Notable Variability in Opioid-Prescribing Practices After Common Orthopaedic Procedures

Sophia A. Traven, MD; Daniel L. Brinton, PhD; Shane K. Woolf, MD; Lee R. Leddy, MD; Michael B. Gottschalk, MD; Harris S. Slone, MD


J Am Acad Orthop Surg. 2021;29(5):219-226. 

In This Article


Study Design

A retrospective analysis of claims data for privately insured subjects from the Truven Health MarketScan database for the years 2015 and 2016 was conducted. This administrative claims database collects health insurance claims across the continuum of care (eg, inpatient, outpatient, and outpatient pharmacy) and enrollment data from more than 100 large employers and health plans across the United States who provide private healthcare coverage for more than 250 million patients. This database also includes a variety of fee-for-service, preferred provider organizations, and capitated health plans.

Patients were identified by Current Procedural Terminology (CPT) codes for 10 of the following most common orthopaedic procedures: carpal tunnel release (29848 and 64721); ACL reconstruction (29888); arthroscopic meniscectomy (29880 and 29881); bimalleolar ankle fracture ORIF (27814); distal radius fracture ORIF (25607, 25608, and 25609); arthroscopic RCR (29827); single-level ACDF (22551); and primary total shoulder (23472), hip (27130), and knee arthroplasties (27447). Patients were included if they had continuous insurance enrolment for 1 year before their procedure and 6 months postoperatively to allow for the identification and quantification of opioid prescriptions. Only opioid-naive patients were eligible for inclusion in the study. Any patient who had previously filled an opioid prescription between 365 and 14 days before their procedure was therefore excluded from the study. An opioid prescription filled between 0 and 14 days before the procedure was not an exclusionary criterion to account for patients undergoing outpatient surgical fixation of an acute distal radius fracture or ankle fracture. Any patient who underwent a secondary procedure within the postoperative 90-day global period was also excluded.

Statistical Methods

Patients were grouped by procedure and region of residence. MarketScan defines four regions of residence including the Northeast, South, Midwest, and West. The definitions of these regions are included in the Appendix section, Supplemental Digital Content 1, Data on narcotic prescriptions filled during the postoperative period were then collected and converted into OME. Data on the median OME and interquartile ranges (IQRs) prescribed for the initial prescription and 90-day total (which includes initial prescription plus refills) were calculated and reported by procedure and region. Inter-regional prescription variability was calculated using the Kruskal-Wallis test for nonparametric data. The percentage of patients who required refill prescriptions was then compared with the initial prescription amount. A posthoc power analysis was conducted for all non-notable data.