For all three cross-substance comparisons, initiation of use for one substance (categorized as never, late or early) was strongly associated with the timing of first use for the other, providing further support that risk for co-occurring substance use is elevated in early users of alcohol (Callas et al., 2004; Ellickson et al., 2003; Schmid et al., 2007; Vieira et al., 2007), cigarettes (Korhonen et al., 2008; Kokkevi et al., 2006; Sartor et al., 2008), and cannabis (Coffey et al., 2000). More specifically, our findings are consistent with other studies reporting that risk for early use of licit or illicit drugs is increased in early initiates of another substance of abuse (e.g., risk for early initiation of cannabis is elevated in individuals who began drinking and smoking cigarettes at a young age [Agrawal et al., 2006; Schmid et al., 2007]). Furthermore, despite the lower overall prevalence of alcohol and cigarette (but not cannabis) use among African-Americans (Ellickson et al., 2004a; Grucza et al., 2008; Heath et al., 1999; Scarinci et al., 2002; Vega et al., 2007), we found the same patterns of cross-substance use as those found in the larger literature, which is based primarily on Caucasians.
