OBJECTIVE

To establish whether nonpharmacologic interventions, such as occupational and physical therapy, were associated with a shorter duration of prescription opioid use after hip or knee arthroplasty.

DESIGN

This retrospective cohort study used data from a national 5% Medicare sample database between January 1, 2010 and December 31, 2015.

SETTING

Home health or outpatient.

PARTICIPANTS

Adults 66 years or older with an inpatient total hip (n=4272) or knee (n=9796) arthroplasty (N=14,068).

INTERVENTIONS

We dichotomized patients according to whether they had received any nonpharmacologic pain intervention within 1 year after hospital discharge (eg, occupational or physical therapy evaluation). Using Cox proportional hazards, we treated exposure to nonpharmacologic interventions as time dependent to determine if skilled therapy was associated with duration of opioid use.

MAIN OUTCOME MEASURES

Duration of prescription opioid use.

RESULTS

Median time to begin nonpharmacologic interventions was 91 days (95% confidence interval [CI], 74-118d) for hip and 27 days (95% CI, 27-28d) for knee arthroplasty. Median time to discontinue prescription opioids was 16 days (hip: 95% CI, 15-16d) and 30 days (knee: 95% CI, 29-31d). Nonpharmacologic interventions delivered with home health increased the likelihood of discontinuing opioids after hip (hazard ratio [HR], 1.15; 95% CI, 1.01-1.30) and knee (HR, 1.10; 95% CI, 1.03-1.17) arthroplasty. A sensitivity analysis found these estimates to be robust and conservative.

CONCLUSIONS

Occupational and physical therapy with home health was associated with a shorter duration of prescription opioid use after hip and knee arthroplasty. Occupational and physical therapy can address pain and sociobehavioral factors associated with postsurgical opioid use.