Abraham Flexner's name, more than any other, has been linked to the transformation of medical education in the first half of the 20th century. Why has the Flexner Report, formally titled Medical Education in the United States and Canada: A Report to the Carnegie Foundation for the Advancement of Teaching, Bulletin No. 41 been perceived to have such an influence on medical education? As has been described previously,2 many of the ideas for curriculum and other reforms contained in the report were not new. Nor was Flexner, at the time of the report, in a position of authority to bring change about. The Flexner Report, however, appeared when health care was improving significantly and when the problems inherent in the system of medical education had been recognized. In this context, it served as an articulate description of the changes needed in medical education and the mechanisms that could be used to bring change about. As such, it acted as a catalyst to stimulate and focus efforts at reform.
Health Care Between 1860 and 1910
In the 50 years prior to the Flexner Report, the ability to deliver health care had changed significantly, based on advances made in the United States and abroad. These included the discovery of the bacteria causing a number of common infectious diseases3; the incorporation of aseptic techniques in surgery4; the introduction of diagnostic laboratory procedures combined with the emergence of clinical pathology as a discipline4; and the introduction of instruments such as the ophthalmoscope, the laryngoscope, and the achromatic microscope.5 These advances led, in turn, to the transformation of the hospital from “an asylum for the indigent” to a “modern scientific institution.” 6(p17)New hospitals were opened and existing ones expanded, equipped, and staffed to perform an increasing array of surgeries and to care for a variety of other conditions.6
For physicians in the community, office visits were replacing house calls.4 It was recommended that general practitioners equip an office with the “new diagnostic tools,” such as the laryngoscope and ophthalmoscope, as well as the microscope.4 Emphasis was being placed on the accuracy of diagnosis and on the application of the new scientific knowledge.
However, the majority of physicians were not prepared for these advances.4 For many, the approach to patient care remained practical and empirical, based on memorization of signs and symptoms of diseases matched to treatments.5 This gap was a result of the system of medical education in which they had been trained. Their teachers in both the basic and clinical sciences often were physicians without experience in the modern methods of laboratory and clinical teaching.1
Medical Education Between 1900 and 1910
The decade before the publication of the Flexner report was an active period in medical education. Between 1900 and 1910, 70 MD-granting medical schools in the United States closed or merged and 50 new schools were founded.7 Medical schools continued to be of various types, including “regular” schools that taught the range of treatment options, and “sectarian” schools that espoused a specific philosophy for treatment, such as homeopathic, eclectic, and physiomedical. Table 1 presents the number of schools, by type as classified by the American Medical Association in 1900, 1905, and 1910. Across all types of MD-granting medical schools, total enrollment was 25,171 students in 1900, 26,147 in 1905, and 21,526 in 1910. In all these years, over 90% of students were enrolled in “regular” medical schools.7
Entrance requirements were highly variable across medical schools. Of the 133 schools in 1910, 99 (74%) required a four-year high school education. Flexner warned that many of the schools with a requirement for completion of high school were accepting “equivalent” credentials that fell far short of the desired standard.1 Only 27 schools (20% of the total) required two or more years of college. Of these, 11 had raised their entrance requirements in 1910, and before 1900, only the Johns Hopkins School of Medicine required two or more years of college.8 The remaining seven of the 133 schools in 1910 (5%) required one year of college, including physics, inorganic chemistry, biology, and modern languages.8 Perhaps not surprisingly, the medical schools requiring some college education as a prerequisite for admission were regular schools associated with a university.8
By 1910, there had been some progress in the overall structure of medical education. In academic year 1889–1890, about three-quarters of the 139 medical schools in operation were using a “repetitional,” or repeating, curriculum, where students studied the same subjects and content every year for two or three years. If available, laboratory courses were offered as optional Spring sessions. By 1900, however, nearly all medical schools had adopted a four-year curriculum with different subjects taught each year, typically referred to as a “graded” curriculum.9 In general, the length of the medical school term also increased markedly after 1900. In 1901, 58 of the 159 medical schools (37%) had a term length of less than 26 weeks; in 1910, the number of such schools was two out of 133 (1.5%). In contrast, 62 schools (47%) had term lengths of 33 weeks or longer in 1910, as compared with 25 (16%) in 1901.7
As part of a 1910 questionnaire from the American Medical Association (AMA) Council on Medical Education, medical schools were asked to provide information about the number of their faculty of all ranks (from professor to instructor/assistant). There were a total of 7,488 faculty in the 122 schools that provided this information. Of these schools, 13 (11%) had a total faculty size of greater than 100, and 59 (48%) had a faculty of 50 or fewer.10 The background and preparation of medical school faculty members was highly variable. Abraham Flexner eloquently deplored the use of busy clinicians to teach.1 There was, according to Flexner, no place in medical school for such a “scientifically dead practitioner whose knowledge has long since come to a standstill” and “whose lectures…like pebbles rolling in a brook get smoother and smoother as the stream of time washes over them.” 1(p57)
However, there was an elite cadre of faculty emerging, many of whom were among the estimated 15,000 U.S. physicians who traveled to Germany for advanced scientific and medical training between 1870 and 1914.3 The experience of working in German laboratories caused physicians who would later become leaders in U.S. medical education to develop a commitment to research.11 The desire to pursue this goal when they returned to the United States was one of the cornerstones of the movement to reform medical education.
