Differences in Diagnosis and ASD Severity Between Latino and White Children

Background: Ethnic minority children are at an increased risk for under-diagnosis and delays in diagnosis of autism spectrum disorder (ASD; Jarquin et al., 2011; Mandell et al., 2002) and appear to have more severe symptoms of ASD (Tek & Landa, 2012).  Latino children seem to be at particular risk for these disparities, due to language barriers, financial limitations, and diminished healthcare knowledge (Liptak et al., 2008; Pew Hispanic Center, 2008).  Presently, there is little research available on ethnic disparities in ASD. 

Objectives: The goal of the present study was to examine differences in diagnostic outcomes and ASD symptom severity between Latino and White children with a diagnosed ASD.  It was hypothesized that Latino children would have greater delays in diagnosis and more severe symptoms of ASD.  Income and maternal autism knowledge were also hypothesized to be significant predictors of delays in diagnosis.  

Methods: Participants included ASD-diagnosed children of Spanish-speaking Latina (n=28) and English-speaking White (n=28) mothers.  Mothers completed background questionnaires, the Social Communication Questionnaire (Rutter, Bailey, & Lord, 2003), and the First Year Inventory-Retrospective (Watson et al., 2007), in their primary language.  Mothers were recruited from an ASD research registry and ASD parent support and events. 

Results: Despite a lack of significant difference in the age at which mothers reported first developing concerns about their child, Latino children were diagnosed later than their White peers on average, at a level approaching significance (t= -1.93, p<.06).  Notably, among Latina mothers greater affiliation with Latino culture was associated with a later age of first concerns (F=5.07, p<.05, R²=.16). There was also a greater delay among Latino children between the time at which mothers reported first having concerns about their child and the time at which children were diagnosed (t= -2.23, p<.05).  Stepwise multiple linear regression analyses indicated that these differences in age at diagnosis (F=6.08, p<.05, R²=.10) and time to diagnosis (F=6.38, p<.05, R²=.11) were better accounted for by household income. Lower current maternal knowledge of ASD was also associated with greater delays in diagnosis (r= -.31, p<.05).

There were no significant differences in ASD severity as measured by the SCQ or the FYI-R. However, Latino children were significantly more likely to be diagnosed with autism (odds ratio=6.92), as opposed to Asperger syndrome, high-functioning autism, or PDD-NOS (χ²=10.10, p<.001). Household income did not predict diagnostic labels. Furthermore, survival analyses indicated that Latino children spoke their first words (χ²=3.11, p<.08) and achieved toilet training (χ²=4.54, p<.05) at later ages on average than their White peers. 

Conclusions: The results of the present study provide additional evidence of delays in diagnosis among Latino children, which are not attributable to delays in parent concerns. Household income appeared to drive these differences. Despite a lack of difference in symptom severity on standard measures, Latino children were more functionally impaired by ASD.  Further research is needed to identify the factors that contribute to ethnically-based delays in diagnosis and functional impairment.  Research is also needed to evaluate whether ethnically-based differential item functioning may occur on parent-report measures of ASD.