QUALITY COMPETITION IN MANAGED CARE
Nancy Beaulieu
Dr. Beaulieu in her dissertation research explored aspects of quality information and quality competition in the managed care health insurance market. In her study of incentives created by overlapping provider networks between health plans, Dr. Beaulieu shows that the existence of overlapping networks reduces the incentives for health plans to invest in quality improvement systems. Her current case study on Health Partners in Minnesota is examining the consequences of making networks more exclusive to a health plan.
Another component of Dr. Beaulieu's dissertation research empirically assesses whether there is a link between plans that offer higher quality along specific dimensions of performance and consumer choice of health plans. She showed that there is an empirical association between performance on quality measures and the choices of health plans of certain segments of the enrollee population. Her "laboratory" was the Harvard University health benefits program. One potentially disturbing implication of this work is that reporting more information may heighten competition to select the best risks using performance on certain quality measures. This is important work and is closely connected to research on risk adjustment and quality competition within managed care.
*QSPAN - CARDIOVASCULAR DISEASE
Barbara McNeil, Heather Palmer, Ed Guadagnoli, Richard Frank, Bruce Landon
Many of the center's investigators from HMS and HSPH are involved in studies on patients who have chronic cardiovascular disease and who are being treated in managed care settings. The first, "Q?SPAN?CD" involves the development and implementation of quality measures for patients with acute myocardial infarction (AMI), congestive heart failure (CHF) or hypertension. The study is in the third of five years and involves a partnership with four large managed care plans (Allina, PacifiCare, Prudential, and United) in six geographically separate areas of the country. It is funded by AHRQ. The second study, "The Role of Financial Incentives and Managerial Practices in the Care of Cardiac Patients", has just begun with funding from AHRQ and the Commonwealth Fund. It will use the quality data obtained from Q?SPAN?CD and will augment the data with information from surveys at the group level on financial incentives and managerial practices in these groups.
For more information please see QSPAN Webpage.
*QSPAN-HEDIS
Arnold Epstein
In recent years the Health Plan Employer Data Information Set (HEDIS) developed
and administered by the National Committee for Quality Assurance has become
the most commonly used set of quality measures for health plans. The current
version of HEDIS includes thirteen indicators that measure the quality of clinical
performance. These data are voluntarily reported by health plans to NCQA which
combines HEDIS results into a publicly available database, Quality Compass.
The most recent version of Quality Compass included data from more than 330
health plans representing 75% of the U.S. HMO population. Investigators at Harvard
University in collaboration with investigators at the RAND Corporation and NCQA
have embarked on a multi?year project to evaluate and refine quality measure
for incorporation into HEDIS. This work involves a broad effort to ensure that
the measures in HEDIS reflect the state of the art in terms of clinical importance,
scientific soundness and feasibility in implementation.
*MAJIC: MAKING ADVANCES IN AVOIDING
JAUNDICE IN INFANT CARE
Heather Palmer
This five-year study, supported by AHRQ (formerly AHCPR), is a 5-year controlled trial of models for quality improvement in large managed care organizations. Partners include the American Academy of Pediatrics (AAP), Henry Ford Health System in Detroit and Blue Cross Blue Shield of Texas, Inc. We are testing the model by applying it to care for infants with hyperbilirubinemia (infant jaundice). Symptoms of jaundice take 3-5 days after birth to fully develop; however, by that time, with today's trends for early discharge, most babies are outside a medical setting and family caregivers can easily miss the symptoms. Baseline and ongoing quality measurement for the project involves administrative (computerized) data, record review and a parent survey. Early findings from the project have influenced the AAP's plans for revising their clinical practice guideline on neonatal hyperbilirubinemia.
For more information please visit MAJIC Webpage.
