Poster #104 - Mother-Child Attachment Security and Child Temperament as Predictors of Sleep Developments During the Preschool Period

• In Event: 2-163 - Poster Session 08
  In Poster Session: PS 08 Section - Health, Growth, Injury

Integrative Statement

Important developments in sleep occur across the preschool period, including increased sleep efficiency and proportion of nighttime sleep as well as decreased daytime and total 24-hour sleep duration. However, little is known of the factors that predict inter-individual differences in these developments. Mother-child attachment security (Bélanger et al., 2015) and child temperament (Ward et al., 2008) have been found to predict child sleep at specific ages. Accordingly, the current longitudinal study investigated the predictive role of these two factors in individual differences in sleep developments at preschool age.

When children (N = 128) were 2 years old, mother-child attachment security was assessed with the Attachment Q-Sort (Waters, 1995) rated by an observer following a 1.5-hour home visit, and mothers reported on three dimensions of their child’s temperament with the Toddler Behavior Assessment Questionnaire (Goldsmith, 1996); activity level, social fear, and proneness to anger. At 2, 3 and 4 years, sleep was assessed for 72 consecutive hours using an actigraph (Mini-Mitter® Actiwatch, Respironics). Actigraphic data were used to derive average nighttime sleep duration, daytime sleep duration, total 24-hour duration, proportion of nighttime sleep, and sleep efficiency over the three-day period at each age.

Growth curves were fitted in Mplus using a multilevel modeling framework. Results showed a linear decrease of daytime, total, and nighttime sleep duration, and a linear increase of proportion of nighttime sleep and sleep efficiency between ages 2 and 4 (Table 1). Next, the extent to which individual differences in initial status and rate of change for each sleep variable could be explained by attachment security and dimensions of temperament was examined. Family socioeconomic status was entered first, followed by predictors (one per model; Table 2). Attachment security and social fear did not significantly predict sleep growth curves. Every one-standard-deviation increase in activity level was associated with a 26.4-minute (.44-hour) decrease in the initial status of total daily sleep duration. In addition, every one-standard-deviation increase in proneness to anger was associated with a decrease in the initial status of the proportion of nighttime sleep (2.31%; marginal), nighttime sleep duration (.32 hours or 19.2 minutes), and sleep efficiency (2.35%). Every one-standard-deviation increase in proneness to anger was also associated with an increase in the annual rate of change of the proportion of nighttime sleep (1.54%; marginal), nighttime sleep duration (.18 hours or 10.8 minutes; marginal), and sleep efficiency (1.47%).

In sum, child temperament, especially anger proneness, was a better predictor of preschool sleep developments than mother-child attachment. It is possible that intrinsic factors are more relevant for sleep early on, when children show very different levels of physiological maturation. Moreover, the results suggest that children with a more difficult temperament enter the preschool period with shorter, less consolidated, and less efficient sleep, but that these aspects of their sleep then increase at a faster rate, possibly allowing them to catch up with their peers and achieve similarly mature sleep, albeit later on. Further research is needed to decipher the implications of these different rates of sleep maturation.

Authors

• Émilie Tétreault, University of Montreal
  Presenting Author

• Célia Matte-Gagné, Laval University
  Non-Presenting Author

• Annie Bernier, University of Montreal
  Non-Presenting Author