Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Helena Temkin-Greener, PhD¹, Alina Bajorska, MS¹, and Dana B. Mukamel, PhD². (1) Department of Community and Preventive Medicine, School of Medicine, University of Rochester, Box 644, 601 Elmwood Avenue, Rochester, NY 14642, 585-275-8713, Helena_Temkin-Greener@urmc.rochester.edu, (2) Department of Medicine, Health Policy and Research Center, University of California, Irvine, 100 Theory, Suite 110, Irvine, CA 92697

This presentation addresses two questions. 1) What is the extent of service use variations in a national, managed care program for frail, chronically ill elderly? 2) Does more care result in better functional outcomes? Experts have suggested that much of the variation in Medicare service use nationally is unwarranted and can be reduced by “reform in financing that would facilitate investment in under funded aspects of care in the non-acute sectors and in the resources needed for active chronic disease management” (Wennberg, 2005). The Program of All-Inclusive Care for the Elderly (PACE) was created with just such a premise in mind. PACE is a community-based managed care program for frail, chronically ill individuals with significant functional impairments. We use longitudinal data containing 42,780 assessment records for 9,987 enrollees in 29 programs, over a period of 3-years. The file contains individual-level data on demographics, physical and cognitive health, diagnoses, treatments, use of hospitals, nursing homes, day center attendance, physical and occupational therapy, and home personal care. Mixed regression and GEE loglinear Poisson models, as well as bootstrap procedure are employed in the study. Controlling for all other site effects, we examine the marginal effect on functional status over time for the 5 service types. Only hospital admissions have a statistically significant (CI, 0.196 – 0.825) impact on ADLs (est. 0.447). Sites that use hospital care more often have worse functional outcomes. We found no other significant relationship between functional change and service use. We discuss operational and policy implications of our findings.

Learning Objectives: