Graduate medical education must take place in an environment of inquiry and scholarship in which residents participate in the development of new knowledge, learn to evaluate research findings, and develop habits of inquiry as a continuing professional responsibility.

The responsibility for establishing and maintaining an environment of inquiry and scholarship rests with the teaching staff ... The staff as a whole must demonstrate broad involvement in scholarly activity. This activity should include active participation ... in clinical discussions, rounds, and conferences in a manner that promotes a spirit of inquiry and scholarship ... in journal clubs and research conferences ... in regional or national scientific societies ... in research ... and offering of guidance and technical support for resident participation in research [and] ... scholarly activities [1].

Going beyond this common emphasis on faculty scholarship, seven Residency Review Committees (those for Colon and Rectal Surgery, Emergency Medicine, Family Medicine, Pathology, Psychiatry, Physical Medicine and Rehabilitation, and Urologic Surgery) encourage resident research during the core residency program [1-6]. Three Residency Review Committees (those for Internal Medicine, Orthopedic Surgery, and Urologic Surgery) now mandate that all core residents participate in research or other scholarly activities. Specifically, as of July 1994, the new Internal Medicine special requirements state:

Prior to the completion of training, each resident must demonstrate some form of acceptable scholarly activity. Scholarly activity may include original research, comprehensive case reports, or review of assigned clinical and research topics [1].

Within internal medicine, the tradition of research by subspecialty fellows is well established and widely accepted. Although the rationale for resident scholarly activity may at first seem inconsistent with the demands of clinical training, I believe that the theoretical basis for this requirement is both rational and compelling. The participation of residents in research—broadly defined to include an array of scholarly pursuits—is one distinction between vocational training and medical education [7]. Resident research might be an indication that a program's service-education dichotomy is appropriately balanced. Residents who engage in research during their clinical education have a unique opportunity and stimulus to develop critical appraisal skills and to acquire habits of intellectual curiosity and life-long learning [8]. They gain a first-hand appreciation of the scientific method and the generation of new knowledge. And they can better discern their own academic interests and potential for a research career.

Irwin Schatz chaired the Residency Review Committee for Internal Medicine during the development of the new special requirements:

With increasing pressures toward loosening our allegiances to the scientific basis of medical practice, it is more important than ever that residents understand the scientific and the scholarly base of internal medicine, and that they participate in it to the degree that it is practical and feasible. Scholarship rather than classic bench research is the key. It is critical for residents to use scholarship in the pursuit of clinical science and for programs to demonstrate this scholarship (Schatz I. Personal communication).

Despite the intellectual appeal of resident participation in research, the practical questions of feasibility and implementation must be answered. Is it reasonable to expect internal medicine residents, who often spend 80 hours a week in clinical activities, to conduct a research project? What are the barriers to this, and what resources are important to ensure success? Can community hospital programs develop resident research programs despite limited resources? In this context, the survey by Alguire and colleagues [9] in this issue provides timely and useful insights for both residency directors and internal medicine faculty.

The survey by Alguire and colleagues [9] was completed by program directors at 271 of the 415 accredited internal medicine residency programs during the 1993-1994 academic year—just before implementation of the new special requirements. Program directors were asked about the "nuts and bolts" of scholarly activity in their residency and about the barriers to and feasibility of a mandatory resident research program. Notable findings included the following:

1. Nearly all surveyed programs expected residents to engage in scholarly activity.

2. Thirty-eight percent of programs had identified a director for resident research. These directors typically spent at least 10% of their time teaching, mentoring, and supervising research activities.

3. University-based and non-university-based programs were remarkably similar in technical resources and in research curricula. Most taught basic research skills in lectures, seminars, and journal clubs. Many still offered this training only on an elective basis.

4. Research mentors, perhaps the single most important component of the successful research equation [10, 11], were available in 96% of surveyed programs. Although "academic preparedness" and protected faculty time were less prevalent among nonuniversity faculty members, lack of resident time and interest were thought to be more important barriers to resident research.

5. Sixty percent of residents were involved in research by their third postgraduate year. Approximately half of the residents' projects included hypothesis-driven research, case series description, population studies, or analytical literature reviews (such as a meta-analysis)—projects that have the greatest potential for presentation or publication.

6. Most program directors rated the educational goals of resident research very highly. Fifty-four percent thought that research should be mandatory, but many believed that compliance with the Residency Review Committee special requirement for research would be difficult.

Should the research requirement be mandatory? Is participation in scholarly activity germane to the training of the future primary care internist? Perhaps the experience in the residency program at the Mayo Graduate School of Medicine will help to illuminate these issues. When the concept of mandatory resident research was introduced in 1988, both faculty members and residents were quick to point out the barriers—the same ones elicited by Alguire and colleagues [9]: Residents were much too busy, and the faculty members were overcommitted. The elective time available was inadequate for universal participation, much less excellence, in resident research. Many faculty members feared that a mandatory research requirement would be detrimental to resident recruitment and would discourage applicants interested in primary care.

Despite the dire predictions, our research program has been an unqualified success. Residents average 2.2 projects in 3 years. Since 1990, 15 residents have won the Paper, Poster, or Clinical Vignette Associates Competition at the American College of Physicians annual meeting. One winner's research was subsequently published as a lead article in The New England Journal of Medicine [12]. An institutionally developed resident research curriculum has been copyrighted and is widely distributed to other residency programs. Resident acceptance, at first lukewarm, has been excellent since the articulation of the research expectation to applicants and new residents. The research requirement has clearly enhanced our ability to attract and recruit Alpha Omega Alpha graduates, many of whom have an interest in primary care. Some residents have discovered an aptitude for research that has led to significant career redirection. Previously, research had been "highly encouraged" but inconsistently experienced. The 1988 mandate that all residents participate in some form of scholarly activity has, in the opinion of the residents and faculty members, uniquely catalyzed the spirit of inquiry and scholarship among the housestaff. I am reminded of those music lessons our parents "mandated" during childhood, the value of which we can only appreciate as adults.

Community hospital [13, 14] and university-based [11, 15] residency programs have reported similar experiences with required resident research. They have identified these essential elements in their success:

1. A dedicated director of resident research, who must organize the curriculum, monitor resident projects, and act as the "chief mentor" for both faculty members and residents.

2. A clear articulation of research goals, expectations, and timetables. Resident projects must be sharply focused and limited in scope.

3. A curriculum in research methods, medical information sciences, and the communication skills necessary to write a paper or present an abstract.

4. Dedicated block time used to initiate a project.

5. The opportunity to present resident scholarly activity to the faculty, either in a local research forum or in an internal housestaff "journal" [16], and at regional or national meetings.

6. Most importantly, enthusiastic and willing faculty mentors who provide the guidance and encouragement that is necessary for residents to succeed in scholarly endeavors.

The question of "outcomes" remains unanswered by Alguire and colleagues [9] and by the anecdotal experience of individual residency programs. Although difficult to define and measure, future studies should attempt to critically analyze the effect of scholarly activity on those attitudes and behaviors we hope to foster in our residents: scholarship, intellectual curiosity, reevaluation of accepted dogma, and the use of evidence-based medicine [17]. These are the same attributes, after all, that characterize the intellectual tradition of internal medicine; they must form the foundation of our future clinical practice.