MedPAC explores physician comparisons

Washington -- One of the first steps toward encouraging doctors to provide more efficient care to their Medicare patients could involve showing them that their colleagues are doing a better job, according to some Medicare policy-makers.

Federal officials and the panel that advises lawmakers on Medicare are starting to explore this concept. By telling individual physicians how they rank among other doctors within the program when it comes to the number and level of services they provide to certain types of patients, the government can convince some of the overutilizers to rein in their habits, they said.

"This is where it is in terms of getting at the cost issue," said Francis Crosson, MD, executive director of the Permanente Federation's medical groups and a member of the Medicare Payment Advisory Commission.

MedPAC recommended to Congress in March that it direct Medicare officials to undertake such an initiative. The process would involve combing through claims submitted by doctors and establishing a confidential system for feeding the results back to the physicians.

Sharing both individual and aggregate data would serve as a powerful educational tool for doctors, said Anne Mutti, a MedPAC consultant.

"They can look to see, for example, if in treating emphysema patients they use three times as much hospital care but don't use as much prescription drugs or home health care as compared to their peers," she said. "They may decide that they would like to better align their care or they may decide not to. But at least they have that information."

MedPAC and the Centers for Medicare & Medicaid Services, which is also looking into the feasibility of launching a doctor-comparison system, have said that the results would not affect Medicare payment -- at least not right away. Mutti noted that if the concept works, it eventually could be folded into a national pay-for-performance program.

Physicians welcome any opportunity to receive good feedback about the care they are providing to their patients as long as it is not done in a punitive manner, said American Medical Association Chair J. James Rohack, MD, a cardiologist in Temple, Texas. But the prospect that doctors will be assessed based on Medicare claims data alone is troubling because such information largely ignores other factors that could be involved, such as severity of illness, individual patient circumstances and evolving medical guidelines, he said.

The diabetic patients who are going to a particular physician, for instance, might be logging more office visits because some of them cannot afford to purchase the expensive drug regimens that help manage their conditions. Or they may be receiving more diagnostic tests from a certain doctor because that physician is following the latest guidelines for diabetic care, he said.

"There's a concern about just taking raw numbers and giving them back to the doctor when they're not adjusted properly," Dr. Rohack said.

"If one is comparing Doctor A against Doctor B with no way to look at whether the services were appropriate or not, then you're driving toward an assumption that people who order more services are automatically overutilizers," he said.

Getting off the ground

MedPAC and CMS already are running into thorny implementation problems in their quest to control Medicare outpatient spending, which underwent a significant increase last year. Treating physicians as part of the solution, rather than part of the problem, will require officials to craft a truly educational initiative, said CMS Administrator Mark McClellan, MD, PhD.

"The main implementation problem is figuring out what types of information to provide to doctors," he said. "We can't just throw data out there that don't mean anything."

Private insurers already utilize sophisticated software to process claims and to group patient spending into individual episodes of care. MedPAC plans to try using such software with Medicare claims to test whether it would allow federal officials to make the necessary case-mix adjustments for producing viable comparisons among doctors.

But while insurance firms already use this technology to determine how many services individual physicians in their plans are utilizing, the process hasn't always been perfect, Dr. Rohack said. Some commercially available software, for example, registers up to only two comorbidities for a patient with a chronic illness, potentially skewing any adjustments for severity of illness.

In addition, investigators could find that Medicare claims do not always give a good indication of which doctor is the primary caregiver for a given patient, said Arnold Milstein, MD, the medical director of the Pacific Business Group on Health and a MedPAC member. A chronically ill senior often sees several doctors over the course of an episode of care, and the identifying physician information on the associated claims might not be enough to single out the lead doctor.

"The main challenge ... is institutional settings in which institutional and provider IDs, rather than individual prescriber IDs, end up being collected," Dr. Milstein said.

With such challenges to implementation, CMS adoption of a physician efficiency comparison tool could be a long way off.

"This is all fantastic, and I wonder if I'll be alive by the time it's finished," said Robert Reischauer, PhD, Urban Institute president and vice-chair of MedPAC.