The Accreditation Council for Graduate Medical Education (ACGME) and the Royal College of Physicians and Surgeons of Canada (RCPSC) are dedicated to maintaining the highest standards in residency education in the United States and Canada, respectively (1, 2). Both agencies made recent changes to their psychotherapy training requirements, emphasizing evidence-based psychotherapies and achieving competence, proficiency, and accountability. Therefore, in addition to the long-established pattern of setting requirements for the process and content of training, the current changes now pay attention to the outcomes of training (3).

ACGME now requires training programs to ensure that residents “develop competence in applying supportive, psychodynamic, and cognitive behavior psychotherapies to both brief and long-term individual practice, as well as to assuring exposure to family, couples, group, and other individual evidence-based psychotherapies” (1). ACGME addresses the doctor-patient relationship more broadly by requiring that residents demonstrate the capacity to show empathy, trust, and respect for their patients. The RCPSC expects residents to achieve differing levels of expertise in the specific psychotherapies, which include proficiency (prime therapist with ongoing supervision) in supportive, crisis intervention, psychodynamic, cognitive behavior, and family (or group) therapies; working knowledge (observing therapy or cotherapist) in group (or family) therapies, dialectical behavioral, interpersonal, and behavioral therapies; and introductory knowledge (via seminars, etc.) in relaxation, motivational interviewing, mindfulness, and brief dynamic therapies (2). Residents are also expected to demonstrate empathy and rapport and develop trusting and ethical therapeutic relationships with patients.

The reasons given for broadening the scope of training in psychotherapy in the United States and Canada include maintaining psychotherapy as a skill set within psychiatry, ensuring accountability in training to competence, establishing a goal of graduating more “sophisticated” general psychiatrists, and advocating for evidence-based practice (1–4). The new requirements clearly imply that psychotherapy is to remain a major focus of training in psychiatry and that competence and proficiency in selected therapies are now the expectation for psychiatric residents. Recognizing this, we will examine the most recent psychotherapy training requirements issued by the ACGME and RCPSC by systematically reviewing the following three questions:

With each question we will examine the empirical literature and arrive at conclusions as to how training under the current requirements can meet the highest standards in curriculum content, evaluation, and method of instruction. Given the emphasis on competency and proficiency, we will conclude with a discussion on how training programs can comply with these new requirements and improve outcomes for trainees. If these new guidelines prompt us to examine and improve our methods of teaching and evaluation, they will prove to be a significant step forward in psychiatric education.

Are the Therapies Selected for Training Evidence Based?

Well-designed and executed randomized control trials with placebo or active comparisons are considered the gold standard in evidence-based medicine (5). Meta-analyses examine focused clinical questions by using explicit strategies and by rigorously appraising randomized control trials. Meta-analyses allow the pooling of large numbers of subjects upon which to calculate effect sizes; they thereby provide the highest level of evidence (Level 1) for specific treatments (6–9).

Prior to examining specific therapies, it is important to begin with the three decades of research showing the therapeutic alliance as an important variable in predicting outcome early in treatment (10). Therapist empathy, respect, genuineness, and attention to goals and tasks in treatment and ruptures in the alliance all predict positive outcomes in psychotherapy (11) and pharmacotherapy treatments (12). Specific psychotherapies that focus on alliance-building skills include the experiential or client-centered therapies (13), Luborsky’s supportive-expressive therapy (14), and emotion-focused therapy (15). Once alliance-building skills are developed, attention in training can focus on specific psychotherapies found to be effective for patients with the most prevalent and disabling conditions (16).

The purpose of this article is not to present an exhaustive review of the psychotherapy outcome literature (17–19) but rather to elaborate a focused review aimed at answering the specific questions raised earlier. We searched PsycINFO and PubMed using the MeSH terms meta-analyses, meta-analysis, and review for the specific therapies selected by the ACGME and RCPSC and for common psychiatric disorders. Only the most recent reviews (2000–2009) were included because these adhere to more rigorous standards, and earlier studies have been reviewed elsewhere (19). A total of 19 meta-analyses were found for a variety of psychotherapies across the major psychiatric disorders. All meta-analytic reviews included in this article reviewed studies that compared active treatments and nonactive treatments (general supportive) with various control groups such as waiting lists. We excluded studies that integrated psychotherapy with medication.

