Health systems Free

There is an avalanche of interest in health systems with countries, donors, international experts and academic institutions all rushing to promote, fund or build capacity to address this wave of interest. As Anne Mills documents in an article in this volume, there is also a growing literature on how to think about health systems, how to use research on health systems to improve their performance, and pleas for greater investment in knowledge about health systems. We are probably riding on the crest of this new wave and it will be important to make use of it to further our knowledge about how to achieve health system effectiveness. With continuing global economic uncertainty, this wave may pass if we do not take advantage of it now.

It is useful to reflect on why there is such interest in this theme now. There are probably many converging causes but the most important, to my mind, is a replay of an old debate between vertical programmes and horizontal, integrated approaches that has experienced swings of interest at least since the Alma Ata Conference in 1978 emphasized the integrated primary health care approach. During the last decade or more, there has been a major focus on vertical disease-specific programmes supported by greatly increased funding from the Global Fund to Fight AIDS, Tuberculosis and Malaria, Global Alliance for Vaccines and Immunisation (GAVI), US President’s Emergency Plan for AIDS Relief (PEPFAR), Bill and Melinda Gates Foundation, among other donors. The major claim of these programmes was that by focusing on clear disease-specific objectives they would be more effective in shorter time periods and they could be held more accountable to donors (since the results could be more measurable) than the difficult and diverse efforts to improve the many elements of a health system through the horizontal integrated approach. While there were evident successes in these programmes, there was also a growing recognition that health system constraints—especially in human resources and logistics systems—were becoming major obstacles to achieving even their limited objectives of increased coverage of HIV/AIDS, tuberculosis, malaria and immunizations.

With the added emphasis on the Millennium Development Goals (MDGs) and the slow progress toward their achievement in many countries, the international community and national governments have turned increasingly toward attempting to reduce the health system constraints—an approach that is closer to the integrated horizontal approach. Both GAVI and the Global Fund have added a separate fund for health systems interventions. In 2008, the G8 called for greater attention to health system strengthening as compatible with the prior focus on vertical programmes (Reich et al. 2008). In 2009, the World Bank convened a seminar on health system strengthening attended by representatives from the World Health Organization (WHO), GAVI and Global Fund. A task force on Healthy Women and Healthy Children calling for a health system funding platform was initiated at the UN General Assembly. In 2010, the WHO and others sponsored the First Global Symposium on Health Systems Research in Montreux, Switzerland. While many observers are trying to argue that the two approaches are now more compatible, at least in rhetorical terms it seems that the pendulum is shifting toward greater attention to health system strengthening.

So how do we, the readers of Health Policy and Planning, take advantage of this interest? Health Policy and Planning is dedicated to the publication of high quality research that, as the name implies, improves decisions on policy and planning. This poses for us a central question: how can we do research that will improve health systems? The answer is complex for it requires attention to the kind of knowledge that is both researchable and leads to practical, implementable recommendations.

The place to start is with a framework of analysis that is both analytical, in that it provides empirically based categories about health systems, and purposive, in that it is oriented toward how to improve the system. Many of the frameworks that are available provide a recognizable set of functional issues whose boundaries help focus attention on specific interactions that can be studied, such as financing, service delivery, human resources, governance (Mills et al. 2001). The WHO Building Blocks framework is a very popular approach that is guiding much of the discussion and research in health systems (WHO 2007). Another approach developed by Harvard School of Public Health faculty is oriented toward using analytical categories in a purposive manner that starts with an ethically derived vision of what we would like to achieve, and works back toward finding the reasons why we have not achieved those objectives. It then focuses our attention on what we can change using evidence on the effectiveness of changing different ‘control knobs’ or policy levers, including financing, payments, organization, regulation and persuasion (Roberts et al. 2004). There is a great deal of overlap in these models and they are compatible in many ways that can guide research toward practical recommendations for policy.

A second question is about levels of analysis. Should we attempt to research whole national systems, with typologies of systems and comparisons of the achievements of the different models? We often hear this orientation used in policy arguments such as the recent arguments over whether the USA should adopt the Canadian single payer model. However, this approach suffers from methodological challenges such as the ‘small number’ problem (which limits the ability to use sophisticated quantitative analysis due to the few comparable national systems) and the many different characteristics and historical trajectories that confound the analysis. Few countries—Chile, China, Colombia, Ghana, Taiwan, Turkey and Eastern European countries—have even attempted broad multifaceted health system reforms and they have tended to develop their own reform models rather than copying others. Nevertheless, there are many country studies of health reforms that provide evidence and much can be gained by detailed evaluation of the effectiveness of these models that can inform policy choice in different contexts.

An alternative means of assessing health systems is to focus on sub-systems or analysis of specific policy changes, focusing on different social health insurance approaches, pay for performance, contracting, decentralization or other reforms, and assessing their performance in terms of achieving health system objectives like the MDGs, reduction in catastrophic expenditures caused by paying for illness, or broader responsiveness and accountability to citizens. The advantage of this approach is that, if the research is well designed, the causality of the impact of the changes on health system objectives can be more clearly demonstrated, and it allows a more focused attention to policy changes that different countries can adopt in an incremental way. This approach can address both systems issues and vertical programme needs. For instance, the work that I am involved in on decentralization—looking at the relationship between choice, capacity and accountability at different levels of a health system—can be used to make policy recommendations on general health system reform as well as specific reforms of logistic systems, vertical immunization or family planning programme effectiveness (Bossert et al. 2007).

The challenge in this approach is to account for the general national context factors—political, economic, social and cultural—which may influence the performance of these initiatives. What works in Chile may not work in China due to many different characteristics of the country contexts. This is a major lacuna in the current research in health systems often covered over by focusing only on the differences in average income levels, assuming for instance that health system changes in low-income countries are all relatively similar. However, governance issues, cultural and ethical differences, past histories with different health systems, all play a role in constraining the lessons learned from one country to another. The recent concern with political economy and political institutionalism may inform future work in this area (Fox and Reich, forthcoming).

In addition, focus on one subsystem or one policy lever is likely to miss the interactions among different subsystems changes that synergistically influence outcomes. For instance, changes in decentralization may be enhanced by changes in social insurance programmes, or the creation of a new cadre of health workers. Focusing on one in order to evaluate its separate impact will miss the role of the combined changes on achieving objectives. However, if we are aware of these limitations, it might be possible to move collectively toward setting a research agenda in which multidisciplinary researchers assess the different subsystems in ways that inform each other. Those of us who focus on organizational issues like decentralization and human resources need to be aware of and incorporate in our studies, the work of experts in economics, ethics, politics and other behavioural sciences, as well as epidemiologists and biostatisticians who help us understand the underlying population and social processes underpinning those systems.

There is also an imbalance of research on functional areas of analysis with a much greater body of knowledge on financing and payment issues than on organizational issues and governance and regulation. In recent years, there has been greater interest in governance, politics and human resources issues, but these areas are so complex that they require much greater attention and more funded research than is currently in the pipeline. This poses an additional requirement for multidisciplinary studies that address these under-researched areas to bring the level of knowledge up to that of the more economic themes.

This is a huge task of course and current funders have little patience for large research projects in health systems. Indeed there is considerable pressure for evaluating the performance of programmes and projects that have been funded in a very short and unrealistic time frame. It suggests that we need to be nimble and develop cadres of multidisciplinary experts to be able to quickly do the solid research necessary but within limited funds and short time horizons to provide well-documented policy recommendations. If we do not, we are likely to have missed the opportunities of this wave of interest as funding for health systems declines due to alternative priorities, donor fatigue or general global economic restrictions.

Funding

None declared.

References