Outcomes from a partially integrated model of managed long-term care—The VNS CHOICE program (VCP)—were compared to those from a fully capitated model—the Program of All-Inclusive Care for the Elderly (PACE). A retrospective cohort design examined discharge, morbidity, and functional status for 3,200 individuals over 18 months. Although the two populations were close in average functional status, the PACE sample was sicker and more cognitively impaired. Higher rates of death were found in the PACE program (22 percent vs. 10 percent over 18 months) and a greater likelihood of decline in instrumental activities of daily living; higher functional disability scores (ADLs) predicted discharge in PACE but not in VCP. When propensity score stratification was used to compare like populations, the impact of program enrolment on outcomes became statistically insignificant, suggesting that a partially capitated model performs at least as well as the fully capitated PACE model.