VAERS is not designed to establish causality, but Rose tries to claim causal relationships using Proportional Reporting Ratio (PRR) analyses as well as Bradford Hill criteria. She then makes up her own notion of a “Composite Causality Score,” which is simply a measure she created but didn’t bother to define and calculated by an AI tool, which are known to be flawed. It’s telling that there is no clear description of the methodology used.
The ONLY legitimate thing that can be done with VAERS data is to generate hypotheses. Some of the reasons for this is because there is no denominator data withing VAERS, there are no controls, the raw data is not vetted, and there is a great deal of reporting bias. She draws temporal association as biological causality, which is a major misuse of VAERS.
Another major flaw is the use of PRR as evidence of vaccines causing myocarditis. However, this compares reporting proportions, not risk. This led to a lot of reporting, whether valid or not in VAERS.
Further, she cites discredited sources, papers from other fringe or predatory journals, and preprints. This does not make a solid base as evidence.
The methods used to “condense” data are not clearly explained or validated, which means that they cannot be reproduced. This is essentially nothing but p-hacking, which is highly unethical in scientific literature.
Finally, the “journal” that this is in is not a legitimate journal It’s from an advocacy group now known as IMA, which used to be the FLCCC. They were pushing fraudulent COVID treatments like ivermectin and hydroxychlorquine.
It’s no wonder that this didn’t go to a normal scientific journal. It’s garbage. Don’t be duped.
