CoChat
Sign Up
Cited by 867Autism Spectrum Disorder ResearchChild Nutrition and Feeding IssuesFamily and Disability Support Research

Prevalence and Early Identification of Autism Spectrum Disorder Among Children Aged 4 and 8 Years — Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022

Open access
Shaw, Kelly A., Williams, Susan, Patrick, Mary E., Valencia-Prado, Miguel, Durkin, Maureen S., Howerton, Ellen M., Ladd-Acosta, Christine M., Pas, Elise T. +44 more·2025·MMWR. Surveillance Summaries
DOI 10.15585/mmwr.ss7402a1PMC PMC12011386
Open this paper in CoChat — free accountOpen PDF

Abstract

Problem/Condition: Autism spectrum disorder (ASD). Period Covered: 2022. Description of System: The Autism and Developmental Disabilities Monitoring Network is an active surveillance program that estimates prevalence and characteristics of ASD and monitors timing of ASD identification among children aged 4 and 8 years. In 2022, a total of 16 sites (located in Arizona, Arkansas, California, Georgia, Indiana, Maryland, Minnesota, Missouri, New Jersey, Pennsylvania, Puerto Rico, Tennessee, Texas [two sites: Austin and Laredo], Utah, and Wisconsin) conducted surveillance for ASD among children aged 4 and 8 years and suspected ASD among children aged 4 years. Surveillance included children who lived in the surveillance area at any time during 2022. Children were classified as having ASD if they ever received 1) an ASD diagnostic statement in a comprehensive developmental evaluation, 2) autism special education eligibility, or 3) an ASD International Classification of Diseases, Ninth Revision (ICD-9) code in the 299 range or International Classification of Diseases, Tenth Revision (ICD-10) code of F84.0, F84.3, F84.5, F84.8, or F84.9. Children aged 4 years were classified as having suspected ASD if they did not meet the case definition for ASD but had an evaluator's suspicion of ASD documented in a comprehensive developmental evaluation. Results: Among children aged 8 years in 2022, ASD prevalence was 32.2 per 1,000 children (one in 31) across the 16 sites, ranging from 9.7 in Texas (Laredo) to 53.1 in California. The overall observed prevalence estimate was similar to estimates calculated using Bayesian hierarchical and random effects models. ASD was 3.4 times as prevalent among boys (49.2) than girls (14.3). Overall, ASD prevalence was lower among non-Hispanic White (White) children (27.7) than among Asian or Pacific Islander (A/PI) (38.2), American Indian or Alaska Native (AI/AN) (37.5), non-Hispanic Black or African American (Black) (36.6), Hispanic or Latino (Hispanic) (33.0), and multiracial children (31.9). No association was observed between ASD prevalence and neighborhood median household income (MHI) at 11 sites; higher ASD prevalence was associated with lower neighborhood MHI at five sites.Record abstraction was completed for 15 of the 16 sites for 8,613 children aged 8 years who met the ASD case definition. Of these 8,613 children, 68.4% had a documented diagnostic statement of ASD, 67.3% had a documented autism special education eligibility, and 68.9% had a documented ASD ICD-9 or ICD-10 code. All three elements of the ASD case definition were present for 34.6% of children aged 8 years with ASD.Among 5,292 (61.4% of 8,613) children aged 8 years with ASD with information on cognitive ability, 39.6% were classified as having an intellectual disability. Intellectual disability was present among 52.8% of Black, 50.0% of AI/AN, 43.9% of A/PI, 38.8% of Hispanic, 32.7% of White, and 31.2% of multiracial children with ASD. The median age of earliest known ASD diagnosis was 47 months and ranged from 36 months in California to 69.5 months in Texas (Laredo).Cumulative incidence of ASD diagnosis or eligibility by age 48 months was higher among children born in 2018 (aged 4 years in 2022) than children born in 2014 (aged 8 years in 2022) at 13 of the 15 sites that were able to abstract records. Overall cumulative incidence of ASD diagnosis or eligibility by age 48 months was 1.7 times as high among those born in 2018 compared with those born in 2014 and ranged from 1.4 times as high in Arizona and Georgia to 3.1 times as high in Puerto Rico. Among children aged 4 years, for every 10 children meeting the case definition of ASD, one child met the definition of suspected ASD.Children with ASD who were born in 2018 had more evaluations and identification during ages 0-4 years than children with ASD who were born in 2014 during the 0-4 years age window, with an interruption in the pattern in early 2020 coinciding with onset of the COVID-19 pandemic.Overall, 66.5% of children aged 8 years with ASD had a documented autism test. Use of autism tests varied widely across sites: 24.7% (New Jersey) to 93.5% (Puerto Rico) of children aged 8 years with ASD had a documented autism test in their records. The most common tests documented for children aged 8 years were the Autism Diagnostic Observation Schedule, Autism Spectrum Rating Scales, Childhood Autism Rating Scale, Gilliam Autism Rating Scale, and Social Responsiveness Scale. Interpretation: Prevalence of ASD among children aged 8 years was higher in 2022 than previous years. ASD prevalence was higher among A/PI, Black, and Hispanic children aged 8 years than White children aged 8 years, continuing a pattern first observed in 2020. A/PI, Black, and Hispanic children aged 8 years with ASD were also more likely than White or multiracial children with ASD to have a co-occurring intellectual disability. Identification by age 48 months was higher among children born in 2018 com

