Sleep duration in infancy predicts autism in girls.

 Saenz J, Yaugher A and Alexander GM (2015) Sleep in infancy predicts gender specific social-emotional problems in toddlers. Front. Pediatr. 3:42. doi: 10.3389/fped.2015.00042

Despite strong evidence linking sleep to developmental outcomes, the longitudinal relationship between sleep and emotional well-being remains largely unknown. To address this gap in our knowledge, the current study examined sleep in infancy, measured via actigraphy, as a predictor of social-emotional problems in toddlers. A total of 47 children (29 males) were included in this longitudinal study. At time one, actigraphy measures of sleep were obtained from 3- to 4-month-old infants. At time two, parents rated their 18- to 24-month-old toddler’s social-emotional well-being using the Brief Infant Toddler Social Emotional Assessment. Results indicated that boys tended to have higher levels of externalizing behaviors than did girls. Additionally, boys with longer sleep durations also showed lower sleep efficiency. In girls, sleep duration in infancy was a significant predictor of autism spectrum disorder behaviors and approached significance as a predictor of externalizing problems in toddlerhood. Our findings are the first to show a relationship between sleep measured in infancy and autism spectrum disorder symptomatology measured in early childhood. They suggest that the etiology of social-emotional problems may differ between genders and raise the possibility that sleep/wake cycles may be differentially related to autism spectrum disorder symptoms in girls and boys...

The overall regression used to predict externalizing problems in girls was not significant, F (2, 15) = 2.73, p = 0.098, R² = 0.27. However, of the predictors investigated, sleep duration was approaching significance as a predictor of externalizing problems, β = −0.46, t (15) = −2.06, p = 0.058 (Table 2). That is, after controlling for sleep efficiency, a 1-min decrease in sleep duration resulted in a 0.005 point predicted increase in a child’s externalizing problems score. The regressions used to predict internalizing problems [F (2, 15) = 0.69, p > 0.05, R² = 0.09] and dysregulation problems [F (2, 15) = 0.95, p > 0.05, R² = 0.11] were also not significant, and no main effects emerged. Lastly, the regression used to predict autism spectrum behaviors in girls was not significant, F (2, 15) = 2.6, p > 0.05, R² = 0.51. However, of the predictors investigated, sleep duration was a significant predictor of autism spectrum behaviors, β = −0.48, t (15) = −2.14, p < 0.05. That is, after controlling for sleep efficiency, a 1-min decrease in sleep duration resulted in a 0.01 point predicted increase in a child’s autism spectrum behaviors score.