2006 State of the State of Health Care Leadership Forum: Experts Urge Caution on Implementing Physician Performance Measurements

Proceed slowly, and do it carefully. That was a key message for more than 200 leaders of health care throughout New England, from three national experts on physician performance measurement systems. Their remarks were delivered on October 26 at the Massachusetts Medical Society’s seventh annual State of the State’s Health Care Leadership Forum.

Kenneth R. Peelle, M.D., MMS president, said, “We don’t oppose transparency. But, we oppose it when it’s done poorly. And mostly, right now, it’s been done poorly.”

He added, “The first test that any reporting and measurement system must pass is the test of accuracy. If the data is not right, then the conclusions you draw from it are wrong.” (Download Peelle presentation --PowerPoint, 2.58 MB)

Robert D. Reischauer, president of the Urban Institute, said pay for performance is coming, “ready or not.” However, he said, evidence of its effectiveness is mixed. “Pay for performance,” he said, “has been propelled more by expectations than by evidence.”

Reischauer supports a recent Institute of Medicine report urging that P4P be implemented in phases, but that there are a lot of challenges to doing it right. “Any effort to do this on the cheap, or too fast, will lead to catastrophe,” he said. The impact of such an event would be a return to the past, where “the only way to control costs would be to lower payment rates on each service.” (Download Reischauer presentation -- PowerPoint, 751 KB)

Elliott Fisher, M.D., professor of medicine at Dartmouth Medical School and a national expert on performance measurement, said a key driver of performance measurement has the wide disparities in the cost and quality of care delivered across the country.

Fisher noted by many measures, spending per Medicare beneficiary varies two to three-fold among regions, with no relationship to the quality of care provided. The principal reason for higher spending, he said, is “overuse of supply-sensitive services,” such as ICU stays, physician visits and specialist consultations.

He added, “The cause [of overuse] is not frank errors, but differences in judgments” influenced by the capacity of a local system, and the local clinical culture.

Fisher concluded, “Improving efficiency will require fostering local organization accountability. Performance measurement, public reporting, payment reform and technical assistance should be aligned toward this goal.” (Download Fisher presentation -- PowerPoint, 3.13 MB)

Elizabeth McGlynn, associate director of the RAND Corporation, said “we don’t have a gold standard” in performance measurement. She said “no single source is adequate to address all the questions.”

McGlynn said that even though most existing approaches use claims data to measure physician performance, “they have some significant problems.” Further, she said, rating physicians on quality and costs at the individual level raise “a number of methodological issues.” Quality measures, she added, tend to be narrow snapshots that tend to reliably measure only small numbers of physicians.

McGlynn asserted that because the data on which scores are based were not intended to measure efficiency of quality measurement, “feedback loops” between physicians and payers are ”terribly important” to improving the quality and accuracy of the data. “We cannot just trust the data,” she said.

She concluded that rather than “torture” claims data to measure physician performance, “we should determine what our information needs are, and then build the systems to get them.” (Download McGlynn presentation -- PowerPoint, 2.93 MB)