An examination of Table 1 illustrates that the number of medical schools decreased in the decade before the Flexner Report. This may be attributed, at least in part, to the standard-setting activities of several organizations.
The Development and Implementation of Standards for Medical Education
The Flexner Report articulated a set of standards for medical education and used them to evaluate existing schools. This was not the first time such a process had been used. For various purposes, previous reviews had been made by the Illinois State Board of Health, the Association of American Medical Colleges (AAMC), and the AMA Council on Medical Education.
The emergence of standards
In 1891, the Illinois State Board of Health issued its seventh, and most detailed to date, report on the medical schools in the United States and Canada. The report included a succinct summary of the defects in medical education at that time:
1. Too little preliminary education, and thus a lack of ability [of students] to grasp scientific principles.
2. Too much didactic work by the teachers.
3. Too little clinical work by the students.
4. Too few tests of practical work.
5. Too short a time of actual work and study. 12(p23)
These defects had been identified, in part, through a review of medical schools in the context of the licensure laws and requirements in each state and territory, especially premedical requirements and the expected length of training. The Illinois State Board of Health report provided information on each medical school, including requirements for admission, the length and structure of the curriculum, and whether the diploma of the medical school was recognized by the local medical licensing board, if one existed.12
The AAMC was formed in 1876. The 22 member schools agreed to a set of standards, including a curriculum of two terms not occurring in the same year. In 1882, the move to raise the requirement for membership to include only those schools having three terms of formal education led to the AAMC's being dissolved. In 1890, 66 medical schools met to reform the organization and to discuss the following requirements that schools must meet to become AAMC members:
1. An examination in English as an entrance requirement.
2. A graded curriculum of three terms of at least six months each.
3. Laboratory teaching in chemistry, histology, and pathology.
4. Written and oral examinations as a requirement for credit in individual courses.13
In the next 15 years, standards for membership increased. In 1905, the AAMC expected that member medical schools require a high school (at a minimum) education for admission or require applicants to have passed an examination in subjects taught in such high schools. Members should also have a four-year medical curriculum of no less than 4,000 hours.13
The AMA was founded in 1847 and soon formed a Committee on Medical Education with the charge to prepare an annual report on the condition of medical education, including the curriculum and the requirements for “practical” (clinical) teaching. The first report of this committee, issued in 1848, included the following series of recommendations:
1. The introduction of clinical teaching in hospitals so that medical students have access to practical experiences.
2. The appointment of physicians and surgeons to hospitals based on “professional and moral worth,” rather than “political or other grounds.”
3. As a requirement for graduation, the preparation of case reports from patients whom the student had observed.14
The Committee on Medical Education, while reporting regularly, admitted that it had “little power to control medical education.”15 The reorganization of the AMA in 1901 led to the formation of the Council on Medical Education in 1904.15 In 1905, the Council on Medical Education articulated an “ideal standard” for medical education that included:
1. Preliminary education sufficient to entitle an individual to enter a university.
2. The completion of a five-year medical curriculum. The first year would be devoted to physics, chemistry, and biology and could be completed at a university or in the medical school. The first two years of the medical curriculum should consist of basic science courses, with laboratory. The last two should occur “in close contact with patients” in dispensaries and hospitals.
