Abstract
OBJECTIVE: Test changes in perceived coping efficacy, negative self-statements, and interpretive biases to threat during treatment as potential mediators of the relationship between randomly assigned treatment conditions and long-term anxiety follow-ups. Age at randomization was also tested as a moderator of mediational relationships. METHOD: Participants included 319 youth (ages 7-17) from the Child/Adolescent Multimodal Study (CAMS) who participated in a naturalistic follow-up beginning an average of 6.5 years after the end of the CAMS intervention. The intervention conditions included cognitive behavioral therapy (CBT; Coping Cat), pharmacotherapy (sertraline), combined CBT and sertraline, and pill placebo. Putative mediators were measured four times during the intervention phase. Follow-up consisted of four annual assessments of current anxiety. RESULTS: Reductions on a measure of interpretive bias to threat over the course of the combined condition intervention, as compared to the placebo condition, mediated anxiety outcomes at the first follow-up visit. This mediated effect was not significant for the CBT-only or sertraline-only conditions when compared to the placebo condition. No other significant mediated effects were found for putative mediators. Age did not significantly moderate any mediated effects. CONCLUSION: Changes in youth-reported interpretive biases to threat over the course of combined youth anxiety interventions, as compared to a placebo intervention, may be associated with lower anxiety an average of 6.5 years following treatment.