Public Disclosure Authorized 69074 The Global HIV/AIDS Program The World Bank Building on Evidence: Public Disclosure Authorized A Situational Analysis of the HIV Epidemic and Policy Response in Honduras Public Disclosure Authorized Public Disclosure Authorized October 2008 Titles in this publication series available at: www.worldbank/aids > publications 1. Lessons from World Bank-Supported Initiatives to Fight HIV/AIDS in Countries with IBRD Loans and IDA Credits in Nonaccrual. May 2005. 2. Lessons Learned to date from HIV/AIDS Transport Corridor Projects. August 2005. 3. Accelerating the Education Sector Response to HIV/AIDS in Africa: A Review of World Bank Assistance. August 2005 4. Australia’s Successful Response to AIDS and the Role of Law Reform. June 2006. 5. Reducing HIV/AIDS Vulnerability in Central America. December 2006. (English, Spanish) 6. Reducing HIV/AIDS Vulnerability in Central America: Costa Rica: HIV/AIDS Situation and Response to the Epidemic. December 2006. (English, Spanish) 7. Reducing HIV/AIDS Vulnerability in Central America: El Salvador: HIV/AIDS Situation and Response to the Epidemic. December 2006. (English, Spanish) 8. Reducing HIV/AIDS Vulnerability in Central America: Guatemala: HIV/AIDS Situation and Response to the Epidemic. December 2006. (English, Spanish) 9. Reducing HIV/AIDS Vulnerability in Central America: Honduras: HIV/AIDS Situation and Response to the Epidemic. December 2006. (English, Spanish) 10. Reducing HIV/AIDS Vulnerability in Central America: Nicaragua: HIV/AIDS Situation and Response to the Epidemic. December 2006. (English, Spanish) 11. Reducing HIV/AIDS Vulnerability in Central America: Panama: HIV/AIDS Situation and Response to the Epidemic. December 2006. (English, Spanish) 12. Planning and Managing for HIV/AIDS Results – A Handbook September 2007 (English, Spanish, French, Russian) 13. Rapid analysis of HIV epidemiological and HIV response data about vulnerable populations in the Great Lakes Region of Africa. January 2008. (English, French) Published with the Great Lakes Initiative on AIDS. 14. HIV/AIDS in Ethiopia - an Epidemiological Synthesis. April 2008. Published with the Ethiopia HIV/AIDS Prevention and Control Office (HAPCO). 15. Blood Services in Central Asian Health Systems: A Clear and Present Danger of Spreading HIV/AIDS and Other Infectious Diseases. May 2008. (English, Russian) Published with World Bank Europe and Central Asia Region. 16. Knowledge, Attitudes and Behavior Related to HIV/AIDS among Transport Sector Workers - A Case Study of Georgia. June 2008. Published with World Bank Europe and Central Asia Region. 17. Building on Evidence: A Situational Analysis of the HIV Epidemic and Policy Response in Honduras. October 2008. BUILDING ON EVIDENCE: A SITUATIONAL ANALYSIS OF THE HIV EPIDEMIC AND POLICY RESPONSE IN HONDURAS World Bank authors: George Ciccariello-Maher, Stephen Forsythe, Eric Gaillard, Joy de, Rosalía Rodríguez-García, and Marcelo Bortman UNAIDS author: Maria Tallarico World Bank Global HIV/AIDS Program Report October 2008 World Bank Global HIV/AIDS Program Reports This series, published by the Global HIV/AIDS Program of the World Bank's Human Development Network, makes interesting new work on HIV/AIDS widely available, quickly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent. Citation and the use of material presented in this series should take into account that it may be provisional. Papers are posted at www.worldbank.org/AIDS (go to “publications”). For free print copies of reports in this series please contact the corresponding author whose name appears the bottom of page iii of the paper. Enquiries about the series and submissions should be made directly to Joy de Beyer (jdebeyer@worldbank). Cover photograph taken near Ivans, Honduras by Sebastian Szyd. © 2008 World Bank © 2008 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. ii Building on Evidence: A Situational Analysis of the HIV Epidemic and Policy Response in Honduras The work that this paper describes was carried out at the request of the government of Honduras, with support from the AIDS Strategy and Action Plan service (ASAP), a service of UNAIDS, hosted at the World Bank. The work was funded through the UNAIDS Unified Budget and Workplan. Abstract: This paper summarizes key findings of the epidemiological, response and policy synthesis analysis carried out in Honduras in 2007, as part of the preparation of the new national HIV/AIDS strategy, PENSIDA III, 2008-2012. It presents the most recent data on HIV prevalence, and the results of models that estimate sources of infection and likely patterns and trends in future prevalence. The paper also describes how these data and projections have been incorporated into the design of Honduras’ response to the epidemic, grounding PENSIDA III’s strategic direction in the data on the epidemic and response in Honduras, building and improving on previous experience. Keywords: HIV, AIDS, Honduras, PENSIDA, epidemic, prevalence, incidence, most at risk groups, men who have sex with men, sex workers, Garífuna, CONASIDA, evidence- based, strategy, World Bank, AIDS Strategy and Action Plan, UNAIDS Disclaimer: The findings, interpretations and conclusions expressed in the paper are entirely those of the authors, and do not represent the views of the World Bank, its Executive Directors, or the countries they represent. Correspondence Details: Rosalía Rodríguez-García, World Bank, 1818 H Street, NW, Washington DC, 20433. tel: (202) 473-8846. fax: (202) 522-1252 email: RRodriguezGarcia@worldbank.org iii Acronyms and Abbreviations AIDS Acquired immunodeficiency syndrome ART Antiretroviral treatment/therapy ARV Antiretroviral CONASIDA The National HIV/AIDS Commission (Consejo Nacional de Atención Integral del VIH/SIDA) CSW Clients of Sex Workers HIV Human immunodeficiency virus MSM men who have sex with men NGO Nongovernmental organization OVC Orphans and vulnerable children PENSIDA National Strategic Plan to Fight HIV/AIDS (Plan Estratégico Nacional de Lucha Contra el VIH/SIDA) PLHIV People living with HIV (includes people living with AIDS) PMTCT Prevention of Mother-to-child Transmission of HIV STI Sexually Transmitted Infection/s SW Sex Worker/s UNAIDS Joint United Nations Programme on HIV/AIDS VCT Voluntary counseling and testing for HIV Targa highly active antiretrovial therapy (HAART) (tratamiento antirretroviral de gran actividad) iv Table of Contents PREFACE AND ACKNOWLEDGMENTS ...................................................................VII INTRODUCTION - THE EVOLUTION OF KNOWLEDGE OF HIV-AIDS IN HONDURAS ............ 