Last spring, in his elegant commencement address to the Harvard Medical School, Dr. Atul Gawande appealed for a dramatic change in the organization and delivery of medical care. His reason, “medicine’s complexity has exceeded our individual capabilities as doctors.” He accepts the necessity of specialization, but he criticizes a system of care that emphasizes the independence of each specialist. Dr. Gawande is not alone in thinking that scientific, technologic, and economic changes require reorganization of care. Larry Casalino and Steve Shortell have proposed Accountable Care Organizations (ACOs); Fisher, Skinner, Wennberg and colleagues at the Dartmouth Medical School have focused on reforming Medicare, and many others have also called for major changes.
I expressed similar concerns in 1974 in my book Who Shall Live?, but at that time I rejected the claim that the problems of medical care had reached crisis proportion. In 2011, however, I agree with those who say the need for comprehensive reform must be marked URGENT. The high and rapidly rising cost of health care threaten the financial credibility of the federal and state governments. The former finances much of its share of health care by borrowing from abroad; the states fund health care by cutting support of education, maintenance of infrastructure, and other essential functions. These are stop-gap measures; neither borrowing from abroad nor cutting essential functions are long-run solutions. The private sector is equally distressed. Surging health insurance premiums have captured most of the productivity gains of the past thirty years, leaving most workers with stagnant wages. Not only is there a pressing need for changes in organization and delivery, but Ezekiel Emanuel and I, in our proposal for universal vouchers funded by a dedicated value-added tax, argue that such changes must be accompanied by comprehensive reform of the financing of medical care (Brookings paper).
But that’s not what I want to talk to you about today. My subject is the urgent need to change the structure of medical education. It seems to me that such change is necessary, and perhaps inevitable, given the revolution in medicine over the past half century, and given the changes in organization and delivery of care that lie on the horizon.
The need for change
Consider the deluge of new medical technologies in recent decades. According to Dr. Gawande, in deciding on interventions for their patients, clinicians now must choose from 6,000 drugs and 4,000 procedures. To be sure, many of the 6,000 are not new chemical entities but rather combination drugs, alternative dosage forms, and other variations. Still, the burden on the clinician to make an appropriate choice is great, especially if, as stated in the Physician Charter, “physicians are required to provide health care that is based on the wise and cost-effective management of limited clinical resources.” Economists have been touting cost-effectiveness for years, but it is a harbinger of change to see organizations representing more than half of all active physicians sign a charter committing them to practice cost-effective medicine.
Along with the new technologies, there has been a proliferation of specialties and sub-specialties. Fifty years ago, there were 18 specialty boards and very few sub-specialties. Now there are 36 specialty boards and 116 sub-specialty certifications, for a total of 152. Does such proliferation provide much or any benefit to patients? The United Kingdom has only 97, while, Canada and France have fewer than half as many. Proliferation of specialties and sub-specialties almost certainly adds to the cost of medical education and the cost of care, while its effect on quality of care has not been systematically investigated. The former chair of medicine at a major academic medical center thinks it has an adverse effect on patient care, but other experts disagree. We just don’t know the answer. If empirical studies conclude that so many sub-specialities are desirable, the training structure that produces them should and could be made more efficient. Medicine is one of the few fields that requires specialists to have more training than generalists. This may have been rational at one time, but may not be today.
Finally, and closely related to the new technologies and increased specialization, there is the soaring cost of medical care. In 1960 U.S. health expenditures, in 2009 dollars, were $864 per person. In 2009, they were $8085. Along with the cost of medical care, the cost of medical education has increased exponentially.
In the face of such revolutionary changes, how has the structure of medical education adapted? It seems that the answer is hardly at all. Fifty years ago, the basic structure was four years of college, four years of medical school, and three years of post-graduate training. Only after 11 years of post-high school graduation was the physician deemed ready to practice medicine. The same is true today, although a much larger percentage than formerly go beyond 11 years to obtain additional specialized training. And in one medical school I know of fewer than 40 percent graduate in 4 years.
