Prescription drug monitoring programs and perioperative opioid prescribing and adverse events.

Introduction: Most states mandate clinicians to review prescription drug monitoring program (PDMP) databases before prescribing opioids, but the association with perioperative outcomes is unknown. Methods: We analyzed 2016-2019 Medicare claims for patients ≥65 undergoing surgery in 21 states that enacted PDMP use mandates in 2017-2018 vs states without mandates during 2016-2019. Procedure-level, difference-in-differences models adjusted for patient demographics, prior opioid use, mental health, prior substance use disorders, comorbidities, surgeon specialty, and procedure type. Outcomes included discharge opioid prescriptions, days supplied, total and daily morphine milligram equivalents, high-risk prescribing (≥7-day supply or opioid-benzodiazepine overlap), and 30-day post-discharge adverse events (overdose, emergency department visit, hospitalization, or death). Results: Among 597 455 procedures for 462 290 patients (mean age 73.3 years; 54.5% female), mandates were not associated with changes in opioid prescribing, high-risk prescribing, or adverse events (difference-in-differences estimate: -0.03% points, 95% CI: -1.0, 0.9). Findings were consistent among patients with prior substance use or in states with stringent mandates. Conclusion: PDMP use mandates were not associated with changes in perioperative opioid prescribing or post-surgical adverse events, suggesting that policymakers should consider pairing mandates with evidence-based, procedure-specific prescribing guidelines.