1 I. CURRENT SNAPSHOT OF THE EPIDEMIC IN HONDURAS ............................ 2 a) Geographical Distribution ...................................................................................... 2 b) Age and Sex Distribution ........................................................................................ 3 c) Prevalence in Specific Populations......................................................................... 5 d) Summarizing the epidemic picture .......................................................................... 7 II. FACTORS IMPACTING A SUCCESSFUL RESPONSE....................................... 7 a) Knowledge and Behavior ........................................................................................ 7 b) Risk Factors............................................................................................................. 9 c) Vulnerability Factors ............................................................................................ 10 d) 2007 HIV Estimates............................................................................................... 11 e) Summarizing Additional Factors........................................................................... 13 III. BUILDING POLICY ON EVIDENCE .................................................................. 13 MOVING FORWARD FROM PENSIDA II ......................................................................... 13 A GLANCE AT PENSIDA III .......................................................................................... 14 REFERENCES................................................................................................................ 18 List of Tables Table 1: Prevalence of STIs in Selected Populations ......................................................... 9 Table 2: Vulnerability Factors in Honduras ..................................................................... 10 List of Figures Figure 1 - Cumulative HIV/AIDS Cases per 100,000 Inhabitants (1985-2005) .................3 Figure 2: Trend in HIV/AIDS Cases by Age Group, Men, 1985-2005...............................4 Figure 3: Trend in HIV/AIDS Cases by Age Group, Women, 1985-2005..........................4 Figure 4: Prevalence of HIV in PMTCT .............................................................................5 Figure 5: Prevalence of HIV in Selected Populations .........................................................6 Figure 6: Condom Use in Selected Populations ..................................................................8 Figure 7: Workbook Method Projection of the HIV, % Adult Prevalence, 1980-2010 ....11 Figure 8: Mode of Transmission Estimate of the HIV and AIDS Epidemic .....................12 v vi Preface and Acknowledgments In 2006, the Honduran Minister of Finance, Rebeca P. Santos, and the Minister of Health, Jenny Meza, requested the assistance of the Global HIV/AIDS Program (GHAP) of the World Bank through the Latin American Office to support the process of formulating a new HIV/AIDS strategy (PENSIDA III, 2008-2012). The Bank responded through ASAP - the AIDS Strategy and Action Plan service which GHAP hosts on behalf of UNAIDS. Honduras has the highest HIV prevalence among countries in Latin America (1.5%), but the request came at a time when new evidence indicates improved condom use and falling prevalence among the groups most at risk. Honduras is determined to continue these trends, also important for sustaining Honduras’ free ARV treatment program. Within 12 months, the initial goal was achieved: Honduras had developed a strong new national HIV/AIDS Strategy, PENSIDA III, which was presented to the people of Honduras by the Minister of Health in the presence of the First Lady. PENSIDA III is firmly grounded in evidence on transmission patterns (transmission during unsafe paid sex and unprotected sex between men predominates), and focuses on achieving and monitoring measurable results in reducing new infections. The team that developed PENSIDA III gained valuable skills and experience in strategic planning, costing, spending analysis, results monitoring, and participatory analytical work. PENSIDA III and the background documents are available only in Spanish. This short report makes the interesting data and strategic thinking more widely accessible, to English-speaking readers. It outlines some key findings of the epidemiological and expenditure analysis underpinning PENSIDA III, and briefly summarizes the main thrust of the new national strategy. Written by George Ciccariello-Maher (Consultant), drawing on the initial draft epidemic and response synthesis by Stephen Forsythe and Eric Gaillard (Consultants), and on the PENSIDA III document, this report was edited by Joy de Beyer (GHAP), and produced under the guidance of Rosalía Rodríguez-García, Marcelo Bortman (World Bank), and Maria Tallarico (UNAIDS). The core PENSIDA III team comprised: Mayté Paredes (Head, Department of STIs/HIV/AIDS), Rudy Rosales de Molinero (Secretary of Labor and Social Security), Xiomara Bú (National Forum on AIDS), Elsa Palou (College of Medicine of Honduras), Vilma Montoya (Cohep), Irma Mendoza (Gerente), Odalys García y Andersy Moncada (Department of STIs/HIV/AIDS), Liliana Mejía and Juan Ramón Gradelhy (UNAIDS), Karla Zepeda and Iris Padilla (WHO/PAHO), Emilia Alduvín (Dfid), Lícida Bautista (Comcavi), Jeffrey Barahona Perdomo (World Bank Consultant to the Secretary of Health), Jose Antonio Izazola and Daniel Aran (cost and expenditure estimates). The PENSIDA III development team was supported by an international team under the direction of Rosalía Rodríguez-García (AIDS Strategy and Action Plan service, Global HIV/AIDS Program), Marcelo Bortman (Latin American and the Caribbean Region, World Bank), and Miriam Montenegro (World Bank, Honduras). Implementation of PENSIDA III is coordinated by the National HIV/AIDS Program (CONASIDA), under the guidance of the Director, Dr. Xioleth Rodriguez. vii viii Building on Evidence: A Situational Analysis of the HIV Epidemic and Policy Response in Honduras Introduction - The evolution of knowledge of HIV-AIDS in Honduras Until recently, existing data provided at best a murky picture of the HIV/AIDS epidemic in Honduras. For example, data collected among pregnant women in the cities of San Pedro Sula and Tegucigalpa during the mid-1990s suggested an alarming prospect of double-digit HIV prevalence by the year 2000, with the epidemic spreading quickly and widely among the general population. Later data, however, showed that this did not happen, and