3. A sixth year as an intern in a hospital or dispensary.16
The council recognized that “this very desirable scheme of requirements” could not be immediately implemented and, therefore, developed a set of standards that included premedical requirements of at least a high school education and participation in a four-year medical curriculum, with each academic year lasting at least seven months.16
The standards are applied
By 1905, both the AMA Council on Medical Education and the AAMC were actively involved in classifying medical schools. In 1904, the Council on Medical Education began reporting the performance of medical school graduates in state medical licensing examinations.15 This evolved, by 1906, into a classification system of medical schools based on the failure rates of graduates in each school.17 In addition, the council conducted its first inspection of medical schools in 1906. The goal was to classify medical schools “on a civil service basis,” with 10 points awarded for each of the following 10 categories:
* performance of graduates in state board examinations,
* requirements for and enforcement of satisfactory premedical education,
* character and length of the medical curriculum,
* medical school buildings,
* laboratory facilities and instruction,
* dispensary facilities and instruction,
* hospital facilities and instruction,
* extent to which the first two years of the curriculum were delivered by faculty devoting their time to teaching and research (i.e., who had no clinical responsibilities),
* extent to which the school was conducted for the profit of the faculty rather than for teaching (in this case, a negative finding was desired), and
* the availability of libraries, museums, charts, and other teaching supports.
Six classifications were used, based on the 100 possible points: A, (90–100) B, (80–90) C, (70–80) D and E, (50–70) and F (50 and below). Schools scoring 70 and above were classified as acceptable.17 Existing medical schools were visited by members of the Council on Medical Education and its secretary, Dr. Nathan Colwell, and the summary results of the survey were reported in 1907. While the report noted that the council had been “exceedingly lenient in marking the poorer schools,” the results were discouraging. Of the 160 schools that were visited, 81 received scores above 70, 47 were rated between 50 and 70, and 32 were rated below 50.17
The council chair, Dr. Arthur Dean Bevan, concluded his report on the inspection with these observations and recommendations:
1. On-site inspection of schools is necessary, in that students can be prepared by drills and quizzes to pass state board examinations. State licensing board examinations should include practical laboratory examinations and, ideally, examinations of patients to assure that students are prepared.
2. A quality medical school requires state aid and private endowment. Medical education cannot be conducted only by the revenue from fees.
3. The state medical licensing boards have, and should exercise, the power to improve medical education. Medical boards should set a deadline for schools to come into compliance with requirements for education, including sufficient preliminary education, laboratory teaching in the basic sciences, and clinical teaching in hospitals and dispensaries.17
Both the AMA Council on Medical Education and the AAMC continued to elaborate their standards in the years leading up to the Flexner Report. The AAMC expanded its requirements for medical school membership in the association, which included a detailed set of entrance requirements for admission to medical school and for curriculum structure, including lecture and laboratory hours for each subject.18 The AMA Council on Medical Education published descriptions of the criteria that were used in its 1906 survey.19
The Flexner Report
By the time Abraham Flexner began his inspection of medical schools in late 1908, the AMA Council on Medical Education and the AAMC had defined many of the criteria that he used. The origins and the content of the report help to place it in the context of this and other work that had gone before.
The origins of the report
In the introduction to the Flexner Report, President Henry S. Pritchett of the Carnegie Foundation for the Advancement of Teaching noted that in November 1908, the leadership of the foundation had authorized and appropriated funding for a study of and report on schools of medicine and law. The study was to help define the relationship between professional education and colleges and universities, which was a previous focus of the foundation.1
The minutes of the AMA Council on Medical Education of December 28, 1908 included a summary of a meeting of the council members with Henry Pritchett and Abraham Flexner. There already had been correspondence with President Pritchett about the “willingness of the Foundation to cooperate with the Council in investigating the medical schools.”20 The minutes stated that
while the Foundation would be guided very largely by the Council's investigation, to avoid the usual claims of partiality no more mention should be made in the report of the Council than [of] any other source of information. The report would therefore be, and have the weight of, an independent report of a disinterested body.20
This approach was followed, in that President Pritchett acknowledged both Dr. Nathan Colwell, secretary of the Council on Medical Education, and Dr. Fred Zapffe, secretary of the AAMC, for their cooperation1 without specifying the information and support provided by those two organizations.
