How many children are identified with autism, at what age, and how well do screening tools work? This page explains the CDC’s surveillance report and NIH-funded studies of digital and clinic-based screening.
Based on CDC MMWR Surveillance Summaries · CDC · NIH media advisory · NIMH · All sources · Checked November 3, 2026
87.8%sensitivity of the SenseToKnow tablet app among 475 toddlers in a research studyNIH · October 2, 2023
Orientation: counting and screening are different jobs
CDC’s ADDM Network counts children already identified with autism in monitored communities. Screening research asks a different question: how well a tool or process picks out toddlers who need a full evaluation. Both matter for families, and both come with limits — surveillance areas are not nationally representative, and a screening result is never a diagnosis. How screening works.
April 17, 2025 · MMWR Surveillance Summaries (CDC) · Surveillance report, 2022 data
1 in 31 eight-year-olds identified with autism across 16 ADDM sites; median age at earliest diagnosis 47 months
What they studied: CDC’s ADDM Network reviewed health and education records for children aged 4 and 8 in 16 U.S. communities for surveillance year 2022.
What they found: Among 8-year-olds, 32.2 per 1,000 children (about 1 in 31) had been identified with autism, ranging from 9.7 per 1,000 in Laredo, Texas, to 53.1 in California. Among 4-year-olds the figure was 29.3 per 1,000. Autism was 3.4 times as common in boys as in girls. The median age at earliest known diagnosis was 47 months, ranging from 36 months (California) to 69.5 months (Laredo).
Why it matters / limits: The report states that the surveillance areas are “not nationally representative”, that data quality depends on the availability and completeness of records, and that private schools and primary care providers are generally not included. The figures describe identification in those communities — not a national count, and not causes.
Read the original source October 2, 2023 · NIH media advisory · Nature Medicine · Screening tool validation
A tablet app, SenseToKnow, identified most toddlers later diagnosed with autism
What they studied: In an NICHD-funded study at Duke’s Autism Center of Excellence, health care providers used the app to screen 475 toddlers aged 17 months to 3 years during well-child visits. The app records a child’s responses to short movies — social attention, facial expressions, head movements, response to name, blink rate and motor skills.
What they found: Of the 475 toddlers, 49 were later diagnosed with autism and 98 with developmental delay without autism. The app showed 87.8% sensitivity and 80.8% specificity, and its accuracy was consistent across sex, race and ethnicity. A positive app screen carried a 40.6% probability of a later diagnosis, versus about 15% for the standard parent questionnaire; combining the two raised it to 63.4%.
Why it matters / limits: NIMH’s 2024 summary of the same study notes the researchers “do not intend for SenseToKnow to be the only data source for diagnosis”, and that more work is needed to establish performance across diverse groups and in home use. It is a research tool, not something families can obtain today.
Read the original source April 21, 2022 · NIMH research highlight · JAMA Pediatrics · Implementation study
Adding structured screening to early-intervention sites raised autism diagnoses, most for Spanish-speaking families
What they studied: NIMH-funded researchers implemented a three-stage screening and diagnosis process — two questionnaires, an observational screening tool, then a psychologist-led evaluation — at three federally funded early-intervention sites in Boston, and compared them with nine similar sites. The sample included 33,326 children aged 14 to 36 months with no prior autism diagnosis.
What they found: After implementation, the study sites showed an additional 8.1 autism diagnoses per 1,000 children per quarter. The increase among Spanish-speaking families (15.4 per 1,000 per quarter) was more than double that among non-Spanish-speaking families (6.9). Boys were more likely than girls to be screened and to screen positive.
Why it matters / limits: The authors say further research is needed to identify which elements of the procedure are necessary, given the cost and effort of a multisite, multistage program. The study shows what one state’s early-intervention system achieved; it is not a tool families can request.
Read the original source April 26, 2021 · NIH media advisory · JAMA Pediatrics · Prototype evaluation
A prototype eye-tracking app distinguished toddlers with autism in a primary-care sample
What they studied: NIH-funded researchers enrolled 933 toddlers aged 16 to 38 months during well-child visits; the children watched short videos of people smiling, making eye contact or talking, while the device camera recorded their gaze for computer-vision analysis.
What they found: The 40 children later diagnosed with autism were much less likely than typically developing children to focus on social cues and to visually track the conversations.
Why it matters / limits: The authors concluded the approach “could be” a viable method, pending confirmation by larger studies. A prototype in a research setting is not a screening tool in use.
Research explained here is educational. A single study — however large or well designed — does not change clinical care on its own; findings need to be repeated by other teams before they shape guidelines. Surveillance figures describe the communities that were monitored, not a national count, and they do not show causes. Laboratory models are not tests or treatments. Funding announcements describe what will be studied, not what has been found. If something here raises a question about your child or yourself, take it to your health care provider or care team.
If you have a concern about your child, you do not need to wait for new tools: ask your child’s provider for a developmental screening at any visit, or contact your state’s early-intervention program directly. How to get an evaluation.
This website provides general educational information. It does not diagnose autism or replace individualized medical, developmental, psychological, or educational advice. Screening is not diagnosis.
External listings do not imply affiliation, endorsement, or guaranteed services. Eligibility, availability, costs, and insurance coverage may change. Research summaries reflect evidence available at the time of review.
