Large longitudinal study finds cannabis use predicts increasing anxiety, but anxious people actually reduce their cannabis use

For the overall sample and men, greater cannabis use predicted greater subsequent increases in anxiety (substance-induced pathway). However, greater anxiety was associated with decreasing, not increasing, cannabis use (contradicting the self-medication hypothesis). For women, the patterns were more complex, with trait-like anxiety predicting increasing anxiety but recent anxiety increases associated with reduced cannabis use.

Data from a longitudinal cohort study (N=2,995) using an accelerated longitudinal design from ages 17-24. Latent difference score models estimated bidirectional relationships between anxiety and cannabis use across 6 waves, separately for men and women.

This study challenges the popular self-medication hypothesis by showing anxious people actually tend to reduce cannabis use, while supporting the idea that cannabis use can increase anxiety over time.

The finding that the self-medication pathway (anxiety driving cannabis use) is not supported, while the substance-induced pathway (cannabis driving anxiety) is, has implications for how we understand and treat co-occurring anxiety and cannabis use.

Self-reported measures. Cannot account for all confounders. Southern California sample. Cannabis use measured by frequency, not dose or potency. Accelerated longitudinal design assumes cohort equivalence.