Substance Use
Not all legalization is equal: exploring how recreational cannabis regulatory approaches shape cannabis use and risk perceptions in US states and DC Ariadne Rivera-Aguirre* Ariadne Rivera-Aguirre Rivera-Aguirre Rivera-Aguirre Rivera-Aguirre Rivera-Aguirre Rivera-Aguirre NYU Grossman School of Medicine
Recreational cannabis laws (RCLs) in the US vary widely. However, most evaluations treat legalization as a binary exposure or use aggregate scores, obscuring how specific policy combinations influence public health. We examined the effect of distinct RCL approaches on cannabis use and risk perceptions overall (ages 12+) and across adolescents (12-17), young adults (18-25), and adults (26+).
Using latent class analysis (LCA) on 2019-2022 cannabis policy data from the Alcohol Policy Information System, we identified four RCL models: No RCL, Pre-commercial (minimal regulation, no retail framework); Full Access (retail sales, home delivery, and on-site retail use); and Dispensary Access (retail sales for off-site use only and prohibiting public use). Linking these to the 2022–2023 National Survey on Drug Use and Health, we estimated effects on past-year, past-month cannabis use and perceived risk of frequent cannabis use using Longitudinal Targeted Minimum Loss-based Estimation (LTMLE) with Super Learner to account for time-varying confounding. We estimated average treatment effects via pairwise contrasts and evaluated hypothetical shifts between Full Access and Dispensary Access models.
Relative to No RCL, all RCL models were associated with higher cannabis use and lower perceived risk, with the largest prevalence differences (PD) among young adults and adults. However, the magnitude of the differences depended on the regulatory approach, particularly among adults (Dispensary Access vs. Pre-commercial: PD=-3.09; 95%CI: -3.66, -2.52). Effects among adolescents were generally small or null. Hypothetical shifts from Full to Dispensary Access yielded small and directionally consistent reductions in use (CIs included the null).
While RCLs are generally associated with increased adult use and decreased risk perceptions, RCL approaches that restrict where cannabis can be consumed and limit on-site cannabis consumption may prevent wider prevalence differences in cannabis use.
