Geographic Variability in Discharge Setting and Outpatient Postacute Physical Therapy After Total Knee Arthroplasty: A Retrospective Cohort Study

Background: Despite the frequency of total knee arthroplasty (TKA) in the Medicare population, little is known about the use of postacute physical therapy among those discharged to home.

Objective: The objectives of this study were to explore factors associated with geographic variability in discharge disposition and outpatient physical therapy utilization for Medicare patients after TKA discharged to home/self-care.

Design: The design of the study was a retrospective cohort study.

Methods: Medicare patients with TKA discharged alive from July 1, 2010, to June 30, 2011, with discharge disposition to home/self-care (HSC), home health agency (HHA), inpatient rehabilitation facility (IRF), or skilled nursing facility (SNF) were selected. Geography was measured with Census region. Outpatient physical therapy utilization was calculated from Medicare Part B claims. Odds ratios for discharge disposition and adjusted means for physical therapy utilization variables by Census region were calculated, accounting for county-clustered data and adjusting for demographics, clinical, and environmental characteristics.

Results: There was significant variation with discharge destination by Census region among 18,278 patients. With discharge disposition analysis, the patients from the West region who were discharged home were the referent group. The patients from the South and Northeast regions had higher odds for discharge to HHAs (adjusted odds ratio [95% CI = 1.80 [1.48-2.19] and 2.20 [1.70-2.84]), SNFs (1.34 [1.08-1.66] and 4.42 [3.38-5.79]), and IRFs (2.36 [1.80-3.09] and 8.83 [6.41-12.18]). For those discharged to HSC, 40.4% received outpatient physical therapy within 4 weeks. Significant differences were found with time to first physical therapy visit (Midwest <South <[West = Northeast]) and length of therapy episode, but not with the number of therapy visits by geographic region.

Conclusions: Geographic region was associated with discharge setting, postacute physical therapy, and outpatient therapy utilization in Medicare beneficiaries after TKA. Differences in outcomes of outpatient therapy should be assessed to better describe the impact of geographic variation in care.