AI snapshot

Written from the paper’s full text.

TL;DR

In 2022, ASD prevalence among children aged 8 years across 16 ADDM Network sites was 32.2 per 1,000 (one in 31), with wide site variation and higher prevalence among several racial and ethnic groups than White children. Early identification by age 48 months was 1.7 times as high among children born in 2018 compared with those born in 2014.

Key findings

  • Among children aged 8 years, ASD prevalence was 32.2 per 1,000, ranging from 9.7 in Texas (Laredo) to 53.1 in California; prevalence was 49.2 per 1,000 among boys and 14.3 among girls (male-to-female ratio 3.4).
  • Prevalence was lower among White children (27.7) than A/PI (38.2), AI/AN (37.5), Black (36.6), Hispanic (33.0), and multiracial children (31.9).
  • Among 5,292 children aged 8 years with ASD and cognitive data, 39.6% had intellectual disability (IQ ≤70); median age of earliest known ASD diagnosis was 47 months.
  • Cumulative incidence of ASD diagnosis or eligibility by age 48 months was 1.7 times as high among children born in 2018 (22.6 per 1,000) compared with children born in 2014 (13.1 per 1,000); suspected ASD among children aged 4 years was 3.1 per 1,000, or one suspected case for every 10 children meeting ASD case definition.

Methods

  • Active surveillance in 16 sites in 2022 among children aged 4 and 8 years, using health and education records; ASD case definition required an ASD diagnostic statement, autism special education eligibility, or qualifying ICD-9/ICD-10 code, with suspected ASD defined for children aged 4 years.
  • Record abstraction was completed for 15 sites for 8,613 children aged 8 years who met the ASD case definition; analyses included prevalence by sex, race and ethnicity, neighborhood median household income, and social vulnerability index.
  • Prevalence was calculated per 1,000 children, with Bayesian hierarchical and random-effects models also used for overall estimates; cumulative incidence by age 48 months and monthly evaluation/identification patterns were compared between birth cohorts.

Limitations

  • The paper states that ADDM surveillance populations are not nationally representative, overall estimates do not capture full community variability, and data quality depends on availability and completeness of records, which vary by site and source.
  • The paper also states that sources reviewed generally do not include private schools or primary care providers, and that the surveillance definition of intellectual disability is not the same as a clinical diagnosis; IQ in young children may lack stability and adaptive behavior assessments are often absent.
  • Evident from the text: Indiana was excluded from record-abstraction analyses because ASD diagnostic statement data were unavailable, and several site-specific estimates were suppressed because relative standard error was ≥30%.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.15585/mmwr.ss7402a1

Cites

30 of 30 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 21 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    The most common tests documented for children aged 8 years were the Autism Diagnostic Observation Schedule, Autism Spectrum Rating Scales, Childhood Autism Rating Scale, Gilliam Autism Rating Scale, and Social Responsiveness Scale.

    “The most common tests documented for children aged 8 years were the Autism Diagnostic Observation Schedule, Autism Spectrum Rating Scales, Childhood Autism Rating Scale, Gilliam Autism Rating Scale, and Social Responsiveness Scale.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.
  • Findingown data

    ASD prevalence was higher among A/PI, Black, and Hispanic children aged 8 years than White children aged 8 years.

