Improving America's Medical Education System: Perspectives From a State Regulator

Undergraduate Education/Training

The period when health care professionals first receive their training represents the most important and formative period in their entire career. It is during this period of time that they learn not only the critical subject matter to practice in their chosen field, but they assimilate skills, behaviors and norms that will influence how (and how capably) they practice. This includes everything from how to team with other health professionals to how to practice with compassion and empathy (which has been shown to decline as students progress through medical school1), from cultural diversity awareness to the concepts of patient safety and continuous improvement, and from the “meaningful use” of electronic medical records to the elevated risks of addiction and suicide faced by physicians.

When considering the learning environment for these professionals, there are two elements I believe are essential precursors for learning. The first involves the selection of students based on aptitudes and character as well as on education and training in sciences and other prerequisites. The publication of the Flexner Report in 1910 resulted in the establishment of preparatory course-work for medical students that has changed little in a century. While the value of this was indisputable in raising the overall level of medical education and practice in the United States at the time, in recent years, there has been greater conversation about the ability of students entering medical schools with alternative backgrounds to succeed like their more traditionally-prepared counterparts. As Muller and Kase noted in Academic Medicine, there appears to be no disadvantage to students from non-pre-med backgrounds in terms of their ability to perform relative to their peers with science backgrounds, and they may actually have the benefits of greater communications skills, humanism, and interpersonal skills.2

With a looming physician shortage and the prospect of over 30 million new entrants into the health insurance market in 2014 due to the federal Affordable Care Act, I would suggest that every college graduate with the interest, ability and commitment to pursue a career in health care should be encouraged to do so. We should not retain an artificial prejudice for applicants with science backgrounds. In so doing, we may also unintentionally be excluding candidates with a broader, highly desirable mix of “intangibles.” Certainly, this shift does not come without some restructuring of the existing education and training infrastructure.

The second precursor necessary is a dedication on the part of educational institutions to creating and maintaining a “pristine” environment, free of undue bias or influence on students. Much has been made in recent years of the role of the pharmaceutical and medical device industries in supporting medical educational programs. This can range from such minor influences as “drug lunches” and small gifts to major financial institutional support. While this topic has been well covered by the Institute of Medicine in 20093, it bears a brief mention here.

With respect to the role these organizations play relative to individuals, according to a 2004 New England Journal of Medicine review, the pharmaceutical industry, with nearly 90,000 representatives, spent nearly one-third of its revenue on marketing to physicians, a staggering $8,000 to $15,000 per physician.4 The magnitude of the marketing effort is no less spectacular at the undergraduate education level. A recent review of 33 studies5 found that:

• Between 89 and 98 percent of medical students reported having accepted a lunch or a snack from the pharmaceutical industry. Up to 90 percent of students in their clinical years had received textbooks, journal reprints or similar materials from the industry. Such interactions appear to be most frequent in the United States, with 100 percent of students reporting some interaction, and third-year students averaging one interaction per week.

• While 85 percent of students surveyed in one study deemed it inappropriate for a government official to accept gifts, most studies found that students in their clinical training years felt it ethically permissible for them to accept gifts from drug manufacturers.

• Between 48 and 80 percent of students justified accepting gifts by citing financial hardship or saying that others were accepting them.

• While 67 to 92 percent of students held the attitude that pharmaceutical industry information was biased, students nevertheless sometimes reported (22–89 percent) that such literature was an important part of their education; moreover, students generally did not support excluding sales representatives from the learning environment.

• Most studies stated a majority of students felt they were immune to any bias due to promotion from the pharmaceutical industry, yet 62 to 86 percent of students reported desiring more instruction on how to interact with the drug industry.

• The attitude that drug industry interactions were inappropriate was generally demonstrated to be inversely related to the amount of interactions students had. Also, the extent of student interactions with the industry was directly related to positive views on industry literature and inversely related to the ability to self-regulate interactions with the drug industry.

These sometimes contradictory findings illustrate a problem. At an institutional level, a similarly pronounced industry influence exists, with the accompanying contradictions. Even after years of debate about its appropriateness, a pattern of the pharmaceutical industry providing support to medical residency programs continues. A 2010 study of residency programs found that 53.9 percent of survey respondents reported accepting pharmaceutical industry support, 72 percent of which expressed the opinion that such support is undesirable.6 Interestingly, the study looked at the relationship between programs that accepted industry support and pass rates for the certification examination of the American Board of Internal Medicine (ABIM), and found that “for each 1% decrease in the ABIM pass rate, the odds of accepting support increased by 21%.” Some institutions have taken steps toward eliminating the sway that industry firms have on their campuses, but more needs to be done.

