The tracking of physical activity and its influence on selected coronary heart disease risk factors were studied in a 6-year (original survey in 1980, with follow-ups in 1983 and 1986) study of Finnish adolescents and young adults as part of the Cardiovascular Risk in Young Finns Study. The subjects in this analysis were aged 12, 15, and 18 years at baseline. Physical activity was assessed with a standardized questionnaire, and a sum index was derived from the product of intensity, frequency, and duration of leisure time physical activity. Complete data on physical activity index from each study year were available on 961 participants. Significant tracking of physical activity was observed with 3-year correlations of the index ranging from 0.35 to 0.54 in boys and from 0.33 to 0.39 in girls. Tracking was better in older age groups. Two groups of adolescents (active and sedentary groups) were formed at baseline according to high and low values of the index, respectively. Approximately 57% of those classified as inactive remained inactive after a 6-year follow-up. The corresponding value for active subjects was 44% (p < 0.01, active vs. inactive). The long-term effects of physically active and sedentary life-styles were studied by comparing groups of young adults who had remained active or inactive in every three examinations. Serum insulin and serum triglyceride concentrations were significantly lower in active young men. They had a more beneficial high density lipoprotein to total cholesterol ratio and thinner subscapular skinfolds. Among young women, significant differences were seen in adiposity (subscapular skinfold) and in serum triglyceride concentration. Physical activity was also related to less smoking in both sexes and, among young men, to lower consumption of saturated fatty acids and to higher polyunsaturated to saturated fatty acids ratio of the diet. In regression analyses adjusted for the 6-year change in obesity, smoking status, and diet, the change in physical activity was inversely associated with changes in serum insulin and triglycerides in boys. Independent association with triglycerides disappeared when insulin change was added to the model, suggesting that the effect may partly be mediated through insulin metabolism. The authors conclude that the level of physical activity tracks significantly from adolescence to young adulthood. Physical inactivity shows better tracking than does physical activity, and subjects who are constantly inactive express a less beneficial coronary risk profile compared with those who are constantly active.(ABSTRACT TRUNCATED AT 400 WORDS)