/mainpage.cfm?file=1 Larviciding Campaigns in Cambodia against the Dengue .0.htm&module=DMP&language=english. Vector Aedes aegypti.” Tropical Medicine and International Orellano, P. W., N. Vazquez, and O. D. Salomon. 2013. “Cost- Health 12 (9): 1026–36. Effectiveness of Prevention Strategies for American Sundar, S., R. Arora, S. P. Singh, M. Boelaert, and B. Varghese. Tegumentary Leishmaniasis in Argentina.” Cadernos de 2010. “Household Cost-of-Illness of Visceral Leishmaniasis Saúde Pública 29 (12): 2459–72. in Bihar, India.” Tropical Medicine and International Health Ozaki, M., S. Islam, K. M. Rahman, A. Rahman, S. P. Luby, 15 (Suppl 2): 50–54. and others. 2011. “Economic Consequences of Post- Sutherland, C. S., J. Yukich, R. Goeree, and F. Tediosi. 2015. “A kala-azar Dermal Leishmaniasis in a Rural Bangladeshi Literature Review of Economic Evaluations for a Neglected Community.” American Journal of Tropical Medicine and Tropical Disease: Human African Trypanosomiasis (‘Sleeping Hygiene 85 (3): 528–34. Sickness’).” PLoS Neglected Tropical Diseases 9 (2): e0003397. Redekop, W., E. J. Lenk, M. Luyendijk, C. Fitzpatrick, Louis Turner, H. C., A. A. Bettis, B. K. Chu, D. A. McFarland, Niessen, and others. Forthcoming. “The Socioeconomic P. J. Hooper, and others. 2016. “Investment Success in Public Benefit to Individuals of Achieving the 2020 Targets for Five Health: An Analysis of the Cost-Effectiveness and Cost- Preventive Chemotherapy Neglected Tropical Diseases.” Benefit of the Global Programme to Eliminate Lymphatic Reithinger, R., and P. G. Coleman. 2007. “Treating Cutaneous Filariasis.” Clinical Infectious Diseases. Epub December 12. Leishmaniasis Patients in Kabul, Afghanistan: Cost- doi10.1093/cid/ciw835. Effectiveness of an Operational Program in a Complex Turner, H. C., J. E. Truscott, T. D. Hollingsworth, A. A. Bettis, Emergency Setting.” BMC Infectious Diseases 7 (January 30): 3. S. J. Brooker, and others. 2015. “Cost and Cost-Effectiveness Remme, J. H. F., P. Feenstra, P. R. Lever, A. C. Medici, of Soil-Transmitted Helminth Treatment Programmes: C. M. Morel, and others. 2006. “Tropical Diseases Targeted Systematic Review and Research Needs.” Parasites and for Elimination: Chagas Disease, Lymphatic Filariasis, Vectors 8 (1): 355. Onchocerciasis, and Leprosy.” In Disease Control Priorities Turner, H. C., M. Walker, T. S. Churcher, M. Y. Osei- in Developing Countries (second edition), edited by Atweneboana, N.-K. Biritwum, and others. 2014. “Reaching D. T. Jamison, J. G. Breman, A. R. Measham, G. Alleyne, the London Declaration on Neglected Tropical Diseases M. Claeson, D. B. Evans, P. Jha, A. Mills, and P. Musgrove. Goals for Onchocerciasis: An Economic Evaluation of Washington, DC: World Bank and Oxford University Press. Increasing the Frequency of Ivermectin Treatment in Rilkoff, H., E. M. Tukahebwa, F. M. Fleming, J. Leslie, and Africa.” Clinical Infectious Diseases 59 (7): 923–32. D. C. Cole. 2013. “Exploring Gender Dimensions of Ugbomoiko, U. S., V. Dalumo, I. E. Ofoezie, and Treatment Programmes for Neglected Tropical Diseases R. N. N. Obiezue. 2009. “Socio-Environmental Factors and in Uganda.” PLoS Neglected Tropical Diseases 7 (7): e2312. Ascariasis Infection among School-Aged Children in Ilobu, Ruan, Y.-Z., R.-Z. Li, X.-X. Wang, L.-X. Wang, Q. Sun, and Osun State, Nigeria.” Transactions of the Royal Society of others. 2016. “The Affordability for Patients of a New Tropical Medicine and Hygiene 103 (3): 223–28. Universal MDR-TB Coverage Model in China.” International Undurraga, E. A., M. Betancourt-Cravioto, J. Ramos-Castañeda, Journal of Tuberculosis and Lung Disease 20 (5): 638–44. R. Martínez-Vega, J. Méndez-Galván, and others. 2015. Shepard, D. S., L. Coudeville, Y. A. Halasa, B. Zambrano, and “Economic and Disease Burden of Dengue in Mexico.” G. H. Dayan. 2011. “Economic Impact of Dengue Illness in PLoS Neglected Tropical Diseases 9 (3): e0003547. the Americas.” American Journal of Tropical Medicine and Uniting to Combat Neglected Tropical Diseases. 2014. Hygiene 84 (2): 200–7. Delivering on Promises and Driving Progress: Second Report Shepard, D. S., Y. A. Halasa, B. K. Tyagi, S. V. Adhish, D. Nandan, on Uniting to Combat NTDs. http://unitingtocombatntds and others. 2014. “Economic and Disease Burden of Dengue .org/report / delivering-promises-driving-progress Illness in India.” American Journal of Tropical Medicine and -second-report-uniting-combat-ntds. Hygiene 91 (6): 1235–42. Uranw, S., F. Meheus, R. Baltussen, S. Rijal, and M. Boelaert. Shepard, D. S., E. A. Undurraga, and Y. A. Halasa. 2013. 2013. “The Household Costs of Visceral Leishmaniasis Care “Economic and Disease Burden of Dengue in Southeast in South-Eastern Nepal.” PLoS Neglected Tropical Diseases Asia.” PLoS Neglected Tropical Diseases 7 (2): e2055. 7 (2): e2062. Stahl, H.-C., V. M. Butenschoen, H. T. Tran, E. Gozzer, Vazquez-Prokopec, G. M., C. Spillmann, M. Zaidenberg, R. Skewes, and others. 2013. “Cost of Dengue Outbreaks: U. Kitron, and R. E. Gürtler. 2009. “Cost-Effectiveness Literature Review and Country Case Studies.” BMC Public of Chagas Disease Vector Control Strategies in Northwestern Health 13 (November 6): 1048. Argentina.” PLoS Neglected Tropical Diseases 3 (1): e363. Stone, C. M., R. Kastner, P. Steinmann, N. Chitnis, M. Tanner, WHO (World Health Organization). 2010. Working to Overcome and others. 2016. “Modelling the Health Impact and Cost- the Global Impact of Neglected Tropical Diseases: First WHO Effectiveness of Lymphatic Filariasis Eradication under Report on Neglected Tropical Diseases. Geneva: WHO. Varying Levels of Mass Drug Administration Scale-Up ———. 2012. Accelerating Work to Overcome the Global and Geographic Coverage.” BMJ Global Health 1 (April 6): Impact of Neglected Tropical Diseases: A Roadmap for e000021. Implementation. Geneva: WHO. 430 Major Infectious Diseases ———. 2013. “World Health Assembly Resolution 66.12: ———. 2015c. PCT Databank. WHO, Geneva. http://www Neglected Tropical Diseases.” WHO, Geneva. http://www .who.int/neglected_diseases/preventive_chemotherapy .who.int/neglected_diseases/mediacentre/WHA_66.12 /databank/en/. _Eng.pdf?ua=1. ———. 2015d. “Kala-Azar Elimination Programme: Report of ———. 2014a. “Global Health Estimates 2014 Summary a WHO Consultation of Partners.” Geneva, WHO, February Tables: DALY by Cause, Age, Sex, by World Bank income 10-11. http://apps.who.int/iris/bitstream/10665/185042/1 Category, 2000–2012.” WHO, Geneva. http://www.who.int /9789241509497_eng.pdf. /healthinfo/global_burden_disease/en. ———. 2016. “World Health Assembly Resolution 69.21: ———. 2014b. World Health Statistics 2014. Geneva: WHO. Addressing the Burden of Mycetoma.” WHO, Geneva. http://www.who.int/gho/publications/world_health http://apps.who.int/gb/ebwha/pdf_files/WHA69/A69 _statistics/2014/en/. _R21-en.pdf. ———. 2015a. Generic Framework for Control, Elimination WHO-CHOICE. 2012. Choosing Interventions That Are Cost and Eradication of Neglected Tropical Diseases. Geneva: Effective (WHO-CHOICE). Geneva: WHO. http://www.who WHO. http://www.who.int/neglected_diseases/resources .int/choice/en/. /NTD_Generic_Framework_2015.pdf. Worrell, C., and E. Mathieu. 2012. “Drug Coverage Surveys for ———. 2015b. Investing to Overcome the Global Impact of Neglected Tropical Diseases: 10 Years of Field Experience.” Neglected Tropical Diseases: Third Report on Neglected American Journal of Tropical Medicine and Hygiene 87 (2): Diseases 2015. Geneva: WHO. 216–22. An Investment Case for Ending Neglected Tropical Diseases 431 Chapter 18 Drug-Resistant Infections Molly Miller-Petrie, Suraj Pant, and Ramanan Laxminarayan ANTIBIOTIC RESISTANCE which typically have high burdens and rapid spread of The global rise in antibiotic resistance threatens to undo infectious disease, poor nutrition, and increasing rates of decades of progress in treating infectious diseases caused antibiotic consumption in humans and animals, in addi- by bacterial pathogens (Laxminarayan and others 2006). tion to weaker health care systems and sparse standards Resistance to the drugs used to treat malaria, human and regulations that govern access, use, and quality of immunodeficiency virus/acquired immune deficiency antibiotics (Okeke, Klugman, and others 2005). syndrome (HIV/AIDS), and Mycobacterium tuberculosis Lack of access to antibiotics is still a serious concern is also a serious concern, with multidrug-resistant and for most LMICs. Pneumonia kills approximately 1 million extensively drug-resistant tuberculosis now documented children under age five each year, and an estimated worldwide, particularly in China, India, and the Russian 445,000 could be saved with the universal provision Federation (WHO 2014). These diseases and their treat- of antibiotics for community-acquired pneumococcal ments are covered in depth in other chapters in this infections (Laxminarayan and others 2016). When they volume. This chapter deals specifically with antibiotic are available, first-line antibiotic treatments are still rela- resistance in the One Health framework of humans, tively affordable, but newer antibiotics needed to treat animals, and the environment. resistant infections may be out of reach in low-resource Bacterial resistance to first-line, second-line, and settings (Laxminarayan and others 2016; Mendelson and last-resort antibiotics is growing, although rates and others 2015). trends vary by location, organism, and antibiotic (CDDEP At the same time, resistant infections are becoming 2016). Increased travel, trade, and migration mean that a significant cause of death, particularly for children. resistant bacteria can spread faster than ever (Du and Mortality is higher for children with drug-resistant infec- others 2016; Johnson and Woodford 2013). The Centers tions, such as methicillin-resistant Staphylococcus aureus for Disease Control and Prevention (CDC) considers (MRSA, a common skin and soft tissue infection) and Clostridium difficile (C. difficile), carbapenem-resistant infections caused by extended spectrum beta-lactamase- Enterobacteriaceae, and cephalosporin-resistant Neisseria producing bacteria (Kayange and others 2010). Though gonorrhoeae urgent threats to health in the United States data are limited, it is estimated that resistant sepsis infec- (CDC 2013). tions kill approximately 214,000 neonates each year, Drug-resistant infections are associated with higher primarily in India, Pakistan, Nigeria, the Democratic morbidity, mortality, and health expenditures (Okeke, Republic of Congo, and China (figure 18.1) (Laxminarayan Laxminarayan, and others 2005). The burden of resistance and others 2016). The prevalence of drug-resistant falls heavily on low- and middle-income countries (LMICs), infections in children is growing: the prevalence of Corresponding author: Ramanan Laxminarayan, Center for Disease Dynamics, Economics & Policy, Washington, DC, United States; ramanan@cddep.org.ss. 433 Figure 18.1 Estimated Neonatal Sepsis Deaths Caused by Bacteria Resistant to First-Line Antibiotics in High-Burden Countries 100 80 Number of deaths (thousands) 60 40 20 0 China Congo, Dem. Rep. Nigeria Pakistan India Gram-negative bacteria Gram-positive bacteria Total Source: Laxminarayan and others 2016. Note: Bars represent maximum and minimum values from Latin Hypercube Sampling model. carbapenem-resistant Enterobacteriaceae infections in Many infections, both drug-resistant and drug- children increased from 0 percent in 2000 to 5 percent in susceptible, are acquired in hospitals. Surgical site infec- 2012 in the United States (Logan and others 2015). tions account for one-third of all health care–associated Drug-resistant typhoidal and nontyphoidal Salmonella infections (HAIs) worldwide; they are the leading cause infections are also on the rise, with multidrug resistance of HAIs in low-resource settings (Allegranzi and others detected in 50 to 75 percent of nontyphoidal Salmonella 2011). Antibiotic prophylaxis can reduce the risk of isolates tested and up to 89 percent of Salmonella Typhi infection in patients undergoing surgery and chemo- isolates tested in Sub-Saharan Africa (Al-Emran and therapy, but antibiotic resistance threatens to undo these others 2016; Kariuki and others 2015). The Typhoid benefits (Teillant and others 2015). A 30 percent decrease Fever Surveillance in Africa Program also reported high in antibiotic efficacy would lead to an estimated 120,000 levels of resistance in invasive Salmonella to first-line additional infections and 6,300 additional deaths per treatments, in addition to multidrug-resistant strains year in the United States, according to a recent modeling (Baker, Hombach, and Marks 2016). However, incidence study (figure 18.2). Other HAIs, particularly sepsis, are a of resistant Salmonella infections varies widely across danger for both neonates and their mothers, who are the African region, indicating that national surveil- increasingly seeking birth and delivery care at hospitals lance systems are needed to guide treatment decisions at the advice of health care providers. Increasing rates of (Al-Emran and others 2016). Invasive Salmonella infec- drug resistance in these infections amplify this risk. tions are responsible for approximately 600,000 deaths a Despite limited national-level data and a lack of stan- year, particularly in children in low-resource settings. dardized methods for collecting resistance data, it is clear Case fatality rates from multidrug-resistant Salmonella that resistance is declining in some instances. Map 18.1 Typhi infections in South Asia are 10 percent— shows the global prevalence of MRSA. MRSA rates are approaching those of the pre-antibiotic era (Okeke, decreasing in Canada, Europe, South Africa, and the Laxminarayan and others 2005). United States (CDDEP 2016; EARS-Net 2014; Martin 434 Major Infectious Diseases Figure 18.2 Number of Additional Infections per Year Expected in the United States if Antibiotic Efficacy Decreased by 30 Percent Colorectal surgery Cesarean section Hysterectomy Transrectal prostate biopsy Spinal surgery Surgical abortion Appendectomy Total hip replacement Pacemaker implantation Cancer chemotherapy Hip fracture surgery 0 5,000 10,000 15,000 20,000 25,000 Number of additional infections per year Source: Teillant and others 2015. and others 2015). However, steep increases have been antibiotics. These drug-resistant bacteria can then spread detected in several LMICs, including India (box 18.1), to humans through direct contact with animals and and what data are available in Africa indicate that rates infected food, water, or waste (Marshall and Levy 2011). are also high in some populations in the region Emerging resistance mechanisms, such as the MCR-1 (Abdulgader and others 2015). Even less information is gene conferring plasmid-mediated resistance to colistin, available on MRSA and other bacterial infections in ani- a last-resort antibiotic, have been detected in both mals in LMICs. humans and animals and have the potential to render Resistance of Enterobacteriaceae to one of the last- even more infections untreatable (Liu and others 2016). resort antibiotics, carbapenems, is low but increasing in Overall, there is a significant lack of surveillance data on many LMICs (CDDEP 2016; Lerner and others 2015). resistance in humans and animals in most LMICs and in Extended spectrum beta-lactamases, which are resistant many high-income countries (HICs); these data are crit- to most beta-lactam antibiotics, are also increasingly ical to guiding policy making and clinical care. prevalent worldwide, including in Africa and Latin America (Datta and others 2012; Lu and others 2012; PAHO, forthcoming; Storberg 2014). Resistance to DRIVERS OF ANTIBIOTIC USE AND quinolones has been widely detected in enteric infections, RESISTANCE such as those caused by Escherichia coli (E. coli), in Sub- Saharan Africa, where quinolone use has increased since Use in Human Health Care generics became available in the early 2000s (Chattaway Every use of an antibiotic, whether appropriate or inap- and others 2016). Sexually transmitted infections, such as propriate, exerts selection pressure, giving resistant Neisseria gonorrhoeae, are increasingly resistant to all bacteria an advantage and accelerating the develop- available treatment options (Buono and others 2015). ment of resistance. Consumption of antibiotics is rising Resistant bacteria can also be detected in the guts of globally, but particularly in LMICs, as a result, in livestock or in water or soil that has been exposed to part, of rising incomes and increased access to drugs. Drug-Resistant Infections 435 Map 18.1 Percentage of Methicillin-Resistant Staphylococcus aureus Isolates, by Country, Most Recent Year, 2011–14 IBRD 42558 | DECEMBER 2016 0–19 60–79 20–39 80–100 40–59 No data Sources: Abdulgader and others 2015; CDDEP 2016 (https://resistancemap.cddep.org); PAHO, forthcoming; WHO 2014. Note: Where available, data from hospital-associated methicillin-resistant Staphylococcus aureus (MRSA) and invasive isolates are used. In their absence, data from community- associated MRSA or all specimen sources are included. Only countries that reported data for at least 30 isolates are shown. Depending on the country, at least one of the following drugs was used to test for MRSA: cefoxitin, cloxacillin, dicloxacillin, flucloxacillin, methicillin, and oxacillin. Intermediate-resistant isolates are considered resistant in some calculations, as in the original data source. Data from Abdulgader and others (2015) collected before and during 2011 were included as 2011. Box 18.1 Antibiotic Resistance in India India has high levels of both antibiotic consumption agriculture that same year. Consumption of faro- and antibiotic resistance. Antibiotic use is increasing in penem, an oral antibiotic structurally similar to car- conjunction with improved livelihoods and expanded bapenems, has increased by 154 percent since it was access to drugs, providing benefits for human health. approved for use in 2010 (Gandra and others 2016). Still, over-the-counter sales without a medical provid- Strains of Escherichia coli, Klebsiella pneumoniae, er’s prescription, weak regulation, and poor incentives Salmonella Typhi, and MRSA (methicillin-resistant for rational use are contributing to inappropriate use Staphylococcus aureus) have high resistance to most of antibiotics. Owing in part to poor health infrastruc- antibiotics, including carbapenems. Improved regula- ture, poor-quality drugs, and a high burden of disease, tion, behavior change for patients and providers, India’s rates of resistant infections are rising. incentives for improved prescribing practices, and increased surveillance have the potential to reverse India was the top consumer of antibiotics in 2010 current trends while preserving and increasing gains and one of the top global consumers of antibiotics in in access (Laxminarayan and Chaudhury 2016). 