Kinda like recording the CTs of billions of PCR tests over the last 18 months. Can you imagine what THAT data would be telling us? We would have had the most comprehensive set of data on testing for any 'disease' in human history. I wonder why it wasn't done ... [sarc]
more relevant is the literature from verifiable voting: without revealing much about who voted for whom, use cryptography to ensure that the voters can check that their votes were correctly counted. The same idea can be used for ensuring that tallies of medical outcomes, e.g. deaths after Covid or Covid vaccines, outcomes in medical trials e.g. death/ICU admission are verifiable and not fudged by pharma companies or governments too invested in some narratives. there were twitter reports of trial participants claiming that the trialists refused to acknowledge/register their adverse effects: nobody knows whether they were bots, lying humans, or honest humans. example project: https://www.cs.cornell.edu/projects/civitas/
We need to be getting participants in such systems to understand the importance of learning to use private-public key pairs. I recall feeling a certain amount of intimidation doing so for the first time, myself, even though number theory was my deepest area of interest during my undergraduate math studies. Heck, even setting up 2FA intimidates a lot of people the first time around.
This is one of those situations in which gamification [aimed at children] could certainly help. Playing games that involve the protection and use of information as parts of quests---habit training---would go a long way.
Thank you for commenting on the superset of ideas. I did think briefly about talking about voting systems, but perhaps the comment section was the best place for it to pop up for this particular article. Cheers.
I had never really thought much about crypto or blockchain before but after all the crazy stuff that's been going on with the covid pandemic, I now see why we badly need it.
It's happening in de-fi. Maybe it's a much simpler system because it is basically accounting and there are many fewer input devices. I imagine a de-med system would be complex, with multiple input type devices per user, and so much more info to track besides stats, but it would have such a great benefit. I fear that the push for improved medical recording that gained so much emphasis under the Obama administration will lead to huge centralization and control. The system you describe would contribute to securing our freedom. Hopefully.
Oh, they might do that. But it would be hard to disambiguate and would remain dissociated from your identity (if you so chose)---together in a query. And there might very well be professional data "curators" who do that. So...great, right?
I found your stack via Larry Turner's stack. Drat. My life is now officially consumed by Substack and my family are starting to look at me askance. Much of it goes way over my head but I endeavour to understand, out of a desperate bid for self-preservation within a storm of darkening forces. I look forward to reading what I can. That said, and on this piece, I am at the point of wanting to devolve from currency altogether. I am too old for this bitcoin stuff, and it leaves me anxious and perplexed. We have a friend who has jumped in, feet first. He is consumed by it and talks in foreign Blockchainese. I'm tending more towards the point of wanting to withdraw all of my money (what there is of it) altogether, and stuffing it under my mattress. To be used only on a rainy day. And then to launch into a life of self-sufficiency and payment-in-kind. I can understand swopping a cabbage or two for a bottle of honey better than I can the world of blockchains!
Not sure why it is so scandalous to start measuring for vaccine efficacy at 14 days. Immunologically it is known that vaccines take time to kick in immune protection since, unlike treatment like say monoclonal antibodies, they stimulate the immune system to act so you don't have immediate protection. Looking at Kaplan-Meier curves you can see the placebo and vaccine groups are nearly identical for 10-14 days when you can see the vaccine protection kick in and the differences manifest:
Now, if they ignored adverse events between days 0-14 after vaccine dosing, then I am 100% with you -- that would be inappropriate and lead to substantial bias that would suppress the type of effects you are convinced are occurring -- but they didn't -- they do count all adverse events during that time period.
"Not sure why it is so scandalous to start measuring for vaccine efficacy at 14 days. Immunologically it is known that vaccines take time to kick in immune protection since, unlike treatment like say monoclonal antibodies"
In all historical risk-benefit analyses, everything that happens from intention to treat onward is part of the treatment. You're a Professor of Biostatistics---you shouldn't need anyone to explain why that's the case. And in the case of an ongoing therapy, removing cases from those days at every window (not just the first) seems...just a wee bit defensive---particularly given that in the vast majority of nations larger outbreaks occured after vaccination programs began, suggesting something more than just a warming up of the B cell reactivity.
