Post-discharge suicide among high-risk psychiatric inpatients: Risk factors and warnings signs

Abstract

The present study explored the predictive value of a novel suicide risk assessment with regard to suicide deaths within three years of discharge from an acute inpatient unit. The contributions of the different components of the assessment to overall determination of suicide risk were also explored. A total of 380 acute psychiatric inpatients were analysed in a 3- year prospective study. At admission, the physician on duty performed a suicide risk assessment based on national guidelines and clinical state variables. A dichotomous variable reflecting high/low suicide risk level was determined by the clinician and recorded in the chart. Information on the number and causes of deaths was obtained from the Norwegian Cause of Death Registry. Eleven (2.9%) patients died by suicide within three years after discharge, eight high-risk patients and three low risk patients (OR = 8.7, 95% CI: 2.2–33, p = 0.002). In multivariable analysis, recent suicidal ideation (SI) (OR = 3.6, 95% CI: 1.5–8.7, p = 0.004), and affective disturbance (OR = 4.3, 95% CI: 1.5–12.6, p = 0.008) were significantly associated with rating of high suicide risk. The sample size was rather small, which in conjunction with the low base rate of suicide, reduces power and limits generalizability of the findings. We found that a novel suicide risk assessment at admission to an acute psychiatric service was a powerful predictor of suicide post-discharge. The components of the risk assessment that were statistically significant with high suicide risk were affective disturbance and recent SI.