436 Major Infectious Diseases Human antibiotic consumption is also driven by the bur- incidence of C. difficile infection, which takes hold when den of infectious disease, as well as by economic, behav- antibiotic treatment has destroyed the normal gut flora. ioral, environmental, and structural factors. Expanded C. difficile is responsible for an estimated 14,000 deaths insurance coverage and increased physician density per year in the United States (CDC 2013). increase the consumption of antibiotics (Klein and As shown in map 18.2, it is estimated that antibiotic others 2015; Zhang, Lee, and Donohue 2010). Decision consumption increased more than 30 percent in 71 fatigue and patient demand also increase antibiotic pre- countries between 2000 and 2010, reaching approxi- scribing (Dosh and others 2000; Linder and others 2014). mately 70 billion standard units (single-dose units) in Antibiotic use is also correlated with season, increasing in 2010 (Van Boeckel and others 2014). This increase was the winter (December through February in the northern primarily in first-line classes of antibiotics, including hemisphere and August through September in the south- penicillins and cephalosporins, which together make up ern hemisphere) when the incidence of infectious disease more than half of global consumption. Use of last-resort is higher (Sun, Klein, and Laxminarayan 2012). In addi- antibiotic classes such as carbapenems and polymixins tion to driving resistance, antibiotic use increases the also increased. Although the consumption of antibiotics Map 18.2 Percentage Change in Antibiotic Consumption per Capita, by Country, 2000–10 IBRD 42559 | DECEMBER 2016 -50– -30 0–10 -29– -20 11–20 -19– -10 21–40 -9–0 41–80 >80 No data Source: Van Boeckel and others 2014, based on IMS MIDAS. Note: Data for Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama were available only as a group classified as Central America. Similarly, data for Benin, Burkina Faso, Cameroon, the Republic of Congo, Côte d’Ivoire, Gabon, Guinea, Mali, Senegal, and Togo were grouped and classified as French West Africa. The data for these countries represent the estimates for the corresponding regional groupings to which they belong. For countries with no data for 2000, the values for the earliest years for which data were available after 2000 were used to calculate the percentage changes. These countries and initial years are Algeria (2002), Bangladesh (2007), Croatia (2005), the Netherlands (2005), and Vietnam (2005). Much of the increase in antibiotic consumption in South Africa can be attributed to the use of co-trimoxazole as prophylaxis for HIV (human immunodeficiency virus) patients as recommended by the World Health Organization. Drug-Resistant Infections 437 declined in some HICs from 2000 to 2010, annual per greater in hospitals, which are home to a rotating popu- capita consumption in the United States is still one of the lation of seriously ill patients treated heavily with highest in the world, at approximately 22 standard units medications, including antibiotics. Infections treated in per person (map 18.2). Consumption grew fastest in hospitals may originate at the point of care (HAIs) or in middle-income countries, including Brazil, China, India, communities, where patients may have already been Russia, and South Africa. treated unsuccessfully with antibiotics. Human consumption of antibiotics is often inap- Suboptimal prescribing of antibiotics is common propriate in both LMICs and HICs, where antibiotics in both LMICs and HICs. In Nepal, an estimated 10 percent are purchased over the counter or prescribed incor- to 40 percent of antibiotic use is inappropriate; in Vietnam, rectly by a physician when they are not needed and will one-third of hospital prescriptions were deemed inappro- not be effective. Diarrheal and respiratory infections priate (Paudel, Sharma, and Das 2008; Shankar and others are frequently treated with antibiotics in the absence of 2007; Thu and others 2012). In six hospitals in the United diagnostics, even if such practice brings no benefit to States, initial antibiotic therapy was changed after five days patients; both consumers and prescribers may lack the in only one-third of patients, although 58 percent of awareness, education, or incentive to use antibiotics patients tested negative for bacterial infection (Braykov correctly. and others 2014). The use of carbapenems is rapidly An estimated 80 percent of antibiotics globally are increasing in hospitals (figure 18.3) (CDDEP 2016). purchased outside of hospitals (Kotwani and Holloway In some countries (for example, China), doctors or hospi- 2011). Although many of these antibiotics are purchased tals profit from the volume of antibiotics sold (Currie, Lin, without prescriptions, increased regulation to restrict and Meng 2014). sales may not be an appropriate solution for communi- ties that lack access to antibiotics. Interventions that target incentives linked to consumers, prescribers, and Use in Food Animal Production retailers and that educate the public and health care pro- At least two-thirds of all antibiotics, including those viders will be required to change consumption patterns important for human medicine, are estimated to be used in communities. in livestock production at subtherapeutic levels to pro- mote growth and prevent disease, and at therapeutic levels to treat disease (Laxminarayan, Van Boeckel, and Use in Hospitals Teillant 2015). Antibiotic use in animal agriculture is The volume of antibiotic consumption is greater in com- widespread and is estimated to increase by more than munities, but the clinical consequences of resistance are two-thirds between 2010 and 2030 (Van Boeckel and others 2015). This widespread use creates strong selec- tion pressure, encouraging the emergence and develop- Figure 18.3 Per Capita Carbapenem Use in the Hospital Sector, ment of resistance (Laxminarayan, Van Boeckel, and 2005–10 Teillant 2015). In animal production systems, antibiotics have long 80 been used in place of improved sanitation and hygiene to prevent disease. As incomes rise in LMICs, the Carbapenem sales per 1,000 population 70 in hospital sector (standard units) demand for animal-source food products will grow 60 (Gelband and others 2015). To meet this demand, more 50 farmers are using intensive production systems, with more animals raised in smaller spaces, leading to 40 greater reliance on antibiotics to prevent disease and promote growth. In the coming decades, antibiotic use 30 for food production is predicted to grow, as shown in 20 figure 18.4, with Brazil, China, India, Russia, and South Africa expected to experience an estimated 98 percent 10 increase collectively (Laxminarayan, Van Boeckel, and 0 Teillant 2015). Consumption is also increasing in some HICs. Figure 18.5 illustrates increasing antibiotic use in 05 06 07 08 09 10 20 20 20 20 20 20 United States Norway China Vietnam meat production in the United States, where antibiotic use per unit of meat has increased every year from 2009 Source: Based on IMS MIDAS. to 2014 (CDDEP 2016). 438 Major Infectious Diseases Figure 18.4 Estimated Growth in the Consumption of • Optimizing the use of antibiotics through steward- Antibiotics in Livestock, Top 10 Countries, 2010-30 ship and appropriate prescriptions 35,000 • Reshaping economic incentives to encourage research on new antibiotics while preserving existing drugs. 30,000 25,000 Some HICs, such as the United Kingdom and the Antibiotic use (tons) 20,000 United States, have implemented comprehensive national action plans. Fewer LMICs have such plans, but 15,000 in those LMICs that do, the plans have proved successful. 10,000 Progress in Ethiopia, Nepal, South Africa, and Vietnam 5,000 shows that action plans are both possible to formulate 0 in LMIC settings and are urgently needed; box 18.2 describes South Africa’s plan. ina St ited il ny ia ain ra an ico ce da az Ind a n de ssi na ex Ch Sp es n Br rm Fra Un tio Ca Fe Ru M at Ge 2010 2030 COST-EFFECTIVE INTERVENTIONS Source: Van Boeckel and others 2015. Although data on antibiotic use and resistance in LMICs are limited, new surveillance networks and laboratory- strengthening programs are being established. Online Figure 18.5 Antibiotic Use in Meat Production in the United databases like the Center for Disease Dynamics, States, 2009–14 Economics & Policy’s ResistanceMap1 are filling the gap 0.4 by providing data on use and resistance for a significant number of countries. Evidence on effective interventions 0.3 is limited by the diversity of resistance drivers and the Antibiotics per 1,000 kg of meat produce (kg) long timelines required to observe an effect on antibiotic resistance rates. A review found that the evidence base 0.2 for specific national interventions is insufficient to inform policies (Dar and others 2015), but areas for 0.1 action have clear potential benefits (Holloway 2011; Jamison and others 2013). 0 2009 2010 2011 2012 2013 2014 Water, Sanitation, and Hygiene Sources: USDA 2015; U.S. Department of Health and Human Services 2015. Note: kg = kilograms. In the United States, the burden of infectious diseases declined before the introduction of antibiotics, largely because of improvements in water and sanitation. NATIONAL POLICIES TO CONSERVE Many LMICs use antibiotics as a substitute for these ANTIBIOTIC EFFECTIVENESS improvements, but medications alone will not achieve the same reductions. Handwashing, clean water, and The World Health Assembly passed the Global Action excreta disposal reduce diarrheal diseases by 48 percent, Plan on Antimicrobial Resistance in May 2015 (WHO 17 percent, and 36 percent, respectively (Cairncross and 2015). The plan calls on member countries to establish others 2010). Handwashing also reduces respiratory national plans to address antibiotic consumption and infections by at least 16 percent (Rabie and Curtis resistance within the next two years. Plans are expected 2006). Handwashing by health care providers reduces to be context specific and ideally will encompass five HAIs and the development and spread of resistant strategic objectives: infections (Allegranzi and Pittet 2009; De Angelis and others 2014). • Improving awareness through education or aware- Sanitation and water quality require expensive infra- ness campaigns structure, but the savings from averted diseases offset • Strengthening existing evidence through surveillance most investments in water, sanitation, and hygiene and monitoring (Clasen and Haller 2008). Water supply and sanitation • Reducing the risk of infections through prevention interventions are cost-effective in all regions (Hutton, measures Haller, and Bartram 2007). Hand hygiene is highly Drug-Resistant Infections 439 Box 18.2 Comprehensive National Action Plans: The Example of South Africa The South African Department of Health launched hygiene, vaccination, and water and sanitation; a national strategic framework in 2014 (Department establishment of hospital antibiotic stewardship of Health, South Africa 2014) that adopted recom- programs; and review of antibiotics approved for mendations set by the World Health Organization use in animal feed. and the International Committee of the World Organisation for Animal Health to curb the threat Early detection and strong surveillance are essential of antibiotic resistance. Key elements of the frame- for capturing trends and creating evidence-based work include strengthening, coordinating, and public policies. South Africa’s national strategy institutionalizing interdisciplinary efforts through framework aims to strengthen the existing surveil- national and governance structures; optimizing lance system by developing an early warning system surveillance and early detection of antibiotic resis- for potential disease outbreaks and by tracking tance; enhancing infection prevention and control; trends and resistance patterns. In the long term, it and promoting appropriate use of antibiotics in will also use local, regional, and national resistance animal and human health. Several interventions patterns to optimize prescribing and report on have been proposed to ensure success, including resistance rates in food-producing and companion amendments to medical curricula; improved hand animals. cost-effective and the most affordable intervention at The introduction of the pneumococcal vaccine in the US$3.35 per disability-adjusted life year (DALY) averted United States reduced infection with both penicillin- (Cairncross and Valdmanis 2006). resistant and multidrug-resistant strains of Streptococcus pneumoniae—incidence of each type dropped more than 50 percent (Grijalva and others 2007; Kyaw and Vaccination others 2006). Similarly, in South Africa, introduction of The potential reduction in antibiotic use attributable to the vaccine reduced infection with penicillin- vaccination is considerable. An estimated 11 million days resistant strains by 67 percent and infection with of antibiotic treatment could be avoided by the universal trimethoprim-sulfamethoxazole-resistant strains by introduction of the pneumococcal vaccine (Laxminarayan 56 percent (Klugman and others 2003). Vaccines also and others 2016). Influenza in the United States is nearly reduce antibiotic resistance by reducing overall selection perfectly predicted by antibiotic sales data (Polgreen, pressure (Gelband and others 2015; Kwong and others Laxminarayan, and Cavanaugh 2011). An estimated 2009; Polgreen, Laxminarayan, and Cavanaugh 2011). 20 percent reduction in influenza would reduce antibi- Vaccines are being developed for commonly resistant infec- otic prescribing by 8 percent in the United States; a tions, such as those caused by Staphylococcus aureus, E. coli, universal vaccination program in Ontario, Canada, was and Klebsiella pneumoniae, as well as C. difficile, but have found to reduce influenza-associated antibiotic prescrib- faced significant challenges (Gelband and others 2015). ing by 64 percent (Kwong and others 2009; Polgreen, The cost-effectiveness of introducing new vaccines Laxminarayan, and Cavanaugh 2011). Influenza vacci- will differ by background burden of disease and existing nation also averts secondary bacterial infections, which cold chains; most vaccines are highly cost-effective, even have been associated with up to 40 percent of influenza though assessments rarely include the savings from cases that require hospitalization (Falsey and others reducing drug-resistant infections (Maurice and Davey 2013; McCullers 2014). The introduction of vaccines for 2009; Rheingans and others 2014). In India, health care viral and bacterial diarrheal infections, such as rotavirus cost savings offset the cost of vaccination against and cholera, are likely to achieve similar reductions in Haemophilus influenzae in all states (Clark and others antibiotic use (Ganguly and others 2011; Okeke 2009). 2013). Similarly, an assessment of cost-effectiveness in Vaccination can reduce the overall disease burden LMICs found that the Haemophilus influenzae type b and the incidence of drug-resistant infections. conjugate vaccine was cost saving or highly cost-effective 440 Major Infectious Diseases (Griffiths, Clark, and Hajjeh 2013), while the pneumo- program linked hospital payments to targets for reduced coccal vaccine was highly cost-effective in 68 of 72 devel- antibiotic use, alongside the introduction of prescribing oping countries (Sinha and others 2007). For diarrheal regulations, audits, and inspections. Following the pro- infections, the rotavirus vaccine provided economic gram, prescribing declined between 10 percent and benefits at diverse prices, while cholera vaccination had 12 percent (Xiao and others 2013). economic benefits in specific contexts (Rheingans and There are no studies that directly assess the cost- others 2014). effectiveness of incentives or regulations to reduce anti- biotic use. However, lessons can be drawn from several historical examples: persons with free medical care in the Antibiotic Stewardship and Infection Control United States used 85 percent more antibiotics than Programs in Hospitals did persons who paid for a portion of care, while cost Stewardship and infection control have been broadly sharing reduced both inappropriate and appropriate proposed as the most effective interventions for reducing antibiotic use (Foxman and others 1987). A free antibi- antibiotic use in all settings (Dar and others 2015). otics program introduced in the United States in Components of stewardship programs such as prescrip- 2006 increased antibiotic prescribing, with prescribers tion audits, peer comparisons, and guidelines have shifting to drugs covered under the program (Li and successfully reduced antibiotic use (Meeker and others Laxminarayan 2015). A prescription audit study in 2014; Meeker and others 2016). Hospital antibiotic stew- China found that financial incentives determined ardship programs aim to provide the correct dose of the prescribing patterns: physicians were more likely to most appropriate antibiotic at the right time; they have prescribe antibiotics when patients indicated they would also been shown to reduce antibiotic use by 11 percent to purchase the antibiotics at hospitals (Currie, Lin, and 38 percent and to be associated with modest reductions Meng 2014). in antibiotic-resistant infections (Kaki and others 2011). Antibiotics are often used as a substitute for infection prevention and control in hospitals. Infection control Reduction of Antibiotic Use in Agriculture programs reduce antibiotic use, in addition to reducing National regulation is one way to encourage reductions the prevalence and spread of HAIs. They are particularly in antibiotic use in food animal production. The use of important in LMICs, where rates of HAIs are two to antibiotics to promote growth has been banned in the three times higher than in Europe and the United States European Union since 2006. Although some farmers (Allegranzi and others 2011; Davey and others 2005; simply shifted to using antibiotics for prevention rather WHO 2011). Bundled interventions and handwashing than for growth, the bans resulted in decreasing or con- have been shown to reduce both HAIs and antibiotic use sistently low use of antibiotics in agriculture in many (Gelband and others 2015). European Union countries, including Sweden and Multifaceted infection control programs have been Denmark (Cogliani, Goossens, and Greko 2011). Few shown to be cost-effective in the United States (Dick LMICs have introduced such bans. and others 2015). Although evidence from LMICs Many countries have been reluctant to ban agricul- is limited, the higher burden of HAIs in LMICs indi- tural use of antibiotics because of potential adverse cates that these interventions would be highly effective economic impacts on their livestock sectors, but research there as well. The cost-effectiveness of antibiotic stew- shows that the impact is minimal in farming systems ardship programs may depend on the prevalence of that are already optimized with respect to the genetic antibiotic resistance. Extensive evidence indicates that potential of herds, hygiene, nutrition, and herd health stewardship programs reduce the length of hospital (Laxminarayan, Van Boeckel, and Teillant 2015). Some stays and expenditures on antibiotics (CDC 2016), and countries have banned or discouraged the use of certain that antibiotic stewardship teams can be cost-effective antibiotic growth promoters (table 18.1). The U.S. Food in improving care for bacteremia in the United States and Drug Administration issued guidelines in 2013 ask- (Scheetz and others 2009). ing the pharmaceutical industry to restrict the use of antibiotics as growth promoters (Laxminarayan, Van Boeckel, and Teillant 2015). Vaccination can also pre- Incentives for Rational Use vent disease and antibiotic use in animals, as it does Delinking the prescribing of antibiotics from financial in humans. incentives or creating financial incentives for improved Evidence from HICs suggests that the economic prescribing can reduce the use of antibiotics (Carbon and impact of removing growth promoters is low, but the Bax 1998; Song and others 2014). For example, a Chinese impact may be different in LMICs where antibiotics are Drug-Resistant Infections 441 Table 18.1 Regulation of Antibiotic Use in Livestock in OECD Countries Country Ban on antibiotic growth promoters Prescription required Australia No, but some AGPs banned (fluoroquinolones, avoparcin, Yes, prescription required for most veterinary and others) antibiotics Canada No, voluntary phaseout notice issued in April 2014 that No, plan to strengthen veterinary oversight mimics the U.S. FDA approach developed in line with the U.S. FDA approach Chile No data No data European Union member states Yes, all AGPs banned in 2006 Yes Israel No data No data Japan No Yes Mexico Yes, AGPs banned in 2007 with exceptions (avoparcin, Yes vancomycin, bacitracin, tylosin, virginiamycin, and others) New Zealand Yes, for critically and highly important antibiotics as listed Yes, for antibiotics identified as having by the WHO and the OIE potential for resistance problems Korea, Rep. Yes, AGP use discontinued in 2011 until a veterinary Yes, veterinary oversight system in oversight system can be put in place development Turkey No data No data United States No, voluntary guidelines to withdraw the use of No, under new U.S. FDA guidance, use of medically important antibiotics as promoters released by medically important antibiotics to be under the U.S. FDA in 2013 oversight of licensed growth veterinarians Sources: Table adapted from Laxminarayan, Van Boeckel, and Teillant 2015. Australian Commission on Safety and Quality in Health Care 2013; European Union 2003; Government of Canada 2014; MAF New Zealand 2011; Maron, Smith, and Nachman 2013; USDA 2011, 2013; U.S. FDA 2013. Note: AGP = antibiotic growth promoter; OECD = Organisation for Economic Co-operation and Development; OIE = World Organisation for Animal Health; U.S. FDA = U.S. Food and Drug Administration; WHO = World Health Organization. still used in place of hygiene and other measures to opti- cost-effectiveness of such campaigns has not been for- mize production. In China, the economic costs of a ban mally assessed, especially in LMICs. could reach billions of dollars but would still be signifi- Education without incentives or oversight may be less cantly less than the economic burden of resistant infec- successful at changing behavior. Interventions targeted tions (Laxminarayan, Van Boeckel, and Teillant 2015). toward prescribers and consumers that combined edu- cation with managerial oversight have been more successful at reducing inappropriate antibiotic use than Education and Awareness have educational materials alone (Holloway 2011). In Campaigns have successfully reduced human antibiotic Asia, health care provider interventions to increase use in HICs, but evidence from LMICs is sparse (Huttner appropriate use have typically been more successful than and others 2010). Awareness campaigns in France and have other interventions to reduce use, unless they have Belgium reduced antibiotic prescribing by 27 percent been accompanied by oversight mechanisms such as and 36 percent, respectively, and both programs saw peer review (Holloway 2011). some reduction in resistant pneumococci (Sabuncu and others 2009). Some LMICs have begun to initiate awareness campaigns coinciding with the CDC’s Get Surveillance of Antibiotic Use and Resistance Smart About Antibiotics Week and the World Health Surveillance of antibiotic use and resistance in humans Organization’s (WHO) World Antibiotic Awareness Week and animals is needed to inform clinical decision making (Global Antibiotic Resistance Partnership 2015). Few and national policies. Although few LMICs have com- national-level campaigns have specifically targeted use prehensive national surveillance programs, private sector in animals. As with antibiotic stewardship programs, laboratories often collect detailed data on resistance and awareness campaigns that reduce use also reduce are an underused resource in many data-scarce areas. expenditures on antibiotics, and the campaigns in These laboratories provide valuable additions to public Belgium and France were associated with significant sector resistance data in countries such as India and savings (Huttner and others 2010). However, the South Africa, where more than 98 percent and 80 percent, 442 Major Infectious Diseases respectively, of accredited medical laboratories are in the Partnership, which develops local capacity in LMICs to private sector (CDDEP 2016; Gandra, Merchant, and design and implement national antibiotic resistance Laxminarayan 2016). plans, has identified six key interventions to curb antibi- Regional surveillance networks collect data in Latin otic resistance (figure 18.6). America (Red Latinoamericana de Vigilancia de la The following recommendations should be part of Resistencia a los Antimicrobianos [ReLAVRA]), Asia any national plan: (Asian Network for Surveillance of Resistant Pathogens [ANSORP]), Central Asia and Eastern Europe • Reduce and eventually phase out subtherapeutic anti- (Central Asian and Eastern European Surveillance of biotic use in agriculture. Improved sanitation and Antimicrobial Resistance [CAESAR]), and Europe hygiene at the farm level would reduce the need for (European Antimicrobial Resistance Surveillance prophylactic antibiotics. Antibiotic use in animal Network [EARS-Net]). ResistanceMap, a global repos- agriculture should be reduced, focusing on involving itory of antibiotic resistance and use data, incorpo- farmers and the agricultural industry in carefully rates these and other data from surveillance networks phasing out the use of growth promoters and pre- and private laboratories in LMICs into a visualization mixed animal feeds (Laxminarayan, Van Boeckel, and interface that can be used to assess national patterns Teillant 2015). and trends. • Adopt incentives that encourage antibiotic overuse and misuse to incentives that encourage antibiotic steward- New Drugs and Antibiotic Alternatives ship. Making sure that payments are not linked to prescribing, as well as introducing rewards for com- The pipeline of new antibiotics is relatively robust, with pliance, may improve prescribing patterns. 