"Now, if they ignored adverse events between days 0-14 after vaccine dosing, then I am 100% with you -- that would be inappropriate and lead to substantial bias that would suppress the type of effects you are convinced are occurring -- but they didn't -- they do count all adverse events during that time period."
Except you DON'T know that, that's the whole point!
And we've had the Type II COVID-19 conversation before. Are you intentionally ignoring it, or not understanding that a disease can be caused by something other than a single source, and that source could itself be the therapy, and induce onset during the first 14 days. If we ignore that window, we're giving a free ride to potentially dangerous therapies. In particular...in this case where the spike protein is being used as part of the therapy, and evidence is increasingly suggesting that the spike protein has always been the primary difference between SARS-CoV-2 and other CoVs that cause far simpler symptoms.
But the entire point of this article is that your bullshit defense of making conclusions without properly collecting the data (like autopsies/virtopsies) would be entirely moot. There is no reason to be defensive of this solution EXCEPT to protect a centralized system that intends to manipulate the data.
It's telling that you had nothing at all to say about that aspect of distributed information technology. Whatever disagreements we might have...wouldn't it be better (meaning to the health of people, not authorities) to choose a system that bypasses those debates?
But since you're keen in temporal relationships, please comment on the WHO trials on hydroxychloroquine, Jeffrey. Let us know you apply your principles equally. I'll wait.
Yet, you're perfectly fine with the game of pretend where you compute mortality efficacy numbers without first observing that data (or mentioning that it might change the numbers dramatically if we call vaccine injuries due to spike protein "Type II COVID-19"). These scales aren't balanced. What's that you're doing with your thumb?
I never said anything about mortality efficacy numbers. I was talking about efficacy vs infection.
As I said, all adverse events including death should of course be counted from treatment time.
And efficacy vs infection numbers should be presented for all time periods but it is fine with me to split out for times just after vaccination for infections that may have started before vaccination or before protection kicked in.
And if in clinical trial it is higher in that time frame they should explore what is going on.
I'm not "upset". I'm "recentering the conversation around what is clearly important, which is finding out what the true risk-benefit looks like, and informing the public about what is important" instead of...dancing like a moth around the flame that is their narrative.
No they don't, they attribute deaths in this group to the "unvaccinated". This is an intention-to-treat issue but worse since they are not only ignoring deaths in the "newly vaccinated" group but attributing them to the opposite group
I agree deaths after vaccination should of course be counted in days 0-14 or any time in any analysis when evaluating vaccines.
There arena lot of sloppy analyses being done on observational data that get misleading results because of unaccounted bias — always a risk with observational data.
Some of these analyses oversell the benefit of vaccines and/or downplay risks, but many other analyses are more carefully done and adjust for many key potential biases and yield more meaningful results.
In terms of deaths, if one wants to stratify by days since vaccination and present deaths in the first 14 days separate from those later, there is nothing wrong with that. But certainly the deaths occurring in the first 14 days should also be evaluated carefully, and certainly not looped in with “unvaccinated groups”.
The careful analyses I’ve seen tend to split out these tune periods, not ignore then and not lump in with unvaccinated, but also not lumped in with those after vaccination protection kicks in.
The best analysis I have seen is Mark Reeder's which accounts for inappropriate matching as well as exclusions from treatment (due to illness) as well as the temporal effect of the 5 weeks delay between the vaccinated and unvaccinated groups on a downward incidence curve. https://zenodo.org/record/5243901 If you have a different analysis or data set please link to it
Wasn't the attack rate in all of the trials under 2% for each trial? I know it was for Pfizer's 6 month study. Would you really expect to have enough power to detect a difference over a 2 week period if the attack rate is under 2% over 6 months?
If you look at the Kaplan Meier curves from the clinical trial in the kinks I sent you see the placebo and vaccine lines are on top of each other for 10-14 days and the the vaccine line immediately flattens out while the placebo continues climbing at the same rate
Great idea!
Mandate autopsies.
Too simple a solution?
Or too 'inconvenient'?
