Predictors of Emergency Department Usage Among Children Ever Diagnosed with Autism Spectrum Disorder

Background:  Children with autism spectrum disorder (ASD) have high rates of healthcare utilization (Kogan et al., 2008), potentially a product of the high co-occurrence of ASD and other medical and psychiatric conditions (Levy et al., 2010).  Many families of children with ASD have difficulty accessing specialty medical and related services (Warfield & Gulley, 2006) and may depend on emergency department services.  Previous research has indicated an increased use of the emergency department for psychiatric crises (Kalb et al., 2012) and injury (McDermott, Zhou, & Mann, 2008) among children with ASD, but little is known about predictors of emergency department services.  This study explores predictors and parent-reported reasons of emergency room (ER) visits among a national population of children with ASD.

Objectives: Compare the likelihood of any ER visits in the last year between children with and without ASD. Examine reported reasons for last ER visit.

Methods: Data come from the 2011-2014 National Health Interview Survey (NHIS).  NHIS is a nationally representative household survey that collects data continuously throughout the year of the noninstitutionalized US population. Respondents (usually the parent) answer on behalf of one randomly selected child in the family. Children aged 2-17 years were included in this analysis (n=46,316), which included children ever diagnosed with ASD (n=689) and children never diagnosed with ASD (n=45,602). 

Estimates were calculated using SUDAAN to account for the NHIS complex sample design. Logistic regressions were used to examine the association between ASD and any ER visits in the past year.  Model 0 was an unadjusted model, Model 1 adjusted for survey year, child and family demographics, and Model 2 additionally adjusted for co-occurring child conditions (seizures, food/digestive allergies, asthma, attention-deficit/hyperactivity disorder, frequent diarrhea/colitis), as well as survey year and demographics. 

Among those who had any ER visits in the past year, parents endorsed multiple reasons (up to eight) for the child’s most recent ER visit. Children were assigned to mutually exclusive categories summarizing their reason for their last ER visit; seriousness of the medical problem or access to healthcare. If a parent endorsed both a seriousness and access reason, the reason was coded as being due to the seriousness of the medical problem.

Results:  Children ever diagnosed with ASD (23%) were significantly more likely than children without ASD (17%) to have had an ER visit in the past year (p<0.05) (Model 0).  This association remained significant after adjusting for survey year and demographics (Model 1). However, after adjusting for co-occurring child conditions (Model 2) this association was no longer significant. Approximately 70% of parents of children ever diagnosed with ASD reported that the most recent ER visit was due to the seriousness of the medical problem and approximately 27% of parents reported reasons only related to healthcare access.

Conclusions: Differences between ER use of children with and without ASD may be explained by differences in the presence of co-occurring medical and psychiatric conditions. Parents of children with ASD were most likely to report their child’s most recent ER visit was due to the seriousness of the medical problem.