A network analysis of depressive and internet addiction symptoms among bullying-victimized adolescents: a comparison between left-behind and non-left-behind children

Objective School bullying is an important risk factor for mental health problems in adolescents. Depressive symptoms and internet addiction often co-occur among adolescents who have experienced bullying; however, the associations among specific symptoms remain unclear. This study aimed to investigate the comorbidity network characteristics of depressive symptoms and internet addiction among bullying-victimized adolescents, and to compare differences in network characteristics between left-behind and non-left-behind adolescents. Methods This study employed a cross-sectional design and included 15,984 bullying-victimized adolescents selected through stratified random sampling. Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D), and internet addiction was measured using the 9-item Internet Addiction Scale. A mixed graphical model (MGM) was used to construct the symptom network of depressive symptoms and internet addiction. Node centrality and bridge centrality indices were calculated, and network stability was assessed using bootstrap methods. The Network Comparison Test was conducted to compare differences in network structure and global strength between left-behind and non-left-behind adolescents. Results Among bullying-victimized adolescents, the overall predictability of network nodes was 0.27. Risky persistence and Tolerance showed the highest strength centrality in the symptom network. The principal bridge symptoms were Risky persistence, Escape coping, and Depressive affect. The network model demonstrated good stability. No significant differences were observed between left-behind and non-left-behind adolescents in terms of either the overall network structure or global strength (all P > 0.05). Conclusion Compared with interventions targeting only inter-symptom relationships, exposure to school bullying itself may be a more critical and primary intervention target. Left-behind and non-left-behind children exhibited highly similar symptom network patterns; therefore, intervention strategies should not rely solely on identity-based distinctions, but rather focus on bullying exposure and symptom severity to implement more precise interventions. These findings provide symptom-level empirical evidence for developing integrated intervention strategies for adolescents exposed to school bullying.