Conspiracy theory websites and social media channels have leapt on a new piece of research that, they say, shows Covid-19 shots have killed more than 400,000 people in the U.S. Furthermore, they say adverse events reporting systems underestimate side effects by some 2000%.

In reality, the claims in the paper, and those on social media, are mistaken and highly misleading. Ample scientific evidence shows that approved jabs are both safe and effective against Covid-19.

Not only is this new research self-published and unreviewed, its first author — a neurobiologist who normally studies neuroimaging — has no clear experience in epidemiology.

But that hasn’t stopped anti-vaccine activists falsely promoting it as evidence that vaccine-related deaths are being covered up.

What does the research claim?

The authors tracked monthly and weekly all-cause mortality data alongside rising vaccination numbers in the US and Europe. They compared these data with similar all-cause mortality figures from 2020, before vaccines were widely available. The U.S. data was also broken down by age.

Attributing increases or decreases in mortality to jabs, the researchers estimated between 130,000 and 180,000 people had died in the US as a result of vaccination between February and August 2021: an enormous figure that does not match any peer-reviewed scientific studies.

The authors use the same figures to estimate that adverse events monitoring systems are underestimating deaths by a factor of 20 — a claim anti-vaccine advocates have used to falsely suggest as many as 400,000 people have been killed by the jabs.

Serious flaws in the research

Several serious flaws undermine this research. Firstly, it fails to account for the fact the peaks in Covid-19 cases — and therefore deaths — did not follow exactly the same pattern in 2020 and 2021.

Both regions saw Covid-19 mortality increase dramatically in spring 2020 — in line with the first wave — and again in the 2020/21 winter months.

By comparing the months of both years like-for-like, the paper won’t properly reflect the impact Covid-19 itself had on overall mortality. It’s likely the research falsely attributes deaths from Covid-19 and other causes to vaccines.

This underlines a more fundamental problem with the paper. It fails to properly distinguish correlation from causation. The authors jump from terms like “correlated with” to “vaccine-associated” without providing further evidence to support a link. Just because event B follows event A, this does not mean event A was the cause.

There’s no evidence in this paper to suggest the deaths it describes were caused by vaccines, nor that adverse events reporting systems are massively underestimating mortality. Further investigation would be required to establish a clear link and, so far, no reliable research has found one. Yet social media posts are rife with claims this new study shows the shots have killed hundreds of thousands of people.

Conflating correlation with causation is a common tactic of those who spread disinformation. Anti-vaccine activists, for example, often plug the gap between the two with their own unverified claims.

They also often describe “adverse events” as “adverse reactions” and “side effects” to misrepresent data published in surveillance systems like the Centers for Disease Control’s Vaccine Adverse Events Reporting System. But to the casual internet user, the difference between these terms may not be obvious.

This doesn’t mean that studying correlation can’t be useful. Adverse events reporting systems, for example, are designed to do just that. They allow scientists to spot when an adverse event seems to be occurring more than normal in vaccinated people. Researchers can then perform further analysis to determine whether or not there is a link between the two events, or if something else is to blame.

But before this extra research takes place, adverse events data alone can’t prove causation.

As the CDC states: “While very important in monitoring vaccine safety, VAERS reports alone cannot be used to determine if a vaccine caused or contributed to an adverse event or illness. The reports may contain information that is incomplete, inaccurate, coincidental, or unverifiable.”

Unreviewed and unverified

This paper, which was uploaded to academia network ResearchGate by author Spiro Pantazatos on 8th December, does not appear to have been accepted for publication in any peer-reviewed journal.

This means it has not been reviewed or corrected by other experts in the field: a basic standard for scientific research. It’s not even clear if the paper has been submitted for review by any journal.

In spite of this, ResearchGate metrics show it’s been read more than 368,000 times as of January 10th. And it’s been shared far and wide as factual by anti-vaccine activists on social media.

Pantazatos’ employer, Columbia University, criticised the paper when contacted by Health Studio. A spokesperson said: “While Columbia University respects the academic freedom of its faculty, the peer reviewed published data do not substantiate the claims in this article.”

Although Pantazatos is a published researcher, he works in the US university’s psychiatry department and usually studies neurobiology and neuroimaging. He does not appear to have published any peer-reviewed research on epidemiology or vaccines.

His co-author, Herve Seligmann, is currently listed as an “independent researcher” on ResearchGate. He has numerous publications in the field of genomics research and appears to have worked at several universities up until 2018.

Seligmann seems to be involved in several Covid-19 projects, one of which falsely links rising case numbers to vaccination campaigns.

What’s really happening?

Contrary to the claims of this paper, statistics show an inverse correlation between vaccination and Covid-19 mortality rates around the world. This means that where vaccination rates are increasing, rates of death from the disease are going down.

In the US, for example, unvaccinated people have died of Covid-19 at a greater rate than their vaccinated counterparts for many months.

Real-world data like this supports numerous peer-reviewed clinical trials and laboratory studies showing that Covid-19 vaccines are very good at preventing severe disease and death.