Perinatal sleep health, stress, and emotion dysregulation: A repeated measures study

• In Event: Importance of multilevel perinatal research for advancing child development and family health equity

Abstract

Emotion dysregulation during the perinatal period is associated with psychological and physical morbidity in the pregnant person. Studies with non-pregnant adults have demonstrated that sleep disturbances increase emotion dysregulation (Mauss et al., 2013). However, these associations have received limited attention during postpartum, despite the fact that almost one out of every four mothers will develop new-onset postpartum sleep disturbance (Sivertsen et al., 2017). Our first aim was to examine the relation between sleep and emotion dysregulation during early postpartum. Our second aim was to investigate whether this relation varied based on multiple domains of life stress—namely, health, interpersonal, and financial.
Participants were 82 women aged 19-38 who were enrolled in a longitudinal study examining the intergenerational transmission of emotion dysregulation. During their 3rd trimester of pregnancy, women completed the UCLA Life Stress Interview (LSI; Hammen et al., 1987). The LSI is a semi-structured interview assessing life-related stress over the past 6 months. In the present study, the “Health,” “Partner,” and “Finance” domains were included in analyses. Between 6 and 16 weeks postpartum, participants completed approximately 14 days of daily surveys assessing sleep health (Consensus Sleep Diary [CSD]; Carney et al., 2012) and emotion dysregulation (State Difficulties in Emotion Regulation Scale; Lavender et al., 2017). Multilevel models were run in HLM8. Daily experiences with emotion dysregulation were entered as the outcome. Nightly subjective sleep quality was entered as a within-person predictor, and the “Health,” “Partner,” and “Finance” domains of the LSI were entered separately as between-person predictors.
Within-person variance accounted for approximately 40% of the variability in state emotion dysregulation (ICC = .60). On average, higher CSD sleep quality ratings were associated with lower emotion dysregulation (B = -2.70, p = .02). The daily association between better sleep quality and lower emotion dysregulation was significant (B = -.75, p = .01) and varied across people, χ2(80) = 113.18, p = .01. Subsequent models including LSI domains indicated that the daily association between sleep quality and emotion dysregulation varied significantly across levels of prenatal financial stress (B = -.58, p = .04). On nights when new mothers reported poor sleep quality, they were more likely to endorse high emotion dysregulation, and this daily negative association was significant only for women who experienced high levels of financial stress during pregnancy (p = .002). There was a significant conditional main effect of prenatal health-related stress on postnatal emotion dysregulation such that at the average level of postnatal sleep quality, more prenatal health-related stress predicted higher postnatal emotion dysregulation (B = 2.79, p = .004). Prenatal health-related stress did not interact with postnatal sleep quality to predict emotion dysregulation (B = -.39, p = .16), and prenatal partner-related stress exhibited neither a significant main (B = 2.24, p = .12) nor interaction effect (B = -.18, p = .54) on postnatal emotion dysregulation. Our findings could help illuminate those at greatest risk for the link between poor sleep and emotion dysregulation postnatally, thereby informing targeted efforts to reduce maternal health disparities.

Authors

• Parisa R. Kaliush, University of Utah
  Presenting Author

• Liz Conradt, Duke University

• Paula G. Williams, University of Utah

• Sheila E Crowell, University of Utah