Exploring the effect of Colorado's recreational marijuana policy on opioid overdose rates.

OBJECTIVES

Opioid overdose death rates have continued to spike exponentially from the start of the 21st century, creating what is known to be one of the worst public health crises in the United States. Simultaneously, as more states began passing medical cannabis laws (MCLs), the idea that marijuana was the solution to the opioid crisis began to spread nationwide. As some states have maintained strict medical marijuana policies, others-such as Colorado-have expanded their statutes to allow recreational marijuana sales within their state. Researchers have been able to provide sense of the public health implications resulting from MCLs, but little is known about the effects of this marijuana policy expansion. This preliminary study will focus on exploring the statewide effects of Colorado's recreational marijuana policy on the state's opioid overdose death rates.

STUDY DESIGN

Because Colorado has existing panel data for opioid overdose death rates, we can use statistical software to define and create an optimal control group to adequately resemble Colorado's outcome variable of interest. This process known as the synthetic control method can provide a valid counterfactual for Colorado's opioid overdose outcomes in the absence of this policy-a Colorado that did not expand marijuana policy to the point recreational dispensaries were established.

METHODS

Opioid overdose death rate data from the Centers for Disease Control and Prevention's Wide-ranging Online Data for Epidemiologic Research (WONDER) will be used to construct a synthetic control unit composed of a donor pool of states resembling Colorado's regulatory environment pertaining to marijuana before legalization. The synthetic control unit allows for a comparative observation of overdose rate trends in Colorado and its synthetic counterpart for the years 1999-2017, all while including a set of predictor variables for robustness checks. A difference-in-difference estimate will then help us observe the effects of the treatment given to Colorado. Inference tests will be conducted to evaluate the method's predictive power and significance of the results.

RESULTS

The results of the synthetic control model and its outcomes showed that the estimated negative 5% drop in overdose death rates was deemed insignificant on conducting a placebo in-space analysis, meaning there is not enough evidence to prove that opening recreational dispensaries as a result of recreational marijuana legislation was instrumental in reducing Colorado's ongoing opioid crisis depicted through opioid overdose deaths.

CONCLUSION

Owing to the lack of additional post-treatment data and captured lagged effects, it is too soon to dismiss this policy as inadequate in combating the opioid epidemic. Once additional post-treatment data become available, the study can be reproduced to obtain more robust results and achieve a clearer understanding of the policy implications shown.