Comparative health policy analysis: Clusters of writing
Troy Media – by Ted Marmor
Health policy in the OECD world is, at the same time, a matter of insistent national debate, a frequent topic of descriptive, statistical portraiture for international organizations, a sometime subject of publication in the comparative journals, and only very infrequently in its cross-national comparative form, the object of book-length treatment.
For many years, readers had to turn to a 1972 treatment of Swedish, British and American medical care developments in the post-World War II period for acute, well-informed judgments. There were many other individual country studies, but few if any that employed a systematic, comparative method of policy analysis.
In contemporary debates about Dutch health care, for instance, there appears little evidence of detailed understanding about German – or American – policy experience with health care reform in the 1990s.
What is true for medical care applies just as well to other fiscally-important areas of the welfare state. So, for example, American discussions of disability policy in the early 1980s drew very little from Dutch experience, though there were knowledgeable scholars in both countries who sought to have influence.
By the end of the 1980s, political scientists – particularly North American ones – had become interested in comparing relations between the medical profession, as a particular kind of interest group, and the state. Their theoretical focus was by and large on the institutions of government and the different ways in which they shape health care politics. Slowly, the field began to produce genuinely comparative political analyses of substantial industry and competence.
The 10 years and more since then have witnessed a rapid expansion of cross-national health policy literature. The quality of these works varies enormously – whether measured by the standard of intellectual rigor, theoretical perspective, descriptive accuracy or concern for systematic policy learning across borders. There are, roughly speaking, four separable but not mutually exclusive categories of such writing.
The first includes the well-known statistical, largely descriptive documents that provide data on a number of countries assumed to constitute a coherent class. It also includes more specialized surveys that deal with public opinion, health care and health policy. In that way they supply much of the basic information that policy commentators explore.
The OECD Health Data series has become a staple of both academic and more applied analyses alike. These studies typically neither provide behavioral hypotheses nor test explanations for why certain patterns exist. Nor do they, generally speaking, explicitly deal with the promise and pitfalls of cross-border learning.
In a wider sense, the recent efforts to rank systems, countries or institutions by means of benchmarking techniques belong to this group, too. In a much-discussed report, the WHO used its comparative data to rank the performance of national health care systems (WHO 2000).
The second category of comparative studies – by far the largest number – includes collections of international material that we label as “parallel” or “stapled” national case studies. These are usually country reports bound together, accompanied by an editorial introduction and summary conclusion. For the most part, the authors are intent on setting out “how things work” in whichever country they are writing about. They are mostly descriptive, but with some assessment of performance and the flagging of issues prompting political concern. As such, they represent a qualitative correlate of the quantitative statistical studies described above. Done carefully, they are an invaluable resource for cross-national understanding. In many cases, they leave readers to find what is relevant and, as far as policy learning is concerned, leave them to do the work.
Thirdly, there are books about a number of individual countries that employ a common framework of analysis, usually addressing a particular theme in health policy, for example competition or privatization. That means, in principle, that comparative generalizations are possible, though not all such works actually draw them.
Fourthly, there are cross-national studies with a fundamental theoretical orientation that take up a specific medical care theme or question as the focus of analysis. One of the interesting features of this fourth category of comparative studies is that there appears a necessary trade-off between theoretical depth and the number of nations studied. The disciplined treatment of broad topics by a single author almost inevitably addresses a more limited set of countries.
In this latter category, Carolyn Tuohy’s Accidental Logics (1999) offers both a theoretical and empirical analysis of policy change and continuity in three English-speaking nations. The book addresses a limited range of countries but combines theoretical sophistication with command of the relevant factual data and causal analysis in addressing the quite different patterns of policy change during the post-World War II years in the UK, Canada and the United States.
The likelihood of major policy changes, for Tuohy, differs according to each nation’s particular “institutional mix.” By that, she means the degree of governmental hierarchy, market forces and professional collegiality in medical decision-making and the “structural balance” between the state, medical providers and private financial interests. Directed at understanding, Tuohy’s work is of clear relevance to policy makers concerned with questions of timing for reform initiatives.
Works in this fourth category of scholarship typically use comparative methods to explore and to explain policy developments. Their practical limitations for policy makers include the relatively restricted range of countries studied and, to some degree, their considerable reliance on the theoretical perspective known as historical institutionalism.
There is some irony in the fact that the most careful cross-national analyses tend to have reinforced a sense of the contingency and specificity of the way things work out at different times in different places. This kind of comparison seems to ignore (if not implicitly deny) the cross-national exchange of information and ideas in health policy that is so much part of the very intellectual environment in which it has been produced.
The most powerful studies are at the same time the most academic; the practical learning which might result from comparison is largely left implicit. Often, those books do not reach the desk of policy makers. There is much less here that speaks directly to the policy maker seeking to use evidence and experience from elsewhere in any straightforward way.
Nonetheless, in the course of little more than a decade, the comparative analysis of health policy became a specialized field of academic inquiry, highly developed and successful in its own terms, but limited so far in its policy impact. So, we turn back to the evaluative question: how should we evaluate the purposes and performance of comparative policy research?
Perhaps the most important lesson we can draw from the overview of the current literature is that the development of a serious body of comparative work takes more time and effort than health policy makers are willing to spend. They feel pressures to take action and feel they cannot wait.
At the same time, policy errors based on misconceptions of the experience abroad can be costly. The eagerness of some health ministers to embrace and import policy models from the U.S. like the managed care models, the benchmark methodology or the medical savings idea without a proper assessment of how those ideas and models worked out in practice may lead to policies that will require repair action soon, can force politicians to reverse policies altogether and can erode the popular support for health policy altogether.
The unwillingness of some politicians to delay action in order to study experience with similar policy elsewhere contrasts sharply with the practice of some Asian countries that have spent much time and attention before adjusting certain measures to their own national policy environments.
The good news is that the last two decades have brought a large body of comparative study that can serve as the base for the next generation of studies that take the above warnings into account.
The statistical data are there, the materials are there, the experience in drawing portraitures of individual countries is there, and all of those are necessary conditions for the next phases of policy learning about causalities and the transfer of policy experience.
Tomorrow: Summary and conclusions