In CHECKBOOK magazine, we try to give you as much information as possible to help you in selecting and working with the most important type of service provider in the health care field: physicians.

We tell you which physicians are most recommended by their peers and how physicians are rated by patients. We advise that you check for hospital affiliations, medical school teaching responsibilities, use of electronic medical records, and other indicators of quality.

In addition, we consistently recommend that you select physicians who are "board certified." What does board certification mean, and why is it important?

In the U.S., there are 24 medical specialty boards, such as, the American Board of Thoracic Surgery and the American Board of Internal Medicine. These boards certify physicians in various specialties and subspecialties. For instance, the American Board of Internal Medicine certifies physicians in the specialty of internal medicine and in 17 subspecialties and areas of special qualifications-including cardiovascular disease, gastroenterology, geriatrics, and hematology.

To become certified, a physician must spend several years (in some cases, more than six years) after medical school getting supervised, in-practice training.

In addition, all specialty boards require passage of a written exam, completed without assistance, usually administered interactively by computer in a secure testing facility. Some specialties also require an oral exam. The exams are intended to assess medical knowledge and clinical judgment.

Doctors whose certificates are time-limited must successfully complete re-certification requirements or they can't call themselves board-certified. The requirements for re-certification, like the requirements for initial certification, include an unassisted written (computer-administered) exam, intended to measure clinical knowledge and judgment. Since the individual specialty boards develop their own exams for certification and re-certification, the validity of the exams as measures of physician competence varies by board. On the exam for re-certification in general internal medicine, the largest specialty, the pass rate has been in the 85 to 92 percent range in recent years, but the rates in other specialties may be higher or lower.

In the good exams, a test-taker is confronted with real-life situations where knowledge of current medical guidelines, along with good judgment, can be expected to lead to correct answers. Here is the type of case you might find on an exam: a 22-year-old male college student came to the doctor's office after fainting for less than a minute during wrestling practice; the patient had suffered previous episodes of lightheadedness and at age 10 was diagnosed with a heart murmur. The test-taker is given the results of several in-office examinations. Then the test-taker is expected to make a diagnosis.

• Show evidence of knowledge and judgment-typically by passing the same types of unassisted, written (computer-administered) exams that have been the basis for certification for many years.
• Show evidence of a commitment to lifelong learning and involvement in a periodic self-assessment process. Boards have identified or created learning materials and computer-based interactive tools that a physician can use to learn about the newest developments in his or her field. The physician is expected to self-administer tests to identify knowledge gaps.
• Show evidence of self-evaluation of performance in practice. For example, the American Board of Internal Medicine (ABIM) asks that a physician periodically pull data from case records for patients with a specific condition like diabetes or asthma, submit the data to the board for evaluation, get back from the board a report comparing the physician's practice patterns to national guidelines, develop a plan for improvements, and then measure whether the improvements have worked. Through this process, a physician might discover, for instance, that he or she has information on LDL cholesterol results for too few diabetes patients and this awareness might prompt the physician to take steps to do better. Physicians are also encouraged to do surveys of patients and to use the results to guide quality improvement.

Except for the written tests of knowledge and judgment, re-certification is not dependent on the results-the scores-from the assessment activities. It is enough that the physician does the self-assessments and develops plans for improvements-regardless of how bad or good the physician looks in the assessments.

So, what useful information does the board certification system provide you?

You can check whether any physician you are considering is board certified by checking the website of the American Board of Medical Specialties, the umbrella organization for the 24 individual specialty boards, at www.abms.org. If the physician you are considering is not board certified, he or she will not be listed.

Unless you have a compelling reason to do otherwise, it is hard to see why you would choose a physician who is not board certified. But be aware that board certification is not a very discriminating measure. About 87 percent of physicians in the U.S. are certified.

Also, be aware that for many physicians being certified does not assure you that he or she has kept up-to-date: there are still many physicians in practice who were certified for life, before the change to time-limited certification and periodic re-certification took effect. All of the specialty boards allow physicians who were certified before time-limited certification began to call themselves "board certified."

