For medical educators, the work of designing curricula that develop students’ understanding in the field of behavioral sciences is both expansive and exciting. There has been no clear agreement about which of these concepts are best taught in the preclinical years and which are left to the students’ third-year psychiatry rotation. While there are recurring calls for a greater integration of behavioral health concepts into all areas of clinical care, the actual task of bringing integrated behavioral science knowledge and perspectives into physiological systems or organ-based instruction has been challenging.

This is likely due to several factors: (1) Behavioral health content experts are seldom available to jump into a series of teaching events at multiple junctures in the preclinical curriculum (2). Students often report that other biomedical content requires greater explanation and therefore more instructor time (3). Students often have a limited understanding of both the complexity and relevance of these core behavioral health concepts to the practice of all areas of medicine.

As epidemiologists increasingly point to behavioral interventions as some of medicine’s most powerful tools for reversing dangerous health trends in the U.S. (1), the importance of educating tomorrow’s physicians in the areas of behavioral science has never been greater. It is important to describe and promote curricular models that offer opportunities for the integration of behavior sciences into the "rest" of medical education.

In this article, a method of structuring problem-based learning (PBL) that may maximize the infusion of behavioral science content into all cases presented to students is discussed. In the absence of content specific-groupings of cases, students are obligated to consider a full range of learning issues, both in terms of potential diagnoses (in the hypothesis generating portion of the PBL process) as well as the treatment considerations and complex patient life circumstances woven into each of the cases. In short, any case can provide fodder for the learning of psychiatric issues. Clearly some cases will do this in a more direct way (e.g., the case of a 24-year-old woman who, for the past 8 months and in the absence of any known medical condition, has been hearing voices commenting on her behavior), while other case scenarios are more subtle in their invitation to consider behavioral science etiologies (e.g., a 67-year-old homeless woman with constipation, cold intolerance, lethargy and unremitting sadness and hopelessness in the months following a thyroidectomy).

PBL can be characterized as a collection of carefully constructed problems presented to small groups of students. The problems usually consist of descriptions of sets of observable phenomena or events that need explanation (2). While some medical schools have opted for a hybrid model of PBL (integrating both case-based and didactic instruction), several medical schools have maintained the inquiry driven student-centered small group learning that uses cases as the primary source of the students’ learning agenda. It is this latter model that has been employed in the case design described here.

Medical schools using a case-based model of PBL have usually chosen to aggregate or cluster the curriculum’s cases by theme, content area or organ system. t1 illustrates several examples.

The clustering of cases by theme or subject area permits several clear advantages. Content experts can be amassed to construct focused learning events; labs and demonstrations can be set up in advance to illustrate challenging concepts and relationships; recurrent themes in close proximity can encourage a deeper level of inquiry into the information being presented.

However, in the absence of an identified theme or organ system creating the case cluster, interesting possibilities emerge. In this study, it was hypothesized that the absence of an identified "psychiatry" or "behavior" unit of cases would encourage a persistent consideration of these dimensions of health and illness across each of the three preclinical years of the program.

The program uses 77 cases aggregated into 11 units. Each unit consists of seven cases spanning a period of 7 weeks, with one case covered per week. An accelerated case rate of two cases per week is conducted during the final unit (11) as students prepare for United States Medical Licensing Examination (USMLE) Step 1. At the close of each tutorial session, students’ learning issues (defined by the program as: A question formulated by students in the PBL process that requires a specific answer to be researched and discussed in the following session of class) were entered into a database by the group’s faculty tutor. Learning issues were then characterized as belonging to one or more of the following 23 subject areas: anatomy, behavioral science, biochemistry, biostatistics, cell biology, clinical medicine, diagnostics, embryology, ethics, genetics, health policy, histology, human rights, immunology, microbiology, neuroanatomy, nutrition, pathophysiology, pathology, pharmacology, physiology, spirituality, and other.

The total number of learning issues captured across all classes numbered 2,220. Only learning issues that were covered in the small group were counted in the analysis. This is the aggregate from 77 cases completed by the third-year students, 56 cases completed by the second-year students and 28 cases completed by the first-year students.

For each class, the percentage of learning issues classified as "behavioral science" were counted as a percentage of total learning issues generated by that class. First-, second- and third-year students were shown to have no significant difference in the percentage of behavioral science learning issues they generated: 8.2%, 7.0% and 8.7% respectively (SD between 20 groups = 3.2).

A qualitative assessment of the students’ learning issues provides the following observations: (1) Beginning with their first case, all student groups included psychiatric etiologies to explain symptom clusters presented in each of their cases (2). Students acquired a behavioral health working vocabulary well in advance of the first predominantly psychiatric cases (3). The problems identified by students working through a PBL case began to include patients’ coping responses to illness and loss earlier than had been anticipated by the cases’ editors (4). The interdependent relationship between health care access and health maintenance behaviors surfaced early in the students’ inquiry (prior to their fourth consecutive case) and remained a consistent theme in many of the subsequent cases.

Previous studies have sought to describe the role played by various factors that affect small-group tutorial learning in problem-based curricula. Factors frequently described include: the structure of the cases themselves, the cognitive processes of the students, intrinsic motivational influences, the influence of the tutor on the small group, as well as the role of assessment within the curriculum (3, 4). To our knowledge, the impact of identifying a specific focus for a given set of PBL cases has not previously been described.

Problem-based learning is frequently identified as an instructional methodology that is amenable to more complete integration of basic, clinical and social sciences. Despite this theoretical advantage, case authors and curriculum evaluators frequently note the difficulty of "guiding" students into areas of behavioral sciences. In theme-based PBL curricula, it is possible that the behavioral dimension of many of the health care problems is consistently deemphasized when the curriculum identifies a specific "brain/mind/behavioral" unit where these issues are expected to be made central. In the nontheme-based PBL curriculum presented, behavioral science learning issues present at a near-constant rate throughout the students’ 3 years of study.

There are many additional factors that may account for the fairly consistent rate of behavioral science learning issues noted between separate groups of students in this particular program. A comparison of the prevalence of behavioral science learning issues in other medical schools that use problem-based learning as the principal method of knowledge acquisition would allow us to better understand the impact of placing cases into an identified thematic cluster. Such comparative data were not obtained in this study, however. School admissions practices (creating a selection bias), the physical presence of the program within a school of public health, assessment patterns, and explicit and implicit program values may provide greater influence than the absence of identified case cluster themes. Given the growing need to integrate behavioral science concepts into all areas of medical education, a better understanding of programs’ successes in this area seems warranted.