Overview

CMS has launched several demonstration programs to find out whether chronic care management programs developed in the private sector can be effective for the Medicare population. A pilot program at the Washington University School of Medicine suggests that barriers to effective treatment and prevention faced by high-risk Medicare patients can be addressed through a holistic care management approach.

Issue: Congress authorized a Medicare Coordinated Care Demonstration (MCCD) in 1997 to determine whether coordinated care programs can improve outcomes for chronically ill, fee-for-service Medicare beneficiaries without increasing the program's costs. In January 2001, the Centers for Medicare and Medicaid Services (CMS) selected 15 MCCD pilot sites in urban and rural areas around the country to test a variety of disease and case management approaches for these patients. This case study examines the early experience of one MCCD program.

Objective: The Washington University MCCD program coordinates social and medical interventions for high-risk Medicare beneficiaries, with the aim of improving their functional outcomes and reducing unnecessary emergency visits and hospitalizations.

Organizations: The Washington University School of Medicine, located in St. Louis, Mo., hosts a MCCD program in partnership with the Washington University Physicians Network (WUPN) and StatusOne, an American Healthways' company. WUPN is an independent physician association (IPA) that includes more than 900 full-time faculty physicians at Washington University School of Medicine and approximately 500 community-based physicians. American Healthways is a publicly traded company based in Nashville, Tenn., that provides disease management and care enhancement services to more than a million people in the United States and U.S. territories.

Implementation Date: Patient enrollment began in August 2002.

Target Population: The Washington University MCCD program targets non-institutionalized fee-for-service Medicare beneficiaries living in the St. Louis metropolitan area who have at least one chronic medical condition and who are at high risk of incurring substantial medical costs within the next year. Enrollees are high users of acute services whose health status is often compromised by psychosocial, educational, financial, and sensory limitations.

Key Measures: The program tracks changes in patients' functional status and acuity over time. Functional status is measured using a questionnaire similar to the Medical Outcomes Study Short-Form (SF-36) Health Survey. Acuity is assessed by the care manager on a scale from one to five, indicating the likelihood that a patient's medical condition will deteriorate or that they will need hospitalization.

Process: To design the program, Washington University drew on its experience with a prototype care management program for commercial and Medicare managed care patients in a local integrated health system that had been developed in partnership with StatusOne [1]. About 150 high-risk patients are identified monthly by a proprietary predictive modeling algorithm created by StatusOne that uses retrospective claims data on patients treated by WUPN. Program managers conduct ongoing community outreach to encourage physicians and ancillary service providers both inside and outside the WUPN network to refer high-risk patients to the program. Patients are invited to participate in the program following an "opt-in" approach in which patients give the MCCD program written authorization to contact their health care providers and coordinate their care.

Mathematica Policy Research, Inc., which has a contract with CMS to evaluate the program, randomly assigns those who enroll to a treatment or control group. Care managers notify physicians when their patients enroll in the program and all physicians receive, monthly, a list of enrolled patients. About 90 percent of care management is done through telephone calls. Patients with less complex cases are assigned to care managers at StatusOne's remote telecenter. More complicated cases are handled locally by care managers at Washington University through a combination of telephone and face-to-face contacts.

Care managers contact physicians and office staff on a daily basis to discuss matters of care coordination and obtain their input for the care plan. Key interventions carried out by the care managers (who are specially trained, registered nurses) include:

• optimizing patients' functional status by maximizing physical and occupational therapy, providing patient teaching on self-care, and supporting family-administered care;
• helping patients and families apply for and obtain financial and social support services through government programs and community organizations such as Medicaid, pharmacy assistance, and programs to help ensure a safe and healthy home environment (e.g., assistance with utility bills);
• reminding patients of, and helping them to prepare for, their doctors' appointments;
• facilitating physician–patient communication by attending physician appointments with patients to help ask questions and clarify their understanding of diagnoses and treatment plans;
• collaborating with home care providers to make them aware of patients' medical and psychosocial issues and ensure that patients are following instructions;
• communicating with covering physicians, when a patient's primary care physician is out of town, to suggest approaches that have worked for the patient in the past;
• expediting necessary hospital admissions by working with physicians to directly admit patients and thus avoid waits in the emergency room; and
• communicating changes in treatment plans to all involved caregivers and troubleshooting other problems through frequent telephone contact.

