Dyadic predictors of outcome in a cognitive-behavioral program for patients with generalized anxiety disorder in committed relationships: a "spoonful of sugar" and a dose of non-hostile criticism may help.

The present study tested whether pre-treatment levels of partner hostility and non-hostile criticism predicted outcome in an individual cognitive-behavioral therapy package for generalized anxiety disorder (GAD). Eighteen patients with a principal or co-principal diagnosis of GAD were randomly allocated to a treatment condition (n=8) or a delayed treatment condition (n=10). In addition, the patients and their partners were videotaped discussing the patients' worries. These videotapes were later coded for levels of partner hostility and non-hostile criticism directed at the patients. Treatment resulted in statistically and clinically significant change at post-test. Finally, partner hostility predicted worse end-state functioning whereas partner non-hostile criticism predicted better end-state functioning.