Although the concept of competency-based education has been around for many years, it didn't find its way into medical education until the 1990s. The Royal College of Physicians and Surgeons of Canada began work on their CanMeds 2000 project in the mid-1990s, which ultimately delineated key competencies for medical specialists (1). The Association of American Medical Colleges (AAMC) subsequently undertook their Medical School Objectives Project (MSOP) (2,3), which in due course identified four attributes deemed essential for the physician of the 21st century. Throughout the 1990s, many medical specialty societies struggled with attempts to define and measure competencies for their members (4—9).
Within this context, in 1997, the Accreditation Council for Graduate Medical Education (ACGME) made a commitment to using educational outcomes as an educational tool. This led to their Outcome Project (10) that was partially supported by funding from the Robert Wood Johnson Foundation. After a comprehensive literature review, extensive vetting by various stakeholders, and lengthy deliberations, the ACGME, with the support of the American Board of Medical Specialists (ABMS), identified six areas of competence that all programs must address. The ACGME then mandated that each Residency Review Committee (RRC) incorporate these general competencies into their program requirements no later than June 2001. Starting in July, 2002, site visitors were expected to evaluate a program based on the development and implementation of a plan designed to define and measure these competencies.
The six general competencies are: patient care, medical knowledge, practice-based learning and improvement, interpersonal and communication skills, professionalism, and systems-based practice.
Coincidentally, at about the same time that the ACGME committed to the inclusion of the standard six competencies into the requirements of all RRCs, the psychiatry RRC was beginning the process of reviewing its program requirements for general psychiatry. The committee was acutely aware of the ACGME process but was also independently aware of needs within the specialty itself. As was true for most RRCs, requirements for accreditation were principally based on structure, function, and content of the programs. There was little emphasis on measuring the outcomes of training, but rather the emphasis was on what took place, for how long (i.e., timed rotations), and the degree of participation. Measurement of outcomes was limited to testing using the psychiatry residency in-training examination (PRITE), mock oral board exams, and reports of supervisors. However, there were no specific requirements for outcome measurements of designated skills or performance competencies.
As the RRC engaged in the process of revision, it became clear that this revision would involve more than changing existing content but would rather include new concepts of competency measurement never before appearing in the requirements. It was quickly agreed upon to accept and include the general competencies already adopted by the ACGME. Since the psychiatry RRC was the first RRC to revise its general program requirements following the adoption of core competencies by the ACGME, psychiatry took the opportunity to lead the way.
This emerging emphasis on performance and outcome measurements throughout graduate medical education stimulated the psychiatry RRC to question whether competencies specific to psychiatry needed to be included in the revised requirements in addition to the ACGME/ABMS core competencies. The answer to this question had broad and far-reaching implications for the entire field and affected many individuals and organizations beyond the RRC. Prior revisions to the general requirements had always involved the field broadly but had done so by requesting feedback after the RRC had made the major revisions. Traditionally, the field received a major draft and was asked to respond. Responses were always taken very seriously, but they were responses to a work that had already been developed by the RRC. As this revision was being undertaken in the context of landmark conceptual changes, the RRC believed that a similar landmark change in the process itself was needed. Involvement of principal stakeholders was necessary from the very beginning of the revision process, not in the middle or at the end. Even before the revision process began, some of these stakeholders expressed concern over the effectiveness of training in areas already required by the RRC. Specifically, concern over graduating residents' abilities to carry out various psychotherapeutic procedures had become increasingly obvious.
The RRC identified the American Association of Directors of Psychiatric Residency Training (AADPRT) and the American Association of Chairs of Departments of Psychiatry (AACDP) as key stakeholders in addition to parent organizations of the RRC (The American Board of Psychiatry and Neurology and the American Psychiatric Association (APA)).
The process began with the establishment of communication between the leadership of the RRC and the leadership of these critical organizations. There was particular emphasis on the AADPRT, which from the beginning took a leadership role in helping to develop the agenda. Representatives from the various organizations were invited to a retreat in April, 1997 to seek input on the delineation of specialty specific competencies for possible inclusion in the next edition of the Program Requirements for General Psychiatry. Subsequently, a collaborative process of revising these requirements in an era becoming dominated by competency concerns was initiated.
Psychotherapy quickly became central to all of the discussions regarding what additional specialty specific competency requirements should be part of the revised requirements. Concern over an apparent atrophy of psychotherapy skills among recent graduates had been growing in the field. Although the concern was not evidence-based, it was widespread. In addition, there was survey data from the AACDP to support this. In 1998, Jerald Kay queried all psychiatry chairpersons about the numerous proposals from the field regarding resident competency requirements in specific psychotherapies (Kay unpublished). More than 80% of the chairpersons of departments of psychiatry were in favor of requiring general psychotherapy competence, and the majority also believed that the psychotherapy requirements should be made as specific as possible. Fourteen psychotherapies were assessed, with the highest priority assigned by respondents to psychodynamic, supportive, cognitive behavioral therapy (CBT) and brief dynamic psychotherapies. Combined or integrated treatment was not assessed. Additionally, there was a perception voiced by many examiners of the American Board of Psychiatry and Neurology (ABPN) Part II oral exam that too many candidates were unable to conduct an empathic interview; and while candidates were knowledgeable of issues concerning pharmacotherapy, they were unable to demonstrate sufficient knowledge regarding the indications for, conduct of, and potential problems of psychotherapy for the patients they were interviewing. Candidates often would say that they would refer the patient to a social worker or psychotherapy.
Furthermore, various resident groups were noticing that adequate clinical experience and supervision in traditional psychodynamic psychotherapy was less available than had been reported by earlier generations of trainees.
Early discussions by the RRC identified a group of five psychotherapies, which the field believed to be the most commonly used in clinical psychiatry. These five therapies were: brief therapy, cognitive-behavior therapy, combined psychotherapy and pharmacotherapy, psychodynamic therapy, and supportive psychotherapy. Initially, it was believed that requiring competency in all five psychotherapies could become a burden on training programs since each program might not have the resources necessary to teach and supervise all of these areas. In the first draft, the RRC stated that trainees must demonstrate competency in at least three of the five areas, while requiring familiarity and experience (but not full competency) with the other two. The response from the field was swift and definitive. No one wanted trainees to graduate without competence in their favored therapy. In response, the RRC revised their final document in order to mandate competence in all five psychotherapies.
The RRC recognized that model curricula and specific outcome measures would have to be developed and refined in the field rather than dictated from above. Thus, subsequent to the publication of the revised program requirements, various RRC members have been actively involved in communicating and interacting with program directors, chairs, and key psychiatric educators. The AADPRT, APA, the Association for Academic Psychiatry (AAP), and other groups have held workshops and symposia on the topic. Best practices and model curricula are beginning to emerge as was predicted.
What does the future hold? The RRC is now working on revisions to the program requirements for the various subspecialties (except for child and adolescent psychiatry) and will then, once again, begin to work on the general program requirements. The impact of the competency requirements, especially psychotherapy requirements, must be assessed as legitimate questions regarding the cost and effectiveness of these recent mandates remain unanswered.