The goals of change
A reasonable goal for structural reform might be to reduce that basic period from eleven to nine years. This can be done by cutting time off the front end or the back end of the process or both. About the front end, I note that there are now 33 medical schools that combine college and medical training in six years. Could there be more such schools? What is known about the quality of care delivered by physicians from these programs compared with the graduates of conventional medical schools? Very little. Most other developed countries combine college and medical school in one program that is typically less than eight years long. Are their physicians inferior to American physicians?
It might be argued that foreign medical schools can admit students directly from high school because the educational achievement of those high school graduates is greater than that of American high school graduates. This is probably true on average, but there are certainly some American high school graduates with educational achievement equal to those who graduate from foreign high schools. Why couldn’t American medical schools consider for admission applicants who, through appropriate examinations and interviews, appear to be as well qualified as the college graduates the schools are now admitting, regardless of how many post-high school years the student has completed? I understand that thoughtful leaders in medicine are studying various possibilities for accelerating admission to medical school for qualified candidates. That’s great. But I hope they realize that the health care system is entering the “ICU;” prompt, decisive action is needed.
In order to reduce time at the back end, schools might consider accelerating choice of specialization. Dr. Gawande notes that there was a time when “doctors could hold all the key information patients needed in their heads and manage everything required themselves.” He says that in such a world it made sense for physicians to prize “autonomy, independence, and self-sufficiency.” But that time is gone forever. What remains is a structure of medical education based on those outmoded assumptions. For Dr. Gawande, who is as handy with a metaphor as with a scalpel, the bottom line is “we train, hire, and pay doctors to be cowboys. But it’s pit crews people need.”
A proposed new structure
If Dr. Gawande is correct, what does this imply for the structure of medical education? Isn’t it time to give up the conventional wisdom that pouring more and more knowledge into each physician about more and more subjects will produce a better system of medical care? Far from rejecting specialization, embrace it sooner. For the purpose of stimulating discussion, I propose the following structure for medical education:
- Two years of medical education taken by all students. This common curriculum would consist of 50 percent basic science with an emphasis on competencies that would be useful to every physician. Subsequent exposure to basic science would depend on its relevance to the student’s prospective career.
- One-third of the time would be devoted to an introduction to clinical care of individual patients, making as full use as possible of modern technologies that have been successful in training programs in industry, the armed forces, and other settings.
- One-sixth of the time would be used to cover key aspects of the health of populations and the organization and delivery of care, with emphasis on a team approach to enhance health. It is important for all physicians, regardless of prospective careers, to understand how each element fits into a health care system.
Upon completion of the two years, each student would select a track which launches him or her into the world of specialization. Here is an example of what the tracks might look like:
- Leaders of primary care teams, possibly sub-divided into adult care, pediatric care, and geriatric care.
- Clinical specialists in medicine, hospital based and ambulatory.
- Clinical specialists in surgery and other procedural specialties.
- Possibly another track for those headed for specialities such as radiology and pathology that treat medical and surgical patients.
- A track for students whose major interest is research, possibly similar to current MD-PhD programs but with explicit recognition that the trainees are not preparing to be clinicians.
The content of the training program would differ depending on the track. For example, students training to be leaders of primary care teams would be exposed to more statistics, epidemiology, preventive medicine, and management skills than those in the other tracks. They would learn how to deploy nurse practitioners, physician assistants, and other non-physicians most effectively.
Is it feasible for students to make specialty decisions sooner than they do in the present structure? Before you answer with a resounding “no”, let me tell you a “tale of two schools.”
A tale of two schools
Just a stone’s throw from the Stanford School of Medicine (if you have a good arm) is the Stanford School of Engineering. The latter school accepts students after they have completed two years as undifferentiated Stanford undergraduates. Prospective students of engineering are encouraged to take a wide variety of courses during their first two years at Stanford, but are also advised to make sure they are getting a good start toward engineering through courses in mathematics and science. At the beginning of their junior year the engineering students declare which of 17 fields they plan to specialize in. The fields range (alphabetically) from Aeronautics and Astronautics to Product Design and include such well-known specialties as Chemical, Civil, Electrical, and Mechanical Engineering.
