BACKGROUND: Pediatric anxiety disorders are highly prevalent and associated with significant functional disabilities and lifelong morbidity. Cognitive-behavioral therapy (CBT), sertraline, and their combination are effective treatments, but little is known about how these treatments exert their effects.

METHODS: Using network intervention analysis (NIA), we analyzed data from the largest randomized controlled treatment trial of pediatric anxiety disorders (Child/Adolescent Anxiety Multimodal Study, NCT00052078, clinicaltrials.gov/ct2/show/NCT00052078) and outlined the causal symptom domain-specific effects of CBT, sertraline, and their combination over the course of the 12-week treatment while taking into account both specificity and overlap between symptom domains.

RESULTS: All active treatments produced positive effects with the most pronounced and consistent effects emerging in relation to psychological distress, family interference, and avoidance. Psychological distress was consistently the most and physical symptoms the least central symptom domain in the disorder network.

CONCLUSIONS: All active treatments showed beneficial effects when compared to placebo, and NIA identified that these effects were exerted similarly across treatments and primarily through a reduction of psychological distress, family interference, and avoidance. CBT and sertraline may have differential mechanisms of action in relation to psychological distress. Given the lack of causal effects on interference outside family and physical symptoms, interventions tailored to target these domains may aid in the building of more effective treatments. Psychological distress and avoidance should remain key treatment focuses because of their central roles in the disorder network. The findings inform and promote developing more effective interventions.