Family physician Kevin B. Martin, MD, offers his patients top-tier medical care in Auburn, Wash. Or maybe he doesn't. Or maybe he didn't at one time, but he does now. It all depends on which health plan you ask – and when.

Martin is one of many doctors who have come away shaking their heads after payers have rated the efficiency and quality of their practices based only on claims data and then have shared the results with patients. The goal of the approach, which some call “tiering and steering,” is to cut costs and encourage quality. Martin just calls it frustrating.

“I am the same doctor,” Martin says. “I give the same care with the same instruments and the same hands in the same rooms regardless of the payer. I'm not a good doc one day on one plan and the dregs another day on another plan.”

How could one payer have designated Martin as one of its top physicians while another sent a letter to his patients informing them that his practice had not met its standards of quality and efficiency (a designation that his practice successfully protested)? It comes down to claims data and how it is used. This article describes how and why insurers are using this data and the potential effect on family physicians.

The evolution of tiered networks

Health insurers' practice of rating physicians' performance based on the cost and quality of the care they provide, often referred to as “physician profiling” or “economic credentialing,” is nothing new. In the past, payers used the practice to justify terminating high-cost physicians from their networks. More recently, payers have used computer programs to analyze physicians' claims data and assess both the quality of their performance and their cost-efficiency relative to their peers'. (See “Payers with tiered networks.”) In some cases, this information is incorporated into a pay-for-performance bonus program. (See “What Family Physicians Need to Know About Pay for Performance,” FPM, July/August 2006.)

A growing number of payers are also using the data to guide the development of “tiered networks” that encourage patients to choose selected providers. Payers use their cost and quality ratings to divide physicians into two or more groups (“tiering”) and make the ratings apparent to patients, for example, by putting a star next to the names of the “better doctors” in their plan directories. “Steering” – offering patients lower co-payments or co-insurance percentages for office visits with “high-performing” physicians – is an emerging strategy that health plans have not commonly applied to their primary care networks.

How profiling is done

Payers rely on their computers to generate physician profiles. Most use one program to analyze physician cost and another program to assess physician quality. Both types of software rely solely on claims data.

To look at physician costs, payers use “grouper” software, which groups a patient's claims over an episode of care. All costs attributed to the diagnosis that defines the episode of care are typically included, including outpatient, inpatient, radiology, lab work and prescription drugs. One deficiency in the grouper methodology is that it does not reflect the number of episodes of care that may have been prevented by primary care interventions. For example, if an intervention prevents complications of diabetes or the onset of congestive heart failure, its impact isn't measured.

To assess physician quality, payers use software that applies evidence-based guidelines and consensus-based quality standards to claims data to determine physicians' adherence to such measures. For example, the software would assess whether at least two A1C tests were done in a 12-month period for patients with diabetes mellitus. Health plans typically establish their tiers based on differences in the cost of care among physicians. They use quality ratings less often, primarily because quality is much more challenging to evaluate with claims data.

Concerns about claims data

Many physicians feel that claims data is insufficient for measuring performance. “The payers assume that the data collected accurately reflect the patient's clinical picture,” says Martin, “but that isn't always the case.” This is just one of a number of concerns associated with using claims data for performance analysis. Other issues include the following:

• The volume of claims analyzed can be too low to produce statistically valid results.

• Claims data can be incorrect. For example, a patient may be attributed to a physician he or she has never actually seen. Or both genders may be included in the denominator when a gender-specific service is analyzed.

• Claims data are incomplete. Payers may not have all of the pertinent claims data because of how lab and radiology tests are billed and pharmacy and mental health “carve-outs.”

• Claims data don't reflect clinical outcomes or how well patients adhere to physicians' treatment plans.

Bruce Bagley, MD, the AAFP's medical director of quality improvement, cautions family physicians against seeing only the negative in the use of claims data. “Although these are all valid concerns, the current methodology offers data for physician quality evaluation where there was none before. This is very threatening, but the unemotional response is that we need to look at the data and see if we need to do something about it in terms of improving care.” The majority of the activity reflected in the claims data is within the physician's control, Bagley says.

Physician incentives

Some payers provide financial rewards to physicians designated as high-performing. For example, United has a pilot program called Practice Rewards that offers a fee schedule increase of 5 percent to practices meeting certain criteria. The program includes a majority of the physicians who have received United's “Q&E” (quality and efficiency) designation.

In addition, health plans believe that physicians designated as high-performers will increase their market share of patients, which will lead to more volume and more income for their practices. However, many primary care physicians have already maximized their patient panels and are not prepared to accept more patients. They want to be able to take better care of the patients they already have rather than having to take on more.

There may come a time when physicians with the lowest quality and efficiency ratings will be pressed to demonstrate to the payer why they should be included in the payer's network, or they may be offered deeply discounted fees that make it difficult for them to accept the plan's contract and be financially successful.

Stock your practice with data

Accurately rating physician performance is important, but to date the methodologies used have contained many flaws. Until better methods are developed, efforts to reduce health care costs will continue to drive payers to use claims data to assess physician performance. Physicians will be ranked, and patients will be offered financial incentives to use the “high-performing” physicians.

The best advice for family physicians is to be prepared. Measure your practice's performance. Respond to any payer performance reports you receive if you disagree with them. It may go without saying, but effective use of a fully functional electronic health record system will help you to track recommended care for your patients. It will also give you the data you'll need to challenge payers. You might not be able to stop them from ranking you, but you might succeed, like Martin's group did, in challenging a ranking with which you disagree.