Background The purpose of the study was to examine the association between physical activity and 23-year incidence of coronary heart disease morbidity and mortality. This cohort study continues to follow 8006 Japanese-American men who were 45 to 68 years of age and living on Oahu, Hawaii, in 1965, for the development of coronary heart disease morbidity and mortality.Methods and Results The Framingham physical activity index was calculated by summing the product of average hours spent at each activity level and a weighting factor based on oxygen consumption. Study subjects were divided into tertiles of physical activity index at baseline. Relative risks and 95% confidence intervals (CI) for incidence of coronary heart disease morbidity and mortality were obtained using the Cox model. After age adjustment and using the lowest physical activity index tertile as a reference group, the relative risk for coronary heart disease incidence for the highest tertile of physical activity was 0.83 (CI, 0.70 to 0.99). After adjusting for age, hypertension, smoking, alcohol intake, diabetes, cholesterol, and body mass index, the relative risk was 0.95 and CI included 1 (CI, 0.80 to 1.14). For coronary heart disease mortality, the age-adjusted relative risk was 0.74 (CI, 0.56 to 0.97) and 0.85 (CI, 0.65 to 1.13) after risk factor adjustment.Conclusions The results suggest that the impact of physical activity index on coronary heart disease is mediated through its effects on hypertension, diabetes, cholesterol, and body mass index. These findings support the hypothesis that physical activity is inversely associated with coronary heart disease morbidity and mortality and suggest that physical activity interventions in middle-aged men, by improving cardiovascular risk factor levels, may have significant public health implications in the prevention of coronary heart disease. (Circulation. 1994;89:2540-2544.)