    “ASD prevalence was higher among A/PI, Black, and Hispanic children aged 8 years than White children aged 8 years, continuing a pattern first observed in 2020.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.
  • Findingown data

    Among children aged 8 years in 2022, ASD prevalence was 32.2 per 1,000 children (one in 31) across the 16 ADDM sites.

    “Among children aged 8 years in 2022, ASD prevalence was 32.2 per 1,000 children (one in 31) across the 16 sites, ranging from 9.7 in Texas (Laredo) to 53.1 in California.”
    • Also stated in: § ASD Prevalence Among Children Aged 8 Years · plain
      “The overall observed ASD prevalence was 32.2 per 1,000 (one in 31) children aged 8 years and ranged from 9.7 in Texas (Laredo) to 53.1 in California (Table 1).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.
    • Rests on an assumption: Children with ASD who are not captured by the available health and education records are negligible or do not substantially bias prevalence estimates.
  • Findingown data

    Overall ASD prevalence among children aged 8 years was lower among non-Hispanic White children (27.7 per 1,000) than among A/PI (38.2), AI/AN (37.5), Black (36.6), Hispanic (33.0), and multiracial children (31.9).

    “Overall, ASD prevalence was lower among non-Hispanic White (White) children (27.7) than among Asian or Pacific Islander (A/PI) (38.2), American Indian or Alaska Native (AI/AN) (37.5), non-Hispanic Black or African American (Black) (36.6), Hispanic or Latino (Hispanic) (33.0), and multiracial children (31.9).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.
  • Findingown data

    Children with ASD born in 2018 had more evaluations and identification during ages 0-4 years than children with ASD born in 2014 during the 0-4 years age window.

    “Children with ASD who were born in 2018 had more evaluations and identification during ages 0–4 years than children with ASD who were born in 2014 during the 0–4 years age window, with an interruption in the pattern in early 2020 coinciding with onset of the COVID-19 pandemic.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.
  • May combine two claims. 25
  • Classified as probably not a substantial claim. 13

  • May combine two claims.

    Findingown data

    Intellectual disability was present among 52.8% of Black, 50.0% of AI/AN, 43.9% of A/PI, 38.8% of Hispanic, 32.7% of White, and 31.2% of multiracial children with ASD.

    “Intellectual disability was present among 52.8% of Black, 50.0% of AI/AN, 43.9% of A/PI, 38.8% of Hispanic, 32.7% of White, and 31.2% of multiracial children with ASD.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.

  • May combine two claims.

    Findingown data

    Of the 8,613 children aged 8 years with ASD, 68.4% had a documented diagnostic statement of ASD, 67.3% had documented autism special education eligibility, and 68.9% had a documented ASD ICD-9 or ICD-10 code.

    “Of these 8,613 children, 68.4% had a documented diagnostic statement of ASD, 67.3% had a documented autism special education eligibility, and 68.9% had a documented ASD ICD-9 or ICD-10 code.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.

  • May combine two claims.

    Findingown data

    Variability was observed in where children were identified (health or education setting), the frequency of cognitive and IQ testing, the proportion of children with IQ ≤70, and the use of ASD diagnostic tests.

    “Variability was also observed in where children were identified (i.e., in a health or education setting), the frequency of cognitive and IQ testing, the proportion of children with IQ ≤70, and the use of ASD diagnostic tests.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Prevalence of suspected ASD was 3.1 per 1,000 children aged 4 years, which translated to one child suspected of having ASD for every 10 children identified with ASD among children aged 4 years.

    “Prevalence of suspected ASD was 3.1 per 1,000 children aged 4 years, which translated to one child suspected of having ASD for every 10 children identified with ASD among children aged 4 years”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Among children aged 4 years with ASD, the most common documented autism test was CARS (38.0%), followed by ADOS (31.7%), ASRS (21.1%), TELE-ASD-PEDS (8.7%), SRS (8.1%), GARS (6.2%), and ADI-R (3.8%).

    “The most common autism test documented among children aged 4 years with ASD was the CARS (38.0%), followed by the ADOS (31.7%), ASRS (21.1%), TELE-ASD-PEDS (8.7%), SRS (8.1%), GARS (6.2%), and ADI-R (3.8%).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The pattern of more evaluations and ASD identification among children born in 2018 resumed by June 2020 and varied by site.

