Commentary: Flexner Redux 2010: Graduate Medical Education in the United States

Abstract

The author posits that if Abraham Flexner could be brought back to write a report to address the same question that his 1910 report answered—How are doctors being prepared for practice?—he would not focus on the state of medical students' education, because today's medical students no longer enter practice on graduation from medical school. Instead, he would ask how the country's graduate medical education (GME) system is preparing residents for clinical practice. Such a report is needed because (as concluded by a variety of sources, including recent reports by the Institute of Medicine) GME training must be changed as part of the effort to improve medical care quality. Despite this finding, professional organizations that regulate how GME is designed and conducted have been unwilling to make substantive changes in the ways residents are being prepared for practice. The author reviews the reasons for this, emphasizes the need for a new Flexner-type report to make the public aware of the training deficiencies so the public could exert pressure on government officials to take action, describes groups that have already recognized the problem and proposed solutions (such as having individual programs held accountable for their residents' performance in practice), and makes recommendations, such as (1) having the Medicare program, the single largest payer of GME costs, require certain needed reforms in residents' training, and (2) modernizing state laws so that residents completing training would have to demonstrate that they are adequately prepared to enter practice before being granted licenses.

In 1910, the Carnegie Foundation for the Advancement of Teaching issued what is popularly known as the Flexner Report, clearly one of the most influential reports ever published by the foundation.¹ Within a few short years the report had had the effect of bringing to a close a long-standing debate within the profession about how medical schools should prepare doctors for entry into clinical practice. And the report continues to have an impact even today on the nature of the undergraduate medical education programs conducted by the country's medical schools.

Because next year will mark the 100th anniversary of the publication of the report, it seems likely that medical schools and professional organizations will sponsor lectures or hold conferences devoted to how the report has affected medical education in this country. And various journals, including Academic Medicine, will almost certainly publish papers focusing on various ways the report has affected the education of today's medical students. I suspect that many in the academic community who will participate in these activities will view them as an opportunity to provide a contemporary update to Flexner's report—that is, to propose topics related to medical students' education that Flexner might choose to address if he were to rewrite today the report he authored a hundred years ago.

But if Flexner could be brought back to rewrite his report today, I do not think he would focus on the state of medical students' education, assuming that his mandate was consistent with the one the Carnegie Foundation gave him long ago. In that case, he would spend no time examining how medical students are being educated; he would concern himself instead with how the country's graduate medical education (GME) system is preparing resident physicians for entry into clinical practice.

To appreciate why this would be the case, it is important to understand the circumstances that led Henry Pritchett, the president of the Carnegie Foundation, to commission Flexner to write a report.

In 1906, the American Medical Association's (AMA's) recently formed Council on Medical Education (CME) conducted a survey of the country's medical schools. The results of the survey documented what the council's leadership already knew: There were extraordinary deficiencies in the way a number of schools, primarily those that were proprietary in nature, were educating students. Having been frustrated for many years by their inability to gain agreement within the profession on how to correct the deficiencies documented by the survey, the council's leadership decided that the changes that were needed would not occur until the general public became aware of the inadequate ways that some medical schools were preparing their students for entry into practice. Given that, the council's leadership approached Henry Pritchett to see if he would be willing for the foundation to play a role in educating the public about the shortcomings that existed.^2,3

Pritchett embraced fully the strategy proposed by the council's leadership. He believed strongly that because members of the public were put at risk by encountering poorly trained practitioners, the public needed to gain an understanding of the inadequate education being provided by some medical schools. Pritchett made his views on this quite clear in his introduction to the Flexner Report:

The interests of the general public have been so generally lost sight of in this matter that the public has in large measure forgot that it has any interests to protect. And yet in no other way does education more closely touch the individual than in the quality of medical training that the institutions of the country provide. Not only the personal well-being of each citizen, but national, state, and municipal sanitation rests on the quality of the training which the medical graduate has received. The interest of the public is to have well-trained practitioners in sufficient number for the needs of society.^{1(p xv)}

Pritchett was convinced, as was the CME leadership, that if members of the public understood the situation, they would then exert pressure on the profession and on public officials to bring about the changes needed to improve the education of doctors. He also agreed with the CME leadership that a report from an independent, nonprofessional organization would be more effective in achieving that objective than a report issued by the AMA. Accordingly, he agreed to have the foundation take the lead in preparing such a report.

Pritchett recruited Abraham Flexner, a nonphysician, to prepare a report on the state of medical students' education based on findings Flexner would make during site visits to each of the schools in existence at the time. There was nothing in the resulting report that was not already known by the CME leadership and by Pritchett. Further, none of Flexner's recommendations in the report about how medical schools should be organized to provide a quality education were original, because Flexner used the Johns Hopkins University School of Medicine as his model, and a number of the schools in existence at the time had for years been using similar approaches for educating their students.

The Flexner Report focused on medical schools and how they were educating their students for one simple reason: At the time the report was commissioned, the vast majority of medical students entered practice directly on graduation from medical school. Thus, to educate the public about how doctors were being prepared for entry into clinical practice, the report had to be focused on medical schools. But the situation is entirely different today: Medical schools are not responsible for preparing doctors for practice and have not been for decades. Thus, if a foundation decided to use the 100th anniversary of the publication of the Flexner Report to commission a study based on the principle used in commissioning the original report—that is, how doctors are being prepared for practice—that foundation would not be concerned with how the nation's medical schools were educating today's medical students but, instead, would focus on the country's GME system. But would the foundation's leaders decide that such a report is needed after they had time to look into the current situation?

