Plasma dehydroepiandrosterone sulfate (DHEA-S) and testosterone levels both decline with age in healthy men. Features of the metabolic syndrome also show age-related deteriorations. We examined the relative contribution of age and declining androgen levels to features of the metabolic syndrome in men. In a sample of 130 nonsmoking men from the Quebec Family Study, we tested the hypothesis that age-related decreases in DHEA-S and testosterone levels would explain most of the variance in alterations of the metabolic profile associated with aging. As expected, we found that plasma DHEA-S and testosterone levels were negatively associated with age. Significant negative correlations were found between androgen levels and adiposity measures, body fat distribution, and metabolic risk variables. Statistical control for age eliminated correlations with DHEA-S, whereas age-adjusted associations between testosterone and most adiposity and metabolic variables remained significant. The percentage frequency of men characterized by 3 or more features of the metabolic syndrome increased with decreasing testosterone (8.9%-44.2%, chi2 = 15.89, P < .0005 ) and DHEA-S levels (8.9%-41.5%, chi2 = 13.02, P < .005). Logistic regression analyses showed that men in the upper tertile of testosterone levels had a lower risk of being characterized by 3 or more features of the metabolic syndrome (odds ratio = 0.24, P < .04) independent of age, whereas tertiles of DHEA-S levels were not related to the metabolic syndrome independent of age. In conclusion, results suggest that age per se is an important correlate of the associations between DHEA-S and metabolic variables, whereas the association of plasma testosterone levels to features of the metabolic syndrome appears to be independent of age.