Several studies have shown a beneficial effect of physical activity and exercise on health. A higher degree of or an increase in physical activity or physical fitness is associated with a lower risk of mortality or incidence of cardiovascular diseases [1±8], independent of several cardiovascular risk factors including smoking, obesity, blood pressure, serum lipids, socioeconomic status and diabetes [3±5].Physical activity also has a beneficial effect on many cardiovascular risk factors [2]. A higher degree of physical activity has been associated with a lower incidence of Type-II (non-insulin-dependent) diabetes mellitus [9] and a lower degree of insulin resistance [10,11]. Fasting concentrations of (immunore- Diabetologia (2001) Abstract Aims/hypothesis. To investigate the effect of changes in physical activity on changes in metabolic cardiovascular risk factors and to investigate what factors affect the association between physical activity and cardiovascular mortality. Methods. Of the 1860 men who were 50 years of age and who were without pre-existing cardiovascular disease participating in a population-based study, 898 were re-examined 20 years later. Altogether 231 died from cardiovascular diseases during the followup (mean = 22.6 years). The examinations which the men underwent at 50 and 70 years of age included assessment of physical activity (self-reported at four alternative levels), anthropometry, measurements of fasting concentrations of glucose, specific insulin, proinsulin, split proinsulin and lipids. Results. During the 20 years, 31 % increased their amount of physical activity while 51 % continued the same amount of exercise. Increased physical activity was associated with significant changes in several important metabolic variables, including fasting glucose, proinsulin and HDL cholesterol, independent of body weight changes. The risk of cardiovascular disease for men performing moderate, regular and athletic physical activity was 25 % (p = 0.127), 34 % (p = 0.022) and 71 % (p = 0.009) lower, respectively, compared with sedentary men. The association was attenuated by adjustment for baseline measurements of insulin, proinsulin and split proinsulin. Additional adjustment for other cardiovascular risk factors did not further attenuate the association. Conclusion/interpretation. Increased leisure time physical activity between the ages of 50 and 70 years, in the absence of active intervention, is associated with improved glucose, insulin and lipid metabolism in men. The concentrations of insulin, proinsulin and split proinsulin could mediate much of the association between a sedentary lifestyle and increased risk of cardiovascular mortality. [Diabetologia (2001)