Sex-disaggregated pooled serious injury prevalence among school-going adolescents aged 13–17 years across 44 Global School-based Student Health Survey datasets, 2015–2023: a meta-analysis

Background Serious injuries are an important cause of morbidity among adolescents globally, particularly in low- and middle-income countries (LMICs). The Global School-based Student Health Survey (GSHS) provides standardized and comparable data across LMICs and, in some cases, high-income countries. Recent multi-country estimates of serious injuries among school-going adolescents are limited. The objectives of this study were to estimate the pooled prevalence of self-reported serious injuries among school-going adolescents across the included GSHS surveys and to analyze variations by geographic region and survey period. Methods A meta-analysis was conducted using the World Health Organization (WHO) GSHS data from 44 national surveys conducted during 2015–2023, for which data were available as of 12 March 2026. The outcome studied was self-reported, GSHS-defined serious injury sustained one or more times during the previous 12 months. Design-based sex-specific prevalence estimates, standard errors, and their 95% confidence intervals were computed. Pooled prevalence estimates were computed using random-effects meta-analysis with logit transformation and restricted maximum likelihood (REML) estimation, and the estimates were back-transformed into percentages. Heterogeneity was assessed, and sensitivity, subgroup, and leave-one-out analyses were conducted. Results Across the 44 included national GSHS surveys, the pooled prevalence of serious injury among male individuals was 50.2% (95% CI: 46.7–53.7%), and the 95% prediction interval (PI) was 28.3–72.0%. Among female individuals, the pooled prevalence was 37.7% (95% CI: 33.9–41.6%), with the 95% prediction interval being 16.7–64.7%. There was high heterogeneity for both sexes (male, I 2 = 96.5%; female, I 2 = 97.0%), indicating substantial between-survey variability. Geographic region and survey period were not statistically significant moderators for either sex. For male individuals, the pooled prevalence was highest in the Pacific and Africa and lowest in Latin America. For female individuals, the pooled prevalence was also highest in the Pacific and Africa and lowest in Latin America. Leave-one-out analysis showed that no single country dominated the pooled estimates. Sensitivity analysis produced materially unchanged findings. Conclusion Across the included national GSHS surveys, serious injury was common among school-going adolescents aged 13–17 years, with a numerically higher pooled prevalence among male individuals. The findings support the need for context- and sex-specific injury prevention strategies in schools, including, where relevant, road safety measures, violence and bullying prevention, and stronger adolescent injury surveillance.