A Population-based Cross-sectional Analysis of Extended-release Opioid Dispensing Incidence, Prognostic Factors, and Variation after Total Joint Arthroplasty

BACKGROUND: Extended-release opioids (EROs) are not recommended for acute postoperative pain, yet their prescribing persists. This study examined the incidence, predictors, and variation in ERO dispensing after total hip and knee arthroplasty. METHODS: This was a population-based cross-sectional study of adults undergoing primary total hip or knee arthroplasty between 2013 and 2022 using linked administrative databases in Ontario, Canada. The primary outcome was fulfillment of an ERO prescription within 7 days of discharge. Multilevel logistic regression estimated associations between patient, surgical, anesthetic, and hospital factors and filling an ERO prescription. Variation was quantified using variance partition coefficients (VPCs) and median odds ratios (MORs) with 95% CIs, based on random effects. RESULTS: Among 258,642 knee and hip arthroplasty procedures, 31,224 (12.1%) patients filled a new ERO prescription after discharge. Male sex (odds ratio [OR], 1.14; 95% CI, 1.09 to 1.19), preoperative opioid exposure (Opioid Naïve-Exposed-Tolerant score, 2; OR, 1.21; 95% CI, 1.15 to 1.27; Opioid Naïve-Exposed-Tolerant score, 3; OR, 1.38; 95% CI, 1.20 to 1.58), and American Society of Anesthesiologists status of III (OR, 1.07; 95% CI, 1.01 to 1.12) increased the odds of filling a new ERO prescription. Neuraxial anesthesia (OR, 0.79; 95% CI, 0.74 to 0.84), peripheral nerve block (OR, 0.84; 95% CI, 0.79 to 0.89), and acute pain service involvement (OR, 0.77; 95% CI, 0.70 to 0.85) were protective against filling a new ERO prescription. Substantial variation was found across hospitals (VPC, 46%; 95% CI, 0.4 to 0.54; and MOR, 9.3; 95% CI, 6.57 to 15.27) and surgeons (VPC, 26%; 95% CI, 0.24 to 0.26; and MOR, 5.3; 95% CI, 4.63 to 6.11), with minimal anesthetist-level variation (VPC, 1%; 95% CI, 0.010 to 0.011; and MOR, 1.4; 95% CI, 1.36 to 1.46). Patient-level factors explained a minority of variation. CONCLUSIONS: One in 10 patients fills an ERO prescription after total hip or knee arthroplasty, a practice with high variation that is predominantly driven by institutional and surgical practice patterns rather than patient factors. Future research should explore institutional stewardship, standardized discharge protocols, and multidisciplinary engagement to reduce unnecessary postoperative exposure to EROs.