Case Study: Cutting Costs and Improving Quality of Care: Not All Programs Likely to Succeed But Targeting Can Help

Medicare Coordinated Care Demonstration Reveals Features that May Bring Success

The award-winning article . . . provides new insights into the delivery of health care and advances knowledge of the field.

- AcademyHealth, 2010 Article-of-the-Year Award
Project Facts
  • This study used a randomized control trial design to evaluate the impact of each of 15 separate Medicare Coordinated Care Demonstration programs on expenditures, service use, quality of care, and patient experience of beneficiaries, and the satisfaction of their usual care physicians.
  • The analysis incorporated multiple data sources including Medicare claims data on hospital admissions and expenditures, as well as results of a telephone survey of patients regarding health behaviors, outcomes, and satisfaction with care (with a mean response rate across programs of 95%), and a survey of the patients’ primary physician regarding their ratings of the care management programs.
  • The results through the initial four years revealed that only two of the fifteen programs reduced hospital admissions, and none of the programs generated net savings. Among the 11 program that were extended for at least 6 years, two reduced hospital admissions for all enrollees and another two reduced hospitalizations for a high-risk subgroup; however, none generated net savings to Medicare.
  • Reports from the project described key elements of the successful programs and offered suggestions for designing future programs to optimize cost savings and quality of care.
  • This pioneering research was featured in multiple peer reviewed journals including Health Affairs and JAMA, and received awards including AcademyHealth’s 2010 article-of-the-year and the National Institute for Health Care Management Foundation’s 2010 research award for excellence in original and creative health care research.


The Issue

Medicare spends a disproportionate amount on beneficiaries with multiple chronic illnesses, primarily due to their high rates of hospitalizations and readmissions. Past studies had suggested that care coordination and moving towards patient-centered care could help reduce hospitalizations and save money among this patient population. However, Mathematica’s evaluation of the Medicare Coordinate Care Demonstration was the first large-scale rigorous study of the effects of care coordination programs in the Medicare fee-for-service setting.

The Approach

The U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Studies, contracted with Mathematica in 2000 to conduct an evaluation of the Medicare Coordinated Care Demonstration to assess whether care coordination programs for Medicare recipients with multiple chronic conditions could lower costs or improve the quality of care and beneficiary and provider satisfaction while not increasing costs in a fee-for-service setting. The evaluation included analysis of Medicare claims data from the four-year demonstration as well as data from surveys of patients and physicians conducted by Mathematica. In addition, Mathematica conducted an extended evaluation for 11 of the 15 programs which received extensions on their original end-dates; this included an implementation analysis, an impact analysis based on claims data, and a synthesis of these data to draw inferences about which program features contributed to any improvement in health outcomes or cost. Mathematica continues to evaluate the one remaining program.

The Impacts

This study was one of the first of its kind to use the methodological gold-standard—the randomized control trial design—to evaluate whether care coordination and disease management programs can lower costs and improve the health of beneficiaries with chronic illness covered by fee-for-service Medicare. This groundbreaking research revealed that the success of such programs is not guaranteed, and reflects specific approaches and targeting to patients at risk. Only 4 of the 11 participating programs that CMS extended past the initial four years significantly reduced hospitalizations, and 2 did so only for high-risk beneficiaries. When care management fees were included, these four programs were essentially cost neutral to Medicare, but none generated net savings.  Frequent in-person contact (in addition to telephone contact) by care coordinators with patients and their providers, robust medication management, and comprehensive transitional care when patients are discharged from hospital were among six features common to the four programs. These features were generally missing from the demonstration programs that did not reduce hospitalizations. The study’s results suggest that incorporating these approaches into medical homes, accountable care organizations, and other policy initiatives could reduce hospitalizations and improve patients’ lives. However, the approaches would save money only if care coordination fees were modest and organizations found cost-effective ways to deliver the interventions.

To learn more about how Mathematica Policy Research can provide solutions to your challenges, contact: Communications@mathematica-mpr.com

This case study is for informational purposes only. Mathematica Policy Research, a nonpartisan research firm, provides a full range of research and data collection services, including program evaluation and policy research, survey design  and data collection, research assessment and interpretation, and program performance/data management, to improve public well-being. Its clients include federal and state governments, foundations, and private-sector and international organizations. The employee-owned company, with offices in Princeton, N.J.; Ann Arbor, Mich.; Cambridge, Mass.; Chicago, Ill.; Oakland, Calif.; and Washington, D.C., has conducted some of the most important studies of education, disability, health care, family support, employment, nutrition, and early childhood policies and programs.

About the Project

As required under the Balanced Budget Act of 1997, the U.S. Secretary of Health and Human Services commissioned the Medicare Coordinated Care Demonstration to pilot and evaluate care coordination programs in the Medicare fee-for-service setting. The Centers for Medicare & Medicaid Services (CMS) contracted with Mathematica to evaluate the Medicare Coordinated Care Demonstration, which tested whether care coordination and disease management programs: (a) lower Medicare expenditures, or (b) increase the quality of health care services and beneficiary and provider satisfaction without increasing expenditures for beneficiaries  with chronic illnesses.

The programs were operated by a range of organizations, targeted different types of beneficiaries with chronic illnesses, and used different approaches. Only one of them embedded care managers in the primary care practices of the beneficiaries, though some  of the others located the care managers in the same hospitals or health systems where beneficiaries received care and/or provided incentives for their usual care providers to interact with the care manager. The final report provided the most rigorous estimates ever presented on the effectiveness of care coordination interventions in a Medicare fee-for-service setting.  Mathematica was subsequently awarded a grant from the Robert Wood Johnson’s Health Care Financing Organization to conduct additional research.