Notice that the choice of specialization is made two years after high school graduation. I may have said that too rapidly. Let me repeat it. Two years after high school, engineering students at Stanford commit themselves to one of 17 specialties. At MIT students must choose their specialty at the end of their freshman year. The heavens do not fall. The SAT scores of the engineering students suggest that they are intellectually about equal to the Stanford medical students. The School of Engineering helps students learn about the various specialties by offering 20 seminars on different subjects with enrollment preference given to freshmen. Examples of seminar subjects are: “Bioengineering Materials to Heal the Body,” “Digital Dilemmas,” “Water, Public Health and Engineering,” and “What is Nanotechnology?”. An additional 12 seminars are offered on other subjects with enrollment preference given to sophomores. Examples of their titles are: “Electric Automobiles and Aircraft,” “Environmental Regulation and Policy,” “Medical Device Innovation,” and “The Flaw of Averages.” These seminars provide an opportunity to work closely with faculty. In addition there are many one-unit seminars that provide exposure to key issues and current research in various fields. At the end of four years at Stanford, approximately 80 percent of the engineering students graduate with a bachelor’s degree and enter the workforce to practice their specialty. Students who go on for a fifth year typically do so in order to earn a master’s degree.
There are of course, many differences between engineering and medicine. Biologic systems are probably more complex than the systems engineers work with, and causal relations are less firmly established. An alleged difference is that physician decisions affect life and death, but the same could be said for many engineers. The men and women responsible for our bridges and tunnels, the design of our airplanes and cars, the safety of our water supply, and many similar functions are surely making decisions that affect life and death. One of the biggest differences is that engineers specialize from the start of their training; they are not expected to know about all aspects of engineering. They typically work in team settings. They are, to use Dr. Gawande’s words, “pit crews” not “cowboys”. Collectively, they get the job done. Perhaps the biggest difference is that when a medical student chooses a specialty, he or she is usually choosing a life-time occupation. For an engineering student, life-time occupation is not as closely linked to choice of specialty training. One reason for persistence by physicians in a certified specialty is that diminished competition affords the specialist the opportunity to earn a “monopoly rent.”
As an example of how specialty training in medicine does not have to take as many years as tradition demands, consider Dr. Robert Chase’s experience in training plastic surgeons at Stanford. When he began his program, plastic surgery required completion of residency in general surgery followed by another residency program in plastic surgery. The combination took a minimum of seven years and more often eight or nine. Drawing on his experience as chief resident in general surgery at Yale, a two-year fellowship in plastic surgery at the University of Pittsburgh, and active duty in the Valley Forge Army hospital, Dr. Chase was pretty sure he could train plastic surgeons in no more than 6 years and often in four or five. To this end he developed an integrated program that started residents headed for plastic surgery side by side with residents headed for general surgery.
The idea was rejected by the American Board of Plastic Surgery, but he pursued it anyway. Fortunately, the first residents to complete the program did so well at both the written and oral examinations that the Board gave tentative approval to the program. Today there are 27 truly integrated programs similar to Stanford’s, and another 62 that combine general and plastic surgery; only 27 of the traditional programs remain. It would be surprising if similar shortening could not be accomplished in other fields of medicine and surgery. What is required is an exceptional clinician-teacher who is willing to confront the established powers and prospective specialists who are willing to commit sooner to their specialty.
Arguments against and obstacles to restructuring
Until now, medical education has proceeded under the premise that “Keeping one’s options open” is a free good. It is not; and the costs to the individual and society increase every year. Those who set the rules and requirements must consider the possibility that what their generation had to endure may not be the best path for the future. Many of the existing rules and requirements seem to be based only on “tradition”. The same academic physicians who would not prescribe a drug without determining efficacy and safety, have no hesitancy in prescribing the structure of medical education without any studies that examine the appropriateness of that structure relative to alternatives.
Changing the structure of medical education will not be easy, even for those who are enthusiastic about the goal. Opponents will be numerous, and the arguments varied. Many of the most popular ones are not persuasive. Consider the cliché, “If it ain’t broke, don’t fix it.” The existing structure may not be “broke”, but it provides the intellectual foundation for a medical care system that is causing the rest of the country to go broke. Some will say that my suggestions are “controversial.” I agree. For more than fifty years I have observed and participated in attempts to reform college curricula, and I can tell you that reforms that are not controversial are inconsequential. Some will want to take credit for the gain in life expectancy of 8.4 years over the past half century. But other developed countries with different systems of medical education and medical care have achieved even greater gains and are at a higher level, while their per capita spending on medical care is 35 to 50 percent less.
