16363
Understanding Associated Features of Autism Spectrum Disorder and Their Relationship to DSM Diagnostic Criteria
Children with autism spectrum disorder (ASD) have many features beyond diagnostic symptoms that contribute to phenotypic heterogeneity. Little is known about the underlying structure of these associated features (AF) and how they relate to diagnostic criteria.
Objectives:
Our goal was to examine the latent structure of common ASD AF and the associations between those features and DSM-IV-TR diagnostic symptoms and domains.
Methods:
Children were identified from the Metropolitan Atlanta Developmental Disabilities Surveillance Program (MADDSP). MADDSP is a population-based surveillance system that monitors the prevalence of ASD and other developmental disabilities in metropolitan Atlanta, GA. We included data from the two most recent surveillance years: 2006 and 2008.
Case-finding in MADDSP is based on detailed abstraction of evaluation reports noted in special education and health records. Abstracted information is concatenated and then reviewed by trained clinicians who apply a standardized coding scheme based on DSM-IV-TR criteria to determine ASD case status. Clinician reviewers also code whether each child has any of 13 ASD AF: eating/drinking difficulties, sleeping difficulties, mood difficulties, uneven cognitive skills, aggressive behaviors, oppositional behaviors, delayed motor skills, Attention-Deficit/Hyperactivity Disorder (ADHD) characteristics, unusual fear response, unusual sensory response, self-injurious behaviors, seizures or seizure-like behaviors, and temper tantrums. Latent class analysis was used to classify the structure of the AF.
Results:
1,075 children met the MADDSP ASD surveillance definition, of which 84.3% were males and 55.8% had a co-occurring intellectual disability. The sample was 40.6% Non-Hispanic white, 40.5% Non-Hispanic black, 7.9% Hispanic, 9.2% other or non-specified race, and 1.8% missing race.
The AF most commonly reported were ADHD characteristics (90.4%), mood difficulties (78.5%), and unusual sensory response (75.7%). Three latent classes were distinguished by the probability of having each AF: Class 1 (high probability; 39%) had the highest probability followed by Class 2 (medium probability; 46%) and Class 3 (low probability; 15%). The only exception to this pattern was that Class 1 and Class 2 had an equally high probability of ADHD characteristics. The mean number of AF was 9.7 for Class 1, 6.8 for Class 2, and 3.2 for Class 3 when children were assigned to classes based on posterior probabilities.
Unusual sensory response and temper tantrums were associated with all 12 DSM-IV-TR symptoms from among all three diagnostic domains. Eating/drinking difficulties, mood difficulties, ADHD characteristics, and unusual fear response were each associated with nine DSM-IV-TR symptoms from among all three diagnostic domains. The association between other AF and diagnostic symptoms varied.
Conclusions:
Children with ASD have many AF that can be classified as high, medium, and low in probability. The presence of AF may complicate diagnostic and treatment decisions and should thus be considered when working with children with ASD (Close et al., 2011). ADHD characteristics, mood difficulties, and unusual sensory response were most prominent in our sample and associated with a majority of ASD diagnostic symptoms. These results support the DSM-V decision to allow dual ADHD and ASD diagnoses and highlight the need for future research on the overlap between ASD, ADHD, and mood and sensory disorders.