Article Title

Reducing Opioid Exposure Following Hip and Knee Surgery: Results of a Randomized, Pragmatic, Pharmacist-Led Intervention

Publication Date



pharmacy, quality of care, clinical trials, pharmaceuticals, costs


Background: Opioids have a role in managing pain for orthopaedic patients, but there is compelling evidence that this exposure should be kept as short as possible. Setting patient expectations for pain and its treatment has been recommended for orthopaedic surgeons as part of minimizing opioid exposure.

Methods: Patients scheduled to undergo total hip arthroplasty (THA) or total knee arthroplasty (TKA) were randomized, weekly, to usual care or intervention. Intervention patients received mailed materials 2 weeks before and after surgery, plus telephone intervention from specially trained pharmacists if they filled opioid prescriptions in the 28–90 days following surgery. Materials were developed using qualitative methods involving patients and orthopaedic clinicians. In order to direct resources toward the riskiest patients, we ranked patients according to predicted risk of persistent opioid use and selected the top 60%. Our primary outcome was the morphine equivalents (MEQ) dispensed in the 90 days following surgery, modeled using a natural log transformation (lnMEQ). We report the relative percentage decrease in geometric means between study groups, with a P < 0.05 cut-point for statistical significance. Our a priori analytic plan specified testing for effect modification on treatment group by: 1) THA or TKA, 2) baseline quartile of opioid use, and 3) quartile of predicted opioid persistence. Informed consent was waived, consistent with pragmatic trial principles.

Results: A total of 561 patients were randomized (286 usual care, 118 THA, 275 intervention, 107 THA); mean age was 66 years and 60% were female. The intervention failed to reduce lnMEQ in the entire surgical population (P > 0.50), but the intervention reduced lnMEQ by a larger percentage in the THA patients than the TKA patients (P < 0.003). Among THA patients, those randomized to intervention used 62% less MEQ (P < 0.003) than usual care. We found no statistically significant difference in opioid use among TKA patients (P > 0.1). We found no effect modification by baseline opioid use or predicted opioid persistence.

Conclusion: Our pharmacist-led intervention to reduce opioid use had clinically meaningful impact on the use of opioids in the 90 days following THA, but not TKA.




June 22nd, 2017


August 10th, 2017