Abstract
BACKGROUND: Alcohol and substance use disorders are important predictors for suicidal behavior. However, the role of individual substances as proximal risk factors for suicidal behavior and the mechanisms through which substance use affect risk are not entirely clear. We examine whether the frequency of substance use and whether biological markers in the HPA axis and inflammatory pathways are associated with clinical risk factors of suicidal behavior of aggression, impulsivity, hopelessness, and poor sleep. METHODS: The sample consisted of psychiatric inpatients, aged 15-30 years, admitted for suicide attempt (n = 38), suicidal ideation (n = 40); and healthy controls (n = 37). We measured hair cortisol concentrations, glucocorticoid receptor (GR) sensitivity, stimulated production of interleukin- or IL-6, C-reactive protein, and mRNA expression of GR, SKA2, FKBP5, TNF-α, and IL-1β. RESULTS: Smoking was associated with increased aggression [β = 2.9, 95% CI (-0.03, 6), p = 0.05], impulsivity [β = 3.1, 95% CI (1.6, 4.6), p < 0.001], and poor sleep [β = 0.5, 95% CI (0.03, 0.95), p = 0.04] even after controlling for demographics and group. Similarly, TNF-α mRNA was associated with impulsivity [β = 0.07, 95% CI (0.01, 0.1), p = 0.02] and hopelessness [β = 0.03, 95% CI (0.004, 0.05), p = 0.03]. Smoking tobacco (r = 0.32, p < 0.001) was positively associated with TNF-α mRNA. LIMITATIONS: Study limitations include the cross-sectional design, retrospective assessment, and relatively small sample size. CONCLUSIONS: Future longitudinal studies are needed to test whether inflammatory markers mediate the relationships between smoking, clinical risk factors, and suicidal behavior; and to examine whether smoking cessation could reduce the risk for suicidal behavior in at-risk patients.