In many medical schools, classroom-based instruction of medical students (e.g., formal lectures) is primarily the responsibility of faculty members. Residents, for whom the primary responsibility is patient care, teach medical students through incidental moments and by role modeling in shared patient encounters (1). In some psychiatric departments, however, chief residents also provide formal, classroom-based instruction (2, 3).

Team-based learning (TBL) (4) is a novel modality for classroom teaching. In traditional TBL, a class is divided into teams of five to eight students at the start of a course, semester, or clerkship to maximize student heterogeneity. For example, a class of 20 students may be divided into four teams of five students. These teams are kept the same for the duration of the course, and team members evaluate their peers. A reading, such as a review article on anxiety disorders for the session on anxiety disorders, is assigned for each session. Sessions start with a readiness assessment test generated from the reading which is completed individually and then by the teams, and answers are reviewed as a class. These are followed by application-focused exercises, in which the class is given a case and the teams are asked to reach a consensus on questions about the case (e.g., a patient who has multiple anxiety symptoms with questions about the appropriate diagnosis). After reaching consensus, the teams reconvene as a class and simultaneously reveal their answers. This, then, becomes the material for the class discussion, during which key teaching points are reviewed. A single facilitator manages both the small group and the classroom components of the lesson.

Team-based learning has been used in a number of settings in medical education. At the University of Texas Medical Branch in Galveston, for example, TBL was introduced into the psychiatry clerkship. After this intervention, student opinions of group-based instruction improved, as did student scores on the United States Medical Licensing Examination subject exam in psychiatry (5). This is the only published report on using TBL in third-year psychiatry clerkships. No reports have commented on the role of trainees in facilitating TBL sessions.

During the third year of medical school, the University of Michigan divides each medical student class into eight sets of 20–25 students. Each set of students rotates through the 6-week-long clerkship in psychiatry. In addition to clinical experiences, the clerkship includes hour-long didactic sessions on nine different psychiatric topics. Given the potential efficacy of student-centered, group-based instruction and the potential to include psychiatric trainees (residents and fellows) as session facilitators, we developed TBL instructional modules for implementation during these didactic sessions. Before this change, clerkship didactic sessions were developed and implemented by various faculty content experts with varying quality of instruction.

Hour-long TBL-style modules were developed for three of nine lecture topics—mood disorders, anxiety disorders, and consultation-liaison psychiatry—by the authors in collaboration with content area experts in the department, a strategy that helps ensure accurate implementation of educational innovations (6). The application-focused exercises for mood disorders were developed with permission from University of Texas-Medical Branch materials. Other application-focused exercises were developed de novo. Each module contained a detailed lesson plan with instructions on facilitating TBL sessions.

Facilitators were assigned to administer each module on the basis of their area of interest and availability. Students were not asked whether they wanted a trainee or faculty facilitator. Facilitators were asked to review the lesson plan and module before the session and could discuss any questions.

The modules were introduced to the clerkship during the 2007–2008 academic year. Eight sets of students participated. The mood disorders module was introduced with the first set of students, but evaluation data are only available from students in the second to eighth sets. The anxiety disorders module was introduced to the second set. The consultation-liaison psychiatry module was introduced to the sixth set of students.

Because the modules were introduced in a rolling fashion, we elected not to preassign students to TBL groups or employ peer evaluations as would be done in traditional TBL. Introducing peer evaluations would have required changing the grading structure partway into the academic year. The students were asked to form four teams of five to six students at the start of each session. Facilitators were explicitly instructed to enforce this group size but not to ensure heterogeneity of experience among each group’s members. The sessions otherwise adhered to the traditional TBL model described earlier.

At the end of each session, students assessed this educational intervention using anonymous feedback forms. Questions were phrased affirmatively and rated on a 5-point scale (1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree). The goal for this assessment was to inform changes to the modules.

Two of the sessions on anxiety disorders were conducted by psychology trainees. Because the intention of this analysis is to compare student experiences with psychiatric trainees and faculty members, the feedback data from those two sessions were excluded. Uninterpretable responses (e.g., multiple numbers circled) were also excluded.

Data were summarized and are presented using means and standard deviations. Student’s t tests were used to compare scores of psychiatric trainees and faculty. This project was considered exempt by the University of Michigan medical institutional review board.

Student ratings were generally positive (Table 1). Grouping the data by whether the facilitator was a psychiatric trainee or faculty member continued to yield positive results, with all but one score higher than 4. Among the nine comparisons (three questions for each of the three modules), only one showed a difference—a modest preference for the faculty facilitators (p=0.04).

This educational intervention was generally well received by the students. The student ratings indicated agreement that the sessions were effective and provided useful information. Students also agreed that they would prefer modified TBL instruction over other forms of instruction and that they would like more sessions using TBL. This finding is in line with those reported by Levine et al. (5). Through this intervention, we met the goal of implementing TBL by psychiatric trainees without notable compromise of student impressions of the lesson. Mean ratings of sessions facilitated by psychiatric trainees were generally not different from those of sessions facilitated by faculty.

There are several limitations to the interpretation and generalizability of this study. Stratification of the data to permit psychiatric trainee and faculty comparisons leads to dramatic differences in the number of respondents in each category, potentially introducing bias. In addition, the number of facilitators for each module was small, so the results may be skewed by the performance of one or a few faculty members or trainees. Further, the data represent students in a single academic year at a single educational site. Finally, survey-based research has the limitation of sampling only those respondents who elect to complete the surveys. Use of different assessment techniques may overcome this limitation.

Although facilitators were given the lesson plan and materials to implement the modules, they were not observed during the sessions. They could have deviated from the lesson plans. Moreover, the use of standalone lesson plans without formal instruction in facilitation may not have sufficiently prepared the facilitators. There is no reason to believe that this would influence trainees or faculty differently. Future research should take this into account.

Further research could also focus on the experiences of the facilitators. Anxiety about speaking in front of a classroom, especially when the lecturers are trainees who are uncertain about their knowledge base, may be a key barrier to trainee participation in lecturing. Being excluded from the development of the lesson plan may exacerbate the anxiety of the facilitator. However, facilitators may find comfort in the fact that, with TBL, they do not have to hold the attention of the class for an entire hour or longer.

Implementation of TBL-style modules in the didactic series for the third-year clerkship in psychiatry was well received. Moreover, student opinion of the sessions, with one exception, was equivalent regardless of whether the facilitator was a trainee or a faculty member. This study, then, constitutes support for the use of TBL modules as a method for involving psychiatric residents and fellows in classroom-based instruction of medical students.

The authors would like to acknowledge Mona Goldman, Ph.D., who assisted with the statistical analysis and preparation of this manuscript. At the time of submission, Drs. Ravindranath and Gay reported no competing interests. Disclosures of editors are published in each January issue.