Increased Emergency Department Use for Mental Health Problems Among Children with Autism Spectrum Disorders: A Population-Based Study

Background: Little is known about the use of emergency department (ED) services among children with an Autism Spectrum Disorder (ASD). Given the high rates of psychopathology in this population, coupled with a severe shortage of child and adolescent psychiatrists in the United States, these children may be at an increased risk of utilizing ED services for mental health problems (Gabel & Sarvet, 2011; Simonoff et al., 2008).  To date, no line of research has explored this question, especially in a nationally representative sample.

Objectives: 1) To compare the prevalence of ED visits between children with and without ASD where the primary indication of the visit was a psychiatric concern; and 2) to examine the number and type of psychiatric diagnoses in mental health related ED visits by children with ASD.

Methods: Pediatric data from the 2008 National Emergency Department Sample (NEDS), the largest all-payer ED database, were examined. The weighted study sample, which represents all ED visits in the United States, consisted of 17,517,397 visits for children ages 3 to 17 years (M = 10.0; SD = 4.72 years). All diagnoses in the NEDS are coded using the ICD-9-CM. ASD-related visits were identified using 299.XX. ED visits were considered primarily psychiatric in nature if the principal diagnosis represents a DSM-IV-TR Axis I psychiatric disorder. If the primary diagnosis was an ASD (n=418), the secondary diagnosis was used. A survey weighted logistic regression model, which controlled for child, ED visit, and hospital-related characteristics, was used to examine if ED visits among children with ASD were more likely to be for psychiatric reasons compared to visits by children without ASD. Among these psychiatrically-related ED visits, a similar, albeit poisson, model was used to determine if visits by children with and without ASD differed in the number of psychiatric comorbidities reported.

Results: Children with ASD had a total of 59,187 ED visits (weighted prevalence of .003%).  Of all ASD-related visits, 13% were related to a primary psychiatric concern, as compared to 2% of all visits by children without ASD. After adjustment for socio-demographic variables, visits by children with ASD were over 8 times more likely to be for psychiatric reasons compared to visits by children without ASD (OR: 8.38, 95%: 7.89 – 8.89). Psychiatric visits by children with ASD involved a greater number of psychiatric diagnoses compared to similar visits of children without ASD (β = .09, 95%: .07 -.12) (all p < .001). The most common principal diagnoses reported among these visits in children with ASD were behavioral (43%), mood (37%), anxiety (11%), and psychotic (8%) disorders as well as suicide and intentional self-inflicted injury (1%).

Conclusions: This is the first study to demonstrate high rates of ED use for psychiatric evaluation in children with ASD. Consistent with the ASD psychiatric comorbidity literature, children with ASD visiting the ED for mental health problems were more psychiatrically complex than their peers. These findings underscore the acuity of psychiatric problems seen in this population and the urgent need for stronger community-based mental health resources.