The average length of United States medical school psychiatry clerkships has been gradually declining over the past 30 years, from 6.4 weeks, in 1982 (1), to 6 weeks, in 1999 (2), to 5.5 weeks, in 2010 (3). Alexander and Bostwick report being recently put in the unenviable position of having the shortest required psychiatry clerkship in North America (4). We can all empathize with the frustration that this reduction might entail, given that their 3-week clerkship is half the length recommended by the Association of Directors of Medical Student Education in Psychiatry in its 2006 position statement (5). Alexander and Bostwick are to be commended for moving beyond their frustration to examine the impact of the reduction in clerkship length on their students’ shelf scores in a thoughtful and scholarly fashion. Their challenging situation offers us an opportunity to think about the meaning and impact of clerkship length.

On the most concrete level, clerkship length might be viewed as an indicator of worth by a number of different stakeholders, from deans to department chairs to teaching faculty to students. Medical school faculty committee meetings devoted to divvying up time within an impacted curriculum can generate intense “turf battles” and hard feelings. Although these disputes are often fueled by financial considerations, the significance of the implied value judgment that accompanies those distributions should not be minimized. Clerkship directors and faculty faced with a reduction in clerkship length might feel as if they have been demoted, and that a value judgment has been made against them and their field. Psychiatrists are particularly vulnerable to this assumption, given the uphill battle that we and our patients have faced for full acceptance within the house of medicine. Recognizing that time-allocation and prominence of placement in the curriculum are important aspects of the “hidden curriculum,” (6) psychiatric educators must be concerned about the underlying message that a shortened clerkship sends to students and colleagues. The importance of these factors may be reduced but will likely not be eliminated by such alternate models for clinical training as the longitudinal clerkship experience (7). The longitudinal clerkship experience is an example of an alternate model for clinical training.

What is the impact of psychiatry clerkship length-reduction on students’ acquisition of knowledge? Alexander and Bostwick reasonably turn to shelf examination scores to address this question. As their manuscript’s title indicates, “shorter psychiatry clerkship length is associated with lower NBME psychiatry shelf exam performance.” As they acknowledge, their ingeniously-obtained data do not allow them to move beyond that association to make any conclusions about causality.

First, their comparisons are limited by the lack of randomized assignment; the difference in shelf exam performance between the groups could be explained by the greater interest in psychiatry held by the students who chose the additional 3-week exposure. Second, a significant difference in shelf exam scores was found only in comparing those students who spent 6 weeks on their psychiatry clerkship with students who spent 3 or 4 weeks. Third, students’ performance on the shelf exam was not significantly different between those students who spent 4 weeks from those who spent 3 weeks in their psychiatry clerkship.

It is possible that the data reflect a genuine reduction in acquired knowledge when the clerkship length is reduced to 4 weeks from 6. Furthermore, it is possible that the lack of difference between the exam scores of students who spent 4 weeks on the clerkship from those who spent 3 weeks reflects a genuine lack of difference in knowledge acquisition. It may be that there is a floor effect, and that students will “crack the books” sufficiently to learn the knowledge tested on the shelf even as their clinical experience is reduced. This conclusion is not of much comfort to those of us concerned with these future physicians’ eventual recognition and appropriate treatment of psychiatric disease. Niedermier and colleagues recently described a successful didactic curriculum revision that improved their students’ performance on USMLE Step 2 CK scores on the Psychiatry and Mental Disorders Subtests despite a 50% reduction in the duration of their psychiatry clerkship, but they astutely observe that the ultimate goal is to “adequately and efficiently prepare students to address fundamental clinical challenges that they would later encounter as physicians, not to achieve high test scores.” (8)

Although acquisition of knowledge is only the “tip of the iceberg,” it is currently the most easily-assessed component of medical students’ learning. Medical educators continue to grapple with the need for reliable and valid outcome measures, particularly for clinical reasoning and interviewing skills. The NBME shelf examinations do not purport to assess students’ clinical reasoning and interviewing skills—only their knowledge. The Clinical Skills portion of Step 2 has been developed by the NBME to address that gap in assessment. There is no equivalent national assessment tool specific to psychiatry.

What is the impact of psychiatry clerkship length-reduction on students’ acquisition of clinical skills? This an essential question, but one for which we do not currently have the data to answer. One would guess that a substantially shorter psychiatry clerkship experience would result in clinical deficits. It would be extremely useful for the field to develop the tools to assess this hypothesized effect, further highlighting the need for research to develop better outcome measures in medical student psychiatric education.

What is the impact of psychiatry clerkship length-reduction on students’ attitudes toward psychiatry? Alexander and Bostwick do not address students’ assessments of their shorter psychiatry clerkship experiences. This information would be of interest in developing a fuller understanding of how the length of the psychiatry clerkship might affect students and their perceptions of psychiatric patients and psychiatrists, as well as their interest in pursuing a career in psychiatry (9, 10). Clearly, more studies, with stronger methodologies (ideally prospective and multisite) are needed to address all of these questions.

What can be done to counter potentially detrimental changes in the curriculum that occur outside of our control? Alexander and Bostwick answer this question with an inspiring example: gather data and share it with your colleagues.

The author acknowledges the valuable input of the following ADMSEP Council members in the preparation of this manuscript: Drs. Greg Briscoe, Amy Brodkey, Tamara Gay, Susan Lehmann, Brenda Roman, Darlene Shaw, John Spollen, and Nutan Vaidya.