Interaction structures as predictors of outcome in a naturalistic study of psychodynamic child psychotherapy.

OBJECTIVE: The first aim of this study was to identify interaction structures (IS), which refer to clusters of items characterizing the course of psychotherapy in terms of reciprocal interaction patterns between the therapist and the child, secondly to assess their trends over the course of treatment, and finally investigate which IS predict outcome in long-term psychodynamic child psychotherapy.

METHOD: The sample included 52 children with externalizing and internalizing problems. 192 sessions were rated with the use of the Child Psychotherapy Q-Set (CPQ). Outcome was assessed with the Child Behavior Checklist (CBCL) Total Problems, and Children's Global Assessment Scale (CGAS).

RESULTS: A principal components analysis resulted in four IS, theoretically named Therapeutic Alliance, Children's Emotion Expression, Child -Centered Technique and Psychodynamic Technique. Multilevel Modeling (MLM) analyses indicated significant linear growth on Psychodynamic Technique. Multiple regression analyses indicated that Psychodynamic Technique positively predicted changes on CBCL Total Problems and CGAS. In contrast, Child-Centered technique negatively predicted change on CBCL Total Problems, however follow up interaction analyses showed that externalizing children who received more Child-Centered technique showed more improvement on this scale.

DISCUSSION: Findings point to empirically derived components of psychodynamic child psychotherapy and provide preliminary answers about which aspects may facilitate change. Clinical or methodological significance of this article: This is one of the first studies to identify core treatment processes in psychodynamic psychotherapy for children with externalizing and internalizing problems. We were able to identify empirically derived change processes that can be used in future research and provided preliminary answers about which aspects may facilitate change, taking into account specific pre-treatment sample characteristics, which provided further evidence that therapist techniques occur as part of a complex interaction with the children's characteristics. These findings highlight the importance of an investigative strategy to study "what works for whom", using a naturalistic process-outcome research design, in order to understand what specifically promotes therapeutic change.