BOSTON, April 24, 2026 – A new study examined geographic patterns in autism diagnosis by primary care providers among Medicaid-enrolled children in 29 states. The study found that Nevada, South Carolina and Connecticut had the highest rates of autism diagnosis by primary care providers. Findings from the study will be presented at the Pediatric Academic Societies (PAS) 2026 Meeting , taking place April 24-27 in Boston. This study is part of a larger research project, Addressing Structural Disparities in Autism Spectrum Disorder through Analysis of Secondary Data (ASD3) , which is funded by the National Institutes of Health.
An autism spectrum disorder (ASD) diagnosis by a medical provider is often required for disability and therapy services access. Given long wait times and limited access to specialty care providers who typically diagnose ASD, some states have expanded primary care provider ASD diagnostic training. Yet little research has examined geographic patterns in primary care providers ASD diagnosis.
"Autism is typically diagnosed by specialists, but various factors such as long wait times can delay this process,” said the ASD3 investigative team. “Primary care providers can offer an additional pathway to early diagnosis, which is important for improving access to needed autism services and supports. Our study looked at Medicaid-enrolled children ages 1-5 in 29 states between 2017-2019, and found that 29% were diagnosed by primary care providers rather than specialists. Rates of autism diagnosis by primary care providers varied significantly between counties, states and regions. We believe that local practice patterns and specialist availability may contribute to this geographic variation."
The sample included 36,263 children across 933 county-sets in 29 states. Overall, 29% of children were diagnosed with ASD by a primary care provider. Regional percentages varied from a median of 20.0% among county-sets in the Midwest to 36.4% in the West. Within-region variation was also high, with interquartile ranges (IQRs) varying from 23.8% in the Midwest to 27.5% in the West. Within-region variation was also high when grouping county-sets by state. For example, the Northeast had both the third highest state (Connecticut, 53.1%) and the third lowest state (New Hampshire, 14.3%) as measured by their median county-set. The neighboring states of South Carolina (60.4%) and Georgia (17.9%), both in the South region, ranked second highest and fifth lowest, respectively. Within-state heterogeneity (IQR) was as low as 6.2% (Connecticut), 9.6% (New Hampshire), and 12.7% (Georgia) and as high as 35.7% (Colorado), 37.6% (Illinois), and 42.3% (Wyoming). Population density showed a slight trend with the densest quartile of county-sets having a median primary care provider diagnosis rate of 24.2%, and 28.6% for the least dense quartile.
Additional information is included in the below research abstract. The PAS Meeting connects thousands of leading pediatric researchers, clinicians and educators worldwide. View the full schedule in the PAS 2026 program guide . For more information about the PAS Meeting, please visit www.pas-meeting.org .
###
About the Pediatric Academic Societies (PAS) Meeting
The Pediatric Academic Societies (PAS) Meeting connects thousands of leading pediatric researchers, clinicians, and educators worldwide, united by a shared mission: connecting the global academic pediatric community to advance scientific discovery and promote innovation in child and adolescent health. PAS is a partnership of four premier pediatric associations: the American Academy of Pediatrics (AAP) , the Academic Pediatric Association (APA) , the American Pediatric Society (APS), and the Society for Pediatric Research (SPR) . For more information, visit www.pas-meeting.org . Follow us on X , Facebook, and Instagram .
---
Abstract: Autism Diagnosis by Primary Care Providers: Geographic Variation Among Medicaid-Enrolled Children in 29 States
Presenting Author
Corey A. Gorgas, MA, Research Coordinator, Northern Arizona University
Organizations
Northern Arizona University
Oregon Health & Science University
SRI International
Topic
Developmental and Behavioral Pediatrics: Autism
Background
An autism spectrum disorder (ASD) diagnosis by a medical provider is often required for disability and therapy services access. Given long wait times and limited access to specialty care providers who typically diagnose ASD, some states have expanded primary care provider (PCP) ASD diagnostic training. Yet little research has examined geographic patterns in PCP ASD diagnosis.
Objective
Describe geographic variation in PCP ASD diagnosis among Medicaid-enrolled children.
Design/Methods
We used Transformed Medicaid Statistical Information System Analytic Files (TAF) to analyze the percentage of ASD diagnoses made by PCPs (versus specialists) at the county-set-level (contiguous groupings of counties containing at least 50,000 residents) from 1-1-2017 to 6-30-2019. Eligible children were 1 to 5 years old at the time of ASD diagnosis, and were Medicaid-enrolled for at least 11 of the 12 months prior with no ASD diagnoses. We required an additional ASD diagnosis within 6 months of the initial PCP diagnosis to reduce misclassification.
Results
The sample included 36,263 children across 933 county-sets in 29 states. Overall, 29% of children were diagnosed with ASD by a PCP. Regional percentages varied from a median of 20.0% among county-sets in the Midwest to 36.4% in the West. Within-region variation was also high, with interquartile ranges (IQRs) varying from 23.8% in the Midwest to 27.5% in the West. Within-region variation was also high when grouping county-sets by state. For example, the Northeast had both the third highest state (Connecticut, 53.1%) and the third lowest state (New Hampshire, 14.3%) as measured by their median county-set. The neighboring states of South Carolina (60.4%) and Georgia (17.9%), both in the South region, ranked second highest and fifth lowest, respectively. Within-state heterogeneity (IQR) was as low as 6.2% (Connecticut), 9.6% (New Hampshire), and 12.7% (Georgia) and as high as 35.7% (Colorado), 37.6% (Illinois), and 42.3% (Wyoming). Population density showed a slight trend with the densest quartile of county-sets having a median PCP diagnosis rate of 24.2%, and 28.6% for the least dense quartile.
Conclusion(s)
Results demonstrate that PCP ASD diagnosis is geographically widespread and already occurring outside of major medical centers and established training programs. Despite clear variations when grouping county-sets by region and state, the often-high IQRs hint that local practice patterns may play a large role. The weak trend of increasing PCP diagnosis as population density decreases additionally provides some evidence of limited rural specialty care access.
Co-Authors
David C. Folch, PhD, Associate Professor, Northern Arizona University
Christina Charlesworth, MPH, Principal Health Services Researcher, Oregon Health & Science University School of Medicine
Kelsey Watson, MPH, Oregon Health & Science University School of Medicine
Vincent Sutherland, MS, Research Assistant, Northern Arizona University
Veronica I. Underwood Carrasco, Senior Clinical Research Assistant, Oregon Health & Science University
Adrienne D. Woods, PhD, MS, Senior Education Researcher, SRI International
Katharine E. Zuckerman, MD MPH, Professor, Oregon Health & Science University School of Medicine
Olivia J. Lindly, PhD, Associate Professor, Northern Arizona University, Health Sciences Department
Tables and Images
Figure 1. Percent of Medicaid-enrolled children diagnosed with ASD between ages 1-5 whose diagnosis was made by a PCP.
Percentages calculated by county-set, from a secondary analysis of 2017-2019 Transformed Medicaid Statistical Information System Analytic Files.
Figure 2. Boxplots of county-set ASD PCP diagnosis percentages grouped by state.
Results are ordered by the median value within each state; dot size reflects the number of diagnoses.