Everyone loves the Yale System.
So why can’t they all agree?

The debate over exams this spring centered on two key questions: how should medicine be taught in the 21st century, and how should a student’s progress be measured?

By John Curtis
Photographs by Terry Dagradi

When Dean David A. Kessler, M.D., first came to Yale in 1997, he made an observation that seemed remarkable. The medical school had a deputy dean for research responsible for $211 million in grants and contracts, a deputy dean for clinical affairs overseeing 650 academic physicians, but no deputy dean for education. Whose job was it to think about the teaching of the 485 medical students who were beginning their careers as doctors and scientists at Yale?

This administrative gap was emblematic of a disparity in American medical education that began with large-scale investment in research following World War II, continued with the passage of Medicare in the 1960s and the resulting clinical expansion, and became pronounced with the advent of managed care in the 1990s: teaching at medical schools often took a back seat to research and patient care, with time and resources frequently cobbled from the other two missions. For a faculty member, it was fine to be a gifted educator but it was prowess in the lab and clinic that earned promotions and tenure at Yale and elsewhere.

The school’s world-class professors and top students aside, Kessler and others on the faculty felt that teaching could be improved and that it deserved more attention. The dean wrote to alumni in March 1999, announcing the formation of an ad hoc committee on medical education that would “examine not only what we are teaching now, but also what we should be teaching to prepare physicians for the 21st century.” Kessler went on to note that other medical schools had undertaken similar programs, raising important questions about the best ways to educate doctors. “Few schools, however, have tackled the larger questions related to the academic content of the curriculum,” he added. “We intend to do just that, and to reaffirm and fortify the Yale System in the process.”

Within a year, the dean had a set of recommendations from the committee’s faculty and student members, and by the year 2000 most of those suggestions had been, or were in the process of being, implemented. A deputy dean for education was appointed. Basic science courses were consolidated and streamlined where appropriate. Increasingly, courses combined material from the basic and clinical sciences. And, in an effort to reinforce one of the elements of the core educational philosophy that has been in place at Yale since the 1920s, the new deputy dean had tried to build more free time into the first two years, a period traditionally reserved in the curriculum for basic science instruction and one which had become increasingly cluttered over the decades as new knowledge was added to the canon and old assumptions became obsolete.

While these efforts pleased many, an unintended consequence of the changes that ensued provoked an uproar among students and alumni. The debate echoed a controversy that surfaced in the mid-1980s, when for the first time in the history of the Yale System of medical education, students had been required to take final exams in the basic science courses.

Exams were again the issue. Since they were introduced in the 1980s, exams have been anonymous and, except for qualifiers, optional. And they have in recent years been available online, so students can take them at home and on their own schedule. The revamped curriculum of 2001-2002 concentrated many of the basic science courses in the first year, rather than the first two years as they had been. That meant more exams in the first year, and a perception among students that they faced one test after another. At the same time, second-year students grew concerned about a faculty decision, since rescinded, to require attendance and mandatory self-assessment exams in the second-year “modules”—interdisciplinary courses that integrate key concepts in the clinical and basic sciences. Some students saw these changes as a threat to the Yale System, which follows a graduate school model for medical education and considers students mature and independent scholars capable of learning on their own.

Under the banner of the “Yale System Preservation Initiative,” nine students wrote to nearly 5,000 medical school alumni, asking their endorsement of a petition that all self-assessment exams remain optional. More than 500 alumni wrote back (See “The Yale System Lives! Long Live the Yale System”), most in favor of the student position. And exams were a major topic of discussion when the Association of Yale Alumni in Medicine (AYAM) executive committee gathered in June. The mood around the table was sympathetic to the petition and skeptical of too many changes. “Are our people being rejected [from top residency programs] because they weren’t assessed properly?” asked Arthur C. Crovatto, M.D. ’54, HS ’61. The question was a rhetorical one, given the school’s record of consistently impressive Match Day placements.

