OBJECTIVE: To examine the association of physical activity and body mass index (BMI), and their combined effect, with the risk of total, cardiovascular disease (CVD) and cancer mortality. DESIGN: Prospective follow-up study. SUBJECTS: In all, 22 528 men and 24 684 women aged 25-64 y at baseline having 7394 deaths during a mean follow-up of 17.7 y. MEASUREMENT: A self-administered questionnaire data on smoking, socioeconomic factors, physical activity and medical history, together with measured height, weight, blood pressure and serum cholesterol using standardized protocol. RESULT: Physically active subjects had significantly lower age-adjusted mortality from cardiovascular, cancer and all causes compared with sedentary ones. Further adjustment for smoking, systolic blood pressure, cholesterol, BMI, diabetes and education affected the results only slightly. Obese subjects (BMI >= 30 kg/m(2)) had significantly higher cardiovascular and total mortality than the normal weight (18.5 <= BMI <= 25 kg/m2) subjects. Part of increased mortality among obese subjects was mediated through obesity-related cardiovascular risk factors. BMI had an inverse association with cancer mortality among men and almost significant direct association among women. Total mortality was also increased among the lean (BMI < 18.5 kg/m(2)) subjects. However, less than 0.3% of deaths were attributed to low body weight, whereas in men 5.5% and in women 17.7% of deaths were attributed to obesity. CONCLUSION: Regular physical activity and normal weight are both important indicators for a decreased risk of mortality from all causes, CVD and cancer. Physical activity had a strong independent effect on mortality, whereas the effect of BMI was partly mediated through other obesity-related risk factors.
