Meta-Analysis: The Effectiveness of Universal School-Based Mental Health Programs for Children and Adolescents

Objective To estimate the effectiveness of school-based mental health programs on internalizing and externalizing symptoms and examine how these effects vary by program characteristics. Method MEDLINE, EMBASE, PsycINFO, ERIC, and Scopus were searched for studies on the effects of universal intervention or prevention programs for internalizing and externalizing symptoms. Programs were implemented during the school day with students as the primary recipients. Random-effects meta-analyses estimated percent difference in post-program mean symptoms, comparing treatment and control groups, adjusting for baseline differences and informant. Results A total of 21,679 non-duplicate records were retrieved, yielding 59 relevant studies (232 sample estimates, N=83,092, 53.8% female) across 22 countries. Results indicated marginal reductions for internalizing (standardized mean difference=–0.12, 95% CI=[–0.19,–0.06]) and externalizing symptoms (–0.08, [–0.14, –0.02]). Moderator analyses revealed stronger program effects for children with higher baseline internalizing symptoms (–0.17, [–0.26, –0.08]), but less so for externalizing symptoms (–0.10, [–0.20, –0.01]). Effects on internalizing symptoms were greatest for cognitive–behavioral therapy programs (–0.25, [–0.35, –0.15]). Results showed evidence of program effects when facilitated by mental health practitioners (–0.18, [–0.26, –0.10]). For externalizing symptoms, programs delivered in elementary (–0.10, [–0.18, –0.01]) and middle (–0.11, [–0.22, –0.002]) school were associated with marginal reductions, whereas effects for high school were inconclusive (0.03, [–0.18, 0.24]). Marginal reductions were observed in samples with higher socioeconomic status (SES) (–0.11, [–0.19, –0.03]), with slightly stronger effects in higher-SES elementary samples (–0.14, [–0.28, –0.01]). Conclusion School-based mental health programs demonstrate some effectiveness as universal interventions for internalizing and externalizing symptoms. Current programs may be more effective for children with elevated internalizing symptoms, suggesting potential benefits as targeted initiatives. Plain language summary The authors examined the effectiveness of universal mental health programs in schools. Results from 59 studies with 83,092 children showed programs were marginally effective. For internalizing symptoms, greater effects were found for children with higher baseline symptoms and when facilitated by mental health experts; however, these differences were marginal. For externalizing symptoms, some effects were found for younger children and higher socioeconomic status. Current universal approaches for school mental health programs should be reconsidered to support all children. Study registration information School-based mental health interventions: A meta-analysis of the characteristics and effectiveness of interventions across school-aged children; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024528751.