Study claiming 74% of deaths due to COVID-19 vaccines contains signs of potential biases, uses flawed methodology

Key takeaway

Numerous studies indicate that COVID-19 vaccinations reduce the risk of death and serious illness. While COVID-19 vaccines are associated with some serious side effects, such as myocarditis and thrombosis, the risk of experiencing these side effects is low and the net positive benefits of vaccination outweigh these risks.

Reviewed content

Verdict:

Unsupported

Claim:

A review of autopsy cases found that 74% of deaths were linked to COVID-19 vaccines

Verdict detail

Inadequate support: The claim relies on a review of individual autopsy case reports and temporal associations to draw an alleged link between COVID-19 vaccination and death. However, this methodology doesn’t provide adequate support for the claim made. Further, numerous studies have shown that COVID-19 vaccines reduce the risk of serious illness and death from COVID-19.

See our method of evaluation

Full Claim

“Explosive Study Once Removed by Lancet within 24 Hours, Now Peer-Reviewed and Public: Reveals 74% of Deaths Directly Linked to COVID-19 Shot”

In late June 2024, social media posts on Facebook claimed that “73.9% of deaths” from a collection of 325 autopsy case reports were “significantly linked to COVID-19 vaccination” (examples here and here). These posts referenced a review authored by Hulscher et al. titled “A Systematic REVIEW of Autopsy findings in deaths after covid-19 vaccination”, published in Forensic Science International on 21 June 2024. [Editor’s note: The study was later withdrawn from the journal by the journal editors, who cited numerous concerns regarding errors and misrepresentation.]

The review was also shared by The Gateway Pundit, a website known for propagating COVID-19 misinformation, on 23 June 2024. The Gateway Pundit claimed an “Explosive Study Once Removed by Lancet within 24 Hours” was “Now Peer-Reviewed and Public: Reveals 74% of Deaths Directly Linked to COVID-19 Shot”.

Seven of the nine authors of the review are affiliated with The Wellness Company, which sells “vaccine injury” consultations and supplements that claim to boost immunity against COVID-19. Several of the authors are also known for spreading COVID-19 misinformation.

As we explain below, these potential biases, along with methodological flaws within the review, provide inadequate support for the claim that COVID-19 vaccines are significantly linked to mortality.

“Explosive study” referenced by The Gateway Pundit contained numerous flaws

The “study once removed by Lancet” refers to a 2023 preprint—a study that hasn’t been peer-reviewed—uploaded on 5 July 2023 to the Social Science Research Network (SSRN), a preprint server associated with The Lancet. The preprint, titled “A systematic review of autopsy findings in deaths after COVID-19 vaccination”, was authored by the same individuals as the 2024 Forensic Science International review. The preprint was removed from the SSRN server one day after it was uploaded, citing that its “conclusions [were] not supported by the study methodology”.

For context, this preprint was initially mistaken for a peer-reviewed study by numerous social media posts, which claimed “Lancet Study on Covid Vaccine Autopsies Finds 74% Were Caused by Vaccine – Study is Removed Within 24 Hours”. Science Feedback explained why the preprint didn’t provide adequate support for establishing a causal relationship between COVID-19 vaccines and death in a previous claim review.

First, we found that the preprint showed signs of potential biases, including a vague definition as to why certain studies were included in the review and a lack of clarity regarding why other studies were excluded.

Second, the three physicians—Roger Hodkinson, William Makis, and Peter McCullough—who reviewed the autopsy case reports to determine whether COVID-19 vaccination was a cause of death are known for spreading COVID-19 misinformation. This is concerning given the potential lack of objectivity in their analyses of the autopsy reports.

Third, the preprint contained methodological flaws, like relying on case reports, to confirm a causal association between vaccination and death. Case reports aren’t ideal for evaluating causal associations because they’re anecdotal in nature, don’t contain controls to mitigate confounding factors, and typically use very small sample sizes. Numerous studies included in the preprint, for example, were case reports of just one patient.

In this case, to determine if a finding is generalizable to a broader population, one would ideally use a different study design, like a case-control study or a randomized clinical trial, to compare patient outcomes between comparable populations of vaccinated and unvaccinated individuals.

Finally, even if a person dies after receiving a COVID-19 vaccination, relying on temporal associations to establish causality is a flawed approach, as Science Feedback explained in a previous Insight article. In other words, just because a person died after receiving a COVID-19 vaccine, it doesn’t mean that their death was caused by the vaccine.

2024 review contained many of the same methodological flaws and biases as the 2023 preprint

While the 2024 review did address some of the issues in the 2023 preprint, many of the methodological flaws and biases we identified in our previous claim review still hold true for the 2024 review.

For example, like the 2023 preprint, the 2024 review looked at 325 autopsy case reports and one necropsy case report and determined that 240 of the 326 deaths (“73.9%”) were “significantly linked to COVID-19 vaccination”.

Yet the fundamental issue with the 2023 preprint’s methodology remains in the 2024 published study. Its conclusions still rely on temporal associations, which don’t provide sufficient evidence for a causal relationship between COVID-19 vaccination and death. As the authors acknowledged themselves, confounding variables such as “illnesses, infection, drug interactions, and other factors not accounted for, could have played roles in the causal pathway to death”.

We also know from other studies that COVID-19 vaccination doesn’t raise the risk of all-cause death, as Science Feedback reported in previous claim reviews. On the contrary, vaccines reduce the risk of serious illness and death from COVID-19.

Finally, three of the authors—Hodkinson, Makis, and McCullough—conducted what they defined as a “contemporary and independent review” of autopsy cases and attributed 240 of the deaths from the autopsy cases to COVID-19 vaccination.

However, this contradicts the original assessments made by the scientists evaluating cause of death for these autopsy reports. Authors from the largest study in the review, comprising 121 of the 325  autopsy cases evaluated, found “[n]o relation between the cause of death and vaccination”. Other studies included in the review also found no proof of a causal relationship between COVID-19 vaccines and death.

Conclusion

A 2024 review published in Forensic Science International employed flawed methodology and relied on potential author biases to draw an association between COVID-19 vaccination and death. These methodological flaws and biases mirrored those from a preprint by the same authors that was retracted from The Lancet’s SSRN server in 2023.

Update (16 Aug. 2024):

We updated our review to indicate that the study by Hulscher et al. was later withdrawn from the journal. The journal editors cited concerns including “[i]nappropriate design of methodology” and “[e]rrors, misrepresentation, and lack of factual support for the conclusions” as reasons for the withdrawal. This information has been added to the first paragraph of the review.