Abstract/Summary

A large proportion of emotional problems begin in adolescence and negatively impact quality of life into adulthood. There have been multiple teacher-delivered, classroom-based programs created to reduce symptoms of internalizing problems amongst young people. This meta-analysis and systematic review aims to examine the effectiveness of teacher-delivered interventions for depression, anxiety, post-traumatic stress disorder (PTSD) and obsessive-compulsive disorder (OCD) symptoms in adolescents, and a range of factors that may impact outcomes. Database searches were conducted from PsycInfo, Medline (PubMed), Scopus, the Cochrane Library and the British Educational Index (from database inception to January 2020). Quality assessment of studies was conducted using the EPHPP Quality Assessment Tool. Fifty-two intervention studies were identified that quantitatively assessed, via controlled design, intervention effects on internalizing disorder symptoms. Two meta-analyses found teacher-delivered interventions were significantly better than control conditions at improving depression (g = -0.12), anxiety (g = -0.13) and PTSD symptoms (g = -0.66) in students. Improvements were only maintained at follow-up for anxiety symptoms and no effect sizes reached a ‘small’ threshold. However, the effect sizes were ‘moderate’ within the context of universal prevention programs for young people. No interventions measured OCD outcomes. Overall, the findings suggest that teachers may not be the optimal deliverers of mental health interventions. Improved outcomes were associated with interventions that lasted up to 16 weeks, had program sessions of 45-90 minute duration, and included two or more days of training for teachers. Future studies should aim to improve reporting quality on number of sessions, teacher training and fidelity of intervention. Increased reporting of outcomes from adolescents with high versus low baseline mental health scores would enable a better understanding of for whom interventions are most effective.