How to lie with statistics: Mendel Singer edition
Meet Mendel Singer. He is the vaccine enthusiast that wants to vaccinate your children. He also likes to lie with statistics.
For example, Mendel wrote an opinion piece about putting the virus in “perspective”. He claimed that vaccine side effects are one in a million. A number that he conjured out of thin air. He then posted a chart that would be deceptive if it was clever:
Oh boy. Where to even begin?
Well, the obvious connection he is trying to make is that getting struck by lightning is twice as likely as having a vaccine side effect. Even if he meant a SAE, he is miraculously wrong.
Then he abruptly switches reference periods from per year to per day, not exactly subtle. He also switches the reference denominator from 1 million to 3 million. We can only assume that he did this to make car crashes seem rarer than they are because he is building to the last point in the chart. If, instead, he had noted that in 2018 the rate of deaths due to car crashes was 117 per million (not just injuries!), people might lose sight of the point. As to why he brought up shark attacks, maybe he just thinks that is an interesting statistic?
Finally, he conflates deaths with the virus to deaths from the virus. Yes, at the time, the CDC listed deaths in that age group as 40 per million, but we know that these numbers are heavily skewed due to a poor (deliberate?) incentive structure.
Singer then goes on to claim that the death rate in the US is higher than it has ever been. He claims that the death rate in 2020 was 10 per 1000 (it wasn’t), and the previous high was 9.5 in 1000. Of course, he neglects to point out that we are nearing the end of the boomer generation and death rates are going to continue to go up at least through the next decade. UN projections show the death rate steadily increasing into the 2050’s. Those projections are not increasing because of the virus. They are increasing because of population demographics. His conclusion is equally demented as he claims that vaccination will:
Help us get back to normal; and,
Prevent the virus from circulating and mutating.
Time has proven him wrong, but it does not take Nostradamus to understand that point 1 was never going to be true, and point 2 could not have been avoided with leaky vaccines.
Hopefully, that will provide my readers with some perspective on how this man manipulates numbers to serve his vaccine enthusiasm. Because he is the lead author on one of the main papers cited by the media, taught to doctors, and proliferated through social media. If you have heard people say that myocarditis is more common from the virus than through vaccination, they are citing this man.
“If you’re focused on heart inflammation, the safer bet is to take the vaccine.
Mendel Singer
Of course, assuming that people reject the vaccine on heart inflammation alone is asinine, but it is certainly one of many considerations. His paper claims that males between the ages of 12-17 are getting myocarditis at an adjusted rate of 876 per million cases.
This seems suspiciously high and his assumed risk of myocarditis following vaccination (76.5 per million) suspiciously low for this age group.
He got his 76.5 per million from the Advisory Committee for Immunization Practices, a Center for Disease Control working group. CDC, of course, is more interested in hiding AEs and SAEs than reporting them. The same CDC that claims anaphylaxis only occurs at a rate of 0.2 to 0.5 per million vaccine doses. Let’s just say they have incentives to under count AEs and SAEs and leave it at that.
Back to Singer. He arrived at his number by examining the TriNetX Research Network, which has aggregated electronic health records for many of the largest healthcare organizations in the US. Ok.
He excludes people with a prior history of myocarditis. Also Ok.
He also requires patients to have visited a healthcare practitioner 1 month to 2 years prior to the index date (ie. a positive diagnosis), and 6 months to 2 years prior to that. Ostensibly, he claims this is to ensure a “reasonable diagnosis history and relationship” with the Health Care Organization. Not Ok.
This would exclude a large amount of individuals with uncertain population dynamics. On the premise of reducing bias in his analysis, he has injected a huge amount of bias. We do not know the way the bias trends, but I can only assume (given his ridiculous chart) that he is looking for an answer here. He then says he excluded people with a negative test and a positive diagnosis who did not have a positive test within 14 days. What he does not say is he did not exclude people that did not have a positive test at all. In other words, a positive diagnosis without confirmation is enough for him despite the fact that hospitals were given strong incentives to relate everything to the pandemic.
He found 6 cases of myocarditis in 6,846 male patients between 12-17. 6. In total for males and females between 12-19, he found 20 cases, 12 of which were 19-82 days after the positive diagnosis (he did not get any more specific, nor did he provide a graph to show on which end of that spectrum the majority of the cases were). His argument is that myocarditis can have a delayed onset. That would be a fine argument, but he is comparing this to the CDC estimate of myocarditis after the vaccination, which does not allow for a delayed onset of myocarditis. If it isn’t within the first few days of the vaccine, it is not getting counted. Period. In other words, Singer is doing Singer things.
Even worse, he did not use any controls. He had access to a massive health database, minimized the group he was analyzing in the most arbitrary way, and did not even control for things like body mass index, at the very least? His adjustment for actual cases of the virus was also only 3.7 (adjusted case rate is 213 per million), which is (again) low considering he was going by cases that ended up being diagnosed in the medical system. Actual cases were likely much, much higher.
I suppose I can also point out the obvious:
Those children with natural immunity do not stand to benefit from the vaccine at all, so the vaccine is all downside risk.
Everyone gets vaccinated. Meaning everyone is exposed to the risk, not just those who catch the virus.
The vaccinations last for a finite time period. The risk of getting the virus in this period is either low enough not to be worth exposing oneself to the risk1, or high enough that a large proportion of the population has only downside risk (1).
Oh yes, there will be boosters!
I have to wonder if Mendel Singer got his MPH and PhD at clown college because YIKES!
For example, let’s pretend his numbers are correct and there is an (myocarditis) incidence of 213 cases per million for males 12-17 that catch the virus. Let’s also imagine that this not dependent on factors like BMI. And let’s imagine that the CDC numbers of 76.5 per million are accurate! Up until this point, he assumed that only 1 in 10 people caught the virus in over a year of the pandemic (slightly less, actually, but let’s round up to his benefit). We can go even further and assume that natural immunity does not exist and people are equally likely to catch the virus. Now, all of the assumptions are extremely biased in his favor. Even with his completely doctored numbers, no pun intended, the benefit does not outweigh the risk. The expected incidence of myocarditis from catching the virus in any given year would be 21.3 per million. The expected incidence of myocarditis from a full vaccine dosage is 76.5 million. Thus, the vaccine would have to remain highly effective (100% effective to be exact) for more than three and a half years for the vaccine to be worth it or the boosters would have to have absolutely 0 chance of causing myocarditis. Considering, by his numbers again, the second dose is much more risky than the first dose, how could anyone expect that assumption to hold? Singer literally lied with statistics and he still did not prove that the benefit is worth the risk in children. Amazing.