The purpose of this study is to investigate whether there is an association between exercise participation and self-rated health and whether this association can be explained by common genes and/or common environmental influences. In a sample of 5,140 Dutch adult twins and their non-twin siblings from 2,831 families, exercise participation (sedentaries, light or moderate, vigorous exercisers) and self-rated health were assessed by survey. To investigate the etiology of the association, bivariate genetic models using structural equation modeling were applied to the data. The correlation between exercise participation and self-rated health is significant but modest (r = 0.20). Exercise participation and self-rated health are both heritable (around 50% of the variance of both phenotypes is explained by genetic factors). The genetic factors influencing exercise participation and self-rated health partially overlap (r = 0.36) and this overlap fully explains their phenotypic correlation. We conclude that the association between exercise and self-rated health can be explained by genes predisposing to both exercise participation and self-rated health. These genes may directly influence both phenotypes (pleiotropy). Alternatively, genes that affect exercise or self-rated health may indirectly influence the other phenotype through a causal relationship. We propose that identification of the genes that cause differences in exercise behavior will help resolve the issue of causality. © 2006 Springer Science+Business Media B.V.