Beyond Suicidal Ideation: Identifying High-Risk University Students Through Depression, Sleep Disturbance, and...

Background: Suicide prevention strategies in university settings largely rely on detecting explicit suicidal ideation. However, students experiencing severe psychiatric distress may not endorse suicidal thoughts and therefore remain unidentified by ideation-centered screening models. This study aimed to identify and clinically characterize university students with high depressive symptoms, poor sleep quality, and elevated impulsivity who deny suicidal ideation in order to examine whether they represent a vulnerable yet overlooked subgroup. Methods: This cross-sectional secondary analysis included 814 undergraduate students from the National Sleep Research Resource (ANSWERS dataset). Participants were classified into three groups based on median splits of depressive symptoms (CES-D), sleep quality (PSQI), impulsivity (UPPS-P), and the presence or absence of suicidal ideation in the past three months: Invisible (high symptoms without ideation), Visible (high symptoms with ideation), and Healthy (low symptoms without ideation). Group differences were examined using ANOVA and chi-square tests. Multivariate logistic regression was conducted to assess independent predictors of suicidal ideation. Results: The Invisible group comprised 11.8% of the sample. Compared with Healthy participants, these individuals showed poorer sleep quality and higher levels of thwarted belongingness and perceived burdensomeness (all p < 0.001). Cannabis use was most prevalent in the Invisible group (54.2%), exceeding both Visible and Healthy groups (p < 0.001). In adjusted analyses, depressive symptoms (OR = 1.10, 95% CI: 1.08-1.12) and sleep disturbance (OR = 1.06, 95% CI: 1.01-1.12) independently predicted suicidal ideation, whereas impulsivity did not. Conclusions: A clinically meaningful subgroup of students experience severe psychological distress without endorsing suicidal ideation yet show behavioral and interpersonal vulnerability. These findings highlight a limitation of ideation-focused screening and support broader, symptom-informed mental health assessment strategies in university populations.