Relationship Between Medical Comorbidity and Problem Behavior in Children with Autism Spectrum Disorder

Thursday, May 11, 2017: 5:30 PM-7:00 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
A. Stedman1, K. A. Smith2 and M. Siegel3, (1)Spring Harbor Hospital, Westbrook, ME, (2)Maine Medical Center, Portland, ME, (3)Maine Medical Center - Tufts School of Medicine - Spring Harbor Hospital, Westbrook, ME
Background: Medical comorbidities, such as gastrointestinal and neurological disorders, in children with autism spectrum disorder (ASD) are relatively common and may have a negative impact on existing behavior problems related to an ASD diagnosis. The prevalence and type of medical comorbidities within the severely affected ASD patient population remains unclear. An improved understanding of medical comorbidities and their relationship to problem behavior has significant implications for treatment.

Objectives: To examine the prevalence of medical comorbidities within the Autism Inpatient Collection (AIC) sample, and to examine the relationship between common medical comorbidities and behavioral problems among severely affected children with ASD.

Methods: Data for this analysis were collected as part of a prospective study (AIC) examining phenotype, behavioral, and genetic data of children with ASD recruited from six specialized inpatient psychiatric units. Three-hundred and fifty children with ADOS-2-confirmed ASD enrolled in the AIC between February 2014 and October 2015 were included in this analysis. Medical comorbidities for each participant were collected from Axis III discharge diagnoses and categorized by ICD-10 code. To quantify severity of behavioral problems, the Aberrant Behavior Checklist-Irritability subscale (ABC-I) was completed at admission by a primary caregiver. Data analysis consisted of frequency and proportion calculations for ICD-10 categories, as well as t-tests to examine the relationship of medical comorbidities and ABC-I scores.

Results: The average age of the total sample was 12.9 years (SD=3.3, range 4-21), 79% were male, 79% Caucasian, and 93% non-Hispanic/non-Latino, with an average length of hospital stay of 25.6 days (SD=23.8, range 3-163). Sixty-percent of the sample (n=211) was diagnosed with at least one Axis III medical condition. Of those, 23.7% (n=83) had one diagnosis, 17.1% (n=60) had two, 10.0% (n=35) had three, 6.6% (n=23) had four, and 2.6% (n=9) with five diagnoses. Medical comorbidities were categorized by ICD-10 diagnostic groups. Of the 446 total diagnoses, the following categories were most prevalent: endocrine/nutritional/metabolic diseases (20.4%, n=91), diseases of the digestive system (14.8%, n=66), diseases of the respiratory system (11.7%, n=52), and diseases of the nervous system (11.4%, n=51) (see Figure 1). There were no significant differences in mean admission ABC-I scores between patients with any Axis III diagnosis (mean=27.96, SD=9.24) compared to those without (mean=26.47, SD=9.49), p=0.182. Further investigation by type of comorbidity (using ICD-10 categories with greater than 50 diagnoses in the sample as a cut-off) indicated subjects with at least one nervous system comorbidity (such as epilepsy, sleep disorder, or migraines) had significantly higher ABC-I scores (mean=29.98, SD=9.97) than those with no medical comorbidities (mean=26.47, SD=2.49), p=0.04.

Conclusions: Endocrine, gastrointestinal, respiratory, and nervous system disorders were found to be the most prevalent medical comorbidities in this inpatient sample. Behavioral severity for subjects with at least one nervous system problem was significantly higher on average than those with no medical problems, and merits further investigation. Targeting treatment of seizure and sleep disorders and other neurologic problems may play a role in decreasing the severity of problem behaviors.