Recreational marijuana dispensaries showed a trend toward lower opioid deaths, but results were not significant

The average treatment effect was approximately -6.49 opioid deaths per capita for Colorado, -2.89 for Washington, and -4.8 for Oregon after recreational dispensaries opened. However, none of these reductions reached statistical significance. Robustness checks confirmed the consistent negative direction but continued non-significance.

Researchers used the synthetic control method to construct counterfactual versions of Colorado, Washington, and Oregon using state-level data from the CDC, Bureau of Labor Statistics, US Census Bureau, and other sources. These three states were chosen because they were the first to legalize recreational marijuana, providing the longest post-treatment observation periods.

The opioid epidemic remains a major public health crisis. Whether legal cannabis access reduces opioid deaths is a high-stakes policy question. This study finds a suggestive but inconclusive downward trend, adding nuance to claims on both sides of the debate.

The consistent direction of effect across three independent state analyses is notable even though no individual result was significant. This pattern is common in policy research where effects are real but modest relative to the noise in state-level data. The findings neither confirm nor refute the hypothesis that cannabis access reduces opioid deaths.

Synthetic control method relies on finding good donor states, which may be imperfect. Only three states were analyzed. Opioid death rates are influenced by many concurrent policy changes (naloxone access, prescribing guidelines, fentanyl emergence) that complicate attribution. The post-treatment period may not be long enough to capture delayed effects.