What’s Next For Physician Compare?

Screen Shot 2015-01-23 at 9.33.17 PMOf the many hidden gems in the Affordable Care Act, one of my favorites is Physician Compare.  This website could end up being a game changer—holding doctors accountable for their care and giving consumers a new way to compare and choose doctors.  Or it could end up a dud.

The outcome depends on how brave and resolute the Centers for Medicare and Medicaid Services (CMS) is over the next few years.  That’s because the physician lobby has been less than thrilled with Physician Compare, and, for that matter, with every other effort to publically report measures of physician performance and quality.

I’d give CMS a C+ to date.   Not bad considering it’s the tough task.  The agency has been cautious and deliberate.  But after the many problems with Hospital Compare, Nursing Home Compare, Home Health Compare, and Dialysis Facility Compare—not to mention the shadow of healthcare.gov’s initial rollout—that’s understandable.  They want, I hope, to get this one right from the get-go.  And competition from the private sector looms.

Congress mandated that CMS establish Physician Compare by Jan. 1, 2011 and that an initial content plan be submitted by Jan. 1, 2013.  CMS met those deadlines, albeit with a rudimentary site that launched in late December 2010.  The agency updated its plans in 2013 and 2014, even as it added more content and functionality to the site.

The law requires the site to have “information on physician performance that provides comparable information on quality and patient experience measures.”  That’s to include measures collected under the Medicare Physician Quality Reporting System (PQRS), Medicare’s main quality reporting vehicle, and assessments of:

  • patient health outcomes and the functional status of patients
  • continuity and coordination of care and care transitions, including episodes of care and risk-adjusted resource use
  • the efficiency of care
  • patient experience and patient, caregiver, and family engagement
  • the safety, effectiveness, and timeliness of care

Notably, Congress set no deadline for the site to meet those specifications or be fully operational.

So what’s posted so far?   The centerpiece of the site is a searchable directory of some 850,000 Medicare providers.  That includes most of the practicing doctors in the U.S. with the exception of pediatricians and other physicians who don’t treat Medicare patients.  This database predates the ACA and Physician Compare but its functionality, reliability and accuracy (a big complaint from physician groups) is being gradually enhanced.

Each doctor has his or her own profile page—a significant foundation that could accommodate quality and patient experience data in the future.

Consumers can also search three additional databases on the site.  They identify doctors and other clinicians who participate in (a) PQRS; (b) the Electronic Prescribing Incentive Program; and (c) the electronic health record (EHR) incentive program (also called the meaningful-use program).  About 350,000 physicians and other clinicians participate in the latter.

The bad news: these databases are separate and their content is not integrated.  That makes searching for information on a particular group practice or individual doctor cumbersome and time consuming.  And the databases aren’t user-friendly.  On the plus side, for researchers and health administrators, the databases are downloadable.

In 2014, CMS added performance data from 2012 on diabetes and heart disease care for 66 group practices with 25 or more clinicians.  These entities serve Medicare patients and participate in yet another quality improvement program called the Group Practice Reporting Option. While 66 doesn’t sound like a big number, many of these group practices are, in fact, among the largest in the country.  In aggregate, they care for millions of people.  Among them: Mayo Clinic facilities in Minnesota, Arizona, and Jacksonville, Fla.; the Cleveland Clinic in Ohio; the Sutter Medical Foundation in California; and the Henry Ford Health System in Michigan.

Also in 2014, CMS added to the site performance data for five measures of diabetes care and one measure of coronary artery disease care for 146 of the 360 accountable care organizations (ACOs) serving people enrolled in Medicare. About six million Medicare beneficiaries are enrolled in ACOs.

Coming in 2015 and 2016, CMS pledges:

  • A database of clinicians who participate in Million Hearts, an HHS campaign to enhance heart disease preventive screening
  • A database of doctors who pass maintenance of certification (MOC) exams.  MOC tests assess how up-to-date doctors are in their specialty areas
  • An update with 2013 data on all the quality measures now available on groups and ACOs
  • Expansion of the number of quality measures for physician groups, including 20 PQRS measures reported in 2013
  • A tab on individual physician profile pages indicating participation in the varying quality improvement programs mentioned above.
  • “If technically feasible,” patient experience data for group practices of two or more doctors participating in a Medicare ACO
  • Data on a limited number individual physicians from several disease-specific treatment tracking registries that CMS has sanctioned
  • Composite scores that combine individual quality-of-care measures (for example, of diabetes care)
  • Expansion of the use of the five-star ratings scale to more and smaller physician groups.  The star ratings are already being used for physician groups and ACOs.

There are reasons to be both pleased and frustrated with this progress.  Accurately and fairly measuring physician performance and quality-of-care is simply not easy to do.  Debate continues to rage over whether the care rendered by individual doctors can and should be assessed—since much care today (other than routine) is rendered by teams of providers.  And there are certainly technical barriers remaining to accurate measurement.  For one, a lack of standardization haunts the field.