Any learning health care system should be founded on the use of research and information that is not only of the highest quality, but also free of any kind of bias — commercial or otherwise. There is a compelling body of evidence that physician prescribing behavior is influenced by industry marketing. Therefore, it is imperative that health care teaching institutions not only provide students a milieu in which to impartially acquire clinical knowledge, but that they provide students the analytical tools to separate rigorous research from cleverly packaged promotional material.

Of particular interest to me, in my role as a regulator, is whether or not health care professionals, as part of their education or training, gain an understanding of the legal and regulatory context in which they practice. According to the American Association of Medical Colleges (AAMC) 2007 Medical School Graduation Questionnaire, well over half of all responding graduates (61.4 percent of 12,512 respondents) felt that the time in medical education dedicated to licensure and regulation was inadequate.7 According to the 2008 AAMC Liaison Committee on Medical Education (LCME) Medical School Questionnaire, the topics of “medical jurisprudence” and “medical licensure/regulation” comprised only 1.07 percent and .35 percent, respectively, of time in overall medical school curricula. By comparison, for example, the topics of “cultural diversity” and “culturally-related health behaviors” comprised 2.07 percent and 1.74 percent respectively.8

What are the implications of a lack of focus on these areas of physician practice? Consider the work of Maxine A. Papadakis et al., who found in her 2005 New England Journal of Medicine (NEJM) article9 that prior unprofessional behavior in medical school was strongly associated with later state board disciplinary action (three times the odds). The types of behavior included “severe irresponsibility” (more than eight times the odds) and “diminished capacity for self-improvement” (more than three times the odds). These results echo an earlier Papadakis study published in the March 2004 Academic Medicine.10

Another example of the critical impact of the undergraduate medical education experience on physicians' career patterns comes from Tamblyn, et al., who looked at physician-patient communication examination scores as a predictor of future complaints in medical practice.11 Of a study group of 3,424 physicians who took the Medical Council of Canada clinical skills examination, Tamblyn et al. found that low clinical examination scores related to physician-patient communication significantly correlated with an increased relative risk (1.38) of complaints retained for investigation. Complementary findings, though not as strongly correlated, were reported in the Papadakis 2005 NEJM article.12

There is, appropriately, a great emphasis placed on attaining technical skill and expertise in schools and training programs for health care professionals; however, in medical regulation, we continue to see professionals stumble in matters ranging from failing to provide medical records to patients to engaging in sexual misconduct with patients. Perhaps the entire spectrum of negative conduct cannot be entirely “taught out” of students, but avoidance of certain behaviors and promotion of other behaviors, through awareness of the broader context in which they practice, needs much greater prominence in school and training program curricula.

The challenge, therefore, is to substantively integrate all these other critical elements into a curriculum that is already stretched tight. A two-hour lecture on the licensing and regulatory process at the end of a medical student's fourth year is wholly insufficient and too late in the process. It is equally insufficient to cover “soft” topics, such as physician-patient communication, but not require that students demonstrate real proficiency in the skill. The subjects need to be an ongoing focus of instruction, consistently and deliberately modeled to students, and evaluated with the same rigor as more traditional subjects.

In response, we, as regulators, are attempting to establish a greater collaborative presence with educational programs, one that is appropriate for state medical boards to have.13 One example of this is the “Partners in Professionalism” program established by the Ohio Medical Board. The board has worked with educational institutions in its state to present information to students about medical regulation. This includes conducting a medical board meeting at the schools for students to observe. Several other state boards have established similar programs, and in Washington, the Osteopathic Board is working to do so with the local osteopathic medicine institution.14

Finally, one of the most fundamental obstacles in creating a learning health care system is to ensure that the ability for students to actually get a health care education is affordable and attainable. The financial costs of medical education or other health care training can be enormous, and even prohibitive; the average debt for medical students graduating in 2009 was $132,000, with one in four owing over $200,000.15 Further, these costs drive decisions that ripple throughout the rest of our health care system. In the case of physicians, we have seen an overall flight from primary care in favor of specialties that are increasingly more lucrative.16,17 Part of the decision, from a student's point of view, is to financially recover the costs of his or her education. Similarly, when a new physician chooses where to locate, the ability to repay student debt through establishing a successful practice (that is, an affluent urban or suburban area versus a rural/underserved area) is a primary consideration.

