COVID-19 Vaccines And Excess Mortality - Why Does Scientific Correction Take Such A Long Time?

In 2024, I wrote about a really bad paper that had just been published. It was causing a massive hubbub because the authors of the study were claiming that COVID-19 vaccines had likely caused a lot of deaths across the world in 2021 and 2022, and that COVID-19 itself was mostly harmless. While that was a common claim during the pandemic, this was a paper published in the leading journal BMJ Public Health, lending the ideas significant credibility.

The paper itself was terrible. The authors had simply lifted data from a previously published work called the World Mortality Dataset by Karlinsky and Kobak. They had then used Excel to poorly recreate some of the graphs from the original paper, and then added in their own opinions.

Karlinsky and Kobak showed mathematically that COVID-19 vaccines were not related to excess mortality, and that instead it was COVID-19 infections that were causing the vast majority of extra deaths. Instead of using this mathematical demonstration, or even discussing it, the BMJ Public Health paper implied that vaccines and lockdowns were the cause of excess mortality. The authors then spent several pages implying that governments were conspiring to keep this a secret, and saying that we should be doing massive autopsy studies to determine how many of the deaths were really caused by vaccines.

It was, as I said, a truly woeful paper. On the 11th August 2026, 874 days after the paper was published, it was retracted. You can read the full retraction notice here.

Obviously, the paper was terrible and COVID-19 vaccines save lives, but I’ve written about that many times before. There are additional things that are fascinating about this whole process that the BMJ notice lays out in detail. Let’s look at what happened with the paper: