THAT daily dose of drivel about all things covid, the Global Health Now newsletter of the Bloomberg School of Public Health at Johns Hopkins University, recently linked to an article from the equally comical newsletter CIDRAP (the Center for Infectious Disease Research and Policy at the University of Minnesota) claiming that ‘Covid vaccination saved more than 5,000 US lives in seven months in 2023-24, CDC estimates’. The important word there is ‘estimates’.
‘How can this be?’ I hear the astute readership of TCW ask, given that we know the covid vaccines barely work. TCW was quick off the mark to publish an explanation by Neville Hodgkinson that the absolute risk ratio (ARR), which is the true measure of your likelihood not to become infected if you take a vaccine compared with not taking it, is very small for the covid vaccines. Less than 1 per cent in fact, and this is borne out by an article, rarely referred to except in these pages, in the Lancet.
The answer to the question above is that the covid vaccines probably did no such thing. Turning to the original source for the report, a paper from a combined US and Mexican team of researchers published in Vaccine titled ‘Estimating Covid-19 associated hospitalizations, ICU admissions, and in-hospital deaths averted in the United States by 2023–2024 Covid-19 vaccination: A conditional probability, causal inference, and multiplier-based approach’ shows how the conclusion about the covid vaccines was reached.
The title explains a lot: this was done by modelling data. Some of it, as explained in the paper, was ‘hypothetical’. Specifically, to ensure that the authors are not misrepresented: ‘To estimate the number of Covid-19-associated hospitalizations averted by 2023–2024 Covid-19 vaccination, we compared the total number of people with a Covid-19-associated hospitalization in the presence of 2023–2024 Covid-19 vaccinations (referred to as factual) to the hypothetical total number of people with a Covid-19-associated hospitalization without the presence of 2023–2024 Covid-19 vaccinations . . .’
To crunch the numbers, the outcomes ‘were estimated via a novel multiplier model that utilized causal inference, conditional probabilities of hospitalization, and correlations between data elements in Monte Carlo simulations’. The outcome, reported in the abstract to the paper, claims a possible ‘107,197 Covid-19-associated hospitalizations averted . . . and 6,749 Covid-19-associated in-hospital deaths averted’.
To be fair to the authors of the paper, the analysis seems rigorous and uses sensitivity analysis to test their model under a range of conditions. Like good scientists, they indicate the limitation that the ‘Monte Carlo simulation procedure may have had inaccuracies’ due to the estimates used as inputs to the model. They also suggest that their modelling is novel and only that it ‘may improve Covid-19 burden averted estimates’ but that further research is needed ‘to understand the methodological properties’.
Thus, the researchers express a suitable caution about their results that is entirely absent from the CIDRAP article. In large bold print, this says that the authors of the Vaccine paper claim ‘More older adults obtaining Covid-19 vaccinations would further reduce Covid-19-associated outcomes. Increased 2023–2024 Covid-19 vaccination rates would reduce Covid-19-associated outcomes in other age groups as well.’
In fact, the authors of the Vaccine paper do not actually say this. CIDRAP has combined phrases from two sentences, one which ends a paragraph and the other which begins the next one. The authors make no such claims, merely saying that if their results are correct then these are the outcomes which are ‘suggested’. CIDRAP’s claims would be a suitable candidate for the Desperate Marketing column of Private Eye.
Finally, what do the data in the Vaccine paper really tell us regarding the efficacy of the covid vaccines? Using ChatGPT, to which I uploaded the paper, I asked it to extract the data on ARRs for hospitalisation and death and, subsequently, the number needed to treat (NNT) to avoid one covid death. The outcomes are even less encouraging than in previous articles in these pages.
According to ChatGPT the ARR for hospitalisation is 0.00683 per cent; the ARR for death is 0.00053 per cent; and the NNT is 188,644 people who would need to be vaccinated to prevent one additional death due to Covid-19. If, after all the effort of vaccinating millions of people, 5,000 deaths were really averted that is possibly something to be happy about.
However, I searched in vain for any mention in either the Vaccine paper or the CIDRAP piece of covid vaccine side-effects or deaths related to the covid vaccines. Given that very few of us know anyone who died of covid but that many of us do know someone who has been damaged by the covid vaccines, I think we can make up our own minds whether continuing mass covid vaccination programmes is worthwhile.