*ASTHMA TREATMENT IN MEDICAID MANAGED
CARE ORGANIZATIONS
Howard Bailit, Joan Buchanan
Managed care organizations, hospitals and physicians and state and city health departments have formed an alliance to more effectively prevent and treat Medicaid eligible (and other) Hartford children who have asthma. The staff of all city clinics and practices were trained to screen for asthma, classify children by severity level and assign them to standardized treatment plans. Available in a central data base, this information is updated continually and used to provide 24 hour nurse response to parent calls and to monitor the delivery of medical care. In addition, parents and children receive special asthma education, and community health aids visit the homes of children with uncontrolled asthma. To evaluate this demonstration project, a pre?post, control group design is used to assess changes in health status, utilization, expenditures, and patient, parent and provider satisfaction. If successful, this asthma control system will continue with alliance support after completion of the project. (Bailit, Buchanan)
*EFFECT OF HEALTH PLANS ON HYPERTENSION
AND DIABETES CARE
Ed Guadagnoli, John Ayanian, Nancy Keating, Bruce Landon, Mary Beth Landrum
Collaborating investigators from the Department of Health Care Policy at Harvard Medical School, the Healthcare Education and Research Foundation of St. Paul, Minnesota, and the University of Massachusetts, Boston propose to examine the quality of care provided to patients treated in three managed care settings located in the Minneapolis/St. Paul area. Investigators will evaluate whether four features of managed care organizations ? financial arrangements with providers, strategies for the management of care, methods of delivery of care, and the degree of cost?sharing borne by patients ? influence the quality of care received by patients with hypertension or diabetes, two highly prevalent conditions for which attempts have been made to define valid indicators of quality of care. The results of this study will identify financial, delivery, and management features associated with the delivery of high quality care, and thereby, assist policy?makers, plan personnel, and providers in their efforts to improve quality, and consumers and purchasers in their selection of health plans.
*IMPROVING CARE COORDINATION AND PATIENT
OUTCOMES
Jody Hoffer Gittell
This study looks at patient care coordination in nine Boston, New York City and Dallas hospitals. The Care Coordination Study tests the impact of organizational practices on coordination among care providers, and the impact of coordination on the quality and efficiency of outcomes for joint replacement patients. The findings provide guidance for designing organizational practices - such as case management, patient rounds, supervision and hiring - to improve coordination among care providers, and ultimately to improve the quality and efficiency of patient outcomes.
*QUALITY MANAGEMENT PRACTICES IN MEDICAID
MANAGED CARE
Arnold Epstein, Bruce Landon
Enrollment of Medicaid beneficiaries in managed care programs is increasing rapidly. Rapid expansion of Medicaid managed care has raised concerns about the capacity and willingness of health plans enrolling Medicaid beneficiaries to provide high quality care. To date little information is available on the types of health plans caring for Medicaid beneficiaries and their quality management practices. After researchers randomly selected the District of Columbia and 11 states, we surveyed all 154 health plans providing pre?paid general medical care to Medicaid beneficiaries during June, 1997. Of the 130 plans that responded, 50 percent were "Medicaid plans" that predominantly served the Medicaid population (>75 percent of enrollees). Investigators found that, relative to commercial plans, Medicaid plans are relatively new and small. While they are similar to commercial plans in many aspects of quality management, they are more likely to target programs to the specific needs of the Medicaid population. Neither commercial nor Medicaid plans have notably high records for success in quality improvement.
*USE OF EFFECTIVE CARDIAC DRUGS AFTER
MYOCARDIAL INFARCTION
John Ayanian, Mary Beth Landrum, Bruce Landon, Paul Cleary, Barbara McNeil
In three related studies, center-affiliated investigators are evaluating the use of cholesterol-lowering therapy and other effective cardiac drugs for secondary prevention among managed-care enrollees and fee-for-service patients who have survived an acute myocardial infarction. One study is evaluating beliefs and experiences regarding cholesterol-lowering therapy among commercially insured enrollees of a large, national managed-care company. A second study is comparing these beliefs and experiences among Medicare beneficiaries enrolled in HMOs or fee-for-service care in California, Florida, Massachusetts, New York, and Pennsylvania. In a third study, rates of use of beta blockers, aspirin, angiotensin-converting-enzyme inhibitors, and cholesterol-lowering drugs are being compared for Medicare HMO and fee-for-service enrollees in these same states.
*QUALITY OF CARE FOR COLORECTAL CANCER
John Ayanian, Alan Zaslavsky, Ed Guadagnoli
This study is assessing the treatment and outcomes of patients with colorectal cancer using linked data from the California Cancer Registry (CCR), hospital discharge abstracts, Medicare enrollment data, and a survey of recently diagnosed patients. The study is using advanced statistical methods to analyze the care of approximately 70,000 patients diagnosed statewide during 1994 through 1998, including their stage at diagnosis, timeliness of care, adherence to treatment guidelines, and survival. CCR records have been linked to Medicare enrollment files to identify specific health plans and fee-for-service care for patients 65 and older and to hospital discharge abstracts and U.S. Census data for all patients to gain added information for risk-adjustment. In addition, 2000 patients in northern California are being surveyed approximately 8 months after diagnosis to obtain personal appraisals of their cancer care and health-related quality of life. This study will yield important new information on multiple measures of care for colorectal cancer, thereby providing a model for new uses of cancer registries to assess the quality of cancer care by patient and provider characteristics and among specific health plans.