Of the 338 adults who joined a sleep study at Korea University, 129 said yes to at least seven of thirteen questions about lifetime spells of unusually elevated mood or energy. None of them carried a bipolar disorder diagnosis; anyone with one had been screened out before enrollment. Yet those yes answers turned out to mark something consequential. Across questionnaires measuring anxiety, depression, fatigue, and the regularity of daily routines, people above that threshold scored consistently worse, whether their sleep was wrecked or perfectly ordinary.
The analysis, published in the journal Brain and Behavior by Sujin Kim, Chul-Hyun Cho, and colleagues at Korea University College of Medicine, draws on the SOMDAY study, a project that characterizes insomnia through questionnaires, wearable devices, brain imaging, and biomarkers. This paper used the questionnaire arm: 252 people with insomnia, defined by frequent symptoms over three months and a high score on the Insomnia Severity Index, plus 86 normal sleepers, all recruited from the community between March 2023 and October 2024.
Psychiatrists use the term bipolarity for manic or hypomanic tendencies that sit below the diagnostic threshold, on a continuum running from ordinary mood variability to full bipolar disorder. The team measured it with the symptom checklist from the Korean version of the Mood Disorder Questionnaire, counting how many of thirteen lifetime symptoms each person endorsed. Seven or more meant high bipolarity, a bar that 38 percent of participants cleared. The researchers then crossed the two classifications, insomnia versus normal sleep and high versus low bipolarity, and compared the four resulting groups on 14 validated clinical scales, statistically adjusting for age and sex and correcting for the number of tests run.
Insomnia status dominated, as expected. It was linked to worse scores on 13 of the 14 scales, with the largest gaps in overall sleep quality, where it accounted for about 23 percent of the variation, and in disturbed biological rhythms, at about 21 percent.
The more novel result is what bipolarity did on its own. After the effects of insomnia were statistically set aside, high scorers on the bipolarity screen still showed elevations on 10 of the 14 scales. The strongest links were with anxiety, with anxiety sensitivity (a fear of bodily sensations such as a pounding heart, measured by the Body Sensations Questionnaire), with depression, and with seasonal swings in mood. The effects were small to moderate, explaining roughly 3 to 6 percent of the variation in each case, but they held up after correction and appeared in insomnia patients and normal sleepers alike.
Notably, the two factors did not amplify each other. No interaction between insomnia and bipolarity survived statistical correction, which suggests their contributions simply stack: high bipolarity added about the same extra burden whether or not a person had insomnia. One exception hinted at more, a fatigue measure where bipolarity seemed to matter mainly in the insomnia group, but that pattern failed correction and the authors treat it strictly as a lead for future work. There was also a telling null result. The Pittsburgh Sleep Quality Index, the core sleep quality measure and the scale most strongly tied to insomnia itself, showed no independent link to bipolarity. In the authors' reading, bipolarity connects to the psychological and circadian landscape around a sleep problem rather than to the core sleep complaint.
The study has real limits, which the authors lay out plainly. It is cross-sectional, so it cannot say whether bipolar traits drive distress or some shared vulnerability produces both. The questionnaire is a screening tool, not a diagnosis, and the simplified scoring the team used, which skips the instrument's usual requirements that symptoms cluster together and cause impairment, likely inflates how many people count as high bipolarity. The smallest group, normal sleepers with high bipolarity, held just 24 people, leaving limited power to detect subtler interactions. And the sample came from one Korean community, so the pattern needs testing elsewhere.
Why it matters
Sleep clinics routinely measure sleep, mood, and daytime function, but they rarely screen for bipolar spectrum traits, especially in people who have never had a manic episode severe enough to reach a psychiatrist. This study suggests that a brief checklist can flag a subgroup carrying extra anxiety, depression, and circadian disruption that standard sleep measures alone would miss.
The finding that these elevations showed up even among normal sleepers points beyond the sleep clinic, toward primary care, where sleep complaints are common and bipolarity is almost never assessed. The authors propose that patients flagged this way might eventually benefit from insomnia therapy augmented with mood and circadian rhythm components, though they are careful to note that this remains a hypothesis for future trials, not yet a demonstrated treatment effect.