I do not mean to imply that these decisions are purely mercenary; on the contrary, choosing a career with some regard to earning potential is practical, especially when the up-front costs are large. While health care has traditionally been seen as a calling (with a different standard than other vocations), it is undeniable that the last four decades have reshaped the profession as a business. This reality leaves us, societally, with issues with which to contend.

Nationally, we have attempted to counter this trend with programs such as the National Health Service Corps (NHSC), which compensates new physicians and other health professionals for practicing in underserved areas. Funding for the Corps was expanded by $290 million through the Affordable Care Act, which is intended to triple the number of participants to 16,000 by 2015. Such programs, I believe, are instrumental in promoting health care access, but they have the added benefit of opening up the field of health care to students who would otherwise be unable to afford to consider it for lack of means. However, it is fair to question whether the additional funding for the NHRS is nearly enough. There is a compelling need, even as the portion of gross domestic product dedicated to health care skyrockets, to broadly re-examine how we remove the significant financial disincentives to students wishing to enter a health care profession.

Again, to some these issues may seem too peripheral to medical regulation. I submit that they are highly relevant: When medical students potentially steer away from their preferred specialty area because of challenges in repaying debt, it creates adverse outcomes for access to care. The nation's physician shortage, especially in primary care, has potentially devastating implications for both medical quality and patient safety.

Graduate Medical Education

Perhaps the most debated issue of the past decade in medical education has been the changes the Accreditation Council for Graduate Medical Education (ACGME) has implemented to the hours that residents may work. Residency hours have been scrutinized from the standpoints of residents' perspectives, patient outcomes, adverse events, compliance and hospital mortalities. Studies have evaluated self-injuries to fatigued residents, motor impairment, attentional failures and whether resident fatigue should even be the focus of informed consent for patients.

To detail the many studies that have been conducted since the ACGME first addressed the residency hour question in 2003 is unnecessary for the purposes of this commentary. However, it is worth noting that in 2007, at the request of Congress, the Institute of Medicine (IOM) considered whether further limitations of residency hours were warranted. The IOM report was a key driver in the formation and work of the 2010 ACGME Task Force.18

The IOM report's thoroughness on the subject seems to crystallize around a single, perfect message. First and foremost, the function of a medical residency must be to create a controlled environment for hands-on clinical training, where new physicians can practice in an increasingly independent manner while minimizing the risk of patient safety errors. The very existence of the medical residency establishes that new physicians are not yet prepared for full and unrestricted licensure, and their practice must reflect that reality. This is the reason for tailoring hours by specialty, the residents' competency, the complexity of patients, and other criteria; it is also the reason to structure resident hours to facilitate patient handoffs and reduce the risk of excessive fatigue.

As a regulator of health professionals such as physicians, I am struck by how differently various states approach residencies. While the goal of the residency is the same — that is, preparing young physicians for fully independent practice — the length of time required by various state boards can vary from one to three years. Indeed, for graduates of U.S. accredited medical and osteopathic schools (i.e., excluding international medical graduates), 72 percent of state boards (47) require one year of post-graduate training, 22 percent (14) require two years, and 5 percent (4) require three years.19

While we must recognize each state's authority to regulate licensure, it does not follow that in some states physicians can learn enough in one year for independent practice while in others they need three years of training. In my view, there is no reason that state boards should not be able to better synchronize across states the requirement for the optimal number of years of residency in order to obtain licensure.

Initial Licensure and Early Practice

In discussing the need for changes to the pre-med curriculum, Ezekiel Emanuel writes in his 2006 Journal of the American Medical Association (JAMA) article:

“By 1905, many recognized that the old heroic treatments — purging, bleeding, cathartics, and proprietary medicines — were ineffective and dangerous. Simultaneously, new science-based practices of bacteriology, antiseptic surgery, and vaccinations were proving effective. The Flexner report codified the need to systematically integrate these scientific advances into the training of physicians.”20

While we can perhaps fault Flexner in hindsight for being too rigid in terms of adherence to a scientific background as the only acceptable preparatory training for a career in medicine, we have to also acknowledge that his message was entirely correct for his day. At a period when medicine was still artisanal, he was able to catapult the profession forward by demonstrating the imperative that the medical profession must universally embrace the latest science, technology and research. That message is still important for health care, perhaps even more so today.

