In spite of a very large literature, there is no consensus on the causes of the reduction in mortality in the 19th century. Less studied is what drove mortality decline in the 20th Century, and once again, there are wide disagreements. Higher mortality rates in poor countries are attributed by some to lack of health care, and by some to low incomes. The differences in mortality rates between socioeconomic groups within countries provide perhaps the most contested ground of all. In principle, there ought to be a coherent account for differences in mortality over historical periods, between countries, and within countries at a moment of time. This does not mean that there should be a single factor that explains everything but that if different factors operate at different times, there should be a coherent story of why.

• socioeconomic status
• public health
• behaviors
• medical technology
  

These come in many different manifestations, for example, socioeconomic status includes income, yet relative income is in many respects very different from absolute income, with the former associated with material as opposed to psychosocial effects on health. Socioeconomic status also includes the effects of education, wealth, social-class, and sometimes even gender, place of residence, and race and ethnicity. Public health focuses on measures that work through sanitation, water, air quality the environment, and vector control. There are also interactions between these factors; higher incomes are needed to fund the technology or sanitation systems; both behavior and technology respond to new scientific knowledge; and behavior is at least in part determined by socioeconomic conditions.

Most writers acknowledge the importance of all four, while emphasizing one over the others. The four broad explanations have strong political overtones. An emphasis on increasing incomes is often associated with pro-growth, individualistic, or “neoliberal” economic policies. By contrast, public health is seem as a collective enterprise that is often at odds with pro-growth economic policies. But there is also a camp that recognizes the importance of income, but argues for income redistribution as an instrument of health promotion. And an emphasis on technology favors medical as opposed to social determinants, with implications for medical provision and research.

Each of these four themes can be thought of as the rows of a matrix whose columns are the types of empirical evidence, historical, cross-country and over groups within a country. Much of the literature can be fitted into one or other of the twelve cells of the matrix, with not much crossing of cell boundaries. So that a good deal of the literature argues for a particular explanation in a particular context, for example for income and the associated nutrition as an explanation for historical patterns of mortality decline, or for public health measures as the primary factor accounting for the mid-20th century decline in mortality in poor countries, for technology as the cause of recent declines in mortality from cardiac disease in rich countries, or for the beneficial effects on mortality of better control over one’s life brought by rank in the British civil service. Many of these accounts are convincing on their own terms, yet they do not provide an understanding of how each fits into the broad picture, or whether they work outside the context in which they were developed.

Much could be learned by bringing together scholars from these different perspectives, and asking them to address, not the episodes emphasized in their own work, or at least not directly, but to talk about how their own explanations would deal with the other accounts of other episodes. We will ask each participant to circulate a few key readings prior to the meeting. Perhaps because of the many disciplines involved, literature on one set of arguments or episodes is often little cited in other literatures. For example, the demographic and epidemiological literatures are largely disjoint. Such a process may lead to a resolution that there are different explanations in different contexts, but we would then ask why that should be the case, and how to determine in new circumstances which explanation is likely to be dominant. But the discussions may also lead to new insights from one or another explanation in unfamiliar contexts.

Organization

The meeting will held on Friday, July 16th 2004 at the Center for Health and Wellbeing at Princeton, which will handle all of the logistics, (Susan Rizzo, srizzo@princeton.edu is the contact person.) Participants will assemble in time for a group dinner the previous day, July 15th. The organizers are David Cutler, Harvard, Angus Deaton, Princeton, and Richard Suzman of the National Institute on Aging, which is sponsoring the meeting. Three important strands in the discussion will be represented by Robert Fogel, Chicago, Sir Michael Marmot, UCL, and Sam Preston, Pennsylvania.

TOP