Both high peak oxygen consumption (VO(2)) and high levels of leisure time physical activity (LTPA) are associated with a reduced risk of cardiovascular morbidity and mortality. We examined the contributions of LTPA and peak VO(2) to the risk of coronary events (CEs) in healthy younger (< or = 65 years, n = 522) and older (>65 years, n = 167) men from the Baltimore Longitudinal Study of Aging. LTPA derived from self reports of time spent in 97 activities were converted into METs-minutes/24 hours and grouped into high (> or = 6 METs), moderate- (4 to 5.9 METs), and low-intensity LTPA (<4 METs). Cardiorespiratory fitness was determined by measuring peak VO(2) during a maximal treadmill exercise test. Over a mean follow-up of 13.4 +/- 6.3 years, CEs occurred in 63 men. After accounting for coronary risk factors, proportional-hazards analyses showed a relative CE risk of 0.53 (p <0.0001) for a SD increase in peak VO(2) in younger men and 0.61 (p = 0.024) in older men, whereas total LTPA was unrelated to coronary risk in either age group. When the 3 LTPA intensity levels were substituted for total LTPA in the model, peak VO(2) remained the only predictor of events in younger men, whereas high-intensity LTPA (RR = 0.39 for tertile 3 vs tertiles 1 and 2, p = 0.016) and peak VO(2) (RR = 0.61/SD increase, p = 0.024) were of similar importance in older men. Thus, in healthy younger men, higher cardiorespiratory fitness but not LTPA predicts a reduced risk of coronary heart disease, independent of conventional risk factors. For older men, high-intensity LTPA and fitness appear to be of similar importance in reducing coronary risk.