Research to support universal coverage reforms in Africa: the SHIELD project

With the release of the 2010 World Health Report (World Health Organization 2010), universal coverage has been placed high on the global health policy agenda. We understand universal coverage to involve two key elements: first, ensuring financial protection for all from the costs of health care; and second, enabling access to needed health care for all citizens. Pre-payment financing mechanisms (such as tax and various forms of health insurance), rather than direct out-of-pocket payments by individuals, are required to ensure financial protection. Indeed, most of the attention in recent health policy debates has been on the alternative ways of generating financial resources for health care, and to some extent on the pooling of these financial resources, to provide financial protection. Considerably less attention has been paid to the second element of universal coverage—how to improve access to, and the use of, quality health services. To date, relatively limited progress towards universal coverage has been made in African countries. There is generally a heavy reliance on out-of-pocket payments in Africa, with some countries (such as Guinea, Nigeria and Cameroon) funding 70% or more of total health care expenditure through this means (WHO National Health Accounts database). A number of African countries (such as Uganda, Zambia and South Africa) have implemented initiatives to reduce out-of-pocket payments, particularly through removing some or all user fees at public sector facilities, but these policies have generally not been accompanied by parallel initiatives to increase domestic pre-payment funding of health services. The two African countries that are commonly regarded as having taken the most dramatic steps towards providing universal financial protection through pursuing increased domestic pre-payment financing mechanisms are Ghana and Rwanda. Where progress has been made towards universal financial protection in Africa, the emphasis has been on introducing or expanding insurance schemes. In very few cases have there also been efforts to increase tax funding of health care. A key challenge that plagues African countries seeking to pursue increased financial protection through contributory insurance schemes is how to ensure that those outside the formal sector are included under these arrangements. This points to potentially the most important single area of current and future debate on achieving universal financial protection in the African context, and hence needed evidence from research; is it most equitable and efficient to seek to cover those outside the formal sector through contributory insurance schemes (with subsidies for the poorest) or through tax funding, some of which may be generated through dedicated taxes that seek to secure revenue from those outside the formal sector? Another area that requires urgent attention is how best to promote access to needed health care on a universal basis. A resolution on universal coverage adopted at the 2005 World Health Assembly highlighted the importance of research to create an evidence base for health care financing reforms, and called for the development of ‘methodologies better to measure and analyse the benefits and cost of different practices in health financing, covering collection of revenues, pooling and provision or purchasing of services, taking account of economic and sociocultural differences’ (World Health Organization 2005). The Social Health Insurance for Equity in Less Developed countries (SHIELD) project sought to respond to this call, by critically evaluating the health systems of three African countries (Ghana, South Africa and Tanzania) through an equity lens, and the extent to which expanded or new financing mechanisms could address the equity challenges faced by these countries. It adopted a system-wide perspective, considering both the public and the private health sectors, and evaluating financing as well as service use elements of these health systems. This supplement presents the key findings of the research. The first paper (Mills et al.) provides a brief overview of the three countries' health systems and highlights some of the major equity challenges facing them. It outlines the context within which the SHIELD study was undertaken. The next three papers (Akazili et al.; Mtei et al.; Ataguba and McIntyre) focus on a core objective of the SHIELD project; a detailed analysis of the distribution of the current health care financing burden and of health care benefits across socio-economic groups in Ghana, Tanzania and South Africa, respectively. These papers shed light on the relative progressivity of alternative ways of generating funds for health services, including out-of-pocket payments and different forms of tax revenue in all three countries, a dedicated health tax in Ghana (the National Health Insurance levy which is an additional 2.5% on value added tax), private voluntary health insurance contributions in South Africa, and mandatory health insurance contributions for formal sector workers and district-based health insurance contributions by those outside the formal sector in Ghana and Tanzania. These papers present the first system-wide findings on health care financing progressivity in Africa and will undoubtedly contribute to the debate on how to fund universal coverage, particularly in the context of large informal sectors in African countries. The three papers also evaluate the distribution of benefits from using health services across different groups. While