This July, the latest revision of the Accreditation Council for Graduate Medical Education Program Requirements for Residency Education in Psychiatry will take effect (¹). The Residency Review Committee for Psychiatry has made a number of changes consistent with the competency model of graduate medical education, including recasting learning objectives in the form of the six core competencies and emphasizing the multifaceted evaluation of residents in terms of competency. Lim and Rohrbaugh (²) have argued that one of these changes, a reduction in the time requirements for inpatient psychiatry rotations, will be detrimental to residency education. In order to evaluate this argument, I propose that we answer these questions: What is the full range of changes psychiatry residencies will need to implement to stay in compliance with the Program Requirements? How do we prepare residents for the evolving nature of hospital-based psychiatry? How should residencies evaluate whether their inpatient clinical experiences are, in fact, meeting residents’ educational needs? Ultimately, I will argue that residencies should focus on the quality of inpatient experiences at least as much as the quantity.

Program directors will note that significant changes to the Program Requirements have occurred, affecting clinical experiences, didactics, personnel, call, and especially the evaluation of residents (Appendix 1). The most dramatic change is largely organizational—namely, incorporating clinical experiences into the Patient Care competency and didactics into the Medical Knowledge competency. More likely to result in substantive changes are 1) the new minimum time commitments for program directors and associate program directors, based on the size of the residency; 2) the new requirements for evaluation of residents, including an annual clinical skills examination; 3) the inducement to develop new elective tracks for residents interested in academic psychiatry; and 4) the heightened emphasis on research literacy and research opportunities.

Though the wording of many clinical experiences has changed, the requirements are now generally less specific. No changes have been made to the amount of time required for clinical experiences (or to the number of clinical experiences that have a time requirement), with the exception of inpatient psychiatry (Appendix 2). Formerly, at least 9 months of inpatient psychiatry were required, and 12 months were "highly desirable." Now the minimum is 6 months and the maximum is 16 months, down from 18 months.

Changing the time-based requirements may allow for greater flexibility in other aspects of residency education. For example, greater time for scholarship and research involvement, particularly earlier in the residency, would allow for a more meaningful, longitudinal research experience than the current model of PGY-4 research electives. As summarized by Yager et al. (³), the Institute of Medicine has identified numerous obstacles to the development of academic psychiatrists, at least two of which are germane here. First, opportunities to conduct research are fragmented across various levels of training (medical school, residency, and postdoctoral fellowship). For example, a resident starting a project in a PGY-4 elective might not be able to complete it prior to graduation unless a postdoctoral fellowship is available locally. Second, clinical requirements "are excessive and prevent tailored training." Though not all stakeholders have agreed with this premise (⁴), the Institute of Medicine has suggested that residents could fulfill clinical requirements at "an accelerated pace, with competency being used as the measure." Program directors have at their disposal at least one model of a research track that attempts to address these obstacles (⁵).

But why do we have timed requirements in the first place? Though time-based rotations have been the norm, they are not necessarily consistent with a competency-based approach to graduate medical education. Historically, the Accreditation Council for Graduate Medical Education (ACGME) "determined the necessary duration of rotations" and "simply required that programs document residents’ completion of the required rotations and clinical experiences" (⁶). Time has been considered a proxy for skill. Sometimes, as a program director, I feel more like an accountant than an educator, developing meticulous spreadsheets tracking the number of days, weeks, or months each resident spends on each rotation. I am sure most educators have witnessed residents developing appropriate clinical skills well within the time allotted for the rotation, while others still appeared to "need a bit more work" even at the end of the rotation. Residencies should judge educational experiences on the basis of meaningful outcomes for learners (quality) and not on time constraints dictated by service requirements (quantity). Indeed, the new ACGME requirements will make program directors focus on evaluation of competency, with the hope that these assessments will lead to improvements in the residents’ performances and in the clinical experiences themselves (⁶). Though I am not suggesting that we do away with time-based criteria entirely, focal changes (e.g., the reduction of inpatient requirements) will give program directors greater flexibility in assessing when residents have successfully completed a rotation.