Not surprisingly, front-line doctors remain skeptical of the whole enterprise. And their trade groups argue vigorously that much physician performance measurement is still not ready for prime time.  In an August 2014 comment letter to CMS, for example, the American Medical Association (AMA) said it would “adamantly oppose the multiple proposals to extensively expand the Physician Compare website, as serious and fundamental flaws and errors remain unaddressed.”

But it’s now clear—and a good thing—that this train has left the station, and that consumers have increasing interest in comparative information on doctors.  A survey of 2,137 adults conducted in 2012 and published in the Journal of the American Medical Association (JAMA) found that one in four people had consulted a physician-rating site when picking a primary care doctor that year.  Five percent had rated a doctor online.  Healthgrades.com, the largest hospital and physician-rating site, says about one million people visit its site every day.

Indeed, one reason CMS ought not to tarry too long with Physician Compare is that physician-rating sites are fast cornering the market.  While consumer reviews of restaurants, hotels, movies and books are fun and useful, reviews of doctors have to date been more specious.  That’s because most of the ratings (on RateMDs.com, ZocDoc.com, Vitals.com, yelp.com and others as well as HealthGrades) are based on low numbers of reviews (usually fewer than ten), meaning the ratings for most doctors are far from statistically robust if not downright misleading.  And there’s still debate on whether the ratings can be manipulated by doctors despite companies’ claims that they protect against this.

It’ll be great once millions of people are reviewing their doctor online, using good survey instruments.  And Physician Compare plans to incorporate such data.  But even so, consumer reviews and patient experience reports don’t assess actual outcomes.  You need to measure what doctors  (and health systems) actually achieve.

Adding to the worry: compared to the big consumer doc review sites, only a relatively small number of consumers so far have ever used the still-fledgling physician performance rating sites sponsored by states, the Robert Wood Johnson Foundation’s Aligning Forces communities, and other local efforts.  In contrast to Healthgrades, for example, these initiatives have not been widely marketed, and word of their existence has spread slowly.

That’s another lesson for CMS.  If the government wants Physician Compare to succeed, it’ll have to market it—and much better than it has Hospital Compare.  Search engine optimization will be critical as will ads social media marketing and buzz.

But my advice would be: don’t do that before you fix the site itself.  Take a page from the terrific job (in the end) that HHS (and contractors) did on healthcare.gov, even as you check out the latest user-friendly templates on the most popular and well functioning websites.  Make it appealing and engaging, and explain what’s on the site and how the data should be used and interpreted.

CMS has to be a fair arbiter and steward in the ongoing debate about accurate physician performance measurement.  But at the end of the day, the public interest is the priority and will be best served if the agency moves aggressively forward with the site, meeting both the letter and spirit of the law.  Physician Compare will be a work in progress for years to come, like many websites, but 2015 should be the year it starts to make an impact.


This piece draws on a policy brief published in December 2014 on the Health Affairs web site and written by the author.  Steven Findlay is an independent journalist and editor who covers medicine and healthcare policy and technology.     

21 replies »

  1. Do you have any evidence to show that above-mentioned methods of measurements (PQRS, MU, MOC, ACO) have demonstrated what patients consider high-value care? The clear answer is “No”. So you advocate using instruments yet to be proven accurate to denigrate physicians who provide service with their patient’s best interest at heart. And, BTW, if this site can identify a better quality surgeon for a consumer, will Medicare pay that surgeon more for that particular procedure? The fundamental problem with quality in medical care comes down to price-fixing. Unless, higher quality means higher revenue for the service, you will never get higher quality. You will get useless metrics, but that will not translate to better patient-care.

  2. Why are consumer, payer and evaluator not the same person? Perhaps then we wouldn’t need an expensive govt agency! If they were the same, bad doctors would be fired by that person and they would wither away. And the converse would be true too.

  3. Shucks, I misspelled his name. Should have been Findlay, not Findley. See why we need journos? Sorry Mr. Findlay, my bad. Maybe we need a Maintenance of Typing Skills too that I can pay into, eh?

  4. Thanks for the article. I thought it pulled the issue together very fairly and with balance. If I were grading journalists, I’d give it an A….and of course it won’t kill me if the story isn’t perfect….whereas my doctor might be killing me slowly (or rapidly) and so I’d kinda like to see how he compares on some basics..

  5. I better appreciate what you have said here and the goals are admirable, but how does one accomplish this goal? It is very difficult. If mortalities are publicized too much surgeons will be incentivized to do only the easy cases and let the hard one’s die. Quantifying risk is an answer, but is so much in its infancy I don’t think it is accurate enough in the majority of cases. How do you judge an Internist? Every patient comes with whole different set of facts so there isn’t enough of a comparison to garner statistical evidence.

