Context: Treatments for obsessive-compulsive disorder (OCD) in children and adolescents.
Objective: Evaluate the comparative efficacy of behavioral and pharmacologic treatments.
Data sources: Six databases and ClinicalTrials.gov registry; search last updated on 5/15/2024.
Study selection: Dual screening augmented by Abstrackr machine learning algorithm.
Data extraction/analysis: Participant characteristics, intervention details and risk of bias.
Results: 71 randomized controlled trials (RCTs). In the random effects network meta-analysis of OCD symptom severity, assessed by the Children's Yale-Brown Obsessive Compulsive Scale Total (CY-BOCS), exposure and response prevention therapy (ERP) is more effective than waitlist (net mean difference [NMD] -10.5; 95% confidence interval [CI]: -12.6, -8.4) and probably more effective vs. behavioral control (NMD -5.3; 95% CI: -8.0, -2.7). Remote ERP is more effective than waitlist (NMD -9.4; 95% CI: -11.9, -7.0) and as effective as in-person ERP. Selective serotonin reuptake inhibitors (SSRIs) are more effective than placebo (NMD -4.4; 95% CI: -6.1, -2.6). Clomipramine is probably more effective than placebo (NMD -4.5; 95% CI: -6.8, -2.1). ERP is probably more effective than SSRI (NMD -2.7; 95% CI -5.4, -0.0) and combined ERP and an SSRI are probably more effective than SSRI alone (NMD -3.0; 95% CI: -5.1, -1.0). Overall, treatments including ERP (ERP+SSRI, ERP and remote ERP) comprise the 3 highest ranked interventions.
Limitations: Non CY-BOCS outcomes were sparsely reported.
Conclusions: ERP, delivered in-person or via telehealth, SSRIs and clomipramine are all effective treatments. ERP, alone or in combination with an SSRI, is probably more effective than SSRI alone.