UNIVERSAL HEALTH COVERAGE STUDY SERIES No. 26 Please cite this paper as: Cotlear, D. and Rosemberg, N. 2018. “Going Universal in Africa: How 46 African Countries Reformed User Fees and Implemented H ealth Care Priorities”. Universal Health Care Coverage Series No. 26, World Bank Group, Washington, DC. 2 UNIVERSAL HEALTH COVERAGE STUDY SERIES No. 26 Going Universal in Africa: How 46 African Countries Reformed User Fees and Implemented Health Care Priorities Daniel Cotlear and Nicolas Rosemberg The World Bank, Washington, DC, 2018 3 4 TABLE OF CONTENTS Preface to the second round of the Universal Health Coverage Study Series ................................ 7 Acknowledgement .......................................................................................................................... 8 About the Authors ........................................................................................................................... 9 Executive Summary ...................................................................................................................... 10 1. Introduction ........................................................................................................................... 11 2. Conceptual Framework and Methodology ............................................................................ 12 3. Limitations............................................................................................................................. 13 4. Paths toward the Reduction of User Fees .............................................................................. 13 5. Setting Priorities for Health Interventions—the Basic Package and Beyond ....................... 16 6. Setting Priorities for Whom to Cover—Subpopulation Coverage Programs ........................ 18 7. Issues of Implementation in the Subpopulation Coverage Programs .................................... 20 8. Decentralization..................................................................................................................... 24 9. Acknowledging that health systems in Africa are not a blank slate -- An illustration based on the cost of a broader universal benefit package ........................................................................ 24 10. Summary and Conclusions ............................................................................................. 27 Annex 1 Acknowledgments .......................................................................................................... 29 Annex 2 Additional Data .............................................................................................................. 30 Figures Figure 1. The Universal Health Coverage Cube ........................................................................... 12 Figure 2. User-Fee Policies around the Turn of the Millennium and their Reform, Africa ......... 14 Figure 3. Countries Charging User Fees in Public Clinics and Hospitals (%) ............................. 15 Figure 4. Composition of the Basic Package (% of countries with no user fees for each intervention) .................................................................................................................................. 17 Figure 5. Countries with Health Coverage Programs for Specific Subpopulations (%) .............. 19 Figure 6. Programs with Output-Based Payments, Africa ............................................................ 22 Figure 7. Autonomy Granted to Managers of Public Health Facilities ........................................ 23 Tables Table 1. Selected Economic and Health Indicators of the African Countries in the Study .......... 13 Table 2. Average Coverage of Population Programs.................................................................... 23 Table 3. Cost of prioritized Benefit Packages: Three scenarios depending on country choice ... 26 5 Abbreviations CBHI Community-based Health Insurance DAH Development Assistance for Health DCP3 Disease Control Priorities 3rd Edition HIV/AIDS Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome HMICs Higher-Middle-Income Countries LICs Low-Income Countries LMICs Lower-Middle-Income Countries MDGs Millennium Development Goals PMT Proxy-Means-Test SDGs Sustainable Development Goals SHI Social Health Insurance THE Total Health Expenditure UBP Universal Basic Package UHC Universal Health Coverage UNICO Universal Health Coverage Studies Series 6 Preface to the second round of the Universal Health Coverage Study Series All over the world countries are implementing pro-poor reforms to advance universal health coverage. The widespread trend to expand coverage resulted in the inclusion of the “achieving universal health coverage by 2030” target in the Sustainable Development Agenda. Progress is monitored through indicators measuring gains in financial risk protection and in access to quality essential health-care services. The Universal Health Coverage (UHC) Studies Series was launched in 2013 with the objective of sharing knowledge regarding pro-poor reforms advancing UHC in developing countries. The series is aimed at policy-makers and UHC reform implementers in low- and middle-income countries. The Series recognizes that there are many policy paths to achieve UHC and therefore does not endorse a specific path or model. The Series consists of country case studies and technical papers. The case studies employ a standardized approach aimed at understanding the tools –policies, instruments and institutions- used to expand health coverage across three dimensions: population, health services and affordability. The approach relies on a protocol involving around 300 questions structured to provide a detailed understanding of how countries are implementing UHC reforms in the following areas: • Progressive Universalism: expanding population coverage while ensuring that the poor and vulnerable are not left behind; • Strategic Purchasing: expanding the statutory benefits package and developing incentives for its effective delivery by health-care providers; • Raising revenues to finance health care in fiscally sustainable ways; • Improving the availability and quality of health-care providers; and, • Strengthening accountability to ensure the fulfillment of promises made between citizens, governments and health institutions. By 2017, the Series had published 24 country case studies and conducted a systematic literature review on the impact of UHC reforms. In 2018 the Series will publish an additional15 case studies, A book analyzing and comparing the initial 24 country case studies is also available: Going Universal: How 24 Developing Countries are Implementing UHC Reforms from the Bottom Up . Links to the Series and the book are included below. Daniel Cotlear, D. Phil. Manager and Editor Universal Health Coverage Study Series Links: http://www.worldbank.org/en/topic/health/publication/universal-health-coverage-study-series http://www.worldbank.org/en/topic/universalhealthcoverage/publication/going-universal-how- 24-countries-are-implementing-universal-health-coverage-reforms-from-bottom-up 7 Acknowledgement The authors wish to acknowledge the comments and guidance received from Magnus Lindelow, Christoph Kurowski, Moulay Driss Zine Eddine El Idrissi, Netsanet Walelign Workie, and Laurence Lannes. The report uses a database of health financing policy implementation built with contributions from World Bank Group country-facing staff supporting 46 African countries. Their names are listed in Annex 1. 8 About the Authors Daniel Cotlear is Lead Economist in the Global Practice, Health, Nutrition and Population, of the World Bank. He supports countries implementing policies to advance Universal Health Coverage (UHC). The Universal Health Coverage Study (UNICO) series published 24 country case studies analyzing how policy makers around the world are implementing UHC, and a synthesis of the findings from those countries. His most recent publications include the book, Going Universal: How 24 Developing Countries are Implementing UHC Reforms from the Bottom Up, co-authored with four World Bank colleagues, and two Lancet articles discussing UHC in Latin America. New projects include a review of UHC policies in 46 African countries, a new round of UNICO studies in 17 countries, and a survey of the impact of population aging in Latin America. Daniel holds a Doctor of Philosophy from Oxford University, and a bachelor’s degree from Pontifical Catholic University of Peru. Nicolas Rosemberg is a Consultant to the World Bank. He focuses on health financing topics and on the promotion of equity in health. His work includes the review of universal health care policies in 46 African countries, a study on propoor health coverage programs in Tanzania, and a Public Expenditure Review of the health sector in Zambia. Before joining the Bank, Nicolas worked as a Consultant at the Inter-American Development Bank and the GIZ. He holds a Master’s in Public Policy from the Hertie School of Governance in Berlin, Germany, and a Bachelor’s degree in Economics from the University of Buenos Aires. 