We examined the relation between physical activity, physical fitness, and all-cause mortality in a national population-based study of Canadians. We followed men and women ages 20-69 years who had participated in the Canada Fitness Survey between 1981 and 1988. We assessed risk factors for 6,246 men and 8,196 women using multivariate Poisson regression analysis. At baseline, all subjects were asymptomatic according to self-reported screening questions for cardiovascular disease. Men who expended > or =0.5 kilocalories per kilogram of body weight per day (KKD) experienced a 20% decline in risk of mortality [rate ratio (RR) = 0.82; 95% confidence interval (CI) = 0.65-1.04] when compared with subjects expending <0.5 KKD. We observed a 30% decline in risk of mortality among women expending > or =3.0 KKD relative to those expending <0.5 KKD (RR = 0.71; 95% CI = 0.45-1.11). Similar patterns of risk were evident for both men and women when analyses were restricted to participation in nonvigorous activities. Those who perceived themselves to be of less than average fitness were at increased risk of mortality (male RR = 1.64, 95% CI = 1.21-2.22; female RR = 1.66, 95% CI = 1.21-2.26). Subjects with undesirable cardiorespiratory fitness levels were more likely to die, compared with those having recommended fitness levels (RR = 1.52; 95% CI = 0.72-3.18). Fifty-three per cent of men and 35% of women reported participating in a vigorous activity. The relation between daily energy expenditure and risk of mortality in these subjects could not be evaluated, as there were few deaths. Nonetheless, our results among participants reporting no vigorous activities support the hypothesis that there is a reduction in mortality risk associated with even modest participation in activities of low intensity.