Retrospective analysis of treated prevalence, psychiatric comorbidity prevalence, and healthcare utilization and expenditures in commercially insured children diagnosed with autism spectrum disorder
Autism spectrum disorders (ASDs) are the leading cause of disability in children under five, and affect 1 in 59 children in the United States. The aims of this study were to explain geographic variation in the treated prevalence of ASD, compare the prevalence of selected psychiatric conditions between ASD and non-ASD patients, and to estimate incremental healthcare utilization and expenditures. A commercial claims database was utilized to address all aims. Prevalence rates were generated for 819 ZIP3 regions, and a generalized linear model (GLM) was used to explain geographic variation in prevalence. Aims two and three were addressed in a separate cohort of ASD patients (n=17,787) matched to a non-ASD control group (n=35,574). Three sets of GLMs were fit to compare age-related prevalence rates of psychiatric conditions, utilization, and expenditures between ASD and non-ASD groups. The overall prevalence of ASD was 6.84 per 1,000 children 3-18 years. Significant positive relationships with prevalence were identified for fully-insured plan density (p=0.018), median income (p<0.001), and private school enrollment density (p=0.015); and significant negative relationships were observed for percent white (p=0.012), and urbanicity (p=0.003). Stronger positive age-prevalence relationships were identified for anxiety (p<0.001), conduct and behavior (p<0.001), mood (p<0.001), and psychotic (p=0.003) disorders in ASD patients compared to non-ASD patients. Greater all-cause utilization was identified in the ASD group for office, occupational/physical therapy (OT/PT), speech therapy, emergency department (ED), and prescriptions; and greater mental health-related (MHR) utilization was identified for outpatient office visits, ED visits, and prescriptions (all p<0.001). Greater all-cause expenditures were identified for total, outpatient office, OT/PT, speech therapy, ED, prescriptions, and OOP; and greater MHR expenditures were identified for total, outpatient office, inpatient, ED, prescription, and OOP (all p<0.05). In conclusion, geographic disparities in prevalence rates of ASD appear in commercially-enrolled children, and many psychiatric conditions manifest more strongly with respect to age in ASD patients when compared to rates in non-ASD patients. Healthcare utilization and expenditures are also greater for ASD patients. In summary, this study provides novel insight into at least three facets of the ASD patient journey, and implications for patients, providers, and health plans.