Skip to main content

 

[an error occurred while processing this directive]
[an error occurred while processing this directive]
[an error occurred while processing this directive]
[an error occurred while processing this directive]

Taking medicine's measure

CECS data reveal what works, what doesn't
Jack Wennberg
Jack Wennberg (Photo courtesy of CECS)

Typically, researchers frame questions about how to "fix" health care within specific parameters: What does a day in the hospital cost? How can patients be encouraged to demand less care? How can financial incentives be used to encourage doctors to practice in underserved areas?

Dartmouth's Center for the Evaluative Clinical Sciences (CECS) targets more fundamental questions: What are the right medical interventions? How well do they actually work? How often does the right thing - from the perspectives of biomedical science, the patient's preferences and societal interests - actually get done?

For the past 15 years, the CECS has been illuminating the disconnect between the perception - that the United States has the best health care system in the world, and the reality - that the system does expensive things too often, useful things too infrequently and makes too many mistakes.

The Dartmouth Atlas of Health Care project, which shows regional variations in medical practice, outcomes and types of procedures, is one of the Center's most visible products.

The basis of the work is an almost unfathomable amount of data, most of it from the Medicare program, that documents what's going on in American medicine - who's getting what and when, and how much it costs. Analysts feed information into complex computer systems that sort, count, make statistical adjustments and spit out some astounding conclusions.

Map of distribution of hospital-employee registered nurses

Map of primary care physician workforce

Map showing rates of cardia catheterization among Medicare enrollees
These maps illustrate: (A) the distribution of hospital-employed registered nurses; (B) the primary care physician workforce; and (C) rates of cardiac catheterization among Medicare enrollees, as compared to the national average rates. Areas with higher than average rates are in darker colors. The Dartmouth Atlas uses these thematic maps to illustrate variations in the distribution and utilization of health care resources in the United States. (Maps courtesy of CECS)

Medicare, for example, spends almost three times as much per enrollee per year in Miami as it does in Minneapolis. You're twice as likely to have back surgery in some areas as in others. How you die depends on where you live.

CECS reflects the core belief of John E. "Jack" Wennberg, the Peggy Y. Thomson Professor for Evaluative Clinical Sciences, who launched the Center in 1989. Michael Zubkoff, Professor and Chair of the Department of Community and Family Medicine, supported Wennberg in his fledgling enterprise. "One of Jack's strengths," Zubkoff said, "is that he can convince very good people to come here - and then he lets them flourish." Wennberg was recently voted the 14th most powerful physician-manager in the United States by readers of Modern Physician magazine.

Among the experts Wennberg has recruited are Paul Batalden, Professor of Pediatrics and Community and Family Medicine, an international leader in health care quality improvement; and Elliott Fisher, Professor of Medicine and Community and Family Medicine, whose research has reached the startling conclusion that an excess use of some health care services can make you sicker.

Other members of the team include Jonathan Skinner, John French Professor of Economics, whose research includes finding out why black Americans have worse health outcomes than white Americans; Ann Barry Flood, Professor of Community and Family Medicine, whose work in medical sociology is helping the city of Nashville redesign systems of care; and Hilary Llewellyn-Thomas, Professor of Community and Family Medicine, a sought-after expert on helping patients become more engaged in making decisions about their health care.

CECS is distinguished by the diversity of its undertakings and the excellence of its component parts. That excellence is based on Wennberg's belief that an ailing health care system can't be diagnosed, let alone treated, without a deep understanding of its complexities.

And on any given day, CECS faculty members can be found addressing Congressional committees, meeting with White House advisors and giving keynote speeches. Many bring CECS data to undergraduate classes in Economics, Sociology and Government. It is unusual for a week to go by without some reference to CECS work in the national media - from the New York Times and the "NBC Nightly News" to the Annals of Internal Medicine.

It wasn't always so. Thirty years ago, no American medical journal would publish Wennberg's papers on small area analysis, a method he developed of measuring the distribution and utilization of health care resources in the United States. His conclusions - that the health care you receive depended more on where you lived and who you consulted than on shared medical knowledge applied in a systematic way - were controversial to say the least.

And 30 years ago, no one thought to apply continuous quality improvement to the delivery of health care. The acceptance both research agendas have received is grounded in Wennberg's commitment to "get the science right."

Getting the science right takes many forms at the CECS - from validating statistical methods and documenting outcomes, to finding ways of improving care and eliminating errors.

CECS' impact on health care policy can be seen in improvements of clinical care, policies governing funding of programs such as Medicare and in a heightened understanding by patients of their own choices. The collective work of a group of extraordinary individuals, supported by an institution willing to take a chance, is paying off and may well point the way to a better, safer and more cost-effective health care system.

View The Dartmouth Atlas.

By MEGAN MCANDREW 

Questions or comments about this article? We welcome your feedback.

Last Updated: 5/30/08