Discrepancies between our findings and those reported elsewhere may be attributable to differences in the phenotypes under study. Whereas the current study addressed the timing of first use, the vast majority of genetically informative studies have operationalized initiation more broadly as ever vs. never use, which may not capture the risk specific to early onset. Although one of the few known studies to address age at first use found evidence for primarily environmental influences on initiation of alcohol and cigarette use (Stallings et al., 1999), the very limited literature in this area and the demographic differences between our sample and the older, all-Caucasian twin cohort used by Stallings and colleagues suggest that the nature of risk for early vs. late onset use remains an open question.
