Background:: This study aimed to compare the predictors of suicides among psychiatric inpatients and recently discharged patients and to examine the association between the length of stay and suicides. Methods:: Data from psychiatric inpatients were extracted from the National Health Insurance databank and merged with information from the Cause of Death data using unique identification numbers. Poisson regression analyses were used to estimate the incidence rate ratio for inpatient and post-discharge suicides, which included the variables of sex, age, psychiatric diagnosis, and number of admissions in the preceding year. The associations between length of stay and inpatient and post-discharge suicide were examined using multivariate Poisson regression analyses that were adjusted for these variables. Results:: A diagnosis of affective disorders and a higher number of previous admissions increased both inpatient and post-discharge suicides. Patients older than 15–24 years had a significantly lower inpatient suicide risk but were more likely to die by suicide post-discharge. The risk of suicide both during the inpatient stay and post-discharge statistically significantly decreased when the length of stay was longer. Limitations:: As there were variations in societal, environmental, and facility-level factors that might have influenced the association between length of stay and suicide, the generalization of our findings to different settings may be impeded. Conclusions:: This study provides additional evidence that hospitalization helps to reduce the suicide risk. It suggests that medical professionals should be alert to inpatient suicide and that they should determine the optimal length of stay considering post-discharge suicide.
ASJC Scopus subject areas