The reciprocal relationship between alliance and symptom improvement across the treatment of childhood anxiety.

This study examined changes in the therapeutic alliance and in self-reported anxiety over the course of 16 weeks of manual-based family treatment for child anxiety disorders. Eighty-six children (51.3% female; aged 7.15-14.44; 86.2% Caucasian, 14.8% minority) with a principal diagnosis of separation anxiety disorder, generalized anxiety disorder, and/or social phobia, and their parents, received family treatment for anxiety disorders in youth. Child, therapist, and parent ratings of therapeutic alliance and child ratings of state anxiety were measured each session. Latent difference score growth modeling investigated the interacting relationship. Therapeutic alliance change, as rated by the mother and by the therapist, was a significant predictor (medium effect) of latter change in child anxiety (with greater therapeutic alliance leading to later reduction in anxiety). However, changes in child-reported anxiety also predicted latter change in father- and therapist-reported alliance (small-to-medium effect). Prospective relationships between child-reported therapeutic alliance and child-reported symptom improvement were not significant. Results provide partial support for a reciprocal model in which therapeutic alliance improves outcome, and anxiety reduction improves therapeutic alliance.