9 Executive Summary African countries have made significant health financing reforms in recent decades. During the 1980s and 1990s, most countries established user fees for publicly provided services as an instrument to finance health care and to balance the demand and supply for care. After the turn of the millennium, there was a significant change in sentiment leading many countries to reduce user fees and replace them with new financing instruments to compensate for the resulting loss of revenue and the greater use of services. The reduction in user fees became a key element in the regional strategy to advance toward universal health coverage. This study created a database identifying health financing policies in 46 countries in Sub-Saharan Africa and North Africa and provides a systematic portrayal of the health financing strategies existing in the region. The findings reveal that to advance UHC, most African countries are using two strategies to compensate for the reduction of fees charged at the point of service and that 82 percent of countries are decentralizing health care financing to subnational governments. The two strategies consist of the implementation of: (i) a narrow universal basic package (UBP) consisting of prioritized interventions for which no user fees are charged; and (ii) subpopulation coverage programs, which are designed to provide access to a wider set of benefits than those provided by the UBP with zero or low user fees. Contributory programs have been established for the higher-income subpopulations in two-thirds of African countries. Non-contributory (i.e. fully-subsidized) programs have been created for the poor and other vulnerable populations, in 80 percent of African countries. The study concludes with a discussion about the implications of these findings for planning and implementing reforms to advance universal health coverage; specifically, it shows that the cost of achieving UHC would be much higher than is estimated unless substantial reforms are introduced when injecting additional funds. 10 Introduction This paper describes the health financing policies used today by African countries to expand health coverage. It identifies key health financing policies used by African countries and measures the existence of regional patterns in the use of these policies. The paper does not attempt to identify “best practices,” nor does it try to measure the effective coverage of the policies or their impact. Rather, it aims to add value to the existing literature by providing a systematic portrayal of the health financing policies that are in place across the region. The study concludes with a discussion about the implications of its findings for planning next steps to advance universal health coverage; specifically, it argues that the debate regarding the cost of achieving UHC needs to include a discussion about the health system reforms required for a successful implementation of the recommended benefit packages. The study contributes to the dialogue about health financing in a context where a major development paradigm is shifting. The development agenda in health has transitioned from the relatively narrow focus of the Millennium Development Goals (MDGs) to a broader one in the Sustainable Development Goals (SDGs). Universal health coverage (UHC), defined “as a situation where all people who need health services receive them without undue financial hardship,”1 has been embraced by the SDGs agenda. The new agenda goes beyond the MDGs in three ways. First, it maintains the MDGs’ concern with mothers and children, but expands its focus on health and well-being to every stage of life. Second, the desired range of health benefits is wider and explicitly includes “needed promotive, preventive, curative and rehabilitative health services.”2 Third, financial risk protection for the health care costs associated with illness is explicitly recognized as part of the SDG agenda. The context has also shifted as there is an expectation, made explicit in the SDG agenda, that most of the projected increase in health funding in Africa must come from a higher allocation of domestic resources to health as future flows of development assistance for health are not expected to grow at the rapid pace they did between 2000 and 2010 (see Annex 2, figure A2.1). Furthermore, the recent framework for action for UHC in Africa states that government health spending as a share of total government spending has decreased by half in the continent, and that only four countries met the Abuja Declarations target3 of 15 percent of general government spending in 2014.4 The rest of the paper is organized as follows. Section 2 describes the conceptual framework and methodology used in the study. Section 3 describes the different paths chosen by African countries to expand health coverage. Sections 4 and 5 describe the two key instruments used in that journey: the universal basic package of health services and the subpopulation health coverage programs. Section 6 reviews in detail some of the technical instruments required for the successful implementation of these policies. Section 7 uses the findings of the paper to discuss the cost estimations of implementing a broader benefit package to advance universal health coverage. The paper concludes with a summary of the findings and their implications. 11 Conceptual Framework and Methodology This paper uses the concept of “coverage” linked to the “UHC cube” popularized by the World Health Organization 2010 World Health Report.5 The cube identifies three dimensions of “health coverage”: Population: Who is covered? Services: Which services are covered? And Affordability: How much of the cost is public or prepaid and how much is p paid at the point of service (see Figure 1). Figure 1. The Universal Health Coverage Cube Source: WHO 2010. The findings of the report are derived from a data collection exercise designed specifically for this study and conducted in 46 African countries, including countries in North Africa and Sub-Saharan Africa. Several key characteristics of these countries are described in Table 1 (the full list of countries is presented in Annex 2, Table A2.1). The study includes 24 low-income countries (LICs), 15 lower-middle-income countries (LMICs), and 7 higher-middle-income countries (HMICs). Any study about Africa must recognize the vast income differences among these three groups. For example, per capita income in African LMICs is three times the per capita income in African LICs, and income in HMICs is almost 10 times higher. The average total fertility rate and the under-five mortality rate in African countries are high compared with countries of similar income in other regions. However, within Africa, these health outcomes improve significantly as income levels grow. 12 Table 1. Selected Economic and Health Indicators of the African Countries in the Study Average GDP Per Capita, PPP Average Total Average Mortality Number of Income Level (constant 2011 international $) Fertility Rate Rate, Under-5 Countries (2014) (2014) (per 1,000) LIC 24 1,435 5.1 84 LMIC 15 4,569 4.2 65 HMIC 7 15,260 3.0 48 Total 46 4,561 4.5 72 Note: HMIC = higher-income country; LIC = lower-income country; LMIC = lower-middle-income country; PPP = purchasing power parity. A questionnaire was designed to identify health financing policies and instruments used to expand health coverage in the African continent. Following the UHC-cube framework discussed above, questions focus primarily on policies and programs designed to reduce financial barriers to access. The questionnaire also enquired about planning documents for UHC and about the decentralization of health care functions to subnational