Harvard Medical School
A Randomized Controlled Trial Evaluating the Impact of the New Pathway Curriculum
Despite many calls over the last fifty years for reform of the unintegrated, lecture-dominated curricula characteristic of American medical schools, few schools have embarked on change. Researchers in the Office of Educational Development at Harvard Medical School believed that the reluctance to reform the curriculum was in part attributable to the absence of any assessment of non-traditional forms of medical education that included a true control group. The purpose of this project was to carry out just such an assessment.
In 1985, Harvard initiated the New Pathway curriculum. This curriculum emphasized active, problem-based, student-directed learning in small groups. The program's learner-centered instructional strategies stood in contrast to the passive, large group lecture method in use at most medical schools. In the New Pathway, basic science and clinical medicine were integrated throughout the four years, with students interacting with patients from the first week, and returning to basic science even in the fourth year. Each New Pathway participant developed a multi-year relationship with a generalist preceptor. The program was based on the notion that learning is facilitated when knowledge, attitudes and skills are taught together in relation to real clinical problems.
Students in the New Pathway Program were randomly selected from a list of volunteers. Those not chosen went through the traditional program with the non-volunteers. Thus three groups were available for comparison: those selected for the New Pathway Program, those who volunteered for the program and were not selected, and those who opted for the traditional program.
The project involved a comparison in six major areas between the volunteers who were selected for the New Pathway Program and those who volunteered but were not selected. The six areas were: 1) entry characteristics and study participation; 2) the cognitive domain: factual knowledge and learning behaviors; 3) clinical skills, including reasoning and problem solving; 4) psychosocial orientation; 5) students' personal experience of the old and new curricula; and 6) career choice. Students entering in 1985 and 1986 were followed for four years to assess both their responses to their educational experiences and the differences in outcomes for the two groups that might be associated with the differences in the design of their education.
The students who volunteered for the New Pathway Program but were not selected provided a true control group. The two groups were of equal size and well matched on a wide range of factors including demographic characteristics, college grade-point average, research experience, Medical College Admission Test scores and several psychosocial characteristics.
More importantly, the control group could be presumed to have applied to medical school with the same orientation toward medical education as the experimental group: they had, after all, volunteered for the new program and had been assigned to the traditional one as the result of a random selection process. Interestingly, though, by the time they arrived on campus the non-selected volunteers (control groups) showed a lower level of preference for discovery-style learning and a more positive orientation toward teacher-directed learning. This statistically significant difference may have resulted from chance or could have reflected an early adaptation to the methods and expectations of the curriculum to which these students had been assigned.
The researchers used a wide variety of bases for comparison of the two groups (34 in all) including demographic data, standardized test data, psychological measurement instruments, standardized patient assessments, cognition and reasoning tests, student statements, interviews with students and student participation in various sorts of learning activities. They were thorough and got surprisingly good cooperation from their subjects.
Though no formal evaluation of the assessment activity which constituted the project itself was undertaken, the project's completeness, care of design, and careful interpretation of findings give it substantial surface credibility.
The project's confirmation that the New Pathway Program was producing medical school graduates with levels of knowledge and clinical skills comparable to those in a traditional program at this highly selective medical school is an important finding. Such a conclusion can serve to reassure skeptical faculty that, at least in the case of the very brightest, a curriculum that is more student-centered, fosters more active learning, provides better integration of basic science with clinical practice and places more emphasis on the social and psychological aspects of medical practice produces at least as high a degree of technical competence as the more traditional program. The demonstration that students develop significantly higher levels of humanism, as measured by a broad array of assessments at the end of year two, should affirm that enhancement of psychosocial competence can be achieved without measurable loss of technical biomedical competencies. These findings should help answer the question "Can we teach humanism?" with a strong "yes," in years one and two.
If the program's other outcomes (see below) are seen as advantages, this research can reinforce the efforts of medical curriculum reformers. It is too soon to know whether this work will contribute to such an agenda or not. One indication may be that, at least at Harvard, the entire curriculum has shifted to a more student-directed, problem-based format since the program has been in effect. Faculty who have had experience with the program tend to adopt its strategies.
While students participated at high levels during their first two years, data from the last two were more sparse. Students did not complete some of the instruments that required time and thought, or participate in certain exercises, such as standardized patient interaction or clinical problem-solving. Participation of larger numbers might have demonstrated more marked differences between the two groups, providing stronger affirmation of tendencies that are based more on convergence of findings than on sharp differences in data.
Major Insights And Lessons Learned
In addition to demonstrating an equivalence in knowledge and clinical competence between students in the traditional and non-traditional programs, the program evaluation revealed that students in the non-traditional program learned and employed more psychosocial skills, including more advanced skills in relating to patients during the first two years of study, and had more positive attitudes about the psychosocial aspects of their relationships with patients. Whether these effects persist to the end of the clinical study years could not be established due to the reduced size of the sample. New Pathway students preferred the student-directed learning environment and developed an ability to study for understanding rather than engaging primarily in rote memorization. The comparative lack of structure in the new program was more stressful, but students were more challenged and less bored by it. They also knew and were better known by their faculty during the preclinical years.
The New Pathway Curriculum was the subject of a NOVA program on the Public Broadcasting System. Since completion of the project, faculty who participated in this study of the New Pathway received a grant from the Robert Wood Johnson Foundation to evaluate their new national initiative "Preparing Physicians for the Future."
The Office of Educational Development at Harvard Medical School has developed an internal program evaluation that is supported by its operating funds and additional extramural support.
Two research reports, several smaller papers, the FIPSE final report, and course materials are available from:Gordon T. Moore
Office for Educational Development
Medical Education Center
260 Longwood Avenue
Harvard Medical School
Boston, MA 02115