> At best, that’s really noisy data.
Yes. That's exactly why each of the confidence intervals have a negative and a positive extreme. If there is a real effect, it's below the noise level.
The RR CI must be >= 0. The calculation is based on the incidence in a population and the incidence in a sub-population, which are themselves fractions of positive integers (e.g., 1/1000 and 1/100). I don't think RR can ever be negative, or at least I'm not sure how you would get that without messing up your data or calculations?
RR of >1 means there is excess risk, and <1 means anti-excess risk which is a word I cannot think of right now (just, less risk than the baseline anyway). But RR is also just a measure of risk over baseline to begin with, so a 1.5 RR on something that happens 1/10,000 times is ... not that concerning? I wouldn't be concerned, anyway. Depending on what the risk is, I guess.
So, no, I don't think we would expect RR CI to be centered around zero at all.
The fact that the main effect is so small and crosses from neg to pos is by far more concerning. I would not submit results like that for publication.
Sanity checking results shouldn't be so difficult.
Note, however, that almost all the effects blamed on the vaccine are also effects we see with the virus. Does this not strongly suggest that what we are actually seeing is how the body reacts to the spike protein? If so, that means that if the patients got the real thing they would have the same or worse outcome?
Realistically, the comparison shouldn't be vaccine vs nothing, but vaccine vs infection--and by that yardstick they aren't even in the same ballpark.
Moreover, if there is no effect you expect that some of the intervals are almost centered but you also expect that many of them are very biased. Moreover^2, if there is no effect, You even expect that in a 5% of them the 95% confidence interval does not include 0.
This is a nice trick to detect fake data in blab reports from students. The intervals are too centered.
This is the conclusion by the authors of the paper.
This is the reality of measuring rare events among moderately large groups of people. It’s also possible that any number of unexpected cofounders could have occurred for a small group of people receiving the placebo doses in one segment, such as a small number of placebo doses having an unexpected issue. Impossible to say, which is why large numbers of people are necessary in trials exploring relatively rare events like this.
Regardless, there doesn’t appear to be cause for concern among any of their observed serious events during the study period.
For those about to downvote me: my general (but unscientific) opinion is that it doesn't make sense for the covid vaccine to be riskier than covid itself. But as a society we have shown that even "technical" people like researchers and scientists are willing to fudge, ignore, or misinterpret data for various political or personal purposes, not to mention that doing so in domains outside of public health is common practice in academia. So I believe it's only prudent and rational to be extremely skeptical of anything that looks "off".
https://dailysceptic.org/2022/05/23/concerns-of-fraud-in-pfi...
Another similar problem in which a trial participant appears to have died a day or two after taking the shot, but it was ruled unrelated (by Pfizer!):
https://igorchudov.substack.com/p/pfizer-study-subject-c4591...
Small samples? The confidence intervals look pretty wide.
As I understand it, these are results from two different trials done separately by Pfizer and Moderna, with slightly different protocols and reporting standards:
"Moderna reported SAEs [serious adverse effects] from dose 1 whereas Pfizer limited reporting from dose 1 to 1 month after dose 2."
"For reasons that are not documented in the trial protocol, Moderna included efficacy outcomes in its SAE tabulations, while Pfizer excluded them. As a result, Moderna’s SAE table did not present a traditional SAE analysis but rather an all-cause SAE analysis. "
Thus, only numerical comparisons between treatment and control within the same trial are valid (assuming proper randomization), not between the two trials.
Pfizer reported serious adverse effects in 127 vaccine and 93 placebo. Moderna reported serious adverse effects in 206 vaccine and 195 placebo.
=> Headline: "Serious adverse events of special interest following placebo vaccination in randomized trials in adults"
The placebo groups for the pfizer cohort and moderna cohort differed in baseline rate by the same magnitude (~25/10000) as the observed effect.
At best, that’s really noisy data.