Psychiatric and psychosocial determinants of suicidality: Findings from a national mental health survey in Korea

Abstract

Background: Various psychiatric disorders are linked to suicidal behaviors, but evidence from nationally representative populations is limited. Few studies have identified factors distinguishing individuals with suicidal ideation from those who attempt suicide. Using national mental health survey data, this study examines the associations between multiple mental disorders and suicidality and identifies key clinical and psychosocial correlates of suicidal acts among ideators. Methods: Data were drawn from 5511 adults (representing 42.0 million nationwide) in the National Mental Health Survey of Korea 2021, a multi-stage stratified survey with sampling weights applied to all analyses for national representativeness. Psychiatric disorders included major depressive disorder (MDD), dysthymic disorder, anxiety disorders, obsessive-compulsive disorder (OCD), posttraumatic stress disorder (PTSD), and other common conditions. Psychosocial factors included resilience, social support, self-harm, social isolation, loneliness, and quality of life. Multiple logistic regression estimated odds ratios (ORs) for suicidality and examined factors distinguishing suicidal ideation from attempts, adjusting for demographic and health-related covariates. Results: In the weighted sample, 4.53 million adults (12.02 %) exhibited suicidality, including suicidal ideation, planning, or attempts. MDD (OR = 3.25, 95 % CI: 1.78-5.91), nicotine dependence (OR = 5.51, 95 % CI: 2.19-13.91), self-harm (OR = 24.24, 95 % CI: 8.09-72.63), and social isolation (OR = 4.44, 95 % CI: 2.02-9.75) showed the strongest associations with suicidal behaviors, distinguishing individuals with suicidal ideation from those with planning or attempts. Conclusion: MDD, nicotine dependence, social isolation, and self-harm were identified as the strongest factors distinguishing suicidal ideation from suicide attempts, although the cross-sectional design limits causal inference.