In his 1940 autobiography, I Remember, Flexner describes how a book he wrote on colleges came to the attention of Henry Pritchett and led to his selection for the Carnegie study of medical education. He describes correspondence from Henry Pritchett that stated the desire for the study to be conducted from the perspective of the educator, not the medical practitioner.21
Flexner noted that his preparation for the study included reading the reports that Nathan Colwell prepared for the Council on Medical Education; consulting with his brother Simon, a physician and scientist; and meeting with the leadership of the Johns Hopkins School of Medicine. He states that the exposure to Hopkins provided a pattern of ideal American medical education, without which he “could have accomplished very little.”21,22 At this time, the Johns Hopkins medical school included a college-level premedical requirement, supported full-time faculty members in the basic sciences with a commitment to research, and offered full access to the university hospital for clinical teaching.5
The conduct of the study
Flexner began his visits to schools in early December 1908.21 In his autobiography, Flexner states that he had “no fixed method of procedure” and that he “never used a questionnaire.”21 He did, however, review areas that he described as “decisive points” as to the “quality and value” of the medical school:
1. The entrance requirements, including their enforcement
2. The size and training of the faculty
3. The funding from endowment and fees for the support of the school and how the funds were utilized
4. The quality and adequacy of the laboratories used for instruction in the first two years of the curriculum and the qualification and training of the preclinical teachers
5. The relations between the medical school and hospitals, including access to hospital beds for teaching and ability for the school to appoint hospital physicians and surgeons21
In his autobiography, Flexner also described the process he used to evaluate a school so that he could, in a “few hours,” make a “reliable estimate” of whether a school was capable of “teaching modern medicine.”21 Flexner visited 155 schools in the United States and Canada by April 1910. His visits were interspersed with trips to New York to write up his findings and to confer with President Pritchett. In June 1910, the report had been finalized and circulated.22
There is some controversy about the role of Nathan Colwell, the secretary of the AMA Council on Medical Education, in Flexner's travels. The minutes of the Council on Medical Education (December 30, 1911)23 imply that Colwell traveled extensively with Flexner, and in his autobiography Flexner stated that “Dr. Colwell and I made many trips together.”21 However, Flexner later stated that the number of visits he made with Colwell was far lower.22 In any case, the “decisive points” used by Flexner were quite similar to the criteria used in the 1906 Council on Medical Education inspection.
The study results
The Flexner Report consists of two sections. In the first, he articulated a conceptual model of modern medical education that included the following expectations for medical schools, in summary:
1. An applicant who had studied biology, chemistry, and physics at the college level before entry to medical school, in preparation for the study of the laboratory branches (basic sciences).
2. A curriculum where a medical student can “acquire the methods, standards, and habits of science.” 1(p59)The curriculum should include didactic and laboratory teaching in anatomy, histology, and embryology, physiology, and biochemistry in the first year, and pharmacology, pathology, bacteriology, and physical diagnosis in the second year.
3. Access to hospitals and dispensaries where the where the student participates actively in the care of patients, under supervision. Flexner mentions the desirability of the students' referring to the medical literature.
4. Salaried faculty in both the basic and clinical sciences who devote their time to teaching and research.1
The first section also summarized the findings from the study, creating categories of schools based on their entrance requirements.
