Dec 28, 2022·edited Dec 28, 2022Liked by Brian Mowrey
Thanks so much for your input on this one. 🙏👍🏽. That post struck me as well, but I didn't know about the statistical gymnastics. I look forward to your input on the current RSV wave. I don't assume that all of these sick kids are vaccinated. I was concerned after reading the Radigast article on the shift to Igg4, a paper I know you covered months ago. He or she contributes the current waves of rsv and flu to the rise in Igg4 from repeated covid injections. I'm not sure if you've read it, Igor did a recent post on it too.
Thank you - Rintrah doesn't actually present any new info, it's the same study. His fixation on IgG3 is a bit eccentric (not supported by prior consensus), which is fine but he doesn't really transparently tell the reader that fact. So now everyone's abuzz with IgG3 talk. It isn't an antibody class you probably want a lot of since it's very pro-inflammatory. It's like the stressed new worker on the first day of the job.
As far as neutralization, even his cited study admits that their results don't match most other studies "In contrast to our study, in which IgG3 was found to be the most important IgG subclass for neutralizing SARS-CoV-2, a number of smaller studies have found that IgG1 plays a more important role." Despite being "big," this study used pooled plasma which can distort the picture compared to what goes on in an individual. What would be expected based on prior research is that IgG1 and 4 are the best at in vitro neutralizing due to being more highly affinity-matured, but you could easily flip that just by using samples that are taken early after infection.
Again that's not to say Rintrah's theory is wrong, but it's beyond what the research says.
Thank you for the context. It's so valuable to read multiple science substackers to get close to the whole picture (the current picture anyway). You are really good with the details of the studies, and you seem to have experience to add context like this. 🙏👍🏽
Dec 27, 2022·edited Dec 27, 2022Liked by Brian Mowrey
When the article first came out I raised some concerns. I think I made a comment in Naked Emperor's Substack, although my first assessment was wrong because I misinterpreted the graphs (I'd have to look back to see what exactly I wrote).
Overall, I think the main problem were the weird inflection points near the end of the graphs, and I think I raised the same remarks in suggesting that there was no data outside of that inflection point, hence why all of the graphs end up having a wonky curve to them at that point because you don't really have any data points to use in order to create a curve.
Edit: I actually deleted the comment because upon reading it again I realized my interpretation was completely off, so never mind since I reached a conclusion based on bad assessments!
We all get led astray by the badass essments from time to time. I did check to see if you had also made a post about this one that I had missed. I was just so distracted by those deer lymph nodes!
In a morbid or sensual way? Never mind, I shall leave that a mystery!
I should have clarified that I deleted a comment where I copied and pasted what I wrote on Naked Emperor's Substack because for some reason I completely misinterpreted the curves and took the dips to coincide with a move to another curve, which was a completely wrong assessment even if it led to the same conclusions as there being the wonky curves.
After looking at it again I couldn't directly determine the weird curves (although I don't remember if I looked at the Supplemental data at the time), just that the curves seemed weird at those inflection points which may have meant something funky going on. I ended up not looking into it further. 🤷♂️
Send nodes, as they say. The inflections may reflect that for both the from-infection and from-injection time points, there's an initial off-wave honeymoon period which is always going to depress apparent efficacy since it's just a lot of counting false positives on both arms.
Wow, that's a pun I didn't know I needed! I'll put that away with my other tissue-related puns, which is rather sparse.
Interesting if that's the cause, I just suppose that such a sudden inflection and change in curve seems weird since the model used is so dependent on variables and the data used, and I would assume it should take on a smoother appearance I guess. 🤷♂️
I don't think any studies from the US are reliable for evaluating efficacy in kids. They all, either, take state health systems where kids who leave the state are incorrectly assumed to still be there, or use healthcare systems where sick kids are over-represented (like TriNetX), or use test-negative designs which I avoid like the plague.
If you mean Omicron reinfection efficacy, that's going to be a really wonky one. In the first place I think these must be mostly post-vaccine infections, otherwise the confidence intervals would be even wider, so the vaccine has already dropped the ball. Then there's potential artifacts of survivorship bias and things like that. It's really a dog's breakfast as they say. I would predict that what is really happening, regardless of whether any US study could show it, is that very early breakthrough infections probably are not going to generate durable immunity. Just like they don't represent most (later) breakthrough infections in terms of viral load and clearance time, they are also not as good at making immunity, and these early semi-infections are why there is a little bit higher rate of reinfection like in the difference between A and C for 4-8 months in Fig 3 of https://www.medrxiv.org/content/10.1101/2021.12.04.21267114v1
You got a retweet by Jikkyleaks today : )
https://twitter.com/ollys7/status/1608215431365464064
"The AP’s fact check is a study in presenting information in an insincere, unconfident, and deceitful-seeming manner."
This gets my vote as best official definition for corp fact check!
Thank you. I like to imagine the writer being like, "Wait, we're not lying this time? How do I write that?"
