The Effect of Social Drivers of Health on 90-Day Readmission Rates and Costs After Primary Total Hip and Total Knee Arthroplasty

Introduction:The effect of social drivers of health (SDOH) on readmissions and costs after total hip arthroplasty (THA) and total knee arthroplasty (TKA) is poorly understood. Policies such as the Hospital Readmissions Reduction Program have targeted overall readmission reduction, using value-based strategies to improve healthcare quality. However, the implications of SDOH on these outcomes are not yet understood. We hypothesized that the area deprivation index (ADI) as a surrogate for SDOH would markedly influence readmission rates and healthcare costs in the 90-day postprocedural period for THA and TKA. Methods:We used the 100% US fee-for-service Medicare claims data from 2019 to 2021. Patients were identified using diagnosis-related groups. Our primary outcomes included 90-day unplanned readmission after hospital discharge and cost of care, treated as "high cost" if > 1 standard deviation above the mean. The relationships between ADI and primary outcomes were estimated with logistic regression models. Results:A total of 628,399 patients were included in this study. The mean age of patients was 75.6, 64% were female, and 7.8% were dually eligible for Medicaid. After full covariate adjustment, readmission was higher for patients in more deprived areas (high Area Deprivation Index (ADI)) (low socioeconomic status (SES) group OR: 1.30 [95% confidence intervals 1.23, 1.38]). ADI was associated with high cost before adjustment (low SES group odds ratio 1.08 [95% confidence intervals 1.04, 1.11], P < 0.001), although, after adjustment, this association was lost. Discussion:This analysis highlights the effect of SDOH on readmission rates after THA and TKA. A nuanced understanding of neighborhood-level disparities may facilitate targeted strategies to reduce avoidable readmissions in orthopaedic surgery. Regarding cost, although there is some association between ADI and cost, this study may illustrate that ADI for THA and TKA is not sufficiently granular to identify the contribution of social drivers to elevated costs.