Bidirectional associations between cannabis use and depressive symptoms from adolescence through early adulthood among at-risk young men (2016)

OBJECTIVE: Previous studies have established a relationship between cannabis use and affective problems among adolescents and young adults; however, the direction of these associations remains a topic of debate. The present study sought to examine bidirectional associations between cannabis use and depressive symptoms, specifically testing the validity of two competing hypotheses: the cannabis effect hypothesis, which suggests that cannabis use contributes to the onset of later depressive symptoms; and the self-medication hypothesis, which posits that individuals increase their use of a substance to alleviate distressing psychological symptoms.
METHOD: Participants in this study were 264 low-socioeconomic-status males assessed at ages 17, 20, and 22. Cross-lag panel models were fit to test bidirectional associations between cannabis use frequency and depressive symptoms across the transition from adolescence to early adulthood. In addition, analyses were conducted within two high-risk subsamples to examine whether associations between cannabis use frequency (ranging from never used to daily use) and depressive symptoms differed among regular cannabis users (used cannabis more than once per week) or subjects reporting at least mild levels of depressive symptoms.
RESULTS: Cannabis use and depressive symptoms were concurrently correlated. Cannabis use predicted increases in later depressive symptoms, but only among the mild-depression subsample. Depressive symptoms predicted only slight increases in later cannabis use, among the subsample of regular cannabis users.
CONCLUSIONS: Temporal patterns of cannabis use and depressive symptoms provide evidence for the cannabis effect but limited evidence for the self-medication hypothesis. Adolescents higher in depressive symptoms may be vulnerable to the adverse psychological effects of using cannabis. Results are discussed in terms of implications for basic research, prevention, and intervention.