Die therapeutische Beziehung: Differentielle Gestaltung in experimentellen Designs und in der psychotherapeutischen Behandlung

Despite the fact that psychotherapy has proven effective (e.g., Wampold & Imel, 2015), research has provided few answers as to how and why psychotherapy works. Numerous meta-analyses show that about 80% of patients develop better compared to untreated control groups (Wampold & Imel, 2015). At the same time, studies of established psychotherapies show at least one-third to one-half of patients as nonresponders, and as many as 60-80% fail to reach remission or relapse after psychotherapy (e.g., Lambert, 2017). As such, there is a need for improvement. Moreover, research has not yet provided exhaustive answers about how and why psychotherapy works. Evidently, psychotherapies are complex and dynamic processes with a multitude of interrelated influencing variables. Herein, therapeutic relationship formation and, consequently, the therapeutic alliance are of central importance to any psychotherapy – both from a clinical and a scientific perspective. The alliance is considered the most widely studied and robust common factor (Norcross & Lambert, 2018; Wampold & Imel, 2015). In addition, advocates of the contextual model also consider expectations an important common factor for successful psychotherapy (Mulder, Murray, & Rucklidge, 2017). At the same time, answers to the question of what works for whom? continue to be lacking – especially in regard to therapeutic relationship formation in the context of alliance and expectations. Addressing this question is an important goal of psychotherapy process research. In the long run, a more differentiated understanding of change mechanisms can lead to a continuous optimization of the treatment of each patient and overall improved effectiveness. Therefore, the overarching goal of this cumulative dissertation project was to shed more light on the so-called differential therapeutic relationship formation. We define differential relationship formation as a flexible and ongoing adaptation of therapeutic relationship behavior, based on an individual case conception and tailored to the needs and relationship behavior of patients (adapted from Sachse, 2000, based on Caspar & Grawe, 1982; Stucki, 2004). Different perspectives and various methodological approaches were used to pursue three specific objectives by means of four studies: (1) The current state of research on differential relationship formation was illustrated by means of an unsystematic literature review (study I). Among other aspects, the review revealed a lack of experimental studies regarding the question of which therapeutic style is helpful for which patient. (2) Therefore, two experimental designs were developed to test the targeted variation in the therapeutic interpersonal style in realistic settings in terms of experimental relationship design (studies II and III). The review also pointed to interpersonal characteristics of patients as an important component of relationship formation. The relationship should be shaped specifically by pursuing specific treatment goals, such as corrective healing relational experiences in patients with early traumatizing experiences. In this regard, the psychotherapy approach of the Cognitive Behavioral Analysis System of Psychotherapy (CBASP) should be highlighted, which was developed specifically for patients with persistent depressive disorder (PDD), who have oftentimes suffered early traumatizing relationship experiences in terms of childhood maltreatment and display current interpersonal conflicts (McCullough, 2003). (3) Therefore, we wanted to further examine psychotherapeutic relationship formation in the context of alliance as an interpersonal change agent in patients* with PDD receiving CBASP-treatment (study IV). In study I, differential relationship formation in Cognitive Behavioral Therapy (CBT) was theoretically and empirically embedded in an unsystematic literature review. It was discussed against the backdrop of (a) therapeutic phases, (b) needs, motives, and expectations, and (c) early traumatizing relationship experiences. Open questions were raised regarding an individualized and empirically supported therapeutic relationship formation. In the context of individualized case conceptualizations, the following gradations of relationship formation should be considered in psychotherapeutic practice and research: (a) non-specific (in terms of basic therapeutic variables), (b) indicated (against the backdrop of the patient’s individual problems), and (c) specific (pursuing particular treatment goals) relationship formation. Given the research gap described earlier, we investigated experimental relationship formation in study II (ExTheRel) and study III (InterAct). We examined the feasibility of a targeted variation of two different therapeutic styles (relationship-focused vs. problem-focused) on the conceptual basis of the interpersonal circumplex. The implementation proved successful in both studies under realistic, face- to-face conditions in a single counseling session amongst healthy subjects with interpersonal conflicts. The alliance was consistently rated high in study II (N = 64) and subjects reported significant symptom reduction, regardless of the style. In Study III (N = 80), the experimental design was extended to include expectations of therapeutic style. In a 2x2 design, participants received either the expected or opposite therapeutic style. Two weeks after the session, participants felt less impaired by the conflict and perceived high alliance and satisfaction with the session, regardless of therapeutic style, expectations, or their interaction. These two studies on experimental relationship formation provide initial evidence for the possibility of counselors to actively and deliberately vary their interpersonal therapeutic style, forming positive alliances, regardless of counseling style and/or expectations and violations hereof. Moreover, one-time psychological counseling sessions on interpersonal conflicts seem to be effective in healthy samples with respect to impairment and distress caused by these very conflicts. The innovative designs can serve as a basis for further process research in the field of psychological counseling and psychotherapy. The review (study I) further showed that the specific relationship formation is particularly important when interpersonal difficulties should be addressed in psychotherapy. In Study IV, we therefore aimed to investigate the interpersonal mechanism of action of the alliance in the context of specific psychotherapeutic relationship formation in a sample of patients with these very interpersonal difficulties. This project was embedded in a comprehensive study to investigate the effectiveness of an inpatient, individualized CBASP concept (CBASPersonalized) in patients with PDD. The aim of Study IV was to apply a recently proposed model of the alliance by Zilcha-Mano (2017) in light of the interpersonal characteristics of patients in the above sample. The model differentiates (a) a trait-like alliance – the general ability to form satisfying relationships and (b) a state-like alliance – changes in the direct interaction with the therapist. Path and mediation analyses demonstrated a good fit of the model and replicated the frequently demonstrated effect of the state-alliance (therapeutic alliance) on depressive symptomatology. In addition, positive changes in the state-like alliance led to a greater decrease in interpersonal problems (trait-alliance), which in turn had a positive effect on depressive symptomatology. The results can be taken as preliminary evidence for the change mechanism of corrective relational experiences proposed in CBASP-treatment, according to which a positive and sustainable therapeutic alliance improves interpersonal skills, which in turn lead to better treatment outcome. Overall, the study results support the significance of therapeutic relationship formation and alliance from a clinical and scientific perspective and provide first insights in terms of (a) differential relationship formation in clinical practice, (b) possibilities for experimental designs of relationship formation, and (c) underlying interpersonal mechanisms of change and action of psychotherapeutic relationship formation in patients with interpersonal difficulties. At the same time, some limitations should be considered, such as methodological shortcomings and rather small sample sizes. In the long run, however, the studies could serve as a basis for further research designs (including the identification of moderators and mediators) and the development of indication criteria for a personalized relationship formation.