Search

Background: There is substantial geographic variability in local cannabis policies within states that have legalized recreational cannabis. This study develops an interpretable machine learning model that uses county-level population demographics, sociopolitical factors, and estimates of substance use and mental illness prevalences to predict the legality of recreational cannabis sales within each U.S. county., Methods: We merged data and selected 14 model inputs from the 2010 Census, 2012 County Presidential Data from the MIT Elections Lab, and Small Area Estimates from the National Surveys on Drug Use and Health (NSDUH) from 2010 to 2012 at the county level. County policies were labeled as having recreational cannabis legal (RCL) if the sale of recreational cannabis was allowed anywhere in the county in 2014, resulting in 92 RCL and 3002 non-RCL counties. We used synthetic data augmentation and minority oversampling techniques to build an ensemble of 1000 logistic regressions on random sub-samples of the data, withholding one state at a time and building models from all remaining states. Performance was evaluated by comparing the predicted policy conditions with the actual outcomes in 2014., Results: When compared to the actual RCL policies in 2014, the ensemble estimated predictions of counties transitioning to RCL had a macro f1 average score of 0.61. The main factors associated with legalizing county-level recreational cannabis sales were the prevalences of past-month cannabis use and past-year cocaine use., Conclusion: By leveraging publicly available data from 2010 to 2012, our model was able to achieve appreciable discrimination in predicting counties with legal recreational cannabis sales in 2014, however, there is room for improvement. Having demonstrated model performance in the first handful of states to legalize cannabis, additional testing with more recent data using time to event models is warranted., Competing Interests: Declaration of competing interest The authors have no conflicts to declare., (Copyright © 2024 Elsevier B.V. All rights reserved.)

The cannabis legalization movement in the United States has experienced unprecedented success in the past decade due to a wave of grassroots reforms in states across the country. The current legalization movement began in 2012, when Colorado and Washington became the first two states to legalize the use and sale of cannabis for adults aged ≥ 21 years. Since then, the use of cannabis has been legalized in 21 states, Guam, the Northern Mariana Islands, and Washington, DC. Many of these states have explicitly framed the law change as a rejection of the War on Drugs and its harms, felt disproportionately in Black and Brown communities. However, racial inequities in cannabis arrests have increased in states that have legalized cannabis for adult use. Moreover, states working to implement social equity and community reinvestment programs have made little progress toward their goals. This commentary describes how US drug policy, racist in intention, gave way to drug policy that perpetuates racism, even when its ostensible or stated goal is equity. As the United States prepares for national legalization of cannabis, it is critical that we break away from past legislation and mandate equity in cannabis policy. Developing meaningful mandates will require us to acknowledge our history of using drug policy for the racist goals of social control and extortion, study the experience of states that are trying to implement social equity programs, listen to Black leaders and other leaders of color who have developed guidance for equity-focused cannabis policy, and commit to a new paradigm. If we are willing to do these things, we may be able to legalize cannabis in an anti-racist way that will stop causing harm and enable us to effectively implement reparative practices., Competing Interests: Declaration of Interest RTI International has no role in our decision to submit this manuscript. The views expressed are those of the authors alone and do not represent the views of RTI International., (Copyright © 2023 The Authors. Published by Elsevier Inc. All rights reserved.)

Liberalized state-level recreational cannabis policies in the United States (US) fostered important policy evaluations with a focus on epidemiological parameters such as proportions [e.g., active cannabis use prevalence; cannabis use disorder (CUD) prevalence]. This cannabis policy evaluation project adds novel evidence on a neglected parameter-namely, estimated occurrence of newly incident cannabis use for underage (<21 years) versus older adults. The project's study populations were specified to yield nationally representative estimates for all 51 major US jurisdictions, with probability sample totals of 819,543 non-institutionalized US civilian residents between 2008 and 2019. Standardized items to measure cannabis onsets are from audio computer-assisted self-interviews. Policy effect estimates are from event study difference-in-difference (DiD) models that allow for causal inference when policy implementation is staggered. The evidence indicates no policy-associated changes in the occurrence of newly incident cannabis onsets for underage persons, but an increased occurrence of newly onset cannabis use among older adults (i.e., >21 years). We offer a tentative conclusion of public health importance: Legalized cannabis retail sales might be followed by the increased occurrence of cannabis onsets for older adults, but not for underage persons who cannot buy cannabis products in a retail outlet. Cannabis policy research does not yet qualify as a mature science. We argue that modeling newly incident cannabis use might be more informative than the modeling of prevalences when evaluating policy effects and provide evidence of the advantages of the event study model over regression methods that seek to adjust for confounding factors., Competing Interests: The authors have declared that no competing interests exist.