7 new antibiotics approved in 2014 and 37 under devel- • Improve hospital infection control and antibiotic stew- opment (Pew Charitable Trusts 2014); however, new ardship. Antibiotic stewardship programs, infection drugs will always be needed as resistance develops, par- prevention and control, and especially handwashing ticularly for serious threats such as Gram-negative bacte- with soap can reduce infections, antibiotic use, and ria and for use in low-resource settings. The financial resistance while improving patient outcomes. incentives to develop antibiotics are limited compared • Educate health professionals, policy makers, and the with other drugs because of their short course of use and public on sustainable antibiotic use. Although public potentially restricted use, among other barriers. Several awareness is growing that antibiotic resistance presents initiatives—including the United Kingdom’s Review on a threat, there is little awareness of the individual Antimicrobial Resistance; the Driving Reinvestment in actions that can be taken to reduce use. Patients, par- Research and Development and Responsible Antibiotic ents, health care providers, stakeholders, and hospital Use (DRIVE-AB) partnership funded by the Innovative leaders all need to be aware of what they can do to Medicines Initiative; and the Global Antibiotic Research reduce unnecessary use. and Development Partnership, established under • Reduce the need for antibiotics through improved the auspices of the Drugs for Neglected Diseases initiative water, sanitation, and immunization. Disease preven- (DNDi) and the WHO—are seeking to address these bar- tion achieves the dual purposes of keeping people riers to ensure a robust antibiotic pipeline (O’Neill 2016). healthy and saving antibiotic doses. Water, sanitation, Because resistance to all antibiotics naturally hygiene, and vaccination should be core components develops, alternatives to antibiotics present other of any public health system. options. Alternatives include vaccines and improved • Ensure political commitment to meet the threat of diagnostics, as well as antibodies, probiotics, lysins, antibiotic resistance. Without national commitment bacteriophages, immune simulation, and peptides. in the form of implemented action plans, the long- Development of a complete portfolio of these term sustainability of efforts to curb antibiotic alternatives will take an estimated 10 years and cost a resistance will be weakened. Although international minimum of US$2.1 billion (Czaplewski and others 2016). efforts to curb antibiotic resistance have focused largely on national action, international support is also needed, particularly to stimulate private and CONCLUSIONS public sector research to fill knowledge gaps and National action, tailored to local contexts and patterns develop new drugs, diagnostics, and other tech- of resistance, is key to curbing the global threat of anti- nologies, as well as to strengthen laboratories for biotic resistance. The Global Antibiotic Resistance improved surveillance. Drug-Resistant Infections 443 Figure 18.6 Six Strategies Needed in National Antibiotic Policies 1. Reduce, and eventually phase out, subtherapeutic antibiotic use in agriculture. 2. Adopt incentives that encourage antibiotic overuse and misuse to incentives that encourage antibiotic stewardship. 3. Improve hospital infection control and antibiotic stewardship. 4. Educate health professionals, policy makers, and the public on sustainable antibiotic use. 5. Reduce the need for antibiotics through improved water, sanitation, and immunization. 6. Ensure political commitment to meet the threat of antibiotic resistance. Source: Gelband and others 2015. NOTES to Ciprofloxacin in Sub-Saharan Africa.” Clinical Infectious Diseases 62 (Suppl 1): S42–46. doi:10.1093/cid/civ788. World Bank Income Classifications as of July 2014 are as Allegranzi, B., S. Bagheri Nejad, C. Combescure, W. Graafmans, follows, based on estimates of gross national income (GNI) H. Attar, and others. 2011. “Burden of Endemic Health-Care- per capita for 2013: Associated Infection in Developing Countries: Systematic Review and Meta-Analysis.” The Lancet 377 (9761): 228–41. • Low-income countries (LICs) = US$1,045 or less Allegranzi, B., and D. Pittet. 2009. “Role of Hand Hygiene in • Middle-income countries (MICs) are subdivided: Healthcare-Associated Infection Prevention.” Journal of (a) lower-middle-income = US$1,046 to US$4,125 Hospital Infection 73 (4): 305–15. (b) upper-middle-income (UMICs) = US$4,126 to US$12,745 Australian Commission on Safety and Quality in Health Care. • High-income countries (HICs) = US$12,746 or more. 2013. “Australian One Health Antimicrobial Resistance Colloquium.” Australian Commission of Safety and Quality Data from figure 18.3 and map 18.2 are from IMS MIDAS in Health Care, Canberra. International Prescription Data, January 2000–December Baker, S., J. Hombach, and F. Marks. 2016. “What Have We 2010, IMS Health Incorporated. All Rights Reserved. The Learned from the Typhoid Fever Surveillance in Africa statements, findings, conclusions, views, and opinions con- Program?” Clinical Infectious Diseases 62 (Suppl 1): S1–S3. tained and expressed herein are not necessarily those of IMS doi:10.1093/cid/civ675. Health Incorporated or any of its affiliated or subsidiary Braykov, N. P., D. J. Morgan, M. L. Schweizer, D. Z. Uslan, entities. T. Kelesidis, and others. 2014. “Assessment of Empirical 1. https://resistancemap.cddep.org. Antibiotic Therapy Optimisation in Six Hospitals: An Observational Cohort Study.” The Lancet Infectious Diseases 14 (12): 1220–27. Buono, S. A., T. D. Watson, L. A. Borenstein, J. D. Klausner, REFERENCES M. W. Pandori, and others. 2015. “Stemming the Tide Abdulgader, S. M., A. O. Shittu, M. P. Nicol, and M. Kaba. of Drug-Resistant Neisseria gonorrhoeae: The Need for 2015. “Molecular Epidemiology of Methicillin-Resistant an Individualized Approach to Treatment.” Journal of Staphylococcus aureus in Africa: A Systematic Review.” Antimicrobial Chemotherapy 70 (2): 374–81. Frontiers in Microbiology 6 (348): 1–21. Cairncross, S., C. Hunt, S. Boisson, K. Bostoen, V. Curtis, Al-Emran, H. M., D. Eibach, R. Krumkamp, M. Ali, S. Baker, and others. 2010. “Water, Sanitation, and Hygiene for and others. 2016. “A Multicountry Molecular Analysis of 4 the Prevention of Diarrhoea.” International Journal of Salmonella enterica Serovar Typhi with Reduced Susceptibility Epidemiology 39 (Suppl 1): i193–205. 444 Major Infectious Diseases Cairncross, S., and V. Valdmanis. 2006. “Water Supply, Department of Health, South Africa. 2014. Antimicrobial Sanitation, and Hygiene Promotion.” In Disease Control Resistance National Strategy Framework 2014–2024. Priorities in Developing Countries (second edition), Pretoria: National Department of Health. edited by D. T. Jamison, J. G. Breman, A. R. Measham, Dick, A. W., E. N. Perencevich, M. Pogorzelska-Maziarz, G. Alleyne, M. Claeson, D. B. Evans, P. Jha, A. Mills, and J. Zwanziger, E. L. Larson, and others. 2015. “A Decade of P. Musgrove. Washington, DC: World Bank and Oxford Investment in Infection Prevention: A Cost-Effectiveness University Press. Analysis.” American Journal of Infection Control 43 (1): 4–9. Carbon, C., and R. P. Bax. 1998. “Regulating the Use of Dosh, S. A., J. M. Hickner, A. G. Mainous, and M. H. Ebell. Antibiotics in the Community.” BMJ 317 (7159): 663–65. 2000. “Predictors of Antibiotic Prescribing for Nonspecific CDC (Centers for Disease Control and Prevention). 2013. Upper Respiratory Infections, Acute Bronchitis, and Acute Antibiotic Resistance Threats in the United States, 2013. Sinusitis.” Journal of Family Practice 49 (5): 407–14. Atlanta, GA: CDC. http://www.cdc.gov/drugresistance Du, H., L. Chen, Y. W. Tang, and B. N. Kreiswirth. 2016. /threat-report-2013/. “Emergence of the MCR-1 Colistin Resistance Gene in ———. 2016. “Impact of Antibiotic Stewardship Program Carbapenem-Resistant Enterobacteriaceae.” The Lancet Interventions on Costs.” CDC, Atlanta, GA. Infectious Diseases 16 (3): 287–88. CDDEP (Center for Disease Dynamics, Economics & Policy). EARS-Net (European Antimicrobial Resistance Surveillance 2016. “Resistance Map.” CDDEP, Washington, DC. http:// Network). 2014. “EARS-Net Report, Quarters 1–4.” Health www.cddep.org/projects/resistance-map. Protection Surveillance Centre, Dublin. Chattaway, M. A., A. O. Aboderin, K. Fashae, C. K. Okoro, European Union. 2003. “Regulation (EC) No. 1831/2003 of the J. A Opintan, and others. 2016. “Fluoroquinolone-Resistant European Parliament and of the Council of 22 September Enteric Bacteria in Sub-Saharan Africa: Clones, Implications 2003 on Additives for Use in Animal Nutrition.” European and Research Needs.” Frontiers in Microbiology 7: 558. Union, Brussels. doi:10.3389/fmicb.2016.00558. Falsey, A. R., K. L. Becker, A. J. Swinburne, E. S. Nylen, Clark, A. D., U. K. Griffiths, S. S. Abbas, K. D. Rao, L. Privor- M. A. Formica, and others. 2013. “Bacterial Complications Dumm, and others. 2013. “Impact and Cost-Effectiveness of Respiratory Tract Viral Illness: A Comprehensive of Haemophilus influenzae Type B Conjugate Vaccination in Evaluation.” Journal of Infectious Diseases 208 (3): 432–41. India.” Journal of Pediatrics 163 (Suppl 1): S60–72. Foxman, B., R. B. Valdez, K. N. Lohr, G. A. Goldberg, J. P. Clasen, T. F., and L. Haller. 2008. “Water Quality Interventions Newhouse, and others. 1987. “The Effect of Cost Sharing to Prevent Diarrhoea: Cost and Cost-Effectiveness.” WHO/ on the Use of Antibiotics in Ambulatory Care: Results from HSE/WSH/08.02, World Health Organization, Geneva. a Population-Based Randomized Controlled Trial.” Journal Cogliani, C., H. Goossens, and C. Greko. 2011. “Restricting of Chronic Diseases 40 (5): 429–37. Antimicrobial Use in Food Animals: Lessons from Europe.” Gandra, S., E. Y. Klein, S. Pant, S. Malhotra-Kumar, and Microbe 6 (6): 274–79. R. Laxminarayan. 2016. “Faropenem Consumption Is Currie, J., W. Lin, and J. Meng. 2014. “Addressing Antibiotic Increasing in India.” Clinical Infectious Diseases 62 (8): Abuse in China: An Experimental Audit Study.” Journal of 1050–52. doi:10.1093/cid/ciw055. Development Economics 110 (September): 39–51. Gandra, S., A. Merchant, and R. Laxminarayan. 2016. “A Role Czaplewski, L., R. Bax, M. Clokie, M. Dawson, H. Fairhead, and for Private Laboratories in Public Health Surveillance of others. 2016. “Alternatives to Antibiotics: A Pipeline Portfolio Antimicrobial Resistance.” Future Microbiology 11: 709–12. Review.” The Lancet Infectious Diseases 16 (2): 239–51. Ganguly, N. K., N. K. Arora, S. J. Chandy, M. N. Fairoze, J. P. Dar, O., R. Hasan, J. Schlundt, S. Harbarth, G. Caleo, and S. Gill, and others. 2011. “Rationalizing Antibiotic Use to others. 2015. “Exploring the Evidence Base for National and Limit Antibiotic Resistance in India.” Indian Journal of Regional Policy Interventions to Combat Resistance.” The Medical Research 134 (3): 281–94. Lancet 387 (10015): 285–95. Gelband, H., M. Miller-Petrie, S. Pant, S. Gandra, J. Levinson, Datta, S., C. Wattal, N. Goel, J. K. Oberoi, R. Raveendran, and and others. 2015. The State of the World’s Antibiotics, 2015. others. 2012. “A Ten-Year Analysis of Multi-Drug Resistant Washington, DC: Center for Disease Dynamics, Economics Blood Stream Infections Caused by Escherichia coli and & Policy. Klebsiella pneumoniae in a Tertiary Care Hospital.” Indian Global Antibiotic Resistance Partnership. 2015. GARPNet News Journal of Medical Research 135 (6): 907–12. 1 (6). Center for Disease Dynamics, Economics & Policy, Davey, P., E. Brown, L. Fenelon, R. Finch, I. Gould, and others. Washington, DC. 2005. “Interventions to Improve Antibiotic Prescribing Government of Canada. 2014. “Collaborative Efforts to Promote Practices for Hospital Inpatients (Review).” Cochrane the Judicious Use of Medically Important Antimicrobial Database of Systematic Reviews (4): CD003543. Drugs in Food Animal Production.” Government of De Angelis, G., M. A. Cataldo, C. De Waure, S. Venturiello, G. La Canada, Ottawa. Torre, and others. 2014. “Infection Control and Prevention Griffiths, U. K., A. Clark, and R. Hajjeh. 2013. “Cost- Measures to Reduce the Spread of Vancomycin-Resistant Effectiveness of Haemophilus influenzae Type B Conjugate Enterococci in Hospitalized Patients: A Systematic Review Vaccine in Low- and Middle-Income Countries: Regional and Meta-Analysis.” Journal of Antimicrobial Chemotherapy Analysis and Assessment of Major Determinants.” Journal 69 (5): 1185–92. of Pediatrics 163 (Suppl 1): S50–59.e9. Drug-Resistant Infections 445 Grijalva, C. G., J. P. Nuorti, P. G. Arbogast, S. W. Martin, Priorities in Developing Countries (second edition), K. M. Edwards, and others. 2007. “Decline in Pneumonia edited by D. T. Jamison, J. G. Breman, A. R. Measham, Admissions after Routine Childhood Immunization with G. Alleyne, M. Claeson, D. B. Evans, P. Jha, A. Mills, and Pneumococcal Conjugate Vaccine in the USA: A Time P. Musgrove. Washington, DC: World Bank and Oxford Series Analysis.” The Lancet 369 (9568): 1179–86. University Press. Holloway, K. A. 2011. “Promoting the Rational Use of Laxminarayan, R., and R. R. Chaudhury. 2016. “Antibiotic Antibiotics.” Regional Health Forum 15 (1): 122–30. Resistance in India: Drivers and Opportunities for Action.” Huttner, B., H. Goossens, T. Verheij, and S. Harbarth. 2010. PLoS Medicine 13 (3): e1001974. “Characteristics and Outcomes of Public Campaigns Laxminarayan, R., P. Matsoso, S. Pant, C. Brower, J.-A. Rottingen, Aimed at Improving the Use of Antibiotics in Outpatients and others. 2016. “Access to Effective Antimicrobials: in High-Income Countries.” The Lancet Infectious Diseases A Worldwide Challenge.” The Lancet 387 (10014): 168–75. 10 (1): 17–31. Laxminarayan, R., T. Van Boeckel, and A. Teillant. 2015. “The Hutton, G., L. Haller, and J. Bartram. 2007. “Global Cost-Benefit Economic Costs of Withdrawing Antimicrobial Growth Analysis of Water Supply and Sanitation Interventions.” Promoters from the Livestock Sector.” Food, Agriculture, Journal of Water and Health 5 (4): 481–502. and Fisheries Paper 78, Organisation for Economic Jamison, D. T., L. H. Summers, G. Alleyne, K. J. Arrow, Co-operation and Development, Paris. S. Berkley, and others. 2013. “Global Health 2035: A World Lerner, A., A. Adler, J. Abu-Hanna, S. Cohen Percia, M. Kazma Converging within a Generation.” The Lancet 382 (9908): Matalon, and others. 2015. “Spread of KPC-Producing 1898–955. Carbapenem-Resistant Enterobacteriaceae: The Importance Johnson, A. P., and N. Woodford. 2013. “Global Spread of of Super-Spreaders and Rectal KPC Concentration.” Clinical Antibiotic Resistance: The Example of New Delhi Metallo- Microbiology and Infection 21 (5): 470.e1–7. β-Lactamase (NDM)–Mediated Carbapenem Resistance.” Li, S., and R. Laxminarayan. 2015. “Are Physicians’ Prescribing Journal of Medical Microbiology 62 (Pt. 4): 499–513. Decisions Sensitive to Drug Prices? Evidence from a Free- Kaki, R., M. Elligsen, S. Walker, A. Simor, L. Palmay, and Antibiotics Program.” Health Economics 24 (2): 158–74. others. 2011. “Impact of Antimicrobial Stewardship in Linder, J. A., J. N. Doctor, M. W. Friedberg, H. R. Nieva, Critical Care: A Systematic Review.” Journal of Antimicrobial C. Birks, and others. 2014. “Time of Day and the Decision Chemotherapy 66 (6): 1223–30. to Prescribe Antibiotics.” Journal of the American Medical Kariuki, S., M. A. Gordon, N. Feasey, and C. M. Parry. 2015. Association Internal Medicine 174 (12): 2029–31. “Antimicrobial Resistance and Management of Invasive Liu, Y. Y., Y. Wang, T. R. Walsh, L. X. Yi, R. Zhang, and others. Salmonella Disease.” Vaccine 33 (Suppl 3): C21–29. 2016. “Emergence of Plasmid-Mediated Colistin Resistance Kayange, N., E. Kamugisha, D. L. Mwizamholya, S. Jeremiah, Mechanism MCR-1 in Animals and Human Beings in and S. E. Mshana. 2010. “Predictors of Positive Blood China: A Microbiological and Molecular Biological Study.” Culture and Deaths among Neonates with Suspected The Lancet Infectious Diseases 16 (2): 161–68. Neonatal Sepsis in a Tertiary Hospital, Mwanza-Tanzania.” Logan, L. K., J. P. Renschler, S. Gandra, R. A. Weinstein, BMC Pediatrics 10: 39. and R. Laxminarayan. 2015. “Carbapenem-Resistant Klein, E. Y., M. Makowsky, M. Orlando, E. Hatna, N. P. Braykov, Enterobacteriaceae in Children, United States, 1999–2012.” and others. 2015. “Influence of Provider and Urgent Emerging Infectious Diseases 21 (11): 2012–21. Care Density across Different Socioeconomic Strata on Lu, P. L., Y. C. Liu, H. S. Toh, Y. L. Lee, Y. M. Liu, and others. Outpatient Antibiotic Prescribing in the USA.” Journal of 2012. “Epidemiology and Antimicrobial Susceptibility Antimicrobial Chemotherapy 70 (5): 1580–87. Profiles of Gram-Negative Bacteria Causing Urinary Tract Klugman, K. P., S. A. Madhi, R. E. Huebner, R. Kohberger, Infections in the Asia-Pacific Region: 2009–2010 Results N. Mbelle, and others. 2003. “A Trial of a 9-Valent from the Study for Monitoring Antimicrobial Resistance Pneumococcal Conjugate Vaccine in Children with and Trends (SMART).” International Journal of Antimicrobial Those without HIV Infection.” New England Journal of Agents 40 (Suppl): S37–43. Medicine 349 (14): 1341–48. MAF (Ministry of Agriculture and Forestry) New Zealand. Kotwani, A., and K. Holloway. 2011. “Trends in Antibiotic Use 2011. Review and Update on New Zealand Regulatory among Outpatients in New Delhi, India.” BMC Infectious Control of Antimicrobial Agricultural Compounds with Diseases 11 (1): 99. Regard to Antimicrobial Resistance. Wellington: MAF. Kwong, J. C., S. Maaten, R. E. G. Upshur, D. M. Patrick, Maron, D., T. Smith, and K. Nachman. 2013. “Restrictions on and F. Marra. 2009. “The Effect of Universal Influenza Antimicrobial use in Food Animal Production: An International Immunization on Antibiotic Prescriptions: An Ecological Regulatory and Economic Survey.” Global Health 9 (48). Study.” Clinical Infectious Diseases 49 (5): 750–56. Marshall, B. M., and S. B. Levy. 2011. “Food Animals and Kyaw, M. H., R. Lynfield, W. Schaffner, A. S. Craig, J. Hadler, Antimicrobials: Impacts on Human Health.” Clinical and others. 2006. “Effect of Introduction of the Pneumococcal Microbiology Reviews 24 (4): 718–33. Conjugate Vaccine on Drug-Resistant Streptococcus pneumo- Martin, I., P. Sawatzky, G. Liu, and M. R. Mulvey. 2015. niae.” New England Journal of Medicine 354 (14): 1455–63. “Antimicrobial Resistance to Neisseria gonorrhoeae in Laxminarayan, R., Z. A. Bhutta, A. Duse, P. Jenkins, T. O’Brien, Canada: 2009–2013.” Canada Communicable Disease Report and others. 2006. “Drug Resistance.” In Disease Control 41-02 (February 5): 35–41. 446 Major Infectious Diseases Maurice, J. M., and S. Davey. 2009. State of the World’s Vaccines Scheetz, M. H., M. K. Bolon, M. Postelnick, G. A. Noskin, and Immunization. Third edition. Geneva: World Health and T. A. Lee. 2009. “Cost-Effectiveness Analysis of an Organization. Antimicrobial Stewardship Team on Bloodstream McCullers, J. A. 2014. “The Co-Pathogenesis of Influenza Infections: A Probabilistic Analysis.” Journal of Antimicrobial Viruses with Bacteria in the Lung.” Nature Reviews Chemotherapy 63 (4): 816–25. Microbiology 12 (4): 252–62. Shankar, P. R., D. K. Upadhyay, P. Mishra, P. Subish, A. K. Meeker, D., T. K. Knight, M. W. Friedburg, J. A. Linder, Dubey, and others. 