partly as a result of quick preventive action, the Honduran epidemic peaked and entered into decline. But this welcome turn entailed a new series of challenges, and still little was known about the dynamics of the epidemic and most impacted groups. An important task is to determine which sub-populations are most at-risk, and how best to target and provide preventive services and treatment to sub-populations with a relatively high prevalence of HIV and AIDS. This requires more sophisticated data and analysis than previously available. Throughout the course of the first national HIV/AIDS strategy, PENSIDA I and well into PENSIDA II, very little data existed to point policymakers toward the most affected populations, and the behaviors causing most new infections. While many saw these initial responses as satisfactory in many ways, it was clear that more could be done to deal even more effectively with the epidemic. Improvements in policies and programs occur through constant assessment in the light of relevant and up-to-date evidence. This paper describes how this has recently been done with regard to the HIV epidemic in Honduras. Part I presents the most recent data for several key variables, focusing specifically on geographical, age, and population distributions of HIV, and their interrelations. The data offer a snapshot of the current state of the epidemic: HIV prevalence in Honduras is heavily concentrated geographically and in specific at-risk populations. Part II briefly discusses various elements—knowledge and behavior, risk and vulnerability, and projections of the future development of HIV—which provide a richer context for considering the epidemic picture outlined in Part I, for HIV policies and programs. Specifically, we seek to understand and explain current prevalence patterns in the epidemic and also look at the results of models that estimate sources of infection and likely patterns and trends in future prevalence. Part III looks briefly at how the current HIV/AIDS program in Honduras—PENSIDA III (2008-2012)—evolves from and improves on prior policy, building in a number of ways on this empirical snapshot of the epidemic and prior policy efforts, incorporating these data and projections into the design of Honduras’ response to the epidemic. 1 I. Current Snapshot of the Epidemic in Honduras HIV was first detected in Honduras in 1985, and according to national statistics, more than 24,000 cases had been reported as of mid-2007. 1 According to existing data sources, the HIV/AIDS epidemic in Honduras has been—in comparison to neighboring nations— exceptionally severe and sexually-driven. What explains these particularities of the Honduran epidemic? Why has it apparently spread so rapidly and primarily through sexual infection? Three key socio-historic factors—which also drive the epidemic’s distribution geographically and in specific at- risk populations that will be discussed below—are: - The development of a relatively large sex industry, coinciding with the presence of national and foreign armed forces in Honduras during the early 1980s; - the high levels of mobility among the ethnic Garífuna population, especially via the merchant marine and contact with the Garífuna population in the northeastern United States (especially New York); - the rapid increase in HIV prevalence among high-risk groups, specifically, men who have sex with men (MSM 2 ), sex workers (SWs), and the Garífuna population (see Ic). This section discusses the patterns in the epidemic and its specific modes of transmission, by geographical region, age and gender, and prevalence among these at-risk populations. a) Geographical Distribution As a result of the modes of transmission and concentration of HIV among specific sub- populations in Honduras, the epidemic has taken on particular geographic characteristics. Figure 1 shows the strong geographic concentration of the epidemic along two axes of a “T” running East-to-West along the northern Atlantic coast (Cortés, Atlántida, and Colón) and the islands, and North-to-South from the Sula Valley (in Cortés) to the capital Tegucigalpa (in Francisco Morazán) and southward to Valle on the Pacific coast. There are particularly high concentrations of reported HIV/AIDS cases in four of 18 national departments: Cortés (521 cases per 100,000), Atlántida (369), and Francisco Morazán and Islas de la Bahía (each with 298). Together, these four departments constitute 71 percent of all HIV/AIDS cases reported since 1985. A key factor in the geographical distribution of the epidemic in Honduras has been its concentration in two urban areas, whose departments show the highest concentration of cases: San Pedro Sula (in Cortés) and Tegucigalpa (in Francisco Morazán). It is a challenge to determine if the heavily urban character of the epidemic in Honduras has to do primarily with the concentration of SWs, access to health care services and testing, or if other factors come into play. In the sparsely populated western and especially eastern 1 Much of the data in this report is derived from CONASIDA, “III Plan Estratégico de Respuesta al VIH en Honduras: PENSIDA III (2008-2012).” Tegucigalpa, Honduras (December 2007). 2 MSM include male homosexuals, transvestites, transgendered, transsexuals, and bisexuals. 2 parts of the country, HIV rates are markedly lower, but given the role that information plays in determining vulnerability (see IIc), policymakers cannot afford to neglect these areas entirely. Figure 1 - Cumulative HIV/AIDS Cases per 100,000 Inhabitants (1985-2005) Islas de la Bahia 297.5 Atlántida Colón Gracias a Dios Cortès 368.6 Santa Bárbara 226.4 521.1 213.7 Copán 74.7 115.7 Yoro 122.6 Comayagua 48.2 62.4 144.6 34.5 Fco Morazán Ocotepeque 28.2 298.2 Lempira 52.4 Olancho 76.9 Intibuca La Paz El Paraíso Leyenda 202.1 184.4 28 - 151 (10) Valle 151 - 275 (4) Choluteca 275 - 398 (3) 398 - 521 (1) Source: Department of STIs/HIV/AIDS. Health Secretary, Honduras. Other key concentrations are explained by the prevalence of the Garífuna population, including Roatán, a heavily-Garífuna island, to which the ethnic group was forcibly exiled by the English in the late 18th century, and La Ceiba, in Atlántida, a major port and landing-point from the islands, which has the coastal area’s highest Garífuna population. Some developing trends require further explanation. For example, recent data show cities like Nacaome in the southern department of Valle joining Tegucigalpa and San Pedro Sula among those with the highest HIV prevalence, suggesting a potentially worrying geographical shift—perhaps