When physicians emerge, freshly minted, from their education and training, they have had the benefit of constant exposure to teaching and mentoring in a structured environment that is flush with the most up-to-date research. For most physicians, there will be no other time when this is so. As physicians — and all health care professionals — make the transition from training to practice, there must be a deliberate effort on the part of practitioners, educational institutions, licensing boards, professional organizations, and employers to establish a habit of continuous learning and inquiry.

However, personal motivation may not prove to be sufficient, and we must conceive of better mechanisms to ensure that new research reaches those already in practice. As the seminal Institute of Medicine report “Crossing the Quality Chasm” cited in 2001, “the lag between the discovery of more efficacious forms of treatment and their incorporation into routine patient care is…in the range of about 15 to 20 years.”21 Such delays between discovery, learning and application in practice are unacceptable.

Just as with medical residencies, we regulators know that states also take different approaches to licensing physicians. Consider how we gauge whether the student has obtained the essential knowledge in order to safely and effectively practice. Physicians seeking licensure in the United States take either the United States Medical Licensing Examination (USMLE) or the Comprehensive Osteopathic Medical Licensing Examination of the United States (COMLEX-USA), which are uniformly accepted across all state licensing boards. However, states can vary in the number of attempts students have to pass an examination, as well as any limits on duration in which to pass the examination.

One feature of medical regulation is that 36 states have composite boards, while 14 states have separate osteopathic and medical boards. Interestingly, a number of states with composite boards (where one might expect greater uniformity of regulation of allopathic and osteopathic physicians) have different regulations on licensing exams, such as Indiana, where USMLE candidates get three attempts per USMLE step, but the board allows five attempts per COMLEX-USA level. In Arkansas, USMLE candidates get three attempts per USMLE step, but there is no limit on the number of attempts per level of the COMLEX-USA. Idaho allows only two attempts per USMLE step, with no limit on COMLEX-USA attempts per level. In Alabama, there is no limit on COMLEX-USA attempts, but USMLE candidates are limited to 10 attempts across all steps. Finally, in a number of states, including California, Florida, Michigan, Pennsylvania, Tennessee and Virginia, there are no limits on the number of attempts a candidate may make at passing the USMLE or COMLEX-USA.22

In the same way, state requirements vary with regard to time limits within which candidates must have completed all steps or levels of the appropriate examination. Some states require that all must be completed within, for example, a seven-or 10-year period. Others specify no timeline; such is the case for the COMLEX-USA examination in Washington. As with the number of attempts question, it is reasonable to ask why there should not be a single, universally-accepted standard that all state medical and osteopathic boards can agree upon.

However, there are ways in which we are beginning to address some of this historical variability. The USMLE recently enacted a policy restricting individuals to no more than six attempts at a step or its component. The only provision for additional access is a request from a medical licensing authority on behalf of the individual. This policy, based on statistical evaluation of examination pass rates, may serve as an appropriate prompt to ease states toward a consensus standard.

The broader point illustrated here is that medicine, and health care more generally, is a profession that requires exceptional knowledge and precision. If state boards share the common goal to protect public health and safety (as I believe they do) by ensuring that physicians and other health professionals meet minimum standards for competency, and these examinations are a yardstick by which we measure that competency, then does such variability make sense? And does a policy of allowing limitless attempts, with no bounds on duration, to pass these examinations further that goal? Once again, it seems worthy of a national conversation to seek a consensus on these issues.

To be certain, by the time a candidate reaches the level of USMLE Step 3 or COMLEX-USA Level 3, he or she has already invested a small fortune in training and made a great personal investment in the outcome of the examination. Even for some well-qualified students, passing the examination may require multiple attempts. At the same time, it is worth questioning whether students who struggle with passing these key gatekeeping examinations are fully prepared to take on the complex and demanding work of caring for patients. Some boards require the candidate, after a certain number of failing attempts, to take more course-work or do additional post-grad training. These are approaches worth considering more globally, but it is also worth considering whether the rigors of a health care career make it the right choice for such students.