Program managers meet with a WUPN physicians' committee quarterly and several clinical leaders serve as advisors to high-risk care managers. Recently, the university began a pilot in which one care manager provides monthly verbal updates to physicians who have a large panel of patients enrolled in the program. The care manager might inform the physician of barriers to treatment compliance so that appropriate responses can be made, for example, or discuss treatment that is not in accordance with evidence-based guidelines.

Results: The Washington University MCCD has a current case load of about 2,100 patients (divided between the treatment and control groups), making it one of the largest MCCD pilot sites. About 20 to 25 percent of high-risk patients who are invited to participate ultimately enroll in the program. A substantial proportion of enrollees are nonelderly, disabled Medicare beneficiaries. The functional status and acuity of patients enrolled in the program has improved over time, suggesting that it is helping to improve their health outcomes. Anecdotal reports from care managers indicate that it also helps patients and their family members make more educated choices and prevent unnecessary hospital admissions and emergency room visits. Physicians, too, have given positive feedback, saying the program helps them get a fuller picture of their patients' needs. Mathematica Policy Research's evaluation of the MCCD will compare measures of service use and cost, patient adherence and disease-related limitations, patient and provider satisfaction, and quality of care for treatment and control groups. A first synthesis report prepared by Mathematica describes the early experiences of these programs. A second synthesis report, expected in August 2005, will present estimates of program-specific outcomes.

Lessons Learned: The Washington University MCCD program focuses on patient-oriented interventions that aim to overcome treatment and prevention barriers physicians are not aware of or are unable to address clinically. It is "astounding" to learn that some patients face basic unmet needs, such as a house without heat in the winter, but they have not communicated such problems to anyone, says Sandy Graff, R.N., program director. Patient recruitment using the "opt-in" approach required by the MCCD program was time-consuming compared with the "opt-out" approach typically used in commercial disease management programs, in which patients are automatically enrolled unless they specifically ask not to be included.

Creative approaches were needed to locate patients with incomplete contact information. Because patients often discarded or misplaced the initial enrollment mailing, the program's recruitment staff began contacting patients by phone, telling them to expect the mailing, and following up after it was sent to ensure it was received and answer any questions. The program's positive track record and preexisting established relationships with physicians paved the way for a positive reception among providers, who saw the benefits of care management and the value of referring patients to the program. Outreach to community organizations helped to broaden the program's reach. In the future, it would be helpful if MCCD programs could have access to Medicare claims data on the target population, which would enable more accurate identification of high-risk patients and timelier, ongoing evaluation of its effectiveness.

Implications: The program's experience suggests that many high-risk Medicare patients could benefit from an integrated care management intervention that overcomes medical and nonmedical gaps in their care and other barriers. Psychosocial supports were found to be especially helpful for socioeconomically disadvantaged populations, such as some of the inner-city patients served by this program. A holistic approach to care management that integrates medical and social supports "provides an opportunity to keep patients as healthy as possible with the greatest functional status they can achieve," says Graff.

For Further Information: Read Coordinating Care for Medicare Beneficiaries: Early Experiences of 15 Demonstration Programs, Their Patients, and Providers, a report prepared for Congress by Mathematica Policy Research. Contact Sandy Graff, R.N., at the Washington University School of Medicine, at graffs@msnotes.wustl.edu.

Reference
[1] J. P. Lynch et al. (2000) High-Risk Population Health Management—Achieving Improved Patient Outcomes and Near-Term Financial Results. American Journal of Managed Care 6, 781–791.

---

This study was based on publicly available information and self-reported data provided by the case study institution(s). The aim of Fund-sponsored case studies of this type is to identify institutions that have achieved results indicating high performance in a particular area, have undertaken innovations designed to reach higher performance, or exemplify attributes that can foster high performance. The studies are intended to enable other institutions to draw lessons from the studied organizations' experiences in ways that may aid their own efforts to become high performers. The Commonwealth Fund is not an accreditor of health care organizations or systems, and the inclusion of an institution in the Fund's case studies series is not an endorsement by the Fund for receipt of health care from the institution.