Cognitive behavior therapies have been the most extensively investigated psychotherapies, with large effect sizes found for the treatment of unipolar depression and anxiety disorders (20–25); moderate effect sizes for marital distress, anger management, childhood somatic disorders, and chronic pain (23); and low to moderate effect sizes for eating disorders and psychotic symptoms (23, 26). Mindfulness-based cognitive therapy shows some promise for patients with three or more episodes of depression (27), while some mindfulness-based interventions have helped patients with anxiety and chronic pain conditions (28). Although gaining popularity for the treatment of depression, behavioral activation needs further investigation (29).

“Third-wave” cognitive behavior therapies include acceptance and commitment therapy, dialectical behavior therapy, cognitive behavior analysis system of psychotherapy, functional analytic psychotherapy, and integrative behavioral couples therapy. All have been investigated for a variety of conditions, with a recent meta-analysis demonstrating moderate effect sizes for only acceptance and commitment therapy (for depression and anxiety) and dialectical behavior therapy (for borderline personality disorder) (30). Crisis intervention has been found to have a large effect size in treating patients experiencing a variety of acute distressing symptoms related to situational crises (31).

With respect to non-cognitive behavior based therapies, moderate to large effect sizes have been found for interpersonal therapy for depressed patients (32), with some evidence showing a differential response for patients with specific psychosocial stressors and “obsessional” personality traits (33) and fewer dropouts when compared with cognitive behavior therapy (34). Emotion-focused therapy is also helpful in treating depression, with improvements in self-esteem and interpersonal problems in the long term (35). Finally, moderate to large effect sizes have been reported with motivational interviewing for patients with substance abuse (36).

Although sample sizes are small, randomized control trials support the use of brief dynamic therapies for a variety of conditions, but not as initial treatments (37). A recent meta-analysis found large effect sizes for patients with complex mental health problems treated with long-term psychodynamic therapy (38). Psychoanalytically oriented partial hospitalization treatment has been found to be helpful in the treatment of patients with borderline personality disorder, showing an advantage over dialectical behavior therapy in decreasing parasuicidal behavior and hospital admissions (39). Effectiveness research also supports the use of long-term psychodynamic therapies in decreasing symptomatic distress and improving interpersonal problems, social adjustment, and self-esteem (38).

Moderate to large effect sizes support the use of many types of group-based interventions in the treatment of several psychiatric disorders (40), while family- and couples-based adjunctive interventions are important in both child and adult psychiatry, with a recent meta-analysis showing moderate effect sizes across diagnostic groups (41).

The empirical literature supports training residents in alliance-building skills and the specific psychotherapies selected by the ACGME and RCPSC. In considering which therapies should be required, we recommend that these decisions be guided by a combination of the epidemiology of psychiatric disorders, illness chronicity and severity, and the empirical evidence.

How Will We Know if Residents Trained Under the New Requirements Will Be More Competent or Proficient Psychotherapists or General Psychiatrists?

We searched PubMed and PsycINFO for all years with the MeSH terms competence, proficiency, evaluation, therapist rating scale, assessment, and the specific psychotherapies listed for training by the ACGME and RCPSC. Greater emphasis was given to recent papers addressing current psychotherapy requirements (since 2001) and specific assessment instruments relevant for evaluating competence in psychotherapy. This focused review yielded 34 relevant articles.

There has been considerable discussion regarding assessing competence in psychotherapy, with specific principles and methods being outlined by many authors (42–50). On one hand, Yager and Bienenfeld (51) define summative competence as the ability to perform to a specific skill level, which is evaluated at the end of training. Formative competence, on the other hand, is assessed at intermediate steps, with feedback provided along the way so that the final performance is enhanced. Waltz et al. (52) define competence as the therapist’s level of skill in delivering a particular treatment. In this section, we will focus on residents’ formative and summative competence in specific psychotherapeutic skills, because this poses the greatest challenge in evaluation. We will also use competence and proficiency interchangeably, because similar adjectives define them and they are assessed by the same process (53).

