By decomposing the variance in each of the three phenotypes and estimating cross-substance correlations, we were able to test for the contribution of common vulnerability factors to the association between the timing of first use for alcohol, cigarettes, and cannabis and determine that this common vulnerability is primarily genetic in nature. The limited research examining overlap across substances in factors influencing initiation has produced mixed results. In their study using a sample of older Caucasian women, Hopfer et al. (2001) reported that lifetime (ever) use of alcohol and cigarettes was attributable to the same shared environmental factors. Young et al.’s (2006) investigation of alcohol, cigarette, and cannabis use in an adolescent sample revealed modest but significant correlations across substances in both shared environmental and genetic influences, although they defined use as repeated use, thus potentially tapping a different construct than age at first use. Perhaps the most intriguing finding in this area comes from a study of 12 to 25 year-olds by Koopmans et al. (1997), who found high correlations between shared environmental influences on initiation of alcohol and cigarette
