smoking and marijuana use, the magnitude of the associations in adolescence was not statistically different from the magnitude of the association in adulthood, suggesting that age does not significantly moderate these associations. These results, which suggest a chronic comorbidity, are consistent with recent studies that found that the association between smoking and depression is established early and persists (Dierker et al., 2015; Leventhal and Zvolenksy, 2015). In contrast, the magnitude of association between regular HED and depressive symptoms was age-dependent. For both genders, the association was strong during early adolescence, decreased steadily during adolescence, and was no longer significant for either gender after age 18. These observed age trends may reflect that alcohol consumption and HED becomes normative and highly prevalent with age, unlike marijuana or cigarette use; 23% of individuals age 12 or older and 43% of adults 21 to 25 report past-month HED (SAMHSA, 2013). Several potential mechanisms could explain the age-varying association for HED. Selection bias may be at play such that the relatively few individuals engaging in regular HED in early adolescence have a higher risk for depression due to behavioral, familial, and environmental risk factors. Alternatively, adolescents at-risk for depression may be more likely to
