18 to 98 serious adverse events for every person helped by vaccination?

Yet another study throws doubt on vaccine strategy

When I wrote my article that “Pfizer’s own data shows a net loss on hospitalisations” I was nervous about publication. I knew there was a problem with the safety data, and I wanted to quantify that problem using Pfizer’s own data. It looked possible that for every person the vaccine kept out of the hospital, five people might end up in the hospital due to serious adverse events from the vaccine itself. A net loss on our collective health.

Pfizer's own data showed vaccine creates net loss on hospitalisations

Nervous of saying 5 to 1 explicitly, I didn’t say it in the end. I presented a conservative case that the number could be 1.3 hospitalisations created for every person helped by the vaccine. I did this because the “Number Needed to Treat” on hospitalisations, taken from Pfizer’s data, wasn’t a very confident number. But the reasons for that lend themselves to the overall point; not enough people were getting hospitalised with Covid-19 across a sample of 20,000 to confidently assess how many people we’d need to vaccinate to stop a single person falling sick. Whilst this casts some doubt on the exact ‘Number Needed to Treat’ it also suggests that we ought to pay very close attention to the side effects of the medicine since the risk of falling genuinely sick seems to be low.

Put another way, if the risk of serious sickness from Covid-19 is low, and the risk of side effects from the medicine is also low but higher than the risk of illness, then we have an actively harmful product.

Weighing up such issues, I instead presented data from which astute readers could calculate that 5-1 figure if they so wished, but I chose not to say the quiet part out loud. I showed how it’s possible to see serious adverse events running at 1 in every 1,444 people, but we would need to vaccinate 7,230 people to stop one serious case of Covid-19. Could it really be? That the data always looked this bad? Such observations are worrisome. Pressing ‘publish’ is never easy on these matters.

Since that time, I published an article on the peer-reviewed Doshi paper which showed broadly the same thing. Using data from the trials, the paper showed that the vaccines appear to create a net loss in hospitalisations. I wrote about that paper below.

Peer Reviewed: Vaccine trial data shows a negative benefit to harm ratio

The thing to note on that ‘Number Needed to Treat’ figure I originally posted about is that it’s calculated across the whole of society. It’s a single ‘society-wide’ number, it’s not broken down by logical segments at all. Depending on which segment of society you look at, that ‘net health loss’ figure could get better, but it could also get worse.

For example, what if we looked only at old people? We know their risk from Covid-19 is much higher, and so even with the side effects we expect to see, we could still see something close to a net benefit in those segments. I stress “could”, because I haven’t looked, I use it only as an example.

What if we looked at a much younger cohort of society? A cohort we know are subjected to significantly lower risk from Covid-19? Lower risk of sickness essentially amplifies the net harm of the side effects when we’re looking at a societal scale. If I don’t really benefit from the product (since I’m at very low risk), then the explicit risks to me from the product itself become ‘all risk, and little benefit’. To use a crude analogy, if you didn’t have appendicitis, you wouldn’t have your appendix removed because the operation would create a net harm. But if there was a huge industry making billions of dollars selling ‘appendix removal kits’ it’s possible they’d try to convince you to have it removed.

A new pre-print has used this approach to look at the relative risk/benefit rewards for young people on University campuses. The abstract of the paper is actually very succinct, so I’ll post it here in full.

“We estimate that 22,000 - 30,000 previously uninfected adults aged 18-29 must be boosted with an mRNA vaccine to prevent one COVID-19 hospitalisation. Using CDC and sponsor-reported adverse event data, we find that booster mandates may cause a net expected harm: per COVID-19 hospitalisation prevented in previously uninfected young adults, we anticipate 18 to 98 serious adverse events, including 1.7 to 3.0 booster-associated myocarditis cases in males, and 1,373 to 3,234 cases of grade ≥3 reactogenicity which interferes with daily activities. Given the high prevalence of post-infection immunity, this risk-benefit profile is even less favourable”

The observation on myocarditis is interesting because it has potentially deadly consequences. Is it right to subject 30,000 people to the risk of myocarditis so that we might stop one person falling seriously sick with Covid-19?

“Our analysis is conservative given the fact that we did not account for the protective effects of prior infection, which is estimated to be substantive.” Put simply, the risks to this group from Covid-19 are even lower than the risk the authors used because they didn’t factor in how many have already become immune to Covid-19.

Full paper.