College of Health

Highlights

• Emerging adults may be especially sensitive to recreational marijuana legalization (RML)
• Substance use trends from 2008 to 2018 were investigated in a national college sample
• RML was linked to decreased binge drinking prevalence among students age 21 and older
• RML was associated with increased sedative misuse among minors.
• RML effects are not limited to marijuana use, and depend on substance and age group.

BACKGROUND: Young adult college students may be particularly sensitive to recreational marijuana legalization (RML). Although evidence indicates the prevalence of marijuana use among college students increased after states instituted RML, there have been few national studies investigating changes in college students' other substance use post-RML.

METHOD: The cross-sectional National College Health Assessment-II survey was administered twice yearly from 2008 to 2018 at four-year colleges and universities. Participants were 18-26 year old undergraduates attending college in states that did (n = 243,160) or did not (n = 624,342) implement RML by 2018. Outcome variables were self-reported nicotine use, binge drinking, illicit drug use, and misuse of prescription stimulants, sedatives, and opioids. Other variables included individual and contextual covariates, and institution-reported institutional and community covariates. Publicly available information was used to code state RML status at each survey administration.

RESULTS: Accounting for state differences and time trends, RML was associated with decreased binge drinking prevalence among college students age 21 and older [OR (95% CI) = 0.91 (0.87 - 0.95), p < .0001] and increased sedative misuse among minors [OR (95% CI) = 1.20 (1.09 - 1.32), p = .0003]. RML did not disrupt secular trends in other substance use.

CONCLUSIONS: In the context of related research showing national increases in college students' marijuana use prevalence and relative increases following state RML, we observed decreases in binge drinking and increases in sedative use that both depended on age. Findings support some specificity in RML-related changes in substance use trends and the importance of individual factors.