via the high mobility of groups like the Garífuna— from the Atlantic to the Pacific coast, whose causes must be determined and confronted. b) Age and Sex Distribution As in other countries, HIV/AIDS in Honduras primarily affects the younger, economically- and reproductively-active segment of the population, and since the onset of the epidemic in Honduras, those between the ages of 15 and 39 have constituted somewhere between 65 and 70 percent of reported new cases. A notable trend is that the average age of diagnosis has been increasing slowly since the early 1980s. Figure 2 (males) and Figure 3 (females) show that the most frequently occurring age of diagnosis increased in males from 20-24/25-29 in the early 1990s to 30- 3 34 in the late 1990s, and the 35-39 age range seems poised to overtake this group, with an increasing number of diagnoses over the age of 40 years. Females show a similar pattern but with a lag. There are factors that could contribute to earlier diagnosis among women without entailing an actual age difference in HIV incidence—for example, expanded testing under Honduras’ recently-instituted PMTCT program. Thus age-sex differentials remain to be explored fully. Figure 2: Trend in HIV/AIDS Cases by Age Group, Men, 1985-2005 Source: Department of STIs/HIV/AIDS, Health Secretary, Honduras. Figure 3: Trend in HIV/AIDS Cases by Age Group, Women, 1985-2005 Source: Department of STIs/HIV/AIDS, Health Secretary, Honduras. 4 Shifts in the infection ratio between men and women are also evident. In 1994, the ratio was 1.7:1.0, suggesting an early concentration among men who have sex with men (MSM) and the exceedingly-mobile male Garífuna population. But as the epidemic gained ground among sex workers and spread to female members of the Garífuna community, there was a striking process of feminization of the epidemic. In 2005, the gender ratio had dropped to 1.1:1.0, and the majority of newly-infected people are women, with a ratio of 0.6:1.0. Current projections suggest that this trend has peaked (see Part IId), and may begin to recede with the brunt of new infections projected to be among MSM and clients of sex workers (CSW). c) Prevalence in Specific Populations The geographic concentration of the epidemic in Honduras reflects its concentration among specific at-risk populations. This section first discusses a target population not considered at-risk—pregnant women, and then considers the three Honduran communities with the highest HIV prevalence. Pregnant women in prenatal care, prevention of mother-to-child transmission (PMTCT) Early surveillance data were worrying: an HIV rate among women attending prenatal care in San Pedro Sula of 3.4 percent in 1990 and 4.1 percent in 1995, and in Tegucigalpa 0.2 percent in 1991 and 1.0 percent in 1996. (These figures were the basis for early projections of double-digit infection rates within the decade with the possibility of a spread of the epidemic in the general population.) Prevalence data among pregnant women accessing prenatal care collected in 2004 showed a greatly improved picture. Figure 4: Prevalence of HIV in PMTCT 6% 5% 4% San Pedro Sula 3% 2% Tegucigalpa 1% 0% 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 19 19 19 19 19 19 19 19 19 19 20 20 20 20 20 Source: Department of STI/HIV/AIDS, Health Secretary of Honduras, studies on prevalence in PMTCT. 5 In Figure 4, we can see that HIV prevalence in prenatal care clients appears to have declined in both major cities. A broad, 8-city study carried out in 2004 found prevalence to be just 0.46 percent among 5,267 women attending prenatal care, down from 1.35% prevalence in 3,248 women attending prenatal care in six regions in 1998. While it is difficult to identify clear causes for the decline, prevention efforts (education, behavior, etc.) may have played a role in reducing HIV among the general population. Policymakers identified the importance of reducing rates of mother-to-child transmission using the existing low-cost technologies available, and Honduras began implementing a program to prevent mother-to-child transmission in 2001. As a result, mother-to-child transmission rates have been reduced by more than 50% in the municipalities where the program operates. The program has also collected additional data on prevalence among pregnant women attending prenatal care, which shows a decline from 0.81% in 2001 (5,000 women) to 0.57% in 2005 (45,000 women). However, these data are likely to be biased by the fact that the program began in urban areas, where prevalence is higher, and the later data include also rural and peri-urban areas where prevalence tends to be lower. Sex workers, men who have sex with men, and Garífunas Since the onset of the epidemic, three groups have been consistently identified as particularly at-risk and have showed a high prevalence of HIV. Each group exhibits various intersecting risk and vulnerability factors (see IIb and IIc) that policymakers must keep in mind. Figure 5 shows available prevalence data for these three most heavily- affected populations (as well as for pregnant women attending antenatal care). Figure 5: Prevalence of HIV in Selected Populations Sex Workers Garifuna MSM Pregnant Women 14% 12% 10% 8% 6% 4% 2% 0% 1998 2001 2006 Source: Studies of HIV Prevalence and other STIs and behavior. 1998 and 1999, Health Secretary. Multicentric Study, Health Secretary/PASCA 2001. EVCV Health Secretary/CDC/GAP/CAP/ USAID, 2007. 6 HIV prevalence increased sharply among MSM—from 8 percent in 1998 to 13 percent in 2001—dropping to 10 percent in 2006. There was a small increase among SW—from 9.9 to 10.1 percent during the same years—then a sharp declined to 4.1 percent in 2006. HIV prevalence in the Garífuna population fell from 8.4 percent in 1998 to 4.5 percent in 2006. Several caveats are in order. First, the decline is statistically significant only for sex workers is. Further, as these data derive from different studies with different methodologies, care must be exercised in its interpretation. In Part II, nevertheless, we will explore a number of factors that could help to explain the apparent trends, factors which are relevant to the formulation of a successful response to the epidemic. The apparent persistence of high HIV prevalence in the MSM population despite reported increased condom use (see IIa) is of concern, and may suggest that other risk factors, most obviously STIs like syphilis (see IIb)contribute to the epidemic’s intractability in this group. Further, this persistence alongside the sharp decline in HIV among SWs may suggest that