Recreational Cannabis Laws Were Linked to Reduced Codeine Dispensing Across US Pharmacies

Using two-way fixed-effects regressions across 11 states that adopted recreational cannabis laws between 2010 and 2019, the study found RCLs led to a reduction in codeine dispensed at retail pharmacies. No significant effects were found for other opioid types. The analysis accounted for opioid prescribing limits and other policies.

Two-way fixed-effects regression using variation from 11 US states that adopted recreational cannabis laws between 2010 and 2019, examining dispensing data across all payers and endpoints.

Unlike prior studies that examined prescriptions from specific payers, this study captured all retail pharmacy dispensing. The finding that codeine, the opioid most associated with non-medical use, specifically declined suggests cannabis may substitute for recreational or quasi-medical opioid use.

The specificity of the codeine finding is telling. Codeine is often used for mild pain and is the most accessible prescription opioid for non-medical use. That cannabis appears to substitute specifically for codeine rather than stronger opioids suggests the substitution effect may be strongest where the clinical need is lowest.

Observational policy analysis cannot confirm causation. Cannot distinguish between recreational and medical substitution. State-level data may mask variation within states. Time period may not capture long-term effects.