    “The pattern of more evaluations and ASD identification among children born in 2018 resumed by June 2020 (Figure 5) and varied by site”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The Georgia site was the only site that met the threshold for statistical precision for AI/AN prevalence; its AI/AN prevalence per 1,000 was 204.5 (95% CI 111.5-345.0).

    “The Georgia site was the only site that met the threshold for statistical precision for AI/AN prevalence; the site-specific prevalence per 1,000 was 204.5 (111.5–345.0).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There were significant differences for Pearson chi-square tests for proportion of Black children with IQ ≤70 among children with ASD versus A/PI, Hispanic, multiracial, or White (each comparison p<0.001); A/PI children with IQ ≤70 versus multiracial or White (each comparison p<0.001); Hispanic children with IQ ≤70 versus multiracial (p=0.02) or White children (p<0.001).

    “Significant differences for Pearson chi-square tests for proportion of Black children with IQ ≤70 among children with ASD versus A/PI, Hispanic, multiracial, or White (each comparison p<0.001); A/PI children with IQ ≤70 among children with ASD versus multiracial or White (each comparison p<0.001); Hispanic children with IQ ≤70 among children with ASD versus multiracial (p = 0.02) or White children”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Among children aged 8 years with ASD in 2022, prevalence per 1,000 was 49.2 for males and 14.3 for females, with a male-to-female prevalence ratio of 3.4.

    “| Total | 49.2 (48.1–50.3) | 14.3 (13.7–15.0) | 3.4 (3.3–3.6)** |”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Similar percentages of girls and boys with ASD had data on cognitive ability (61.0% and 61.7%, respectively).

    “Similar percentages of girls and boys with ASD had data on cognitive ability (61.0 and 61.7%, respectively).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    This pattern was consistent at most (13 of 16) sites, ranging from 1.4 times as high in Arizona and Georgia to 3.1 times as high in Puerto Rico.

    “This pattern was consistent at most (13 of 16) sites, ranging from 1.4 times as high in Arizona and Georgia to 3.1 times as high in Puerto Rico.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Among children with ≥1 IEP available, the top five eligibility categories were autism (77.7%), speech or language impairment (24.7%), health, physical, or other disability (7.9%), developmental delay (6.9%), and intellectual disability (3.6%).

    “the top five eligibility categories were autism (77.7% of children); speech or language impairment (24.7%); health, physical, or other disability (7.9%); developmental delay (6.9%); and intellectual disability (3.6%).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Lower neighborhood MHI was associated with higher ASD prevalence for five sites (New Jersey, Tennessee, Texas [Laredo], Utah, and Wisconsin) and overall, although the overall trend was not monotonic.

    “lower neighborhood MHI was associated with higher ASD prevalence for five sites (New Jersey, Tennessee, Texas [Laredo], Utah, and Wisconsin) and overall (although the overall trend was not monotonic)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    ASD prevalence per 1,000 children aged 4 years in the 16 sites combined was 29.3, which was 0.9 times the overall prevalence among children aged 8 years in 2022.

    “ASD prevalence per 1,000 children aged 4 years in the 16 sites combined was 29.3, which was 0.9 times the overall prevalence among children aged 8 years in 2022”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The median age of most recent cognitive test was 67 months overall and varied by site (range 45 months in Texas [Austin] to 85 months in Puerto Rico).

    “The median age of most recent cognitive test was 67 months overall and varied by site (range = 45 months in Texas [Austin] to 85 months in Puerto Rico)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Higher social vulnerability was observed with higher ASD prevalence at five sites (Maryland, New Jersey, Tennessee, Utah, and Wisconsin) and overall, although overall, Maryland and Tennessee trends were not monotonic.

    “higher social vulnerability was observed with higher ASD prevalence at five sites (Maryland, New Jersey, Tennessee, Utah, and Wisconsin) and overall (although overall, Maryland and Tennessee trends were not monotonic)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    AI/AN-to-White prevalence ratios were significant for Georgia (8.5 [95% CI 4.7-15.5]) and overall (1.4 [95% CI 1.0-1.9]).

