The effect of physical fitness and physical activity on all-cause mortality and mortality due to cardiovascular disease (CVD), cancer and other causes was examined using a population-based representative sample of the Canadian population. A total of 691 deaths occurred among persons age 30-69 during the 7-year follow-up period. Of these, 37, 33 and 30% died of CVD, cancer and other causes, respectively. The effect of each risk factor on mortality was assessed using logistic regression analysis. Adjustment was made for age, sex, smoking and body mass index in the case of mortality due to CVD, all causes and causes other than CVD and cancer. Models for mortality due to cancer included adjustment for age, sex, smoking and alcohol consumption. For all cause mortality, those individuals who did not pass the physical fitness tests had significantly higher risks of death than those that passed. For CVD mortality, subjects whose physical activity was moderate were protected, and those who did not pass the physical fitness tests had substantially higher risks of death due to CVD. The risk of death due to cancer was not significantly related to physical activity or physical fitness. Persons who failed the physical fitness tests had significantly elevated risks of death due to causes other than CVD or cancer. Our findings support the conclusion that persons who are physically fit have an overall reduced risk of death independent of other major risk factors. Moderate levels of physical activity appear to be protective against cardiovascular disease.