2007. “Fluoroquinolone Utilization N. J. Goldstein, and others. 2014. “Nudging Guideline- among Inpatients in a Teaching Hospital in Western Nepal.” Concordant Antibiotic Prescribing: A Randomized Clinical Journal of the Pakistan Medical Association 57 (2): 78–82. Trial.” Journal of the American Medical Association Internal Sinha, A., O. Levine, M. D. Knoll, F. Muhib, and T. A. Lieu. Medicine 174 (3): 425–31. 2007. “Cost-Effectiveness of Pneumococcal Conjugate Meeker, D., J. A. Linder, C. R. Fox, M. W. Friedberg, S. D. Persell, Vaccination in the Prevention of Child Mortality: and others. 2016. “Effect of Behavioral Interventions on An International Economic Analysis.” The Lancet 369 Inappropriate Antibiotic Prescribing among Primary Care (9559): 389–96. Practices: A Randomized Clinical Trial.” Journal of the Song, Y., Y. Bian, M. Petzold, L. Li, and A. Yin. 2014. “The American Medical Association 315 (6): 562–70. Impact of China’s National Essential Medicine System on Mendelson, M., J.-A. Røttingen, U. Gopinathan, D. H. Hamer, Improving Rational Drug Use in Primary Health Care H. Wertheim, and others. 2015. “Maximising Access to Facilities: An Empirical Study in Four Provinces.” BMC Achieve Appropriate Human Antimicrobial Use in Low- and Health Services Research 14 (1): 507–13. Middle-Income Countries.” The Lancet 387 (10014): 88–98. Storberg, V. 2014. “ESBL-Producing Enterobacteriaceae in Okeke, I. N. 2009. “Cholera Vaccine Will Reduce Antibiotic Africa: A Non-Systematic Literature Review of Research Use.” Science 325 (5941): 674. Published 2008–2012.” Infection Ecology and Epidemiology Okeke, I. N., K. P. Klugman, Z. A. Bhutta, A. G. Duse, P. Jenkins, 4. doi:10.3402/iee.v4.20342. and others. 2005. “Antimicrobial Resistance in Developing Sun, L., E. Y. Klein, and R. Laxminarayan. 2012. “Seasonality Countries. Part II: Strategies for Containment.” The Lancet and Temporal Correlation between Community Antibiotic Infectious Diseases 5 (9): 568–80. Use and Resistance in the United States.” Clinical Infectious Okeke, I. N., R. Laxminarayan, Z. A. Bhutta, A. G. Duse, Diseases 55 (5): 687–94. P. Jenkins, and others. 2005. “Antimicrobial Resistance in Teillant, A., S. Gandra, D. Barter, D. J. Morgan, and Developing Countries. Part I: Recent Trends and Current R. Laxminarayan. 2015. “Potential Burden of Antibiotic Status.” The Lancet Infectious Diseases 5 (8): 481–93. Resistance on Surgery and Cancer Chemotherapy Antibiotic O’Neill, J. 2016. Tackling Drug-Resistant Infections Globally: Prophylaxis in the USA: A Literature Review and Modelling Final Report and Recommendations. London: The Review Study.” The Lancet Infectious Diseases 15 (12): 1429–37. on Antimicrobial Resistance. Thu, T. A., M. Rahman, S. Coffin, M. Harun-Or-Rashid, PAHO (Pan American Health Organization). Forthcoming. J. Sakamoto, and others. 2012. “Antibiotic Use in Vietnamese Informe Anual de la Red de Monitoreo/Vigilancia de la Hospitals: A Multicenter Point-Prevalence Study.” American Resistencia a los Antibióticos y de Infecciones Asociadas a la Journal of Infection Control 40 (9): 840–44. Atención de la Salud 2014. Washington, DC: PAHO. USDA (U.S. Department of Agriculture). 2011. “Korea Phases Paudel, K. R., M. Sharma, and B. P. Das. 2008. “Prevalence of Out Antibiotic Usage in Compound Feed.” Report, USDA, Antimicrobial Chemotherapy in Hospitalized Patients in Washington, DC. the Department of Internal Medicine in a Tertiary Care ———. 2013. Feedlot 2011: Part III: Trends in Health Center.” Nepal Medical College Journal 10 (2): 91–95. and Management Practices on U.S. Feedlots, 1994–2011. Pew Charitable Trusts. 2014. Tracking the Pipeline of Antibiotics Washington, DC: USDA. in Development. Philadelphia: Issue Brief, Pew Charitable ———. 2015. “Livestock and Meat Domestic Data.” Washington, Trusts. DC: USDA. http://www.ers.usda.gov/data -products Polgreen, P. M., R. Laxminarayan, and J. E. Cavanaugh. 2011. /livestock-meat-domestic-data.aspx#26084. “Respiratory Fluoroquinolone Use and Influenza.” Infection U.S. Department of Health and Human Services. 2015. Control and Hospital Epidemiology 32 (7): 706–9. “Summary Report on Antimicrobials Sold or Distributed Rabie, T., and V. Curtis. 2006. “Handwashing and Risk of for Use in Food-Producing Animals.” U.S. Food and Drug Respiratory Infections: A Quantitative Systematic Review.” Administration, Washington, DC. Tropical Medicine and International Health 11 (3): 258–67. U.S. FDA (U.S. Food and Drug Administration). 2013. Rheingans, R., M. Amaya, J. Anderson, P. Chakraborty, “Guidance for Industry #213: New Animal Drugs and and J. Atem. 2014. “Systematic Review of the Economic New Animal Drug Combination Products Administered Value of Diarrheal Vaccines.” Human Vaccines and in or on Medicated Feed or Drinking Water of Food- Immunotherapeutics 10 (6): 1582–94. Producing Animals: Recommendations for Drug Sponsors Sabuncu, E., J. David, C. Bernède-Bauduin, S. Pépin, M. Leroy, for Voluntarily Aligning Product Use Conditions with GFI and others. 2009. “Significant Reduction of Antibiotic Use #209.” U.S. FDA, Washington, DC. in the Community after a Nationwide Campaign in France, Van Boeckel, T. P., C. Brower, M. Gilbert, B. T. Grenfell, 2002–2007.” PLoS Medicine 6 (6): e1000084. S. A. Levin, and others. 2015. “Global Trends in Antimicrobial Drug-Resistant Infections 447 Use in Food Animals.” Proceedings of the National Academy ———. 2014. Antimicrobial Resistance: Global Report on of Sciences 112 (18): 5649–54. Surveillance. Geneva: WHO. Van Boeckel, T. P., S. Gandra, A. Ashok, Q. Caudron, ———. 2015. Draft Global Action Plan on Antimicrobial B. T. Grenfell, and others. 2014. “Global Antibiotic Resistance. Geneva: WHO. Consumption 2000 to 2010: An Analysis of National Xiao, Y., J. Zhang, B. Zheng, L. Zhao, S. Li, and others. 2013. Pharmaceutical Sales Data.” The Lancet Infectious Diseases “Changes in Chinese Policies to Promote the Rational Use 3099 (14): 1–9. of Antibiotics.” PLoS Medicine 10 (11): e1001556. WHO (World Health Organization). 2011. Report on the Zhang, Y., B. Y. Lee, and J. M. Donohue. 2010. “Ambulatory Burden of Endemic Health Care: Associated Infection Antibiotic Use and Prescription Drug Coverage in Older Worldwide. Geneva: WHO. Adults.” Archives of Internal Medicine 170 (15): 1308–14. 448 Major Infectious Diseases DCP3 Series Acknowledgments Disease Control Priorities, third edition (DCP3) We thank the many contractors and consultants who compiles the global health knowledge of institutions provided support to specific volumes in the form of and experts from around the world, a task that required economic analytical work, volume coordination, chapter the efforts of over 500 individuals, including volume drafting, and meeting organization: the Center for editors, chapter authors, peer reviewers, advisory com- Disease Dynamics, Economics & Policy; Center for mittee members, and research and staff assistants. For Chronic Disease Control; Centre for Global Health each of these contributions we convey our acknowldg- Research; Emory University; Evidence to Policy Initiative; ment and appreciation. First and foremost, we would Public Health Foundation of India; QURE Healthcare; like to thank our 32 volume editors who provided the University of California, San Francisco; University of intellectual vision for their volumes based on years of Waterloo; University of Queensland; and the World professional work in their respective fields, and then Health Organization. dedicated long hours to reviewing each chapter, pro- We are tremendously grateful for the wisdom and viding leadership and guidance to authors, and fram- guidance provided by our advisory committee to the ing and writing the summary chapters. We also thank editors. Steered by Chair Anne Mills, the advisory com- our chapter authors who collectively volunteered their mittee ensures quality and intellectual rigor of the high- time and expertise to writing over 170 comprehensive, est order for DCP3. evidence-based chapters. The National Academies of Science, Engineering, We owe immense gratitude to the institutional spon- and Medicine, in collaboration with the Interacademy sor of this effort: The Bill & Melinda Gates Foundation. Medical Panel, coordinated the peer-review process The Foundation provided sole financial support of the for all DCP3 chapters. Patrick Kelley, Gillian Buckley, Disease Control Priorities Network. Many thanks to Megan Ginivan, Rachel Pittluck, and Tara Mainero Program Officers Kathy Cahill, Philip Setel, Carol managed this effort and provided critical and substan- Medlin, Damian Walker, and (currently) David Wilson tive input. for their thoughtful interactions, guidance, and encour- World Bank Publications provided exceptional agement over the life of the project. We also wish to guidance and support throughout the demanding thank Jaime Sepúlveda for his longstanding support, production and design process. We would particularly including chairing the Advisory Committee for the sec- like to thank Carlos Rossel, Mary Fisk, Nancy Lammers, ond edition and, more recently, demonstrating his Rumit Pancholi, Deborah Naylor, and Sherrie Brown vision for DCP3 while he was a special advisor to the for their diligence and expertise. Additionally, we thank Gates Foundation. We are also grateful to the University Jose de Buerba, Mario Trubiano, Yulia Ivanova, and of Washington’s Department of Global Health and suc- Chiamaka Osuagwu of the World Bank for providing cessive chairs King Holmes and Judy Wasserheit for professional counsel on communications and marketing providing a home base for the DCP3 Secretariat, which strategies. included intellectual collaboration, logistical coordina- Several U.S. and international institutions contrib- tion, and administrative support. uted to the organization and execution of meetings that 449 supported the preparation and dissemination of DCP3. • Union for International Cancer Control cancer We would like to express our appreciation to the follow- consultation (November 2013, December 2014) ing institutions: • Harvard T. H. Chan School of Public Health, economic evaluation consultation (September 2015) • University of Bergen, consultation on equity (June 2011) • University of California, Berkeley School of Public • University of California, San Francisco, surgery Health, and Stanford Medical School, occupational and volume consultations (April 2012, October 2013, environmental health consultations (December 2015). February 2014) • Institute of Medicine, first meeting of the Advisory Carol Levin provided outstanding governance for Committee to the Editors (March 2013) cost and cost-effectiveness analysis. Stéphane Verguet • Harvard Global Health Institute, consultation on added valuable guidance in applying and improving the policy measures to reduce incidence of noncommu- extended cost-effectiveness analysis method. Elizabeth nicable diseases (July 2013) Brouwer, Kristen Danforth, Nazila Dabestani, Shane • National Academy of Medicine, systems strengthen- Murphy, Zachary Olson, Jinyuan Qi, and David Watkins ing meeting (September 2013) provided exceptional research assistance and analytic • Center for Disease Dynamics, Economics & Policy assistance. Brianne Adderley ably managed the budget (Quality and Uptake meeting, September 2013, and project processes, while Jennifer Nguyen, Shamelle reproductive and maternal health volume consulta- Richards, and Jennifer Grasso contributed exceptional tion, November 2013) project coordination support. The efforts of these indi- • National Cancer Institute cancer consultation viduals were absolutely critical to producing this series, (November 2013) and we are thankful for their commitment. 450 DCP3 Series Acknowledgments Volume and Series Editors VOLUME EDITORS and Infectious Diseases and Global Health and Population of the Harvard T. H. Chan School of Public King K. Holmes Health, where he served as Dean from 1999 to 2009. His King K. Holmes is the Director of Research and research interests have been in immunology, infectious Faculty Development in the University of Washington’s diseases, tuberculosis, leprosy, and vaccines. He served as Department of Global Health. He also serves as Director a consultant to the White House on international health of the Center for AIDS Research and Professor in the policy and has been extensively involved with the World University of Washington’s Department of Medicine, Health Organization, chairing advisory committees on and as Principal Investigator for the University of tropical diseases, leprosy, tuberculosis, and malaria. He Washington’s International Training and Education serves as Co-Chair of the Independent Expert Committee Center for Health, with HIV-related programs in of the Human Heredity and Health in Africa (H3Africa) 20 countries throughout the world. He served as the program; is a member of the Scientific Advisory Board William H. Foege Endowed Chair in Health from 2006 of the Africa Health Research Institute in Durban, South to 2015. Holmes has participated in research on sexually Africa; and chairs the Research Advisory Committee of transmitted infections (STIs) for over 50 years and in the Public Health Foundation of India. research, training, and technical assistance on HIV/AIDS and other STIs globally for over 30 years. Prabhat Jha Prabhat Jha is the founding director of the Centre Stefano Bertozzi for Global Health Research at St. Michael’s Hospital. Stefano Bertozzi is Dean and Professor at the School of He holds Endowed and Canada Research Chairs in Public Health at the University of California, Berkeley. He Global Health in the Dalla Lana School of Public Health has led impact evaluations of large, national health and at the University of Toronto. He is lead investigator of social programs in Africa, Asia, Latin America, and Mexico. the Million Death Study in India, which quantifies the His research focuses on the prevention of HIV/AIDS and cause of death and key risk factors in over two million STIs, as well as on risk behavior among adolescents. He was homes over a 14-year period. He is also Scientific previously the director of the HIV Global Health Program Director of the Statistical Alliance for Vital Events, which at the Bill & Melinda Gates Foundation. His research has aims to expand reliable measurement of causes of death covered a diverse range of projects in health economics and worldwide. His research includes the epidemiology and policy, focusing on the economic aspects of HIV/AIDS and economics of tobacco control worldwide. the health impact of large social programs. SERIES EDITORS Barry R. Bloom Dean T. Jamison Barry R. Bloom is Harvard University Distinguished Dean T. Jamison is Emeritus Professor in Global Health Service Professor of the Departments of Immunology Sciences at the University of California, San Francisco, 451 and the University of Washington. He previously held the Asian Development Bank, several United Nations academic appointments at Harvard University and the agencies, and the International Development Research University of California, Los Angeles. Prior to his aca- Centre, among others, in work conducted in over demic career, he was an economist on the staff of the 20 low- and middle-income countries. She led the World Bank, where he was lead author of the World work on nutrition for the Copenhagen Consensus Bank’s World Development Report 1993: Investing in in 2008, when micronutrients were ranked as the Health. He serves as lead editor for DCP3 and was lead top development priority. She has served as associate editor for the previous two editions. He holds a PhD in provost of graduate studies at the University of economics from Harvard University and is an elected Waterloo, vice-president academic at Wilfrid Laurier member of the Institute of Medicine of the U.S. National University in Waterloo, and interim dean at the Academy of Sciences. He recently served as Co-Chair University of Toronto at Scarborough. and Study Director of The Lancet’s Commission on Investing in Health. Prabhat Jha See the list of Volume Editors. Rachel Nugent Rachel Nugent is Vice President for Global Non- communicable Diseases at RTI International. She was Ramanan Laxminarayan formerly a Research Associate Professor and Principal Ramanan Laxminarayan is Director of the Center for Investigator of DCPN in the Department of Global Disease Dynamics, Economics & Policy in Washington, Health at the University of Washington. Previously, she DC. His research deals with the integration of epidemiolog- served as Deputy Director of Global Health at the Center ical models of infectious diseases and drug resistance into for Global Development, Director of Health and the economic analysis of public health problems. He was Economics at the Population Reference Bureau, Program one of the key architects of the Affordable Medicines Director of Health and Economics Programs at the Facility–malaria, a novel financing mechanism to improve Fogarty International Center of the National Institutes access and delay resistance to antimalarial drugs. In 2012, he of Health, and senior economist at the Food and created the Immunization Technical Support Unit in India, Agriculture Organization of the United Nations. From which has been credited with improving immunization 1991–97, she was associate professor and department coverage in the country. He teaches at Princeton University. chair in economics at Pacific Lutheran University. Charles N. Mock Hellen Gelband Charles N. Mock, MD, PhD, FACS, has training as both a Hellen Gelband is an independent global health policy trauma surgeon and an epidemiologist. He worked as a expert. Her work spans infectious disease, particularly surgeon in Ghana for four years, including at a rural hos- malaria and antibiotic resistance, and noncommunica- pital (Berekum) and at the Kwame Nkrumah University ble disease policy, mainly in low- and middle-income of Science and Technology (Kumasi). In 2005−07, he countries. She has conducted policy studies at Resources served as Director of the University of Washington’s for the Future, the Center for Disease Dynamics, Harborview Injury Prevention and Research Center. Economics & Policy, the (former) Congressional Office He worked at the WHO headquarters in Geneva from of Technology Assessment, the Institute of Medicine of 2007 to 2010, where he was responsible for developing the U.S. National Academies, and a number of interna- the WHO’s trauma care activities. In 2010, he returned to tional organizations. his position as Professor of Surgery (with joint appoint- ments as Professor of Epidemiology and Professor of Global Health) at the University of Washington. His main Susan Horton interests include the spectrum of injury control, espe- Susan Horton is Professor at the University of cially as it pertains to low- and middle-income countries: Waterloo and holds the Centre for International surveillance, injury prevention, prehospital care, and Governance Innovation (CIGI) Chair in Global Health hospital-based trauma care. He was President of the Economics in the Balsillie School of International International Association for Trauma Surgery and Affairs there. She has consulted for the World Bank, Intensive Care from 2013–15. 