Blog

Wampold, B. E., Baldwin, S. A., Holtforth, M. G., & Imel, Z. E. (2017). What characterizes effective therapists. In L.G. Castonguay and C.E. Hill (Eds.) How and why are some therapists better than others? Understanding therapist effects. Washington, DC: American Psychological Association.

The research on therapist effects indicates that some therapists are more effective than others. Previous research showed that therapist characteristics like age, race, ethnicity, gender, and experience are not consistently related to patient outcomes. Neither is therapist competence and adherence to a treatment approach. In this chapter, Wampold and colleagues ask the question: what characterizes effective therapists? The research is complicated because it is difficult to disentangle therapist effects from patient factors. That is, it is possible that some clients (i.e., those who are more motivated, likeable, and psychologically minded) might create favorable conditions for some therapists to be more effective. However, recent advances in statistical methods have allowed researchers to isolate the effects of therapist characteristics from patient factors. Based on this new research, Wampold and colleagues identified four characteristics of effective therapists. (1) The ability to form an alliance across a range of patients. The therapeutic alliance is defined as the agreement on tasks and goals of therapy, and the affective bond between therapist and patient. Alliance is reliably associated with good patient outcomes. Research shows that therapists and not clients are primarily responsible for the alliance-outcome relationship. (2) Facilitative interpersonal skills – which includes verbal fluency, warmth, empathy, and emotional expression. These skills in a therapist are a strong predictor of patient outcomes. (3) Professional self doubt – or healthy skepticism about one’s abilities and skills leading to self-reflective practice has also been found to predict positive patient outcome. (4) Deliberate practice - defined as individualized training activities especially designed to improve specific aspects of an individual’s performance through repetition and successive refinement. The amount of time outside of therapy that therapists engage in improving targeted therapeutic skills predicted patient outcomes.

Practice Implications

Some therapists are better than others - and demographics, professional affiliation, training, and adherence to a manual do not differentiate better therapists. Four factors are emerging as indicators of better therapists. Ability to develop, maintain, and repair a therapeutic alliance is well known to predict patient outcomes and it appears that therapists are largely responsible for the condition of the alliance. Therapists’ ability to be verbal, warm, and empathic is also key to patient outcomes. Professional skepticism about one’s abilities that lead to reflective practice is also an important characteristic in order to continually improve one’s abilities and monitor one’s outcomes. And, finally therapists who spend time outside of therapy deliberately and repetitively practicing skills will achieve better patient outcomes.

Goldberg, S.B., Babins-Wagner, R., Rousmaniere, T., Berzins, S., Hoyt, W.T., Whipple, J.L., Miller, S.D., & Wampold, B.E. (2016). Creating a climate for therapist improvement: A case study of an agency focused on outcomes and deliberate practice. Psychotherapy, 53, 367-375.

There is a lot of evidence that psychotherapy is effective – a result that has been demonstrated in randomized trials and in naturalistic setting. As I have noted numerous times in this Blog, psychotherapy is as effective as medications but without the side effects and with longer lasting results. However, there is room for improvement, especially in the effectiveness of individual therapists. Health care organizations are increasingly interested in quality improvement, which refers to efforts to make changes in practice that will lead to better patient outcomes, better care, and better professional development. One approach to quality improvement in medicine has been through audit and feedback – which involves measuring a clinician’s practice, comparing the clinician’s outcomes to professional standards, and giving the clinician feedback. In psychotherapy, the analogue is routine outcome monitoring in which patient progress is monitored with standardized measures throughout therapy, and therapists receive ongoing feedback on each patient’s progress relative to the average patient with that disorder. We know that therapists tend not to improve in terms of patient outcomes with experience alone, and some authors argue that one of the things that therapists are missing is good quality information about their clients’ progress. What would happen if an agency or organization decided to make it a priority to provide therapists with quality information about client progress? This paper by Goldberg and colleagues is a case study in which an agency deliberately created a culture of quality feedback and professional development to improve therapist expertise, therapist intentional practice, and client outcomes. The case study is of a community mental health agency in Alberta. Over 5,000 clients were seen by 153 therapists over a 7 year period (2008 to 2015) as part of the study. Clients received at least three sessions of therapy (mean = 6.53 sessions, SD = 5.02), and had a range of disorders typically seen in a mental health clinic. Therapists included 49.7% licensed or provisionally licensed professionals at the masters or doctoral level from different professions (e.g., social work, psychology, pastoral counselling), and 50.3% practicum students. Throughout the 7 years of the study, therapists saw an average of 33.52 clients (SD = 26.24). In 2008, the agency required the staff to collect outcome measures of all clients before each session (although patient scores were not tied to staff performance evaluations). This policy change caused a 40% turnover in clinical staff within 4 months (clearly a large minority of therapists did not want to participate in this new clinic directive)! These staff positions were replaced and staffing was stable after that point. In addition to requiring clinicians to provide measures on all patients (although patients could decline to participate), the agency provided monthly clinical consultations with an external consultant as a means of professional development. During these consultation, clinicians were encouraged to bring cases that were not progressing well in order to get feedback on their most challenging patients. Discussions were organized around therapeutic alliance, i.e., clarifying goals and preferences, and ways of facilitating engagement. The overall results showed a significant decline in distress among patients over the course of treatment. Of most interest was that therapists on average showed a significant improvement in their outcomes over time. That is, contrary to research showing that therapists do not improve over time when left to their own devices, therapists in this agency that received feedback and professional education around difficult cases did improve significantly.

