Prescription Opioids Higher Among Knee Arthroplasty Recipients Randomized to Inpatient Rehabilitation

OBJECTIVE: To determine whether the purchase of prescription opioids was lower among people randomized to inpatient rehabilitation (IR) compared with those discharged directly home following total knee arthroplasty (TKA). METHOD: A secondary analysis of a previous clinical trial in which participants were randomized 3 to 5 days after ‐surgery to 10 days of IR and a home program or to a home program alone. The primary outcome for this secondary analysis was the purchase of opioid‐based pain relief up to 10‐weeks after surgery, which was captured via patient diaries. Between‐group differences were analyzed using a χ(2) test and relative risk (RR) (95% confidence interval [CI]). We report this outcome alongside the main outcomes observed at 10 weeks for the original study (6‐minute walk test, index joint pain, and function) for context. RESULTS: At 10 weeks, 158 participants were available for follow‐up; 120 (76%) provided diaries, with 113 providing generic or brand names for the pain relief purchased. In the IR group, 60% (33/55) reported the purchase of opioid‐based medications after discharge compared with 34% (20/58) in the home group (χ(2) = 7.4; P = 0.007); thus, the risk of purchasing opioids for those in the IR group was almost double (RR, 1.7 [95% CI, 1.1‐2.6]). No significant or meaningful between‐group differences in index joint pain, function, or mobility were observed. CONCLUSION: Contrary to what was hypothesized, IR is a strong driver of opioid purchase after discharge from the hospital following TKA.