Low-density lipoprotein is recognized as a primary vascular risk factor. However, recent data favor apolipoprotein (apo)B and apoA-I as risk factors with higher predictive values than conventional lipids. We investigated how leisure-time physical activity relates to the serum apoB/apoA-I ratio in middle-aged men. The results showed that compared with a sedentary lifestyle, moderate physical activity was associated with a decreased apoB/apoA-I ratio (1.01 ± 0.28 vs 0.87 ± 0.24, P < .05) and increased apoA-I levels (1.30 ± 0.20 g/L vs 1.43 ± 0.22 g/L, P < .05), whereas vigorous activity was required to observe a reduction in apoB levels (1.27 ± 0.28 g/L vs 1.14 ± 0.24 g/L, P < .05). A covariate analysis showed that leisure time physical activity was also associated with reduced apoB/apoA-I ratios after adjustment for smoking, systolic blood pressure and waist circumference. Importantly, this association was seen at moderate levels of physical activity, supporting the notion that some activity is better than none.