New research in the International Journal of Epidemiology by three scholars at the University of Michigan — Javier M Rodriguez, John Bound, and Arline Geronimus — extends this vein of research into a new area: infant mortality. The longer-term trend in infant mortality in the United States is clear: infant mortality has declined sharply in the past 80 years.
What Rodriguez and colleagues investigate is whether, relative to this broader trend, infant mortality rates are higher under Republican or Democratic presidential administrations. Drawing on data from 1965 to 2010, they find that infant mortality is, on average, 3 percent higher under Republican administrations (again, relative to the trend). This is after accounting for other factors that could also affect infant mortality, such as trends in education attainment, the unemployment rate, and economic inequality.
This finding has already occasioned some debate. In the same issue of the International Journal of Epidemiology, Ralph Catalano likens this finding to “dog food” and critiques the statistical methods of the authors. Rodriguez and colleagues respond to this critique here. I will spare you the details, except to say that Rodriguez and colleagues tried a variety of statistical methods and found the same result in each case.
The bigger question, however, is whether there is a compelling causal story to back up this correlation. Another scholar who responded to this article, Danny Dorling, is more persuaded by it, but notes that we do not know nearly enough about why the party of the president might have any effect. Rodriguez and colleagues offer this speculative explanation:
For example, it may be that Republicans are more likely to view health disparities as inevitable, whether due to market forces or as a matter of personal responsibility to be addressed through individual health behaviors, not by government. Democrats, on the other hand, may be more likely to view health disparities as a preventable social problem about which something should be done via government intervention. Although this is an oversimplified rendering of the differences in the ideological traditions, it suggests one type of reason why one might expect Republican and Democratic presidents to affect population health differently. For example, whether a government advocates austerity measures or increasing social welfare protections as the response to economic crisis may reflect such ideological differences and influence mortality rates.
At the same time, they are well-aware that much more needs to be done. They conclude by saying:
Further research is needed to determine whether the association we have uncovered is causal, and to identify the mechanisms involved.
And that’s the appropriate tone, I think. There is a correlation here that persists after accounting for some obvious alternative explanations. However, the mere existence of this correlation does not permit any strong conclusions. Perhaps further studies will help us understand more.