452 Series and Volume Editors Contributors Elaine J. Abrams Corey Casper Mailman School of Public Health, Columbia University, Fred Hutchinson Cancer Research Center, Seattle, New York, New York, United States Washington, United States Saeed Ahmed Marcia Castro Baylor College of Medicine–Abbott Fund Children’s Harvard T. H. Chan School of Public Health, Boston, Clinical Center of Excellence, Lilongwe, Malawi Massachusetts, United States Sevgi O. Aral Harrell W. Chesson Centers for Disease Control and Prevention, Atlanta, Centers for Disease Control and Prevention, Atlanta, Georgia, United States Georgia, United States Rifat Atun Justin Cohen Harvard T. H. Chan School of Public Health, Boston, Clinton Health Access Initiative, Boston, Massachusetts, Massachusetts, United States United States Myron S. Cohen Joseph B. Babigumira University of North Carolina School of Medicine, Department of Global Health, University of Chapel Hill, North Carolina, United States Washington, Seattle, Washington, United States Ted Cohen Sarah J. Baird Yale School of Public Health, New Haven, Connecticut, Milken Institute School of Public Health, United States The George Washington University, Washington, DC, United States Chris Cotter University of California, San Francisco Medical Center, Till Bärnighausen San Francisco, California, United States Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States Heidi Crane Department of Global Health, University of Fred Newton Binka Washington, Seattle, Washington, United States University of Health and Allied Sciences, Ho, Ghana John A. Crump Lori A. Bollinger Department of Economics, University of Otago, Avenir Health, Glastonbury, Connecticut, United States Dunedin, New Zealand Donald A. P. Bundy Gina Dallabetta Bill & Melinda Gates Foundation, London, Bill & Melinda Gates Foundation, Seattle, Washington, United Kingdom United States 453 Kristen Danforth Roland Gosling Department of Global Health, University of Department of Epidemiology & Biostatistics, University Washington, Seattle, Washington, United States of California, San Francisco, San Francisco, California, United States Lisa M. DeMaria Berkeley School of Public Health, Berkeley, California, Reuben Granich United States International Association of Providers of AIDS Care, Washington, DC, United States Sake J. de Vlas Department of Public Health, University Medical Simon A. J. Gregson Center Rotterdam, the Netherlands Biomedical Research and Training Institute, Harare, Zimbabwe Charlotte Dolenz Timothy B. Hallett Clinton Health Access Initiative, Boston, Massachusetts, School of Public Health, Imperial College London, United States London, United Kingdom Christopher Dye James R. Hargreaves Office of the Director-General, World Health London School of Hygiene & Tropical Medicine, Organization, Geneva, Switzerland London, United Kingdom Wafaa M. El-Sadr Katherine Harripersaud Mailman School of Public Health, Columbia University, Mailman School of Public Health, Columbia University, New York, New York, United States New York, New York, United States Joanne E. Enstone Kate L. Harris School of Medicine, University of Nottingham, Bill & Melinda Gates Foundation, Seattle, Washington, Nottingham, United Kingdom United States Richard Feachem Bernadette Hensen Global Health Sciences, University of California, San London School of Hygiene & Tropical Medicine, Francisco, San Francisco, California, United States London, United Kingdom Christopher Fitzpatrick Charles B. Holmes Department of Control of Neglected Tropical Diseases, Centre for Infectious Disease Research, Lusaka, World Health Organization, Geneva, Switzerland Zambia Hamish Fraser Susan Horton Brigham & Women’s Hospital, Harvard Medical School, School of Public Health and Health Systems, University Boston, Massachusetts, United States of Waterloo, Waterloo, Canada Patricia J. Garcia Grace John-Stewart Ministry of Health, Lima, Peru Department of Global Health, University of Washington, Seattle, Washington, United States Geoffrey P. Garnett James G. Kahn Bill & Melinda Gates Foundation, Seattle, Philip R. Lee Institute for Health Policy Studies, Washington, United States University of California, San Francisco, San Francisco, Louis P. Garrison Jr. California, United States Department of Global Health, University of John Kinuthia Washington, Seattle, Washington, United States Kenyatta National Hospital, Nairobi, Kenya Hellen Gelband Gwen Knight Disease Control Priorities Network, Washington, DC, London School of Hygiene & Tropical Medicine, United States London, United Kingdom Gabriela B. Gomez Shari Krishnaratne London School of Hygiene & Tropical Medicine, London School of Hygiene & Tropical Medicine, London, United Kingdom London, United Kingdom 454 Contributors Ramanan Laxminarayan Edward Nardell Center for Disease Dynamics, Economics & Policy, Harvard T. H. Chan School of Public Health, Boston, Washington, DC, United States Massachusetts, United States Christian Lengeler Gretchen Newby Swiss Tropical and Public Health Institute, Basel, Global Health Sciences, University of California, Switzerland San Francisco, San Francisco, California, United States Edeltraud Lenk Erasmus University Rotterdam, Rotterdam, the Paul N. Newton Netherlands Lao-Oxford-Mahosot Hospital, Wellcome Trust Research Unit, Vientiane, the Lao People’s Democratic Carol Levin Republic Department of Global Health, University of Washington, Seattle, Washington, United States Uzoma Nwankwo Federal Ministry of Health of Nigeria, Abuja, Nigeria Jenny Liu Global Health Sciences, University of California, San Nancy Padian Francisco, San Francisco, California, United States Berkeley School of Public Health, University of Yoel Lubell California, Berkeley, California, United States Mahidol Oxford Tropical Medicine Research Unit, Rosanna W. Peeling Bangkok, Thailand International Diagnostics Centre, London School David Mabey of Hygiene & Tropical Medicine, London, London School of Hygiene & Tropical Medicine, United Kingdom London, United Kingdom Allison Phillips Kudzai Makomva Global Health Sciences, University of California, San Malaria Elimination 8 Secretariat, Windhoek, Namibia Francisco, San Francisco, California, United States Elliot Marseille Yogan Pillay Health Strategies International, San Francisco, South African National Department of Health, Pretoria, California, United States South Africa Philippe Mayaud Suraj Pant London School of Hygiene & Tropical Medicine, Center for Disease Dynamics, Economics & Policy, London, United Kingdom Washington, DC, United States Margaret McNairy Paul Revill Weill Cornell Medical College, Cornell University, Centre for Health Economics, University of York, York, Ithaca, New York, United States United Kingdom Didier Ménard Eric Rubin Worldwide Antimalarial Resistance Network, Harvard T. H. Chan School of Public Health, Boston, Paris, France Massachusetts, United States Molly Miller-Petrie Center for Disease Dynamics, Economics & Policy, Joshua Salomon Washington, DC, United States Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States Deborah Money Department of Obstetrics & Gynecology, University Michael Santos of British Columbia, Vancouver, British Columbia, Bill & Melinda Gates Foundation, Seattle, Washington, Canada United States Meghan Murray Rima Shretta Harvard T.H. Chan School of Public Health, Boston, Global Health Sciences, University of California, San Massachusetts, United States Francisco, San Francisco, California, United States Contributors 455 John Stover Rochelle P. Walensky Avenir Health, Glastonbury, Connecticut, United States Harvard Medical School, Boston, Massachusetts, United States Jessica Taaffe World Bank, Washington, DC, United States Tim Wells Medicines for Malaria Venture, Geneva, Switzerland Marcel Tanner Swiss Tropical and Public Health Institute, Basel, Richard White Switzerland London School of Hygiene & Tropical Medicine, Allison Tatarsky London, United Kingdom Clinton Health Access Initiative, Boston, Massachusetts, United States Danielle Wideman Berkeley School of Public Health, University of Fabrizio Tediosi California, Berkeley, California, United States Swiss Tropical and Public Health Institute, Basel, Switzerland Stefan Z. Wiktor Department of Global Health, University of Harsha Thirumurthy Washington, Seattle, Washington, United States Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina, United David Wilson States World Bank, Washington, DC, United States Anna Vassal London School of Hygiene & Tropical Medicine, Douglas Wilson London, United Kingdom Department of Medicine, Edendale Hospital, KwaZulu- Natal, South Africa Grigory Volchenkov Vladimir Oblast Tuberculosis Dispensary, Center of Prashant Yadav Excellence for Tuberculosis Infection Control, Vladimir, William Davidson Institute, University of Michigan, the Russian Federation Ann Arbor, Michigan, United States 456 Contributors Advisory Committee to the Editors Anne Mills, Chair Roger Glass Professor, London School of Hygiene & Tropical Director, Fogarty International Center, National Medicine, London, United Kingdom Institutes of Health, Bethesda, Maryland, United States Olusoji Adeyi Amanda Glassman Director, Health, Nutrition and Population Global Chief Operating Officer and Senior Fellow, Center for Practice, World Bank, Washington, DC, United States Global Development, Washington, DC, United States Kesetebirhan Admasu Glenda Gray Minister of Health, Addis Ababa, Ethiopia Executive Director, Perinatal HIV Research Unit, Chris Hani Baragwanath Hospital, Johannesburg, George Alleyne South Africa Director Emeritus, Pan American Health Organization, Washington, DC, United States Demissie Habte Chair of Board of Trustees, International Clinical Ala Alwan Epidemiological Network, Addis Ababa, Ethiopia Regional Director Emeritus, World Health Organization, Regional Office for the Eastern Mediterranean, Cairo, Richard Horton Arab Republic of Egypt Editor, The Lancet, London, United Kingdom Rifat Atun Edward Kirumira Professor, Global Health Systems, Harvard T. H. Chan Dean, Faculty of Social Sciences, Makerere University, School of Public Health, Boston, Massachusetts, Kampala, Uganda United States Peter Lachmann Zulfiqar Bhutta Professor, University of Cambridge, Cambridge, Chair, Division of Women and Child Health, Aga Khan United Kingdom University Hospital, Karachi, Pakistan Lai Meng Looi Agnes Binagwaho Professor, University of Malaya, Kuala Lumpur, Former Minister of Health, Kigali, Rwanda Malaysia Mark Blecher Adel Mahmoud Senior Health Advisor, South Africa Treasury Senior Molecular Biologist, Princeton University, Department, Cape Town, South Africa Princeton, New Jersey, United States Patricia Garcia Anthony Measham Minister of Health, Lima, Peru World Bank (retired) 457 Carol Medlin Jaime Sepúlveda Independent Consultant, Washington, DC, United States Executive Director, Global Health Sciences, University of California, San Francisco, California, United States Alvaro Moncayo Researcher, Universidad de los Andes, Bogotá, Richard Skolnik Colombia Yale University School of Public Health (retired) Jaime Montoya Stephen Tollman Executive Director, Philippine Council for Professor, University of Witwatersrand, Johannesburg, Health Research and Development, Taguig City, South Africa the Philippines Jürgen Unutzer Ole Norheim Professor, Department of Psychiatry, University of Professor, Department of Global Health and Primary Washington, Seattle, Washington, United States Care, University of Bergen, Bergen, Norway Damian Walker Deputy Director of Data and Analytics, Bill & Melinda Folashade Omokhodion Gates Foundation, Seattle, Washington, United States Professor, University College Hospital, Ibadan, Nigeria Ngaire Woods Toby Ord Director, Global Economic Governance Program, President, Giving What We Can, Oxford, United Kingdom Oxford University, Oxford, United Kingdom K. Srinath Reddy Nopadol Wora-Urai President, Public Health Foundation of India, Professor, Department of Surgery, Phramongkutklao New Delhi, India Hospital, Bangkok, Thailand Sevket Ruacan Kun Zhao Dean, Koç University School of Medicine, Istanbul, Researcher, China National Health Development Turkey Research Center, Beijing, China 458 Advisory Committee to the Editors Reviewers Leela Barham Kristian Schultz Hansen Economic Consulting LTD, Royston, United Kingdom Department of Global Health and Development, London School of Hygiene & Tropical Medicine, Zulfqar A. Bhutta London, United Kingdom Division of Women and Child Health, Aga Khan University Hospital, Karachi, Pakistan Cecilia T. Hugo ACTMalaria, Manila, the Philippines Richard E. Chaisson Bloomberg School of Public Health, Johns Hopkins Salim S. Abdool Karim University, Baltimore, Maryland, United States Centre for the AIDS Programme of Research in South Africa (CAPRISA), Columbia University, Arantxa Colchero Durban, South Africa Center for Health Systems Research, National Institute Michael Lynch of Public Health, Cuernavaca, Mexico Division of Parasitic Diseases and Malaria, Centers for James Curran Disease Control and Prevention, Atlanta, Georgia, Rollins School of Public Health, Emory University, United States Atlanta, Georgia, United States Kamini Mendis Helen Fletcher Independent Consultant, Colombo, Sri Lanka Tuberculosis Centre, London School of Hygiene & Praphan Phanuphak Tropical Medicine, London, United Kingdom Thai Red Cross AIDS Research Centre, Bangkok, Thailand Omar Galárraga Brown University School of Public Health, Providence, Anthony Seddoh Rhode Island, United States Health, Nutrition and Population Global Practice, World Bank, Accra, Ghana Glenda Gray Perinatal HIV Research Unit, Chris Hani Baragwanath Laurence Slutsker Hospital, Johannesburg, South Africa Center for Global Health, Centers for Disease Control and Prevention, Atlanta, Georgia, United States Laura A. Guay Milken Institute School of Public Health, George Samuel So Washington University, Washington, DC, Asian Liver Center, Stanford University School of United States Medicine, Palo Alto, California, United States 459 Sunil Suhas Solomon Diana Weil Johns Hopkins University School of Medicine, Global TB Programme, World Health Organization, Baltimore, Maryland, United States Geneva, Switzerland Neeraj Sood Mary E. Wilson Sol Price School of Public Policy and Schaeffer Center Harvard T. H. Chan School of Public Health, Boston, for Health Policy and Economics, University of Southern Massachusetts, United States California, Los Angeles, California, United States Paul Volberding AIDS Research Institute, University of California, San Francisco, San Francisco, California, United States 460 Reviewers Index Boxes, figures, maps, notes, and tables are indicated by b, f, m, n, and t following the page number. A HIV/AIDS among, 8, 38 Abrams, Elaine J., 67 adherence to treatment, 80 Abu-Raddad, L. J., 283 testing services, 68, 69 Acinetobacter spp., 366 AEM. See AIDS Epidemic Model ADCs (AIDS-defining cancers), 45, 46 Afghanistan, cutaneous leishmaniasis in, 422 adherence to treatment Africa. See also specific countries and regions HIV/AIDS, 79–81 febrile illness in, 366 barriers, 79–80 hepatitis in, 404 cost-effectiveness, 81 malaria control in, 348 decentralized care and, 81 tuberculosis incidence rates, 240 guidelines, 79 Africa Centre, 104b improvement approaches, 80–81 African Americans measurement of, 79 dyslipidemia among, 55 mother-to-child transmission (MTCT), 122 as HIV/AIDS key population, 36–37 peer counseling and, 81 African Leaders Malaria Alliance, 316 rationale and coverage, 79 Agence Nationale de Recherche, 104b tuberculosis, 274–75 agricultural usage of antibiotics, 438, 439f, 441–42, 442t adolescents. See also children Ahmed, Saeed, 67 febrile illness among, 365–84 AIDS-defining cancers (ADCs), 45, 46 burden of, 365–66 AIDS Epidemic Model (AEM), 180, 181–85t, 186, 192, cost-effectiveness of interventions, 374–79, 196–97 376–77f AIDS Impact Model (AIM), 186, 187 costs of, 373, 374f Alam, N., 214 diagnosis, 369, 370–71t, 372–73, 378 Alistar, S. S., 97 at district hospital facilities, 371–72 Ambinder, R. F., 47 etiology of, 366 amikacin, 252, 253 at first-level health facilities, 371 anal cancer, 45 future research needs, 377–79 anal sex, 207 incidence rates, 366–68, 368–69t Anderson, S., 35, 36 sepsis management and, 375 Angola, malaria elimination in, 321b surveillance, 375 antibiotics and antimicrobial resistance, 19–20, 433–48 treatment interventions, 370–71t, 371–72 drivers of use, 19, 435–39, 437m 461 food animal production usage, 438, 439f, B 441–42, 442t Babigumira, Joseph B., 385 health care usage, 435–38 bacille Calmette–Guérin (BCG) vaccine, 233, 243, hospital usage, 438, 438f, 441 255–56, 278, 279 interventions to ensure appropriate use, bacterial sepsis, 375 19–20, 439–43 bacterial vaginosis, 46, 52, 145 agricultural usage regulations, 438, 439f, Baggaley, R. F., 222 441–42, 442t Baird, Sarah J., 365, 375 education and awareness campaigns, 441–42 Bangladesh in hospitals, 441 resource allocation modeling in, 197 incentives, 441 sexually transmitted infections in, 214, 215 surveillance, 442–43 tuberculosis in vaccinations, 440–41 community-based care, 266 water, sanitation, and hygiene, 439–40 diagnosis, 263b sexually transmitted infections and, 208 hospital-based care in, 264 antiretroviral treatment (ART), 5 information management system, 275 cardiovascular disease and, 53 research and development, 283 effectiveness of, 91 typhoid in, 373 mortality rates and, 208 vector control interventions in, 419 mother-to-child transmission (MTCT) and, visceral leishmaniasis in, 19, 415 91, 121, 122 Bärnighausen, T., 97 for persons who inject drugs (PWIDs), 163 Baxi, Sanjiv, 29 preexposure prophylaxis (PrEP) and, 93 Baxter, C., 214 resource allocation modeling and, 197, 198 BCG. See bacille Calmette–Guérin vaccine transmission biology and, 92 bedaquiline, 251, 253, 277, 284 tuberculosis treatment and, 253–54 behavioral change Arab Republic of Egypt. See Egypt, Arab Republic of HIV/AIDS, 49 Aral, Sevgi O., 203, 222, 223 sexually transmitted infections and, 8, 207, 210, 219 Argentina, malaria elimination in, 315 behavioral economics, 198–99 Armenia, malaria elimination in, 315 Belarus, HIV/AIDS in, 171, 171f ART. See antiretroviral treatment Bendavid, E., 97 artemisinin-based combination therapies (ACTs), Benin 318, 327–28, 385 HIV/AIDS testing services in, 70 artemisinin resistance, 322–24, 323m sexually transmitted infections in, 216 ascariasis, 416 Bertozzi, Stefano, 1 Asia Pacific Leaders Malaria Alliance, 316 Bill & Melinda Gates Foundation, 148, 163, 187, 198, Attia, S., 93 280, 348, 349 Atun, Rifat, 233, 235 Bingham, A. L., 223 Australia bisexual persons. See lesbian, gay, bisexual, and hepatitis in, 407 transgender (LGBT) persons HIV/AIDS in Blacksell, S. D., 367 localized intervention programs, 163 Blaya, J. A., 273 men who have sex with men (MSM), 163 blood supply safety, 16, 405 treatment as prevention (TasP), 94 Bloom, Barry R., 1, 11, 233 sexually transmitted infections in, 214 Bloom, D. E., 97 tuberculosis interventions in, 280 Blower, S., 95 autopsy procedures, 366 Bollinger, Lori A., 148, 179, 188 Avahan Program, 148, 163, 216 Borghi, J., 221 Avenir Health models, 146, 180, 181–85t, 186–90 Botswana Azerbaijan, malaria elimination in, 315, 321b HIV/AIDS in azithromycin, 216, 420, 421, 422 cardiovascular disease comorbidity, 53 462 Index morbidity rates, 33 tuberculosis, 234, 237–42, 238–39t, 244 mother-to-child transmission, 125 typhoid, 373 treatment as prevention (TasP), 104b visceral leishmaniasis, 416 malaria elimination in, 321b WHO estimates on, 380n1 Boyd, M. F., 337 Burkholderia pseudomallei, 373 Brandeau, M. L., 97 Burkina Faso, HIV/AIDS in, 159 Brazil Burnet Institute, 180 antibiotic use in, 438 Buruli ulcer, 18, 414, 419 febrile illness in, 386 hepatitis treatment access in, 405 C HIV/AIDS in CAESAR (Central Asian and Eastern European dyslipidemia comorbidity, 56 Surveillance of Antimicrobial Resistance), 443 incidence rates, 35 Cambodia syphilis in, 130 dengue in, 414–15 tuberculosis in febrile illness in, 367 diagnosis, 263 malaria in pharmaceutical supply chain, 272 incidence rates, 316 preventive therapy, 257 interventions, 321b research and development, 283 sexually transmitted infections in, 221 brucellosis, 367, 368 tuberculosis diagnosis in, 262 Brunei Darussalam, malaria in, 318 Cameroon Bundy, Donald A. P., 411 Buruli ulcer in, 414 burden of disease HIV/AIDS in antiretroviral treatment (ART) and, 208 dyslipidemia comorbidity, 55, 56 Burkholderia pseudomallei, 373 testing services, 70 cardiovascular disease, 52–53 sexually transmitted infections in, 215 chlamydia, 7, 207 Canada by country income group, 2, 2t antimicrobial resistance in, 440 dengue, 413, 413t, 414–15 HIV/AIDS in, 95 diabetes mellitus, 54 malaria elimination in, 326 dyslipidemia, 55 sexually transmitted infections in, 203, 214 febrile illness, 365–66, 397, 397t tuberculosis in, 280 gonorrhea, 7, 207 cancer hepatitis, 16, 402, 403f, 403m HIV/AIDS comorbidity, 46–50 herpes simplex virus (HSV-1 and HSV-2), 207 anal cancer, 48 HIV/AIDS, 31–33, 31–33f, 31f, 32–33 behavioral interventions, 49 comorbidities, 45–66 cervical cancer, 47 mother-to-child transmission (MTCT), 114–19, chemoprevention of, 49 114t, 116–19m epidemiology of, 46–47 human African trypanosomiasis (HAT), 416 hepatocellular carcinoma (HCC), 48 human papillomavirus (HPV), 7, 207 Kaposi sarcoma (KS), 47 leptospirosis, 373 lung cancer, 48 malaria, 316, 348, 373, 413, 413t lymphoma, 47–48 melioidosis, 373 pathogenesis of HIV-associated neglected tropical diseases (NTDs), 19, 412–15, malignancies, 47 413–14t, 414f prevention strategies, 48–49 Orientia tsutsugamushi, 373 screening for, 49 schistosomiasis, 416 vaccines and, 48–49 sexually transmitted infections, 7–8, 204–7 Canchihuaman, F. A., 9 syphilis, 7, 114–19, 114t, 116–19m, 207 Candida spp., 367 trichomoniasis, 7, 207 carbamazepine, 253 Index 463 cardiovascular disease and HIV/AIDS comorbidity, tuberculosis among 52–54 comorbidities, 245 ART impact on, 53 vaccinations, 255–56 burden of, 52–53 China epidemiology of, 52–53 antibiotic resistance in, 433, 434f, 441, 442 factors associated with, 53 antibiotic use in, 438 Caribbean. See Latin America and Caribbean febrile illness in, 367 Carrara, V., 221 hepatitis in Casper, Corey, 45, 47 cost-effectiveness of interventions, 407 Castellsague, X., 210 vaccinations, 404, 406 Castro, Marcia, 347 HIV/AIDS in Cates, W., 144 burden of, 116 CBT (community-based testing), 69 mother-to-child transmission, 130 Celli, Angelo, 352 testing services, 70 Center for Disease Dynamics, 439 treatment as prevention (TasP), 93, 164 Centers for Disease Control and Prevention (CDC), malaria elimination in, 321b, 328–29 24n1, 30, 75, 433 malaria incidence rates in, 318 Central African Republic, syphilis in, 203 syphilis in, 129, 130 Central Asian and Eastern European Surveillance of tuberculosis in Antimicrobial Resistance (CAESAR), 443 diagnosis, 250 cephalosporins, 208 incidence rates, 240 cervical cancer pharmaceutical supply chain, 272 HIV/AIDS comorbidity, 45, 46, 47 treatment interventions, 254 HPV and, 205 chlamydia incidence rates, 7 burden of, 7, 207 screening for, 49 incidence rates, 203 Chagas disease male circumcision and, 210 cost-effectiveness of interventions, 421 Chlamydia trachomatis, 418 interventions for, 18, 414 chloroquinine (CQ), 350, 385 mortality and morbidity rates, 413, 413t, 416 cholesterol, 55 vector control interventions, 418–19, 424, 427 CHWs (community health workers), 76 vector transmission prevention, 421 circumcision. See voluntary