As stated earlier, when physicians and other health care professionals enter practice, they experience aspects of their work for which most have not been prepared. There is a need for structured education and training on topics such as teamwork, health policy, reimbursement and health financing, and the legal and regulatory framework within which they practice. With regard to the latter, some states, such as Texas, Oklahoma, Mississippi and Maine, require that physician applicants pass a jurisprudence examination as a condition of licensure.23

These exams typically cover elements of the law related to physician practice in that state, such as important concepts of rules or policies promulgated by the board. While boards typically require a 100 percent score to pass, jurisprudence exams are not intended to “trip applicants up,” and are often open book. In some cases, the exams are offered on-line, with a link to the relevant law, rule, or policy provided, so that the physician can read and understand it as he or she answers. I believe jurisprudence exams are valuable teaching tools and should be incorporated more broadly into licensing practices by state boards.

Mid-Career Practice

Over the past decade, one of the primary topics of discussion within the field of health care has been how to ensure that professionals maintain the skills and knowledge needed to adequately care for patients. It has been well established that clinicians lose their edge at an alarming rate without a constant infusion of new information, research and best practices:

• McGlynn et al. looked at health care in 12 metropolitan areas nationally, with the disturbing finding that patients nationally receive appropriate care only 54.9 percent of the time.24 This finding did not vary across preventative care (54.9 percent), acute care (53.5 percent) and care of chronic conditions (56.1 percent). The analysis of the care provided for 6,712 participants was analyzed against “established national guidelines and the medical literature.”

• Norcini et al. evaluated the effects of experience and education on the treatment of acute myocardial infarction (AMI).25 They found a reduction on the mortality rate of 10 percent for every 16 AMIs a physician treated; however, they also found a .5 percent increase in AMI mortality for every year since the physician had graduated from medical school.

• A review of 62 evaluations by Choudhry et al. found that 32 (or 52 percent) associated decreasing performance with increasing years of practice for all outcomes studied.26 Another 13 (or 21 percent) found a similarly inverse relationship between performance and years of practice for some outcomes but no relationship for others. After controlling for multiple variables, the review examined studies related to a wide range of conditions and procedures as well as operative mortality rates and length of hospitalization.

For decades, continuing medical education (CME) has been the primary tool to impart new practice information and skills to physicians, but there have been challenges to the effectiveness of CME, on two fronts. First, traditional didactic methods of delivering CME have been found to be of limited effectiveness in improving performance.27 Second, professionals lack the skill to effectively self-assess their own performance, cannot reliably identify needed subjects of ongoing education and training, and exhibit a general overconfidence in their level of competency — findings which have been demonstrated for decades.28,29 Consequently, the value of CME may be quite limited as a means to ensure continuing professional competency.

The CME community has taken steps within the last decade to improve the quality and presentation methods of CME. Large theater-style lectures have been retooled in favor of smaller and more interactive sessions. In addition, organizations like the American Academy of Family Physicians and the American Medical Association have developed new approaches to continuing medical education that seek a tighter linkage with providing quality care, such as evidence-based CME, point-of-care CME, and performance improvement CME.30 Nevertheless, there has been a call, and a movement, to enact more rigorous methods of assessment and of assurance of continuing competency in order to protect the public.

For physicians, this work began over a decade ago with the American Board of Medical Specialties (ABMS) establishing its Task Force on Certification, the precursor to its Maintenance of Certification (MOC) program, in March 1998. 31 Even earlier, the American Osteopathic Association (AOA) mandated in July 1992 that the AOA Bureau of Osteopathic Specialists (BOS) create a mechanism for recertification of all diplomates by January 1, 1995.32 This requirement was a first step in establishing the Osteopathic Continuous Certification (OCC) program.

Within the regulatory community, the Federation of State Medical Boards (FSMB) convened a committee on Maintenance of Licensure (MOL) in 2003, and in 2004 adopted the position statement that “state medical boards have a responsibility to the public to ensure the ongoing competence of physicians seeking licensure.”33 After adopting five guiding principles to which MOL should adhere in 2008, the FSMB formally approved MOL in April 2010 as a policy priority for implementation with state medical and osteopathic boards across the country.34

Although separate and distinct, all three programs contain parallel elements. Both ABMS and AOA-BOS identified four key “active ingredients” of their continuous certification programs: 1) professional standing; 2) lifelong learning and self-assessment; 3) cognitive expertise; and 4) practice performance assessment. FSMB's MOL program contains three components: 1) reflective self-assessment; 2) assessment of knowledge and skills; and 3) performance in practice. All three entities have also adopted similar guiding competencies.35

In addition, both AOA-BOS and ABMS have effectively ended the practice of lifetime certification. Recertification under BOS occurs every 10 years, while ABMS recertification is required between six and 10 years, depending upon the specialty. The FSMB MOL program anticipates that states would evaluate candidates for re-licensure periodically, with a time frame to be determined.