current trends toward the feminization of the epidemic will change (see IId). d) Summarizing the epidemic picture To summarize briefly: the picture that emerges is of an HIV epidemic primarily sexually- driven, and heavily concentrated in specific communities with vulnerability and risk factors. HIV has been concentrated largely on the Atlantic coast and in the larger cities. While the epidemic still affects young adults disproportionately, there is a rise in age group most affected, especially among men. Women, initially less hard-hit, increasingly have borne the brunt of new diagnoses, many initially among sex workers. It is important to formulate an accurate picture of where most new infections are occurring. But this simple snapshot is too limited and static as a good basis for policy and program decisions. So Part II considers additional factors that seem to underlie the snapshot, and which are important for a successful response. II. Factors Impacting a Successful Response Knowledge and behaviors, risk and vulnerability, and estimates of the epidemic’s likely future course are all crucial for policymaking and program development. This section explores factors that help better understand the empirical trends in the epidemic in geographical areas, age-groups, by gender, and among specific sub-populations. a) Knowledge and Behavior Information is fundamental to all prevention efforts. Knowledge about HIV and AIDS among at-risk populations has risen significantly (with the striking exception of PLHIV surveyed in 2006). For example, among MSM, 76 percent reported knowing that condoms reduce the risk of infection in 1998; by 2001 this had climbed to 97 percent, and 7 was around 99 percent among sex workers. In addition to this increase in awareness, condom use among at-risk populations has increased in recent years (Figure 6). Note that the data in Figure 6 on condom use by sex workers distinguishes between use during sex with clients/new partners (n), and with stable partners (s). Figure 6: Condom Use in Selected Populations Sex Workers (n) Sex Workers (s) MSM Garifuna 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1998 2001 2006 Source: Sierra M.A., Paredes C., Pinel R., Fernández J., Mendoza S., Branson B., Soto RJ. “Estudio Seroepidemiológico de VIH, Sífilis y Hepatitis B en hombres que tienen sexo con hombres de Tegucigalpa y San Pedro Sula, Honduras”, 1998. Secretaria de Salud de Honduras. “Estudio Multicéntrico Centroamericano de Prevalencia de VIH/infecciones de transmisión sexual y Comportamientos en Poblaciones específicas en Honduras”, 2003. “Encuesta Centroamericana de Vigilancia de Comportamiento sexual en hombres que tienen sexo con hombres y trabajadoras del sexo,” SSH/CDC/GAP-USAID, 2007. The increase in condom use was steepest among Garífuna men, jumping from a reported 11 percent in 1998 to 58 percent in 2006. Significant increases have been registered among the MSM population—from 47 percent in 1998 to 67 percent in 2006—and from 80 percent to 95 percent in the same years among SWs with new partners. The one exception seems to be condom use by SWs with stable partners, which actually decreased in 2001 before rebounding in 2006, but remains relatively low at around 30%. Condom use trends could be related to trends in HIV prevalence. The sharp increase in condom use by male Garífunas and near-universal use among SWs with new partners may have contributed to decreased HIV prevalence among these populations. The modest increase in condom use among MSMs may help explain their persistently high rate of infection. HIV testing has increased among MSM—from 45 percent in 1998 to 57 percent in 2001 and 63 percent in 2006—and is much higher than among other groups in the population, consistent with the particular importance of testing and counseling among groups with high HIV prevalence and risk. 8 b) Risk Factors Given the sexually-driven nature of the HIV epidemic in Honduras—accounting for more than 90 percent of all cases between 1985 and 2005—it should come as little surprise that sexually-transmitted infections (STIs) constitute one of the virus’s most significant risk factors. Table 1 shows the prevalence of selected STIs among pregnant women and the three identified at-risk populations: Table 1: Prevalence of STIs in Selected Populations Selected Prevalence of acute sexually transmitted Presence of Population infections Antibodies Syphilis Gonorrhea Chlamydia Herpes Hepatitis B Pregnant 0.5% (1998) 0.4% 6.0% (1998) women 2% (2006) 0.5% (2004) (2006) 2.8% (2004) Men who have 39.6% sex with other 1.2% (1998) Tegucigalpa men 4.6% (2001) 9.2% 12.5% (2006) 16.9% 5.6% Tegucigalpa (2001) (2001) 24.8% SPS (2001) (2006) (2006) 8.9% SPS (2006) 19.8% La Ceiba (2006) Garífunas 3.3% (2006) 0.4% 3.4 % 56.6% ND (2006) (2006) (2006) Sex workers 76.4% Tegucigalpa 15.2% (1998) 8.6% 15.5% (2006) 25.0% 6.8% (2001) (2001) (2001) 58% SPS (2001) 7.1% (2006) 5.2% (2006) (2006) 42% Comayagua (2006) Source: Sierra M.A., Paredes C., Pinel R., Fernández J., Mendoza S., Branson B., Soto RJ. “Estudio Seroepidemiológico de VIH, Sífilis y Hepatitis B en hombres que tienen sexo con hombres de Tegucigalpa y San Pedro Sula, Honduras,” 1998. Secretaría de Salud de Honduras. “Estudio Multicéntrico Centroamericano de Prevalencia de VIH/infecciones de transmisión sexual y Comportamientos en Poblaciones específicas en Honduras,” 2003. ”Encuesta Centroamericana de Vigilancia de Comportamiento sexual en hombres que tienen sexo con hombres y trabajadoras del sexo”, SSH/CDC/GAP-USAID, 2007. Although programs to treat STIs have not been demonstrated to decrease HIV incidence, presence of other STIs – especially ulcerating STIs – enhances the transmission risk for an exposed individual. These data suggest that syphilis rates have been declining among sex workers and are low among the Garífuna population and negligible among pregnant women, but remain high and appear to be on the rise among the MSM population. This suggests an important focus for programs to target MSM. 9 c) Vulnerability Factors There is a useful distinction to be made between risk and vulnerability. While populations deemed at risk are those which already demonstrate a considerable prevalence of HIV—in the case of Honduras, MSMs, SWs, and Garífunas—vulnerability arises from a group’s situation in society. In general, vulnerability indicates a variety of factors—power, information, behaviors—that have the potential to lead to a spread in HIV. Table 2 lists some key vulnerability factors in Honduras, and the populations affected: Table 2: Vulnerability