The Centers for Disease Control and Prevention spent two decades telling parents the same thing the evidence told them: vaccines do not cause autism. Sometime in November 2025, under Health Secretary Robert F. Kennedy Jr., the agency rewrote that page. [4] It now reads: "The claim 'vaccines do not cause autism' is not an evidence-based claim because studies have not ruled out the possibility that infant vaccines cause autism." [1] That sentence is built to sound careful. It does the opposite. It inverts one of the most thoroughly settled questions in medicine.
Give the doubt its strongest form
State the other side fairly, because the wording has a kernel that sounds reasonable. No study proves a negative with absolute finality; science deals in evidence, not metaphysical certainty, and "we have not ruled it out" is technically true of almost anything you choose. [2] Hold onto that, because it is the whole trick. The standard the new language quietly demands, absolute proof that something never happens, is one no scientific question can ever meet. That is exactly why it is the favorite tool of people manufacturing doubt: it cannot be satisfied, so the doubt never has to end.
What the evidence actually is
Here is the evidence the page now calls insufficient. The possible link between vaccines and autism is not understudied. It is among the most studied questions in the history of pediatrics. A 2014 meta-analysis in the journal Vaccine pooled ten studies covering more than 1.2 million children and found no association between vaccination and autism, none for the MMR vaccine, and none for thimerosal, the mercury-based preservative the theory once blamed. [3] Across more than a dozen countries and well over a decade of follow-up, the answer has not moved. The conclusion that vaccines do not cause autism is not an absence of evidence. It is the evidence.
Where the claim comes from
The myth has a birthday and a death certificate. It began with a 1998 paper by Andrew Wakefield that claimed a link; the study was found fraudulent, retracted by The Lancet, and Wakefield lost his medical license. [2] The single paper Kennedy has pointed to as fresh justification was examined by FactCheck.org and found to be deeply flawed, the kind of study whose method cannot carry the weight placed on it. [2] A retracted fraud and a flawed paper are not a counterweight to 1.2 million children. They are the reason the question got studied so exhaustively to begin with.
Why the wording matters
The danger here is not abstract. When the CDC's own website tells parents the safety of vaccines is unsettled, some parents will reasonably hesitate, and hesitation is how measles comes back, which, as the 2025 outbreaks showed, it already has. A government health agency lending its authority to a manufactured doubt does not make the science less certain. It makes the next outbreak more likely.
The honest sentence is the one the CDC used to publish. Vaccines do not cause autism. That is not a slogan or a hope. It is what more than a million children's records show, checked and rechecked by researchers who went looking for a link and did not find one. [3] The page can be rewritten. The evidence cannot.
