Despite the fact that epidemiologic studies demonstrate a consistent covariation between the use of tobacco, alcohol, and coffee, most previous behavioral genetic-studies have determined the contribution of genetic and environmental influences as if the consumption of these substances occurred independently of each other. In this study, we used multivariate structural equation modeling to determine the genetic and environmental overlap in the observed correlations between tobacco smoking and alcohol and coffee drinking in 173 monozygotic and 183 dizygotic male twin pairs (M age = 59 years; range = 52-66 years) who participated in a follow-up cardiovascular examination of the National Heart, Lung, and Blood Institute's Twin Study. Consistent with hypothesized psychoneurogenetic predispositions for the joint use of these substances, the most parsimonious model fitting these data identified a common genetic latent factor underlying the observed associations between smoking, alcohol, and coffee use in this cohort. This factor, herein called polysubstance use, underscores the role of genetic influence on the clustering of these behaviors in the same individual.