Factors in Honduras Factors of Vulnerability Vulnerable groups Affected Mobility Drivers (trucks, taxis, etc.) Garífuna Armed services Migrants Consumption of alcohol/drugs Armed services Youth Drug users Stigma and discrimination MSM Sex workers Lesbians Poor knowledge of HIV Poor Rural residents Disabled Prisoners Peer pressure Youth Low use of condoms Housewives Domestic workers Prisoners Gender inequality Housewives Sex workers Domestic workers Lesbians High prevalence of STIs MSM Sex workers Poverty All Once again, we see the importance of the relationship between data and prevention: all of the at-risk populations which currently show a disproportionately high prevalence of HIV also appear here as vulnerable—a vulnerability which has no doubt contributed to their current at-risk status. Many of the other groups also affected by these vulnerability factors may be potentially at-risk in the future, important for taking into account in designing effective HIV prevention measures. 10 From a policymaking perspective, the concept of vulnerability poses the challenge of making every effort to include vulnerable populations who might find themselves at-risk in the future, without losing sight of the fact that it is above all risk and high incidence in specific communities that drives the epidemic. The difficulty of this challenge is reflected in the history of Honduran efforts to respond to the HIV/AIDS epidemic (see Part III), as prior programs have progressively generalized and expanded the lists of vulnerable target populations to effectively include the entire general population, resulting in a loss of focus in both analysis and provision of resources. d) 2007 HIV Estimates As the complexity of balancing current risk and future vulnerability demonstrates, one of the most urgent tasks of decision makers confronting HIV/AIDS is to anticipate future trends which could entail a shifting of risk toward new, previously vulnerable populations. One of the few tools available for such a task is estimation, and Honduran policymakers used two methods—the UNAIDS “Workbook Method,” and the “Mode of Transmission Model”—to estimate current HIV prevalence levels and predict future tends in Honduras. On the basis of past statistics and PMTCT studies, the participants agreed it made sense to make separate estimates for the Atlantic coast, and the rest of the country. The Workbook Method HIV prevalence estimate for 2007 for the north coast was 0.91 percent (with a range of 0.57 – 1.85 percent), and for the rest of the country, 0.59 percent (with a range of 0.37 – 1.2 percent). Nationally, the estimate was 0.68 percent (range of 0.42 – 1.37 percent). Past trends and future estimates of HIV prevalence in Honduras are charted in figure 7: Figure 7: Workbook Method Projection of the HIV, % Adult Prevalence, 1980-2010 Source: Spectrum model Projection While it is too early to say for sure, several possible causes for the current and predicted decline in HIV prevalence stand out. The death of early HIV sufferers and development of more and better national statistics could have contributed to this trend. However, we 11 should not discount the impact of behavioral changes, and especially the significant shift in knowledge of the epidemic and condom use in affected populations as a primary driving force behind the decline. But the high concentrations of a variety of STIs among these same populations should serve as a warning that, if these groups are not sufficiently targeted in the future, the epidemic could resurge. The second method used to generate 2007 HIV prevalence estimates—Mode of Transmission—sheds additional light not only on the number of expected cases, but also the populations among whom these infections are predicted to occur. The results of the Mode of Transmission estimate appear below, in Figure 8: Figure 8: Mode of Transmission Estimate of the HIV and AIDS Epidemic Source: UNAIDS Mode of Transmission Model (SSH, Honduras, 2007). Other vulnerable populations includes: prisoners, adolescents, women, uniformed officers, and men in the general population. Worryingly, if unsurprisingly, the MSM and SW populations together constitute 48 percent of expected new cases. The Garífuna population is expected to contribute only 5 percent of new infections, given their relatively small numbers, rise in condom use, and substantial decrease in HIV rates in recent years. It is particularly striking that fewer than 1 percent of new cases are expected to be among the sex workers themselves in the future, but that 15.3 percent of all new cases are expected to occur among clients of SWs and 12.4 percent among partners of those clients. Clearly, it is not sufficient to target only the specific at-risk population of SW, efforts must additionally track the expected course of the epidemic through their sexual partnership networks. The 43.8 percent of expected new cases that occur in “other vulnerable populations” include prisoners, adolescents, women, uniformed officials, and men. The very broad categories (women, men) indicates the need to better specify and determine the behaviors and populations expected to contribute nearly half of new cases of HIV in Honduras, to enable effective targeting. 12 e) Summarizing Additional Factors We now have a better basis for reviewing HIV programs and policies, after considering additional factors that have contributed to declines in HIV prevalence—shifts in knowledge and preventive behaviors—and risk factors that are likely to be driving incidence among specific populations. We have seen that an assessment of additional vulnerability factors and the use of modeled estimates of prevalence, incidence and transmission pattern can contribute to our understanding of the likely future development of the epidemic without losing focus on those populations currently most at-risk. III. Building Policy on Evidence This final section looks briefly at recent efforts to confront the HIV/AIDS epidemic in Honduras (PENSIDA II, 2003-2007), its successes and shortcomings, and the degree to which the data summarized above have influenced the formulation of the next stage of the national response to the epidemic: PENSIDA III (2008-2012). Moving Forward from PENSIDA II To briefly summarize the current status of the HIV epidemic in Honduras: earlier predictions of rapid growth of the epidemic—possibly based on poor data—have not proven accurate. Rather, lower HIV prevalence is estimated both among the general population and among most affected and at-risk populations. But the reasons for this are not entirely