    “AI/AN-to-White ratios were significant for Georgia (8.5 [95% CI = 4.7–15.5]) and overall (1.4 [95% CI = 1.0–1.9]).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Among the 11 sites that conducted ASD surveillance in both 2020 and 2022, prevalence was higher in 2022 than in 2020 at nine sites (from 14.0% to 36.7% higher) and overall (22.2% higher; absolute difference of 6.1 more children with ASD per 1,000 children aged 8 years).

    “Among the 11 sites, prevalence was higher in 2022 than in 2020 at nine (from 14.0% to 36.7% higher) and overall (22.2% higher [absolute difference of 6.1 more children with ASD per 1,000 children aged 8 years]).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.
    • Rests on an assumption: Children with ASD who are not captured by the available health and education records are negligible or do not substantially bias prevalence estimates.

  • May combine two claims.

    Findingreasoning

    Puerto Rico was the site with second-highest prevalence among children aged 4 years, but prevalence among children aged 8 years in Puerto Rico was below the ADDM Network average.

    “Puerto Rico was the site with second-highest prevalence among children aged 4 years, but prevalence among children aged 8 years in Puerto Rico was below the ADDM Network average.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The American Academy of Pediatrics recommends that pediatric care providers screen all children for ASD at ages 18 and 24 months.

    “the American Academy of Pediatrics recommends that pediatric care providers screen all children for ASD at ages 18 and 24 months (2,3).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [2] Hyman, 2020, Identification, evaluation, and management of children with autism spectrum disorder., Pediatrics source

      Not checked in the free run.

    • cites [3] Lipkin, 2020, Promoting optimal development: identifying infants and young children with developmental disorders through developmental surveillance and screening., Pediatrics source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Before 2016, the highest ASD prevalence was observed among White children and in children from neighborhoods with higher socioeconomic status.

    “Before 2016, the highest ASD prevalence was observed among White children and in children from neighborhoods with higher socioeconomic status (SES) (7).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [7] Durkin, 2017, Autism spectrum disorder among US children (2002–2010): socioeconomic, racial, and ethnic disparities., Am J Public Health source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Although prevalence has been lower among children aged 4 years than aged 8 years in each reporting year, comparing cumulative incidence of identification by age 48 months has consistently indicated higher rates of ASD identification in younger cohorts.

    “Although prevalence has been lower among children aged 4 years than aged 8 years in each reporting year, comparing cumulative incidence of identification by age 48 months has consistently indicated higher rates of ASD identification in younger cohorts (i.e., children born more recently) (9–11).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [9] Shaw, 2020, Early identification of autism spectrum disorder among children aged 4 years—Early Autism and Developmental Disabilities Monitoring Network, six sites, United States, 2016., MMWR Surveill Summ source

      Not checked in the free run.

    • cites [10] Shaw, 2021, Early identification of autism spectrum disorder among children aged 4 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2018., MMWR Surveill Summ source

      Not checked in the free run.

    • cites [11] Shaw, 2023, Early identification of autism spectrum disorder among children aged 4 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2020., MMWR Surveill Summ source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A report that presented data collected in 2020 found interruptions to early identification during the onset of the COVID-19 pandemic.

    “A report that presented data collected in 2020 found interruptions to early identification during the onset of the COVID-19 pandemic (11).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [11] Shaw, 2023, Early identification of autism spectrum disorder among children aged 4 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2020., MMWR Surveill Summ source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Record abstraction was completed for 15 of the 16 sites for 8,613 children aged 8 years who met the ASD case definition.

    “Record abstraction was completed for 15 of the 16 sites for 8,613 children aged 8 years who met the ASD case definition.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    Early identification of ASD can help children receive services and supports they might need for their development and to improve long-term outcomes.

    “Early identification of ASD can help children receive services and supports they might need for their development and to improve long-term outcomes (4).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] Zwaigenbaum, 2015, Early intervention for children with autism spectrum disorder under 3 years of age: recommendations for practice and research., Pediatrics source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    A Bayesian hierarchical modelling approach and random-effects restricted maximum likelihood method were used to calculate overall prevalence and measures of uncertainty among children aged 8 years to better account for the wide range in prevalence across sites.

    “A Bayesian hierarchical modelling approach and random-effects restricted maximum likelihood method were also used to calculate overall prevalence and measures of uncertainty (95% Bayesian credible intervals and 95% CIs, respectively) among children aged 8 years to better account for the wide range in prevalence across sites; full details are available (Supplementary Methods 2, https://stacks.cdc.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisenone given

    California has a local initiative (the Get SET Early model).