governments. The questionnaire was field-tested in eight countries and distributed to World Bank Group staff assigned to each country. In most cases, the questionnaires were completed in consultation with government officials. A health financing policy database was compiled and reviewed in two consecutive rounds with the informants that provided the country-by-country data. (The names of the individuals who provided the data for each country are listed in the acknowledgements). In addition to the data collection exercise, a non-systematic literature review was conducted and international health financing databases were analyzed to contextualize findings. Limitations This paper provides a portrayal of UHC health financing policies implemented in Africa. It describes the path African countries have chosen in the pursuit of UHC. All over the developing world, there is a gap between health policies on paper and health policies on the ground; this paper makes no attempt to measure that gap, it simply identifies the official policies in place. The study also does not attempt to identify “best practices”. Paths toward the Reduction of User Fees Despite the relatively short history of the global movement toward universal health coverage, African countries have rapidly joined the movement. As of 2016, 44 of the 46 African countries included in this study have officially committed to achieving UHC. Among them, 80 percent have an official Health Sector Strategy document describing their medium-term plans for the sector. Thirteen of these countries have even declared a target year (usually sometime between 2020 and 13 2035) for “reaching UHC.” In addition, 33 African countries are either in the process of developing or have already finalized a Health Financing Strategy.6 During the late 1980s and early 1990s, as populations and social expectations grew while public budgets were often reduced, African countries sought new sources of funding for health and often relied on the systematic imposition of user fees for publicly provided health care.7 The number of countries introducing user fees increased significantly, because user fees were considered a feasible option to raise revenue in weak tax-based systems8 and as an instrument to improve health service access and engage the community.9 Around the turn of the millennium, 41 African countries (89 percent) had established user fees. Since their introduction, user fees have been controversial. Many studies have looked at the impact of user fees on service utilization, and most conclude that user fees negatively influence utilization by imposing a financial barrier to access,10 while their removal increases health service utilization.11 Conclusions are less clear regarding equity. In some cases, user fees were regressive, because they impose a significant burden on the poor.12,13 In other cases, no impact on equity was found.14,15 Research has also been conducted, to a lesser extent, on the impact of user fees on the quality of health care16 and the health status of the population.17,18 A central feature of UHC reforms in Africa today is the reduction of user fees for public health services and for publicly provided drugs. There now exists a broad consensus across most African health policy makers that this is a crucial step to increase access to health care and to reduce the financial risks associated with ill health. Accordingly, 80 percent of countries (37 of the 41 countries that had use fees) have implemented reforms to reduce or eliminate user fees (Figure 2). Figure 2. User-Fee Policies around the Turn of the Millennium and their Reform, Africa 14 Given the large contribution of user fees to total health expenditure, it was difficult for most countries to completely abolish user fees, and instead they took incremental steps to reduce them. Initially, health financing reforms focused on ensuring the provision of prioritized health services (“a basic package”). In line with the health priorities set by the MDGs, exemption policies were designed to target primarily treatment for HIV/AIDS, malaria and tuberculosis, and maternal and child care.19 Figure 3 shows the utilization of user fees in Africa today. Two-thirds of countries are charging user fees at all levels of care, 15 percent of countries have eliminated fees at the primary care level, and 20 percent charge no fees at any level. The utilization of user fees varies by country income level. There exist user fees at the primary care level in 75 percent of LICs, 67 percent of LMICs, and only 29 percent of HMICS. Figure 3. Countries Charging User Fees in Public Clinics and Hospitals (%) 100% 90% 29 80% 70% 65 67 Countries (%) 75 14 60% 50% 40% 30% 57 15 20% 8 27 10% 20 17 7 0% All LIC LMIC HMIC No user fees Fees only at hospital level Fees at all levels The use of incremental steps in the reduction of economic barriers to access requires choosing priorities and answering this fundamental question (Box 1): What health services should be made universal first? Countries responded to this question by creating a basic package for priority interventions, a discussion of which follows. 15 Box 1 Implicit Rationing Many studies reveal that despite the de jure abolition of user fees, health services are still not provided for free, and considerable financial barriers to access remain.a We attempted to collect information regarding the health system’s response to the elimination of user fees in the nine African countries with no user fees (four countries never established them and five abolished them). The LICs and LMICs that abolished all fees report to be struggling with various symptoms of implicit rationing, that is, long waiting lines for services; frequent stock-outs of drugs, sometimes accompanied by the appearance of a black market for drugs; and the establishment of private wings in public hospitals (as a means of bypassing the lines and securing timely access to drugs). These problems are less acute in the high-income and higher-middle-income countries that abolished fees. Unfortunately, we could not collect comparable information for countries that do charge user fees. Measuring the extent of implicit rationing in Africa is a task for future research. Sources: a. Laokri et al. 2013; Ben Ameur et al. 2012. Setting Priorities for Health Interventions—the Basic Package and Beyond The study found a high level of convergence across countries regarding which interventions to cover first. This convergence is reflected in the contents of the basic package—the health benefits or interventions the country decides to make available universally—to the whole population and at no cost at the point of use. Figure 4 shows the existence of four levels of convergence across countries regarding the composition of the benefits package. 16 Figure 4. Composition of the Basic Package (% of countries with no user fees for each intervention) Note: 87 percent of countries have at least one malaria intervention in their basic package. ANC = antenatal care; ARI = acute respiratory infection. The higher level of convergence involves over 90 percent of countries and includes services related to immunizations, HIV/AIDs, and tuberculosis. The second level of convergence, involves about two-thirds of countries and adds interventions related to malaria to the benefits package. While almost 90 percent of countries have a malaria intervention in their basic package, the specific interventions vary by country income group. LICs tend to eliminate fees for prevention (free mosquito nets) but often retain charges for treatment, while MICs tend to include treatment in the basic package but maintain charges for mosquito nets. The third level involves around 60 percent of the countries and introduces services for child care and prenatal maternal care. The fourth level of convergence is for skilled birth attendance. These services are universally free of charge in slightly over half of African countries (including natural birth and cesarean section). An intriguing feature related to this level is that it is found more often in low-income-countries than in middle-income countries. Government officials interviewed for the study suggested that this could reflect the higher cost faced by MICs, where a larger proportion of total births takes place in hospitals. They also suggested that introducing no-fee services in hospitals (which often 17 rely on user fee income for