Difficult Temperament Moderates Links Between Sleep and Behavior Problems in Toddlerhood: Differential Associations by Reporter

• In Event: Poster Session 09
  In Poster Session: PS 09 Topic - Social, Emotional, Personality

Abstract

Difficult temperament (e.g., negative emotionality, poor self-regulation) have been associated with sleep problems, internalizing symptoms, and externalizing behaviors in toddlerhood (Goodnight et al., 2007; Troxel et al., 2013). However, internalizing and externalizing symptoms seem to be differentially linked with sleep indicators (e.g., Quach et al., 2018; Troxel et al., 2013), and most studies rely on parent reports of sleep problems and parent or teacher reports of child problem behaviors, without considering that measurement and links between sleep and internalizing/externalizing symptoms differ depending on the reporter. Child characteristics such as difficult temperament may attenuate previously identified links between sleep and internalizing and externalizing symptoms early in life (e.g., Quach et al., 2018) and this information is critical for informing the type and timing of interventions aimed at reducing behavior problems. Thus, we tested whether associations between multiple actigraphy-based sleep indicators and internalizing and externalizing symptoms were stronger for children with difficult temperament in toddlerhood, taking into account primary (e.g., mothers), parenting partner (e.g., fathers), and secondary caregiver reports (e.g., grandparent, teacher) of temperament and internalizing and externalizing symptoms.
Participants were 420 toddlers from a multi-site, longitudinal twin study assessed at 42 months of age (Mage = 3.63 years; 52.5% male; 90.45% European American). Wrist-based accelerometers (actigraphs) captured sleep activity, nighttime sleep duration and duration variability across 14 nights at 42 months; higher order composites of sleep activity, duration, variability, and timing were also used in analyses (see Staples et al., 2019 for details on actigraph composites). Toddler difficult temperament measured the amount and intensity of negative mood, soothability, and child fussiness (Bates et al., 1998) and was a composite of items rated by a given caregiver’s at 30 and 36 months. Primary, parenting partner, and secondary caregivers reported internalizing and externalizing symptoms with the Child Behavior Checklist at 42 months (Achenbach & Rescorla, 2000). Multiple regression analyses were conducted in Mplus 7.0 to account for nested nature of multi-site data and to handle missing data. Predictor variables were centered at zero and significant interactions were probed at 1 SD above and below the mean of difficult temperament (Preacher et al., 2006). Site, sex, ethnicity, and socioeconomic status were tested in all analyses as covariates.
Table 1 shows key model results. The covariates were non-significant and excluded from final models. Overall, shorter sleep duration, as indexed both separately and as a composite, shorter sleep duration, and more sleep duration variability predicted more internalizing and externalizing symptoms particularly for children with high levels of difficult temperament as reported by primary caregivers. Effects of sleep duration variability on internalizing and externalizing symptoms were moderated by difficult temperament for both parent (internalizing symptoms only) and parenting partner reports. No other sleep x temperament interaction effects were significant, and no models were significant for secondary caregiver reports of temperament and behavior problems. When we collapsed estimates of temperament, internalizing, and externalizing symptoms across reporters, significant interactions were no longer present, suggesting that the interaction between sleep and adjustment are particular to the reporter.

Authors

• Reagan S. Breitenstein, Appalachian State University
  Presenting Author

• John E Bates, Indiana University - Bloomington
  Non-Presenting Author