New study improves criteria to identify sleep problems in autistic children and adolescentsOctober 5, 2021
A study published this September in Sleep Medicine has developed and validated a new cut-off value for the Children’s Sleep Habits Questionnaire for autism (CSHQ-autism), a common tool used to screen for sleep difficulties in autistic children and adolescents up to 17 years of age.
The original CSHQ is a 45-item questionnaire developed for children aged 4 to 10 years that assesses sleep difficulty by instructing parents to rate the frequency with which their child displayed certain sleep behaviors during the previous week. Ratings are combined to create eight subscales that relate to common sleep problems. Clinicians then calculate a total sleep disturbances index, for which a score of over 41 indicates a sleep disorder.
However, concerns have been raised that the tool has limitations that prevent it from capturing sleep difficulties in autistic children and adolescents, and that the original cut-off score of 41 may over-identify sleep problems in autistic children aged 2 to 5 years. As a result, the 23-item, four-factor CSHQ-autism was designed specifically for autistic children—but a cut-off value was not specified.
In this study, researchers aimed to develop and validate age-specific cut-off values for the CSHQ-autism that can more accurately identify sleep problems among autistic children. By analyzing parent-reported sleep problems from a sample of 5,472 children aged 2 to 17 years recruited from 16 Autism Treatment Network sites across North America, researchers identified the following cut-off values:
- For children aged 2-3 years — 34
- For children aged 4-10 years — 35
- For children aged 11-17 years — 33
- For children aged 2-17 years — 35
These cut off-values performed with moderate to high sensitivity (meaning they were able to correctly detect people with sleep problems) and moderate specificity (meaning they could correctly rule out people without sleep problems) across all age groups, indicating that the new CSHQ-autism total scores are effective among autistic children aged 2 to 17 years. In fact, the CSHQ-autism cut-off values performed better than the original CSHQ cut-off value of 41.
While the CSHQ-autism is slightly less likely than the original CSHQ to correctly identify sleep disorders among people who have them, it is more likely to correctly rule out sleep disorders among people who don’t, avoiding false positive results. The CSHQ-autism also has the advantage of being a shorter questionnaire, cementing is as a valid tool to evaluate sleep difficulties in autistic children.