Stricter State Cannabis Laws Linked to Less Cannabis Use Among College Students | Research

The takeaway

More restrictive state cannabis policies were associated with lower rates of cannabis use and cannabis-alcohol co-use among college students, while stricter alcohol policies independently reduced drinking.

Campus health administrators, state legislators evaluating cannabis policy, public health researchers studying young adult substance use.

What the researchers found

Drawing on survey data from over 900,000 college undergraduates (ages 18–24) across 591 four-year institutions in 47 states between 2008 and 2019, researchers examined how state-level cannabis and alcohol policy environments related to substance use patterns.

Using a Cannabis Policy Scale (sum of 17 policies weighted by efficacy) and an Alcohol Policy Scale (29 policies), the analysis found that more restrictive cannabis policy environments were associated with significantly lower odds of any cannabis use (OR 0.97), frequent cannabis use (OR 0.93), and co-use of cannabis with binge drinking (OR 0.94). Cannabis policy restrictiveness was not significantly associated with alcohol-only outcomes.

Conversely, more restrictive alcohol policy environments were associated with lower odds of alcohol use outcomes but did not significantly affect cannabis use. The two policy domains operated independently — each influenced its target substance without significant cross-substance effects.

Why it matters

This is one of the largest studies of how cannabis policy affects a high-risk population. College students are in a critical developmental period and have among the highest cannabis use rates of any age group. The finding that policy restrictiveness has measurable effects on use — even after accounting for alcohol policy — provides evidence that state-level cannabis regulations have real behavioral consequences.

The numbers in context

902,486 undergraduates, 591 institutions, 47 states, 2008–2019. Cannabis policy effects (per unit increase in restrictiveness): any use OR 0.97, frequent use OR 0.93, co-use OR 0.94. Cannabis policy not significantly associated with alcohol outcomes. Alcohol policy not significantly associated with cannabis outcomes.

How the study worked

Cross-sectional survey of 902,486 undergraduates (ages 18–24) from 591 four-year institutions in 47 states, 2008–2019. Time-varying state-level Cannabis Policy Scale (17 policies) and Alcohol Policy Scale (29 policies) scored by efficacy and implementation. Outcomes: 30-day cannabis use, frequent use (20+ days), 30-day alcohol use, 2-week binge drinking (5+ drinks), and co-use. Multilevel models accounting for institution and state clustering.

Who was studied

N=902,486 undergraduates aged 18-24, 47 states, US national survey

What this study cannot tell us

Cross-sectional survey design cannot establish that policies caused the use differences. Self-reported substance use likely underestimates actual use. The 2008–2019 period captures the early legalization era; effects may differ as legal markets mature. The Cannabis Policy Scale weights policies by estimated efficacy, which involves subjective judgment. Four-year college students are not representative of all 18–24-year-olds.

How to read the evidence

Large cross-sectional study with sophisticated multilevel modeling — strong associational evidence from an enormous sample, but the observational design limits causal conclusions.

When this study was published

Published in 2026 with data through 2019. The cannabis policy landscape has changed substantially since then, with many more states legalizing recreational use.

The bigger picture

This study provides the behavioral evidence that connects to the tax standardization work (RTHC-00247) and the California retail authorization study (RTHC-00248). Together they sketch a picture: state cannabis policies measurably influence who uses cannabis and how much, retail access adds a local dimension, and tax structures provide a specific policy lever. For the college-age population specifically, the co-use finding is important — cannabis policy appears to reduce not just cannabis use but the particularly risky combination of heavy drinking and frequent cannabis use.

Replication

Not stated in abstract.

Funding

Not reported in abstract.

Conflicts of interest

Not reported in abstract.

Questions still open

Which specific policies within the 17-component cannabis scale drive the strongest effects? Do these associations hold for non-college young adults? As more states legalize, will the policy-use association weaken as cannabis becomes more culturally normalized regardless of policy?

Read the original research

State cannabis and alcohol policy environments: Associations with college students' use of cannabis, alcohol and both substances.

American journal of preventive medicine, 108275

The American Journal of Preventive Medicine is a reputable journal focusing on public health and preventive medicine research.

View on PubMed ↗Publisher’s paper ↗

Citation

Kerr, David C R; Bae, Harold; Naimi, Timothy S; Subbaraman, Meenakshi S; Hummel, Haley M; Lira, Marlene C. (2026). State cannabis and alcohol policy environments: Associations with college students' use of cannabis, alcohol and both substances.. American journal of preventive medicine, 108275. https://doi.org/10.1016/j.amepre.2026.108275