Physician Compare
Bookmark and Share

Laudable goal, long way to go

The reporting of quality measures on Physician Compare, a Centers for Medicare and Medicaid Services (CMS) website to help patients learn more about physicians, has raised concerns about the website’s integrity and accuracy as well as the potential for the data to be misinterpreted and misrepresentative, especially with regard to dermatologists. As a result, the site’s ability to empower patients to make educated decisions when choosing physicians — although a laudable goal — is being questioned, as well.

Oliver Wisco, DO, co-chair of the American Academy of Dermatology’s Performance Measurement Task Force, summed it up as follows: “It’s a good start and a worthwhile process, but people should be leery at this point.”

As required by the 2010 health system reform law, CMS enhanced its Physician Compare website in December 2010. Individuals can currently search for a physician by specialty, location, gender, and whether or not the Medicare amount is accepted as full payment. Other information available includes group practice locations, hospital affiliation, education, and languages spoken. This information is gleaned from the Medicare Provider Enrollment, Chain, and Ownership System or PECOS. Currently, the site lists providers who participated in the Physician Quality Reporting System (PQRS) in 2009. But beginning Jan. 1, 2013, it will publicly report physician performance derived from data gathered in 2012.[pagebreak]

Inaccurate data

But even before Physician Compare has started to report quality data, the site has been criticized for containing information that is either incorrect or lacking. According to the American Medical Association, common errors included name misspellings, the inclusion of retired or deceased physicians, and inaccurate Medicare participation status.

“Concern about the integrity of the site, at this point, is quite valid considering how many mistakes were found,” noted Marta Van Beek, MD, MPH, chair of the Academy’s Congressional Policy Committee. “The more complex the information, the greater chance there is for mistakes,” she added.

The concern is not as much with Physician Compare introducing new inaccuracies, but with the website displaying inaccurate information pulled from the PECOS and PQRS, said Jack S. Resneck Jr., MD, chair of the Academy’s Council on Government Affairs, Health Policy and Practice, and associate professor of dermatology and health policy at the University of California, San Francisco School of Medicine. “The larger issue is whether CMS will be able to accurately measure and report information on physician quality measures through PQRS,” he said.

Another concern is the ability to review and appeal inaccurate data reports from CMS, Dr. Resneck added. Correcting information on PECOS and PQRS has proven to be a frustrating endeavor. Inaccuracies found in demographic information are easier to correct than inaccuracies in performance assessment, said Louis Kuchnir, MD, president of the Massachusetts Academy of Dermatology.

If the website is not executed with a high degree of accuracy, it has the potential to do more harm than good, Dr. Kuchnir added. “Even if it’s done accurately, the unintended consequences remain to be discovered.”[pagebreak]

Inaccurate interpretation

Those unintentional consequences include quality data being misinterpreted and misrepresentative not only for dermatologists, but for all physicians. “For example, there is concern that if a ‘good’ physician chooses not to participate in performance measurement, he or she may be perceived as a ‘bad’ doctor, which is an unfair conclusion,” Dr. Wisco explained. He believes that unless the site includes guidelines explaining how the data should be interpreted, such as the disclaimer on the Physician Profiles website run by the Massachusetts Medical Society, the data isn’t necessarily useful.

For example, a physician who provides excellent quality care may postpone submitting data to PQRS for a year until he or she purchases an electronic health record to simplify the data submission process, Dr. Van Beek said. meanwhile, that physician may be viewed as providing poor quality care. however, she said, the mere act of reporting doesn’t correlate with the quality of one’s performance — a concept that the average patient using this site may not realize. consequently, the website may give the public a false sense of quality.

Another issue is the lack of quality measures applicable to dermatology in the PQRS. “As a small specialty, we continue to face challenges getting our proposed measures endorsed by organizations such as the National Quality Forum, which CMS often looks to when it considers PQRS measure adoption,” Dr. Resneck said. The Academy, which continues to develop evidence-based measures for inclusion in the PQRS beyond the three melanoma measures already included, anticipates the addition of a fourth measure related to biopsy follow-up in 2012, he said.

Because virtually none of the PQRS measures are relevant to dermatology practices, Robert Swerlick, MD, the Alicia Leizman Stonecipher Professor and Chairman of dermatology at Emory University School of Medicine in Atlanta, doesn’t believe that Physician Compare will have a significant impact on dermatologists in the short term. Without the addition of dermatology measures with hard endpoints, the website lacks information that would allow the public to distinguish one dermatologist from another, he said. Such measures would include, for example, the appropriate use of medications as well as screening for bone health and infectious diseases for patients with chronic skin diseases prescribed corticosteroids; or the rate of post-operative complications and recurrences within acceptable limits for patients who underwent surgery. “That’s the kind of information, along with patient-driven information, that would help the public make informed decisions,” he said.[pagebreak]

“The good news for dermatologists is that we’re probably dealing with low-frequency bad outcomes, but it’s not no frequency,” Dr. Swerlick continued. “So you have to collect a lot of data to see if the endpoints fall outside the standard. This type of measurement happens more readily with hospital-based procedures and is much harder to do in an ambulatory environment.”

