Bipolarity as an Independent Correlate of Clinical Profiles in Insomnia: A Factorial Analysis Across Multiple Clinical Domains

BACKGROUND: This study investigated the independent and interactive effects of bipolarity (subthreshold bipolar spectrum features) on multidimensional clinical profiles in individuals with insomnia, using data from a comprehensive deep phenotyping study. METHODS: A total of 338 participants (252 with insomnia, 86 normal sleeper controls) from the SOMDAY deep phenotyping study were classified into a 2 × 2 factorial design: Sleep Status (Insomnia vs. Control) × Bipolarity (Korean version of the Mood Disorder Questionnaire [K-MDQ] ≥ 7 vs. < 7). Two-way analyses of covariance, adjusted for age and sex, were conducted across 14 validated clinical scales. Benjamini-Hochberg false discovery rate correction was applied, with rank-transformed sensitivity analyses. RESULTS: Sleep Status showed significant main effects across 13 of 14 scales (all false discovery rate-corrected p < 0.05), with the largest effects on sleep quality (Pittsburgh Sleep Quality Index [PSQI], partial η2 = 0.230) and circadian rhythm disturbance (Korean version of the Biological Rhythms Interview of Assessment in Neuropsychiatry [K-BRIAN], partial η2 = 0.210). Bipolarity was independently associated with elevations in 10 of 14 scales, with the strongest effects on anxiety (Generalized Anxiety Disorder-7, partial η2 = 0.059), anxiety sensitivity (Body Sensations Questionnaire, partial η2 = 0.049), and depression (Patient Health Questionnaire-9, partial η2 = 0.041). No interaction effects survived correction, indicating predominantly additive effects. CONCLUSION: Bipolarity, indexed by the K-MDQ, exerts statistically independent main effects associated with greater psychological distress, anxiety sensitivity, and circadian vulnerability, operating additively with insomnia status across both insomnia patients and normal sleepers. These findings support the integration of bipolar spectrum screening into sleep medicine assessment.