Intra-individual Variability in Sleep Duration and Schedule in Middle Childhood: Links to Early Maternal Sensitivity and Education

• In Event: Parent and Child Sleep in the Context of Family Adversity

Abstract

The importance of sleep for healthy child development is well demonstrated. However, most studies focus on one aspect of sleep, namely its average duration. Studies considering sleep parameters other than duration are needed to better understand the role of sleep in child development (Matricciani, 2019). Traditionally, sleep measures are averaged across several days to obtain a more reliable measure. Recently, however, research has begun to examine also the notion of intra-individual variability (IIV) in sleep, which reflects the night-to-night fluctuations in a person’s sleep across a given timeframe (Bei et al., 2016). Research suggests that IIV relates to numerous indices of cognitive and socioemotional maladjustment in children (Becker et al., 2017); it is, therefore, important to understand why some children have more variable sleep than others. This study focused on two sources of stress in infancy, namely low levels of maternal education and poor maternal sensitivity, as predictors of IIV in sleep duration and sleep schedule in middle childhood.

A sample of 162 mother–child dyads were first visited in their homes when children were aged 1 year. The 1.5-hour visits consisted of a series of child-focused research tasks and mother-child interactions. Trained research assistants observed maternal behavior throughout and then scored maternal sensitivity using the Maternal Behavior Q-Sort (Pederson & Moran, 1995). Mothers reported on their total years of education. Approximately 8 years later, when children were in third (M age = 8.76 years) and fourth grade (M age = 10.01 years), they were invited to wear an AW-2 Actiwatch on their non-dominant wrist for seven consecutive nights during the regular school year. At both time points, IIV was computed by calculating the standard deviation across the seven nights of assessment for sleep duration and three aspects of sleep schedule, namely sleep onset, sleep offset, and sleep midpoint (halfway point between sleep onset and offset). Estimates of IIV were then averaged across the two timepoints.

The analyses revealed no significant associations between maternal education or sensitivity and IIV in sleep duration. However, after accounting for child sex and age at each sleep timepoint, children of less educated mothers had more variable sleep onset times (β = -.20, p = .027), offset times (β = -.30, p < .001), and midpoints (β = -.29, p < .001) eight years later. Children of less sensitive mothers also had more variable sleep onset times (β = -.20, p = .032). Given that both education and sensitivity were related to IIV in sleep onset, we investigated the presence of a mediated pathway. The analyses revealed that poorer sensitivity was a significant mediator of the association between lower maternal education and child IIV in sleep onset (BCBCI = -.015 – ¬-.001).

The findings suggest that early stress in the family environment, in the form of low maternal education and poor sensitivity, may initiate a developmental cascade that results in unstable child sleep schedules several years later.

Authors

• Annie Bernier, University of Montreal
  Presenting Author

• Charlotte Dusablon, University of Montreal

• Anna-Francesca Boatswain-Jacques, McGill University

• Elie Yu Tong Guo, Université de Montréal (UdeM)

• Laurianne Bastien, Université de Montréal (UdeM)