Habitual physical activity may appear as a determinant of cardiovascular morbidity because of (i) noncausal associations with other risk factors, (ii) an influence on known risk factors, and/or (iii) some, as yet, not defined mechanisms. The associations of physical activity with other life-style variables, such as smoking and plasma low-density lipoproteins, vary both in direction and strength with population, time, and type of physical activity (eg, work versus leisure time). On the other hand physical activity has been shown to cause an increase in plasma high-density lipoproteins and a decrease in plasma triglycerides and insulin and to predict a lower incidence of hypertension. In some multivariate analyses-of the incidence of coronary heart disease and strokes, physical activity has evolved as an independent predictor of reduced risk. On the type, intensity, and frequency of physical activity as needed for risk reduction, the results are not fully consistent. For delineating the mechanisms for the independent reduction of risk by physical activity, major investments of further research are still necessary.