    “and has a local initiative (the Get SET Early model).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Thirteen sites had data agreements with education sources covering 100% of their study areas; three sites had agreements with education data sources covering less than 100% (Georgia 97.3%, Missouri 68.0%, and Texas [Laredo] 99.7%).

    “Thirteen sites had data agreements in place with education sources covering 100% of their study areas; three sites had agreements with education data sources covering <100% (Georgia [97.3%], Missouri [68.0%], and Texas [Laredo] [99.7%]).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Indiana is included in analyses of prevalence by demographic characteristics, MHI, and SVI but excluded from other analyses that required data from record abstraction.

    “Because Indiana did not have data available from record abstraction, Indiana is included in analyses of prevalence (by demographic characteristics, MHI, and SVI) but excluded from other analyses that required data from record abstraction (e.g., co-occurring intellectual disability or presence of evaluation or diagnosis or age at evaluation or diagnosis).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    California has regional centers throughout the state that provide evaluations and service coordination for persons with disabilities and their families.

    “In addition, California has regional centers throughout the state that provide evaluations and service coordination for persons with disabilities and their families (https://www.dds.ca.gov).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites https://www.dds.ca.gov

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning

    Variability in identification practices could affect receipt of services and suggests opportunities to implement successful identification strategies to ensure children in all communities receive the diagnostic and care services they need.

    “Variability in identification practices could affect receipt of services and suggests opportunities to implement successful identification strategies to ensure children in all communities receive the diagnostic and care services they need.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Prevalence estimates with a relative standard error of 30% or greater were suppressed because of limited statistical precision.

    “Prevalence estimates with a relative standard error ≥30% were suppressed because of limited statistical precision.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    All sites participating in 2022 had access to health and education records.

    “All sites participating in 2022 had access to health and education records (Table 1).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    As part of the Get SET Early initiative, hundreds of local pediatricians have been trained to screen and refer children for assessment as early as possible, which could result in higher identification of ASD, especially at early ages.

    “As part of the initiative, hundreds of local pediatricians have been trained to screen and refer children for assessment as early as possible, which could result in higher identification of ASD, especially at early ages (17).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [17] Pierce, 2021, Get SET Early to identify and treatment refer autism spectrum disorder at 1 year and discover factors that influence early diagnosis., J Pediatr source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    Another reason for differences in prevalence could be whether children have insurance coverage or meet eligibility criteria for access to early intervention services.

    “Another reason for differences in prevalence could be whether children have insurance coverage or meet eligibility criteria for access to early intervention services.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Pennsylvania, the site with second highest prevalence among children aged 8 years, has state Medicaid policy that includes children with physical, developmental, mental health, or intellectual disabilities regardless of parents' income.

    “Pennsylvania, the site with second highest prevalence among children aged 8 years, has state Medicaid policy that includes children with physical, developmental, mental health, or intellectual disabilities regardless of parents’ income (18).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [18] Commonwealth of Pennsylvania, Department of Human Services. Medicaid for children with special needs. Harrisburg, PA: Commonwealth of Pennsylvania, Department of Human Services; 2024. https://www.pa.gov/agencies/dhs/resources/medicaid/medicaid-children-special-needs-ph95.html

      Not checked in the free run.

36 of 36

  • The paper relies on this without support.

    The surveillance ASD case definition accurately identifies children with ASD in the surveillance population.

    “children aged 8 years (born in 2014) or aged 4 years (born in 2018) met the surveillance ASD case definition if they lived in the surveillance area at any time in 2022 and they received 1) an ASD diagnostic statement in a comprehensive developmental evaluation, 2) autism special education eligibility, or 3) an ASD ICD-9 code in the 299 range or ICD-10 code of F84.0, F84.3, F84.5, F84.8, or F84.9.”

  • The paper relies on this without support.

    Children with ASD who are not captured by the available health and education records are negligible or do not substantially bias prevalence estimates.

    “All sites participating in 2022 had access to health and education records (Table 1).”

  • The authors state this as an assumption.

    The methodology assumes that all Puerto Rico residents are Hispanic.

    “This methodology assumes that all Puerto Rico residents are Hispanic.”