a significant part of their costs) may be more complicated than doing it in primary care clinics (where most of the other basic package services are provided). While these four groups of interventions are part of the basic package in more than half of all African countries, few countries have expanded the basic benefit package to the whole population beyond these four groups. Only 20 percent of countries have eliminated user fees for all outpatient medical services provided at the primary care level, only 10 percent of countries have eliminated user fees for inpatient services (beyond birth deliveries), and even fewer offer universal access to basic diagnostics or specialized outpatient services at no charge. While most African countries have chosen a narrow basic package with relatively few benefits, many countries are making significant efforts to expand benefits beyond the basic package. How they do this is described in the next section. Setting Priorities for Whom to Cover—Subpopulation Coverage Programs Beyond the interventions of the universal basic package, most African countries are also expanding available health interventions at no cost at the point of service in a more targeted manner. This coverage expansion is achieved through implementation of health coverage programs designed to eliminate the financial barrier to additional services for specific subpopulations. Some of these programs are contributory, requiring a prepayment by the beneficiaries. Sixty-one percent of African countries have contributory social health insurance (SHI) programs for civil servants and about half of countries have such programs for formal sector employees (Figure 5). In addition, many of the countries with no SHI programs are evaluating creating one. Participation in these programs, when they exist, is compulsory, and contributions are automatically deducted from employees’ paychecks. 18 Figure 5. Countries with Health Coverage Programs for Specific Subpopulations (%) 90 83 80 80 70 60 60 53 Countries (%) 50 43 40 33 30 20 17 10 - Mothers and The poor Civil servants Private sector Local The elderly Children under 18 children under 5 (SHI) employees (SHI) communities (CBHI) Advocates of strengthening contributory programs argue that these schemes improve resource mobilization for health and pool financial and health risks.20 Contributory schemes face important challenges in countries in which a high share of the workforce is employed in the informal sector.21,22 The literature on SHI highlights the complexity of reforms required for its implementation. A significant part of the research analyzes the political and financial feasibility of the introduction of SHI schemes. Some examples include the assessment of the financial preparedness of Lesotho and Swaziland for the introduction of an SHI scheme,23,24 the evaluation of different reform scenarios in South Africa and Kenya,25,26 and discussions around the political acceptability of SHI in Kenya and Nigeria.27,28,29 In Ghana and Gabon, SHI has been combined with schemes to cover the informal sector and the poor.30,31 Few papers, however, discuss the mechanisms for financing exemptions for the poor or the institutional arrangements that promote the corresponding good financial fund management.32,33 Given the above-mentioned difficulty of taxing the informal sector, over 40 percent of countries have opted for the creation of voluntary schemes such as community-based health insurance (CBHI).34 The fervor for CHBI can be explained by two interconnected factors. First, CBHI can mobilize resources from a group of the population that has the ability to pay for health services but is hard to reach by weak tax-based systems. This group often includes nonpoor rural communities and the informal sector. Second, CBHI provides financial and health care protection35 to a sector of the population that would otherwise lack access. 19 The range and scope of CBHI schemes varies enormously. Bennett (2004) developed a typology of CBHI arrangements illustrating the link between these arrangements and the broader health financing landscape and emphasizing the degree of participation of the state. In some countries, the national government (or donor agencies) directly subsidizes the scheme (for example, Tanzania’s Community Health Fund) to cover the premiums for the poor or to ensure its financial sustainability. Bennett argued that even in the absence of state subsidies, most CBHI rely on the state for the provision of certain health services.36 Some evidence suggests that government-run schemes are less successful at enrolling marginalized groups than community run schemes.37 Most African countries have also established noncontributory (that is fully subsidized) coverage programs. As shown in Figure 5, 80 percent of countries have such programs for the poor, and (separately) a similar proportion of countries have programs for mothers and children under 5. Programs for other age groups are much less common—only a third of countries have programs for the elderly and less than a fifth have programs for adolescents. Issues of Implementation in the Subpopulation Coverage Programs The new “subpopulation coverage programs” are different from the old “vertical programs.” The vertical programs financed inputs for single disease activities and operated within the structure of health care providers, encouraging the development of silos in health care provision. The new population programs do not operate within the providers; rather, they are designed as financing vehicles, paying providers for the delivery of (user-fee-free) services to prioritized populations. They do not create silos in health care provision or in supply chain systems. Because they operate outside the organization of the provider, they separate financing from provision. They can be challenging to implement because they require the use of new skill sets, including identification and enrollment of beneficiaries, development of payment systems, and the use of new financial management systems. While the development of these skills in many LICs imposes a significant challenge, once established these skills often allow the use of better incentives and stronger accountability mechanisms. Subpopulation coverage programs vary in their operational complexity and in their coverage. At one extreme, some operate as full-fledged health insurance schemes, such as Ghana’s national health insurance system. That program has explicit benefits packages, clear payment systems, and fee schedules to finance providers, claim management, and audit systems. At the other end of the spectrum, programs (especially those for the poor) sometimes simply consist of official statements decreeing that services should be provided without charge to a certain population together with a financial transfer to pay for the loss of revenues to providers for the implementation of exceptions. In the weakest contexts, these programs are simply unfunded mandates. What follows is a discussion of several technical characteristics of these programs in relation to pooling, targeting, provider payments, and managerial autonomy. Pooling: In Africa, programs for the poor are usually autonomous; however, there are a few countries where the poor are combined with other subpopulations. In Ghana and Gabon, for 20 example, there are health insurance agencies managing schemes that cover poor populations and the formal sector under a social health insurance model. In Ghana, these two subpopulations are covered by the same scheme and the funds for both populations are commingled. In Gabon, the SHI agency manages three separate schemes with different benefit packages. A more common arrangement is to pool resources for the poor and for the informal sector under community-based health insurance schemes, as in Rwanda and Ethiopia. Targeting: Worldwide, there now exist highly developed instruments and methods to identify the poor and channel subsidies to them. While these instruments have existed for several decades for social programs, only recently have they been used for health coverage efforts. There is a global trend to evolve from the use of simpler methods to more rigorous but demanding methods to identify the poor.38 The simplest methods are based on geographic targeting to prioritize poorer regions. Systems targeting individuals within regions range from systems of subsidies administered