Learning psychotherapy : An effectiveness study of clients and therapists

Sammanfattning: Background Many psychotherapy studies with trainees have been conducted, but few have investigated how effective baseline trainee-led psychotherapies are. Baseline trainee-led psychotherapies are often provided by a professional education, and the therapists are often young, untrained and inexperienced. The present study was conducted at the Clinical Psychology Program at Umeå University, in Sweden. The psychology students were in their fourth or fifth year of, in total, five years, and few had practiced therapy before. Clients, students and education providers are interested in the outcome of trainee-led psychotherapies because clients want an effective treatment, and students and the educators want the best education. In research, there is an interest in knowing more about training, how training influences clients’ benefits of therapy, and how training works in regular activity. In the present thesis, we investigate questions related to outcome and how different training factors affect outcome. The overall purpose of the present thesis was to examine 1) the effectiveness of trainee-led therapies in a psychology education setting and 2) if clients’ self-image patterns would predict the outcome 3) if different training conditions covary with treatment outcome 4) how novices develop in their professional characteristics and work involvement styles.Methods and Result The current thesis utilized data from the Swedish naturalistic study Effects of Student Therapies (EUT) at Umeå University. The EUT is a naturalistic psychotherapist research project, which comprises client data from 2003 to 2012. The present study included 235 clients. The mean age of the clients was 31 years (SD = 9.66), and 69% of the clients were women. The clients had mixed psychological symptoms and were well functioning. Psychological symptoms were measured by Symptom Check List 90 (SCL-90; Derogatis, Lipman, & Covi, 1973). The patients’ self-image was measured using the Structural Analysis of Social Behavior (SASB), the introject questionnaire (Benjamin, 1974). All therapists were students at the psychology program. In Paper III, 76 therapists participated. The therapists’ mean age was 28 years (SD = 5.55), and 71% of the therapists were women. Therapists’ professional characteristics and work involvement styles were measured by Development of Psychotherapists’ Common Core Questionnaire (DPCCQ; Orlinsky et al., 1999).Four specific objectives have been addressed. The first objective was to investigate the overall effectiveness of treatment. In Papers I and II, the effect sizes implicated that the therapy outcome was moderate. Paper I showed that 67% of the clients were in the dysfunctional domain before therapy compared to 34% after completed therapy. Further in Paper I, it was found that 42% of the clients had recovered or improved at the end of the therapy, but most of the clients remained unchanged (55%) and a few percent had deteriorated (3%). This result is in line with a Norwegian training study (Ryum, Stiles & Vogel, 2007) but less effective than effectiveness studies have shown with professional therapists (e.g. Hunsley & Lee, 2007). Paper II, where we used a subsample of Paper I’s clients, showed a similar result.The second objective was to investigate if clients’ self-image pattern (attachment group, disrupted attachment group, self-control and self-autonomy) predicted change in psychological symptoms (GSI: global severity index) and personality symptoms (PSI: personality symptom index). The disrupted attachment group or the clients’ negative self-image had the strongest relationship to outcome and explained 8% vs. 10% in outcome (PSI vs. GSI). Self-control explained a further 3% (GSI) and 4% (PSI) of the result, and self-autonomy added 1% in both GSI and PSI. The result indicates that clients with an increased negative self-image, higher self-control, and lower level of self-autonomy before therapy improve more in both psychological symptoms and personality symptoms than clients with a less negative self-image, lower self-control, and higher level of self-autonomy.The third objective was to explore if treatment duration (one or two semesters) and training condition (cognitive therapy and psychodynamic therapy) could affect basic psychotherapy outcome. Paper II demonstrated that clients in all training conditions, cognitive therapy two semesters (CT2), psychodynamic therapy one semester (PDT1) and psychodynamic therapy two semesters (PDT2), had significant changes in self-image patterns and symptoms, except for cognitive therapy one semester (CT1). Analyses using clinically significant change demonstrated that fewer clients in CT1 had recovered and reliably improved compared to the other training conditions (in CT1: 20- 23%, in PDT1: 27- 43%, in CT2: 49- 54% and in PDT2: 35- 41%). Two hierarchical multiple regression analyses demonstrated that clients’ pre-tests characteristics self-image pattern (affiliation: AFF) and psychological symptoms (global severity index: GSI) explained 34% of the results. Treatment duration and training condition demonstrated an interaction effect between duration and theoretical approach, explaining about 2%. The regression lines for self-image pattern AFF and psychological symptoms GSI showed that clients in CT2 and PDT1 improved more than clients receiving CT1 and PDT2.The fourth objective was to examine how novice therapists in psychotherapy training develop in professional characteristics and work involvement styles (healing and stressful work involvement styles). The study was longitudinal and therapists were measured at session 2, 8, 16, 22 and endpoint. Mixed model analyses of the Development of Psychotherapists’ Common Core Questionnaire (DPCCQ) (when controlled for therapists’ age and gender) showed that the therapists’ professional characteristics and work involvement styles changed positively over time in training, except for in-session feelings of anxiety and boredom. The therapists increased most in technical expertise and less in basic relational skill. The result also indicated that the students changed linearly over time.Conclusion The present studies draw attention to the moderate outcome for clients in trainee-led psychotherapy. The novices appear to need time to increase in effectiveness possibly due to the high load of technical training in the beginning of the therapy. However, when exploring different training durations and training conditions, the contexts are shown to influence the outcome. In addition, clients with a more negative self-image pattern, with higher levels of self-control and lower levels of self-autonomy had better outcome, a finding with prognostic value. Finally, the training of students improves both a healing and a stressful involvement style, but in-session feelings of anxiety and boredom are more resistant to change.