Editorial Incidence Rates of Stroke in the Eighties: The End of the Decline in Stroke?

the study of Broderick et al,1 based on the trends in stroke incidence in Rochester, Minnesota, between 1950-1954 and 1980-1984. The incidence of stroke for men in that community declined until 1975-1979 and then increased in 1980-1984, while for women the incidence rates appeared to "bottom out " around 1970-1974 and increased in 1980-1984. The appar-ent increase between 1975 and 1979 occurred in every 10-year age group from 45 to older than 85 years. However, by 1980-1984, approximately half of the incident strokes occurred among individuals 75 years of age or older. The mortality and case-fatality rates following a stroke continued to decline, which is consistent with trends in stroke mortality in the United States and many other countries.2 The first question to be resolved is whether the trends to 1980-1984 are an artifact of case-ascertainment, criteria of diagnosis, changes in the characteristics of the population at risk, or, possi-bly, errors in the measurement of the size of the population at risk. The cases were all collected in the same way, reviewed by the same nurse-abstractor, and had criteria of diagnoses that were consistent over time. It is extremely unlikely that between 1975-1979 and 1980-1984 there was a substantial change in the ascertainment of stroke cases within this commu-nity. It is possible that strokes diagnosed in the 1950-1954 period were missed, but if this were the case, the trends to 1975-1979 would not have shown the decline in the incidence rates. The selected increase in the use of CT scans in the 1970s and 1980s could have resulted in a greater ascertainment of strokes, but the authors wisely excluded strokes diagnosed only on the basis of CT scan or autopsy findings, without prior clinical evidence of stroke. They believe, however, that the use of CT scans may have increased physicians&apos