Sensitivity of the child behaviour checklist sleep items and convergent validity with the Sleep Disorders Scale for Children in a paediatric ADHD sample

Abstract

Objective: The Child Behavior Checklist (CBCL) is a commonly used measure of child and adolescent functioning, which includes seven items that can be aggregated to provide a purportedly valid measure of sleep functioning. The objective of this study was to examine the convergent validity of the CBCL in a paediatric ADHD population and to evaluate the sensitivity of the instrument when benchmarked against the Sleep Disorders Scale for Children (SDSC).

Methods: The parents of 215 individuals (ages 6-17 years, 86% male) completed the CBCL and SDSC as part of a battery of measured administered as part of a specialised ADHD service located in Perth, Western Australia. All participants had a diagnosis of ADHD confirmed by a paediatrician or psychiatrist prior to attending the service.

Results: The CBCL Sleep Composite Scale was strongly correlated with the SDSC, but reported below adequate internal reliability. Receiver Operating Characteristic (ROC) suggests that a cut-off score of 4 may have good diagnostic accuracy compared to SDSC.

Conclusions: The CBCL Sleep Composite Scale may be reasonable to use if no purpose-developed sleep screening tool is available. The CBCL sleep items demonstrated good convergent validity, however, did not otherwise demonstrate acceptable psychometric properties that would endorse its use in an ADHD sample. The development of a specific measure of sleep in children with ADHD children is recommended.

Keywords: ADHD; CBCL; Hyperactivity; SDSC; Sleep difficulties; Sleep problems.

Fig. 1

Receiver Operating Characteristic (ROC) Curve depicting the sensitivity and specificity of the Child Behavior Checklist seven-item Sleep Subscale in identifying children experiencing clinically-significant sleep difficulties as scored on the Sleep Disorders Scale for Children (SDSC). Area Under Curve (AUC) = 0.83 (95% CI [0.77, 0.88]).