BioMed Central Page 1 of 6 (page number not for citation purposes) Health Research Policy and Open Access Commentary Comparative health systems research in a context of HIV/AIDS: lessons from a multi-country study in South Africa, Tanzania and Zambia Suraya Dawad and Nina Veenstra* Address: Health Economics and HIV/AIDS Research Division, University of KwaZulu-Natal, Durban, South Africa Email: Suraya Dawad - dawads1@ukzn.ac.za Nina Veenstra* - veenstran@ukzn.ac.za * Corresponding author Abstract Comparative, multi-country research has been underutilised as a means to inform health system development. South-south collaboration has been particularly poor, even though there have been clearly identified benefits of such endeavours. This commentary argues that in a context of HIV/ AIDS, the need for regional learning has become even greater. This is because of the regional nature of the problem and the unique challenges that it creates for health systems. We draw on the experience of doing comparative research in South Africa, Tanzania and Zambia, to demonstrate that it can be useful for determining preconditions for the success of health care reforms, for affirming common issues faced by countries in the region, and for developing research capacity. Furthermore, these benefits can be derived by all countries participating in such research, irrespective of differences in capacity or socio-economic development. Introduction HIV/AIDS is having a major impact on health systems in sub-Saharan Africa (sSA), most of which have in the last three decades also undergone a string of health care reforms aimed at improving efficiency, effectiveness and equity. Reforms have meant substantial change in both what is done and how it is done, in other words, change in both policies and institutions [1]. They have been com- plicated by the additional layer of complexity which HIV/ AIDS has brought. This dual challenge of health care reform and HIV/AIDS faced by countries in the region suggests scope for joint learning. More generally speaking, there have also been calls for more comparative, multi- country research that will contribute to health system development [see for example [2]]. A recent review of health systems research publications demonstrated that only 10% of such papers make reference to multiple coun- tries [3]. The review concluded that south-south collabo- ration in health systems analysis is particularly poor. This commentary advocates for comparative, multi-coun- try research to assist countries in sSA in developing their health systems in a context of HIV/AIDS. It does this in two ways. Firstly, it examines the use of comparative, multi-country health systems research. It considers the insights we might obtain from such research and, in par- ticular, the rationale for joint learning to manage the impacts of HIV/AIDS. Secondly, this paper elaborates on the lessons learned from our experience of conducting comparative research in South Africa, Tanzania and Zam- bia. This research was designed to elicit challenges in the functioning and development of health systems in rela- tion to HIV/AIDS. Although only a pilot study, the Published: 30 October 2007 Health Research Policy and Systems 2007, 5:13 doi:10.1186/1478-4505-5-13 Received: 21 May 2007 Accepted: 30 October 2007 This article is available from: http://www.health-policy-systems.com/content/5/1/13 �� 2007 Dawad and Veenstra licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Health Research Policy and Systems 2007, 5:13 http://www.health-policy-systems.com/content/5/1/13 Page 2 of 6 (page number not for citation purposes) research revealed some unexpected benefits through its focus on these three countries. Why comparative research? 'Comparative health systems research' could be defined a number of ways and so it is necessary to clarify the defini- tion employed for the purposes of our research and now for that of this commentary. Firstly, 'comparative' implies a comparison. Interestingly, past research indicates com- parisons have not always been across countries ��� in some instances policy and programme analysis has looked at whether lessons are transferable from the education sector to the health sector, for example [4]. While such approaches are not without merit, this commentary focuses on more commonly encountered endeavours to develop shared learning across countries in a particular region. Secondly, by 'health systems research' we are look- ing specifically at research on the functioning of health management structures and processes which have been the subject of health system reforms. This is in contrast to research which concentrates on health facilities and their role in service delivery. Using this definition, there is some experience to go by, since multi-country, comparative health systems research has been used to investigate a diverse range of health sys- tem issues in sSA including decentralisation, community financing and regulation of private sector activity, to name just a few [see [5-7]]. Other studies have looked at a broader range of health system characteristics in various contexts to see how these influence clinical outcomes [see for example [8] in relation to maternal health outcomes]. The question that we have to answer, before turning more specifically to the challenge of HIV/AIDS, concerns the contribution of comparative research to our insights on health system functioning. In other words, what is the rationale for doing health systems research across more than one country, given the potentially difficult logistics and much greater expense? In most cases, the rationale for choosing to study a number of countries in the region is that the variation in context allows one to better understand the necessary pre- conditions for the success of national policies. In the case of community financing, for example, contextual factors, the design of different schemes, processes used in imple- mentation, or the mix of actors involved, may all explain the pattern of equity impacts observed [6]. Other studies have pointed out how the aims and types of health care reforms in sSA have often been similar, but that these are implemented in different contexts and using different approaches, providing unique opportunities to under- stand factors influencing the success of reforms [9]. Such reasoning concurs with a 'realist' approach to evaluation which replaces the question of 'what works?' with the question of 'what is it about this programme or policy that works for whom in what circumstances?' [10]. This approach therefore focuses on the mechanisms through which a programme, policy, or reform strategy might work in relation to associated contexts. While the 'realist' approach to evaluation constitutes the most common rationale for doing comparative research, there are other important benefits that can be derived. In particular, comparative research has been identified as a method to build cross-country research capacity [3]. Examples have been documented of research networks whose priority it has been to strengthen national research by combining and sharing knowledge and experience [11]. This might seem like a less important objective of comparative research, but without adequate research capacity, policy makers will not have access to sound information on which to base decisions and the potential for shared learning will be lost. Researchers stand to ben- efit from such research by developing their skills around elements such as research design, implementation and analysis. For this to occur, however, research has to be conducted by researchers within their respective countries [3]. Given these benefits of comparative research, we argue that this approach is ideally suited to examining the inter- face between HIV/AIDS and health system functioning. This is because HIV/AIDS is arguably the most challenging contextual factor that health systems are currently facing. The epidemic increases the demand for health care, while impeding household's ability to access care and worsen- ing the human resource crisis. These stresses may be felt higher up in the system through various means including the establishment of new structures, the centralisation of management, more vertically structured programmes and an increase in donor activity. We hypothesize that how a health system copes (or indeed doesn't cope) with such stresses depends on a number of health system factors operating in tandem with other contextual factors. For example, the structure of the health system and the nature and extent of health reforms undertaken may all affect the ability of the health system to respond appropriately. Socio-economic factors, on the other hand, can impose additional restrictions on health systems, not least of all through limiting the amount of resources allocated to them. Additional motivation is provided for conducting com- parative research if we consider how little experience we have in dealing with HIV/AIDS. The pandemic is a threat to health systems development unlike any other previ- ously experienced, partly because of its long wave nature and widespread impacts. Capacity building is urgently needed to facilitate management strategies that are