Temporal trends in opioid prescription fills following cancer-directed surgery.

Abstract

11099 Background: Postoperative pain is a common complication of cancer surgery. Opioid prescribing has declined dramatically since the early 2010’s; however, little is known about trends in opioid prescribing following cancer-directed surgeries. Methods: Using administrative data for 100% Medicare fee-for-service beneficiaries enrolled in parts A, B, and D, we identified initial episodes of cancer-directed surgeries from 2012-2021 among adults who survived > 30d after surgery and were discharged home. We used Part D claims to identify early postop opioid fills, defined as prescriptions filled in the 30d after surgery for outpatient surgeries, and in the 30d following hospital discharge for inpatient surgeries. We described opioid fills, median dose, pill counts, and the proportion with subsequent fills overall and annually. Results: We identified 981,702 episodes of cancer directed surgeries, most often for breast (38%), colorectal (15%), prostate (13%), and lung (10%) cancers. Patients’ mean age was 73(SD,8), 36% were male, 83% White, 8% Black, 4% Hispanic, and 2% Asian. Overall, 66.6% of patient episodes had ≥1 postoperative opioid fill, declining from 70.7% in 2012 to 59.3% in 2021. Among episodes with ≥1 opioid fill, the median dose of the first opioid fill declined from 225 (IQR:150,300) morphine milligram equivalents (MMEs) to 100 (IQR:75,150) MMEs between 2012-2021, and the total dose of all fills in 30d fell from 250 (IQR:150,450) MMEs to 112.5 (IQR:75,210) MMEs. Median pill counts for the first short-acting opioid fill declined from 30(IQR:30,40) to 18(IQR:10,25) over the study. Among episodes with ≥1 short-acting opioid fill, the proportion with subsequent short-acting opioid fills declined from 30.7% to 17.8%. Conclusions: Medicare beneficiaries undergoing cancer-directed surgeries have experienced steep declines in prescription opioid medication fills in the postoperative period. Future work is needed to understand potential impacts of declining postoperative opioid prescribing on outcomes such as uncontrolled pain, pain-related emergency department visits, persistent opioid use, and development of opioid use disorders. Support: R01CA279414.