clear, and include factors as varied as the quality of prior data, the impact of PMTCT programs and behavioral changes among at-risk groups, contending with risk factors such as the prevalence of a variety of other STIs among different at-risk and vulnerable populations, and persistence of risky behaviors in some groups. The picture is complex, but does suggest that prevention policies may have had an impact. How has this review of the evidence on the current state of the epidemic influenced contemporary policymaking and program decisions? After implementing two prior HIV/AIDS programs in Honduras—PENSIDA I (1999-2002) and PENSIDA II (2003- 2007)—the National AIDS Commission (CONASIDA) assembled over 100 government officials, academics, NGO and civil society representatives, people living with HIV, and participants from the private sector and international partners. to evaluate the successes and failures of PENSIDA II, with an eye toward the development of a third national program: PENSIDA III (2008-2012). The Comprehensive Assessment of PENSIDA II was conducted through a participatory strategic process and facilitated by a “knowledge transfer” approach that provided guidance to the national team who led and managed the process. The thoughtful and thorough assessment of the strengths and shortfalls of PENSIDA II in the light of the latest evidence on the epidemic provided a strong basis for a renewed and honed response to the epidemic. While several areas of achievement were noted—specifically the promotion of preventive sexual and reproductive health (reflected in increased condom use), the reduction in 13 mother-to-child transmission of HIV, and the participatory nature of the program— PENSIDA II was deemed to have lacked the necessary resources and organization to represent a coherent response to the epidemic. Two interrelated shortcomings seem to have contributed to this. The first was in the area of monitoring and evaluation: PENSIDA II fell short both in its identification of key measurable criteria for success and in the absence of baseline figures against which to gauge success. The second significant critique had to do with the populations targeted: PENSIDA II identified 11 different prioritized sub-populations—ranging from at-risk populations with high prevalence to some in which prevalence remains negligible—suggesting that the program’s priorities were not particularly well-focused. In a context in which infection rates among the general population are low and declining, but in which some at-risk populations are experiencing high and seemingly-intractable epidemics, this lack of focus was in need of revision. Such concerns aside, it must nevertheless be noted that PENSIDA II made considerable progress in addressing the HIV/AIDS epidemic in Honduras. While measurable criteria for success were somewhat lacking, and while some think that PENSIDA II should have devoted more than the 33 percent of resources that were targeted to MSM and SW populations (41 percent of resources were for programs for the general population), the program’s overall success, as seen in falling prevalence rates among critical target populations, is no small feat. A Glance at PENSIDA III The lessons learned during the course of PENSIDA II and in assessing PENSIDA II against the epidemic and response data played a fundamental role in formulating the strategic orientation of PENSIDA III. On the one hand, PENSIDA III seeks to continue the effective nature of Honduras’ response to the epidemic, but at the same time it needs to be more than a vision and a direction—arguably the case with PENSIDA II—and serve as a realistic strategic plan for responding to the HIV/AIDS epidemic during the next five years. The plan of action for PENSIDA III reflects this learning process. In response to both the state of the epidemic in the country and the perceived shortcomings of PENSIDA II, PENSIDA III seeks to improve monitoring and evaluation capacity and, above all, to place a particular and proportional emphasis on those groups where prevalence of HIV and AIDS are highest, and those others with the highest degree of vulnerability. Defining the baseline One clear shortcoming of the PENSIDA II program was the lack of monitoring and evaluation, much of which had to do with the absence of baseline measures against which to determine progress. Five of the ten impact objectives identified in PENSIDA II could not even be measured, and nearly half of the program’s result objectives suffered from the same lack of data and conceptual clarity. The lack of baseline data was in many ways 14 resolved during the course of PENSIDA II, as new modes of data collection were instituted. It is on the basis of these new and improved data sources that PENSIDA III has formulated its new objectives, which are clearly defined relative to existing baseline measures. All the indicators in the current program refer to a baseline, so achievements should be measurable in future. PENSIDA III has formulated both specific “impact” goals and broader, behavior-based “result” goals. The impact goals are: at the very least to maintain the national prevalence rate; reduce prevalence in vulnerable populations with respect to the baseline; reduce STIs in the same communities; lower the mother-to-child transmission rate; reduce the transmission rate of congenital syphilis; reduce the rate of HIV infections associated with blood transfusions; reduce the prevalence of HIV-TB co-infection; and halve the mortality rate of HIV-positive people on antiretroviral treatment (ART). In order to realize the impact goals, PENSIDA III sets the following result goals: to increase knowledge of HIV transmission and prevention; to increase the median age of initiating teen sexual activity; to further increase condom use among the general population and those populations with the highest vulnerability; to increase the use of clean needles by intravenous drug users; to increase sexual and reproductive health education; to increase post-exposure prophylactic treatment; to increase voluntary HIV testing and counseling; to increase knowledge of and access to mother-to-child transmission prevention; to increase access to TB counseling, testing, and prevention among those with HIV; to increase accessibility to integral care and ART; and to increase treatment for children with HIV. The result goals also include seeking to: eliminate stigma and discrimination toward those who are HIV-positive and seek medical services; institutionalize public policy toward HIV, with a focus on human rights, gender