The relative contributions of a low and high body mass index (BMI [weight in kilograms divided by height in meters squared]) to all-cause and cause-specific mortality are still controversial. To examine mortality rates in relation to BMI in a prospective cohort study of 48,287 Dutch men and women aged 30 to 54 years at baseline from 1974 to 1980. During an average 12-year follow-up, 1319 deaths occurred. Relative risks (RRs) were calculated from the Cox proportional hazard model by using a BMI between 18.5 and 24.9 kg/m2 as the reference category. All-cause mortality was significantly increased in obese men (BMI, > or = 30 kg/m2; RR, 1.5; 95% confidence interval [CI], 1.1-2.0) and in underweight men (BMI, < 18.5 kg/m2; RR, 2.6; 95% CI, 1.8-3.9) but not in women. The increased risk in underweight men could be attributed to deaths within the first 5 years of follow-up and to lung cancer mortality among smokers. Coronary heart disease (CHD) mortality was about 3-fold higher among obese men and women. About 21% and 28% of CHD mortality in men and women, respectively, could be attributed to being overweight (BMI, > or = 25 kg/m2). The RR (but not the absolute risk) for CHD among obese men was still significant after adjustment for the presence of smoking, hypertension, hypercholesterolemia, and diabetes mellitus at baseline, and it was more pronounced for CHD among nonsmokers than among smokers (RR, 7.1; 95% CI, 2.3-21.7; and RR, 2.7; 95% CI, 1.5-4.7, respectively). Total mortality was increased in obese and underweight men but not in women. The increased mortality in overweight men was mainly attributable to CHD and, in underweight men, to early mortality and especially lung cancer mortality among smokers.