ASD Co-Occurring Conditions and Change of ASD Diagnosis

Background:  Researchers have recently begun to understand the importance of co-occurring developmental and psychiatric conditions in the context of proper diagnosis and treatment of autism spectrum disorders (ASDs).  Previous studies have shown high rates of co-occurring developmental and psychiatric disorders in children with an ASD, while stability of ASD diagnosis seems to vary (75 – 95%).  The need for a better understanding of the extent to which ASD co-occurring conditions are associated with ASD diagnosis and diagnostic change is warranted.

Objectives: This study aims to investigate associations between common co-occurring conditions in ASDs and ASD diagnosis change across three developmental stages: young children (ages 3 – 5), children (ages 6 – 11) and adolescents (ages 12 – 17).

Methods: The US 2007 National Survey of Children’s Health dataset was obtained. Two study groups across three developmental stages were defined based on parent-reported data: 1) reported current ASD diagnosis (n=154 for ages 3-5; n=373 for ages 6-11; n=386 for ages 12-17), and 2) past but no current ASD diagnosis (n=53 for ages 3-5; n=189 for ages 6-11; n=211 for ages 12-17).  Co-occurring conditions of interest included ADHD, learning disability, developmental delay, speech problems, hearing problems, anxiety, depression, behavioral or conduct problems, and seizures or epilepsy.  Multinomial logit models were carried out to examine the association between co-occurring conditions and a current or past ASD diagnosis.  Odds ratios were computed by taking the odds of each co-occurring condition for the current ASD group and comparing against the odds of the condition for the past ASD group.  Confounding variables adjusted in multivariate regression models included race, ethnicity, maternal education, and individualized education plan.

Results: Results from multivariate analysis indicated that co-occurring conditions distinguishing the two study groups are: 1) current moderate/severe learning disability (Adjusted Odds Ratio [AOR]=11.07, 95% Confidence Interval [95%CI]: 2.48-49.49), current moderate/severe developmental delay (AOR=9.20, 95%CI: 1.90-44.42), and past anxiety (AOR=13.46, 95%CI: 1.08-167.20) in ages 3-5; 2) past speech problem (AOR=3.85, 95%CI: 1.09-13.67), current moderate/severe anxiety (AOR=3.51, 95%CI: 1.28-9.65), and past hearing problem (AOR=0.23, 95%CI: 0.06-0.83) in ages 6-12; and 3) current moderate/severe speech problem (AOR=3.91; 95%CI: 1.56-9.75), current mild seizure/epilepsy (AOR=10.48; 95%CI:2.25-48.87), and past hearing problem (AOR=0.11; 95%CI: 0.01-0.93) in ages 13-17.  Due to small numbers (n<5) of some co-occurring conditions in the past age 3-5 ASD group, results of this age group need to be interpreted with caution.

Conclusions: This analysis showed that children with current ASD diagnosis are significantly more likely to have speech problems in the past and current moderate/severe anxiety, but less likely to have hearing problems, as compared to same age children with past ASD diagnosis.  On the other hand, adolescents with current ASD diagnosis have more current, moderate/severe speech problems and current, mild seizures but fewer past hearing problems.  Some co-occurring conditions are associated with the change of ASD diagnosis in children and adolescents with an ASD diagnosis.  Effort is needed to clarify whether the change of ASD diagnosis (persistent vs. “grew out” of the diagnosis) is due to etiologic differences or diagnostic determination.