Medical Cannabis and Opioid Prescription Discontinuation Among Chronic Pain Patients on Long-term Opioid Therapy.

Nguyen T.; Li Y.; Greene D.; Stancliff S.; Rosa N.; Rosenberg E.; Kelly K. · Journal of addiction medicine · 2026

**Background & Methods** This was a retrospective cohort study using New York State Medical Cannabis Program and Prescription Monitoring Program data (2017-2019) examining 8,358 chronic pain patients on long-term opioid therapy. Patients were stratified by medical cannabis use duration (>30 days vs. ≤30 days) and followed for 90 days, with opioid discontinuation defined as a ≥45-day break in prescriptions. Multivariate time-to-event analyses were stratified by baseline morphine milligram equivalent (MME) dosage. **Key Findings** - Medical cannabis users demonstrated significantly higher opioid prescription discontinuation rates (15.5%) compared to non-users (7.6%), representing a 2-fold increased likelihood (P<0.001). - Discontinuation risk varied substantially by baseline opioid dosage: patients with baseline MME <50 showed the most dramatic early effect (adjusted hazard ratio [aHR] 2.87 in first 30 days, declining to 1.81 thereafter), while those with MME 50-<90 maintained elevated risk throughout follow-up (aHR 3.13). - No significant association was observed among higher-dose patients (baseline MME ≥90), suggesting limited effect of medical cannabis in reducing opioids among those with established high-dose therapy. **Dosage & Administration** Not reported. The study classified cannabis use based on duration (>30 days vs. ≤30 days) rather than specific doses or routes of administration. **Safety & Adverse Effects** Not reported. The abstract does not discuss adverse effects, safety profile, or drug interactions between medical cannabis and opioids. **Evidence Quality** This retrospective observational study has moderate evidence limitations. While the large sample size (N=8,358) and stratified analysis by dosage provide valuable real-world data, the lack of randomization prevents causal inference. The study cannot distinguish between medical cannabis facilitating opioid discontinuation versus selection bias (patients motivated to discontinue being more likely to initiate cannabis). The brief 90-day follow-up period limits assessment of long-term sustainability of opioid discontinuation. Generalizability may be limited to New York State populations. The study would benefit from prospective randomized controlled trials to establish causation and optimal patient selection criteria.

More recent cannabis research