Methods/Statistics
Impact Of Marijuana Legalization on Alcohol Consumption and Alcohol Use Disorder: Substitution, Complementarity, or Both? Rikki Fan* Rikki Fan Aaron White
Background. Marijuana policies are rapidly evolving. The legalization policy may potentially influence marijuana use. In addition, marijuana policies may impact use and consequences of alcohol and simultaneous use of alcohol and marijuana (SAM) due to potential spillover effects.
Methods. State-level aggregate data from National Survey on Drug Use and Health (NSDUH) 2002-1018 were used to conduct pretest-posttest control group design analysis. The nine states with passage of recreational marijuana legalization (RML) between 2012 to 2016 were the treatment group, the 32 states who had never passed RML up to 2021 served as the control group. Two years preceding and two years succeeding the institution of RML were designated as the points where pretest (baseline) and posttest (follow-up) values were extracted respectively. The following measures were considered: 30-day marijuana use; initiation of marijuana past year; alcohol use past month; and alcohol use disorder past year; recent simultaneous use of alcohol and marijuana. The analysis was stratified by three age groups, i.e., 12-17 years, 18-25 years, and 26 years or older. The mixed regression model was constructed to examine RML effects with adjustment for pretest, cohort effect, and status of medical marijuana legalization (MML).
Results. States instituting RML exhibited higher baseline of 30-day prevalence and 1-yr incidence of marijuana use which were also associated with higher proportion of states who had previously instituted MML in RML states compared to non-RML states. Although marijuana use increased in both RML-states and non-RML states, RML-states exhibited increases to a greater magnitude for all age groups. Among teenagers, the most recent SAM decreased among non-RML states over time. Among adults, the most recent SAM increased among both non-RML and RML states over time; after controlling for the baseline, the magnitude of increase was higher among RML states than that among non-RML states.
Conclusion. The further increased marijuana use after RML institution was superimposed over the effect of MML. The relatively higher rate of SAM among RML-instituted states appear to be mainly attributed to increased marijuana use. Neither complementarity nor substitution hypothesis is supported in the context of liberalization of marijuana policies.