Associations between emotion dysregulation and multiple sleep domains in bipolar disorder: differential patterns across sleep quality, insomnia and daytime sleepiness

Background:Sleep disturbances are a core feature of bipolar disorder (BD) and contribute to functional impairment, reduced quality of life and increased relapse risk. Emotion dysregulation (ED), which is highly prevalent in BD, has been investigated as a key mechanism underlying sleep problems, particularly insomnia [1]. However, its differential association with different sleep domains -such as sleep quality, insomnia severity, daytime sleepiness and chronotype -remains insufficiently explored.

Aims: To investigate the association between ED and multiple sleep-related dimensions in a clinical sample of euthymic patients with BD, with a focus on sleep quality, insomnia severity, daytime sleepiness and chronotype.

Methods: A cross-sectional analysis was conducted using baseline data derived from the BEGIN study, a longitudinal study in an adult sample of euthymic patients with BD conducted in the Bipolar and Depressive Disorder Unit of the Hospital Clinic of Barcelona. ED was assessed using the Difficulties in Emotion Regulation Scale (DERS-S-SF), considering both total score and subscales. Sleep-related measures included sleep quality (Pittsburgh Sleep Quality Index, PSQI), insomnia severity (Insomnia Severity Index, ISI), daytime sleepiness (Epworth Sleepiness Scale, ESS) and chronotype (Morningness–Eveningness Questionnaire, MEQ). Partial correlations between sleep variables and DERS subscales were computed controlling for age and gender, with false discovery rate correction for multiple comparisons. Multiple linear regression models were conducted to examine the association between DERS total score and each sleep outcome. Separate models including all DERS subscales simultaneously were also fitted. The multiple linear regression model with ESS as dependent variable was additionally adjusted for current pharmacological treatments.

Results: A total of 88 euthymic patients with BD were included in the present study (42 % female, mean age: 51.16 yo +/- 13.02). Higher levels of ED were significantly associated with poorer sleep quality. DERS total score was positively correlated with PSQI after controlling for age and gender (partial r = 0.434, adjusted p < 0.001). In regression models, ED significantly predicted poorer sleep qualityPSQI (B = 0.346, p < 0.001), explaining a modest but significant proportion of variance (R² = 0.166, adjusted R² = 0.137, p = 0.002). The model including all DERS subscales was overall significant (R² = 0.199, adjusted R² = 0.129, p = 0.011), with the inattention subscale of DERS emerging as a significant predictor (B = 1.03, SE = 0.46, t = 2.22, p = 0.030). For insomnia severity, the regression models were significant for both DERS total score (R² = 0.25, adjusted R² = 0.21; p < .001) and subscales (R² = 0.262, adjusted R² = 0.173, p = 0.010). In contrast, no significant associations emerged between DERS measures and chronotype (MEQ) and DERS measures and daytime sleepiness (ESS).

Conclusions: In euthymic BD, ED was associated with poorer sleep quality and insomnia severity, but not with chronotype and daytime sleepiness. This pattern suggests that ED may shape how patients experience and maintain sleep, rather than reflecting a broader alteration of circadian preference or sleep propensity. Integrating emotion regulation strategies into sleep-focused interventions may help address a clinically relevant pathway of residual vulnerability in BD.

References

1. Samea F, Mortazavi N, Reimann GM, Ebneabbasi A, Zarei M, Khazaie H, Goldstein-Piekarski AN, Spiegelhalder K, Baglioni C, Sepehry AA, Tahmasian M. Insomnia and emotion dysregulation: a meta-analytical perspective integrating regulatory strategies and dispositional difficulties. Sleep Med Rev. 2025 Aug;82:102111. doi: 10.1016/j.smrv.2025.102111. Epub 2025 Jun 4. PMID: 40554329.