By John Reichard, CQ HealthBeat Editor
October 29, 2007 -- A leading national health insurer announced Monday that it has agreed to seemingly tough state oversight of what could be a hot new tool in the consumer marketplace: ranking doctors based on their efficiency and quality.
The agreement between Cigna and New York State Attorney General Andrew M. Cuomo involves only one state and one insurer. But with private insurers and the federal government eyeing the use of the rankings to restrain rising health care costs, the deal could be a national model. Cuomo and Cigna touted it in those terms Monday and the American Medical Association followed suit, with president-elect Nancy H. Nielsen saying it "will influence other states to implement careful and independent oversight and evaluation of physician performance measurement projects."
The practice among insurers of ranking doctors is growing rapidly. But the pact "marks the first such agreement between a major insurer and a state attorney general," according to a press release from Cuomo's office.
With a number of studies concluding that much of U.S. health care spending is wasted on unproven procedures, analysts see a big opportunity to get more bang for the health care buck by focusing on the prescribing patterns of doctors. They hope that data on quality and efficiency will bring new "value" to health care spending by steering patients to more skilled providers and by encouraging their more wasteful colleagues to adhere to practice norms to produce the best medical outcomes at the lowest possible costs.
But doctors warn that poorly designed systems could lead them to avoid costly patients who could reduce their rankings.
This summer, Cuomo sent letters to Cigna and other insurers expressing concern about their ranking methods. An Aug. 16 letter to Cigna noted the insurer's plans to market to employers a network of specialists picked based on their performance on quality and efficiency measures. "Employers who have selected the Cigna Care Network may create financial incentives, such as reduced copayments or deductibles, to encourage their employees to use that network," the letter noted. But depending on how the rankings are constructed, "consumers may be encouraged to choose doctors because they are cheap rather than because they are good."
The letter cited a number of potential flaws with ranking systems. For example, using data based on payment claims might omit necessary clinical information contained in medical charts. "It may be necessary to audit or validate claims data, even on a random sampling basis," Cuomo's office said. In addition, the claims database may be too small to generate reliable rankings, and the number of patients per physician may be too small to yield meaningful results. "Because several physicians may treat the same patient during the course of a single episode of care, it may be unfair to attribute to one of these several physicians all care rendered by those in the group," the letter added.
"Given these risks, we were surprised to learn that you initiated the Cigna Care Network without disclosing the data you used to rank the doctors, even to the doctors themselves. As a result, doctors and consumers have no ability to bring errors in the rankings to your attention so that they may be corrected," the letter said.
Cuomo said Monday that the agreement with Cigna culminated in a series of negotiations addressing key problems with rankings. "By working together with Cigna, consumer advocates, doctors, and medical societies, we have been able to create a new template for ranking programs that the entire industry can follow."
Under the agreement, Cigna said it will ensure that rankings are not solely based on cost. The deal provides that "in information for consumers and public reporting, measures of cost-efficiency and measures of quality of performance shall be calculated separately and disclosed as such. To the extent the individual scores for quality of performance and cost efficiency are combined for a total ranking, the proportion of each measure shall be clearly disclosed." Where available, Cigna said it would rely on measures developed by the National Quality Forum, a public-private group that issues measures based on a consensus-setting process involving providers, insurers, consumers, and government officials. Where NQF standards aren't available, the insurer said it would rely on those developed by other consensus-forming groups or other "bona fide nationally recognized guidelines."
To judge efficiency accurately, the agreement provides for the use of computer software that takes into account the medical complexity of the cases treated by an individual physician. In addition, Cigna "shall compare physicians within the same specialty within the appropriate geographical market," the agreement states. The insurer said it would disclose the number of cases on which ratings are based in order to assure accuracy in sample sizes. It will explain how it decides to which doctor a patient will be "attributed" for ranking purposes when more than one doctor is involved in treating the patient. It will disclose to doctors how the rankings are designed and give them a chance to appeal incorrect rankings. And the company said it would pay for a "Ratings Examiner" responsible for overseeing compliance with the agreement and for reporting to the AG every six months on compliance.
Under the agreement, the Ratings Examiner must be an independent "national standard setting organization." And Cigna Chief Medical Officer Jeffrey Kang said the insurer "will also contribute up to $100,000 to an independent organization to develop better means of communicating to consumers all aspects of the ranking program in a clear and straightforward manner."
AMA's Nielsen said "this agreement is important because it establishes a process that seeks to guard against some of the risks inherent in physician performance programs run by health insurers. A lack of scrutiny has allowed health insurers to unfairly evaluate a physician's individual work using an insufficient number of patient cases, questionable quality measures, and poor adjustments for risk."