As the debate over exams has unfolded at Yale, medical educators around the country continue to rethink the model that has ruled since the time of Abraham Flexner almost a century ago—two years of basic science followed by two years of clinical instruction. In the mid-1980s Harvard introduced its New Pathways curriculum, which interwove the clinical and basic sciences, combined basic science courses and defined a core of knowledge to be mastered to avoid information overload. In 1997 the University of Pennsylvania announced Curriculum 2000, which integrated the basic sciences and clinical medicine and encouraged self-directed and lifelong learning. And at medical schools across the country and beyond, innovative teachers have sought a way to help medical education keep pace with a revolution in medicine.

In class from 9 to 5
Although he’s a graduate of Brown and has spent most of his teaching career at Columbia University’s College of Physicians and Surgeons, Herbert S. Chase Jr., M.D., considers the Yale System, with its emphasis on critical thinking as well as core knowledge, essential to his vision of medical education. Nevertheless, when he arrived at Yale in 2000 as the newly appointed deputy dean for education, he was surprised. “I didn’t realize how far the actual practice of the Yale System deviated from the philosophy,” he told alumni at the June meeting of the AYAM executive committee. “The first thing I found was that there was no free time. The students were in class from nine to five almost every day.”

The biggest threats to the Yale System, Chase said, came from an ever-expanding curriculum trying to keep up with advances in medical knowledge. Chase saw his first task as freeing time in the preclinical years that could be devoted to thesis preparation, the pursuit of individual interests or simply the unstructured exploration of medicine and science.

Since his arrival, classroom time has been reduced by 25 percent, the ratio of small-group sessions to lectures has increased and exams have been placed on the Web so students can take them on their own schedule. The old model—learning how the body works in the first year and studying disease in the second—has changed. “Now, in April of the first year you start learning about abnormal human physiology,” Chase said. “I think it has been a spectacular success. Students end the year not only knowing how the entire body works, but they also have a pretty good foundation in the mechanisms of disease, pathology, immunology and genetics.”

Chase has encouraged the use of concise study guides, identified educational resources on the Internet and consolidated redundant course material. Cell biology, physiology and biochemistry—formerly three distinct courses—are taught together as Molecules to Cells to Tissues to Systems, known as MCTS. “Now students have two free afternoons every week to do what their hearts desire,” he said. Despite this, there are still problems to be ironed out. “Even though class time is less, the content is the same,” said Christoph Lee, now in his second year. “We see lecturers, more often than not, running over because they are trying to cover the same amount of material in a shorter period of time.” Most students spend their free time studying for exams, added classmate Brenda Ritson, and at least a few are wondering whether the Yale medical school they applied to is the same one they’re attending. Nicholas Countryman, a third-year student whose grandfather graduated with the Class of 1944, had heard about the Yale System’s merits for years before arriving as a student in 2000. If the Yale System is allowed to erode, he asked, “What is going to be unique about Yale next year or 60 years in the future?”

An age-old question
The discussion on how best to teach—and how to assess learning—is not a new one. In its earliest days, the Yale System under Dean Milton C. Winternitz, M.D., rejected the traditional yardsticks of student achievement. Examinations and grades would undermine the very educational atmosphere the Yale System was meant to create—a collaborative, almost intimate world in which leading clinicians and scientists engaged and inspired a select group of bright, motivated students. But the system depended on a social contract. If students were to have the independence to design their own medical studies in the preclinical years, they would also be expected to rise to an unprecedented level of responsibility. Faculty, in turn, would have to spend the time necessary to follow and evaluate the students’ progress. “The Yale System is predicated on teachers wanting to teach and students wanting to study—and being mature enough to seek help,” said former Dean Gerard N. Burrow, M.D. ’58, HS ’66, whose history of the medical school was published in October by Yale University Press.

In the early days, the only required tests were qualifying exams administered at the end of the second and fourth years; the only debate was whether these qualifiers would be developed in-house or replaced by the boards, officially known as the United States Medical Licensing Examination, or USMLE. According to the minutes of the school’s curriculum committee, the faculty tried both methods before settling on the boards in 1942.