Two possible objections to changes discussed in this lecture are that they threaten the deeply held (albeit antithetical) visions of the physician as scientist and the physician as humanist. The threats are real, but the visions are increasingly unreal. American medical education is at a cross-road: Shall it continue to strive to produce scientists-humanists or recognize that what society needs most at this time are competent professionals, capable of providing leadership and supervision for the more than 15 million individuals now employed in the delivery of health services. The challenge to the leaders of medical education is to figure out what kind of admission policy and what structure and content of medical education, undergraduate and post-graduate, will produce such professionals at a reasonable cost It could be correctly argued that the cost of medical education is a relatively small part of the total cost of medical care, so why change medical education? The reason is that a restructured admission policy, earlier specialization, and shorter period of training can contribute to producing a different physician, one better suited for a team approach to remedying the cost, access, and quality problems now evident in American health care.
The obstacles to change will be partly external to the medical education establishment and partly internal. Consider, for example, the dense network of laws and regulations that now govern the practice of medicine. Some are federal, most are state, and often differ from one state to another. Those that are worth preserving should be federalized. These laws and regulations have been passed with the present structure of medical education in mind. Change in that structure will require changes in the existing legal framework. Many of the laws were enacted with the stated purpose of “patient protection”, but as is true in so many industries, they often wind up giving providers protection from competition.
Consider also how malpractice attorneys will leap on health outcomes that fall short of ideal and try to tie these lapses to changes in medical education. We badly need a better system of dispute resolution to replace malpractice suits. Consider also, how large insurance companies and hospitals will resist change, not necessarily because the change would harm them in the long run, but because change is usually disruptive and costly in the short run.
Perhaps the biggest obstacle to change will be within the medical education establishment which includes not only the medical schools but also post-graduate training programs and the bodies that control certification for 152 specialities and sub-specialties. Are all these necessary? Restructuring will undoubtedly require some faculty to change what they do and some faculty may be redundant. Many specialty and sub-specialty boards will need to change their criteria, as in the case of plastic surgeons. In some areas it may be difficult at first to find medical educators well-equipped to meet the needs of students in the new structure. For instance, where will medical schools find instructors to train the students who have opted for the track of leaders of primary care teams?
Finally, there is the chicken or the egg problem. There are medical leaders who see the need for significant changes in the financing, organization, and delivery of care. But they feel stymied by the absence of physicians with the preparation and attitudes necessary to be most effective in the new systems of care. There are leaders in medical education who see the need for significant changes in structure and content, but wonder where the graduates of the new programs will find appropriate employment.
All these obstacles suggest that restructuring may be impossible. But I draw some hope from an observation made by Alexis de Toqueville who said, “The United States moves from the impossible to the inevitable without ever stopping at the probable.”
This is the end of my jeremiad. If I have offended any readers, I apologize. That was not my intent. I have, for many decades, studied the American health care system, focusing on the high cost, the inequalities in access, and the lapses in quality of care. I concluded that these problems will not yield to piecemeal reforms. What is needed is comprehensive change in the financing, organization, and delivery of care.
But I have not paid much attention to medical education. Dr. Gawande’s Harvard commencement address made me realize that reform of the health care system must be accompanied by a restructuring of medical education. Hence this lecture. Perhaps my suggestions for restructuring are off the mark. Some may have better ideas as to how it should proceed. If so, all to the good. If I have convinced you of the urgency of the task and stimulated you to address the problem, my effort will not have been in vain. I greatly appreciate the opportunity you have afforded me, and I thank you for your patience.
I await your questions with interest and a reasonable amount of apprehension.
Victor Fuchs is the Henry J. Kaiser, Jr., Professor of Economics and of Health Research and Policy, emeritus, an FSI senior fellow and a core faculty member at Center for Health Policy/ Center for Primary Care and Outcomes Research at Stanford University. He has written extensively on the cost of medical care and on determinants of health, with an emphasis on the role of socioeconomic factors.