Background: Adolescent involvement in religious organizations has been hypothesized to protect against early age at first drink. However, the correlation between adolescent religiosity and later age at first drink may be confounded by environmental or genetic differences between families. This study tests whether, after controlling for shared environmental and genetic confounds using a behavior genetic design, the association between individual levels of religiosity and earlier age at first drink is still evident. Method: Twin and sibling pairs were drawn from the National Longitudinal Study of Adolescent Health, a nationally-representative sample of US adolescents. Age at first drink was measured as how old adolescents were when they first had a drink of beer, wine, or liquor. Religiosity was measured using four items concerning frequency of religious activities and importance of religious beliefs. Using twins and siblings who were discordant for religiosity, analyses tested whether religious adolescents had a later age at first drink than their non-religious co-twins/co-siblings. Results: Religious adolescents did not differ from their non-religious siblings in their mean age at first drink. Results from survival models indicate that environmental differences between families completely account for the correlation between religiosity and later age at first drink. Conclusions: Results suggest that individual religious involvement is a proxy variable for family or cultural environments that are salient for when adolescents initiate alcohol use. Future research is needed to identify specific protective environments in religious families. These results have implications for both public policy and etiological theory.