Some limitations of this study are worth noting. First, the evaluations focused only on European American subjects as the meta-analyses were restricted to similar samples. Second, these data did not have a quantitative “CPD-like” measure of cannabis use. Third, as the sample was enriched for alcohol dependence, rates of cigarette smoking and cannabis involvement are higher than those noted in the general population. This enrichment, particularly as one of the contributing samples was ascertained for nicotine dependence (COGEND), may have influenced our ability to detect these associations. Importantly, the associations with cannabis involvement should be viewed as preliminary and pending replication.