equity, violence, poverty, and human insecurity; maintain the composite index of national HIV and AIDS policy; increase national and international spending on HIV and AIDS efforts in Honduras; and establish a mechanism for public accountability and rendering of accounts for HIV and AIDS programs. Given the experiences of PENSIDA II and acknowledged need for a strategic plan, PENSIDA III would be incomplete if it failed to make the transition from defining only what to do, to also consider how to do it. The lessons of PENSIDA II and the national evidence point in one central direction: concentrate resources and effort on the most vulnerable and at-risk communities. Reaching target populations PENSIDA III’s strategic orientation sets out from the following observation, which applies with special force to the Honduran experience: “All countries have adopted a broad approach to the prevention and control of the HIV virus. The list of potential objective groups has increased to include the entire population. This strategy must be revised to assure that the limited available resources be assigned to critical groups in 15 order to prevent transmission of the virus: sex workers, men who have sex with men, prisoners, and mobile populations.” 3 Given the high concentration of HIV prevalence and predicted transmission among specific at-risk groups in Honduras, it may seem surprising that PENSIDA II devoted some 41 percent of resources to the general population, compared with 33 percent combined to the MSM and SW populations. While such an approach may be consistent with the evolution of the epidemic in the country, the decline in HIV in the general population and high concentration among specific populations indicates that it is time to adjust this approach. Accordingly, PENSIDA III seeks to prioritize populations likely to have the greatest impact on the next 5 years of the epidemic. Using existing data and studies on HIV, existing prevalence of STIs, and knowledge of behavior, potential target populations were divided into three groups: 1.) Highest-priority – a. MSM b. Prisoners c. People living with HIV d. SWs e. Garífuna 2.) Particularly vulnerable but with low HIV prevalence – a. Adolescents and young people (aged 10-24) b. Pregnant women c. Orphans d. Maquila workers e. Uniformed officers 3.) Vulnerable groups on whom more research is necessary – a. Street children b. Various categories of women within the general population (housewives, domestic workers, victims of abuse, and maquila workers) c. The sexually-diverse (lesbians, transgendered, transvestites, bisexuals, transsexuals) d. Other ethnic groups (Misquitos and Tolupanes) e. People with different capacities f. Mobile populations (truck drivers, migrants, taxi drivers) g. Drug users 3 Bortman M, Saenz L, Pimenta I, Isern C, Rodríguez A, Miranda M et al. Reducing HIV/AIDS Vulnerability in Central America: Honduras: HIV/AIDS Situation and Response to the Epidemic. 1-52. 2006. Washington DC, The World Bank. 16 Through this new system of categorizing both at-risk and potentially vulnerable populations, PENSIDA III seeks to walk the fine line between excessive focus and excessive generality. While the majority of resources will be devoted to the first category of populations, this categorization allows the third category—about which little accurate information exists—to remain on the national research agenda. Funding PENSIDA III The PENSIDA III team needed to know what activities would cost and whether resources would be sufficient to implement the activities planned. This gave the impetus to an analysis of actual spending under PENSIDA II, and a careful costing exercise to estimate the cost of the planned program and any resource gaps. In line with the program’s emphasis on aggressively targeting specific populations, estimated costs per population were calculated on the basis of: 1.) The estimated unit cost of reaching the population 2.) The size of the target population 3.) The current number receiving services under PENSIDA II 4.) The estimated number to receive services under PENSIDA III Before even considering projected costs, PENSIDA III’s estimates of service delivery demonstrate the program’s commitment to target populations. For both the SW and MSM communities, the program devotes resources to both education and promoting condom use. PENSIDA III aspires to increase indicators of knowledge about HIV transmission from 54 percent (2006) to 93 percent among SWs by 2012, and condom use from 86 percent to 95 percent during the same period. The goals for the MSM population are to increase knowledge from 28 percent (2006) to 75 percent (2012) and increase condom use from 67 percent (2006) to 90 percent (2012). For the Garífuna population, the focus is above all on education, with the aspiration of increasing knowledge about the epidemic from 20 percent (2006) to 80 percent (2012). PENSIDA III foresees a progressive increase in funds allocated for these and other target populations, eventually doubling the funds directed toward target populations: from $100,000 to $200,000 annually for Garífunas; from $300,000 to $600,000 annually for SWs; and from $800,000 to $1.6 million for MSMs. The differences are not a question of relative priority, but are driven by the size of the target population in question and the estimated unit costs of reaching an individual in that population (the estimated unit costs for reaching SWs, for example, are four times that of Garífunas). While PENSIDA II’s lack of a structured budget makes detailed comparison difficult, PENSIDA III certainly foresees a marked increase in overall program funding. Resources made available for PENSIDA II totaled some $21 million annually between 2005 and 2007, PENSIDA III looks ambitiously toward a gradual increase from $30 million in 2008 to $49 million in 2012. 17 References CONASIDA. Analisis de la Situacion y de la Respuesta de Honduras ante la epidemia del VIH/SIDA y en el marco del PENSIDA-II periodo 2003-2007. Tegucigalpa, Honduras; 2007. CONASIDA. III Plan Estratégico Nacional de Respuesta al VIH y SIDA en Honduras. PENSIDA III, 2008 – 2012. República de Honduras, Comisión Nacional de SIDA, Tegucigalpa, Honduras; Diciembre, 2007. Synthesis of the HIV Situation in Honduras and an Analysis of the Results from PENSIDA II. Unpublished Draft, September 14, 2007 18 W WW. WO RL DB AN K . O RG /AI DS For more information, please contact: The Global HIV/AIDS Program World Bank Group 1818 H St. NW Washington, DC 20433 Tel:   +1 202 458 4946 Fax: +1 202 522 1252 wbglobalHIVAIDS@worldbank.org