Every two years, the CDC publishes its Autism and Developmental Disabilities Monitoring Network (ADDM) prevalence report. This is the dominant source for autism statistics in the United States. The headline numbers — "1 in 31," "1 in 36," and earlier — drive most autism public-discussion framing. Most people who quote these numbers do not understand what they actually measure. Here is how to read the ADDM report without getting manipulated.

What ADDM actually is

ADDM is a multi-site surveillance system. The CDC funds a small number of sites (currently 16, varying by year) across the United States to identify 8-year-olds with autism diagnoses in their medical, educational, and clinical records. The prevalence number is derived from those sites and extrapolated as a national estimate.

What "1 in 31" actually measures

The 2025 ADDM release reported 1 in 31 8-year-olds identified with autism in the surveillance year. This is the measured prevalence rate at age 8 across the ADDM monitoring sites. It is not the national autism rate at all ages. It is not the lifetime prevalence. It is not the rate at which autism is increasing in the population — it is the rate at which autism is being identified by the systems ADDM measures.

The three patterns the report reveals

  • Identification rates have climbed dramatically. Earlier ADDM reports showed 1 in 150 (2002), then 1 in 88 (2008), then 1 in 54 (2016), then 1 in 36 (2022), then 1 in 31 (2025). Each increase reflects broader diagnostic criteria, more recognition of autism in girls and racial minorities, and better surveillance. The underlying rate of autism in the population is almost certainly not changing at this pace.
  • Boys-to-girls ratio is shrinking. Earlier ADDM showed ~4:1. The 2025 report is closer to 3:1. This is the diagnostic system catching up to the reality, not a change in underlying biology.
  • Racial disparities are persistent and shrinking, but still present. Black autistic kids have historically been under-identified in ADDM relative to white kids. The 2025 report shows continued narrowing of this gap, but it remains real.

How the data gets manipulated

  • "Autism epidemic" framing uses the rising identification rate as evidence of a biological epidemic. This is wrong. The rate of identification is rising. The underlying biology is not the question ADDM was designed to answer.
  • "Causation" claims attached to ADDM data — connecting prevalence increases to environmental factors, dietary changes, vaccines, or anything else — are not supported by what ADDM measures. ADDM is a surveillance system, not a causation study.
  • Adult autism prevalence claims derived from child surveillance are extrapolations, not measurements. Adult prevalence research uses different methodologies and arrives at numbers closer to 1 in 45 in some studies and 1 in 100 in others, depending on methodology.

What the report actually tells autism families

That autism is more common than the diagnostic system used to acknowledge. That the system is improving at identifying autism in girls and in historically underdiagnosed communities, though slowly. That a child diagnosed today is being identified by a system more accurately calibrated than the one that missed many adults of previous generations.

Find the report: cdc.gov/ncbddd/autism/addm.html. Read the full report PDF, not just news coverage. The Discussion section is where the report itself acknowledges what the numbers do and do not show.

Source briefs (internal): webearish-audit-2026-05.md

Disclaimer: educational content from autistic adults and the autism family community. Not medical or legal advice. Consult a qualified professional for medical and legal decisions specific to your situation.