at the local level by local authorities, community leaders, or personnel of health facilities based on their “local knowledge,” to more refined proxy means tests (PMTs), which are managed at the national level by specialized agencies charged with developing “targeting registries” for the use of various social programs. In Africa, most countries use the simpler methods, but following the global trend, almost a third of countries are now using PMT systems. The PMT systems are more demanding in capacity and are proportionately more common in MICs than in LMICs. Eighteen African countries are also implementing incentive programs for the poor, consisting of cash transfers or some type of voucher to incentivize the use of health services (usually for maternal and child services). The targeting registries developed by social safety net programs can help governments identify the poor and better allocate health subsidies to the most vulnerable sectors of the population. Provider Payments: Many of the programs that cover specific subpopulations pay health care providers for services provided to their members. These payments sometimes involve linking money with services produced and/or with coverage and quality achieved in predetermined indicators. Our survey shows that almost two-thirds of African countries are experimenting with these type of payments, although almost 75 percent of these programs are pilots with limited regional coverage. Ministries of health are the main purchaser of services in 58 percent of the cases, and ministries of social welfare and ministries of finance each fulfill this task in 4 percent of the cases. After national ministries, the most common purchaser of services are local-level third- party payers (23 percent of the programs). Donor organizations play this role in 12 percent of cases. Fee-for-service is the preferred provider payment mechanism, and it is used in 60 percent of the reported programs. While most of these schemes focus on reproductive, maternal, newborn, and child health services, some provide a more comprehensive package and include even tertiary- level care (for example, Burundi) (Figure 6). 21 Figure 6. Programs with Output-Based Payments, Africa Managerial Autonomy: When health coverage programs transfer funds to providers, providers need some autonomy to respond to incentives and to develop the capacity to use the funds. Even though the autonomy of the facilities at the lower level was initially introduced by the Bamako Initiative of 1987,39 the introduction of programs that pay for outputs may be accelerating the granting of autonomy to managers—in 28 of the 29 countries where programs that pay providers for outputs are in place, facilities also enjoy a certain degree of autonomy. The study found that 87 percent of African countries are now granting some form of autonomy to public providers to allow them to make use of the transfers. This autonomy is exercised in several different ways. The survey inquired about the use of revenue, human resources management (hiring and firing staff), procurement, and asset management. It was found that procurement is the most common manifestation of facilities’ autonomy. In most cases (13 of 19 countries for which information was available), facilities have the capacity to procure some items outside the official supply chain, and in some cases, they even have their own bank accounts (for example, the Republic of Congo). Procurement was followed by use of revenue and management of assets. These often include medical and nonmedical equipment, vehicles, and the facility’s building and land. Human resources management is rarely a responsibility of lower-level facilities; only 43 percent of the facilities had the autonomy to hire personnel, and only 23 percent had the autonomy to fire personnel (Figure 7). 22 Figure 7. Autonomy Granted to Managers of Public Health Facilities Panel A Panel B Are public health facilities Areas of autonomy granted autonomy in some 100 88 areas of management? 90 80 80 73 Countries (%) 70 60 50 43 No 13% 40 30 23 15 20 10 Yes - 87% Coverage: One of the biggest surprises of the study was the finding that few countries have readily available data regarding the coverage achieved by their subpopulation programs. Countries have established subpopulation coverage programs, but in most cases, they fail to systematically monitor their coverage. The difficulty encountered in obtaining information about the programs in most countries surveyed suggests that often there is no systematic effort to track active program membership, program revenue and expenditure, or capacity to provide effective access to the promised benefits. To get a sense of the coverage achieved, we asked each country team to estimate, in collaboration with government officials, a likely coverage range for each program. Estimates were obtained for only 18 countries and have many methodological weaknesses, but they are presented in Table 2 as an illustration for the purposes of discussion. Table 2 shows the average coverage for the 18 countries that had any type of program. The results suggest that SHI coverage is generally high for civil servants and lower for the private sector. Countries with CBHI have low levels of national coverage, except for Rwanda, which reports high enrolment. There is a low average coverage, but high variance for the programs for the poor. Two countries—Rwanda and Gabon—report very high coverage, one country has medium coverage, three countries have low coverage, and six have very low or extremely low coverage. Table 2. Average Coverage of Population Programs Non- Non- SHI for Private Community- SHI for Civil contributory contributory Sector Based Health Servants Programs for Programs for Employees Insurance the Elderly the Poor 86% 49% 29% NA 30% Note: *The average is estimated only for countries with positive values of coverage, that is, zeros are not included. It assumes countries are at midrange of coverage scale estimations. 23 Decentralization There is a massive process of decentralization in Africa—82 percent of African countries have recently decentralized health care functions. In 34 of the 37 countries with decentralized health care functions, the process of decentralization is not complete. In most countries, the process of health reform is planned by health sector decision makers independently of the process of decentralization, which often involves a very different group of local and national decision makers. The decentralization process often includes a decision to allocate a certain share of fiscal revenues to subnational governments and correspondingly transfer to them the responsibility for running district and regional hospitals and community health care services. In many countries, these new inter-fiscal transfers take the form of “block grants”; in that context, the subnational government has the liberty to decide what share of the grant to assign to health care and what share to assign to other sectors. Having transferred the funds assigned for health care, ministries of finance often lack the means to monitor local-level allocations to the health sector. Acknowledging that health systems in Africa are not a blank slate -- An illustration based on the cost of a broader universal benefit package The study found that in the last two decades, most countries in Africa have developed health financing systems that are fragmented and include: a sub-system providing a narrow universal basic package; various contributory and non-contributory subpopulation programs covering a broader package of services; and multiple subsystems of services financed by subnational governments. We found that fragmentation is widespread throughout the region, implying that there are strong political forces driving the establishment of this configuration. In this section, we argue that any plans to implement change need to acknowledge that the starting point is one of fragmentation. Assuming for planning purposes that health systems are a blank slate or that there exists a single decision maker that controls all financing in the country may be misleading. In this section, we illustrate this point using as an example the debate surrounding the cost of achieving UHC. Attempts to estimate the cost of achieving universal coverage began during the MDG era, with the WHO Commission on Macroeconomics and Health estimating that a basic service