News Release

Rigorous Evaluation Of "Hospitalist" Model Needed To Determine Impact Of This New Medical Specialty, According To UCSF Health Policy Researchers

Peer-Reviewed Publication

University of California - San Francisco

In today's increasingly competitive health care environment, many changes are occurring to improve patients' clinical outcomes and to decrease costs. One change includes the recent introduction of the hospital-based generalist physician, known as the hospitalist -- doctors who specialize in the care of hospital patients. The question remains does this new specialty provide improved health care?

A new report by UC San Francisco health policy researchers says a rigorous evaluation of the system is needed to address these concerns and assess the full impact of the hospitalist model.

"Measuring the quality and outcomes of health care is a complex process," said Jonathan Showstack, PhD, MPH, UCSF professor of medicine and health policy and associate director of the UCSF Institute for Health Policy Studies. "Rigorous evaluations must be conducted to provide convincing evidence concerning health status outcomes, the satisfaction of patients, providers, and trainees, and the costs of care."

Showstack and colleagues discuss the many components involved in evaluating the impact of hospitalists in the supplement of the February 16 issue of the Annals of Internal Medicine.

The hospitalist is a departure from the way doctors have traditionally functioned in the American academic hospital. Traditionally, the care of hospital patients has been managed by residents, who are medical school graduates in the late stages of training in academic medical centers. Residents work under the supervision of faculty members who rotate in that role infrequently, often just one month each year, according to the researchers. Patients' primary care doctors traditionally manage their care in non-teaching hospitals.

The emergence of hospitalists is based on the idea that inpatient care specialists will be better skilled, more efficient and more available to hospitalized patients than primary care physicians or residents.

The hospitalist model, said Showstack, has generated controversy. Some research suggests that the model is associated with lower costs and with comparable or better patient outcomes. Concerns have been expressed, however that patients will be dissatisfied with hospital care because of lack of physician continuity and that referring physicians will be dissatisfied because of the loss of control of their hospital patients, he added. There have also been concerns that costs may actually be higher for the hospital and, perhaps most importantly, health status outcomes may be affected adversely by the transfer of responsibility of a patient from their primary care physician to an inpatient generalist physician.

To provide evidence of the value of the hospitalist model, Showstack and colleagues recommend quality of care should be evaluated through the measurement of both processes and outcomes. In addition, an evaluation of the hospitalist model requires an adequate research design, Showstack said, which includes the following components:

  • A precise definition of the model.

  • Intervention and control groups.

  • Collection of patient clinical and demographic information.

  • The specification of both process and outcome measures.

  • Use of statistical techniques appropriate to the question(s) being asked and the data that are collected.

  • A design that distinguishes between outcomes attributable to the introduction of hospitalists and other changes in medical treatments and the organization of care.
Co-authors are Patricia Katz, PhD, UCSF assistant professor of medicine and health policy, UCSF Institute for Health Policy Studies, department of medicine and Ellen Weber, MD, FACEP, UCSF associate clinical professor of medicine, associate director of division of emergency medicine, department of medicine.

###



Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.