The boards appear to have been the only required exams until the mid-1980s, when in a single year, 17 students failed Step 1 of the USMLE. “That provoked a great deal of concern among the basic science faculty,” former Dean Leon E. Rosenberg, M.D., HS ’63, said in a recent interview. As a result, Rosenberg said, he and Robert H. Gifford, M.D., HS ’67, who was then the associate dean for student affairs, decided to implement qualifying exams in the basic science courses. The basic science faculty, he said, were as solidly behind this decision as alumni and students were opposed to it.

“There was quite a lot of unhappiness,” Rosenberg said when contacted at his office in the Department of Molecular Biology at Princeton. “The alumni felt that this was an attempt to demolish the Yale System, which, of course, it was not. The students felt that they were being punished for the performance of their predecessors. They also were concerned that because Bob Gifford and I were not products of the Yale System, we did not find the matter of the Yale System as hallowed as they did, which was not true.”

To preserve the independent spirit of the Yale System, the new exams would be anonymous. They would not be graded. Students would only come to the attention of faculty if they failed more than two of the so-called minimal competency exams.

“The rule was that if you failed, it was your obligation to make yourself known to the director of the course and find a way, in collaboration with the director, to pass,” said Nancy R. Angoff, M.P.H. ’81, M.D. ’90, HS ’93, associate dean for student affairs, and a student in one of the first classes subject to the new requirements. “It could be by taking the exam again or taking an oral exam or writing a paper or analyzing articles. You had to find a way to show you were competent in that area.”

A perceived change
Since then students have been required to take qualifying exams in basic science courses and, as before, have had the option of taking periodic self-assessments to gauge their own progress in those courses. (Mandatory evaluations have always been part of clinical instruction in the third and fourth years.)

Under the 2001-2002 curriculum, however, students found required exams demanding more of their attention. Although the number of basic science exams had dropped from 18 to 13 (and fell to 10 this academic year), the interdisciplinary nature of the new courses meant that the exams mixed questions from various fields. A single test might require a review of topics in physiology, biochemistry and cell biology. And, with more classes concentrated in the first year, the scheduling of exams left students with the impression that there was always another test for which to prepare.

Students were also concerned about exams in the second-year interdisciplinary modules, which were in conflict with the national boards. According to Margaret J. Bia, M.D., professor of medicine, who directed the second-year clinical modules for four years and is now director of clinical training, “board fever” typically has struck early in the second semester as students abandoned the classroom to prepare for Step 1 of the USMLE. By semester’s end, attendance in the modules was down to about a third of the class. Bia considers the modules the most important courses in the first two years of medical school; they integrate the clinical and basic sciences, offer a case-based overview of organ systems and are taught, at least half the time, in interactive workshops with practicing clinicians as instructors. “It’s the time when students are encouraged to think about disease in a pattern they will use over and over again in their medical lives. These are the courses in which the pathophysiology of important diseases is explored and discussed,” she said. Faculty members also were putting in hours of preparation for students who never benefited because they didn’t come to class.

Bia said a crisis was mounting because “with so many students not attending lectures or workshops, the faculty had no way of knowing whether they were learning this important material. So we created a series of self-assessment exams. These exams were also a learning tool, as students were given annotated explanations to all the questions after they submitted their answers.” About a quarter of the class either refused to take the self-assessment exams for the modules in the winter of 2002 or scored in the 20s out of 100, she said.

Bia made the ungraded exams mandatory, which she now regrets. “I made an absolutely strategic error in making these changes without including a representative group of students to advise us,” she said. “It would have been better for them, for the faculty and for the curriculum had we had their input on these changes from the beginning. That being said, I hope this doesn’t preclude module self-assessment exams in the future, as they’re a great learning tool for the students and provide an opportunity for the faculty to see if students are learning the material.”