package would cost the equivalent of US$ 71 per capita in current dollars.40 A few years later, McIntyre and colleagues proposed a minimum target for UHC spending of five percent of GDP per capita, which would imply around US$ 86 per capita.41 In recent years, there have been three large-scale efforts to identify the health interventions that should be prioritized in the expansion of the benefits package: The Lancet Commission on Investing in Health; 42 the Disease Control Priorities 3rd Edition (DCP3);43 and the 2017 WHO high level taskforce that produced a report about financing for the achievement of the health SDGs.44 The three efforts also projected the number of deaths averted per year by the expansion of the benefit package, with estimations ranging between 2 million and 4.5 million for LICs and between 4.2 million and 6.1 million for LMICs. Finally, the 24 efforts projected the incremental cost of the expanded benefit package, with estimates ranging between US$30 and US$64 billion per year for LICs and US$61 and US$190 billion per year for LMICs. These estimates assume coverage for the entire population with a prioritized set of benefits and low or zero payment at point of service. DCP3 estimates that countries would need to increase priority package financing by 3 percentage points of GDP (3.1 percent in LICs and 2.9 percent in LMICs). However, this is not the same as the cost of achieving UHC. The cost of implementing the prioritized benefit package would be the same as the cost of the prioritized benefit package if all additional health financing is earmarked toward the expansion of the new benefit package for everyone in the population. How realistic is this assumption? The answer partly depends on the source of the additional financing. In the MDG era, most of the increase in health expenditures in LICS and LMICs was externally financed. In the SDG era, the expectation is that most of the projected increase in expenditures will be financed domestically, heightening the governments influence on the specific use of public health expenditures. Is it realistic to assume that governments would channel all the incremental health financing to subsidize the prioritized benefit package? Would the population coverage programs, the private practitioners and the referral hospitals not receive any part of the additional money? Or is it more realistic to assume that – in the absence of significant reforms-- governments would allocate new funds in the same proportions as today? Table 3 estimates the sensitivity of the total cost of implementation to the use of different assumptions regarding the allocation of health expenditure increases. Total health expenditures per capita today are $40 in LICs and $140 in LMICs. LICs and LMICs are currently spending per capita $16 and $49 respectively on the prioritized package. This implies that the allocation of total health expenditures to the prioritized package is currently 40 percent in LICs and 35 percent in LMICs. The UBP is financed through a combination of external grants and domestic revenues; as most external financing is targeted to the UBP, the share of domestic resources is smaller than the share of total resources. There are three possible scenarios regarding the share of additional health expenditure that would be used for the recommended priority package: Scenario 1: The incremental total health expenditure (THE) is earmarked entirely toward the new benefit package; Scenario 2: The incremental THE is allocated in the same proportion as today; Scenario 3: The incremental domestic funding for health is allocated in the same proportion as today. 25 Table 3. Cost of prioritized Benefit Packages: Three scenarios depending on country choice LICS LMICS HPP EUHC HPP EUHC 1. Total annual expenditures in health ($ per capita)1 $40 $40 $140 $140 2. Total annual cost of benefit package ($ per capita)2 $42 $76 $58 $110 2 3. Incremental annual cost of benefit package ($ per capita) $26 $53 $31 $61 2 4. Total annual cost of benefit package ($ billion) $38 $68 $160 $280 billion billion billion billion 5. Incremental annual cost as a percentage of current GDP2 3.1% 6.4% 1.5% 2.9% 6. Current annual spending on benefit package ($ per capita) $16 $23 $27 $49 [Line 2-Line 3] 7. Current spending in BP/Total spending in health (%) (Line 40% 58% 19% 35% 6/line 1) 8. Current domestic revenues in BP/total health expenditures 20% 29% 13% 23% 9. GNI per capita3 $830 $830 $2100 $2100 Required total health expenditures as a percentage of GNI: (assuming different shares of the incremental health expenditures are allocated to the benefit package) Scenario 1: 100% to BP 5% 9% 3% 5% Scenario 2: Same share of THE as today 13% 16% 14% 15% Scenario 3: Same share of domestic financing as today 25% 32% 21% 22% Sources: 1 WHO Global Health Expenditure Data base; 2 Jamison and others 2017;3 World Bank Indicators. Acronyms: LICs: Low-income countries; LMICs: Lower-middle-income countries; HPP: Highest priority package; EUHC: Essential Universal Health Coverage package BP: Recommended benefit package: DCP3 recommends focus on HPP for LICs and EUHC for LMICs in recognition of their different income levels. Scenarios: Scenario 1 assumes all incremental health expenditures are injected into the BP; Scenario 2 assumes the BP would receive the same share of any incremental health expenditures as the average today; scenario 3 assumes that LICs (LMICs) finance half (two thirds) of the total health expenditures allocated to the BP from domestic resources. The level of health expenditures required to cover the whole population with the recommended benefit package varies greatly depending on the scenario: in scenario 1 countries would need to spend 5 percent of GNI on health care; in scenario 2 the requirement increases to 13-15 percent of GNI; in scenario 3 it reaches 22-25 percent of GNI (higher than any country in the world!). We focus on HPP for LICs and on EUHC for LMICs in the table following DCP3’s suggestion that LICs in the short term should aim for a narrower benefit package than LMICs. The conclusion from this exercise is not that UHC implementation is beyond the reach of LICs and LMICs. The conclusion is that to advance toward UHC, major reforms will be needed to make sure that a larger share of the incremental health expenditures are allocated to the prioritized benefit package than what countries are allocating today. To understand the difficulty of these reforms, it is important to remember that 60-65 percent of total health expenditures is spent outside of the UBP by powerful stakeholders, each with their own priorities. The non-UBP spending includes: 26 central government financing of national referral hospitals; subnational governments expenditure toward regional hospitals; various contributory and non-contributory subpopulation programs financing services for their members; and significant out of pocket payments by users. The implication is that greater focus must be given to planning reforms needed to increase the share of incremental health expenditures used efficiently and equitably. Research and policy dialogue must be directed at understanding how an expanded priority benefit package can be implemented under the existing configuration or through incremental reforms to the existing configuration. Note that it is not obvious what the best intervention would be if we recognize that much of the fragmentation took place during the period when external financing was at its height. An obvious place to start would be to work with the subpopulation coverage programs to help them implement the prioritized benefit packages. For equity reasons, it may be desirable to strengthen the operations of the programs that provide coverage to the poor. For fiscal sustainability, it would be important to work with the contributory programs to help them become more efficient and less dependent on tax subsidies. Summary and Conclusions This study describes health financing reforms in African countries, with an emphasis on user-fee policy reforms and decentralization. Countries in the region have implemented significant health financing reforms in recent decades. During the 1980s and 1990s, most countries sought to establish user fees for publicly provided health care. After the turn of the millennium, there was a significant change in sentiment, and most African countries began to reduce user fees charged by