Get access to over 1,300 articles and reader Q&As.
In a recent episode of The Tucker Carlson Show, Robert F. Kennedy, Jr, the Secretary of Health and Human Services, claimed that the CDC had evidence in the late 1990s that the hepatitis B vaccine caused a 1135% increase in autism rates, and they covered it up.
This is not true. Not the connection between vaccines and autism, not the claim of a cover-up, nothing. But beyond that, I want to unpack where this even comes from. How does a claim like this end up getting amplified in this way? And what is the defense against it?
Where did this claim come from?
In 1999, a researcher named Thomas Vaerstraeten, who was working at the CDC at the time, presented an abstract at a meeting of the Epidemiological Intelligence Services. The abstract utilizes data from health records, which contain information on immunizations and neurological diagnoses. Based on these data, the authors of the abstract show that autism diagnosis rates are higher among children whose vaccine history indicates a higher dose of thimerosal before the age of one month.
This particular abstract has come up periodically over the years in the anti-vaccine rhetoric, which is likely why RFK Jr. was aware of it. He has some of the details wrong – the abstract suggests an increased risk of 760%, not 1135%, and the paper isn’t about hepatitis B vaccines specifically. But it is fairly clear that this is the data he’s referring to.
That’s the background. What’s the counterargument?
Point 1: This claim is not supported by actual research
In scientific research, abstracts like this one are often a first look at data, designed to point towards future work. They are not peer-reviewed and usually provide very limited detail. In this case, we learn almost nothing about where the data is from, how they measured autism diagnosis rates, how they measured vaccination rates, or differences across different groups.
In this particular case, the abstract compares the groups without any other adjustments for differences (other than gender, year of birth, and geography). There are a lot of other variables you’d want to account for. Perhaps most importantly, people who have more vaccines early on tend to have more well-child visits, and those well-child visits are also where other diagnoses (such as autism) are often made. This is a known and pernicious problem with this particular analysis, and it may be what is going on here.
It is, in fact, impossible to tell from this abstract alone what explains the results. There just isn’t enough information in the abstract. The way science works is that after an abstract, authors will move on to produce a paper in which they usually provide more details, do more careful analysis, and often add more data. The abstract is a first step of the research process.
These researchers did that. In 2003, they published a more complete paper in the journal Pediatrics, which showed no link between autism and thimerosal exposure. Review data based on multiple papers shows a similar result.
The bottom line is that the most complete data on this topic does not show a link between thimerosal exposure through vaccines and neurodevelopmental issues.
Point 2: Current childhood vaccines do not contain thimerosal
To the extent that RFK Jr. was using this fact to imply current vaccines are not safe, it is important to point out that in 1999, the FDA required the removal of thimerosal from childhood vaccines. This was done after pressure related to these claims, and even with no actual evidence, they decided it was worth doing since this isn’t an important vaccine ingredient.
So even if you weren’t convinced by the best data here, the particular claim being made is irrelevant to current vaccines.
Point 3: There is no cover-up
The claim made by RFK Jr. is that the CDC has been covering up this result for decades.
This is not true. There’s no cover-up! The abstract is available online. You can read it. The researchers took that data and added more data, and wrote a paper which was published in a prestigious journal. This is how science works. It is extremely common that when you do further analyses or add more data, results change. (And not, as RFK Jr. claimed, due to pressure from the CDC.) This is especially likely in a setting like this, where the initial results in the abstract were fairly obviously correlation and not causation. In that case, it’s completely unsurprising that once you do a better job addressing these biases, the results change.
Closing thoughts
This entire situation illustrates a core problem with this kind of misinformation online. It’s easy to make a claim like this and, with a sufficient platform, it gets amplified. Most people are not in a position to dive down the rabbit hole of what, exactly, is the source of the claim. Even if you get partly down the rabbit hole, it’s hard for people to know why an abstract like this isn’t the same as a paper, and to understand why results might change. Often, people like RFK Jr. cite “research” as though it is all created equally, when, of course, that is not the case.
I wish I had a solution to make it easier to spot and understand things like this – I do not, but hopefully this helps with this particular one.
Community Guidelines
Log in
Thanks so much for articles like this. It helps people to understand the way medical research is often misrepresented by politicians and on social media. Even if you can’t provide an article to combat every single conspiracy theory or TikTok trend, reading just a few of these articles helps parents like me to appreciate how easily people can find questionable research or misrepresent good data to back up their claims.
This is so helpful!! Thank you for doing this work and sharing this. It really adds to the ridiculousness of the way claims are being made these days and it’s so helpful to have your voice of reason giving us accurate information. Thank you!
Thank you for continuing to go down rabbit holes and lay out facts for us