Background: Obesity and substance use are major concern in young people. This study explored the bidirectional longitudinal relationships between the body mass index (BMI) of young men and their use of (1) 4 classes of nonmedical prescription drugs; (2) alcohol; (3) tobacco; and (4) cannabis.

Methods: Baseline and follow-up data from the Cohort Study on Substance Use Risk Factors were used (N = 5007). A cross-lagged panel model, complemented by probit models as sensitivity analysis, was run to determine the bidirectional relationships between BMI and substance use. Alcohol was assessed using risky single-occasion drinking (RSOD); tobacco, using daily smoking; and cannabis, using hazardous cannabis use (defined as twice-weekly or more cannabis use). Nonmedical prescription drugs use (NMPDU) included opioid analgesics, sedatives/sleeping pills, anxiolytics, and stimulants.

Results: Different associations were found between BMI and substance use. Only RSOD (β = −.053, P = .005) and NMPDU of anxiolytics (β = .040, P = .020) at baseline significantly predicted BMI at follow-up. Baseline RSOD predicted a lower BMI at follow-up, whereas baseline NMPDU of anxiolytics predicted higher BMI at follow-up. Furthermore, BMI at baseline significantly predicted daily smoking (β = .050, P = .007) and hazardous cannabis use (β = .058, P = .030).

Conclusions: These results suggest different associations between BMI and the use of various substances by young men. However, only RSOD and NMPDU of anxiolytics predicted BMI, whereas BMI predicted daily smoking and hazardous cannabis use.