Policymakers debate how to profile Medicare physicians

Washington -- Lawmakers and federal officials have expressed interest in profiling physicians based on how many resources they use to treat their patients, part of an eventual move toward paying doctors based on the quality and efficiency of their care. But exactly how Medicare and other payers set up such report cards will determine whether such an incentive system will work as intended, experts said.

Developing more accurate information about the comparative value of different physicians' care can help lower high rates of health care spending growth while improving quality, said Tom Miller, resident fellow with the American Enterprise Institute, which hosted a Dec. 2 discussion on physician profiling in Washington, D.C.

Patients need better information, not necessarily more of it, and they also need value and not just quality, he said. The term value can be assessed in various ways, but it most commonly involves both higher quality and lower total cost of care, said Miller, who also co-authored a paper on measuring physicians' performance published in the September/October Health Affairs.

Employers are pushing payers to identify high-quality and low-cost care, but physicians generally have been opposed to the current methods of assessing performance, Miller said. Specifically, doctors are critical of inconsistent approaches by different plans in the same market; efficiency scores that are susceptible to events outside of a doctor's control; failure to assess broader measures of performance, such as clinical judgment; and a lack of any assurance that money saved will be directed to improving care quality elsewhere.

Increasing physician acceptance by instituting measurements they trust and tolerate could boost the availability of quality information, Miller said.

"You can have the best information in the world, but if people don't get it, don't understand it and can't use it, it's not going to mean a whole lot," he said. "We've got to find people who are delivering information who are trusted, and in many cases it's not your friendly employer or insurer."

Determining how to measure efficiency

Physicians play a central role in driving health care spending amid evidence that some of the spending might not be warranted, according to a September report by the Government Accountability Office.

Because of this, GAO recommended that the Centers for Medicare & Medicaid Services develop a physician profiling system that uses Medicare expenditures as the basis for measuring efficiency. Such a profiling system would include:

• Adjustments for differences in patients' health status.
• Empirically based efficiency standards.
• An education program that explains to physicians how the system works and how their results compare with peers.
• Financial and other incentives for individual doctors to improve the efficiency of care.
• Methods for measuring the impact of doctor profiling on program spending and physician behavior.

The GAO also questioned whether the government has been using all the tools at its disposal.

"Generally, we adopt the attitude that physicians are the key decision-makers," said A. Bruce Steinwald, an independent consultant and former senior executive on GAO's health care team in Washington, D.C., and a participant in the AEI forum. "There are plenty of people out there who say we have good quality metrics, but we just don't apply it."

In its report, GAO evaluated the per-capita methodology for profiling physicians --which measures a patient's resource use over a fixed period and attributes it to physicians providing his or her care -- to assist CMS with development of a physician feedback program. Episode-based measures, on the other hand, look at resource use on patients experiencing a particular episode of illness, such as a stroke or heart attack.

Using Medicare claims data from 2005 and 2006 and applying a per-capita methodology, GAO found that specialists' resource use remained stable, despite high patient turnover. "This stability suggests that per-capita resource use is a reasonable approach for profiling specialist physicians because it reflects distinct patterns of a physician's resource use, not the particular population of beneficiaries seen by a physician in a given year," the report concluded.

Connecting profiling with payment

Physicians may need to face the fact that profiling systems are going to be imperfect, and that any associated payment incentives may not always match actual performance levels, said Robert M. Krughoff, president of Consumers' Checkbook/Center for the Study of Services. Checkbook/CSS is a Washington, D.C.-based organization that rates health care professionals as well as services such as auto repair shops, electricians and movers.

"In general, we have to have a tolerance for some error," Krughoff said. "It seems to me we have to be prepared to take some risks on that front. If the risk is some doctor gets a lower pay-for-performance reward, that's a small risk, and it should average out over time. Sometimes you'll do well and sometimes you'll do badly."

Checkbook/CSS in July released results from a pilot study it conducted with physicians in Denver, Kansas City, Mo., and Memphis, Tenn., that developed physician report cards through patient surveys. The program demonstrated that low-cost, rigorous survey methods can produce efficient ratings and reports on most doctors, Krughoff said.

Checkbook/CSS sued the federal government in 2006 for access to Medicare claims data on individual physicians that it wanted to make public. A federal appeals court in January 2009 ruled in favor of keeping the information private after CMS and other groups, including the American Medical Association, fought the disclosure. The AMA argued that such raw data would be of little use to patients in choosing doctors and could compromise physician privacy.