Search This Blog

Wednesday, August 16, 2023

Inpatient Discharge Data Confirm Increase in Autism Prevalence

In The Politics of Autism, I discuss evaluationdiagnosis, and the uncertainty of prevalence estimates.

"Epidemiologic Patterns of Autism Spectrum Disorder in Pediatric Inpatients in the United States, 1997–2019," by Stanford ChihuriAshley BlanchardCarolyn G DiGuiseppiGuohua Li

Journal of Autism and Developmental Disorders (2023) Cite this article

The reported prevalence of autism spectrum disorder (ASD) has more than tripled in the past two decades in the United States, due in part to improved screening and diagnostic techniques. Epidemiologic data on ASD, however, are largely limited to population-based surveillance systems. We examined epidemiologic patterns in ASD diagnoses among inpatients aged 1–20 years, using data from the Kids’ Inpatient Database (KID) from 1997 to 2019. ASD cases were identified using ICD-9-CM and ICD-10-CM codes. Of 9,267,881 hospital discharges studied, 110,090 (1.19%) had a diagnosis of ASD. The prevalence of ASD was higher among males compared to females (1.53% vs. 0.54%) and was highest among non-Hispanic Whites (1.28% vs. 0.95% in non-Hispanic Blacks, 0.94% in Hispanics, and 1.18% in Other races). ASD prevalence increased from 0.18% to 1997 to 3.36% in 2019 (Z= -273.40, p < 0.001). The absolute increase was higher among males compared to females (0.26–4.90% vs. 0.08–1.77%) and among non-Hispanic Whites (0.18–2.88%) compared to non-Hispanic Blacks (0.23–2.72%), Hispanics (0.14–2.60%), and Other races (0.19–2.97%). The epidemiologic patterns of ASD based on inpatient data are generally consistent with reports from the community-based autism surveillance system. Our findings indicate that KID and other health services data might play a complementary role in ASD surveillance.


In a national study of pediatric inpatient discharges, there was a marked increase in ASD prevalence, with more rapid increase among younger birth cohorts, males, and non-Hispanic Whites. The prevalence of ASD followed a similar pattern to the ADDM Network- reported prevalence over time. Prevalence estimates for all age groups are important for informing public policy, allocating resources and services, and raising awareness. Our findings demonstrate the potential utility of hospital discharge record data for ASD surveillance.