Practice Implications

The findings of this study indicate that psychotherapists can improve over time if they receive quality information about client progress, and if they receive professional development that is tied to this information (i.e., concrete suggestions for ways of working with difficult clients). In other words, it is possible for therapist to develop expertise over time under some conditions. A significant challenge in this case study was that a number of therapists left the agency due to the quality improvement efforts. Some therapists are sensitive to or feel threatened by outcome monitoring. However, therapists who remained or who were subsequently hired by the agency showed a reliable increase in their expertise and client outcomes as a result of deliberate intentional practice, quality feedback about client progress, and concrete professional development focused on the therapeutic alliance.

Goldberg, S.B., Rousmaniere, T., Miller, S.D., Whipple, J., Nielsen, S.L., Hoyt, W.T., & Wampold, B.E. (2016). Do psychotherapists improve with time and experience? A longitudinal analysis of outcomes in a clinical setting. Journal of Counseling Psychology, 63, 1-11.

Do therapists get better in providing psychotherapy as they gain more experience? This is a long standing question in psychotherapy, and most studies that compare therapists of different experience levels have not provided encouraging findings. This large longitudinal study in a practice setting by Goldberg and colleagues is unique because they follow therapists over a number of years during their careers. That is, the authors do not focus on outcome differences between therapists with different levels of experience, but rather they see if a therapist improves over time as the therapist accrues experience. Data were collected on 170 therapists and 6,591 patients over 18 years in a large practice in the U.S. Patients were distressed adults who attended an average of 8 sessions (range = 3 to 153) across 13 weeks. Over the 18 years of the study, on average therapists saw 39 patients, saw their first patient of the study after their 5^th year post graduate school, and had been working at the practice for about 5 years. On average patients got better, so that their psychological symptoms declined significantly over the course of treatment (i.e., 50% reliably improved). These rates of improvement are similar to benchmarks set in clinical trials. Contrary to expectations, therapists tended to have slightly poorer patient outcomes as the therapists gained experience. This result remained significant even when patient baseline severity, therapist caseload size, and other factors were controlled. However, more experienced therapists tended to have fewer early unplanned terminations (< 2 sessions) than less experienced therapists.

Practice Implications

This is the first large longitudinal study that followed therapists over several years of their career. Therapists became less effective over time, although the magnitude of the deterioration was very small. At the very least, one can say that patients did not achieve better outcomes as their therapists became more experienced. The authors note that the results of this study are in contrast to a large therapist survey in which most practitioners reported that their skills improved with passing time, and in contrast to another study in which therapists tended to over-estimate their effectiveness and under-recognize failing cases. Ways for therapists to improve their skills and patient outcomes might include: engaging in regular progress monitoring, targeted learning of fundamental therapeutic skills, training with standardized patients, and setting aside time for reflection and clinical consultation.

Chow, D. L., Miller, S. D., Seidel, J. A., Kane, R. T., Thornton, J. A., & Andrews, W. P. (2015). The role of deliberate practice in the development of highly effective psychotherapists. Psychotherapy, 52(3), 337.

