Antivaxx claim #1: “Death rate from variant COVID virus six times higher for vaccinated than unvaccinated, UK health data show”
Note: this is the first article of a series of articles dedicated to finding common antivaccine claims surfacing social media platforms and some irreputable news websites, blogs and forums, particularly the ones that are supposedly supported by evidence, and checking for their accuracy.
A recent claim accompanied with a data table from a British governmental agency has been circling around twitter and other social media platforms, which was observed to be a part of an effort put by some social media personalities and users to promote their anti-vaccine beliefs.
The data table in question is captured from the most recent technical briefing update released by Public Health England, an agency of the UK’s department of health and social care, to provide recent information and data about the different SARS-CoV-2 variants spotted in the UK. The document can be found in this following link:
The relevant data table that’s used to support the claim that vaccinated individuals are more likely to die from the delta variant than the unvaccinated ones is found on page 12.
The claim is based on the calculation that the death rate among fully vaccinated people 14 days or more after receiving the second dose is 0.636 % (26 / 4087 * 100) as compared to only 0.096 % among the unvaccinated. It was also calculated that the death rate 21 days or after receiving only the first dose is 0.106 %, which is lower than the death rate calculated for fully vaccinated people. These data and calculations were immediately used, without any considerable amount of thinking, to trigger the hyperactivity of antivax social media accounts and some conspiracy theory websites to publish ridiculous claims such as that quoted in the title of this article. An individual with little statistical knowledge would take a look at the claim, input the numbers on their own calculators, and get either panicked or excited (depending on their prior vaccination stance) when they think they verified the accuracy of the claim. Meanwhile, a critical thinker would be wise enough to not jump to any conclusion yet.
The problem with the way antivaxxers used the data is that they failed to account for alternative interpretations and biases that are very well-studied in statistical fields. Think carefully of this question: are age group and comorbidities necessary to make conclusions based on death rate from these data? The answer should be YES. It matters because we know that the vaccination effort started by vaccinating the older population in the UK and the age range eligible for vaccination decreased over time. It’s very possible that the higher death rate shown in these data with those infected with the among the vaccinated is due to their older age and presence of pre-existing conditions. Similarly, the death rate among the unvaccinated is due to the likelihood that the biggest portion of them are young. We can’t tell for sure, and that’s why additional data regarding age and comorbidities are required before we can make sort of dangerous claims, but based on what we already know about vaccination and demographics, this speculation is plausible. Additionally, we know that vaccines aren’t 100% effective, so those cases and possible subsequent death resulting from the COVID-19 are expected. The fact that the cases and hospitalization numbers among the unvaccinated are bigger than in the vaccinated demonstrates that vaccines are indeed effective to a large degree. At the end, we should always keep in mind when examining data and interrupting it that correlation doesn’t equal causation, and that once cannot make such misleading and harmful conclusions when they clearly don’t have any statistical knowledge begin with.