“22 years ago, it was, uh, 1 in 20,000... But now it's 1 in, 1 in 12.”

Public remarks before the start of the second term, repeated in 2025

The reported rise in autism prevalence is driven mainly by broader diagnostic criteria, expanded screening, and greater awareness, not by some new artificial cause. Researchers and the CDC are clear that older studies used narrow definitions that captured only the most severe cases, while today's 'autism spectrum disorder' definition encompasses a far wider range of traits. There may be some true increase, but not remotely to the extent Trump implies.

Trump's own numbers are wrong. He has claimed autism went from '1 in 20,000' to today's 1 in 31, but the CDC estimated about 1 in 150 children had autism in 2000, not 1 in 20,000. The 1-to-5-per-10,000 figures he invokes come from 1960s and 1970s studies that used deliberately narrow definitions of the condition. Comparing those antique, restrictive counts to today's broad spectrum diagnosis is not a like-for-like comparison. Critically, autism is not a recent invention: it exists in people over 50 and develops in the womb, well before any childhood vaccine. The strongest evidence the vaccine-skeptic camp cites for a 'true' increase, the UC Davis MIND Institute report, was never peer-reviewed and was criticized as drawing 'unwarranted' conclusions. Studies that have isolated single factors attribute roughly one-quarter to one-third of the rise to diagnostic changes alone. "Reported prevalence rates have increased significantly over time, but researchers say that's due to increased screening and awareness, broader diagnostic criteria, and more limited definitions of the condition in older studies." — FactCheck.org (Sept. 23, 2025)