Abstract
PURPOSE: Identifying temporal pattern of associations between bullying perpetration and substance use among youth is useful for optimal targeting of intervention and prevention. We examined the mutual influences on bullying perpetration and substance use and test the cumulative effects of bullying perpetration and substance use over time. METHODS: We used data from the Population Assessment of Tobacco and Health cohort study, a nationally representative sample of youth, on bullying perpetration and substance use at three points in time over 3 years. We analyzed the associations of bullying perpetration with alcohol, cannabis and cigarette use using binary logistic regressions. The 13,068 youths aged 12-17 years at baseline (2013-2014) were included in the analytical sample. RESULTS: Cross-sectional associations between bullying perpetration and substance use were found at baseline (odds ratio [OR](alcohol) = 1.62, 95% confidence interval [CI] 1.33-1.97; OR(marijuana) = 1.28, 95% CI .99-1.66; OR(cigarette) = 1.86, 95% CI 1.44-2.39). Cumulative episodic and frequent bullying perpetration over three data waves predicted an increased likelihood of substance use at follow-up (ORs between 1.18, 95% CI 1.02-1.36 and 2.13, 95% CI 1.66-2.74). Conversely, both episodic and frequent substance use predicted an increased probability of bullying others at follow-up (ORs between 1.50, 95% CI 1.20-1.86, and 2.10, 95% CI 1.69-2.61). CONCLUSIONS: Associations between bullying perpetration and substance use appears to be bidirectional. In addition, episodic bullying perpetrators displayed intermediate levels of risk, suggesting a graded association between bullying perpetration pattern and substance use.