public providers. This study produced systematic data on health financing policies covering 46 countries in Sub-Saharan Africa and in North Africa, and used the data to identify the prevailing strategies followed in the region to reduce user fees. The study found that many countries followed three approaches to overcome financial access barriers and to improve responsiveness: implementation of a narrow universal basic package of services that are available to all with no user fees; establishment of subpopulation coverage programs developed to finance access to health services with zero or low user fees beyond the basic package; and decentralization of health care financing to subnational governments (82 percent of countries in the region are decentralizing health care financing). All 46 countries have established a narrow basic package of prioritized interventions for which no user fees are charged, prioritizing communicable diseases and maternal and child care. Currently there is no general trend to expand the universal basic package beyond the existing interventions, through many countries have added specific benefits to their basic package. Some countries have added all primary care consultations, others inpatient care, while others have included other benefits. Nevertheless, most countries are actively trying to expand benefits beyond those included in the universal basic package. The instrument favored in most African countries to achieve this consists of the implementation of subpopulation coverage programs. There are two types of subpopulation 27 programs from a financing standpoint. Contributory programs, have been created for the higher- income subpopulations in two-thirds of African countries. These programs are generally established in the formal sector, where payroll deductions can easily be used to operationalize mandatory participation. Voluntary contributory programs in the informal sector exist in over 40 percent of the countries. Noncontributory (fully subsidized coverage) programs have also been created in 80 percent of African countries; designed to cover the poor and other vulnerable populations. The paper concludes with a discussion illustrating the importance of recognizing that further steps towards achieving UHC in Africa will have as point of departure a fragmented health system and not a blank slate. Progress towards UHC will require not only more money, but also important policy reform. It is not obvious what policy reforms are feasible and effective to minimize the cost of implementing the prioritized benefit packages; this should become a high priority of work for researchers and policy makers. 28 Annex 1 Acknowledgments Country Contact Person Humberto Albino Cossa, Joao Carlos Duarte Pacheco Blasques de Oliveira, and Carmen Angola Carpio Benin Ibrahim Magazi and Maud Juquois Botswana Sheila Dutta Burkina Faso Haidara Ousmane Diadie Burundi Moulay Driss Zine Eddine El Idrissi, Alain-Desire Karibwami, and Richard Shugugu Cameroon Paul Jacob Robyn, Dr. Elise Virginie Owono Longang, and Jean Claude Taptue Fotso Central African Paul Jacob Robyn, Gervias Chamberlin Yama, and Jean Claude Taptue Fotso Republic Chad Dr. Hamid Jabar, Aissatou Diack, and Alain-Desire Karibwami Comoros Voahirana Hanitriniala Rajoela Congo Gyorgy Bela Fritsche Cote d’Ivoire Ibrahim Magazi and Professor Samba Mamadou Democratic Alain M Iyeti, Hadia Samaha, and Luc Laviolette Republic of Congo Djibouti Emre Ozaltin Egypt Moustafa Abdalla Ethiopia Roman Tesfaye, Jane Chuma, Netsanet Waleign Workie, and Anne Margreth Bakilana Gabon Moulay Driss Zine Eddine El Idrissi, and Paul Jacob Robyn Gambia Rifat Hasan Ghana Anthony Seddoh, Francisca Ayodeji Akala, and Huihui Wang Guinea Dr Sekou Conde, Ibrahim Magazi, and Christopher H. Herbst Guinea-Bissau Edson Correia Araujo Kenya Jane Chuma Lesotho Omer Ramses-Zang Sidjou, Kanako Yamashita-Allen, and Mikail Dastgir Liberia Rianna L. Mohammed and Munirat Iyabode Ayoka Ogunlayi Madagascar Jumana Qamruddin and Voahirana Hanitriniala Rajoela Malawi Collins Chansa Mali Amadou Diallo and Aissatou Diack Mauritania Mohamed Vadel and Moulay Driss Zine Eddine El Idrissi Mauritius Dominic S. Haazen Morocco Dorothée Chen and Houcine El Akhnif Mozambique Humberto Albino Cossa Niger Djibrilla Karamoko and Abdou Galo Karimou Nigeria Benjamin Loevinsohn and Fatimah Abubakar Mustapha Rwanda Netsanet Walelign Workie Senegal Moussa Dieng, Christophe Lemiere, and Maud Juquois Seychelles Netsanet Walelign Workie Sierra Leone Francisca Ayodeji Akala, Alix Bonargent, and Eleonora Genovese South Africa Paolo Belli and Thulani Clement Matsebula Kukwaj Nyawello, Nicolas Collin Dit De Montesson, Noel Chisaka, and Anne Margreth South Sudan Bakilana Sudan Karima Saleh Swaziland Edit V. Velenyi and Kanako Yamashita-Allen Tanzania Mariam Ally, Huihui Wang, Gayle Martin, and Son Nam Nguyen Togo Haidara Ousmane Diadie Tunisia Dr. Sonia Khayat and Dorothée Chen Uganda Peter Okwero and Jane Chuma Zambia Netsanet Walelign Workie and Collins Chansa Zimbabwe Ronald Upenyu Mutasa and Chenjerai N. Sisimayi 29 Annex 2 Additional Data Table A2.1 Economic and Health Indicators of Countries in the Study GDP Per Capita, PPP Mortality Rate, Total Fertility Country Income Level (constant 2011 Under-5 (per Rate (2014) international $) (2014) 1,000) Angola HMIC 6,956 6.1 157 Benin LIC 1,937 4.8 100 Botswana HMIC 15,200 2.8 44 Burkina Faso LIC 1,546 5.5 89 Burundi LIC 734 5.9 82 Cameroon LMIC 2,836 4.7 88 Central African LIC 544 4.3 130 Republic Chad LIC 2,074 6.2 139 Comoros LIC 1,364 4.5 74 Congo LMIC 5,988 4.9 45 Cote d’Ivoire LMIC 3,108 5.0 93 Democratic Republic LIC 711 6.0 98 of Congo Djibouti LMIC 3,120 3.2 65 Egypt LMIC 10,049 3.3 24 Ethiopia LIC 1,431 4.4 59 Gabon HMIC 18,537 3.9 51 Gambia LIC 1,556 5.7 69 Ghana LMIC 3,894 4.2 62 Guinea LIC 1,165 5.0 94 Guinea-Bissau LIC 1,336 4.8 93 Kenya LMIC 2,819 4.3 49 Lesotho LMIC 2,517 3.2 90 Liberia LIC 804 4.7 70 Madagascar LIC 1,371 4.4 50 Malawi LIC 1,115 5.1 64 Mali LIC 2,188 6.2 115 Mauritania LMIC 3,694 4.6 85 Mauritius HMIC 17,737 1.4 14 Morocco LMIC 7,146 2.5 28 Mozambique LIC 1,080 5.4 79 Niger LIC 902 7.6 96 Nigeria LMIC 5,639 5.7 109 Rwanda LIC 1,585 3.9 42 Senegal LMIC 2,215 5.1 47 Seychelles HMIC 25,207 2.3 14 Sierra Leone LIC 1,919 4.6 120 South Africa HMIC 12,436 2.4 41 South Sudan LIC 1,926 5.0 93 Sudan LMIC 3,882 4.4 70 Swaziland LMIC 7,911 3.3 61 Tanzania LIC 2,421 5.1 49 Togo LIC 1,337 4.6 78 Tunisia HMIC 10,749 2.2 14 Uganda LIC 1,689 5.8 55 Zambia LMIC 3,710 5.4 64 Zimbabwe LIC 1,709 3.9 71 Source: World Development Indicators. Note: PPP = purchasing power parity. 30 Figure A2.1 Share of Development Assistance for Health (DAH) in Total Health Expenditure (THE) Source: World Bank 2016. 31 Notes 1 WHO 2010. 2 http://www.who.int/healthsystems/universal_health_coverage/en/. 3 The Abuja Declarations and Frameworks for Action on Roll Back Malaria was a pledge made in 2001 by members of the African Union during a conference in Abuja, Nigeria. In it, the member nations pledged to increase their health budget to at least 15 percent of the state’s annual budget, and requested Western donor countries to increase their support. 4 World Bank; World Health Organization; JICA 2016. 5 WHO 2010. 6 WHO 2016. 7 Yates 2010. 8 Gilson 1997. 9 Robert and 2013. 10 Druetz et al. 2015. 11 Leone et al. 2016. 12 Leone et al. 2016. 13 Deininger and Mpuga 2004. 14 Audibert and Mathonnat 2000. 15 McKinnon, Harper, and Kaufman 2015. 16 Philibert et al. (2014) find that the elimination of user fees had no negative impact on patient’s perception of the quality of health services in Burkina Faso. 17 Tanaka (2014) offers evidence on the improvement in weight-for-age Z-scores in South African children after the elimination of user fees. 18 Boyer et al. (2010) observe a positive effect of user fee removal on antiretroviral therapy adherence and improved clinical results in Cameroon. 19 Ben Ameur et al. 2012. 20 Evans and Pablos-Méndez 2016. 21 Lagomarsino et al. 2012. 22 Tetteh 2012. 23 Mathauer, Doetinchem et al. 2011. 24 Mathauer, Musango et al. 2011. 25 Mcintyre, Doherty, and Gilson 2003. 26 Carrin et al. 2007. 27 Carrin et al. 2007. 28 Onoka et al. 2013. 29 Onoka et al. 2013. 30 Mathauer, Schmidt, and Wenyaa 2008. 31 Kotoh and Van der Geest 2016. 32 Hsiao et al. 2006. 33 Carrin and James 2004. 34 Mladovsky and Mossialos 2008. 35 Ekman 2004. 36 Bennett 2004. 37 Mebratie et al. 2013. 