Perhaps most importantly, all three programs feature enhanced methods of assessment, training, and evaluation that the learning has been incorporated into the physician's practice — features most traditional CME programs have lacked. This could include comparisons against national quality benchmarks, self-assessment tools, examinations, patient and peer surveys, computer-based simulations, and/or mentorships. While the ultimate goal of creating a learning health care system may be to instill in health professionals, from their initial training onward, an intrinsic and perpetual drive to learn, these assessment and evaluation mechanisms provide motivation to professionals, in the form of continued ability to practice, and greater social insurance toward consistently safe and appropriate care.

There is an important difference between MOC/OCC and MOL. While many physicians hold board certification, some physicians practice in a specialty where no certification exists. Some physicians choose not to seek board certification. Still others hold certification in one specialty but choose to practice in a different field. For these reasons, the reach of MOL must encompass all physicians, not just those with board certification. That said, there appears to be broad consensus that, for many physicians, MOC/OCC activities may be considered by states as sufficient to meet the requirements of MOL, and reasonably so.

Implementing this aspect of a learning health care system will not come without challenges. From the perspective of state licensing boards, some are not even authorized by law to engage in continuing competency, although the public assumes it is being done.36 Some statutes only specifically authorize “continuing education” and not “continuing competency.” In addition, capturing additional demographic and practice information in the application and renewal processes will involve more resources, as will performing periodic analysis to determine whether the necessary continuing competency activities have been successfully completed. This is due to the fact that, for many states, renewal of licenses has become a largely automated process, with little human interaction. Finally, it will be critical to have national organizations step forward in implementing continuing competency to sponsor the educational and evaluative activities and to safeguard the data generated. If such activities were to fall to the states, implementation costs could be prohibitive for many boards.

To summarize, it is evident that physicians (and I would postulate, many other health professionals) lose skills the further they progress in their career from their initial training. It is also clear that traditional methods of lecture-style CME have not had the effect of maintaining physicians' currency, which leads to a degradation in the overall quality of care, and that even enhanced CME may not be sufficient to remedy it alone. Continuing competency initiatives have the potential to reduce these impacts through a more thorough assessment of skill deficits, more effective methods to provide the needed information, and a conscious effort to ensure that the information actually is incorporated into practice. However, continuing competency efforts will require a major investment of time, effort and resources by all parties in order to be successful.

Late-Career Practice, Inactive Licensure, and Re-Entry to Practice

A final category in our multi-stage educational system, with great impact on regulation, is our system for helping ensure the competency of health professionals who have been away from practice and subsequently seek to re-enter the work force. While this movement often occurs in the latter stages of a professional career, it is also common in cases in which a health professional has had children and takes extended family leave for parenting, or a return after a serious illness or injury. However, many state licensing boards have encountered this issue in the context of retired physicians who, for financial or other reasons, return to practice after several years. A recent study of 62 physicians who voluntarily left practice and subsequently had their clinical skills evaluated found the average age of the participants to be 53.7 years.37

Given the rate of change of the vital knowledge to safely and effectively practice as a health care professional, as well as the preceding discussion of continuing competency, there are concerns that such professionals may not be prepared to resume practice. Grace et al. found that, of re-entering physicians who completed a clinical skills evaluation, “the majority (67%) were found to have educational needs requiring moderate to considerable reeducation or updating, and another 6.5% showed educational needs that were broad enough to recommend education in a residency program to prepare for a return to practice.”38

Both the American Medical Association and the Federation of State Medical Boards have identified this policy issue as a priority in recent years, and medical licensing boards have individually promulgated regulations that may include successfully passing an examination of general medical knowledge such as the SPEX39 or the COMVEXUSA40, additional supervised training, a clinical evaluation or other conditions prior to licensure. Given one estimate of 10,000 physicians per year re-entering active practice41, it is imperative that a learning health care system not only provide an accurate assessment of a practitioner's knowledge base prior to re-entry, but that it also provide avenues for obtaining the necessary skills and information to return to practice.

The work of the IOM's Consensus Committee on The Learning Healthcare System in America comes at an important time for health care in the United States. We are in a period of transition that will require new thinking about the ways we go about educating our health care workforce — and medical regulators should be vocal leaders in the national conversation that lies ahead.