In 2014, Tracey and colleagues caused a stir when they claimed that there was no evidence of expertise in psychotherapy (see my July, 2014 blog). They defined expertise as increased quality of performance that is gained with additional experience – and they concluded that psychotherapy research has not provided evidence that therapist performance improves with experience. The issue is important because differences between therapists account for over 5% of patient outcomes. This seems small, but it is larger than variance in outcomes accounted for by the use of empirically supported treatments (0% - 4%), and almost as large as the variance accounted for by client-rated alliance (5% - 15%). Across a wide variety of professions (e.g., music, medicine, chess, sports), professionals’ engagement in deliberate practice results in improvement and superior performance. However, there is little evidence of this in psychotherapy. In this article by Chow and colleagues, the authors look specifically at “deliberate practice” defined as individualized training activities to improve one’s performance through repetition and refinement. To be effective, deliberate practice has to be focused on achieving specific targets and guided by conscious monitoring of outcomes over a long period of time. The authors collected a sample of 69 therapists who worked across a number of organizations and practice areas, and these therapists provided data related to 4,850 patients. Seventeen of the 69 therapists who treated 1,632 clients also provided data on professional development activities. Therapists were multidisciplinary (i.e., counsellors, psychologists, marital therapists, social workers, psychotherapists) with an average of over 8 years of experience, who worked mainly in private practice or within the national health service in the U.K., and who primarily treated adult patients with depression or anxiety disorders. Patient outcomes were measured repeatedly with a valid standardized scale, and deliberate practice was self reported by therapists using a measure that asked about the frequency and time therapists engaged in 25 activities outside of work aimed at improving therapeutic skills. On average, clients improved by the end of treatment and the effect was large (d = 1.22). As expected therapists differed in their patient outcomes (i.e., some therapists were reliably more effective than others). Therapist demographic variables, theoretical orientation, years of experience, and practice setting were not related to patient outcomes. However, the amount of time in deliberate practice activities was associated with a reduction in client distress. Compared to the less effective therapists (2.62 hrs/wk in deliberate practice), the best performing therapists (7.39 hrs/wk in deliberate practice) spent about 2.81 times more time on deliberate practice. Therapists rated the following deliberate practice activities as the most relevant to their patients’ outcomes: reviewing challenging cases, attending training workshops, reflecting on past sessions, and reflecting on what to do in future sessions.

Practice Implications

Although this is a single study with a relatively small sample of therapists, it is one of those rare studies to assess the effects of therapist deliberate practice on patient outcomes. As is the case with other professions, reviewing one’s performance can play an important role in identifying errors, altering course, and remediating problems. As Tracey and colleagues indicated, therapists need good quality information in order to learn from their errors and make adjustments so that clients can improve. Quality information might be available from progress monitoring (i.e., continuous feedback to therapists about client outcomes), which has been shown to improve client outcomes especially for at-risk cases. Chow and colleagues go further to suggest targeted learning by using standardized clients within training and supervision contexts. Deliberate practice is not only for newer or less experienced therapists, since experienced therapists also vary in their ability to engage and help clients. Highly effective therapists spend more time engaging in activities outside of their practice specifically aimed at improving their performance.

Hill, C. & Knox, S. (2013). Training and supervision in psychotherapy. In M.E. Lambert (Ed.), Bergin and Garfield’s Handbook of Psychotherapy and Behavior Change, 6^th Edition (pp. 775-811). New York: Wiley.

Research on training and supervision in psychotherapy has proven to be very difficult to conduct. Part of the difficulty with the research is that the process under study is highly complex with many interacting variables. Therapists and supervisors have different personal qualities, patients have different levels of problems, training programs differ, supervision styles differ, and therapists and supervisors differ in terms of experience, case load, knowledge, and training background. Nevertheless there exists a moderately large literature on training, supervision, and continuing education in psychotherapy. However, the findings so far have been mixed and somewhat disappointing. In their chapter in the Handbook, Hill and Knox (2013) tackle the difficult task of summarizing this literature and giving some coherence to the findings. Is training and supervision effective? Hill and Knox tentatively conclude that the answer is “yes”. They provide some evidence that novice therapists can be trained in helping skills, that trainees improve over the course of training, that supervision enhances trainees’ awareness of self and others and improves their autonomy, and that experienced therapists, including those in the community can be trained to use manuals. Despite these positive findings, the existing literature also provided some sobering results. These less supportive findings include: that nonsupervised therapists did not differ from supervised therapists on therapy alliance and patient outcomes, that supervision sometimes has negative effects on trainees and their patients, that therapist experience may not be related to better patient outcomes, and that some highly facilitative non-professionals can be just as effective as trained therapists. What contributes to making training and supervision effective? The research in psychotherapy training and medical education is clear on this question: hands-on experience is key to learning a practice-based skill such as psychotherapy. Practice is the most helpful component of skills training. In medical education research, systematic reviews have shown that traditional didactic learning (i.e., classroom style lectures) had no significant impact on physician behaviors or patient outcomes. However, interactive programs (especially supervised rehearsal of skills) did have a significant positive impact on physician behaviors and patient outcomes. Furthermore, psychotherapy supervisees reported that supervisors who were open, empathic, and who provided supportive nurturance in the context of a good supervisory alliance were most helpful to trainees to develop and improve their clinical skills.

