The Effect of Medical Cannabis Laws on the Use of Pain Medications Among Commercially Insured Patients

• In Event: Innovations in Causal Inference: Quantile Mixtures, Empirical Bayes Aggregation, and Synthetic Controls

Abstract

Chronic pain is a major trigger for opioid prescribing in the U.S. Further, as of 2025, 39 U.S. states and the District of Columbia have adopted medical cannabis laws (MCLs), and chronic pain is the most common condition qualifying for medical cannabis access. In this article, using comprehensive claims data of commercially insured individuals, capturing approximately 15 – 20 million patients per year (2007-2020), we assess nationwide changes in opioid analgesic and non-opioid pain medications dispensed in response to state cannabis access policies, using evidence from 22 state-specific synthetic control case-studies. The methodology allows us to construct a stable counterfactual in the pre-policy period for each case and to accommodate heterogeneity of effects across adoption timing. In all except one of the 22 examined states we find reductions in the rate of patients receiving any dispensed prescriptions of opioids (extensive margin) following the legalization of MCLs, several of which are statistically significant at conventional levels. Among treated states, the rate of patients with opioids prescriptions following MCL adoption decreased between 9.7% to 15.8opo% on average relative to the pre-policy baseline. With individual decreases ranging from 10% and up to 28% across statistically significant individual state-cases. This paper extends the current literature, by focusing not only on finding plausibly causal effects, but also further exploring how these changes have emerged by different patient characteristics such as their cancer diagnoses status, or sociodemographics such as gender, age and race.
We also find a significant reduction in the intensive margin, measured as both the average daily supply of dispensed prescriptions and as the average number of dispensed prescriptions per patient. Considering other non-opioid pain medications, we find some evidence of complementarity between cannabis and non-steroidal anti-inflammatory agents (NSAIDs), as the rates of patients prescribed NSAIDs and the average number of NSAID prescriptions per patient dispensed increased (albeit with lower daily supply per prescription) in response to state cannabis legalization, which we interpret as a favorable impact of expanded cannabis access in light of the lower risks associated with NSAID use relative to prescription opioid use. Our study highlights the large positive spillovers of MCLs on prescription opioid utilization among commercially insured Americans across U.S. states. Findings suggest that cannabis may offer an alternative for pain management with opioid analgesics. Thus, policymakers should consider MCLs as an alternative tool for pain management, specifically when other policies that constrain supply of opioids might push some Americans to more harmful substances.

Author

• FELIPE ANDRES LOZANO-ROJAS, University of Georgia
  Presenting Author