Background/Purpose: After total knee arthroplasty (TKA), most patients receive community-based rehabilitation in the home health (HH) and/or outpatient (OP) settings. The purpose of this study was to examine the relationships between sociodemographic characteristics and functional outcomes within the HH and OP rehabilitation settings after TKA, and the extent to which rehabilitation utilization modifies those relationships.

Methods: We analyzed 5 years of data (2017-2021) for patients who received TKA within a large health system and post-operatively received HH (n = 4,816) and/or OP rehabilitation (n = 2,103). We used regression models to examine main effects of 4 sociodemographic variables (race, sex, area deprivation, and rurality) on functional improvement achieved within HH and OP rehabilitation. In HH, function was measured using a composite measure of the activities of daily living items from the OASIS-D. In OP rehabilitation, function was measured using the patient-reported Knee Outcome Survey. For each regression model, we performed unadjusted analyses and then adjusted for age, hospital, hospital length of stay, non-home discharge, marital status, and Elixhauser comorbidity index. Finally, we added an interaction term to each regression model, examining the effects of the interaction of each of the 4 sociodemographic variables with rehabilitation utilization on the functional outcomes within that setting. Our measure of rehabilitation utilization was “intensity” of care, calculated as the number of visits divided by the number of days in the HH or OP episode. 

Results: HH: In both unadjusted and adjusted models, we found that patients who were male sex (p< .0001 for both models) and White Non-Hispanic race/ethnicity (p< .0001 for both models) had significantly better functional outcomes at discharge from HH than female patients and Black Non-Hispanic patients. We found no significant associations between area deprivation or rurality and HH functional outcomes. Intensity of HH was not a significant predictor of function. When adding interaction terms, we did not find any significant effects of the interactions between sociodemographic variables and HH utilization on function.

OP: There was a significant main effect of OP intensity as a predictor of function (p=.001). Male (p< .0001 for both models) and White Non-Hispanic (p = .008 unadjusted model, p = .035 adjusted model) patients had greater improvement on the Knee Outcome Survey than female and Black Non-Hispanic patients. Higher area deprivation (p = .021) and rurality (p = .033) were both associated with poorer function in unadjusted models, but not after adjusting for demographic and clinical variables. Additionally, the interactions of OP intensity with area deprivation (p=.037) and rurality (p=.019) were significant predictors of function.

Conclusion: After TKA, men and non-Hispanic White patients demonstrate better functional outcomes within the HH and OP settings. The roles of area deprivation and rural residence in post-TKA function are less clear. The intensity of rehabilitation received has minimal influence in functional outcomes in HH but may be of higher importance in OP rehabilitation.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/disparities-in-rehabilitation-outcomes-after-knee-arthroplasty/