Decline in autopsies for deaths attributed to cerebrovascular disease.

BACKGROUND AND PURPOSE: United States national autopsy rates have declined in recent years. In the present study, changes in autopsy rates for deaths due to stroke are examined and compared with changes in autopsy rates for all deaths. METHODS: National Center for Health Statistics data on United States national autopsy rates were examined for the years 1955, 1958, and 1972-1988. RESULTS: Since at least 1955, nonstroke deaths were more than twice as likely to be autopsied as deaths due to cerebrovascular disease. The annual autopsy frequency for all deaths, for deaths due to stroke, and for deaths due to each stroke subtype declined precipitously after 1972. Since 1982, less than 5 percent of deaths attributed to stroke have been documented by autopsy. Information obtained at autopsy was frequently ignored in the determination of cause of death on the death certificate. CONCLUSIONS: Careful consideration of the value of autopsy for education, research, and quality assurance is urgently needed. Unless the present problems in obtaining, processing, disseminating, and using autopsy data are adequately addressed, the autopsy rate will continue to decline.