Predictors of voter support for the legalization of recreational cannabis use and supply via a national referendum

Uneven effects of twenty years of Chile's cannabis policy implementation in cannabis onset

BACKGROUND

In Chile, Laws 19366 and 20000, implemented in 1995 and 2005 respectively, regulated and sanctioned cannabis' personal use, cultivation and trafficking.

METHODS

We use thirteen biannual cross-sectional national surveys data from 1994 to 2018 to examine the effect of Laws 19366 and 20000-using the rate of individuals incarcerated per 100000 population due to drug-related crimes as proxy-on the age of onset of cannabis use over time. We estimate the effect of these policies using a mixed proportional hazards framework that models the transition to first cannabis use in 47,832 individuals aged 12-21.

RESULTS

Overall, changes in these laws did not affect the transition to first cannabis use. However, increases in the rate of individuals incarcerated were associated with decreases on the age of onset of cannabis use in females and individuals living in affluent neighborhoods or in specific regions.

CONCLUSION

We find no evidence of cannabis policy changes affecting the age of onset of cannabis use across all individuals aged 12-21. Policy effects associated with decreases in cannabis onset age in females and individuals from affluent neighborhoods or specific regions can be explained by using theoretical frames that recognize specific dynamics of cannabis supply and demand.

Using publicly available data to predict recreational cannabis legalization at the county-level: A machine learning approach

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.