Thanks so much for your input on this one. 🙏👍🏽. That post struck me as well, but I didn't know about the statistical gymnastics. I look forward to your input on the current RSV wave. I don't assume that all of these sick kids are vaccinated. I was concerned after reading the Radigast article on the shift to Igg4, a paper I know you covered months ago. He or she contributes the current waves of rsv and flu to the rise in Igg4 from repeated covid injections. I'm not sure if you've read it, Igor did a recent post on it too.
https://www.rintrah.nl/the-trainwreck-of-all-trainwrecks-billions-of-people-stuck-with-a-broken-immune-response/
It was good to read Modern Discontent post saying in the age of omicron, the shift to igg4 may not be a train wreck.
https://moderndiscontent.substack.com/p/a-few-shower-thoughts-on-igg4-and/comments
Thank you - Rintrah doesn't actually present any new info, it's the same study. His fixation on IgG3 is a bit eccentric (not supported by prior consensus), which is fine but he doesn't really transparently tell the reader that fact. So now everyone's abuzz with IgG3 talk. It isn't an antibody class you probably want a lot of since it's very pro-inflammatory. It's like the stressed new worker on the first day of the job.
As far as neutralization, even his cited study admits that their results don't match most other studies "In contrast to our study, in which IgG3 was found to be the most important IgG subclass for neutralizing SARS-CoV-2, a number of smaller studies have found that IgG1 plays a more important role." Despite being "big," this study used pooled plasma which can distort the picture compared to what goes on in an individual. What would be expected based on prior research is that IgG1 and 4 are the best at in vitro neutralizing due to being more highly affinity-matured, but you could easily flip that just by using samples that are taken early after infection.
Again that's not to say Rintrah's theory is wrong, but it's beyond what the research says.
Thank you for the context. It's so valuable to read multiple science substackers to get close to the whole picture (the current picture anyway). You are really good with the details of the studies, and you seem to have experience to add context like this. 🙏👍🏽
Looks to me like it was just a propaganda piece designed to scare parents into thinking they need to boost up their kids
Exactly.
When the article first came out I raised some concerns. I think I made a comment in Naked Emperor's Substack, although my first assessment was wrong because I misinterpreted the graphs (I'd have to look back to see what exactly I wrote).
Overall, I think the main problem were the weird inflection points near the end of the graphs, and I think I raised the same remarks in suggesting that there was no data outside of that inflection point, hence why all of the graphs end up having a wonky curve to them at that point because you don't really have any data points to use in order to create a curve.
Edit: I actually deleted the comment because upon reading it again I realized my interpretation was completely off, so never mind since I reached a conclusion based on bad assessments!
We all get led astray by the badass essments from time to time. I did check to see if you had also made a post about this one that I had missed. I was just so distracted by those deer lymph nodes!
In a morbid or sensual way? Never mind, I shall leave that a mystery!
I should have clarified that I deleted a comment where I copied and pasted what I wrote on Naked Emperor's Substack because for some reason I completely misinterpreted the curves and took the dips to coincide with a move to another curve, which was a completely wrong assessment even if it led to the same conclusions as there being the wonky curves.
After looking at it again I couldn't directly determine the weird curves (although I don't remember if I looked at the Supplemental data at the time), just that the curves seemed weird at those inflection points which may have meant something funky going on. I ended up not looking into it further. 🤷♂️
Send nodes, as they say. The inflections may reflect that for both the from-infection and from-injection time points, there's an initial off-wave honeymoon period which is always going to depress apparent efficacy since it's just a lot of counting false positives on both arms.
Wow, that's a pun I didn't know I needed! I'll put that away with my other tissue-related puns, which is rather sparse.
Interesting if that's the cause, I just suppose that such a sudden inflection and change in curve seems weird since the model used is so dependent on variables and the data used, and I would assume it should take on a smoother appearance I guess. 🤷♂️
Be careful you dont get Epsteined
You mean the writer who got in trouble for saying we shouldn't call Jill Biden "Dr"?
no the dude with the Island
Ah, everyone knows that was suicide. The guards even called the hotline shortly beforehand.
I'm still wondering how they convicted JISSlane without identifying the pedos she was supplying.
I don't think any studies from the US are reliable for evaluating efficacy in kids. They all, either, take state health systems where kids who leave the state are incorrectly assumed to still be there, or use healthcare systems where sick kids are over-represented (like TriNetX), or use test-negative designs which I avoid like the plague.
If you mean Omicron reinfection efficacy, that's going to be a really wonky one. In the first place I think these must be mostly post-vaccine infections, otherwise the confidence intervals would be even wider, so the vaccine has already dropped the ball. Then there's potential artifacts of survivorship bias and things like that. It's really a dog's breakfast as they say. I would predict that what is really happening, regardless of whether any US study could show it, is that very early breakthrough infections probably are not going to generate durable immunity. Just like they don't represent most (later) breakthrough infections in terms of viral load and clearance time, they are also not as good at making immunity, and these early semi-infections are why there is a little bit higher rate of reinfection like in the difference between A and C for 4-8 months in Fig 3 of https://www.medrxiv.org/content/10.1101/2021.12.04.21267114v1