Her colleague Frank J. Bia, M.D., FW ’79, felt that a clash was inevitable as the curriculum began to interweave the clinical and basic sciences. “In the 25 years I have been here, this represents a major shift, putting real emphasis on clinical medicine during the first and second years,” said Bia, professor of medicine and laboratory medicine. “Once you start doing that, however, you must deal with perceptions of the Yale System. The modules are the point where the clinical and basic sciences meet. Now you’re learning information that is directly applicable to the wards. Self-assessment becomes critical. How can you argue that doing a good history and physical exam, being observed doing it and being critiqued are a violation of the Yale System because they’re mandatory?

“There is this misguided notion that you can translate the Yale System into clinical medicine when it comes to the clinical skills that are involved. You cannot learn clinical skills in isolation. Faculty and students have to be held responsible and accountable for both teaching and learning these skills.”

Assessing assessment
Around the country medical schools are looking at ways of assessing students, including peer review, the use of “standardized patients” in mock clinical situations, direct observation and written exams. The Liaison Committee on Medical Education (LCME), the accrediting body for medical schools, requires “formative and summative evaluation of student achievement in each course and clerkship” but discourages tests that condition students to memorize facts for the short term. Evaluations should measure students’ knowledge and the development of the skills, behaviors and attitudes essential to the practice of medicine.

“The emphasis here is on providing the means for students to measure their own progress in learning,” said David P. Stevens, M.D., vice president for standards and assessment at the Association of American Medical Colleges and secretary of the LCME. “There are many ways to do this that allow for anonymity but do not necessarily call for an official, identified letter or numerical grade.”

A number of things have happened since the debate began in March.

First, Chase rescinded the requirement that module exams be mandatory. He has also taken steps to remove the conflict between the modules and the boards. This academic year, modules began in September instead of October and will end earlier, creating a seven-week break so students can study for the boards.

The challenge of assessing students without resorting to exams remains, however. “We still need a means by which to evaluate our students,” said Stuart D. Flynn, M.D., professor of pathology and surgery and the new director of the second-year modules. “How do you evaluate individual students in the preclinical years without the mindset of giving examinations? I think there are ways to do it, and this represents a wonderful challenge for the school. It would require faculty or some kind of small group to assess individual students periodically, with the goal being to assure a certain level of competency to allow advancement to the next level of training. That results in a lot of one-on-one time between students and faculty.” Finding a solution, he added, will require serious discussions among faculty, administrators and students.

The main vehicle for the ongoing conversation is a rejuvenated committee on educational policy and curriculum, which dates to 1989. In its original format, said Emile L. Boulpaep, M.D., professor of cellular and molecular physiology, it was made up of course directors and had only two subcommittees, for the basic and clinical sciences. “Now we have a third area, curriculum design—how the teaching is being delivered and all the things that have to do with evaluation of students and the educational process,” he said. Those three subcommittees oversee a dozen “education working groups,” in which students elected this spring serve alongside faculty.

In May, Kessler, Chase and the students leading the Yale System Preservation Initiative wrote to alumni to bring them up-to-date on the recent events. Both Kessler and Chase strongly reaffirmed their support of the Yale System and their commitment to preserving it. As Chase subsequently told alumni leaders at the June AYAM meeting, “The philosophy is safe. We all believe in it. That’s why we’re all here.” In their letter, students welcomed the administration’s decision to delay the evaluation format pending further discussions and to include students in decision-making committees.

“It’s nice to know that what we think matters,” said Michele Flagge, a second-year student who was the first to notice the curricular changes in 2001-2002. “It was never our intention to be rabble-rousers who wanted to change the establishment. Our main purpose was to heighten awareness of what the traditions of the Yale System were. We opened up the dialogue, which was great.”

Kessler, who came to Yale five years ago with the intention of bolstering the educational mission, agrees. “The debate about the Yale System is important for the institution,” he said. “I think it’s healthy for the institution. I think we have all learned from it.” YM

John Curtis is the associate editor of Yale Medicine. Terry Dagradi is a photographer with Med Media Group at the School of Medicine.