This leads to our second question: how do we prepare residents for hospital-based psychiatry? Given the changing nature of inpatient care and the rise of the hospitalist movement (⁷), we could use this opportunity to rethink the current model of training residents for inpatient care. Inpatient psychiatry units have played a central role in medical education since at least the 1930s, serving as a "small university" for physicians and other health care professionals (⁸). Financial and regulatory pressures have led to dramatic changes, however, including decreased access to inpatient services, shorter lengths of stay, higher acuity and medical comorbidity, and the "revolving door phenomenon" (⁸). Our colleagues in internal medicine have arguably done a better job of recognizing the need to train hospitalists—that is, physicians who work specifically in this setting. Residency tracks and fellowship programs for hospitalists now exist and focus on domains not usually emphasized in residency training, including health economics, subacute care, and quality assurance and improvement (⁹).

Though there are few data regarding psychiatrists who practice mostly or exclusively in hospitals, it appears that many inpatient psychiatric units, especially in academic settings, are increasingly relying upon hospitalists. The inpatient psychiatric setting has undergone dramatic regulatory changes, for example with regard to the use of seclusion and restraints (¹⁰), and is ripe for quality improvement research (¹¹). Thus, it behooves us to consider how to train residents to be hospitalists. No amount of time on inpatient units in the first or second postgraduate year will prepare residents adequately, especially for more sophisticated systems and quality improvement issues.

I would propose a "Hospitalist" elective track to address this need. As noted above, the revised Program Requirements explicitly allow for the development of new elective tracks within psychiatry residencies (¹). Like other residents, those in the hospitalist track would complete a 6-month PGY-1 inpatient clinical experience that emphasizes interviewing skills, psychiatric diagnosis, management of acutely ill patients, working within an interdisciplinary team, and consolidation of professional identity. The hospitalist track would add inpatient electives in the PGY-3 and -4 years that would allow residents to 1) lead an interdisciplinary team, 2) begin to understand health care economics and the regulatory framework of inpatient care, 3) coordinate care among inpatient and outpatient providers, 4) develop and implement a quality improvement project, and 5) develop other administrative skills. The evaluation of the educational merits of such a track could include comparing how well prepared for inpatient practice are residents who complete the hospitalist track versus residents in a conventional training program who have spent the same amount of time on inpatient psychiatry rotations.

Which leads to the final question: once we’ve determined educational objectives for inpatient psychiatry, how do we know our training programs actually "deliver"? As Lim and Rohrbaugh (²) point out, there is an astonishing lack of research on the nature and duration of the inpatient clinical experience necessary to train residents appropriately. I would add to their literature search reports on an educational intervention to improve residents’ charting (¹²), the use of a structured approach to gathering a family history (¹³), and the effect of the elimination (in the early 1970s) of the internship on inpatient psychiatry training (¹⁴). Note that all but one of the studies cited here and by Lim and Rohrbaugh were published at least 10 years ago; that is, prior to the substantial changes in inpatient psychiatric practice described above. A recent review (¹⁵) of educational research on inpatient rotations for internal medicine residents also found no studies assessing the "core" inpatient experience, though the authors described 13 studies of specific interventions that could serve as models for educational research on inpatient psychiatric units.

At a bare minimum, we should survey program directors about the current state of inpatient training (e.g., duration, setting, overlap with other requirements such as geriatric psychiatry, challenges faced) and about plans to change inpatient training (including in response to the reduction in months the ACGME requires). Educators could use any number of "educational epidemiology" strategies to empirically test the assertion that there is an optimal number of months of inpatient experience (¹⁶). For example, a prospective cohort study could compare the educational outcomes (e.g., medical knowledge pertinent to inpatient psychiatry, psychiatric interviewing skills, pass rate on the Boards examination, and patient satisfaction with the care provided by psychiatrists in practice) of residents who have had different durations of inpatient experience. Certainly there are ample opportunities for educational research in this area.

So, is the reduction of inpatient psychiatry requirements a danger or an opportunity? I recently learned that, despite many popular references to the contrary, the Chinese word for "crisis" is not composed of the words for "danger" and "opportunity" (¹⁷), thus losing from my collection of chestnuts one that might have been relevant here. Be that as it may, psychiatry educators seem to be perpetually in crisis: not enough resources, not enough money, not enough time—and a new set of ACGME requirements every 2 to 3 years to add to our plight. But the real danger here is not that our residents will spend less time on inpatient psychiatric units but that the time spent will not be as educationally valuable as it should be. We should use this opportunity to examine the educational practices on our inpatient units and to design and test curricula that will prepare residents for modern inpatient psychiatry.