32 38 Cotlear et al. 2015. 39 “The Bamako Initiative was a formal statement adopted by African health ministers in 1987 in Bamako, Mali, to implement strategies designed to increase the availability of essential drugs and other healthcare services for Sub-Saharan Africans) 40 WHO 2003 41 McIntire and others2017 42 Jamison et al. 2013. 43 Jamison et al. 2017. 44 WHO 2017. 33 References Audibert, M., and J. Mathonnat. 2000. “Cost recovery in Mauritania: Initial lessons.” Health Policy and Planning 15 (1): 66–75. doi:10.1093/heapol/15.1.66. Ben Ameur, A., V. Ridde, A. R. Bado, M. Ingabire, and L. Queuille. 2012. “User fee exemptions and excessive household spending for normal delivery in Burkina Faso: The need for careful implementation.” BMC Health Services Research 12 (1): 412. doi:10.1186/1472- 6963-12-412. Bennett, S. 2004. “The role of community-based health insurance within the health care financing system: a framework for analysis.” Health Policy and Planning 19 (3): 147–158. Boyer, S., F. Eboko, M. Camara, C. Abé, M. E. O. Nguini, S. Koulla-Shiro, and J. P. Moatti. 2010. “Scaling up access to antiretroviral treatment for HIV infection: the impact of decentralization of healthcare delivery in Cameroon.” Aids 24 (Suppl. 1): S5–S15. Carrin, G., C. James, and World Health Organization. 2004. Reaching universal coverage via social health insurance: key design features in the transition period. Geneva: World Health Organization. Carrin, G., C. James, M. Adelhardt, O. Doetinchem, P. Eriki, et al. 2007. “Health financing reform in Kenya: Assessing the social health insurance proposal.” South African Medical Journal 97 (2): 130–135. Cotlear, Daniel, Somil Nagpal, Owen Smith, Ajay Tandon, and Rafael Cortez. 2015. Going Universal: How 24 Developing Countries are Implementing Universal Health Coverage Reforms from the Bottom Up. Washington, DC: World Bank. doi:10.1596/978-1-4648- 0610-0. Deininger, K., and P. Mpuga. 2004. Economic and welfare effects of the abolition of health user fees: Evidence from Uganda. Washington, DC: World Bank. doi:10.1596/1813-9450- 3276. Druetz, T., F. Fregonese, A. Bado, T. Millogo, S. Kouanda, S. Diabaté, and S. Haddad. 2015. “Abolishing Fees at Health Centers in the Context of Community Case Management of Malaria: What Effects on Treatment-Seeking Practices for Febrile Children in Rural Burkina Faso?. PLoS ONE 10 (10): e0141306. Ekman, B. 2004. “Community-based health insurance in low-income countries: a systematic review of the evidence.” Health Policy and Planning 19 (5): 249–270. Evans, T., and A. Pablos-Méndez. 2016. “Shaping of a new era for health financing.” Lancet 387 (10037): 2482–2884. Gawande, Atul. 2011. The Checklist Manifesto: How to Get Things Right. New York: Macmillan. Gilson, L. 1997. “The lessons of user fee experience in Africa.” Health Policy and Planning 12 (4): 273–285. 34 Hsiao, W., R. Shaw, A. Fraker, P. Hanvoravongchai, M. Jowett, D. Pinto, and S. Ramachandra. 2006. “Social Health Insurance for Developing Nations.” World Bank Institute Development Studies, World Bank, Washington, DC. Jamison, D.T., Summers, L.H., Alleyne, G., Arrow, K.J., Berkley, S., Binagwaho, A., Bustreo, F., Evans, D., Feachem, R.G., Frenk, J. and Ghosh, G., 2013. Global health 2035: a world converging within a generation. The Lancet, 382(9908), pp.1898-1955. Jamison, D.T., Alwan, A., Mock, C.N., Nugent, R., Watkins, D., Adeyi, O., Anand, S., Atun, R., Bertozzi, S., Bhutta, Z. and Binagwaho, A., 2017. Universal health coverage and intersectoral action for health: key messages from Disease Control Priorities. The Lancet. Kotoh, A. M., and S. Van der Geest. 2016. “Why are the poor less covered in Ghana’s national health insurance? A critical analysis of policy and practice.” International Journal for Equity in Health 15 (1): 1. Lagomarsino, G., A. Garabrant, A. Adyas, R. Muga, and N. Otoo. 2012. “Moving towards universal health coverage: health insurance reforms in nine developing countries in Africa and Asia.” Lancet 380 (9845): 933–943. Laokri, S., M. K. Drabo, O. Weil, B. Kafando, S. M. Dembélé, and B. Dujardin. 2013. “Patients are paying too much for tuberculosis: A direct cost-burden evaluation in Burkina Faso.” PLoS One 8 (2): e56752. doi:10.1371/journal.pone.0056752. Leone, T., V. Cetorelli, S. Neal, and Z. Matthews. 2016. “Financial accessibility and user fee reforms for maternal healthcare in five sub-saharan countries: A quasi-experimental analysis.” BMJ Open 6 (1): e009692. Mathauer, I., J. O. Schmidt, and M. Wenyaa. 2008. “Extending social health insurance to the informal sector in Kenya: an assessment of factors affecting demand.” International Journal of Health Planning and Management 23: 51–68. Mathauer, I., L. Musango, S. Sibandze, K. Mthethwa, G. Carrin. 2011. “Is universal coverage via social health insurance financially feasible in Swaziland? South African Medical Journal 101 (3): 179–183. Mathauer, I., O. Doetinchem, J. Kirigia, and G. Carrin. 2011. “Reaching universal coverage by means of social health insurance in Lesotho? Results and implications from a financial feasibility assessment.” International Social Security Review 64 (2): 45–63. McIntyre, D., J. Doherty, and L. Gilson. 2003. “A tale of two visions: the changing fortunes of social health insurance in South Africa.” Health Policy and Planning 18 (1): 47–58. Mcintyre, D., Meheus, F. and Røttingen, J.A., 2017. What level of domestic government health expenditure should we aspire to for universal health coverage? Health Economics, Policy and Law, 12(2), pp.125-137. McKinnon, B., S. Harper, and J. S. Kaufman. 2015. “Who benefits from removing user fees for facility-based delivery services? Evidence on socioeconomic differences from Ghana, 35 Senegal and Sierra Leone.” Social Science & Medicine 1982: 135: 117–123. doi:10.1016/j.socscimed.2015.05.003. Mebratie, A., R. Sparrow, G. Alemu, and A. S. Bedi. 2013. “Community-Based Health Insurance Schemes.” ISS Working Paper Series/General Series 568 (568): 1–47. Mladovsky, P., and E. Mossialos. 2008. “A conceptual framework for community-based health insurance in low-income countries: social capital and economic development.” World Development 36 (4): 590–607. Onoka, C., O. Onwujekwe, B. S. Uzochukwu, and N. Ezumah. 2013. “Promoting universal financial protection: constraints and enabling factors in scaling-up coverage with social health insurance in Nigeria.” Health Research Policy and Systems 11 (20). Philibert, A., V. Ridde, A. Bado, and P. Fournier. 2014. “No effect of user fee exemption on perceived quality of delivery care in Burkina Faso: a case-control study.” BMC Health Services Research 14 (1): 1. Robert, E., and V. Ridde. 2013. “Global health actors no longer in favor of user fees: A documentary study.” Globalization and Health 9 (1): 29. doi:10.1186/1744-8603-9-29. Stenberg, K., Hanssen, O., Edejer, T.T.T., Bertram, M., Brindley, C., Meshreky, A., Rosen, J.E., Stover, J., Verboom, P., Sanders, R. and Soucat, A., 2017. Financing transformative health systems towards achievement of the health Sustainable Development Goals: a model for projected resource needs in 67 low-income and middle-income countries. The Lancet Global Health, 5(9), pp.e875-e887. Tanaka, S. 2014. “Does Abolishing User Fees Lead to Improved Health Status? Evidence from Post-Apartheid South Africa.” American Economic Journal: Economic Policy 6 (3): 282– 312. Tetteh, E. K. 2012. “Responding to the challenges of social health insurance in African countries.” Development Southern Africa 29 (5): 657–680. WHO (World Health Organization). 2003. Investing in health: A summary of the findings of the Commission on Macroeconomics and Health. Geneva: World Health Organization. ———. 2010. The World Health Report – Health Systems Financing: The Path to Universal Coverage. Geneva: World Health Organization. http://www.who.int/whr/en/index.html. ———. 2016. Public Financing for Health in Africa: from Abuja to the SDGs. Geneva: World Health Organization. World Bank. 2016. UHC in Africa: A Framework for Action. Washington, DC: World Bank 2016. World Bank; World Health Organization; JICA; The Global Fund to Fight AIDS, Tuberculosis and Malaria; and the African Development Bank. 2016. UHC in Africa: A framework for Action. Washington DC: World Bank. Yates, R. 2010. “Women and children first: an appropriate first step towards universal coverage.” Bulletin of the World Health Organization 88 (6): 474–475. 36 37 38