Recreational Cannabis Legalization Was Linked to Lower Retail Codeine Distribution

A 2023 study in Health Economics found that states adopting recreational cannabis laws experienced a substantial decline in codeine distribution through retail pharmacies. The analysis used data from the Drug Enforcement Administration’s Automation of Reports and Consolidated Orders System, or ARCOS, which tracks shipments of controlled substances to pharmacies, hospitals and other recipients. Researchers Shyam Raman, Johanna Catherine Maclean, W. David Bradford and Coleman Drake examined opioid distribution in all 50 states and the District of Columbia from 2010 through 2019. Their analysis compared 11 states that adopted recreational cannabis laws with states that had not yet done so, while accounting for other opioid-related policies and state characteristics. After recreational cannabis laws took effect, codeine distribution to retail pharmacies fell by an estimated 25.7%. The decline grew over time, reaching approximately 37% four years after implementation. However, the study did not find statistically significant changes in the distribution of opioids overall or in hydrocodone, morphine or oxycodone specifically. That distinction is important. ARCOS measures shipments to pharmacies and other facilities—not individual prescriptions, patient behavior, opioid misuse or overdose deaths. The study was also observational, so it identified an association rather than proving that cannabis legalization caused the decline. The authors suggested that the result may reflect reduced demand for codeine, which is more frequently used nonmedically than some other prescription opioids, but they did not establish that cannabis replaced opioids or reduced opioid use disorder. The findings are broadly consistent with some earlier research, although estimates vary by population and outcome. A 2019 study of more than 1.5 billion opioid prescriptions reported lower prescribing after recreational and medical cannabis laws, while a JAMA Internal Medicine study of Medicaid data found that adult-use cannabis laws were associated with a 6.38% reduction in opioid prescriptions for Medicaid enrollees. These studies used different data sources and methods, so their results are not directly interchangeable. Other research has produced more limited or mixed conclusions. For example, a 2019 JAMA Network Open analysis found little evidence that medical marijuana laws reduced individual-level nonmedical prescription opioid use or opioid use disorder. Together, the literature suggests that cannabis policy may influence some forms of opioid prescribing or distribution, but it does not yet demonstrate a broad reduction in opioid misuse, addiction or overdose. The Health Economics study therefore offers a potentially important public-health signal, not a definitive answer. As more states adopt or modify cannabis laws, researchers will need to examine patient-level prescribing, nonmedical use, treatment access, illicit opioid use and overdose outcomes to determine whether lower pharmacy distribution translates into measurable improvements in public health.