benefit incidence analyses have previously been undertaken in a number of African countries, these earlier studies have focused only on the distribution of public subsidies for health care. In contrast, the SHIELD project evaluated the distribution of benefits from using both public and private health services (McIntyre and Ataguba 2010). In addition, previous studies implicitly assume that the need for health care is the same across different socio-economic groups (e.g. that the appropriate distribution of benefits would be a 20% share for each quintile). However, given that equity in service use is most frequently defined as utilization being in line with need, these papers compare the distribution of service benefits with the distribution of a measure of need for health care. Combined, the financing and benefit incidence analyses provide insights into the extent of income and risk cross-subsidies within the health systems of the three countries. Macha et al. then explore in depth, through both quantitative and qualitative methods, the factors affecting the distribution of the burden of financing and the benefits of care. This paper particularly focuses on factors influencing out-of-pocket payments and membership of voluntary insurance schemes (whether by formal sector workers as in South Africa or by those outside the formal sector as in Ghana and Tanzania). This provides insights into factors impeding or facilitating progress towards universal financial protection in the three countries, particularly in relation to the effectiveness of exemption policies and efforts to provide such protection to those outside the formal sector. In addition, it provides a detailed analysis of the community and health system factors that influence the distribution of benefits from using health services and, thus, addresses the much neglected issue of how to promote universal access to needed care. Thereafter, Goudge et al. review evidence on the willingness of citizens in the three countries to support income and risk cross-subsidies in the health system. Universal health systems are built on social solidarity. An important factor influencing the ability of a country to achieve universal coverage is the extent of social solidarity and citizens' understanding of and willingness to support health system cross-subsidies. Any intended health system reform needs to take account of the views of its citizens. Another key objective of the SHIELD project was to identify and critically evaluate options for the likely future development of financing mechanisms in the study countries, particularly in relation to their financial sustainability, potential equity impact and their feasibility given attitudes of key stakeholders. The final four papers address this objective. Two of these papers are more methodological in nature and aim to illustrate in some detail the process we used for undertaking stakeholder analyses in South Africa and Tanzania (such policy analyses had been undertaken previously in Ghana and were, therefore, not required) and spreadsheet modelling of policy options. We believe that it is critical to explore the implications of alternative paths of health financing reform through modelling before final policy decisions on the most appropriate path are taken. We also believe that technical analyses are not by themselves adequate. They should be complemented by policy analyses, particularly to understand the views of key stakeholders about alternative health financing reform paths and how these may influence the feasibility of pursuing such reforms. Thus, the Gilson et al. paper illustrates how stakeholder analysis of health care financing reforms can be undertaken and also presents some findings on stakeholder views on the reform options currently being considered in South Africa and Tanzania. Thereafter, McIntyre and Borghi provide guidance on how to draw on the stakeholder analyses to inform the policy options to be evaluated, and how to address the modelling challenges likely to be faced by researchers in data scarce contexts. The final two papers by Borghi et al. and McIntyre and Ataguba present the findings of spreadsheet models, which consider the resource requirements and potential funding sources of a universal coverage option and compare this with the status quo and with only partially expanding coverage. These papers focus on the Tanzanian and South African experiences, respectively. As Ghana is already firmly set on the path of moving to universal coverage through a National Health Insurance (NHI), a similar exercise was not appropriate and more focused modelling of policy options in that country is planned. The South African paper also considers the impact on financing and benefit incidence of the alternative policy options. While there remains a massive research agenda to inform universal coverage reforms in Africa, the findings of the SHIELD project presented in this special issue not only provide evidence of value to current reform debates, but also illustrate the range of technical and policy analyses that should be undertaken to comprehensively assess equity in existing health systems and appropriate reforms to promote universal coverage. This research was funded by International Development and Research Centre (Grant number 103457) and the European Commission (Sixth Framework Programme; Specific Targeted Research Project no: 32289). DM is supported by the South African Research Chairs Initiative of the Department of Science and Technology and National Research Foundation. The authors declare having no conflict of interest.