The Working Alliance in Manualized CBT for Generalized Anxiety Disorder: Does It Lead to Change and Does the Effect Vary Depending on Manual Implementation Flexibility?

Objective: The investigation of session-to-session effects of working alliance on symptoms and coping experiences in patients diagnosed with generalized anxiety disorder. In addition, investigating these effects dependent on whether therapists are primed to work with patients strength (resource priming) or to adhere to the treatment manual (adherence priming). Method: Data was drawn from a randomized controlled trial in which 57 patients were randomly assigned to either the resource priming condition or the adherence priming condition. Within- and between patient associations were disentangled using dynamic structural equation modeling. Results: The total score of the working alliance, as well as all its overlapping components (i.e., goal agreement, task consensus, bond) showed significant within-patient effects on next session coping experiences. More specifically, better alliance scores in one session were followed by more coping experiences in the subsequent session. With regard to anxiety symptoms, an association was found only with the working alliance total score as well as for the bonds component, but not for the goals and task components of the working alliance. The priming condition (resource priming vs. adherence priming) had no influence on the within-patient alliance–outcome association. Between-patient alliance associations were only present with coping experiences, but not with anxiety symptoms. Conclusion: The findings provide further empirical evidence for the hypothesis that the working alliance may be a robust facilitative factor for change in CBT treatments for generalized anxiety disorder, which evolves irrespective of the strictness with which therapists adhere to the treatment manual. The present study shows that for a homogeneous sample of patients suffering from generalized anxiety disorder, fluctuations in patients' ratings of the working alliance predict anxiety symptoms and coping experiences in the subsequent session. This alliance–outcome association was more pronounced for patients with higher average alliance scores over the course of the treatment but was not influenced by the strictness with which therapists adhered to the treatment manual. [ABSTRACT FROM AUTHOR]