Time series analysis of therapy process in groups of clients with personality disorders

Abstract

During recent years the treatment of personality disorders has increasingly come into focus. As the psychotherapeutic interventions are still limited with respect to these demanding disorders there is a considerable need for further efficacy and, particularly, therapy process investigations. A promising approach is the development of integrated psychotherapy which combines cognitive-behavioral interventions with further change mechanisms such as the clarification of maladaptive schemata. On top of that, interventions should maintain a focus on emotional aspects of the therapeutic alliance (emotion regulation). In a study on “Schema-focused Emotive Behavioral Therapy” (SET), 93 patients with personality disorders of clusters B (predominantly narcissistic and Borderline PS) and C (avoidant and dependent PS) were randomised into one arm with SET group therapy and a control arm with manualised social skills training (SST). Therapies lasted for approximately 30 two-hour sessions.

Therapy process was closely monitored using therapy session reports both from the patients' and therapists' perspectives. Therapy outcome was assessed prior to and after therapy.

We report here on the change mechanisms, which were derived from modeling therapy session reports with novel time series methods (vector autoregression based on the estimation of session-to-session changes). It was found that the two therapy approaches differed with respect to change mechanisms. In SET (yet not in SST) therapy groups, clarification and insight reduced feelings of being rejected and disrespected, which was a major concern of many patients. In addition to this, a contrast was found between the prototypical therapy processes of cluster B and C patients. In conclusion, these results lay the ground for a disorder-specific application of integrated psychotherapy in personality disorders.