male medical circumcision Chambers, R., 334 (VMMC) Charania, M. R., 217 clofazamine, 253 Chersich, M., 215 Clostridium difficile, 433, 437, 440 Chesson, Harrell W., 203, 223 Cochrane Library, 193, 210 chikungunya, 366, 367, 419, 421 Cohen, Justin, 315 children. See also mother-to-child transmission Cohen, Ted, 233 (MTCT) Coker, R., 235 febrile illness among, 385–400 Colombia costs of treatment, 394, 395–96t Chagas disease in, 421 decision-analytic model for, 388–90, 388–89f sexually transmitted infections in, 219 diagnosis, 394, 395–96t Commission on AIDS in Asia, 197 mortality rates, 397, 397t Commission on Macroeconomics and Health, 199 study methodology, 386–94, 387t, 391–94t community-based interventions hepatitis vaccinations for, 404, 406 HIV/AIDS, 163 HIV/AIDS among mother-to-child transmission (MTCT), 124 adherence to treatment, 79, 80 testing services, 70, 75 linkage to care and treatment, 75, 76 sexually transmitted infections, 216–17, 221–22 retention in care, 78 tuberculosis, 264–69, 266f, 267t testing services, 68, 69 efficiency and effectiveness of, 265–66, 267t malaria among, 322, 348 human resource challenges to, 267 sexually transmitted infections among, 205 private sector challenges, 267–68 464 Index recommendations, 292 human African trypanosomiasis (HAT), 422 targeting, 268–69 malaria control, 350–51 community-based testing (CBT), 69 neglected tropical diseases (NTDs) treatment community health workers (CHWs), 76 interventions, 420–22, 420t comorbidities point-of-care tests (POCTs), 378 HIV/AIDS, 45–66 preexposure prophylaxis (PrEP), 137, 149 bacterial vaginosis, 52 sexually transmitted infection interventions, 217–22, cancer, 46–50 218–19t cardiovascular disease, 52–54 syphilis, mother-to-child transmission (MTCT), diabetes mellitus, 54–55 125–29, 126f, 128–29t dyslipidemia, 55–57, 55t treatment as prevention (TasP), 96–102, 98–99t febrile illness, 368, 368t, 372 treatment interventions, 20–22, 21f female reproductive health, 50–51 tuberculosis, 281–89, 285–86t, 288f hepatitis C, 198 vaccines and vaccinations, 20 herpes simplex virus, 51 visceral leishmaniasis, 374, 422 human papillomavirus (HPV), 51 yaws, 422 malaria, 348 Côte d’Ivoire noncommunicable chronic comorbidities HIV/AIDS in, 78–79 (NCCs), 52–57 malaria elimination in, 334 other sexually transmitted infections, 51–52, 115 soil-transmitted helminthiases in, 421 pelvic inflammatory disease, 51–52 syphilis in, 203 sexually transmitted intestinal and enteric tuberculosis in, 257 infections, 52 cotrimoxazole, 254 tuberculosis, 234, 239, 245 Cotter, Chris, 315 infection control and, 3 Crane, Heidi, 45 concentrated HIV/AIDS epidemics, 158, 159–60, Crump, John A., 365, 366, 375 159–60f Cryptococcus spp., 367 condom use Cuba, HIV/AIDS in, 114, 125 HIV/AIDS and, 50, 137, 143–44, 161, 163 cutaneous leishmaniasis, 18, 421–22 resource allocation modeling and, 198 cysticercosis, 2 sexually transmitted infection prevention via, 210 Copenhagen Consensus Center, 280 D cost. See also cost-effectiveness of interventions DAA. See direct acting antiviral medicines of febrile illness, 373, 374f, 394, 395–96t Dallabetta, Gina, 137 of hepatitis, 406–7, 406t DALYs of HIV/AIDS febrile illness, 366, 376, 378, 386 prevention interventions, 146–50, 148t, 149–50f HIV/AIDS, 33–34 retention in care, 78–79 prevention interventions, 149, 151 of malaria control interventions, 350–51 treatment as prevention (TasP), 96 of neglected tropical diseases (NTDs) treatment malaria, 351 interventions, 420–22, 420t neglected tropical diseases and, 413, 421, 423, Costa Rica, malaria elimination in, 315 425, 427 cost-effectiveness of interventions resource allocation modeling and, 180, 199 dengue, 421 sexually transmitted infections and, 8, 205, 208, 221 febrile illness, 374–77, 376–77f, 378–79 syphilis and, 129 hepatitis, 406–7, 406t tuberculosis, 284 HIV/AIDS, 76–77, 81 Danforth, Kristen, 29 local epidemics, 170–73, 171–73f, 172t Database of Abstracts of Reviews of Effects, 210 mother-to-child transmission (MTCT), 125–29, Davis, K. R., 144 126–27t, 126f Decision Makers’ Program Planning Tool (DMPPT) prevention, 137–56 model, 189–90, 196 retention in care, 78–79 Deen, J., 367 testing services, 70–75, 71–74t delamanid, 251, 253, 277 Index 465 DeMaria, Lisa M., 1 malaria, 353–56, 354–55f Democratic Republic of Congo long-term chemoprotection via, 355–56 antibiotic resistance in, 433, 434f transmission-blocking medicines, 355 human African trypanosomiasis in, 415, 422 mass drug administration (MDA), 327–28 malaria control in, 348, 359 neglected tropical diseases (NTDs), 18, 416–18, malaria mortality rates in, 316 420, 420t tuberculosis in, 264 sexually transmitted infections, 9 Demographic and Health Surveys, 39 treatment as prevention (TasP), 103–4 DemProj, 186 tuberculosis dengue drug toxicities and interactions, 253 burden of, 414–15 research and development, 275–80 community-based care, 376 Dua, V., 333 cost-effectiveness of interventions, 421 Dye, Christopher, 233, 241 diagnosis, 372 dyslipidemia, 55–57, 55t incidence rates, 365, 367, 368 mortality and morbidity rates, 413, 413t E vector control interventions, 419, 424, 427 East Africa, HIV/AIDS in. See also specific countries vector transmission prevention, 421 comorbidities in, 47 de Vlas, Sake J., 215, 411 incidence rates in, 33, 35 diabetes mellitus testing services in, 70 febrile illness comorbidity, 368, 368t East Asia and Pacific. See also specific countries HIV/AIDS comorbidity, 54–55 hepatitis in burden of, 54 incidence rates, 16 epidemiology of, 54 mortality rates, 401 factors associated with, 54–55 HIV/AIDS in as tuberculosis risk factor, 246–47 comorbidities, 46 diagnosis concentrated epidemics, 159 dengue, 372 incidence rates, 33 febrile illness, 369, 370–71t, 372–73, 378, 394, key populations, 37 395–96t malaria elimination in, 332 malaria, 327–29, 350 Eastern Europe and Central Asia. See also specific sexually transmitted infections, 206 countries tuberculosis, 248–50, 249b, 261–64, 263b, 292 antibiotic use in, 443 typhoid, 369 hepatitis in diarrhea, 366, 378 incidence rates, 16 Digbeu, N., 214 prevalence, 401 diptheria, 22 HIV/AIDS in direct acting antiviral (DAA) medicines, 401, 405, 407 allocative efficiency for interventions, 173, 173f Djibouti, malaria in, 316 incidence rates, 33 Dolenz, Charlotte, 315 key populations, 36 Dominican Republic morbidity rates, 33 HIV/AIDS in, 160 East-West Center, 180 malaria elimination in, 331 Eaton, J. W., 95–96, 97, 101 sexually transmitted infections in Ebola, 2, 8, 22, 206, 287 community-based interventions, 221 eChasqui, 273 structural interventions, 217 ectopic pregnancies, 51 DOTS strategy, 234, 240–41, 240b, 254, 291–92 efavirenz, 253 Dowdy, D. W., 283 Egypt, Arab Republic of doxycycline, 375, 376 hepatitis in dracunculiasis (Guinea worm), 415 incidence rates, 16 drug interventions. See also antibiotics and prevalence, 401–2 antimicrobial resistance; vaccines and malaria elimination in, 315 vaccinations; specific medications electronic medical records (EMRs), 273–74 466 Index El-Sadr, Wafaa M., 67 study methodology, 386–94, 387t, 391–94t El Salvador, syphilis in, 203 HIV/AIDS comorbidity, 368, 368t, 372 Embase, 210 intervention packages, 14, 14–15t, 370–71t, 371–72 End TB Strategy, 235, 291 Federal Drug Administration (FDA, U.S.), 276 Enstone, Joanne E., 137 financing enteritis, 52 in health system framework, 235f, 236 enzyme-linked immunosorbent assay (ELISA) test, 119, malaria elimination and eradication, 330–36, 331t, 131n5 333–35f, 335–36b Epidemiological Modeling (EMOD), 180, 181–85t, tuberculosis intervention programs, 280–81, 186, 192 280–81f, 289–90, 290t Epstein-Barr virus, 47 Fitzpatrick, Christopher, 411 Eritrea fluoroquinolone, 276 HIV/AIDS morbidity rates in, 33 Fofana, O., 283 tuberculosis incidence rates in, 240 Fontanet, A. L., 210 Escherichia coli, 366, 367, 435, 436b, 440 food animal production, antibiotic usage in, 438, 439f, Estonia 441–42, 442t tuberculosis diagnosis in, 249 France, tuberculosis intervention financing in, 280 tuberculosis treatment in, 264 Fraser, Hamish, 233 ethambutol, 251, 253 Fraser, N., 95 Ethiopia Frenk, J., 235 antimicrobial resistance in, 439 Furness, B. W., 208 HIV/AIDS in Futures Institute, 186 dyslipidemia comorbidity, 56 morbidity rates, 33 G trichiasis in, 413–14 Galárraga, O., 146 tuberculosis treatment in, 264 Galeo, S., 289 European Medicines Agency, 329, 356 Gallup, J. L., 333 Gandhi, N. R., 259 F Gantt, S., 47 Family Health International, 187 Garcia, Patricia J., 113 family planning, 124 Garnett, Geoff P., 137 FamPlan model, 186 Garrison, Louis P., Jr., 385 Feachem, Richard, 315 Gates, B., 334 febrile illness, 365–400 Gavi, the Vaccine Alliance, 24n10, 190, 220 in adolescents and adults, 365–84 gay persons. See lesbian, gay, bisexual, and transgender burden of, 365–66 (LGBT) persons cost-effectiveness of interventions, 374–79, GBD. See Global Burden of Disease 376–77f Gelband, Hellen, 1, 385 costs of, 373, 374f generalized HIV/AIDS epidemics, 158, 160–61 diagnosis, 369, 370–71t, 372–73, 378 GeneXpert, 272, 282 at district hospital facilities, 371–72 Geng, E. H., 39 etiology of, 366 geographic targeting, 167–70, 168–70m, 169f at first-level health facilities, 371 Georgia, malaria elimination in, 315 future research needs, 377–79 Germany, tuberculosis intervention financing in, 280 incidence rates, 366–68, 368–69t Ghana sepsis management and, 375 Buruli ulcer in, 414 surveillance, 375 febrile illness in, 386 treatment interventions, 370–71t, 371–72 HIV/AIDS in in children, 385–400 concentrated epidemics, 160 costs of treatment, 394, 395–96t men who have sex with men (MSM), 160 decision-analytic model for, 388–90, 388–89f malaria elimination in, 12, 333 diagnosis, 394, 395–96t sexually transmitted infections in, 216 mortality rates, 397, 397t GHD. See Global Health Decisions model Index 467 Glaxo-Smith-Kline, 104, 356 syphilis in, 129, 130 Global Action Plan on Antimicrobial Resistance, 439 tuberculosis in Global Antibiotic Resistance Partnership, 443 community-based care, 266 Global Burden of Disease (GBD), 8, 33, 425. See also information management system, 274 burden of disease Hallett, Timothy B., 95–96, 137 Global Drug Facility, 270 Hargreaves, James R., 137 Global Fund to Fight AIDS, Tuberculosis, and Malaria Harripersaud, Katherine, 67 on HIV/AIDS interventions, 20, 138, 190, 196 Harris, Kate L., 137 on malaria interventions, 318, 335, 349 HAT. See human African trypanosomiasis on tuberculosis interventions, 240, 280 HCC (hepatocellular carcinoma), 45 Global Health Cost Consortium, 198 HDL (high-density lipoprotein) cholesterol, 55 Global Health Decisions (GHD) model, 180, 181–85t, Health Action, 394 186, 192–93 health care-associated infections (HAIs), 434, 438, 441 Global Health Estimates database, 380n1 Health Metrics Network, 190 Global Health Observatory Data Repository, 203 Heimbeck, J., 256 Global Health Workforce Alliance, 190 Heller, P. S., 235 Global Laboratory Initiative, 240 helminthiases, 416, 421 Global Malaria Action Plan (GMAP), 348, 349, 350 Henley, C., 215 Global Malaria Eradication Program (GMEP), 318, Hensen, Bernadette, 137 320, 330–31 hepatitis, 14–16, 401–10. See also specific strains below Global Plan to Stop TB, 241, 258 diabetes mellitus comorbidity, 54 Global Resource Needs Estimate (GRNE) model, HIV/AIDS comorbidities, 48 187, 196 incidence rates, 401–2, 402f, 402t glucocorticoids, 253 interventions, 16, 17t, 402t, 404–5 Goals model, 180, 181–85t, 188–89, 189f access to, 405, 407 Golden, M. R., 215 blood supply safety, 16, 405 Goldie, S. J., 220 costs and cost-effectiveness, 406–7, 406t Goldmann, E., 289 harm reduction programs, 405 Gomez, Gabriela B., 114–15, 233 health care injection safety, 16, 405, 407 gonococcal antimicrobial resistance, 208 sanitation, 16, 404 gonorrhea vaccination, 16, 49, 404–5, 404f, 406 antimicrobial resistance and, 208 mortality rates, 16, 402, 403f, 403m burden of, 7, 207 transmission of, 403–4 incidence rates, 203 hepatitis A income inequality and, 223 incidence rates, 401 Gosling, Roly, 315 sanitation improvement interventions, 404 governance in health system framework, 235f, 236 transmission of, 403 Grading of Recommendations Assessment, vaccine for, 401, 404, 406 Development, and Evaluation (GRADE), hepatitis B 103, 121 blood supply safety and, 405 Granich, Reuben, 29, 95, 97, 104 burden of, 207 Grant, P. M., 97 healthcare injection safety and, 405 Greece, malaria elimination in, 331 incidence rates, 401 Green Light Committee, 240, 270 persons who inject drugs (PWIDs) and, 405 Gregson, Simon A. J., 137 prevention interventions, 209 Grossman, M., 217 transmission of, 403 Guinea, tuberculosis in, 287 treatment access for, 405 Guinea worm infections, 2, 19, 337 vaccine for, 209, 213, 222, 401, 404, 406 hepatitis C H blood supply safety and, 405 HAIs. See health care-associated infections cost-effectiveness of interventions, 407 Haiti direct acting antiviral (DAA) medicines for, 401 HIV/AIDS in, 130 healthcare injection safety and, 405 468 Index HIV/AIDS comorbidity, 198 cervical cancer, 47 incidence rates, 401 chemoprevention of, 49 mortality rates, 402 epidemiology of, 46–47 persons who inject drugs (PWIDs) and, 405, 407 hepatocellular carcinoma (HCC), 48 transmission of, 403–4 Kaposi sarcoma (KS), 47 treatment access for, 405 lung cancer, 48 hepatitis D lymphoma, 47–48 incidence rates, 401 pathogenesis of HIV-associated malignancies, 47 mortality rates, 402 prevention strategies, 48–49 transmission of, 403 screening for, 49 vaccine for, 401 vaccines and, 48–49 hepatitis E cardiovascular disease comorbidity, 52–54 incidence rates, 401 ART impact on, 53 sanitation improvement interventions, 404 burden of, 52–53 transmission of, 403 epidemiology of, 52–53 vaccine for, 401, 404–5 factors associated with, 53 hepatocellular carcinoma (HCC), 45, 48 care continuum, 67–90, 68f herpes simplex virus (HSV-1 and HSV-2) adherence, 79–81 burden of, 207 linking to and engagement with care and HIV/AIDS cormorbidities and, 46 treatment services, 75–77 incidence rates, 7 retention in care, 77–79 male circumcision and, 210 testing services, 67–75, 71–74t high-density lipoprotein (HDL) cholesterol, 55 comorbidities, 45–66 high-income countries bacterial vaginosis, 52 antibiotic use in, 438 cancer, 46–50 diabetes mellitus incidence rates in, 54 cardiovascular disease, 52–54 hepatitis in diabetes mellitus, 54–55 cost-effectiveness of interventions, 407 dyslipidemia, 55–57, 55t treatment access, 405 febrile illness, 368, 368t, 372 HIV/AIDS in, 138 female reproductive health, 50–51 cancer comorbidity, 46 hepatitis C, 198 noncommunicable chronic comorbidities, 52 herpes simplex virus, 51 treatment interventions, 29 human papillomavirus (HPV), 51 mortality rates in, 2, 2t malaria, 348 sexually transmitted infections in, 208 noncommunicable chronic comorbidities structural interventions, 217 (NCCs), 52–57 Histoplasma capsulatum, 367 other sexually transmitted infections, 51–52, 115 HIV/AIDS, 5–9, 29–232 pelvic inflammatory disease, 51–52 90-90-90 targets, 30, 34–35, 39–40 sexually transmitted intestinal and enteric adherence to treatment, 79–81 infections, 52 barriers, 79–80 tuberculosis, 234, 239, 245 cost-effectiveness, 81 DALYs, 33–34 decentralized care and, 81 diabetes mellitus comorbidity, 54–55 guidelines, 79 burden of, 54 improvement approaches, 80–81 epidemiology of, 54 measurement of, 79 factors associated with, 54–55 peer counseling and, 81 dyslipidemia comorbidity, 55–57, 55t rationale and coverage, 79 burden of, 55 adolescents, 8 epidemiology of, 55 biology of transmission, 92–93 factors associated with, 55–56 cancer comorbidity, 46–50 febrile illness comorbidity, 368, 368t, 372 anal cancer, 48 female reproductive health comorbidity, 50–51 behavioral interventions, 49 family planning and, 50 Index 469 infertility and, 50–51 implementation of interventions, 121–24 intervention packages, 5, 5–6t male partner involvement and, 123–24 key populations, 36–37 noncommunicable chronic comorbidities (NCCs), 90-90-90 targets among, 37 52–57 defining, 36–37 cardiovascular disease, 52–54 injecting drug users, 8, 36, 68, 160, 160f diabetes mellitus, 54–55 men who have sex with men (MSM), 8, 36, 45, dyslipidemia, 55–57, 55t 68, 160, 160f prevention interventions sex workers, 8, 36, 68, 159–60, 159f barrier methods, 144 surveillance, 39 categorizing, 138–39 linking to and engagement with care and treatment cost-effectiveness of, 137–56 services, 75–77 costs of, 146–50, 148t, 149–50f barriers, 75–76 defining, 138–39 cost-effectiveness, 76–77 history of, 137–38 guidelines, 75 intervention endpoints, 139 measurement of, 75 male circumcision, 7, 8, 144–45 rationale and coverage, 75 preexposure prophylaxis (PrEP), 7, 8, 145, 145f uptake improvement approaches, 76 prevention cascades, 140, 140f, 146, 147f local epidemics, 157–78 systematic reviews of, 140–43, 141f, 142–43t allocative efficiency for interventions, 170–73, targeting, 150–51, 151f 171–73f, 172t vaccines, 146 concentrated epidemics, 158, 159–60, 159–60f vaginal or rectal microbicides, 145–46 cost-effectiveness of interventions, 170–73, provider-initiated testing and counseling (PITC), 68 171–73f, 172t resource allocation modeling, 179–202 generalized epidemics, 158, 160–61 AIDS Epidemic Model (AEM), 180, 181–85t, 186, geographic targeting and hotspot mapping, 192, 196–97 167–70, 168–70m, 169f AIDS Impact Model (AIM), 186, 187 injecting drug users, 8, 160 Avenir Health models, 180, 181–85t, 186–90 men who have sex with men (MSM), 8, 160, 160f behavioral economics and, 198–99 mixed epidemics, 158, 161 comparison of models, 180–93, 181–85t, 198 program science approach, 165–67 controversies in, 199–200 sex workers, 8, 159–60, 159f Decision Makers’ Program Planning Tool tailoring responses to, 161–70, 161–62f, 164–65f, (DMPPT) model, 189–90, 196 166m, 167f, 168–70m Epidemiological Modeling (EMOD), 180, transmission dynamics and, 157–61, 158m 181–85t, 186, 192 malaria comorbidity, 348 external validity of, 198 microepidemics, 35–36, 35–36f field experience with, 194–97 surveillance, 39 Global Health Decisions (GHD) model, 180, morbidity rates, 32–33, 32–33f 181–85t, 186, 192–93 mortality rates, 31–32, 31f Global Resource Needs Estimate (GRNE) model, mother-to-child transmission (MTCT), 7, 113–36 187, 196 adherence to treatment and, 122 Goals model, 180, 181–85t, 188–89, 189f ART and, 121, 122 implementation of, 198 assessment of interventions, 124–25 OneHealth Tool, 180, 181–85t, 186, 190, 191f burden of, 114–19, 114t, 116–19m Optima HIV model, 180, 181–85t, 190–92, 197 case studies, 129–30, 130t policy implications, 194–97 community engagement and, 124 Resource Needs Model (RNM), 180, 181–85t, consequences of, 117–18 187–88, 188f cost-effectiveness of prevention, 125–29, role of, 180, 197 126–27t, 126f unit cost resources and, 198 cross-cutting issues for, 123, 123t What Works Reviews (WWR), 193–94, 193f, 195f effectiveness of interventions, 119–21, 120t retention in care, 77–79 family planning and, 124 barriers, 77–78 470 Index cost-effectiveness, 78–79 diagnosis, 249 guidelines, 77 incidence rates, 240 improvement approaches, 78 Horton, Susan, 1 incentives for, 78–79 hospitals measurement of, 77 antibiotics usage in, 438, 438f, 441 patient costs and, 78–79 febrile illness treatment in, 371–72 rationale and coverage, 77 Hotez, P. J., 420 treatment supporters, use of, 78 hotspot mapping, 167–70, 168–70m, 169f sociocultural barriers to interventions, 7 HPTN (HIV Prevention Trials Network), 91 surveillance HPV. See human papillomavirus 90-90-90 targets, 39–40 Hsiao, W. C., 235 challenges in, 37–40 Humair, S., 97 expansion of, 30 human African trypanosomiasis (HAT) key populations, 39 cost-effectiveness of interventions, 422 loss to follow-up, 38–39 incidence rates, 19, 415 microepidemics, 39 mortality rates, 416 prevalence and incidence rates, 37–38 treatment interventions, 18, 423 syphilis comorbidity, 115 human herpes virus (HHV-8), 47 testing services, 67–75, 71–74t human papillomavirus (HPV) barriers, 69 burden of, 7, 207 community-based testing, 70, 75 cervical cancer caused by, 47, 205 cost-effectiveness, 70–75, 71–74t HIV/AIDS cormorbidities and, 46 early infant diagnosis, 75 male circumcision and, 210 guidelines, 68 prevention interventions, 209 measurement of, 68–69 vaccinations for, 48–49, 209, 213, 222 rationale and coverage, 67–68 hygiene, antibiotics usage and, 439–40 uptake improvement approaches, 69–70 hypertension, 54 voluntary counseling and testing, 70 treatment as prevention (TasP), 8, 91–112 I biological studies, 103–4 IDUs (injecting drug users). See persons who inject clinical studies, 104 drugs (PWIDs) cost-effectiveness, 96–102, 98–99t IGRAs (Interferon-Gamma Release Assays), 255 ecological studies, 94 immune reconstitution inflammatory syndrome effectiveness of, 93–96, 96f (IRIS), 251 measurement challenges, 94–95 immunizations. See vaccines and vaccinations modeling cost-effectiveness, 100–101 incidence rates modeling population-level effectiveness, 95–96 dengue, 365, 367, 368 pharmacological studies, 103–4 febrile illness, 366–68, 368–69t policy implications, 101–3 gonorrhea, 203 population-based studies, 94, 104, 104b hepatitis, 401–2, 402f, 402t public health studies, 104 hepatitis D, 401 serodiscordant couples, 93, 102b hepatitis E, 401 targeting of, 164 herpes simplex virus (HSV-1 and HSV-2), 7 tuberculosis comorbidity, 234, 239, 245 human African trypanosomiasis (HAT), 19, 415 HIV Infant Tracking System (Kenya), 76 malaria, 315 HIV Modelling Consortium, 188 syphilis, 203 HIV Prevention