Knowing how recently a physician has been certified or re-certified is important, since there is substantial evidence that physician performance gets worse over time. A 2005 article published in the Annals of Internal Medicine reviewed 62 studies that had examined the relationship between age or time in practice and various measures of quality-physician knowledge, adherence to recognized care guidelines, and medical outcomes. Of these studies, 45 reported decreasing performance on some or all quality measures over time, and only two reported improved performance.

Physicians who were certified for life can if they wish seek voluntary re-certification, but not many have done so. Those who don't point to the cost, the time required, the risk of failure, and other factors. For the relatively few physicians who have voluntarily become re-certified despite having a life-long certification, that diligence and self-scrutiny may be a meaningful indicator of quality.

It would be desirable for the specialty boards to create and advertise a concept like "recently certified" or "certification updated" so consumers could easily distinguish physicians whose certification status is based only on having been certified for life from those who are actively engaged in self-assessment and practice improvement. Unfortunately, the www.abms.org website, which reports on all specialties' certifications, doesn't provide information on when a physician was certified or whether the physician has voluntarily become re-certified. Some of the individual specialty boards do provide that information on their websites, but some provide virtually no consumer information. The ABMS and all of the individual boards should have the information on their websites.

Are there additional types of comparative information specialty boards could provide consumers on individual physicians? Possibly.

First, it might be possible in the future for the boards to release to the public physicians' scores on the written exams of knowledge and judgment-at least information like "top 10 percent," "top 25 percent," or "top half." The boards can't unilaterally begin to report data that they have historically promised physicians would be kept confidential. But changes in confidentiality policies might be possible in the future. And even in the near term, the boards might consider releasing scores for physicians who give permission for such release. Disclosure of test scores would certainly put physicians out ahead of most professionals; it is not common to have one's personal test scores made public. But the advantage of releasing these scores is that they can be expected to reflect on aspects of quality that other measures of physician performance may miss.

Second, the boards might consider, for the future, working with physician leaders to change the self-assessment process into a process that has the combined purpose of self-assessment and public assessment. There are already various efforts underway-led by government agencies, health plans, employers, and consumer organizations-to increase public reporting of physician quality measures-for instance, measures from medical records showing whether a physician consistently gives all the appropriate tests and treatments to diabetes patients, or measures from patient surveys of how well the physician communicates. Some of the specialty boards are involved in these public measurement efforts, but more might be done to use the same data and analyses for public reporting and physician self-improvement purposes.

If measures are to be useful for public reporting, they will have to be standardized and independently collected. Currently a physician can self-select the diabetes cases he or she abstracts for the specialty board and can even cherry-pick to look good, or can decide to survey only patients who are likely to give good reports. There is not much harm in that when the data are being used only for self-assessment; worst case, when the physician gets a performance report back from the board, he or she might simply say, "Wow, I did that badly even though I was cheating."

For public reporting, the system would have to change. The specialty boards would have to work with health plans, Medicare, and others to develop systems to collect data from medical records or claims records, for example, and to develop a nationwide standardized patient survey-so that the resulting measures could be used not only for physician practice improvement but also for public reporting. Having forward-thinking specialty boards involved in the development of measures might help to move the measurement process forward and to assure that measures are well designed.

Public reporting of standardized measures would not only help you choose good physicians, but would also enable health plans to reward high-rated physicians and would facilitate public and peer recognition of top performers. All that might reinforce incentives for the kind of physician quality that is the mission of the specialty boards.

If the boards don't get actively involved in public reporting efforts, however, it is still important to remember that what is being done, and the progress that has been made toward continual maintenance of certification, is important. Even without more public reporting, board certification provides tools for quality improvement, channels through which well-motivated physicians can fulfill a desire for professional improvement, and a way for each physician to demonstrate to patients and the public that he or she is committed to professional development.

It is important to realize that the commitment to professionalism and the desire to help others-not public scrutiny-have probably been the most important forces for quality in health care over the years. So fostering these motivations is a good thing.

Return to top of page