Changes in Sleep Related With Behavioral Problems Between Childhood and Adolescence

During the transition from childhood to adolescence, a study published in JAMA Psychiatry identified profiles of sleep problems which were related with behavioral and emotional problems.

Investigators from the University of Melbourne in Australia sourced data for this study from the Adolescent Brain Cognitive Development (ABCD) study which was an observational cohort study conducted in a community setting in the United States.

Caregivers of children (N=10,313) completed the Sleep Disturbance Scale for Children (SDSC) and the 113-item Child Behavior Checklist (CBCL) instruments at baseline and at a 2-year follow-up between 2016 and 2020. The change in sleep problems and their relationship with psychopathology symptoms during the transition from childhood and adolescence were evaluated. Sleep pathology was defined as a SDSC t score of 70 or greater and internalizing, externalizing, and total problems were defined as SDSC t scores of 65 or greater.

At baseline, the children were mean age 119 (SD, 7.51) months, 52.4% were boys, 54.0% were White, they had a BMI of 18.7 (SD, 3.72) kg/m², and 42.7% had a parent with psychopathology.

At baseline, 7.9% met the criteria for sleep problems, 8.9% for internalizing problems, and 8.8% for externalizing problems. At follow-up, the proportion who met the criteria for sleep (7.8%) and internalizing (8.8%) problems were similar but the proportion who met the criteria for externalizing symptoms had decreased from baseline (7.3%).

The data at baseline best fit a 4-profile solution (intraclass correlation coefficient [ICC], 0.97; P <.001).

Overall, 25.2% fit the low disturbance profile, 16.0% had sleep onset or maintenance problems, 42.3% had mixed disturbance, and 16.5% had high disturbance. At follow-up, the 4 classes comprised 30.3%, 32.6%, 22.1%, and 15.0%, respectively.

Between baseline and follow-up, remaining in the same profile was most likely (55%-83%). For other transitions, the most common were a high disturbance to sleep problems (21%), a mixed disturbance to low disturbance (20%), and a low disturbance to sleep problems (18%).

Having sleep disturbance at baseline was most strongly related with high disturbance on the somatic problems subscale (odds ratio [OR], 1.59; 95% CI, 1.47-1.73) and high disturbance on the anxiety and depression subscale (OR, 1.50; 95% CI, 1.42-1.58). Trends were similar at follow-up.

During the transition to adolescence, higher internalizing scores increased the likelihood of transitioning from mixed to high disturbance profiles (OR, 1.09; 95% CI, 1.02-1.16; P =.02) and decreased the likelihood of transitioning from high to mixed disturbance profiles (OR, 0.90; 95% CI, 0.83-0.98; P =.01) compared with transitioning into the low disturbance profile.

The results of this study are likely not generalizable for children with severe neurodevelopmental disorders as they were excluded from this analysis.

Study authors concluded, “In this study, sleep problems were heterogenous and highly comorbid in late childhood and early adolescence and varied in their developmental course. The identification of discrete sleep profiles suggests that interventions should target specific patterns of sleep problems as an alternative to nonspecific targeting of sleep difficulties.”