Abstract
BACKGROUND: Attention deficit hyperactivity disorder (ADHD) is often present in people with intellectual disability (ID) and autism. However, few ADHD measures have been developed specifically for individuals with these conditions. There is little literature exploring how well ADHD measures are performing at picking up specific symptoms at the item level. METHODS: Analyses were conducted on data from 122 children aged 7-15 years old with diagnoses of both ADHD and ID enrolled in the Hyperactivity and Special Educational Needs trial. Parents and teachers completed ratings of ADHD symptoms on the Aberrant Behavior Checklist (ABC) hyperactivity subscale and the revised Conners' Rating Scales hyperactivity scale and ADHD index. Cronbach's alpha was used to examine the reliability of these measures. Item response theory explores the performance of individual items. Multiple indicators, multiple causes models were used to test for measurement invariance by ID severity, co-occurring autism traits and child age. RESULTS: The reliability of parent and teacher reports of ADHD symptoms on the Conners' and ABC was acceptable across the range of ID. Item performance was generally good, and information was provided across the continuum of ADHD traits. Few items on either measure were non-invariant (i.e., item endorsement generally did not differ based on other child characteristics). When non-invariance was found, the effect was small. CONCLUSIONS: Both the parent-reported and teacher-reported versions of the Conners' hyperactivity scale and ADHD index and the ABC hyperactivity subscale appear to function well in the current sample of children with co-occurring ADHD and ID.