What does the evidence say about vaccines and autism?

Major scientific reviews conclude that vaccines do not cause autism. In December 2025 the WHO Global Advisory Committee on Vaccine Safety reaffirmed its earlier conclusions from 2002, 2004 and 2012 that there is no evidence of a causal relationship. The claim itself began with a 1998 paper in The Lancet that was later retracted. The picture is not entirely uniform in official US messaging: a CDC webpage was updated to say that the statement "vaccines do not cause autism" is not an evidence-based claim, because studies have not ruled out a contribution from infant vaccines. [1] [2]

According to a Johns Hopkins interview, the concern was first raised in 1998 by Andrew Wakefield, a gastroenterologist at the Royal Free Hospital in London, in a Lancet paper on 12 children. Parents of eight of them reported autism after the MMR vaccine. The study had no control group, and a case series can only describe cases, not show cause. The cases were not consecutive but cherry-picked. The paper was retracted and several co-authors withdrew their names. Wakefield nonetheless told the press that vaccines had caused autism, which the study could not show. [3]

The same source offers reasons the idea spread. Autism often becomes apparent at the age when many vaccines are given, so some children will show signs after vaccination by chance alone. Parents looking for explanations, and advocacy groups sharing similar stories, linked the two through timing rather than cause and effect. [3]

Once the MMR claim drew attention, large studies compared vaccinated and unvaccinated children and found no association. The hypothesis then shifted to thimerosal, a preservative used in some multidose vaccines but never in MMR, and then to the number of vaccines given at once. Studies of each found no relationship. Johns Hopkins counts 16 well-conducted, large population-based studies from different investigators and countries. [3]

The WHO committee's 2025 review covered publications from 2010 to August 2025. Of 31 primary studies, 20 from 11 countries, including the most rigorous, found no association, as did all five meta-analyses. Eleven studies suggested an association, but the WHO judged them to have significant methodological problems, very low strength of evidence and high risk of bias; nine came from a single US research group. A second review of aluminium-adjuvanted vaccines found no link in ten randomized trials and seven large cohort studies. A Danish registry study of children born from 1997 to 2018 found no association with autism either. Two ecological studies reported a link with aluminium exposure, but the WHO rated them at critical risk of bias and unable to show causality. [1]

The CDC page, now carrying a Data Quality Act rationale, says scientific studies have not ruled out that infant vaccines contribute to autism and that the earlier statement was used to prevent vaccine hesitancy. It says HHS has launched an assessment of autism causes. On MMR, it acknowledges that the IOM and AHRQ reviews find a high strength of evidence for no association. It then says the IOM gave most earlier studies no weight in 2012 for methodological limitations, and argues that the rest are retrospective, cannot prove causation and do not account for vulnerable subgroups. [2]

The WHO committee takes the opposite view of the same body of work, weighting high-quality, low-bias studies most heavily and concluding that the evidence strongly supports no causal link. The WHO position is the one backed by the systematic reviews in these results. The CDC page's wording is a recent change of stance rather than new data presented in these results. [1] [2]