business
Medical societies demand insurers rethink doctor cost ratings
■ Researchers find current systems to be inaccurate and unlikely to lower health system costs.
By Victoria Stagg Elliott — Posted Aug. 2, 2010
- WITH THIS STORY:
- » How accurate is cost profiling of physicians?
- » Related content
The American Medical Association -- along with 47 state medical societies -- is calling for an outside entity to formally re-evaluate cost-profiling programs because recently published data indicate that those programs may misclassify many physicians.
"There are serious flaws in the attempts to rate individual physicians based on economic criteria," said AMA President Cecil B. Wilson, MD.
The July 19 letter, sent to 47 insurers nationwide, follows a series of studies by RAND Corp. researchers, funded in part by the Dept. of Labor. Those studies concluded, as RAND outlined in a June 21 brief, that current physician cost-profiling records "are not ready for prime time" because of how often cost-profiling systems come to the wrong conclusions.
RAND research in the March 18 New England Journal of Medicine, using data from four Massachusetts health plans to establish a mock cost-tiering system mimicking some now in use, found that up to 22% of physicians were misclassified. A second RAND study, in the May 18 Annals of Internal Medicine, found that between 17% and 61% of physicians would be assigned to different cost categories based on the formula for attributing care-related expenses.
"These tools are pretty young," said John Adams, PhD, an author of the RAND papers and a senior statistician with that organization. "We have to have a serious conversation about how to make them better and if they are really up to the challenge."
To rein in health costs, insurers have established physician tiering or cost-profiling programs to steer patients to clinicians who are believed to provide less-expensive care.
The letter from the AMA and state medical societies called for insurers to demonstrate that physician rating programs used to set up "tiered" or "limited" networks are reliable, accurate and valid. They also should drive quality improvement efforts.
"The studies by the RAND Corp. demonstrate beyond any reasonable doubt that these tiering programs are flawed to their core," said Alice A. Coombs, MD, president of the Massachusetts Medical Society and a critical care specialist and anesthesiologist from Sharon, Mass. "These systems do not accurately measure a physician's performance and should not be used to report on the cost performance. They mislead patients and unfairly impugn the reputation of a great many physicians."
MMS was one signer of the letter. Separately, the medical society has sued the Massachusetts' Group Insurance Commission, which purchases health coverage for many government employees, over its physician profiling program. That case, filed in May 2008, is ongoing.
Medical societies are concerned that a rating can be skewed if a doctor cares for a sicker group of patients or performs very few of a particular procedure. Because the ranking can affect how much is paid out of pocket for care, these programs also may encourage patients to sever long-standing doctor-patient relationships.
"They force patients to make decisions based on co-payments that are associated with a system of tiering of physicians that has poor reliability," Dr. Coombs said. "And if a physician cares for a patient population that is incredibly sick, the cost of caring for sicker patients is going to naturally be higher. ... To place a physician in a tier based on cost alone, that, in and of itself, is unfair."
The letter to insurers is the latest action on the part of physician organizations challenging profiling programs incorporating cost data.
In April, the California Medical Assn. stopped participating in a quality reporting program developed by a coalition of insurers and purchasers in that state because of its reliance on claims data. The "Blue Ribbon" program launched June 1.
"It is still something that we are opposed to and don't think is supported by the data that they are using. We don't think it is an accurate rating system," said CMA spokesman Andrew LaMar.
Insurers disagree
Those speaking for the insurance industry agreed that physician profiles should be reliable and improve health care quality but also said cost was not usually the only evaluation measure. They expressed concern that the models used in the RAND research did not necessarily reflect programs actually used by insurers.
America's Health Insurance Plans responded with its own letter from CEO Karen Ignagni. "Our members agree with the AMA on the need for providing reliable and useful information to providers to stimulate quality improvement and to support patient decision-making. ... We believe that there is need for further research on methods used to measure and assess performance of physicians and other providers. AHIP and its members have a long history of working collaboratively with providers and other stakeholders, and we are committed to continuing these collaborations to ensure reliable and accurate provider performance information."
AMA policy is strongly opposed to using primarily cost measures to develop tiered and narrow physician networks that deny patients access to or attempt to guide them toward certain physicians. The AMA encourages using physician data to improve the quality of patient care and the efficient use of resources in delivering medical services.