Because the majority of the PQRS measures are not applicable to dermatologists, anyone comparing the number of measures reported by a dermatologist versus an internist would definitely notice the difference, Dr. Van Beek said. “It would give a false impression that the dermatologist isn’t providing a high quality of care.”

Mary Maloney, MD, chief of dermatology at the University of Massachusetts Medical School in Worcester, knows firsthand the consequence of lacking a robust number of patient cases to accurately reflect quality of care. Although her experience is related to tiered insurance products for state employees, the Physician Compare website has similar ramifications, she said.

As a Mohs surgeon, Dr. Maloney treats few medical dermatology patients. “I work hard to keep costs down. I prescribe generics. I don’t schedule extra visits,” she said. “I do all the things the insurance companies ask me to do.” And yet Dr. Maloney has been placed in either tier two or three, depending on the insurance company. When she was finally able to reach an individual at the insurance companies, Dr. Maloney was told that they don’t have enough information about her treatment of medical dermatology patients to know if she meets all their cost-efficiency and quality ratings. Additionally, the insurance companies lack criteria for Mohs surgery. Dr. Maloney’s placement in a lower tier requires patients to pay a higher co-pay to see her. “They don’t have enough information about you, so the assumption is that you must be bad,” Dr. Maloney said, adding, “I haven’t found one dermatologist yet who is in tier one, and I have asked many.”

People assume that physicians in a higher tier off er better quality of care than those in a lower tier, Dr. Van Beek noted — just as they will assume that physicians who don’t report quality measures on Physician Compare provide lower quality care. To confuse either of these with quality will have horrible consequences for physicians, she said.[pagebreak]

Risk adjustment

Currently, PQRS measures are not risk-adjusted for different patient populations. “A major concern is that unadjusted quality measures would penalize those physicians who care for the sickest or most vulnerable patient populations,” Dr. Resneck said. For example, dermatologists in academic centers tend to see a lot of complicated patients who are referred.

“Some physicians see a subset of patients who are sicker than those seen by other physicians,” Dr. Van Beek said. “No matter what quality of care they provide, those physicians will likely have worse outcomes because their patients are sicker to begin with.”

Furthermore, Dr. Van Beek and others are not convinced that CMS or any private insurance companies currently have the statistical knowledge to risk adjust measures. Dr. Maloney pointed out that it would be very labor-intensive and costly to do that type of risk assessment across all of medicine, let alone dermatology. She believes that eventually CMS and insurance companies will first focus on risk adjusting measures for chronic conditions. “But that won’t filter down to dermatology for a long time,” Dr. Maloney added. Meanwhile, the concern is that a lack of risk-adjusted measures will discourage physicians from seeing complicated patients.

Wrong focus

Another concern is that the measures used to evaluate dermatologists will rely too heavily on patient satisfaction and/or compliance because tools to measure these factors are readily available. Although Dr. Swerlick believes that it’s good for physicians to get feedback from their patients, holding physicians accountable largely on the basis of patient satisfaction is not. He noted that unhappy patients are often more motivated to respond to surveys. “They may not be unhappy with the care, but unhappy because they have a chronic condition that is difficult to treat,” Dr. Swerlick said.

Similarly, patient compliance does not necessarily correlate with the quality of care provided. While it’s important for physicians to explain why patients should follow their treatment regimen, Dr. Van Beek said, other factors often interfere with patient compliance. For example, some patients can’t afford their medications, don’t have the time to appropriately follow through with treatment, or have work obstacles that interfere with their ability to see their physician. As a result, their disease often fares worse than for those who can afford their medications as well as have the time to stick to the regimen and see their physician for follow-up. “All of these factors will result in low patient satisfaction and compliance, which has nothing to do with the care being provided,” she said. [pagebreak]

Laudable goal, difficult task

Many dermatologists agree, in principle, that physicians should measure for quality and the public should have access to that information. “But when you start to drill down to the logistics of how to get this done, it’s really a very daunting task,” Dr. Swerlick said.

Physician Compare’s impending move from being a provider directory to a disseminator of quality reporting information doesn’t appear as if it will improve the scenario for either physicians or patients, he said. “Patient advocate groups are very disappointed with the site because it essentially provides nothing more than you can find in the Yellow Pages,” he said, adding, “There’s only one thing worse than no data, and that’s bad data.”

Dr. Kuchnir concluded, “There’s a belief that data have been collected, or could be collected, that will help patients make wiser decisions about who their specialist should be. But patients tend to rely on referrals from peers or their primary care physician to make those decisions, and that’s hard to improve upon even with sophisticated data mining. One day that website may be the way to pick a physician, but it has a long, long way to go.”