The High Prevalence of Autism Spectrum Disorders Among Children with Intellectual Disabilities

Background:  Recent studies have estimated the prevalence of Intellectual Disability (ID) at 7 in 1,000 children (Boyle et al., 2011); however prevalence estimates relating to the rate of co-occurring autism spectrum disorders (ASD) have varied markedly based on the definition, assessment instruments, and study populations.  Despite these discrepancies, several consistent trends have emerged.  First, the prevalence of ASD appears to be inversely related to the severity of ID, such that ASD rates increase at lower levels of cognitive ability.  Second, as cognitive functioning decreases, the gender gap between males and females in ASD prevalence also narrows.  Last, it appears that many children with ID may have undiagnosed or unidentified comorbid diagnoses (Saemudsen et al., 2010), which showcases the importance of obtaining accurate prevalence rates for ASD in youth with ID in order to inform clinical diagnosis and ensure appropriate service utilization.     

Objectives:  (1) To estimate the prevalence of ASD in a population-based epidemiological sample of youth with ID; diagnosed through record review methodology. (2) To compare characteristics of youth with comorbid ID/ASD to youth with ID only.

Methods:  Data for the present study came from SC-ADDM, which is one of several sites collaborating with the CDC to conduct ID surveillance in the United States.  For each of the study years, all 8-year-old children with ID in the study area (the Coastal and Pee Dee regions of SC) were identified through screening and records abstraction at multiple educational and clinical sites.  The de-identified records were linked with ASD surveillance data also collected by SC-ADDM to determine the prevalence of ASD in youth with ID.

Results:  Among youth born in 1994, 1996, 1998, and 2000, 1,818 met criteria for ID as defined by IQ < 70 on standardized instruments.  Of these 1,818 youth, 38% (n = 698) also met ADDM case definition for ASD.  Children with comorbid ASD/ID were more likely to be male (75% vs. 64%, χ² (1) = 20.7, p < 0.05) and were more likely to have moderate to severe intellectual disability (43% vs. 16%, χ² (1) = 67.5, p < 0.05) compared to those with ID only.  Youth with ID/ASD were also significantly more likely to have low birth weight (χ²(1) = 5.7, p < 0.05). 

Conclusions:  This is the first study in almost twenty years to evaluate ASD prevalence and characteristics in a sample of youth with ID.  Results suggest that almost 40% of children with ID have comorbid ASD, suggesting very high risk for ASD in this population. Increased ASD screening efforts may be needed in this group to ensure appropriate identification and intervention. Strengths include a large epidemiological sample utilizing independent review and assessment of ASD and ID, regardless of previous community diagnosis, which allows for more accurate prevalence estimates even for youth without formal clinical diagnoses.  Obtaining accurate prevalence estimates of ASD in youth with ID is needed to inform clinical practice and ensure appropriate service utilization.