Background: People who are physically active live longer, but it is unclear whether this is because of physical activity in the distant or more recent past. Methods: We assessed activity levels in 5209 men and women in the Framingham Heart Study from 1956 to 1958 and again from 1969 to 1973. We included individuals who were alive and without cardiovascular disease in the period 1969 to 1973. The primary outcome was death from all causes during the 16 years after the 1969 to 1973 assessment. Secondary outcomes were incidence and mortality rate of cardiovascular disease. We used Cox proportional hazards regression to calculate the relative risk of being sedentary, both unadjusted and controlling for smoking, weight, systolic blood pressure, cholesterol, glucose intolerance, left ventricular hypertrophy, chronic obstructive pulmonary disease, and cancer. Results: The overall 16-year mortality rate was 37% for men and 27% for women. When both distant and recent activity levels were included along with major cardiovascular disease risk factors, for recent activity the most active tertile had lower overall mortality rate than the least active tertile for men (risk ratio 0.58, 95% confidence interval, 0.43-0.79) and women (risk ratio 0.61, 95% confidence interval, 0.45-0.82). For distant activity there was no difference in overall mortality rate between the most and least active tertiles either for men or for women. Adjusting for major cardiovascular disease risk factors had little effect on the results. Conclusions: The reduction in overall mortality rates is more associated with recent activity than distant activity. These results suggest that for sedentary patients, it may never be too late to begin exercising. (Am Heart J 1999;138:900-7.)