What is the evidence base supporting best practice supervision guidelines? A systematic review and multilevel meta-analysis of the association between supervision and therapist performance
Abstract
Clinical supervision is widely considered to be integral to effective clinical training, with best practice guidelines of each mental health discipline proposing common supervision elements for effective supervision. However, it is unclear to what extent these common supervision elements are supported by empirical evidence. The aims of this program of research are to a) describe relevant supervision models including theories and goals of supervision, b) detail common and distinct elements across supervision guidelines from three primary psychotherapy disciplines (i.e., psychology, counselling, and social work), c) review methods and preliminary findings from an initial scoping review of supervision studies, e) conduct a systematic review and meta-analysis evaluating the relationships between commonly proposed supervision elements and relevant outcomes, and f) discuss limitations of existing empirical literature on clinical supervision, and (g) propose future directions for supervision research. First, we reviewed professional supervision guidelines for psychologists, counselors, and social workers and identified 17 common supervision elements across these best practice supervision guidelines. Next, we conducted a scoping review, screening 1517 published articles, and coding 28 studies from 26 articles. These supervision studies examined the relationship between common supervision elements and numerous supervisor-, therapist-, and client-level outcomes. Some supervision elements were examined more often than others (e.g., therapist outcomes). In addition, some supervision elements were more consistently associated with positive outcomes (e.g., fostering a collaborative supervisor-supervisee relationship, encouraging therapist self-evaluation). Next, we conducted a systematic review and meta-analysis to improve understanding of the associations between common supervision elements and meaningful therapist outcomes. We searched for studies that included supervision provided to mental health professionals with pre- and post-supervision measures for at least one relevant therapist outcome (i.e., adherence, competence, self-efficacy). Of 346 full-text articles screened for eligibility, 29 studies from 32 articles met inclusion criteria and contained sufficient data for quantitative analysis. Across these studies, we identified 110 effect sizes from 38 groups receiving supervision with at least one common supervision element, and we also identified 1 group receiving supervision with no common supervision elements, and 7 groups receiving no supervision (including 6 groups that received training). Results indicate that the average pre-post effect sizes for the 38 groups receiving supervision with at least one common supervision element was statistically significant and in the large range, indicating improvement in therapist outcomes following supervision (g = .787). Most of the hypothesized moderators were unrelated to therapist improvements. Contrary to hypothesis, the more common supervision elements incorporated into supervision, the less therapist improvement was achieved pre to post supervision. We also examined the impact of each of the 17 identified common supervision elements separately. None of the common supervision elements were related to greater improvement in therapist outcomes following supervision, when compared with supervision not containing the given common supervision element. Concerningly, the lack of three common supervision elements were each associated with more therapist improvement pre to post supervision: beneficence for the client, technology considerations, and general feedback. Our results suggest that without studying common supervision elements from best practice guidelines systematically (i.e., evaluating the impact of specific common supervision elements on meaningful therapist outcomes), it is challenging to identify which supervision elements will maximize the benefit of supervision. Gaps in existing supervision research and future directions for evaluating supervision are discussed.
Degree
Ph. D.