Physician Compare is a little gem hidden in the giant clam shell of the health reform bill. It has tremendous potential to dramatically change the way Americans choose their doctors — in the future. But that potential won’t be realized without bold leadership from government officials (at CMS) charged with building the site and the commitment of the physician community to full transparency and accountability.

The site, for now, offers up a useful directory of doctors and other providers — 932,000 of them — who serve Medicare beneficiaries. The health reform law specifies that quality measures consumers can understand and base choices on be posted on the site starting in 2015.

We have a lot of work to do before then. Among the major challenges:

• The directory must be accurate. At a public comment session on Physician Compare last October, CMS officials declined to divulge the error rate in the current provider database — names, addresses, or specialty listed incorrectly — but the error rate is believed to exceed 10%. If that’s true, it will undermine trust in the site and get Physician Compare off to a poor start. CMS this year should launch a comprehensive review of the directory’s accuracy and report the results. When quality and performance data get added to the site, doctors will need a clear review process that adheres to the standards set out by the “Patient Charter for Physician Performance Measurement, Reporting and Tiering Programs” developed by the Consumer-Purchaser Disclosure Project in 2009.
• Solving the “low numbers” problem. Physician groups raise legitimate concerns about the validity of some clinical measures as applied to individual doctors, instead of at the group practice level. The numbers on some measures will simply be too low to allow for statistically meaningful results. Work on this problem is under way, including assessment of whether electronic health record-derived data on individual doctors will be ready for prime time in 2015. The process must, of course, be fair to doctors. However, the physician lobby cannot be permitted to undermine Physician Compare with endless arguments about methodological issues, risk adjustment and statistical validity. Perfect measures will never exist.
• Getting to quality measures that matter to consumers. The “meaningful use” measures developed for phase 1 (2011) of the Medicare EHR incentive program and those being developed for phase 2 (2013) are focused on areas poorly measured in the past but of great importance to most Americans: patient safety; care coordination; patient and family engagement; and appropriate treatment. The government needs to build on this work to offer consumers an easily understood “dashboard” of measures. Consumers need bottom-line information and data on the outcomes of care and on how a doctor stacks up against his or her peers.
• Should people be allowed to rate their doctors? Yes. A batch of websites now permits consumers to “rate” (and whine about) their doctors. The volume of such ratings is increasing rapidly, but most doctors rated on such sites don’t yet have enough reviews to make the results statistically valid or meaningful, and the input can be manipulated. Even so, there’s potential here. The government should experiment with creating a section on Physician Compare that allows consumers to provide reviews of their own physician(s) using a standardized online survey tool. Initial results from this effort could be analyzed closely before they are publicly posted. If executed with care, such consumer reviews could become an important adjunct to other data and information on physician quality and behavior.
• Educating the public. The government needs to start developing a marketing plan for Physician Compare now. The agency has shown — with its rollout of the Medicare Part D prescription drug program — that it knows how to get information about new tools into the hands of millions of beneficiaries. Still, Hospital Compare, CMS’ website that presents comparative hospital data, was never marketed, and the vast majority of Americans (90% in one recent survey) do not know about it. One new model: healthcare.gov, the government’s new site aggregating advice and data on health insurance. More than four million people have visited the site since its October 2010 launch.

Physician Compare is being launched as a confluence of forces propels us into a new era of health care accountability. Implemented well, it has the potential to be a game changer for consumers.

Will Physician Compare deliver accurate, actionable information to seniors and change health care behavior constructively … or will it waste a lot of money chasing a fundamentally flawed hypothesis and damage the fragile trust needed for meaningful reform? I think it’s headed for a muddy rut.

We struggle with two competing beliefs.

One holds that if consumers review specific measures of individual physician performance, they will reward superiority with revenue-generating service opportunities. Meanwhile, professional pride and business imperatives will motivate the underperformers to catch up. Makes sense in an Adam Smith kind of way.

The other belief contends that individual doctors are obviously important but not often the sole determinants of overall quality, access and affordability of care. Rather, health outcomes are determined by the system that supports the individual doctor’s work with sophisticated information systems, instant communications channels, care management personnel, constant measurement, direct feedback, support for improvement and aligned incentives. A public, individual reporting scheme will backfire if it is so inaccurate that physicians disregard the content, if it offers measures of little utility to actual patients and if it attempts to use competition to alter an environment where primary care physicians are rapidly growing scarce.

CMS has two horses in the race — the individual reporting intuition with Physician Compare and the system integrity notion with accountable care organizations. It’s hard to ride two horses, but I believe they can be harnessed to the same cart.