The literature has shifted recently toward more unbiased assessment of competence, with the use of more objective source material, such as audio or videotapes of therapy sessions, and standardized assessment instruments such as therapist rating scales (54–56). Weekly supervision with the resident that involves listening to recorded sessions and providing feedback permits a formative evaluation of competence by the supervisor. Summative competence can be assessed at the end of training by a second objective evaluator. Formative and summative competence should focus on both general and specific psychotherapy skills, with the latter being based on clearly defined criteria of competence specific to the therapy being conducted. Such criteria have been established for most of the psychotherapies involved in randomized control trials. These instruments can be utilized in training institutions, keeping in mind that they are simply tools and should be used in conjunction with other assessment methods (43). We will explore only some of these assessment instruments relevant to the current training requirements.

Table 1 lists the scales that have been used most frequently to assess therapists’ general and specific psychotherapeutic skills. These scales have been empirically validated and have good construct validity and good interrater reliability (57–76). Raters should receive training prior to using these scales, and reliability should be assessed. Scales can be used in supervision to provide weekly formative feedback to encourage summative competence. When rating audiotapes for summative competence, it is helpful to rate an early- and late-session tape so that gains in training can be assessed. Although this may seem cumbersome, it is an excellent way to assess improvement of specific therapy skills and to provide trainees with objective confirmation of their progress (50). However, this is only one form of assessing competence in psychotherapy, and other methods should be considered (43, 56). The supervisor’s evaluation of the resident’s general and specific psychotherapy skills and the resident’s ability to deal with alliance ruptures and transference and countertransference issues are all important components of the overall evaluation. In addition, multiple evaluations from multiple sources are the best approach to ensuring an accurate and comprehensive summative evaluation.

The recent mandates to demonstrate competence in the different psychotherapies call for a reevaluation of available assessment methods. The psychotherapy literature provides instruments to assess empathic skills, the therapeutic alliance, and specific competence in the different psychotherapies important in treatment outcome. Incorporating these tools in training will not only help demonstrate and quantify resident competence in a variety of therapies but will provide trainees with objective reinforcement of their growing skills. These instruments are available, have been empirically validated in outcome studies, and can be incorporated in training programs (50).

Do the New Requirements Provide Guidelines on the Best Methods for Teaching Psychotherapy to Encourage Competence or Proficiency?

Although it may be argued that prescribing training methods does not fall under the mandate of either the ACGME or RCPSC, both agencies aim to maintain standards in education that encourage competence in many areas of psychiatric education, including the psychotherapies. We searched PubMed and PsycINFO for all years with the MeSH terms psychotherapy, training, evaluation, competence, proficiency, and learning. Only the 12 papers that investigated a specific method of training (e.g., role playing, audiotapes, videotapes), utilized treatment manuals, and utilized therapist competency scales were selected for discussion in this review.

Carl Rogers and the client-centered school are credited with being the first to utilize in-session audiotapes to research therapists’ behaviors in relation to outcome (77). As an outgrowth of this research, “microcounseling” was developed and found to enhance therapists’ skills (78). In microcounseling (a teaching method in which moment-to-moment feedback is given to trainees about their activity heard on audiotapes), attention is paid to the therapist’s delivery of empathy, warmth, respect, positive regard, reflection of content, and feeling and to the therapist’s vocal quality, intonation, and tone. This training method is helpful in teaching residents the empathic skills and techniques necessary to develop a positive therapeutic alliance.

Following Rogers, Truax and Carkhuff (79) identified three key steps that improve learning relationship/psychotherapeutic skills: modeling, rehearsal, and feedback. Modeling requires the expert to demonstrate specific therapy skills, rehearsal allows the learner to practice what has been modeled, and feedback provides an opportunity for the expert to comment on the observed behavior so that this can be incorporated into new learning. Dismantling studies attempting to identify which components of learning are essential in training produced conflicting results, yet support for the triad remains robust.

The finding that learning does not occur without feedback and that feedback enhances performance has been supported by subsequent research (80–85). It is especially important to provide feedback early in training, when behaviors closely approximate the specific therapy. For these reasons systematic feedback and reinforced practice have become well-known, empirically grounded principles of learning that can be utilized to produce improvements in therapists’ competence across all therapies.

