Efficacy, time-course, and adverse events of antidepressants in pediatric social anxiety disorder: a systematic review and meta-analysis of randomized controlled trials

Antidepressants are established as effective treatments for social anxiety disorder (SAD) in adults and for anxiety disorders in children and adolescents. However, evidence specific to pediatric SAD remains limited, and disorder-specific effects, the time course of social anxiety (SA) symptom improvement, and adverse event (AE) profiles have not been systematically evaluated. We conducted a meta-analysis of randomized controlled trials (RCTs) to investigate the efficacy, time course, and AEs of antidepressants in children and adolescents with SAD. A systematic search of PubMed and CENTRAL identified RCTs comparing antidepressants with placebo in participants younger than 18 years with SAD. Between-group differences in treatment response at the final assessment point were evaluated using the Clinical Global Impressions-Improvement (CGI-I ≤ 2). Continuous outcomes of SA symptoms were extracted at each time point and analyzed using standardized mean differences to assess changes over time. Time-course effects were examined to characterize the trajectory of symptom improvement using mixed-effects meta-regression and linear mixed-effects models. Between-group differences in AEs were evaluated using pooled risk differences (RDs). Three RCTs including 669 participants were included. Antidepressants were associated with a higher likelihood of response compared with placebo (odds ratio = 4.05, p = 7.52 × 10 − 4 ). Longitudinal analyses demonstrated that improvements in SA symptoms increased over time ( p = 1.15 × 10 − 5 ), with effect sizes reaching approximately − 0.7 at later time points, following a nonlinear trajectory with diminishing gains after approximately 8–12 weeks. Antidepressants were associated with increased risks of several AEs, including nausea/vomiting (RD = 0.071; 7.1%, p = 0.025), asthenia (RD = 0.086; 8.6%, p = 1.37 × 10 − 3 ), and suicidal ideation (RD = 0.023; 2.3%, p = 7.59 × 10 − 3 ), whereas no increase was observed in other assessed AEs, including self-harm behavior or suicide attempts. These findings suggest the potential efficacy of antidepressants in pediatric SAD, with treatment gains slowing after 8–12 weeks, while the available data did not indicate an increase in self-harm behavior or suicide attempts.