Body mass index and mortality: a twelve-year prospective study in Korea.

The relation between body mass index (kg/m2) and noncoronary mortality is not well established. To study this relation, a population with low coronary mortality may be especially useful. We conducted a 12-year follow-up study of 235,398 Korean men 40-64 years of age. Study subjects had undergone health examinations in 1986 (baseline) and 1990. We excluded subjects with substantial weight loss during this period. There were 13,387 deaths, including 600 deaths from coronary events, between 1990 and 1998. We estimated the relation of body mass index to the risk of death after adjusting for common risk factors. There was a positive relation between body mass index and coronary mortality, but this relation was attenuated after serum total cholesterol, blood pressure, and fasting serum glucose were taken into account. A J-shaped relation with cerebrovascular mortality was also attenuated after adjustment. Even after this adjustment and exclusion of early deaths between 1990 and 1994, the relation of body mass index to all-cause (U-shaped), cancer (J-shaped), and noncancer noncoronary noncerebrovascular (inverse J-shaped) mortality remained. Both high and low body mass index were related to increased mortality among these Korean men.