Social anxiety is a common mental disorder among adolescents and is associated with detrimental long term outcomes. Therefore, this study investigated the efficacy of two possible early interventions for adolescent social anxiety and test anxiety. An internet-based cognitive bias modification (CBM; n = 86) was compared to a school-based cognitive behavioral group training (CBT; n = 84) and a control group (n = 70) in reducing symptoms of social and test anxiety in high socially and/or test anxious adolescents aged 13-15 years. Participants (n = 240) were randomized at school level over the three conditions. CBM consisted of a 20-session at home internet-delivered training; CBT was a 10-session at school group training with homework assignments; the control group received no training. Participants were assessed before and after the intervention and at 6 and 12 month follow-up. At 6 month follow-up CBT resulted in lower social anxiety than the control condition, while for CBM, this effect was only trend-significant. At 12 month follow-up this initial benefit was no longer present. Test anxiety decreased more in the CBT condition relative to the control condition in both short and long term. Interestingly, in the long term, participants in the CBM condition improved more with regard to automatic threat-related associations than both other conditions. The results indicate that the interventions resulted in a faster decline of social anxiety symptoms, whereas the eventual end point of social anxiety was not affected. Test anxiety was influenced in the long term by the CBT intervention, and CBM lead to increased positive automatic threat-related associations.
Trial registration: TrialRegister.nl NTR965.