A JOURNAL OF REPLICATED RESULTS.

A JOURNAL OF REPLICATED RESULTS. In the article by Robert Samuelson I posted on yesterday, he questions whether increased health expenditures have much effect on health. He says: “But the connections between being healthy and more health spending are loose. The health of most people reflects personal habits and luck.” I find this implausible, but maybe I am overly impressed by the differences between what I saw of health care in 1975 and in 2005. I posted here on an article by David Brown which pointed out that in 1975 the chances of dying in the days immediately after a heart attack was 27 percent, whereas today it’s about 6 percent. I don’t think that this improvement can be characterized as reflecting “personal habits and luck.” What is striking to me, however, is that the effect of health expenditures on longevity is an important issue. It also seems to me that, as with so many important issues, there are only a few scattered studies. I think that the reason for the scarcity of studies is that the first study in a policy area is easily published. Later studies don’t seem to be pathbreaking, and so they are harder to publish and don’t further careers. There is less incentive to do follow up studies so that larger samples can be developed and results verified. My brother Elmer used to argue that there should be a journal of negative results because failed experiments contribute to knowledge. As I posted on here, there are now several journals of negative results. I think there should be a journal (or journals) of replicated results to reward studies which follow on promising results.