In the second section of the report, Flexner provides a brief description of each medical school, with his findings about the areas he considered as decisive points. The descriptions are pithy and, in many cases, damming. For example:
The school has nothing that can be fairly dignified by the name of laboratory. 1(p287) [Toledo Medical College]
The school building is wretchedly dirty. Its so-called laboratories are of the worst existing type: one neglected and filthy room is set aside for bacteriology, pathology, and histology: a few dirty test tubes stand around in pans and old cigar boxes.1(p237) [Maryland Medical College]
Flexner noted in his autobiography21 that he had the ability to be frank in his assessments. The Council on Medical Education agreed that it was valuable for Flexner to have been so explicit. The Council minutes of December 30, 1911 stated that
the drastic report by the Carnegie Foundation was what made it possible for the Council's classification [of medical schools] to be published with very little comment where otherwise it would have produced storms of criticism.23
Flexner also awarded praise where he believed that it was due and pointed to institutions that that met his standards. For example, he reported that the laboratory facilities of the Medical Department of Johns Hopkins University were “in every respect unexcelled”1(p235) and that St. Louis University had “excellent teaching laboratories” and had made provision for research. 1(p255)
Forces for Change Following the Flexner Report
There were several interrelated forces that combined to raise standards to those articulated by the “reformers,” including Flexner, the AAMC, and the AMA Council on Medical Education, and to bring about the demise of medical schools that could not meet the higher requirements. Those forces included the costs of making the required improvements and the requirements of state medical licensing boards.
Supporting the costs of reform
Abraham Flexner, in his typically cogent way, wrote that “medicine is expensive to teach.” 1(p142) He believed that medical schools could not be financed strictly from student fees. Endowment and support from a university were necessary for all schools, not just the elite. Increasing requirements from the profession and the state medical licensing boards led to rising costs. For example, the Council on Medical Education standards that were the basis of the 1906 review included the expectation that there be full-time salaried professors in the basic science disciplines, along with assistants to support them, as well as an appropriate physical plant and well-equipped laboratories.17 Flexner wrote that medical schools with no other source of support than fees should combine with other schools or go out of business.1 Schools relying only on fees had begun to do that even before the Flexner Report.
The role of state medical licensing boards in raising medical education standards
In his report, Flexner began the chapter on state medical licensing boards by stating that the “state boards are the instruments through which the reconstruction of medical education will be largely effected.” Flexner recommended that admission to examinations for licensure should be granted only to individuals whose fitness is based on an appropriate preliminary education and graduation from a “recognized or reputable medical school.” In schools that are “bad beyond a reasonable doubt,” the board should refuse to admit the individual to licensure. This would result in the closure of “unfit” schools. 1(p167)
The central role of medical licensing boards had been recognized by others as well. A 1904 report on medical education published in the Journal of the American Medical Association stated that a way that medical schools could be “compelled to live up to standards” was by “efficient and rigid inspection” by the Confederation of State Medical Examining and Licensing Boards (a predecessor to the current Federation of State Medical Boards). The unfavorable review of a medical school could result in the refusal of medical boards to recognize its diploma.24
By 1910, most states had the ability to deny licensure to the graduates of substandard schools, and the rest of the states soon achieved that capability.25 All but three states required an examination for licensure by 1905. In the 1890s, a few states began to require medical schools to offer a curriculum of a certain length, and by 1910, 22 states required a four-year curriculum.25 In addition, by 1910, seven state licensing boards had adopted premedical requirements of one or two years of college courses,26 and by 1914, 25 states had adopted a requirement that included one or more years of college education.27 The raising of standards by state medical licensing boards had national, not just local, implications. In 1914, 21 of the 107 existing medical schools were not recognized in from 24 to 32 states,27 and by 1920, the diplomas of schools rated Class C by the Council on Medical Education were not recognized by from 37 to 42 licensing boards.28 The combination of the rating system and the action of state medical licensing boards led to the closure of many schools, since “no intelligent student would knowingly spend his time and money in a low-grade college” where the diploma was not recognized by many states.28
Medical Education in the Decade After the Flexner Report
By 1920, there were about one half the number of medical schools than had operated in 190028 (see Table 2). The remaining schools had standards for admission and for curriculum that were significantly higher than in 1910. Of all the schools in 1920, 77 of 85 required a minimum of two years or more of college for admission and about one third of graduates in 1920 held a college degree.28 This is compared to 27 out of 133 schools (20%) requiring any college in 1910.8 More than two thirds of schools in 1920 had terms lasting between 33 and 36 weeks,28 compared with 44% in 1910 and 32% in 1905.7 In 1920, according to the latest ratings by the AMA Council on Medical Education, 70 U.S. medical schools had a Class A rating (acceptable medical schools), seven schools had a Class B rating (schools needing general improvements to be made acceptable), and 10 schools had a Class C rating (schools requiring a complete reorganization to be acceptable).28 This was a significant improvement from the results of the 1906 survey, where half of the existing schools were, graded leniently, classified as acceptable.17
The standards for preliminary education and the required number of hours of lecture, laboratory, and clinical education in the curriculum had been specified by discipline and published by the AAMC by 1908. The initial model included a 4,000-hour curriculum with 1,750 hours of lecture, 1,010 hours of laboratory, and 1,240 hours of clinical work over four years.13,18 In 1919, the perception that such detail was too prescriptive led the AAMC to change the absolute hours for each subject area to a percentage of hours for each in the total curriculum. For example, anatomy, embryology, and histology should occupy from 14–18.5% of total curriculum hours.29 As Flexner noted in his 1925 study of medical education, the curricula in most medical schools had adapted to this national standard.30 While Flexner referred to the standardization as unfortunate,30 it was an almost inevitable outcome of the process of reform.
