Associations of regular leisure-time physical activity and changes in leisure-time physical activity with risk of death were studied in 7,023 healthy men and women aged 20-79 years in Copenhagen, Denmark. Physical activity was estimated in both 1976-1978 and 1981-1983. Men consistently engaging in a moderate or high degree of physical activity, respectively reported at both examinations, had significantly lower risks of death than men reporting low activity at both examinations. Adjusted relative risks were 0.71 (95% confidence interval (CI): 0.57, 0.88; p = 0.002) and 0.61 (95% CI: 0.48, 0.76; p < 0.001), respectively. Similar relative risks were found in women: 0.64 (95% CI: 0.52, 0.79; p < 0.001) and 0.66 (95% CI: 0.51, 0.85; p = 0.001), respectively. Men who increased their leisure-time physical activity from low to moderate or high had a significantly lower risk of death than men reporting low physical activity at both examinations (relative risk = 0.64, 95% CI: 0.50, 0.81; p < 0.001). In this study, maintaining or adopting a moderate or high degree of physical activity was associated with lower risk of death across a wide range of ages in both sexes.