  • The paper relies on this without support.

    Research has not demonstrated that living in certain communities puts children at greater risk for developing ASD.

    “Research has not demonstrated that living in certain communities puts children at greater risk for developing ASD.”
4 of 4

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Maximum identification increase is misstated as 300%

    Minor

    The reported maximum ratio of 3.1 times corresponds to 210% higher, not 300% higher. A 300% increase would be four times the earlier rate.

  • Conflicting percentages for children with all three ASD identification elements

    Minor

    The abstract reports 34.6%, whereas the Results section reports 34.7% for the same group of children across the 15 sites.

2 of 2

Cites

Flagged first, then by citations

1 of the 31 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. Institute of Medicine (US) Committee on Understanding Premature Birth and Assuring Healthy Outcomes. Behrman RE, Butler AS, eds. Sociodemographic and community factors contributing to preterm birth [Chapter 4]. In: Preterm birth: causes, consequences, and prevention. Washington, DC: National Academies Press (US); 2007:124–47.

Cited by

Sorted by citations

Related

Authors
Shaw, K. A., Williams, S., Patrick, M. E., Valencia-Prado, M., Durkin, M. S., Howerton, E. M., Ladd-Acosta, C. M., Pas, E. T., Bakian, A. V., Bartholomew, P., Nieves-Muñoz, N., Sidwell, K., Alford, A., Bilder, D. A., DiRienzo, M., Fitzgerald, R. T., Furnier, S. M., Hudson, A. E., Pokoski, O. M., Shea, L., Tinker, S. C., Warren, Z., Zahorodny, W., Agosto-Rosa, H., Anbar, J., Chavez, K. Y., Esler, A., Forkner, A., Grzybowski, A., Agib, A. H., Hallas, L., Lopez, M., Magaña, S., Nguyen, R. H. N., Parker, J., Pierce, K., Protho, T., Torres, H., Vanegas, S. B., Vehorn, A., Zhang, M., Andrews, J., Greer, F., Hall-Lande, J., McArthur, D., Mitamura, M., Montes, A. J., Pettygrove, S., Shenouda, J., Skowyra, C., Washington, A., & Maenner, M. J.
Affiliations
  • Centers for Disease Control and Prevention (US)
  • University of Wisconsin–Madison (US)
  • Johns Hopkins University (US)
  • University of Utah (US)
  • University of California San Diego (US)
  • National Center on Birth Defects and Developmental Disabilities (US)
  • Rutgers New Jersey Medical School (US)
  • Pennsylvania Department of Human Services (US)
  • Washington University in St. Louis (US)
  • University of Arkansas for Medical Sciences (US)
  • Vanderbilt University Medical Center (US)
  • Puerto Rico Department of Health (PR)
  • University of Minnesota (US)
  • The University of Texas at Austin (US)
  • Arizona State University (US)
  • University of Arizona (US)
Corresponding author
Shaw, Kelly A.
ORCID
Venue
MMWR. Surveillance Summaries
Publisher
Centers for Disease Control MMWR Office
Published
2025-04-17
Volume
74
Issue
2
Pages
1-22
ISSN
  • 1545-8636 (ISSN-L)
  • 1546-0738
DOI
10.15585/mmwr.ss7402a1
PMC
PMC12011386
OpenAlex
W4409458169
Semantic Scholar
ce82e952834c1f4b32c7226767552851cc1f0430
Cited by
867
References
31
Citations by year
2026: 615 · 2025: 252
FWCI
491.83
Type
journal article
Language
en
Open access
Licence
public-domain
Locations
  • MMWR Surveillance Summaries (publishedVersion, public-domain)
  • PubMed Central (publishedVersion, public-domain)
  • PubMed (publishedVersion)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Autism Spectrum Disorder Research
  • Health Sciences › Medicine › Psychiatry and Mental health › Child Nutrition and Feeding Issues
  • Social Sciences › Psychology › Clinical Psychology › Family and Disability Support Research
Keywords
autism spectrum disorder, autism prevalence, sex differences, racial and ethnic disparities, ASD subtypes, developmental disabilities, prevalence, early detection, autism screening
MeSH terms
Autism Spectrum Disorder, Child, Child, Preschool, Female, Humans, Male, Population Surveillance, United States, Prevalence, Early Diagnosis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-06.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-06.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review