Divergent Global Trajectories in Adolescent Depression Burden: Childhood Maltreatment Attributable Disability, Socio-Behavioral Gradients, and Symptom Network Topology

Background Childhood maltreatment imposes profound disability burdens through major depressive disorder (MDD), yet global trends, sociodemographic determinants, and symptom-level mechanisms remain inadequately quantified. Methods Integrated analyses leveraged Global Burden of Disease (GBD) 2017–2021 data (204 countries; n = 23,487 sources) and US National Health and Nutrition Examination Survey (NHANES) cycles (n = 900 adults). GBD-estimated disability-adjusted life years (DALYs) and years lived with disability (YLDs) attributable to childhood sexual abuse/bullying victimization employed DisMod-MR 2.1, CODEm, and geospatial frameworks. Longitudinal trends used linear mixed-effects models with Monte Carlo uncertainty propagation. NHANES analyses deployed Gaussian graphical models (GGMs), restricted cubic splines (RCS), and ensemble machine learning (XGBoost/Random Forest) to delineate socio-behavioral correlates, nonlinear exposure-response relationships, and symptom network architecture. Results Adolescents in low Socio-demographic Index (SDI) regions bore the highest sexual abuse-attributable DALY burden (32,090; 95% UI:29,266–35,120) with 4.38%/year growth, while high-SDI regions exhibited rising abuse burden (+ 1.41%/year) alongside declining bullying-attributable disability (− 0.62%/year). Geospatial analysis revealed Egypt (204.48 DALYs/100,000), the US (249.11), and Greenland (395.80) as critical hotspots. US state-level analyses demonstrated alarming divergences: sexual abuse-attributable DALYs increased universally (+ 1.14%/year), eclipsing bullying reductions (− 0.89%/year), with females sustaining 6.3-fold higher abuse-related disability rates by 2021. NHANES stratification identified severe depression concentrated in younger, economically disadvantaged adults (income-to-poverty ratio: 0.73 vs. 3.55; *p < 0.001). Poverty-sedentary behavior interactions synergistically increased severe depression risk (OR = 1.005; 95%CI:1.001–1.009; *p = 0.013). Symptom networks identified depressed mood (strength = 0.92) and anhedonia as central nodes, with suicidal ideation bridging affective and cognitive clusters. Machine learning confirmed PHQ-9 severity as the dominant risk predictor (|SHAP|=0.42), outperforming socio-behavioral factors. RCS models revealed J-shaped sedentary behavior-depression relationships, steepening below poverty thresholds (OR = 1.52; 95%CI:1.38–1.68). Conclusion Childhood sexual abuse drives escalating global depression disability—unmitigated by current public health interventions—with distinct socio-behavioral vulnerability pathways. Symptom network topology and nonlinear exposure-response dynamics identify critical targets for precision prevention. Urgent recalibration of child protection policies is warranted to address this diverging burden epidemic.