  6. Appreciate the comments. “Monitoring” was perhaps not the best word to use in my previous comment. It evokes a “watching over your shoulder” environment…and probably sparked several of the comments. Indeed, some nooks and crannies of medicine do have good systems of checks and balances, and real-time peer monitoring as well as regulatory oversight. Think of hospitals, for example. Treatment decisions are routinely subject to review and second guessing, with many people involved. And states, the feds and the Joint Commission do a lot of good hospital oversight. What this piece is talking about is a system of accountability in which physicians’ care delivery and the results for patients of that care are assessed fairly and accurately….in a way that meaningfully propels care improvement (in health systems as well as individual doc offices) and allows payers and consumers to make choices among health systems and physicians. On a concrete level, if you learn you must have surgery you ought to be able to compare surgeons based on their performance and results as well as other useful parameters (convenience, consumer reviews, referrals from other docs, cost, insurance coverage, the system in which they operate, etc). As I said in the piece, no easy task, and several of the comments hint at this being an impossible task. Not so. There is widespread proof of concept in the U.S. and in other countries. Accountability in medicine is a clear driver of improvement, as it is in many fields. Per one comment, relying on the “wisdom and judgement” of the individual physician will always be a core element of the practice of medicine. But it’s not a strategy for dealing with the now well-documented system-wide problems in the quality of care (with 20 to 30 percent of all care being unnecessary, inappropriate or just plain the wrong thing to have done). As for the evaluation of journalists suggested in several comments, I would concur that there’s no clear systemic approach to this in the media industry. There are almost certainly as many poor performing journalists and media outlets out there as there are poor performing doctors, hospitals and health systems.

  7. Oh please.

    If I published an article arguing that journalists should have a publicly available website that grades their journalism on an A to F scale by a neutral third party for everyone to see, you’d be in an uproar making specious arguments about how “journalism quality cant be measured on that scale”

    In other words, a grading system is OK for doctors but not OK for your profession (or any others for that matter).

  8. So Milleson your argument is that doctors need to be evaluated and journalists don’t?

    Please tell me why.

  9. I appreciate your concerns as we all desire a higher level of knowledge.. When do we have enough? When do we have too much so that the inquiry interferes with what we are getting?

    I find it difficult to believe you think healthcare is “unmonitored” in the way you expressed yourself. Physician actions are among the most monitored actions that exist. They are monitored by the patient, the hospital, the county, the state, the federal government (many monitoring stations there), their peers, attorneys, etc.

    In fact there is a big target on the physicians back that says ‘sue me’ and make millions.

    Please. How can we take you seriously with an attitude that physicians are essentially unmonitored?

    Why not deal in recognizing that there is a lot of monitoring, but the quality of monitoring could be improved. The first step would be for journalists to practice journalism and do independent research to look for the truth instead of satisfying their ideology.

  10. What they measure is kaka, bearing no relationship to the wisdom and judgment of the physician. Feh!

  11. Folly for cases of complex multisystem disease, but probably good for rating the quality of treatment of hangnails.

  12. Could you give an example of a profession that’s monitored the way you would like physicians to be monitored?

  13. Dr. Palmer – interesting thoughts. Agree that medicine is part science/part art. But it shouldn’t be as much unmonitored chance or guesswork as it is now. Although it’s neither easy nor simple, many initiatives are successfully measuring physician performance and treatment outcomes. And there’s every reason to expect we’ll get better at doing it, in a way that will be fair to physicians and meaningful to payers and consumers.

  14. How can you rate something that is half art and half science? How can you rate an effort that often takes years to evaluate and assess the results? How can you rate, fairly, qualities that are partially genetic and inherited? and partly from hard work? How can you rate presidents? How can you rate artists? How about lovers? How would you rate soldiers who give their lives?

    There are too many qualities that are artistic or are inherited that are essential for excellence in doctoring–just as in actong or painting or being a leader or a good husband–to permit fairness or accuracy in real time ranking. We do this type of assessment after someone has died with the aid of historians and biographers who may be old fans or lovers.

    One of the essentials of excellence in direct patient care–you might agree– would be empathy. Does this mean that a doc on the Asperger spectrum can not be a fine doctor? What if he were a whiz scientifically and was a fine diagnostician or surgeon? How would his patients view his work compared to his historica record of real outcomes? Is it fair to rate someone by examining qualities that he has no control over? that may be inherited? Attractiveness? Halitosis? Acne? Deficits in empathy? or I Q?

    This rating scheme has been dreamed by folks who do not understand medicine.

  15. last time I looked, journalists put their mistakes (i.e., “retractions”) on the BACK page, not the front. And bloggers obviously make zero mistaeks . . . [sic]

  16. I guess, Jason, you’ve never heard the saying that journalists put their mistakes on the front page but doctors bury theirs.

    Interesting article, Steve, thanks for sharing.

  17. Why dont I get a “journalist compare” website where I can compare the writing of Steve Findlay against other journalists? Why do journalists get a free pass from quality review measures?

    The federal govt should mandate that all writings by journalist including Steve Findlay get reviewed by neutral 3rd parties and graded on a scale of A to F.

    Only then can the public be reassured that journalists are serving the public good.