Lewis A. Conner Lecture. Contributions of epidemiology to the prevention of stroke.

E pidemiological study and clinical trial research has flourished during the past 40 years. During this period the decline in death rates from stroke in the United States has accelerated. Since 1970 the US stroke mortality rate has declined nearly 60%. There is reason to believe the declining death rate is related to improvements in the risk factor status of the US population. Some of the data presented is derived from the Framingham Study and has been collected by many investigators during the past 40 years. I specifically acknowledge the contributions of William B. Kannel, Ralph B. D'Agostino, Margaret Kelly-Hayes, Carlos S. Kase, Thomas R. Dawber, and William B. Castelli to this effort. Stroke is the third leading cause of death in the United States, accounting for 147 470 deaths in 1989. Stroke is more often disabling than lethal and is the No. 1 cause of neurological disability in adults. Mortality statistics fail to convey the toll of stroke. One must also take into account the extraordinary degree of human suffering experienced by stroke survivors and their families, whose lives are irrevocably altered by this catastrophic illness. To the functionally independent elderly person, stroke often represents a situation worse than death itself. It is estimated that 30% of stroke survivors need assistance in their daily activities, and in the elderly, the attendant loss of function and independence often signals the end of an active life.' Many of the estimated 2 980 000 stroke survivors require chronic health care, including institutionalization.1 Not all stroke occurs in the elderly: one third of stroke victims are younger than 65 years of age, and most of them will never be able to work again. Nevertheless, the remarkable decline in stroke mortality clearly shows that stroke is preventable. Stroke is not an inevitable consequence of aging but rather the consequence of a chain of events set in motion many years before. Furthermore, risk for stroke is predictable. After 36 years of follow-up of the Framingham Study cohort there were 693 strokes and transient ischemic attacks (TIAs). Since 1967, most cases were seen either by me or by another Framingham Study neurologist at