Impaired Physical Performance Predicts Hospitalization Risk for Participants in the Program of All-Inclusive Care for the Elderly

Background: Medicaid spending on the Program of All-Inclusive Care for the Elderly (PACE) has grown rapidly over the last 5 years. Reducing hospitalization rates is a major goal for PACE. However, there is a paucity of research evaluating the relationship between impaired physical performance and hospitalizations in PACE. Objective: This study tested whether physical therapist-assessed physical performance, measured by the Short Physical Performance Battery (SPPB), can be used to identify participants in PACE at risk for all-cause hospitalizations or potentially avoidable hospitalizations (PAH). Design: This was a retrospective cohort study of 1093 participants in PACE facilities in the Denver, Colorado, area. Methods: Data were acquired from linked electronic medical record data and hospitalization claims. Unadjusted and adjusted Cox proportional hazards regression models were used to evaluate the relationship between SPPB scores and the probabilities of both all-cause hospitalizations and PAH. Results: The unadjusted likelihood of hospitalization increased with greater physical performance impairment (for SPPB scores ≥8/12: 12.2%; for SPPB scores of 4/12 to 7/12: 15.7%; for SPPB scores <4/12: 21.1%). Compared with participants with SPPB scores ≥8/12, participants with SPPB scores <4/12 had nearly double the unadjusted hazard for hospitalization (hazard ratio = 1.99; 95% CI = 1.34-2.96). In adjusted Cox regression models, participants with SPPB scores <4/12 remained significantly more likely to be hospitalized (hazard ratio = 1.87; 95% CI = 1.24-2.84). Similar relationships were observed for PAH. Limitations: The use of data from a single network of PACE facilities might limit generalizability to states with different Medicaid guidelines. Conclusions: The findings suggest that impaired physical performance is an independent risk factor for hospitalization among participants in PACE. These findings could help guide the development of PACE program modifications for measuring and intervening on impairments in physical function.