Associations of serum cortisol levels, stress perception, and stressful events with suicidal behaviors in patients with depressive disorders.

Abstract Background Stress is a well-known risk factor of suicide. However, there is a lack of research on the best psychobiological markers of stress for monitoring and predicting suicidal behaviors in depressed patients. We aimed to investigate the individual and interactive effects of serum cortisol levels and stress-related parameters, including perceived stress and life stressors, on suicidal behavior in depressed patients. Methods At baseline, serum cortisol levels were measured and stress-related data, i.e., Perceived Stress Scale (PSS) scores and the number of stressful life events (SLEs), were collected from 1,094 patients with a depressive disorder, of whom 884 were followed up for 1 year. Suicidal behaviors evaluated at baseline included previous suicide attempts and baseline suicidal severity, while those at the 1-year follow-up were increased suicidal severity and fatal/non-fatal suicide attempts. Individual and interactive associations of exposure variables with suicidal outcomes were analyzed using logistic regression models after adjusting for relevant covariates. Results Higher serum cortisol levels were significantly associated with suicidal outcomes at the 1-year follow-up. In addition, higher PSS scores were significantly associated with higher baseline severity, whereas the number of SLEs was not associated with any suicidal outcomes. A significant interactive effect of cortisol level and PSS score on suicidal outcomes at the 1-year follow-up was observed. Higher cortisol levels and PSS scores were associated with increased suicidal severity and more fatal/non-fatal suicide attempts. Conclusions The combination of serum cortisol levels and PSS scores may be useful for monitoring and predicting suicidal behaviors in depressed patients.