Evaluation of Programs of Coordinated Care and Disease Management

Mathematica Policy Research, Inc. (MPR) evaluated 16 independent demonstration sites that provide coordinated care interventions to Medicare beneficiaries with chronic illnesses. The rationale for the demonstration is the lack of coordination among the multiple providers typically serving Medicare beneficiaries with chronic illnesses, as well as the adverse consequences of the lack of coordination for the beneficiaries and for Medicare costs. The demonstration sites, selected in early 2001, offered programs designed to improve both the care that patients receive and patients' knowledge of, and adherence with, recommended self-care and behavior. The study estimated the effects of each site on patients' well-being and satisfaction, in addition to the site's effects on the use and cost of Medicare covered services. This analysis relied on a patient survey conducted 6 to 12 months after enrollment, and on Medicare claims data and any data available from the demonstration sites that could enhance the study. The study included two rounds of physician surveys. In each site, eligible applicants were randomly assigned to treatment and control groups. An extensive process analysis was conducted to describe the interventions in detail, with the key goal being an assessment of those factors that account for program success and failure. The study included case studies of each site, program profiles, interim site-specific memos, two interim summary reports, two reports to Congress (based on the interim summary reports), and a final summary report. This original study enrolled 18,277 beneficiaries. In 2008 Congress extended the study for 2 of the sites, Mercy Medical Center - North Iowa and Health Quality Partners in Pennsylvania, and they will recruit beneficiaries and provide demonstration intervention services through the spring of 2010. Mathematica Policy Research will evaluate the results of this extended demonstration using Medicare claims data and qualitative site visits to the two programs.