For cigarette use, a beta value of 0.87 and an upper confidence interval including 1 provided evidence that the underlying liabilities for initiation and quantity of use may lie on the same continuum. Heavier use would then represent a higher loading on the same liability distribution as initiation. This suggests that there may be substantial overlap in the risk factors involved in both initiation and progression. It does not mean that individuals who initiate smoking will necessarily progress to heavier use. However, nicotine is a highly addictive substance [36] and it is possible that this is reflected in the strong relationship we find between initiation and progression. For marijuana use, we also found evidence for mainly overlapping liabilities [beta value of 0.88 and an upper confidence interval approaching unity (0.99)]. This suggests that the risk factors implicated in initiation may also be important for progression to heavier use. Longitudinal, epidemiological studies of cigarette and marijuana use in adolescents in a US-based sample also suggest considerable overlap in risk factors (e.g. peer substance use and other substance use) for the initiation and progression stages, with only a few stage-specific risk factors found [37,38].
