RESUMO
BACKGROUND: Many late adolescents experience a state of psychological sub-health, requiring early recognition and intervention. This study aims to assess the psychological state of late Chinese adolescents and uncover developmental trend of mental health through network analysis. METHOD: We analyzed data from 9072 Chinese high school adolescents in Shandong Province surveyed in 2020-2021, and divided them into the normal, the suspected, and the abnormal groups based on Symptom Checklist 90 (SCL-90) scores. Network analysis was employed to identify the core symptoms and bridge symptoms across different states. RESULTS: Anxiety and depression were the most central symptoms, without gender differences. Core symptoms, network structure, and network invulnerability varied across different psychological states. The abnormal group exhibited the highest value of natural connectivity, followed by the suspected and normal groups. This pattern extended to bridge networks. While not meeting diagnostic criteria, the suspected group demonstrated abnormalities in network edge invariance and global strength invariance. LIMITATIONS: The cross-sectional design cannot establish causality, and biases in self-report measurements cannot be ignored. CONCLUSION: Compared to traditional scale indicators, network structural characteristics may be a more sensitive assessment indicator.