Robert F. Kennedy Jr. misrepresents CDC study, spreading false claim about hepatitis B vaccine-autism link

During an episode of the Tucker Carlson Show that aired on 30 June 2025, U.S. Secretary of Health and Human Services Robert F. Kennedy Jr. claimed that the U.S. Centers for Diseases Control and Prevention (CDC) had performed a study in 1999 revealing that the autism risk skyrocketed by 1,135% among children vaccinated against Hepatitis B compared to those who weren’t. 

This claim is part of a larger push from Kennedy –former chairman of the antivaccine organization Children’s Health Defense– to question the safety of Hepatitis B vaccines for children. Kennedy further claimed that the CDC had “watered down” the study to cover its conclusion. 

The claim has since spread on social media platforms. We found several TikTok videos containing the claim, including one with more than 240,000 views. 

In the interview, Kennedy mentioned that the study was led by a researcher named Thomas Verstraeten. Verstraeten and colleagues indeed shared results in 1999 assessing the risk of neurological impairment among vaccinated children, but Kennedy’s account is inaccurate on several counts and a large body of research now available shows that vaccines don’t cause autism. We’ll explain below.

1999 results were preliminary, not peer-reviewed

Verstraeten and colleagues communicated their results at the 1999 Epidemic Intelligence Service Conference. Conference abstracts are typically short summaries of ongoing research. They usually don’t go through the same vetting process as peer-reviewed published studies. For that reason, results presented in conference abstracts should be interpreted with some caution as they are likely to change as the research progresses. 

The abstract explains that Verstraeten and colleagues compared the risk of neurological disorders in children who received higher or lower amounts of thimerosal during their first month of life. It reported a 760% increase in cases of autism among children who received thimerosal-containing vaccines at one month of age compared to unvaccinated children. 

Although Verstraeten and colleagues aimed to study thimerosal-containing vaccines in general, the scope of the data used in the abstract effectively limited the analysis to only hepatitis B vaccine exposure. Indeed, in a comment about his work at the 2000 Simpsonwood conference, he explained that the only source of thimerosal exposure in the first month of life is the hepatitis B vaccine.

The findings contributed to public health concern around thimerosal at that time. Thimerosal is an ethylmercury-containing preservative used in some vaccines to prevent contamination. It has been a target of anti-vaccine disinformation campaigns that allege it causes autism and other harmful effects. However, Science Feedback showed that scientific evidence doesn’t support that claim. The evidence also shows that the hepatitis B vaccine doesn’t contain toxic doses of ingredients and is safe for children, as Science Feedback explained in earlier reviews. 

The completed, final study didn’t corroborate preliminary observation

The potential public health implications of their early results led the three authors of the 1999 abstract (Verstraeten, Davies, and DeStefano) to follow up with a more rigorous analysis. In 2003, they published a complete peer-reviewed study comparing the neurological development of children according to their level of exposure to thimerosal at different ages^[1].

They concluded that “No consistent significant associations were found between [thimerosal-containing vaccines] and neurodevelopmental outcomes.” In particular, there was no statistically significant increase in autism among children exposed to thimerosal (Table 4 of the 2003 study, see Figure 1 below).

But it is important to keep in mind that there are several differences in the dataset and methods used for the 1999 abstract and the 2003 study. For example, the 1999 abstract was restricted to children exposed to thimerosal during the first month of their life, whereas the published study from 2003 included exposure at one, three, and seven months of life. 

Furthermore, the 2003 study contained additional inclusion criteria such as low birthweight or congenital conditions that could affect how likely a child would be to receive vaccines or be diagnosed with developmental disorders.

These methodological differences may explain why the autism risk differed so greatly between the 1999 abstract and the 2003 study. It is common in research that preliminary observations aren’t replicated in a subsequent, more robust, and larger study, as researchers progressively refine their approach to mitigate possible biases and limitations.

Verstraeten refuted the claims of the CDC having “watered down” the study in a letter to the editor of the journal Pediatrics, which published the 2003 article. He argued that additional data were included in the 2003 article because initial results were “too prone to potential biases to be the basis for important public health decisions”.

Furthermore, we should point out that the 1999 abstract is publicly available. The methodology and results of the 2003 paper was also discussed by Verstraeten during the 2000 Simpsonwood conference, for which the transcript is publicly available. The fact that this information is publicly accessible clearly speaks against Kennedy’s claim of a CDC cover-up.

In September 2007, the U.S. Senate Committee on Health, Education, Labor and Pensions cleared the CDC and Verstraeten of alleged misconduct related to thimerosal in vaccines, rejecting allegations of a cover-up:

“Instead of hiding the data or restricting access to it, CDC distributed it, often to individuals who had never seen it before, and solicited outside opinion regarding how to interpret it. The transcript of these discussions was made available to the public. The data was also discussed at the Advisory Committee on Immunization Practices, a public forum held on June 21 and 22, 2000.”

Conclusion

In summary, Kennedy’s claim is inaccurate on several grounds. While initial results by Verstraeten et al. showed an increase in autism risk following hepatitis B vaccination, a later study conducted by the same researchers, which was published in a peer-reviewed journal, failed to corroborate that initial observation. These results are further supported by a large body of literature that now exists, showing that neither thimerosal nor vaccines cause autism. Claims of a cover-up are also contradicted by the evidence, given that the 1999 abstract and the proceedings of the 2000 Simpsonwood conference are publicly available.

References

• 1 – Verstraeten et al. (2003) Safety of thimerosal-containing vaccines: a two-phased study of computerized health maintenance organization databases. Pediatrics.