Science A substack about adverse event reports

[crOOk]

I found this substack very interesting

Are the modRNA vaccines enhancing SARS-COV-2 associated disease?

VAERS data suggest that COVID-19 modRNA vaccines introduced myocarditis as a disease risk

knowhatamine.substack.com

Translating report proportions to incidence proportions

Putting an end to our search for the elusive underreporting factor

knowhatamine.substack.com

The underreporting factor is 1073

for incidents of any type occuring within 7 days after dose 1

knowhatamine.substack.com

How many children had their hearts broken?

It will break mine to tell you.

knowhatamine.substack.com

The author also runs the website pervaers.com, which is directed at doctors and patients with an affinity for numbers:​

perVAERS

This is what the CDC is not showing us: Proportional reporting differences of vaccine adverse events between VAERS reports for different types of vaccines.

pervaers.com

 

[Nurse Ratched]

Hi @crOOk

I'm gonna shoot your thread over to our CEP sub forum ( current events and politics ) as your post is related to the Covid vaccines among other topics.

 

[crOOk]

Thank you and sorry for cross-posting!

 

[Nurse Ratched]

Thank you and sorry for cross-posting!

No apologies necessary . It did kind of relate to health and recovery as well so i will keep my eye on this thread and if it isn't going the way i think it should i am going to move it back.

 

[crOOk]

And here is my final article of the core ones I had planned:

How many children had their hearts broken?

It will break mine to tell you.

knowhatamine.substack.com

In it I use logical inference to estimate how many American children suffered heart damage as a consequence of being given the modRNA vaccines.

Spoiler alert: Roughly one million

Yes, I am sure. I have spent 2000 hours on the VAERS data. I know my way around it.

Yes, I may be a bit alarmist, but these numbers don't lie. However am in good spirits that the general 5-year-survival-rate of 50% for myocarditides applies here.

We will find out within the next 4 years. My sincerest apologies go out to everyone whose day I might have ruined with this revelation.

No apologies necessary . It did kind of relate to health and recovery as well so i will keep my eye on this thread and if it isn't going the way i think it should i am going to move it back.

Thank you very much.

 

[Skorpio]

The VAERs is an unblinded data set, and is subject to the nocebo effect. You cannot compare it to the average population. If there was a group that thought they got the vaccine but didn't, self reporting would be a valid comparison.

 

[thujone]

I appreciate the effort gone in to make this and hope that more data lead to more questions and more investigation. Right now there aren't enough questions being asked and a lot of data that would be relevant here simply isn't even being collected.

A lot is said by what's not said.

 

[Skorpio]

I appreciate the effort gone in to make this and hope that more data lead to more questions and more investigation. Right now there aren't enough questions being asked and a lot of data that would be relevant here simply isn't even being collected.

A lot is said by what's not said.

Look, I'm on the fence about keeping this here or moving it into the dive. The fact that it is a large amount of data analysis gives me hope that it can be discussed with concepts of rigor. However if this thread devolves into whataboutism and gish galloping, I am going to move it.

@cr00k do you have access to any placebo controlled data? These analyses would be valid if the vaccinated and non vaccinated groups both thought they were vaccinated.

 

[crOOk]

The VAERs is an unblinded data set, and is subject to the nocebo effect. You cannot compare it to the average population. If there was a group that thought they got the vaccine but didn't, self reporting would be a valid comparison.

@Skorpio
There is no reliable class 1 evidence I know of.

That is why I'm studying the distribution of side effects among spontaneous reports. They are sufficiently randomized for common symptoms to be reported at the same proportion at which they occur.

I reviewed this in my latest article by comparing incidence proportions from a Polish study with report proportions in an age-dose-sex-adjusted VAERS report group.

I'll gladly accept any type of criticism of my methodology, unless it's "you cannot do that".

Being aware what type of catastrophic data foundation all the claims about efficacy and safety rest (6months, 22k healthy adult participants in the case of Pfizer), it seems bold to just discard a million adverse event reports as useless.

Data always tells a story. It's all about how you approach it. You have to understand how it is collected.

So we can keep claiming we have nothing or we can start taking the greatest post-approval pharmacovigilance database seriously. There are many more stories in that database.

I have since adjusted my estimate to 750k US children after removing 80% of myocarditis cases for suspected overrepresentation due to raised awareness. The article isn't ready yet.

All this is is building a statistical model that helps us understand what exactly is going on. I am of course biased after what was done to us, but I am trying to be as honest as I can.

 

[crOOk]

Look, I'm on the fence about keeping this here or moving it into the dive. The fact that it is a large amount of data analysis gives me hope that it can be discussed with concepts of rigor. However if this thread devolves into whataboutism and gish galloping, I am going to move it.

@cr00k do you have access to any placebo controlled data? These analyses would be valid if the vaccinated and non vaccinated groups both thought they were vaccinated.

I do always use a reference group of people who received other vaccines. Report proportions of reference groups are substracted from the respective proportions in their study groups, resulting in "proportional differences".

For pervaers.com I built a pseudo-placebo reference group for every report group. It's complex, but I explained it on https://help.pervaers.com

All in all I calculate 10s of millions of confidence intervals for every update.

 

[Skorpio]

@crOOk
My issue with the VAERs data is that it is not randomized, it is subject to the rapid travel of the idea that the covid vaccine causes myocarditis. This effect operates in a similar manner to Morgellon's disease or gang stalking, where the presence of biasing information causes individuals to alter reporting. I know you touch on that with the Herpes Zoster article, but there was nowhere near the viral spread of that information when compared to myocarditis.

I do want to say, I agree that the covid vaccines cause some increase in the presentation of myocarditis, as it induces a pretty robust inflammatory state. Early reports (less likely to be influenced by memetic information) cite the incidence around 10 to 30 per 100,000 of adolescent males.

In 2015 the incidence of myocarditis was reported to be 2.5 million individuals. These cases are unlikely to be caused by covid vaccines. Due to the large number of individuals vaccinated, a large number of individuals will experience myocarditis unrelated to covid vaccination after being vaccinated.

A meta analysis of placebo controlled vaccine trials reports that 76% of systemic adverse events were reported by placebo arms of the trials. This indicates a strong nocebo effect. This does not mean that the remaining 24% of reports are true reports, just that they are not definate nocebo. I think even scaling your estimation back 80% due to over reporting produces a great overestimate of cases. Furthermore these studies enforced reporting of side effects, which captures a more accurate sample than people who are motivated to self report.

I want to see the null-hypothesis that the increase in myocarditis is due to the nocebo effect disproven before I will believe conclusions drawn from the VAERs data set. The data science axiom of "garbage in, garbage out" holds true here.

I don't think monkeypox vaccine recipients are subject to the same biases about myocarditis, and therefore would not be subject to overreporting. That is comparing somebody who got a drug and was told that they would likely become nauseaous, with a group which got saline and were not told such a thing.

One thing I see in the substack repeatedly is comparing over/under reporting of non-heart related side effects as a validation of heart related reports. These non-heart related side effects are not subject to the nocebo effect as myocarditis is. Therefore, they do not provide an accurate measure of myocarditis overreporting.

I appreciate your commitment to being transparent and honest, and the effort you have put in to this project.