Associations of a perceived parental-rejection experience with self-harm or death thoughts in adolescents: evidence from symptom network analysis and structural equation modeling

Abstract

Background Childhood trauma (CT) is associated with depression, anxiety, and suicide-related thoughts in adolescents. However, the item-level associations linking trauma-related experiences with internalizing symptoms and thoughts of self-harm or death remain unclear. Methods We conducted a cross-sectional analysis of 2,413 adolescents who completed the Childhood Trauma Questionnaire-Short Form (CTQ-SF), Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Disorder-7 (GAD-7). Symptom-level associations were examined using a Spearman-EBICglasso network, with a polychoric network as a sensitivity analysis. CTQ-SF item 8 was interpreted descriptively as a perceived parental-rejection experience. PHQ-9 items 1–8 indicated latent depressive symptoms, GAD-7 items 1–7 indicated latent generalized anxiety symptoms, and PHQ-9 item 9 represented thoughts of being better off dead or self-harm. Four cross-sectional structural equation model (SEM) specifications were compared using maximum likelihood (ML), with MLR and WLSMV sensitivity analyses. Results Childhood trauma, depressive symptoms, and generalized anxiety symptoms formed distinct but interconnected network communities. CTQ-SF item 8 ranked seventh in bridge strength in the primary network and remained among the leading bridge items in the polychoric network. The correlated-factors SEM showed the best primary fit. Under ML, item 8 was associated with depressive symptoms (standardized β = 0.439, p < 0.001), generalized anxiety symptoms (standardized β = 0.377, p < 0.001), and item 9 directly (standardized β = 0.077, p = 0.012). The depressive-symptom association with item 9 was larger (standardized β = 0.562, p < 0.001) than the generalized-anxiety association (standardized β = 0.111, p = 0.073). MLR reproduced this pattern, whereas WLSMV identified a smaller but statistically significant generalized-anxiety association (standardized β = 0.166, p = 0.001). Conclusions A perceived parental-rejection experience emerged as a salient bridge item linking childhood trauma with internalizing symptoms and thoughts of self-harm or death. Depressive symptoms showed the larger and more consistent conditional association with item 9, whereas the generalized-anxiety association was smaller and varied by estimator. These findings support further investigation of family relational experiences in adolescent mental-health assessment and prevention research.