Childhood trauma and suicide ideation and attempts: Examining the mediating role of psychosocial, personality, lifestyle and biological factors.
BACKGROUND Childhood trauma (CT) is robustly associated with suicide ideation (SI) and attempts (SA) in individuals with depressive and/or anxiety disorders. However, pathways underlying these associations and whether they differ between suicide ideation only (SI + SA-) and suicide ideation with attempt (SI + SA+) remain unclear. Variables that may statistically account for CT-SI + SA- and CT-SI + SA+ associations were examined. METHODS Data from 1572 respondents with a 12-month depressive and/or anxiety disorder from the Netherlands Study of Depression and Anxiety were used. The NEMESIS childhood trauma interview (0-8 score) measured CT severity. Single and multiple-mediator models via generalized structural equation modeling (GSEM) were used to investigate whether psychosocial, personality, lifestyle and biological variables showed indirect associations consistent with mediation of the association between CT and suicide outcomes. RESULTS Higher CT scores were associated with SI + SA- (OR = 1.51, 95% CI: 1.32-1.73) and SI + SA+ (OR = 2.68, 95% CI: 2.21-3.27). In single mediator models, personality traits, insomnia, low social support and loneliness statistically accounted for a part of the association between CT and both outcomes. Interleukin-6 statistically accounted for a small portion of the CT-SI + SA+ link. In multiple-mediator models, direct effects reduced by 65.6% (SI + SA-) and 39.1% (SI + SA+) and aggression, insomnia, social support (both outcomes), neuroticism, introversion, hopelessness (SI + SA-) and locus of control (SI + SA+) remained significant. LIMITATIONS The cross-sectional design limits causal inferences. CONCLUSION Multiple personality traits, insomnia and lower social support statistically accounted for a portion of the association between CT and suicide outcomes. Especially aggression, insomnia and social support may represent actionable suicide prevention targets, pending longitudinal validation.