Trials Network (HPTN), 91 trichomoniasis, 203 Hodgkin lymphoma, 47, 48 tuberculosis, 236–37, 237f Holmes, King K., 1, 210 visceral leishmaniasis, 19, 415 Honduras, syphilis in, 203 India Hong Kong SAR, China antibiotic resistance in, 433, 434f, 435, 436b malaria incidence rates in, 318 antibiotic use in, 438, 442 tuberculosis in febrile illness in, 367 Index 471 HIV/AIDS in International Health Partnership and Joint Assessment cancer comorbidity, 46 of National Health Strategies and Plans, 190 geographic targeting of interventions, 167 International Medicines Price Workbook incidence rates, 35 (Health Action), 394 linkage to care and treatment, 76, 77 interventions. See treatment interventions; specific localized intervention programs, 163 interventions prevention interventions, 147, 148 intracytoplasmic sperm injection, 51 testing services, 70, 75 intrauterine insemination, 51 treatment as prevention (TasP), 95, 97 Investment Framework, 196 leprosy in, 415 in vitro fertilization, 51 malaria control and elimination in, 320b, 348 Iran, Islamic Republic of, HIV/AIDS in, 160, 161f cost-effectiveness of interventions, 331, 334 Iraq, malaria elimination in, 315 sexually transmitted infections in, 216 IRIS (immune reconstitution inflammatory tuberculosis in syndrome), 251 community-based care, 266, 268 IRS (indoor residual spraying), 349, 351 diagnosis, 250 isoniazid, 234, 251 financing of interventions, 289–90, 290t ivermectin, 421 hospital-based care in, 264 incidence rates, 240 J information management system, 275 Jamison, D. T., 290 pharmaceutical supply chain, 272 Japan risk factors, 247 tuberculosis comorbidities in, 245 treatment interventions, 255, 282 tuberculosis incidence rates in, 240 vaccinations, 255, 256 tuberculosis intervention program financing in, 280 typhoid burden in, 373 Japanese encephalitis virus (JEV), 367 vector control interventions in, 419 Jha, Prabhat, 1 visceral leishmaniasis in, 19, 415, 419, 422 John-Stewart, Grace, 113 Indonesia Jordan, malaria elimination in, 332 febrile illness in, 367 HIV/AIDS in K mixed epidemics, 161, 162f Kaestner, R., 217 testing services, 70 Kahn, James G., 179 malaria control in, 348 kanamycin, 252, 253 resource allocation modeling in, 197 Kaposi sarcoma (KS), 45, 46, 47 syphilis in, 204 Karim, Q. A., 214 tuberculosis in, 274 Kenya typhoid burden in, 373 HIV/AIDS in indoor residual spraying (IRS), 349, 351 concentrated epidemics, 160 influenza, 367, 368, 375, 440 geographic targeting of interventions, 168 injecting drug users (IDUs). See persons who inject key populations, 36 drugs (PWIDs) linkage to care and treatment, 76 Innovative Vector Control Consortium, 321b, 358 persons who inject drugs (PWIDs), 160 insecticide-treated nets (ITNs), 349, 352 prevention interventions, 148, 149 Institute for Disease Modeling, 180, 192 regional microepidemics, 35–36, 36m Institute for Health Metrics and Evaluation, 31, retention in care, 78 32–33, 237 testing services, 69, 70 integrase inhibitors, 56 treatment as prevention (TasP), 95, 104b Interferon-Gamma Release Assays (IGRAs), 255 voluntary male medical circumcision and, 164 International AIDS Society, 38 malaria vaccine program in, 12 International Association of Physicians in AIDS Care, pelvic inflammatory disease in, 52 75, 79 sexually transmitted infections in, 215, 216 International Epidemiological Databases to Evaluate syphilis screening in, 123 AIDS, 38 tuberculosis in 472 Index diagnosis, 262 leishmaniasis. See cutaneous leishmaniasis; hospital-based care in, 264 visceral leishmaniasis information management system, 275 Lengeler, Christian, 347 key populations for HIV/AIDS, 36–37 Lenk, Edeltraud, 411 90-90-90 targets for, 37 leprosy defining, 36–37 cost-effectiveness of interventions, 422 injecting drug users, 8, 36, 68, 160, 160f incidence rates, 415 men who have sex with men (MSM), 8, 36, 45, 68, treatment interventions, 18, 423 160, 160f tuberculosis and, 242, 246 sex workers, 8, 36, 68, 159–60, 159f leptospirosis, 366, 367, 368, 369, 375 surveillance, 39 lesbian, gay, bisexual, and transgender (LGBT) persons, Khan, M. J., 333 7, 36, 37. See also men who have sex with men kidney disease, 247 (MSM) Kinuthia, John, 113 Lesotho, HIV/AIDS in Kirby Institute, 170 behavioral change interventions, 164, 165, 165f Klebsiella pneumoniae, 436b, 440 generalized epidemics, 161 Knight, Gwen, 233, 287 Levi, J., 34 Korea, Republic of, tuberculosis incidence rates in, 240 Levin, Carol, 113, 220, 347 Kretzschmar, M. E., 96 Liberia, tuberculosis vaccinations in, 287 Krishnaratne, Shari, 137, 140, 141 LICs. See low-income countries Kruk, M. E., 289 linezolid, 253 Kyrgyz Republic liposomal amphotericin B (AmBisome), 422 HIV/AIDS in, 34 Liu, Jenny, 315 malaria elimination in, 315, 318 Lives Saved Tool, 186–87 LLINs. See long-lasting insecticide-treated nets L local HIV/AIDS epidemics, 157–78 Labrique, A. B., 273 allocative efficiency for interventions, 170–73, Lancet Commission on Investing in Health, 198, 171–73f, 172t 427, 428 concentrated epidemics, 158, 159–60, 159–60f Lao People’s Democratic Republic cost-effectiveness of interventions, 170–73, febrile illness in, 367, 375, 376f 171–73f, 172t malaria elimination in, 321b generalized epidemics, 158, 160–61 sexually transmitted infections in, 215 geographic targeting and hotspot mapping, 167–70, Larson, B. A., 129 168–70m, 169f La Ruche, G., 214 injecting drug users, 8, 160 Latin America and Caribbean. See also specific countries men who have sex with men (MSM), 8, 160, 160f antibiotic use in, 443 mixed epidemics, 158, 161 behavioral change interventions in, 22 program science approach, 165–67 Chagas disease in, 419 sex workers, 8, 159–60, 159f HIV/AIDS in tailoring responses to, 161–70, 161–62f, 164–65f, comorbidities, 47 166m, 167f, 168–70m concentrated epidemics, 159, 160 transmission dynamics and, 157–61, 158m key populations, 36 Long, E. F., 97 men who have sex with men (MSM), 160 long-lasting insecticide-treated nets (LLINs), 318, 325, malaria incidence rates in, 318 349, 352, 358–59 sexually transmitted infections in, 214 Lorougnon, F., 214 Latvia, tuberculosis in low-density lipoprotein (LDL) cholesterol, 55 community-based care, 266 Lubell, Yoel, 365, 375 diagnosis, 249 Luecke, E., 141 Laxminarayan, Ramanan, 290, 433 lung cancer, 45, 48 LDL (low-density lipoprotein) cholesterol, 55 Lyamuya, R. E., 39 Lebanon, malaria elimination in, 332 lymphatic filariasis, 2, 415, 418, 423 Leichliter, J. S., 223 lymphoma, 47–48 Index 473 M HIV/AIDS in Mabey, David, 113, 210 adherence to treatment, 79 Madagascar, malaria incidence rates in, 316 ART interventions, 122 Maddali, M. V., 95, 97 behavioral change interventions, 164, 165, 165f Magadi, M. A., 35 intervention incentives, 143 Makomva, Kudzai, 315 linkage to care and treatment, 75, 76 malaria, 11, 14, 315–64 mother-to-child transmission, 122 burden of, 348, 373, 413, 413t surveillance, 38 comorbidities, 348 testing services, 68, 69, 70 control interventions, 13t, 347–64 malaria vaccine program in, 12 costs and cost-effectiveness of, 350–51 sexually transmitted infections in diagnosis, 350 partner notification, 215 effectiveness of, 349–50 structural interventions, 217 environmental determinants, 351–52 treatment interventions, 214 health system factors, 352–53 Malaysia, malaria control in, 352 medicine therapies, 353–56, 354–55f Maldives, malaria incidence rates in, 318 social determinants, 351–52 male circumcision. See voluntary male medical surveillance, 350 circumcision (VMMC) systemic approaches, 351–53 Mali, malaria elimination in, 334 treatments, 13t, 350 Manhart, L. E., 210 vector control, 349, 358–59, 358f Markov model, 390, 390f, 394 elimination and eradication, 315–46 Markowitz, S., 217 artemisinin resistance and, 322–24, 323m Marrazzo, J. M., 144 challenges to, 320–21b, 322–24 Marseille, Elliot, 179, 221 defined, 316–17, 317b Massad, L. S., 51 diagnosis and treatment, 327–29 mass drug administration (MDA), 327–28 epidemiological surveillance, 328–29 Mauritius financing of, 330–36, 331t, 333–35f, malaria elimination in, 321b, 331, 332 335–36b malaria incidence rates in, 318 high-risk populations, 322 Mavedzenge, S. N., 141 interventions, 324–30, 324–25t Mayaud, Philippe, 203, 210, 220 mass drug administration, 327–28 Mayxay, M., 367 milestones and targets, 316, 316t McCarthy, F. D., 333 program management, 329–30 McNairy, Margaret, 67 progress toward, 318–20, 319m, 320t MDA (mass drug administration), 327–28 prospects for, 336–37 MDGs. See Millennium Development Goals research priorities, 337b measles, 2, 22 vector control, 325–27 Médecins Sans Frontières, 102, 274 febrile illness comorbidity, 365 Medication Event Monitoring System, 79 HIV/AIDS comorbidity, 348 MEDLINE, 210 incidence rates, 315 melioidosis, 373 medicine therapies, 353–56, 354–55f MEMA kwa Vijana (intervention program), 216 long-term chemoprotection, 355–56 Ménard, Didier, 315 transmission-blocking medicines, 355 Menon, Manoj, 45 mortality rates, 316 men who have sex with men (MSM) surveillance, 11 HIV/AIDS among, 138 vaccines, 11, 329, 356–58, 357f adherence to treatment, 80 Malaria Program Efficiency Analysis concentrated epidemics, 160 Tool (MPEAT), 336b incidence rates, 92 Malaria Vaccine Funders Group, 356 local epidemics, 8, 160, 160f Malaria Vaccine Initiative, 356 treatment as prevention (TasP), 93, 95 Malaria Vaccine Roadmap, 357 localized intervention programs for, 163, 167 Malawi preexposure prophylaxis (PrEP) for, 163 474 Index resource allocation modeling and, 194 assessment of interventions, 124–25 syphilis among, 203–4, 208 burden of, 114–19, 114t, 116–19m voluntary male circumcision and, 145 case studies, 129–30, 130t MERS (Middle East respiratory syndrome), 2 community engagement and, 124 Mexico consequences of, 117–18 HIV/AIDS in, 148 cost-effectiveness of prevention, 125–29, malaria elimination in, 332 126–27t, 126f Meyer-Rath, G., 100 cross-cutting issues for, 123, 123t microbicides effectiveness of interventions, 119–21, 120t resource allocation modeling and, 194 family planning and, 124 sexually transmitted infections and, 210 implementation of interventions, 121–24 microepidemics, 35–36, 35–36f, 39 male partner involvement and, 123–24 Middle East and North Africa. See also specific countries resource allocation modeling and, 195 brucellosis in, 367 Mozambique HIV/AIDS in HIV/AIDS in comorbidities, 47 prevention interventions, 149 concentrated epidemics, 159 retention in care, 78 key populations, 36 treatment as prevention (TasP), 102 Millennium Development Goals (MDGs), 16, 123, malaria elimination in, 321b 234, 241, 315, 348, 412 sexually transmitted infections in, 221 Miller, W. C., 96 tuberculosis in, 264 Miller-Petrie, Molly, 433 MRSA, 433, 434–35 mixed HIV/AIDS epidemics, 158, 161 MSM. See men who have sex with men modeling for resource allocation. See resource MTCT. See mother-to-child transmission allocation modeling mumps, 2 Moldova, tuberculosis in Murdoch, D. R., 367 community-based care, 266 murine typhus, 367 diagnosis, 264 Murray, Megan, 233 Money, Deborah, 45 Myanmar morbidity rates. See also comorbidities malaria elimination in, 321b, 323 dengue, 413, 413t resource allocation modeling in, 197 febrile illness, 397, 397t tuberculosis in, 264 HIV/AIDS, 32–33, 32–33f mycetoma, 412 neglected tropical diseases (NTDs), 413, 413t Mycobacterium avium, 367 sexually transmitted infections, 205 Mycobacterium bovis, 242, 255 Morocco, malaria elimination in, 315 Mycobacterium kansasii, 256 mortality rates Mycobacterium leprae, 242 antiretroviral treatment (ART) and, 208 Mycobacterium microti, 256 by country income group, 2, 2t Mycobacterium tuberculosis, 233, 242, 244–45, 367, 433 hepatitis, 16, 402, 403f, 403m Mycobacterium ulcerans, 414 HIV/AIDS, 31–32, 31f Mycoplasma genitalium, 210 human African trypanosomiasis (HAT), 416 myocardial infarction, 52, 53 malaria, 316 neglected tropical diseases (NTDs), 412–13, 413t N schistosomiasis, 416 NAATs (nucleic acid amplification tests), 249 sexually transmitted infections, 205 NADCs (non-AIDS defining cancers), 45, 46 tuberculosis, 234, 244 Namibia, malaria elimination in, 321b visceral leishmaniasis, 416 Naranbhai, V., 248 mother-to-child transmission (MTCT) Nardell, Edward, 233 hepatitis, 404 Natunen, K., 220, 222 HIV/AIDS, 7, 113–36 NCCs. See noncommunicable chronic comorbidities adherence to treatment and, 122 Ndowa, F., 220 ART and, 121, 122 needle and syringe programs (NSPs), 161, 163, 407 Index 475 neglected tropical diseases (NTDs), 16–19, 411–48. diabetes mellitus, 54–55 See also specific diseases dyslipidemia, 55–57, 55t burden of, 19, 412–15, 413–14t, 414f non-Hodgkin lymphoma (NHL), 45, 46, 48 defined, 24n6 nonmalarial fever. See febrile illness global targets for control, elimination, and non-nucleoside reverse transcriptase inhibitors eradication, 412, 412t (NNRTIs), 55 proof of concept for, 415–16, 415f, 416m NRTIs (nucleoside reverse transcriptase inhibitors), intervention packages, 16–18, 18t, 416–20, 47, 56 417–19t, 417f NSPs (needle and syringe programs), 161, 163, 407 cost and cost-effectiveness of, 420–22, 420t NTDs. See neglected tropical diseases fairness of, 422 nucleic acid amplification tests (NAATs), 249 individual treatment and care, 419–20 nucleoside reverse transcriptase inhibitors (NRTIs), large-scale prevention to entire communities, 47, 56 18, 416–18 Nundy, S., 289 preventive chemotherapy, 18, 416–18, 420, 420t Nwankwo, Uzoma, 411 return on investment for, 425–27, 426t targeting of, 423–25, 423–24f O morbidity rates, 413, 413t Obiero, J., 145 mortality rates, 412–13, 413t Odeny, T. A., 39 preventive chemotherapy, 412, 417, 417f, 417t Oman, malaria elimination in, 315, 327 vector control, 412, 418–19, 418t, 421 onchocerciasis, 418, 423 Neisseria spp., 366, 433, 435 OneHealth Tool, 180, 181–85t, 186, 190, 191f Nepal OpenMRS platform, 274 antibiotic use in, 438 opioid substitution therapy (OST), 161, 163 antimicrobial resistance in, 439 Optima HIV model, 157, 170, 180, 181–85t, febrile illness in, 367 190–92, 197 malaria elimination in, 332 oral sex, 207 vector control interventions in, 419 Orientia tsutsugamushi, 367, 373 visceral leishmaniasis in, 19, 415 Over, M., 100 Newby, Gretchen, 315 Owens, D. K., 97 Newton, Paul N., 365 Owens, J. P., 283 Next Generation Indoor Residual Spray Project, 321b Owusu-Edusei, K., Jr., 223 NHL (non-Hodgkin lymphoma), 45, 46, 48 nifurtimox-eflornithine combination therapy P (NECT), 422 Padian, Nancy, 29, 146 Nigeria Pakistan antibiotic resistance in, 433, 434f antibiotic resistance in, 433, 434f ascariasis in, 413 hepatitis in HIV/AIDS in incidence rates, 16 cancer comorbidity, 46 prevalence, 402 dyslipidemia comorbidity, 56 HIV/AIDS in geographic targeting of interventions, 167–68, concentrated epidemics, 160 168–69m, 169f geographic targeting of interventions, program science approach, 166–67, 167f 167, 170m malaria in, 348 key populations, 36 cost-effectiveness of interventions, 334 persons who inject drugs (PWIDs), 160 mortality rates, 316 malaria control and elimination in, 333, 334, 348 NNRTIs (non-nucleoside reverse transcriptase sexually transmitted infections in, 217 inhibitors), 55 tuberculosis in non-AIDS defining cancers (NADCs), 45, 46 community-based care, 268 noncommunicable chronic comorbidities (NCCs), hospital-based care in, 264 52–57 information management system, 273, 274 cardiovascular disease, 52–54 palliative care, 254 476 Index PaMZ (pretomanid/moxifloxacin/pyrazinamide), polio, 2, 22, 336–37 277, 284 Population Effects of ART to Reduce HIV Transmission Pant, Suraj, 433 (PopART), 104b Paraguay, malaria elimination in, 315, 318 Population HIV Impact Assessment (PHIA), 68 Partners in Health, 273, 274 Powers, K. A., 96 PARTNER Study, 93 preexposure prophylaxis (PrEP) Peeling, Rosanna W., 113, 130 antiretroviral treatment (ART) and, 93 peer counseling, 81 cost-effectiveness of, 137, 149 pelvic inflammatory disease, 46, 51–52 female reproductive health and, 50 penicillin, 208, 420 HIV/AIDS, 7, 8, 145, 145f Penno, E. C., 375 localized intervention programs, 163 persons who inject drugs (PWIDs) resource allocation modeling and, 194 antiretroviral treatment (ART) for, 163 sexually transmitted infection prevention via, 209 hepatitis and, 405 pregnant women HIV/AIDS among, 138 febrile illness and, 368, 368t concentrated epidemics, 160 hepatitis vaccinations for, 404 interventions targeting, 162–63 malaria among, 322, 348 local epidemics, 8, 160 maternal screening for sexually transmitted treatment as prevention (TasP), 95 infections, 222 preexposure prophylaxis (PrEP) for, 163 syphilis among, 208, 222 Peru tuberculosis among, 253 HIV/AIDS in President’s Emergency Plan for AIDS Relief (PEPFAR), concentrated epidemics, 160, 160f 20, 102, 103, 138, 149, 196 men who have sex with men (MSM), 160 President’s Malaria Initiative, 20, 349 mother-to-child transmission, 130 pretomanid/moxifloxacin/pyrazinamide (PaMZ), sexually transmitted infections in, 216 277, 284 syphilis in, 129, 130 prevention interventions tuberculosis in Chagas disease, 421 community-based care, 264 chemotherapy, 412, 417, 417f, 417t diagnosis, 262 dengue, 421 information management system, 273, 274, 275 HIV/AIDS Peterman, T. A., 208 barrier methods, 144, 210 Pettifor, A., 214 categorizing, 138–39 phenytoin, 253 cost-effectiveness of, 137–56 Philippines costs of, 146–50, 148t, 149–50f leptospirosis in, 373 defining, 138–39 malaria elimination in, 331, 334 history of, 137–38 resource allocation modeling in, 197 intervention endpoints, 139 tuberculosis in, 275 male circumcision, 7, 8, 144–45 Phillips, Allison, 315 preexposure prophylaxis (PrEP), 7, 8, 145, 145f PITC (provider-initiated testing and counseling), 68 prevention cascades, 140, 140f, 146, 147f Plasmodium falciparum, 347–48, 367 systematic reviews of, 140–43, 141f, 142–43t Plasmodium knowlesi, 348 targeting, 150–51, 151f Plasmodium malariae, 348 vaccines, 146 Plasmodium ovale, 348 vaginal or rectal microbicides, 145–46 Plasmodium vivax, 347–48, 367, 386 sexually transmitted infections, 144, 210 pneumonia, 19, 366, 378, 433 treatment as prevention (TasP), 8, 91–112 point-of-care tests (POCTs) biological studies, 103–4 cost-effectiveness analysis, 378 clinical studies, 104 for febrile illness, 369, 372, 374, 378 cost-effectiveness, 96–102, 98–99t for influenza, 375 ecological studies, 94 for malaria, 378 effectiveness of, 93–96, 96f for sepsis, 375 measurement challenges, 94–95 Index 477 modeling cost-effectiveness, 100–101 AIDS Impact Model (AIM), 186, 187 modeling population-level effectiveness, 95–96 antiretroviral treatment (ART) and, 197, 198 pharmacological studies, 103–4 Avenir Health models, 180, 181–85t, 186–90 policy implications, 101–3 behavioral economics and, 198–99 population-based studies, 94, 104, 104b comparison of models, 180–93, 181–85t, 198 public health studies, 104 controversies in, 199–200 serodiscordant couples, 93, 102b Decision Makers’ Program Planning Tool (DMPPT) targeting of, 164 model, 189–90, 196 tuberculosis, 255–61 Epidemiological Modeling (EMOD), 180, 181–85t, in hospitals, 259 186, 192 infection control in congregate settings, 258–61 external validity of, 198 preventive therapy, 257–58 field experience with, 194–97 transmission control, 258–61 Global Health Decisions (GHD) model, 180, vaccinations, 255–56 181–85t, 186, 192–93 primaquine, 328, 350, 353 Global Resource Needs Estimate (GRNE) proctitis, 52 model, 187, 196 proctocolitis, 52 Goals model, 180, 181–85t, 188–89, 189f Promoting Maternal-Infant Survival Everywhere in health system framework, 235f, 236 (PROMISE), 115, 131n7 implementation of, 198 prostitution. See sex workers OneHealth Tool, 180, 181–85t, 186, 190, 191f protease inhibitors, 56 Optima HIV model, 180, 181–85t, 190–92, 197 Proteus mirabilis, 366 policy implications, 194–97 provider-initiated testing and counseling (PITC), 68 Resource Needs Model (RNM), 180, 181–85t, Pseudomonas spp., 366 187–88, 188f PubMed, 193 role of, 180, 197 Punjabi, N. H., 367 unit cost resources and, 198 PWIDs. See persons who inject drugs What Works Reviews (WWR), 193–94, 193f, 195f pyrazinamide, 251, 253 Resource Needs Model (RNM), 180, 181–85t, pyrethroids, 326, 359 187–88, 188f Resources for the Awareness of Population Impacts on Q Development, 186 Q fever, 366, 367, 368 retention in care, 77–79 Quinn, T. C., 92 Réunion, malaria in, 318 quinolones, 208, 435 Revill, Paul, 137 rickettsial infections, 367 R rifampicin, 234, 251, 253, 272 rabies, 416 Rift Valley fever, 367 Radhakrishna, S., 256 RNM. See Resource Needs Model rapid diagnostic tests (RDTs), 350, 385, 386 Roberts, M., 235 RBM. See Roll Back Malaria Partnership Roll Back Malaria (RBM) Partnership, 316, 318, 348–49 receptivity risk maps, 336b Romania, tuberculosis in, 266 Reddy, E. A., 366 Ross, D. A., 141 reproductive health Ross, Ronald, 347 ART and, 50 rubella, 2 family planning and, 50 Rubin, Eric, 233 HIV/AIDS cormorbidities and, 45–46, 50–51 Russian Federation infertility and, 50–51 antibiotic use in, 438 STIs and, 7 HIV/AIDS in República Bolivariana de Venezuela, malaria in, 316 90-90-90 targets and, 34 ResistanceMap, 443 burden of, 116 resource allocation modeling, 179–202 prevention interventions, 148 AIDS Epidemic Model (AEM), 180, 181–85t, 186, tuberculosis in 192, 196–97 diagnosis, 263b 478 Index hospital-based care in, 264 prevention interventions, 8–9, 9–10t, 209–23 incidence rates, 237 behavioral change interventions, 210, 219 infection control, 259, 260b case management, 214, 220 Rutstein, S. E., 221 community interventions, 216–17, 221–22 Rwanda cost-effectiveness of, 217–22, 218–19t