Where does this idea of “overwhelming scientific evidence” in favour of the vaccines reducing severe illness and death come from? Neither were clinical endpoints of the manufacturers’ randomised control trials. So not from there. /1

They come instead from observational studies. But the very reason we do RCTs is that observational studies are of extremely low evidentiary value in complex settings, where confounding factors are multitudinous and often unknown to us. /2

Shockingly, many observational studies I’ve been offered as “overwhelming evidence” fail to mention confounding factors at all. Many adjust for one—age. Because older people (who have higher mortality) are more likely to be vaccinated, adjusting for age favours the vaccines. /3

But there are two other factors that work in the other direction, yet are seldom controlled for—income and health status. Poor people have massively higher mortality than rich (duh) and are less likely to be vaccinated. People who are terminally ill are seldom vaccinated. /4

If your age-adjusted analysis doesn’t mention that these factors (and others) could entirely invalidate your findings, rendering your analysis of extremely low evidentiary value, then your analysis is of NO evidentiary value, because you’re either unsuitably trained or biased. /5

As if this isn’t enough, we’ve now seen multiple reports of scientists going way beyond this kind of statistical duplicitousness, and messing with the underlying data by misclassifying deaths or fiddling with the population denominators. I’ve kind of lost count. /6

I’ve also never seen a regulator or academic institution respond coherently to such criticism, for example, by fixing the problems. Thus I believe there is actually no support for the efficacy claim that anybody should consider suggestive, let alone “overwhelming”. /7

Even if all of the problems were fixed, we’d still not be in a very good position, because there are, with certainty, other confounding factors we haven’t thought of or do not have a means for addressing. /8

I haven’t even begun to attack the meager findings of the manufacturer trials (that the vaccines reduce sniffles), or their major corollary that went unreported—that young and healthy people aren’t at material risk to Covid and don’t really need vaccines. /9

Nor have I covered the biased removal for “protocol violations” from the intervention arms of those trials, which could invalidate even those meager findings. /10

All of these observations have been made by better thinkers than me, but not all at one place and in plain English, hence this thread. /11

To stimulate discussion, I’ll make the proposition clear. I say there is no high quality evidence for the vaccines preventing severe illness and death. /12

My colleague @GirardotMarc has pointed to the absence of adequate mechanistic explanations to buttress such claims. And I coined the term “antibody myopia”, as so many competent scientists had pointed out to me that antibody counts are woefully inadequate measures of immunity./13

Legal cases are swamped by reams of the low quality stuff I’ve been talking about here. Media noise has replaced science, and it is disturbing to see doctors and scientists—even entire faculties and institutions—make sweeping generalizations without any scientific references. /14

The burden of evidence for these claims, let alone for the mandates and violations of multiple ethical codes that they underpin, is high. Yet the evidence provided is derisory. /15

Don’t even get me going on claims regarding transmission reduction. I have to think very hard to imagine an experimental design adequate to make a strong case for that, and nothing in the offing even comes close. /16

Israel, with the considerable natural immunity afforded by three past waves, with a world-leading proportion of triple-jabbed vulnerable people, and facing what is apparently a decidedly mild variant, hit record daily deaths a few days ago. /17

Before you get too excited, note that the last tweet is … an observational study. And therefore of low evidentiary value. It doesn’t prove negative efficacy, pathogenic priming or masking of adverse events. But it is good for refuting a huge range of claims of efficacy. /18

And it acts as at least as a reminder of how observational studies of efficacy should more or less be discarded, leaving us with the poorly designed manufacturer trials that showed very little, and should also be discarded until the data has been brought to light. /19

The FDA & Pfizer went to extreme lengths to avoid that ever happening. When they failed in court, Pfizer’s quarterly filing warned investors of the risk that the clinical trial data may contain inaccuracies. I invite you to study that fact. Observationally, of course. /20

For more philosophical thoughts on how we arrived at this strange place, kick back and listen to this well-prepared interview by @Ike__McFadden. /21

Add 1: Why Pfizer doesn’t care:

Add 2: Double standards applied to off-label treatments:

Add 3: And a new category of manipulation emerges. If the observational data doesn't suit the narrative, stop letting people see it. P-hacking in reverse. Imagine how many inconvenient studies were simply not published in the first place.

Covid data will not be published over concerns it's misrepresented by anti-vaxxers

---

Public Health Scotland will stop publishing data on covid deaths and hospitalisations by vaccination status - over co...

https://www.glasgowtimes.co.uk/news/19...