I’ve been reading his work wherein he shows his reasoning/calculations. You can go find them, but it would take away from snarking.
OK, I did, and they're a mess.
There are a number of issues but let's just focus on one for now.
His whole calculation hinges on something called PTR (Propensity To Report). This number represents the likelihood someone will actually report an adverse event to VAERS.
According to Kirsch, the FDA says this number is 10 for the Covid vaccines. Kirsch isn't buying it. He says the number is really .25.
So Kirsch says people are 40X less likely to report adverse events to VAERS than the FDA says they are. Wow.
He must have some pretty good evidence of this, right? Well...let's see what he bases this claim on:
"For example, there are 3.9 times as many otitis media infections reported this year in VAERS as in previous years, but there are 17X more events reported. If PTR was same as in previous years, we should have seen 17X as many infections this year. Since we saw only 3.9, it appears the propensity to report is .23 this year, i.e., ΒΌ of the normal rate."
Get that? There are a lot few middle ear infections reported as a percentage of all events so...there must be underreporting? Even though the same numbers mean there must be overreporting of other events?
Well, maybe I just don't get his ear infection logic.
Let's look at another way he calculates URF. He relies on a study of post-vaccination anaphylaxis at Mass General Brigham. It found a rate of 2.47 per 10,000 vaccinated.
Kirsch then assumes that is the rate that should be reported to VAERS since anaphylaxis should occur quickly and is likely to be recorded. Since the rate reported to VAERS is significantly lower than this Kirsch concluded the underreporting factor is 41.
But there a a couple of problems. First, the MGB study wasn't based on anything like VAERS. Instead of relying on self-reporting they actively followed up with each of the recipients for three days and asked about particular symptoms. If they reported any two, say itching and swelling, it was recorded as an anaphylactic reaction.
Compare this with VAERS. What are the odds that someone who had itching and minor swelling at the site of the injection the next day would run and report it to VAERS?
This illustrates the difference between a passive system like VAERS, and an active one that follows up and solicits reports, like V-Safe.
Clearly this is a giant flaw in his reasoning.
More importantly, much of this debate has become overtaken by events. We don't have to rely on the torturing of VAERS to draw conclusions any more . We have real world data on hundreds of millions of shots.
If the rate of deaths caused by the vaccines are anything like Kirsch says the actual data will be screaming at us.
But it isn't.
It turns out I'm not the only one wondering about Kirsch's numbers. Jeffery Morris, Director of the Biostatistics Division at Penn, has been trying to have a dialog with Kirsch about his claims.
It seems that Morris actually looked at the data from CMS which clearly indicate the background death rate for people getting the Covid vaccine is almost exactly the same as it was for people getting the flu vaccine. For anyone who's interested he addresses this issue and others he has with Kirsch's methods here:
Response to Steve Kirsch's ad hominem attack, and clarifying the key sleight of hand in his argument
So far my research makes me skeptical of Kirsch's claims.