Objective: Attention Deficit Hyperactivity Disorder (ADHD) is a common co-occurring condition in autistic individuals. ADHD is sometimes first recognised in young adulthood because ADHD symptoms may be misattributed to autism due to superficial overlap in presentation and diagnostic overshadowing. It should be investigated whether ADHD questionnaires are accurate in screening symptoms in young adults with autism. The current study examined this.
Methods: Participants were autistic young adults (N = 119) who took part in the Special Needs and Autism Project (SNAP), a population-based cohort. ADHD research diagnoses were obtained through the parent-informed Young Adult Psychiatric Assessment. Parents and young adults (self-report sample N = 71) completed ADHD questionnaires (Aberrant Behavior Checklist hyperactivity/non-compliance subscale, Conners Adult ADHD Rating Scales ADHD Index, and Strengths and Difficulties Questionnaire ADHD subscale). Receiver operating characteristic analyses were conducted to explore if the questionnaires discriminated ADHD cases from non-cases. To assess whether results varied by intellectual functioning, subgroup analyses were completed for those with an IQ ≥ 70 vs. <70.
Results: Weighted ADHD rates were high. Overall although the measures were performing at or close to adequate levels (area under the curve was 0.66 to 0.79 for parent-report and 0.70 to 0.65 for self-report), no single measure met adequate thresholds for sensitivity and specificity simultaneously. Tool performance was not different for those with an IQ ≥ 70 vs. <70.
Conclusion: No single measure reported adequate performance for distinguishing ADHD from non-ADHD cases in this sample of young autistic adults. Use of current thresholds may lead to under-diagnosis.
Keywords: ADHD; Autism; accuracy; screening; young adults.
© 2023. The Author(s).