Association between fatal opioid overdose and state medical cannabis laws in US national survey data, 2000-2011

Most information on the relationship between medical cannabis laws (MCL) and the risk for opioid overdose fatality has been based on studies with ecological designs. To contribute additional information, we used a novel case-control design and individual-level data from national surveys to assess whether state medical cannabis laws were associated with reduced risk of fatal opioid overdose between 2000-2011.

Data from participants surveyed in the National Health Interview Survey (NHIS) between 1986-2011 were included. For those sampled between 1986-2009, detailed mortality follow-up data were available from the National Death Index up to 12/31/2011. Opioid overdose decedents (n = 791) were classified as cases. Between 2000-2011, all cases arising in a given year were matched to adult controls who were surveyed the same year and eligible for mortality follow-up (n = 723,920). The distribution of exposure to state MCL was contrasted between cases and controls, providing an approximation of the rate ratio of fatal opioid overdose associated with MCLs. Due to a NHIS sample redesign, we stratified analysis using timeframes before and after 2005.

Overall, compared to controls, cases were more likely to be male, middle-aged, non-Hispanic White, separated/divorced; less educated, and have a family income below the poverty threshold. No overall association between state MCLs and the rate of opioid overdose was observed between 2000-2005 (aOR = 1.22, 95% CI: 0.83-1.79) or between 2006-2011 (aOR = 0.87, 95% CI: 0.60-1.25). No significant difference between sampling timeframes was observed (ratio of aOR's = 0.71, 95% CI: 0.49-1.01).

We found no overall protective relationship between state MCLs and opioid overdose. Future research with more recent mortality data and more refined cannabis policy classifications would be useful. The importance of the study is two-fold. First, the findings provide an additional source of information countering claims of a protective effect of MCLs on opioid overdoses, suggesting that other solutions to the opioid overdose crisis are needed. Second, the study offers a potentially useful design to answer important population-level public health questions.