Screening for Autism Spectrum Disorders in Children with Down Syndrome in New York State

Background: People with Down syndrome (DS) are often described as social and friendly. However, families and professionals have long recognized that some individuals with DS have diminished social reciprocity, atypical communication development and unusual routines and habits, i.e., symptoms that are characteristic of an autism spectrum disorder (ASD). Objectives: To determine the prevalence of ASD in children with DS in a large, diverse population and to examine best practices for screening and diagnosing ASDs in individuals with DS and co-morbid intellectual disability. Methods: A tiered approach was employed to screen and diagnose ASD in children with DS. First, children ages 3-13 years with DS born in New York State (outside the 5 boroughs of New York City) were recruited through the New York Congenital Malformation Registry, parent groups and clinics and screened for ASD using the Modified Checklist for Autism in Toddlers (MCHAT), the Social Communication Questionnaire (SCQ), and the Pervasive Developmental Disorder in Mental Retardation Scale (PDD-MRS). Subsequently, a subgroup of screen positive (on any screening measure) and screen negative (on all screening measures) children were evaluated with the Autism Diagnostic Interview–Revised via telephone interview. Of those, geographically available children who were ADI-R positive and an equal number who were ADI-R negative were administered an Autism Diagnostic Observation Schedule (ADOS) and an assessment of their intellectual and language ability. Sensitivity and specificity of the three screening tests were examined relative to the ADI-R diagnostic and current algorithms, the ADOS, and expert clinical consensus using the DSM-IV checklist. Results: The families of 457 children completed the screening, 221 went on for an ADI-R, and 76 completed the ADOS and psychological evaluations. Expert clinical consensus was applied to all children that completed both ADI-R and ADOS. Findings indicated that as many as 29-44% of children with DS screened positive for symptoms compatible with an ASD. Compared to the ADI-R diagnostic algorithm, sensitivity and specificity, respectively, of the screening tests were: MCHAT (0.727 and 0.719), SCQ (0.568 and 0.930) and PDD MRS (0.580 and 0.867). Comparison to the clinical consensus diagnosis yielded similar results. Sensitivity was increased when using the ADI-R current algorithm but specificity was reduced. Conclusions: Many of the participants with DS demonstrated symptoms of an ASD as indicated by elevated scores on the screening and diagnostic assessments, supporting previous reports that some individuals with DS exhibit symptoms consistent with an ASD diagnosis. This finding suggests that primary care practitioners should screen for ASD as part of their routine monitoring of health and development. Positive screens in primary care settings warrant referral for comprehensive evaluations because of the complexity of differential diagnosis of ASD comorbidity with intellectual and language disabilities. However, future research should evaluate the validity of these screening measures to inform best clinical practice for sufficiently sensitive and specific screening and diagnosis for children with DS. Findings have implications for timely and effective treatment of children with DS and co-morbid ASD.