HIV/AIDS in effectiveness of interventions, 210–17, 211–13t cardiovascular disease comorbidity, 53 HIV-related benefits of, 222 mother-to-child transmission, 124 HPV vaccines, 210, 213, 220, 222 prevention interventions, 148 income inequality and, 222–23 tuberculosis in male circumcision, 210, 219–20 community-based care, 266 mass treatment, 216, 221 information management system, 275 maternal screening, 222 microbicides, 213–14, 220 S partner notification and management, Sachs, J. D., 333 214–15, 221 Sahin-Hodoglugil, N. N., 220 periodic presumptive treatment, 215–16, 221 Salmonella enterica, 366, 367, 372, 373, 434, 436b research agenda, 223 Salomon, Joshua, 233, 283 structural interventions, 216–17, 221–22 Samb, B., 235 targeted interventions, 215–16, 221 Samyshkin, Y., 235 sexual behaviors and, 8, 207 sanitation social determinants, 208–9 antibiotics usage and, 439–40 sex workers hepatitis treatment interventions, 16, 404 HIV/AIDS among, 138 Santos, Michael, 137 adherence to treatment, 81 São Tomé and Príncipe, malaria elimination in, 332 concentrated epidemics, 159, 159f SARS (severe acute respiratory syndrome), 2 local epidemics, 8, 159–60, 159f schistosomiasis, 416, 423 localized intervention programs, 163 scrub typhus, 367, 368, 369 treatment as prevention (TasP), 95 SDGs. See Sustainable Development Goals localized intervention programs for, 167 SEARCH (Sustainable East Africa Research on Sharma, N. P., 367 Community Health), 104b Shaw, A. V., 366 Sengupta, A., 289 Shigella spp., 366 sepsis, 375 Shretta, Rima, 315, 347 service delivery in health system framework, 235f, 236 Sierra Leone, tuberculosis in, 287 sexually transmitted infections (STIs), 5–9, 203–32. silicosis, 247 See also specific infections Singapore adolescents, 8 malaria in AIDS mortality and, 208 cost-effectiveness of interventions, 332 burden of, 7–8, 204–7 incidence rates, 318 clinician online education for, 9 syphilis in, 203 clustering patterns, 208–9 tuberculosis in diagnosis, 206 diagnosis, 249 epidemiology of, 207–8 incidence rates, 240 globalization and, 208–9 Situational Analysis of Sexual Health, 209 gonococcal antimicrobial resistance, 208 smallpox, 336–37, 338n5 key populations, 208 SMART (Strategies for Management of in LMICs, 206–7 Antiretroviral Therapy), 53 in lower middle income countries, 8 snakebites, 16 morbidity rates, 205 sociocultural barriers to interventions, 7 mortality rates, 205 soil-transmitted helminthiases (STH), 416, 421 mother-to-child transmission (MTCT), 208 Solomon Islands, malaria elimination in, 321b pharmacy treatment of, 9 South Africa prevalence of, 203–4, 204f, 204m, 205t, 206f antibiotic use in, 438, 442 Index 479 antimicrobial resistance in, 439, 440, 440b Spectrum (modeling software), 186, 187, 200n2 bacterial vaginosis in, 52 sperm washing, 51 HIV/AIDS in Sri Lanka adherence to treatment, 80, 81 febrile illness in, 386 behavioral change interventions, 164 malaria in, 315, 317, 318, 330b, 338n1 burden of, 115 cost-effectiveness of interventions, 331, 332 cancer comorbidity, 46 incidence rates, 318 cardiovascular disease comorbidity, 53 vector control, 327 comorbidities, 46 Standards for Reporting of Diagnostic Accuracy, 372 dyslipidemia comorbidity, 55 Staphylococcus aureus, 367, 433, 436b, 440 generalized epidemics, 160, 161f Steen, R., 215 incidence rates, 35 stigma linkage to care and treatment, 76, 77 HIV/AIDS, 36, 37, 69, 79 mother-to-child transmission, 125, 129–30 leishmaniasis, 422 prevention interventions, 148 sexually transmitted infections in, 206 retention in care, 78 of visceral leishmaniasis, 415 testing services, 70 STIs. See sexually transmitted infections treatment as prevention (TasP), 94, 95, Stop TB Initiative, 240 96, 97, 101, 104b, 164 Stop TB Partnership, 240, 241, 291 voluntary male medical circumcision and, 164 Stover, John, 179, 188 malaria elimination in, 321b, 331 Strategies for Management of Antiretroviral sexually transmitted infections in Therapy (SMART), 53 community-based interventions, 217 Streptococcus pneumoniae, 367, 372 microbicide interventions, 214 streptomycin, 253 targeting of interventions, 215 Sub-Saharan Africa. See also specific countries tuberculosis in antibiotic resistance in, 434 community-based care, 266 brucellosis in, 367 comorbidities, 245 cervical cancer in, 47 costs of treatment, 285, 287 febrile illness in, 373, 385 diagnosis, 250, 262, 263 HIV/AIDS in, 5, 29, 30 hospital-based care in, 264 90-90-90 targets and, 34 incidence rates, 236 adherence to treatment, 79, 80 infection control, 259, 260, 261 ART and, 35 preventive therapy, 257–58 burden of, 115 research and development, 283 comorbidities, 47 treatment interventions, 253, 282 concentrated epidemics, 159, 160 vaccinations, 256 cost-effectiveness of interventions, 67 voluntary male circumcision in, 144 diabetes mellitus comorbidity, 54 South Asia dyslipidemia comorbidity, 56 antibiotic resistance in, 434 geographic targeting of interventions, 167–68 febrile illness in, 367–68 key populations, 36 hepatitis in linkage to care and treatment, 75, 76 incidence rates, 16 local epidemics, 158, 158m mortality rates, 401 mortality rates, 31, 31f HIV/AIDS in mother-to-child transmission, 126, 130 comorbidities, 47 noncommunicable chronic comorbidities, 57 morbidity rates, 33 persons who inject drugs (PWIDs), 160 South-East Asia prevention interventions, 148 febrile illness in, 367–68 regional microepidemics, 35 hepatitis in, 404 retention in care, 77, 78 malaria control in, 348 surveillance, 38 sexually transmitted infections in, 203 testing services, 69, 70 tuberculosis in, 237 treatment as prevention (TasP), 95, 97 480 Index Kaposi sarcoma (KS) incidence in, 47 Sweat, M., 221 malaria control and elimination in, 11, 316, 321b, syphilis 348, 353 burden of, 7, 207 treatment interventions, 350 HIV comorbidity, 115 vector control, 325, 349 incidence rates, 203 mother-to-child transmission, 129, 130 mother-to-child transmission (MTCT), 113, 208 neglected tropical diseases in burden of, 114–19, 114t, 116–19m burden of, 413 consequences of, 118–19 cost-effectiveness of interventions, 421 cost-effectiveness of prevention, 125–29, 126f, resource allocation modeling in, 195, 196 128–29t sexually transmitted infections in, 203 cross-cutting issues for, 123, 123t partner notification, 221 effectiveness of interventions, 121 structural interventions, 217 penicillin treatment and, 121 treatment case management, 220 screening during pregnancy, 222 syphilis in, 122 pediatric burden of, 116–17, 117m tuberculosis in transmission of, 114–15 burden of, 237 Syrian Arab Republic, malaria elimination in, 315, 332 community-based care, 268 diagnosis, 262, 263 T incidence rates, 237 Taaffe, Jessica, 157 treatment interventions, 282 tafenoquine, 327 voluntary male circumcision in, 51, 144 Taiwan, China Sudan, visceral leishmaniasis in, 415 HIV/AIDS in, 94 sufadoxine/pyrimethamine (SP), 385 malaria elimination in, 318, 338n3 Supervie, V., 93 tuberculosis in, 255 surveillance Tajikistan, malaria elimination in, 321b antibiotics usage, 442–43 Tanner, Marcel, 347 febrile illness, 375 Tanser, F., 94 HIV/AIDS Tanzania 90-90-90 targets, 39–40 febrile illness analytical model, 386–97, 391–97t. challenges in, 37–40 See also febrile illness expansion of, 30 HIV/AIDS in key populations, 39 behavioral change interventions, 164, 165, 165f loss to follow-up, 38–39 dyslipidemia comorbidity, 55, 56 microepidemics, 39 testing services, 69, 70 prevalence and incidence rates, 37–38 malaria elimination in, 317 malaria, 11, 328–29, 350 sexually transmitted infections in Sustainable Development Goals (SDGs), 11, 24n4, 234, community-based interventions, 216 241, 412, 425, 427 treatment interventions, 214 Sutherland, I., 256 syphilis in, 117, 121, 122, 129, 208 Svandova, E., 256 tuberculosis in Swanson, R. C., 235 community-based care, 266 Swaziland hospital-based care in, 264 HIV/AIDS in research and development, 283 allocative efficiency for interventions, 171–73, Tatarsky, Allison, 315 172f, 172t TB Alliance, 240, 278 concentrated epidemics, 159 TB Supranational Reference Laboratory Network, 249 generalized epidemics, 161 Tediosi, Fabrizio, 347 treatment as prevention (TasP), 164 telemedicine to support health workers, 274 malaria elimination in, 321b Terris-Prestholt, F., 129, 220 cost-effectiveness of interventions, 332 testing services for HIV/AIDS, 67–75, 71–74t surveillance, 328–29 barriers, 69 tuberculosis incidence rates in, 240 community-based testing, 70, 75 Index 481 cost-effectiveness, 70–75, 71–74t costs and cost-effectiveness, 406–7, 406t early infant diagnosis, 75 harm reduction programs, 405 guidelines, 68 health care injection safety, 16, 405, 407 measurement of, 68–69 sanitation, 16, 404 rationale and coverage, 67–68 vaccination, 16, 49, 404–5, 404f, 406 uptake improvement approaches, 69–70 human African trypanosomiasis (HAT), 18, 423 voluntary counseling and testing, 70 leprosy, 18, 423 tetanus, 22 neglected tropical diseases (NTDs), 16–18, 18t, tetracycline, 208 416–20, 417–19t, 417f Thailand cost and cost-effectiveness of, 420–22, 420t HIV/AIDS in fairness of, 422 concentrated epidemics, 160 individual treatment and care, 419–20 diabetes mellitus comorbidity, 54 large-scale prevention to entire communities, dyslipidemia comorbidity, 55 18, 416–18 geographic targeting of interventions, 168 preventive chemotherapy, 18, 416–18, 420, 420t men who have sex with men (MSM), 160 return on investment for, 425–27, 426t mother-to-child transmission, 127 targeting of, 423–25, 423–24f sex workers and, 138 vector transmission prevention, 418–19, 418t, 421 malaria elimination in, 321b, 332, 334 scale-up of, 22–23 sexually transmitted infections in, 210 services integration and, 23 tuberculosis in, 264 sexually transmitted infections, 8–9, 9–10t, 209–23 Thirumurthy, Harsha, 67 behavioral change interventions, 210, 219 ThoughtWorks, 274 case management, 214, 220 Total Community Treatment, 420 community interventions, 216–17, 221–22 trachoma, 418, 423 cost-effectiveness of, 217–22, 218–19t Trajman, A., 283 effectiveness of interventions, 210–17, 211–13t transgender persons. See lesbian, gay, bisexual, and HIV-related benefits of, 222 transgender (LGBT) persons HPV vaccines, 210, 213, 220, 222 TREAT Asia HIV Observational Database, 46 income inequality and, 222–23 treatment as prevention (TasP), 8, 91–112 male circumcision, 210, 219–20 biological studies, 103–4 mass treatment, 216, 221 clinical studies, 104 maternal screening, 222 cost-effectiveness, 96–102, 98–99t microbicides, 213–14, 220 ecological studies, 94 partner notification and management, effectiveness of, 93–96, 96f 214–15, 221 measurement challenges, 94–95 periodic presumptive treatment, 215–16, 221 modeling cost-effectiveness, 100–101 research agenda, 223 modeling population-level effectiveness, 95–96 structural interventions, 216–17, 221–22 pharmacological studies, 103–4 targeted interventions, 215–16, 221 policy implications, 101–3 targeting of, 22 population-based studies, 94, 104, 104b tuberculosis, 234, 250–55, 252t public health studies, 104 vaccine development and, 23 serodiscordant couples, 93, 102b visceral leishmaniasis, 18, 423 targeting of, 164 Treponema pallidum, 216 treatment interventions trichomoniasis, 7, 145, 203, 207, 210 behavioral, 23 triglycerides, 55 cost-effectiveness of, 20–22, 21f Trypanosoma cruzi, 414 drug development and, 23 tuberculosis, 10–11, 233–313 febrile illness, 14, 14–15t, 370–71t, 371–72 burden of, 237–42, 238–39t hepatitis, 16, 17t, 402t, 404–5 community-based care, 264–69, 266f, 267t access to, 405, 407 efficiency and effectiveness, 265–66, 267t blood supply safety, 16, 405 human resource challenges, 267 482 Index private sector challenges, 267–68 second-line treatment of MDR TB, 251–53 recommendations, 292 surgery, 254 targeting, 268–69 vaccinations, 233, 243 comorbidities, 234, 239, 245–46 universal public financing for treatment of, cost-effectiveness of interventions, 281–89, 289–90, 290t 285–86t, 288f Tuberculosis Impact Model and Estimates, 186 diagnosis, 248–50, 249b, 261–64, 263b, 292 Turkey, malaria elimination in, 315 DOTS strategy, 291 Turkmenistan, malaria elimination in, 315, 329b financing for programs, 280–81, 280–81f, typhoid, 367, 368, 369, 373 289–90, 290t Typhoid Fever Surveillance in Africa Program, 434 global response, 240, 240b, 241f, 242b historical trends, 236–37 U HIV/AIDS comorbidity, 234, 239, 245 Uganda incidence rates, 236–37, 237f HIV/AIDS in information management and, 272–75 adherence to treatment, 80, 81 mortality rates, 234, 244 cancer comorbidity, 46 prevention interventions, 255–61 diabetes mellitus comorbidity, 54 in congregate settings, 258–61 incidence rates, 138 in hospitals, 259 linkage to care and treatment, 76 preventive therapy, 257–58 mother-to-child transmission, 130 transmission control, 258–61 prevention interventions, 148 vaccinations, 255–56 retention in care, 78 research and development, 275–80 surveillance, 38 medication optimization, 276–77 testing services, 69, 70, 75 recommendations, 293 treatment as prevention (TasP), 95, 100, 104b treatment interventions, 277–78 voluntary male medical circumcision vaccines, 278–79 and, 164 risk factors, 246–48, 246–47t malaria in screening, 248–50, 249b cost-effectiveness of interventions, 333 SMS reminders to improve adherence, 274–75 incidence rates, 316 stages and intervention points, 242–44, 243f neglected tropical diseases in, 422 strains of, 244–45 sexually transmitted infections in supply chains and, 269–72, 270–71f mass treatment interventions, 216 telemedicine to support health workers, 274 partner notification, 215 transmission control, 258–61 treatment interventions, 214 air disinfection, 260 syphilis in, 130 cost-effectiveness of, 260 tuberculosis in in hospitals, 259 hospital-based care in, 264 in households, 261 treatment interventions, 282 masks and respirators, 260–61 voluntary male circumcision in, 144 measuring effectiveness of, 259–60 Ukraine principles of, 259 HIV/AIDS in, 160 treatment interventions, 12t, 234, 250–55, 252t tuberculosis in, 264 adherence, 274–75 Ukwaja, K. N., 285 antidiabetic treatment, 254 UNAIDS. See United Nations Programme on ART, 253–54 HIV/AIDS cascade of care, 254–55 United Arab Emirates, malaria elimination in, 315 DOTS strategy, 234, 240–41, 240b, 254, 291–92 United Kingdom drug toxicities and interactions, 253 antimicrobial resistance in, 439, 443 first-line treatment of drug-susceptible TB, 251 sexually transmitted infections in, 207 palliative care, 254 tuberculosis in recommendations, 292 financing intervention programs, 280 Index 483 incidence rates, 236 V research and development, 279 vaccines and vaccinations vaccinations, 255 antibiotics and antimicrobial resistance, 440–41 United Nations Office on Drugs and Crime, 163 bacille Calmette–Guérin (BCG) vaccine, 233, 243, United Nations Programme on HIV/AIDS 255–56, 278, 279 (UNAIDS), 5, 30, 31, 32, 67, 69, 163 cost-effectiveness of, 20 on mother-to-child transmission of HIV, 113, 125 hepatitis, 16, 49, 401, 404–5, 404f, 406 Reference Group on Estimates, Modelling and hepatitis A, 401, 404, 406 Projections, 39 hepatitis B, 209, 213, 222, 401, 404, 406 resource allocation modeling and, 179, 187 hepatitis D, 401 surveillance and, 38 hepatitis E, 401, 404–5 United States HIV/AIDS, 48–49 antibiotic use in, 438, 441 human papillomavirus (HPV), 48–49, 209, 213, 222 antimicrobial resistance in, 439 influenza, 440 cardiovascular disease in, 52 malaria, 11, 329, 356–58, 357f diabetes mellitus in, 54 resource allocation modeling and, 194 dyslipidemia in, 55, 56 tuberculosis, 233, 243, 255–56 hepatitis treatment in, 407 Vanuatu, malaria elimination in, 321b, 332 HIV/AIDS in Vassall, Anna, 233, 281 diabetes mellitus comorbidity, 54 vector control dyslipidemia comorbidity, 56 dengue, 419, 424, 427 female reproductive health and, 51 malaria, 325–27, 349, 358–59, 358f key populations, 36–37 neglected tropical diseases (NTDs), 412, linkage to care and treatment, 76 418–19, 418t, 421 localized intervention programs, 163 Venezuela. See República Bolivariana de Venezuela men who have sex with men (MSM), 163 Verguet, S., 290 regional microepidemics, 36, 37m Vickerman, P., 220, 222 treatment as prevention (TasP), 95, 97 Vietnam HPV vaccinations in, 49 antibiotic use in, 438 infertility treatment in, 51 antimicrobial resistance in, 439 influenza vaccinations in, 440 dengue in, 414–15 malaria control and elimination in, 317, 352 HIV/AIDS in vector control, 326 90-90-90 targets and, 34 sexually transmitted infections in, 203, 208 localized intervention programs, 165, 166m behavioral change interventions, 207 treatment as prevention (TasP), 95, 97 burden of, 206–7 malaria elimination in, 321b, 323 microbicide interventions, 214 sexually transmitted infections in, 217 partner notification, 215 syphilis in, 203 structural interventions, 217 tuberculosis in tuberculosis in community-based care, 268 incidence rates, 236 hospital-based care in, 264 infection control, 259 viral hepatitis. See hepatitis preventive therapy, 257, 258 visceral leishmaniasis research and development, 276 cost-effectiveness of interventions, 374, 422 universal public finance (UPF) for tuberculosis incidence rates, 19, 415 treatment, 289–90 mortality rates, 416 University of California, San Francisco, treatment interventions, 18, 423 180, 192, 199 Volchenkov, Grigory, 233 U.S. Agency for International Development voluntary male medical circumcision (VMMC) (USAID), 187, 196 HIV/AIDS interventions and, 7, 97, Uzbekistan, malaria elimination in, 315, 318 144–45, 164 484 Index resource allocation modeling and, 189–90, on hepatitis vaccinations, 49 194, 196, 198 HIV/AIDS treatment guidelines, 30, 31, 68, 69, 75, sexually transmitted infection prevention, 209, 210 77, 79, 96, 102–3, 138, 163 vulvar intraepithelial neoplasia, 51 HIV prevention cascade, 140 on HPV vaccinations, 48–49 W Integrated Management of Adolescent and Wagner, B. G., 95 Adult Illness (IMAI), 371, 375 Walensky, R. P., 97 International Health Regulations, 1 Walker, D., 146 on malaria, 11–12, 315, 349 warfarin, 253 malaria treatment guidelines, 371 Wasserheit, J. N., 210 on mother-to-child transmission of HIV, 113 Weller, S. C., 144 on neglected tropical diseases, 16 Wells, Tim, 347 NTD Roadmap, 412, 416, 418, 423, 427 West Africa. See also specific countries resource allocation modeling and, 190, 199 hepatitis in on sexually transmitted infections, burden of, 7 incidence rates, 16 on syphilis incidence rates, 117, 122 mortality rates, 401 on treatment as prevention (TasP), 164 prevalence, 401 on tuberculosis, 10, 233, 235, 242b, 269b, 280 HIV/AIDS in, 161 vector management strategy, 418 Wetmore, C. M., 210 World Organisation for Animal Health, 440b What Works Reviews (WWR), 193–94, 193f, 195f Wu, Y., 333 White, Richard, 233 WWR (What Works Reviews), 193–94, 193f, 195f WHO. See World Health Organization Wiedeman, Danielle, 29 X Wiktor, Stefan Z., 401 Xpert MTB/RIF test, 250, 261–62, 281–82, 292 Wilson, David, 95, 157 Wilson, Douglas, 233 Y Wolf, H., 333 Yadav, Prashant, 233 women. See also sex workers yaws family planning and, 50 cost-effectiveness of interventions, 422 febrile illness and, 368, 368t treatment interventions, 2, 18, 420, 423, 427 hepatitis vaccinations for, 404 Ying, R., 100 infertility and, 50–51 Yuen, C. M., 263 malaria among, 322, 348 maternal screening for sexually transmitted Z infections, 222 Zambia reproductive health, 45–46, 50–51 HIV/AIDS in syphilis among, 208, 222 adherence to treatment, 79, 80 tuberculosis among, 253 linkage to care and treatment, 75 World Bank, 170, 180, 190 mother-to-child transmission, 130 World Health Assembly, 16, 24n6, 315, 318, prevention interventions, 148 349, 412, 439 surveillance, 38 World Health Organization (WHO) testing services, 68 on cervical cancer screening, 49 treatment as prevention (TasP), 95, 97, 101, 104b Clinical Staging and Disease Classification malaria control and elimination in, 321b, 352 System, 24n1 syphilis in, 129, 130 on febrile illness, 366, 385, 398 tuberculosis in Global Health Estimates, 31 community-based care, 266 Guidelines Group on Couples HIV Testing and diagnosis, 261, 263 Counseling, 102, 102b Zanzibar, malaria elimination in on hepatitis, 407 cost-effectiveness of interventions, 332 Index 485 financing of interventions, 335b prevention interventions, 146, 147f, 149 Zika virus, 2, 8, 18, 22, 421 retention in care, 77 Zimbabwe surveillance, 38 HIV/AIDS in testing services, 68 adherence to treatment, 79, 80 malaria elimination in, 321b cardiovascular disease comorbidity, 53 sexually transmitted infections in, 214 incidence rates, 35, 35f, 138 tuberculosis in, 272 linkage to care and treatment, 75 Zwerling, A., 284 486 Index ECO-AUDIT Environmental Benefits Statement The World Bank Group is committed to reducing its environmental footprint. In support of this commitment, we leverage electronic publishing options and print- on-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping distances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. We follow the recommended standards for paper use set by the Green Press Initiative. The majority of our books are printed on Forest Stewardship Council (FSC)–certified paper, with nearly all containing 50–100 percent recycled con- tent. The recycled fiber in our book paper is either unbleached or bleached using totally chlorine-free (TCF), processed chlorine–free (PCF), or enhanced elemen- tal chlorine–free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://www.worldbank.org/corporateresponsibility.