Outcomes by Race and Ethnicity Following a Medicare Bundled Payment Program for Joint Replacement.

Importance: The Comprehensive Care for Joint Replacement (CJR) model, a traditional Medicare bundled payment program for lower-extremity joint replacement, is associated with care for patients outside traditional Medicare. Whether CJR model outcomes have differed by patient race or ethnicity outside of traditional Medicare is unclear. Objective: To evaluate outcomes associated with the CJR model among Hispanic patients not enrolled in traditional Medicare. Design, Setting, and Participants: This cohort study used hospitalization data from California's Patient Discharge Dataset for all patients who underwent lower-extremity joint replacement in California between January 1, 2014, and December 31, 2017. In California, 3 metropolitan statistical areas (MSAs) were randomly selected to participate in CJR in April 2016. Hospitals not participating in other Medicare Alternative Payment Models were included in the treated group if they were in these 3 MSAs and in the control group if they were in the remaining 23 MSAs. The data analysis was performed between October 1 and December 31, 2023. Exposure: Comprehensive Care for Joint Replacement program implementation. Main Outcomes and Measures: The main outcomes were hospital length of stay and home discharge rates by race and ethnicity. Home discharge status included self-care, the use of home health services, and hospice care at home. Event study, difference-in-differences, and triple differences models were used to estimate differential changes in health care service use by race and ethnicity for patients in the treated MSAs compared with the control MSAs before vs after CJR implementation. Results: Of 309 834 hospitalizations (patient mean [SD] age, 68.3 [11.3] years; 60.6% women; 14.8% Hispanic; 72.4% non-Hispanic White), 48.0% were in treated MSAs and 52.0% in control MSAs. The CJR program was associated with an increase in home discharge rates for patients without traditional Medicare coverage; however, the increase differed by patient race and ethnicity. The increase was 0.05 (95% CI, 0.02-0.08) percentage points higher for Hispanic patients with Medicare Advantage and 0.03 (95% CI, 0.01-0.04) percentage points higher for Hispanic patients without Medicare compared with their non-Hispanic White counterparts. Conclusions and Relevance: This cohort study shows that CJR program outcomes differed by race and ethnicity for patients covered outside traditional Medicare, with home discharge rates increasing more for Hispanic compared with non-Hispanic White patients. These findings suggest the importance of considering differential outcomes of Medicare payment policies for racial and ethnic minority patient populations beyond the initially targeted groups.