Medical Education Reform: Are There Lessons From the Past?
It is tempting to ask what lessons can be learned from the period of reform in the first two decades of the 20th century. Why was change occurring before the Flexner Report and why did it continue at such a steady pace after the report was published? There appear to be several factors that contributed to the success in introducing the changes recommended by reformers.
The changes appeared to be directly linked to practice and the quality of care
The scientific advances in the latter half of the 19th century made an excellent case for the reforms related to laboratory and clinical teaching, as well as for the need for students to have sufficient premedical preparation to understand the science. Flexner linked the scientific and clinical methods, eloquently illustrating that the thought processes in the laboratory and in clinical diagnosis and management were identical; both were based in the scientific method of hypothesis generation and testing.1,22 New medical instrumentation and laboratory methods that were first introduced in hospitals could also be used in the physician's office, as long as the physician was properly trained. This reinforced the need for a well-equipped medical school and access to a modern hospital for clinical teaching to prepare students for practice in the community.
The changes were concrete
The changes recommended by the Illinois Board of Health, the AAMC, the AMA Council on Medical Education, and Flexner were clear, specific, and easily observable. As Flexner showed in his brief visits to medical schools, inspection and review of records could easily reveal an institution's premedical requirements and whether they were enforced, the length of the curriculum, and the availability of appropriate laboratory and clinical resources to support medical student teaching. Such focused recommendations for change were easy for medical schools and others, such as state medical licensing boards, to understand.
Decision making was centralized
There were relatively few groups that set standards for medical education in the early years of the 20th century. The profession, through the AAMC and the AMA Council on Medical Education, defined the entrance requirements, the curriculum, and the resources that medical schools should have and then evaluated schools based on these standards.
While the state medical licensing boards independently developed their own standards, they generally utilized the same categories of standards as did the “reformers” to determine which schools were acceptable. There appeared to be few if any barriers to licensing boards' introducing new requirements for admission to licensure or to later raising these standards.
Medical school financing mechanisms supported reform
Even the best medical schools needed student fees to operate, but many schools were almost totally dependent on that source of revenue. Such tuition-dependent schools could not meet the rising standards expected by accrediting and licensing bodies. These schools lost students and their fees, leading them to close or merge with more financially sound partners.
Why is Flexner Remembered?
While change in medical education occurred before the publication of the Flexner Report in 1910, Abraham Flexner's contribution was significant. His public disclosure of the poor conditions at many medical schools provided a means to galvanize all the constituencies needed for reform to occur. He could say what other reformers could not, due to their links to the medical education community.
But Flexner was not just an echo of the groups—such as the AAMC and the AMA Council on Medical Education–that informed his work. He was an educator, with strong feelings about teaching and, more importantly, about learning. He sympathized with medical students attending low-quality medical schools and refuted the “poor boy” argument that medical schools with low standards should be available for medical students who could not afford premedical preparation. In his opinion, standards were needed for the protection of the student as well as for the protection of the public. In his later career with the General Education Board of the Rockefeller Foundation, he himself provided funding so that schools could raise standards.22