In California, we’ve had the opportunity to watch these dueling concepts jousting with the California Physician Performance Initiative experiment. It’s a long story, but in brief, the premature publication of remotely collected individual doctor reports — known to be inaccurate by as much as 30% — over the unanimous objections of the delivery system physicians on the advisory board resulted in a black eye and a lawsuit. It damaged precious trust between purchasers, insurers, delivery systems and individual practitioners. No consumer benefited, and potential progress has stalled.

Aside from the inaccuracy of remote data collection, the problem with single doctor reporting is that in modern California, doctors work in teams and within systems that support the complex functions envisioned for truly functional “medical homes.”  Attributing care outcomes to one person simply harkens to a beloved TV fiction.

Bottom line: What would actually help patients choose wisely for their individual medical and personality preferences, as well as stimulate the delivery system to accelerate improvement?

Three suggestions:

• Score systems, not individual doctors, for relevant quality measures;
• Insist that those systems score individual doctors with highly accurate, locally valid methodologies and use that information to motivate change (a central tenet of the ACO concept); and
• Score and report individual doctors for the traits that patients actually want to know: communication style, electronic access, staff demeanor, timeliness, responsiveness to patient requests for convenience, and cultural and linguistic accommodation.

That third step is extremely difficult, but so was sending a rocket to the moon.

For the past several years, we have had few effective tools in our health care toolbox that improve health outcomes and keep care affordable. Purchasers believe that provider-level data is a potentially powerful, yet underutilized tool to begin to change all that. When consumers have data about the quality of their care, they can choose doctors most likely to meet their needs and send a signal about how they value quality. When purchasers have data, they can begin to use benefit designs to reward high performers. And, when doctors get information on how they actually compare to their peers, they can identify opportunities to improve and redesign their practices to better serve their patients.

Medicare’s new Physician Compare website is a promising tool that could ultimately deliver this performance information. The initial website release has only very basic information, but the reform law requires that better data be added each year, including patient assessments, measures of quality of care and a user-friendly tool to allow easy doctor-to-doctor comparisons. There is enormous variation in how well and in how consistently doctors provide the right tests and treatments.

In California, for example, the best doctors perform tests like mammograms and cholesterol checks 60% more often than the lowest performers — and few of us know which category our doctor falls into. At the same time, doctors control about 80% of all health care costs in this country. Inappropriate care is wasteful, expensive and sometimes a matter of life and death. These facts are not lost on consumers or purchasers, who can no longer tolerate — or afford — either trend.

At the Pacific Business Group on Health, our purchaser members are longstanding supporters of physician-level data. Our California Physician Performance Initiative gathers performance information about California doctors. One cutting-edge health plan — Blue Shield — has already shared this information with its members.

Aggregating data from doctors and insurance companies can be a slow process. Fortunately, Medicare is sitting on a large, readily accessible body of data. Now, we strongly encourage CMS to do right by consumers and employers, by swiftly expanding the consumer information available on the Physician Compare website. When the site is fully operational, it can make data a powerful tool in our arsenal to improve quality and control costs.

Publishing health care quality metrics, as Medicare’s new Physician Compare website will in the future, is a step in the right direction toward increased transparency and accountability in health care, and should help to both educate consumers and give them needed tools for more quality-conscious decision-making.

In California, there has been public reporting of physician organization performance in the statewide Pay-for-Performance program since 2004. Results are published on the Office of the Patient Advocate website and include both clinical quality and patient experience ratings using a four-star scale. The ratings are easy to access — consumers can search by county and compare physician organizations within their geographic area — and presented in a consumer-friendly format. Reporting both clinical quality data and patient experience data provides important information about different aspects of the performance of a provider, as consumers are interested not only in clinical ability but also in variables such as doctors’ communication skills and timely access to care.

While public reporting is important in encouraging consumers to choose high-quality providers, it must be carried out in a way that is both meaningful to consumers and non-threatening to providers. Thus, special attention needs to be paid to making sure that reported data are valid, reliable and fair to those being measured. The California P4P program measures performance at the physician organization level, rather than the individual physician level, which ensures a sufficient patient population size for reliable, trusted results. More granular reporting is more challenging and may yield less reliable results, depending on the size of the population that is being measured. Increased resistance to public reporting may result if providers see reported information as an untrue representation of their overall performance.

Ideally, cost and quality data from all payers — both public and private — and all providers would be reported, which would allow for valid, reliable performance comparisons across both health plans and providers, and would give consumers, purchasers and policymakers robust data upon which health care and resource allocation decisions could be made. Some states have initiatives — called All-Payer Claims Databases — that collect these data, but most do not.

Physician Compare is an important step in the right direction, but the availability of all-payer data would provide a more accurate picture of performance and be of greater benefit to health care consumers.