Practice Implications

Practicing clinicians who want to get the most out of continuing education should look for opportunities in which they get hands-on experience and continuous supervision in providing the psychotherapy intervention. Other than acquiring a limited amount of knowledge, didactic training alone without practice will likely have little impact on practice. The research also indicates that supervisors and trainees who are able to develop a good supervisory alliance, and supervisors who are open and empathic are more likely to result in improved psychotherapy skills in trainees and better outcomes in patients. Binder and Henry (2010) describe the importance of “deliberate practice” in psychotherapy training and continuing education that includes: performing a task at an appropriate level of difficulty, receiving immediate and informative feedback from a supervisor, and having the opportunity to repeat the skill and correct errors.

Tracey, T. J., Bludworth, J., & Glidden-Tracey, C. E. (2011). Are there parallel processes in psychotherapy supervision? An empirical examination. Psychotherapy, 49(3), 330-343.

Parallel process was first proposed in the psychodynamic literature as the replication of the therapeutic relationship in supervision. Parallel process is also recognized as an important aspect of supervision in developmental and interactional models of supervision, even though those models do not endorse the unconscious aspects of parallel process. Parallel processes in supervision occur when: (1) the trainee therapist brings the interaction pattern that occurs between the trainee therapist and client into supervision and enacts the same pattern but with the trainee therapist in the client’s role, or (2) the trainee therapist takes the interaction pattern in supervision back into the therapy session as the therapist, now enacting the supervisor’s role. For example, a client comes into therapy seeking guidance because things are not going well in his relationships. He desires structure and direction from the trainee therapist (client’s behaviour is submissive). The trainee therapist attempts to help the client by providing guidance (therapist’s behaviour is relatively dominant). The client then responds with “Yes, but…” to suggestions offered by the trainee therapist (client’s behaviour is non-affiliative). The trainee therapist over time starts to become subtly “critical” of the client (therapist matches the non-affiliative client behavior). The trainee therapist goes into supervision complaining about the client’s “resistence” and the trainee therapist asks for help and direction from the supervisor (trainee therapist increases his submissive behavior in a parallel enactment of the client’s submissive stance). As the supervisor provides some direction (supervisor increases her dominance), the trainee therapist responds with “Yes, but. . .” (trainee therapist increases his non-affiliative behavior). The supervisor engages in more “critical” comments than usual in response to the therapist (supervisor matches the non-affiliative trainee behavior). In this way, the supervision interaction becomes a relative replication of the therapy relationship, captured in the parallel amounts of dominance/submission and affiliation/non-affiliation exhibited by the participants in relation to each other. Tracey and colleagues (2012) studied this phenomenon by coding moment by moment interpersonal interactions using an interpersonal circumplex model (i.e. a model that assesses relative dominance and affiliation) among 17 triads of clients/trainee therapists/supervisors in a series of single case replications. The authors hypothesized that relative dominance and affiliation would be parallel between clients/trainee therapist pairs and corresponding trainee therapist/supervisor pairs in contiguous sessions. Significant results were found for each dyad within the 17 client/trainee therapist/supervisor triads. Therapists in the role of trainee altered their behavior away from their usual in supervision to act somewhat more like particular clients did in the previous therapy session. Supervisors tended to engage in complementary interpersonal responses in the subsequent supervision session. This provided evidence for parallel process at an interpersonal level of interactions. Further, positive client outcome was associated with increasing similarity of trainee therapist behavior to the supervisor over time on both dominance and affiliation. That is, the more therapists acted like their supervisors in the previous supervision meeting on both dominance and affiliation, the better the client outcome.

Practice Implications

This article provides intriguing evidence for an interpersonal model of parallel process. Supervisors may choose to communicate with the trainee about how the trainee therapist and client are interacting, as well as how the trainee and supervisor are interacting. In this way, the supervisor makes the implicit aspects of the parallel process more explicit for the trainee therapist. The trainee then can make choices about how best to proceed based on the new understanding of the interactional pattern at the process and content levels of interaction. For example, a therapist and supervisor can come to understand a block in the supervisory alliance as a parallel to a similar impediment in the trainee therapist-client relationship. A supervisor working through the block in supervision to create a more collegial and affiliative environment may model for the trainee therapist ways in which to effectively and collaboratively work with their client.

Author email: Terence.Tracey@asu.edu