We investigated the association between physical activity and colorectal cancer risk in a cohort of Swedish men. Information on physical activity was obtained at baseline in 1997 with a self-administered questionnaire from 45,906 men who were cancer-free at enrollment. During a mean follow-up of 7.1 years, 496 cases of colorectal cancer occurred. Leisure-time physical activity was inversely associated with colorectal cancer risk; the multivariate hazard ratio (HR) for 60 min or more per day of leisure-time physical activity compared with less than 10 min per day was 0.57 (95% CI 0.41-0.79; P for trend=0.001). Results were similar for colon (HR=0.56; 95% CI 0.37-0.83) and rectal cancer (HR=0.59; 95% CI 0.34-1.02). Home/housework activity was inversely associated with colon cancer risk (HR=0.68; 95% CI 0.48-0.96). No association was observed for work/occupational activity. These results support a role of physical activity in reducing the risk of colon and rectal cancer.