Residency training: beginning with the end in mind.

In this issue of the Journal of General Internal Medicine, Glasheen et al. provide the rationale for targeted training in hospital medicine during residency.1 As part of this rationale, the authors describe the discrepancy between the skills obtained in residency, and those required in the practice of hospital medicine.2 In doing so, they have identified the sentinel issues that face residency reform. How should residency training be modified to meet the requirements of a dynamically changing healthcare landscape? Should residency training define the practice, or should practice define the residency training? There are no easy answers to these questions, except to confirm that residency training must continually change, balancing the needs of the profession with the ideals that give it its identity. As this change occurs, there are important caveats to be considered.

Should hospital medicine be taught at all? The cynic would say that residency training is already training for hospital medicine. It is true that the emphasis on inpatient training during residency is sufficient for the hospitalist who is biding time between her third year of residency and the first year of fellowship. But as the authors point out, residency training is missing the mark for the skills important for the internist who will practice hospital medicine as a career: quality improvement, transitions of care, palliative care, multi-disciplinary team leadership, and care of non-medical patients.

No one debates the importance of these skills. The question is whether it is sufficient to emphasize these as a part of daily inpatient practice (i.e., the evaluation of core competencies on the wards), or whether it should be taught as a focused component of the curriculum (i.e., a rotation). After ten years since the integration of the core competencies into residency training, we have still not seen the wished-for result of adeptness in quality improvement and multi-disciplinary team leadership in our residency graduates. Our residents’ proficiency in patient safety and effective transitions of care is not at the level that it must be. The work-hours reform has decreased the total time on service, but it has also increased the intensity. Time for meaningful instruction and reflection on these skills has been squeezed out. It is not likely that successful instruction and application of these skills will occur given the frenetic pace of the wards.

Mastery of these skills requires active engagement and application. A resident must have time to be a part of a quality improvement project and work as a part of a multi-disciplinary team. He must have the perspective that comes with visiting the nursing home, skilled nursing facility or clinic to which he refers patients, and the time to understand the system that underlies each. The resident must have time for reflection on these systems, and mentorship from a physician who is actively involved in assessing and changing these systems. Focus, active participation and reflection are requisite, which may explain why the “drip approach” of teaching this content as a part of the core curriculum noon conference series has not been successful. It makes sense to devote protected curricular time to learning these skills.

But should the remedy involve a curricular change focused on hospital medicine? Are these skills not just as important for the subspecialist or the ambulatory-based general internist? While the generic principles may be the same, there are features of the inpatient system that are different from the outpatient system. Playing the guitar is a generic skill, but the application to a classical versus a rock-and-roll genre is very different. Allowing the resident to apply these skills to the domain in which she will ultimately practice offers a sense of utility. This in turn provides the motivation necessary for her investment in learning. For this reason, a hospital medicine elective does appear to make sense for the resident who has chosen a career in general internal hospital medicine. A corresponding ambulatory-based elective, emphasizing the same skills, but with a bent towards the features unique to the ambulatory system, makes equal sense.

But where should this curriculum be positioned? There are five options: an elective rotation, a required rotation, a defined track, a fellowship, or merely deferred to on-the-job training. It is apparent that on-the-job training has not been effective in teaching these skills. It is also unrealistic to think that a generation of physicians who were deprived of the training in these skills will be able to train subsequent generations.

A hospital medicine fellowship year has a high opportunity cost, as the difference in salary between a fellow and a full-time hospitalist can be up to $150,000 per year. The number of residents likely to choose this option will be insufficient to effect the change needed in the profession. Analogous to the general internal medicine fellowship, the fellowship option may make sense for those seeking research skills that are unlikely to be acquired as a part of standard hospital medicine practice.3

The most meaningful application of the training described in Glasheen’s article appears to be as a curricular rotation during the residency training. However, residency training has become a zero-sum game: any curricular addition will mean the disposal of something else. The value of elements eliminated must be weighed against the value of proposed additions. Twenty-two of the 36 months in residency training are reserved by the Accreditation Council for Graduate Medical Education (ACGME) requirements (wards, ICU, ER, geriatrics, neurology, consultation).4 The limitations on work-hours and caps on admissions/service size have resulted in many program directors increasing the number of ward and night float assignments, further limiting the degrees of freedom for elective assignments. The result is that there are a maximum of eight to twelve months over a three-year residency that are at the disposal of the program director, eight of which must occur in the ambulatory setting. For this reason, a hospital medicine elective seems most plausible if it encompasses pre-existing program requirements (i.e., geriatrics, medicine consultation), and is presented in a way that ensures compliance with the ACGME core competency requirements (i.e., a train-the-trainer approach whereby participating residents train other residents in systems of care).

The number of available elective months limits the feasibility of a hospital medicine track. Further, the institution of any track must not be done capriciously, and should be entered into only after reaching comfort with the answer to this philosophical question: “At what point in the residency training does pluripotency end, and vocational training begin?” Tracks by definition predispose to early career-choice closure, and any one track provides precedent for other tracks. A hospital medicine track may open the door for subspecialty tracks. Given the competitive nature of the subspecialty application process, residents may perceive that failure to follow a specialty track will disadvantage them in the application process. The result would be an even earlier career-choice closure. Most residents do not enter into residency with a full appreciation of the fulfillment and challenge inherent in the general internal medicine practice. An early career choice closure, as accelerated by the advent of tracks, prohibits the exposure to general internal medicine or any other subspecialty until it is too late to make an informed decision.5 The adverse effect to general internal medicine, both for hospital and ambulatory-based physicians, could be severe.

Importantly, a hospital-medicine elective should not preclude a corresponding ambulatory elective that focuses upon the same skills, but within the ambulatory care venue. Indeed, one could argue that a corresponding ambulatory-care elective would augment a hospital medicine elective, as the coexistence of the two would enable optimal learning and study of transitions of care. For both electives, the key feature is the protected time to learn these skills. The skills cannot be an afterthought or added learning objective on top of patient care. There must be some patient care as a part of the elective to preserve CMS-based funding lines necessary to pay the resident’s salary. However, the emphasis should be on the key skills as outlined in Table 1 of Glasheen, et al.’s article, applied within the respective venues.

But at the end of the day, the success or failure of the instruction in these skills is contingent upon the people who will teach them. The instruction of these skills requires a champion: a person who believes in the importance of a skill and finds fulfillment in its enactment. Further, these people must be trained in medical education, with the skills necessary to teach and motivate effectively. The collaborative vision of the Association of Chiefs of General Internal Medicine, Society of General Internal Medicine, and Society of Hospital Medicine to develop a training course in medical education is an important first step to creating a generation of general medicine educators equipped to enact instruction of these important skills.

The wished-for end is a generation of physicians capable of mastering the system in which they work, enacting changes necessary to ensure patient safety and advance the quality of healthcare. Glasheen et al.’s proposed curriculum is a nice beginning to this end, and this is exactly how we should proceed: beginning with the end in mind.