Long-term outcomes and predictors of internet-delivered cognitive behavioral therapy for childhood anxiety disorders

•Twelve-month outcomes of internet-delivered CBT were investigated.•Baseline data were used to explore predictors of outcome at three-month follow-up.•Treatment gains were maintained at twelve-month follow-up.•Suspected autism spectrum disorder was associated with poorer treatment outcome.•Higher CSR baseline scores were associated with greater improvement.

ObjectiveThis study investigated the long-term outcomes of internet-delivered cognitive behavior therapy (ICBT) for children with anxiety disorders, and potential pre-treatment predictors of treatment outcome.MethodThe sample included eighty-four children (8-12 years old) with anxiety disorders, from both a treatment group and a waitlist control (after participants had crossed over to treatment) of a previous randomized controlled study. Participants were assessed at post-treatment and three- and twelve-months after treatment using a semi-structured interview and parent ratings. Pre-treatment data were used to investigate predictors of treatment outcome at three-month follow-up.ResultsIntention-to-treat analysis showed that treatment gains were maintained at twelve-month follow-up, including clinician rated severity of the principal anxiety disorder, parent rated anxiety symptoms and global functioning, with mainly large effect sizes (Cohen's d = 0.63-2.35). Completer analyses showed that suspected autism spectrum disorder was associated with less change in symptom severity. No other pre-treatment measures significantly predicted treatment outcome.ConclusionThis study suggests that internet-delivered CBT can have long-term beneficial effects for children with anxiety disorders. Predictors of treatment outcome need to be evaluated further.Trial registrationClinicaltrials.gov; NCT01533402.