Poster #46 - Can Race be a Significant Predictor of an Autism Diagnosis?

• In Event: 3-053 - Poster Session 10
  In Poster Session: PS 10 Section - Developmental Disabilities

Integrative Statement

Background: In autism research, there are fewer African American than Caucasian participants, as well as limited research examining the relationship between race and whether or not a person receives an autism spectrum disorder (ASD) diagnosis (Wallis and Pinto-Martin, 2008). However, some studies have indicated that African American children are less likely to be diagnosed with ASD (Jarquin, Wiggins, Schieve, Van Naarden-Braun, 2011). This study investigated initial racial discrepancies that were found between Caucasian and African American participants across four diagnostic conditions, specifically to explore potential racial disparities in patterns of ASD and ID diagnosis in a clinic-referred sample.
Participants: 390 Caucasian and African American participants ages 5 – 23yrs. (M=11.21, SD=3.74) were chosen from neuropsychology clinic referred databases. Participants were classified into four groups: group 1: ASD with Intellectual Disability (ID), group 2: ASD without ID, group 3: ID without ASD, and group 4: non-ID, non-ASD (Table 1). Those without ASD had other medical and developmental disabilities (e.g., epilepsy, ADHD). In this sample, ID was defined by having full-scale IQ and adaptive functioning scores < 75.
Methods: Frequency distributions were conducted on parent-reported race across the four diagnostic conditions. The initial analysis showed abnormally high numbers of African American participants compared to Caucasian participants in groups 1 and 3, but not in groups 2 and 4. Due to these discrepancies, other race categories were excluded from further analysis to focus on differences between African American and Caucasian groups. A chi square analysis was conducted to analyze the relationship between race and ID as well as race and ASD. A logistical regression was then conducted using an autism diagnosis as a dependent variable to determine if race can significantly predict an ASD diagnosis, when compared to the other commonly used predictive factors gender, IQ and adaptive scores.
Results: The frequency distributions in Table 1 show discrepancies in reported race between groups. Results from the chi square analysis showed a significant interaction across all four groups when comparing race and ID (χ² (3) = 25.45, p < .001) as well as race and ASD (χ² (3) = 25.45, p <. 001). Results from the logistical regression showed the overall model between race, gender, IQ and adaptive scores to be significant in predicting an ASD diagnosis (χ2 (4) = 59.41, p < .001). When compared with other significant predictive factors such as gender (β = .836, p < .001), IQ (β = .27, p < .001) and adaptive scores (β = -.060, p < .001), race did not significantly predict an autism diagnosis (β = -.369, p = .173).
Conclusion: Results of this study indicate that racial disparities exist in diagnosis patterns within this clinic-referred sample. However, when other diagnostic factors are accounted for (e.g., gender, IQ and adaptive scores), race is not a significant predictor for ASD diagnosis. Further research should be conducted to determine how race may interact between the predictive factors of gender, IQ and adaptive scores.

Authors

• Maddy Quidzinski, The Catholic University of America
  Presenting Author

• Allison B. Ratto, Children's National Medical Center
  Non-Presenting Author

• Gregory L Wallace, The George Washington University
  Non-Presenting Author

• A. Chelsea Armour, Children's National Medical Center
  Non-Presenting Author

• Alyssa Verbalis, Children's National Medical Center
  Non-Presenting Author

• Madison Berl, Children's National Hospital
  Non-Presenting Author

• Kelsey Csumitta, National Institutes of Health
  Non-Presenting Author

• Lauren Kenworthy, Children's National Medical Center
  Non-Presenting Author