Milne and colleagues (86) trained 20 clinical psychologists, psychiatrists, and nurses in a 40-day course of cognitive behavior therapy with extended supervision for patients with depression or an anxiety disorder. Trainees received an intensive induction period, with didactic seminars that included role playing and modeling, followed by intensive supervision with session videotapes to assess adherence and competence using the Cognitive Therapy Scale as an adherence measure and the Coping Response Inventory to measure patient outcome. Results showed an increase in therapist competence, which was related to improved patient coping. The study lacked control conditions, however, to assess whether time alone or other nonspecific therapist factors mediated patient outcome.

More recent studies have continued to investigate the contributions of modeling, rehearsal, and feedback. In an intriguing study teaching motivational interviewing to licensed substance abuse professionals, Miller and colleagues (87) randomized 140 therapists to five training conditions: clinical workshop only; workshop plus practice feedback; workshop plus individual coaching session; workshop, feedback, and coaching; or a waiting list control group of self-guided training. Results revealed that although all four workshop groups showed effects on motivational interviewing proficiency immediately following the workshop, only those receiving feedback and/or coaching based on audiotape performances more fully retained their clinical proficiency over time. This is a significant finding because it demonstrates the importance of using audiotapes in the supervision of competence.

Sholomakas and colleagues (88) randomly assigned 78 community-based clinicians to one of three training conditions: review of a cognitive behavior therapy manual only, review of the manual plus access to a cognitive behavior therapy web site, or review of the manual, a didactic seminar, and ongoing supervision with audiotapes. Session tapes were reviewed and feedback was given over the phone, while clinician competence was assessed by independent ratings of structured role plays, with the use of an adherence scale. Results revealed that clinicians receiving the manual, seminars, and ongoing supervision demonstrated greater proficiency than those in the web-training or manual-only condition. The ongoing supervision with direct feedback based on therapists’ specific behaviors on audiotapes achieved greater therapist competence once again.

Kivlighan and colleagues (89) compared therapist performance and patient outcome in two training conditions of manual-based interpersonal-dynamic psychotherapy: live supervision or supervision of videotaped performance. Results revealed that the live supervision training encouraged greater use of more relationship and support interventions and enhanced overall learning of the treatment, with patients reporting stronger alliances despite the interruptions. Therapist competence, however, was assessed at session 4 only, making it possible that live supervision is simply a more efficient, rather than a superior, method of training.

This research demonstrates that evidence-based teaching methods enhance competence in psychotherapy, including expert modeling of the therapy, learner rehearsal of skills, and supervisor feedback based on actual, observed (heard/seen) learner (therapist) performance. Neither the ACGME nor RCPSC offers clear guidelines regarding teaching methods for psychotherapy. Given the available evidence, it is not too soon to incorporate these teaching methods into our psychotherapy training programs.

The ACGME and the RCPSC have changed the requirements in psychotherapy, shifting to evidence-based therapies with a focus on competence and proficiency as an outcome of training. We found that the majority of therapies listed in the ACGME and RCPSC requirements (supportive, crisis intervention, psychodynamic, cognitive behavioral, family, interpersonal, group, dialectical behavior, motivational interviewing, relaxation, and brief dynamic therapy) are supported by the most recent meta-analyses as treatments for important psychiatric disorders. Emerging psychotherapies such as mindfulness are supported by recent randomized control trials. Although competence and proficiency are emphasized as outcomes of training, and despite the availability of objective, evidence-based methods to assess competence, use of these methods has not been required. Psychiatric training could benefit by incorporating validated therapist rating scales. Finally, and most important, evidence-based teaching methods have been demonstrated to be effective in catalyzing the development of competence and could be incorporated into the ACGME and RCPSC guidelines. Historically, training residents in psychotherapy has depended largely on process notes, with residents graduating without ever having been observed conducting psychotherapy. One wonders how comfortable we would be undergoing a surgical procedure with a surgeon trained in a similar manner!

The ACGME and RCPSC have made bold moves in broadening the psychotherapy training requirements to incorporate evidence-based therapies in residency training. With these new requirements in place, we hope to graduate more competent psychotherapists and sophisticated general psychiatrists. We recommend that future revisions provide additional guidelines on assessing competence and require or encourage evidence-based methods of teaching the selected psychotherapies. This would improve our standard of training and the delivery of effective health care to our patients and fulfill the missions of the ACGME and the RCPSC.

At the time of submission, the authors reported no competing interests.