Absolutely! Once they learned about how GME is preparing doctors for practice today, they would realize that a report is needed because certain dynamics that limited the adoption of needed reforms by the medical education community in the early 1900s still exist. For example, during the past decade there have been growing concerns within the community that GME programs, at least in some specialties, are not designed and conducted in ways that would best prepare resident physicians to provide high-quality care to the kinds of patients they will encounter on entering practice.^4–6 Indeed, reports issued by the Institute of Medicine^7,8 in 2001 and in 2004 concluded that training approaches need to be changed as part of the overall effort for improving medical care quality. And studies have documented that resident physicians completing their training, as well as physicians who have recently entered practice, believe that they have not been adequately prepared to care for some of the common conditions encountered in the clinical practice of their specialties.^9–13 Despite these findings, the various professional organizations involved in regulating how GME is designed and conducted have been unwilling to make substantive changes in the ways residents are being prepared for practice.^14–16

The reasons for the lack of response by the profession can be traced to the tremendous fragmentation that exists in the regulatory apparatus that governs how residency programs are designed and conducted.¹⁷ In each specialty, there are several different organizations—specialty boards, specialty societies, academic societies—that exert influence over the nature of residency training in their specialty. Despite the fact that clinical practice in each specialty has changed over the years, these organizations have been unwilling to put aside their adherence to traditional views on how individuals entering their specialty should be trained. But equally important, there are reasons to believe that those in the academic community who exert a great deal of influence over how the programs are designed and conducted are unwilling to make changes that will result in the loss of resident coverage on certain clinical services, primarily inpatient services. And finally, the body that accredits GME programs (the Accreditation Council for Graduate Medical Education) is unwilling to overrule the recommendations of its residency review committees (RRCs), the bodies that develop accreditation standards, because doing so would place it in conflict with the specialty organizations that exert a great deal influence over the accreditation standards developed by the RRCs.

The aim of a Flexner-like report on GME would be to make members of the public aware of the deficiencies that exist in how residents are being trained for clinical practice, so that they would exert pressure on government officials to take actions on the public's behalf. That approach was successful in the early years of the 20th century by leading state governments to adopt licensure laws that required physicians entering practice to have completed a defined set of educational experiences before entering practice. The challenge facing those interested in improving GME is to identify the government levers that might be used to ensure that the medical education community makes the changes needed in how resident physicians are being prepared for practice.

In that regard, it is worth noting that government officials have already recognized the problem.¹⁸ In 2002, the Agency for Healthcare Research and Quality (AHRQ) and the Health Resources and Services Administration (HRSA) convened a meeting to initiate a dialogue among medical education researchers, outcomes researchers, and stakeholder organizations about how long-term medical education outcomes should inform the design and conduct of GME programs. The premise for the meeting was that GME programs should be held accountable for the performance of their graduates in practice—that is, for the quality of care their graduates provide.^19,20 The agency officials reasoned that such an approach would likely result in programs taking steps to ensure that their residents had the kind of training experiences that would prepare them to provide high-quality care on entering practice. It seems highly likely that if individual programs were held accountable for the performance of their residents, they would become more actively involved in forcing their respective specialty organizations to take actions needed to change accreditation standards. In 2006, the Association of American Medical Colleges Institute for Improving Medical Education partnered with AHRQ in an effort to develop a research agenda aimed at achieving objectives that emerged from the AHRQ/HRSA meeting. A select group of academic medical centers were invited to send teams to a working conference to determine how such an agenda might be put in place. Because there was no follow-up to that meeting, it would seem that other approaches are needed.

In fact, there are several approaches that government might employ to achieve the goal of ensuring that residents completing training are adequately prepared to provide high-quality care. Because the federal government funds the majority of the costs incurred in training residents in GME programs, it could exert great leverage over how the programs are designed and conducted. For example, the Medicare program, which is the single largest payer of GME costs, could require that those who regulate training experiences in individual specialties must demonstrate that the design and conduct of the programs are based on an analysis of the kinds of patients that those in training will encounter on entering practice. It would not be difficult to acquire the data needed to meet that requirement. Armed with the appropriate data, who could argue responsibly that the changes needed in the design and conduct of the programs were too difficult to implement?

Another approach, although one that would be far more difficult to implement, is to modernize state licensure laws so that residents completing training in a specialty would have to demonstrate that they are adequately prepared to enter practice before being granted a license. This, of course, would mimic to some degree the approach used after the release of the Flexner Report in 1910. In fact, the National Board of Medical Examiners has been in favor since the early 1970s of a licensure requirement for those nearing completion of their residency training.²¹ Recently, a committee charged to conduct a comprehensive review of the United States Medical Licensing Examination (USMLE) recommended such an approach, and that recommendation has been approved by the USMLE Composite Committee. Thus, if individual licensing authorities prove unwilling to adopt new licensure requirements, the federal government could simply make meeting the requirements a prerequisite for physicians to participate in the Medicare program. Such an approach would clearly incentivize the licensing authorities and the state legislatures that generate the laws governing licensure to make the required changes.

To be clear, I am all in favor of taking steps to improve the education of today's medical students. But the major challenge facing the medical education community today—a challenge that needs the active support of those holding leadership positions in academic medicine—is to reform how resident physicians are being prepared for entry into clinical practice. The purpose of doing so is very clear: to ensure that physicians completing GME programs have been prepared to provide high-quality care to the kinds of patients they will encounter when they enter practice. Given all of the attention now being paid to various proposals for reforming the country's health care system to, among other things, provide high-quality care to all citizens, it seems quite appropriate for policy makers to focus some attention on the need for reform of the country's GME system. Just as in Pritchett's and Flexner's day, the public deserves no less.