Differential effects of adult attachment in cognitive-behavioural and psychodynamic therapy in social anxiety disorder: A comparison between a self-rating and an observer rating.

Different measures of attachment are usually weakly correlated. In a subsample of an RCT comparing short-term cognitive-behavioural therapy (CBT) and psychodynamic therapy (PDT), we examined the association between attachment and outcome using two attachment measures. The sample comprises 148 patients with social anxiety disorder who were treated in the SOPHO-NET trial. Pretreatment attachment was assessed using the Adult Attachment Prototype Rating (AAPR) and the Bielefeld Questionnaire of Client Expectations (BQCE). Regression models were used to predict the therapeutic alliance (HAQ) at session 8, the Liebowitz Social Anxiety Scale (LSAS) at the end of therapy and a 6-month follow-up. Attachment groups (secure, avoidant, and ambivalent) classified with the AAPR and the BQCE were not significantly correlated (Cohen's κ = 0.08). Only the BQCE was associated with the HAQ indicating avoidantly attached patients showing lower HAQ scores than securely attached (Cohen's d = 0.722). Regarding the AAPR, we found an interaction effect of treatment and attachment related to the post-treatment LSAS scores. Post hoc tests revealed that securely attached patients in CBT had lower scores than securely attached in PDT (d = 0.922) and, on a trend level, avoidantly attached in CBT (d = 0.782). We conclude that attachment-outcome associations are affected by the applied measure. The identified differential effects suggest that psychotherapists should adapt the interventions on the attachment of their patients.