In defense of death data: an example with multiple sclerosis.

In epidemiologic studies, mortality data can provide large numbers of cases of broad geographic extent, but diagnostic errors and case ascertainment bias may be present. We found that the geographic distribution of MS in the United States, as derived from mortality data, was comparable with that derived from a large, nationwide case-control study. In MS, evaluation of mortality provides an accurate picture of national geographic distribution supportable by more sophisticated methods. Although conclusions drawn from mortality data could be misleading, they may be accurate and are certainly worthy of further study.