Kinda like recording the CTs of billions of PCR tests over the last 18 months. Can you imagine what THAT data would be telling us? We would have had the most comprehensive set of data on testing for any 'disease' in human history. I wonder why it wasn't done ... [sarc]
I'm still researching an autopsy article. It's going to have a bomb or two.
more relevant is the literature from verifiable voting: without revealing much about who voted for whom, use cryptography to ensure that the voters can check that their votes were correctly counted. The same idea can be used for ensuring that tallies of medical outcomes, e.g. deaths after Covid or Covid vaccines, outcomes in medical trials e.g. death/ICU admission are verifiable and not fudged by pharma companies or governments too invested in some narratives. there were twitter reports of trial participants claiming that the trialists refused to acknowledge/register their adverse effects: nobody knows whether they were bots, lying humans, or honest humans. example project: https://www.cs.cornell.edu/projects/civitas/
Indeed.
We need to be getting participants in such systems to understand the importance of learning to use private-public key pairs. I recall feeling a certain amount of intimidation doing so for the first time, myself, even though number theory was my deepest area of interest during my undergraduate math studies. Heck, even setting up 2FA intimidates a lot of people the first time around.
This is one of those situations in which gamification [aimed at children] could certainly help. Playing games that involve the protection and use of information as parts of quests---habit training---would go a long way.
Thank you for commenting on the superset of ideas. I did think briefly about talking about voting systems, but perhaps the comment section was the best place for it to pop up for this particular article. Cheers.
I had never really thought much about crypto or blockchain before but after all the crazy stuff that's been going on with the covid pandemic, I now see why we badly need it.
It's happening in de-fi. Maybe it's a much simpler system because it is basically accounting and there are many fewer input devices. I imagine a de-med system would be complex, with multiple input type devices per user, and so much more info to track besides stats, but it would have such a great benefit. I fear that the push for improved medical recording that gained so much emphasis under the Obama administration will lead to huge centralization and control. The system you describe would contribute to securing our freedom. Hopefully.
What stops one entity from buying the data off you and then reselling to other entities or even just leaking it?
Oh, they might do that. But it would be hard to disambiguate and would remain dissociated from your identity (if you so chose)---together in a query. And there might very well be professional data "curators" who do that. So...great, right?
I found your stack via Larry Turner's stack. Drat. My life is now officially consumed by Substack and my family are starting to look at me askance. Much of it goes way over my head but I endeavour to understand, out of a desperate bid for self-preservation within a storm of darkening forces. I look forward to reading what I can. That said, and on this piece, I am at the point of wanting to devolve from currency altogether. I am too old for this bitcoin stuff, and it leaves me anxious and perplexed. We have a friend who has jumped in, feet first. He is consumed by it and talks in foreign Blockchainese. I'm tending more towards the point of wanting to withdraw all of my money (what there is of it) altogether, and stuffing it under my mattress. To be used only on a rainy day. And then to launch into a life of self-sufficiency and payment-in-kind. I can understand swopping a cabbage or two for a bottle of honey better than I can the world of blockchains!
Not sure why it is so scandalous to start measuring for vaccine efficacy at 14 days. Immunologically it is known that vaccines take time to kick in immune protection since, unlike treatment like say monoclonal antibodies, they stimulate the immune system to act so you don't have immediate protection. Looking at Kaplan-Meier curves you can see the placebo and vaccine groups are nearly identical for 10-14 days when you can see the vaccine protection kick in and the differences manifest:
Would be easier if I could post plots here -- but you can find the Pfizer one on page 30 of the FDA report: https://www.fda.gov/media/144245/download and for Moderna, on page 28 of their FDA report: https://www.fda.gov/media/144434/download
Now, if they ignored adverse events between days 0-14 after vaccine dosing, then I am 100% with you -- that would be inappropriate and lead to substantial bias that would suppress the type of effects you are convinced are occurring -- but they didn't -- they do count all adverse events during that time period.
"Not sure why it is so scandalous to start measuring for vaccine efficacy at 14 days. Immunologically it is known that vaccines take time to kick in immune protection since, unlike treatment like say monoclonal antibodies"
In all historical risk-benefit analyses, everything that happens from intention to treat onward is part of the treatment. You're a Professor of Biostatistics---you shouldn't need anyone to explain why that's the case. And in the case of an ongoing therapy, removing cases from those days at every window (not just the first) seems...just a wee bit defensive---particularly given that in the vast majority of nations larger outbreaks occured after vaccination programs began, suggesting something more than just a warming up of the B cell reactivity.
"Now, if they ignored adverse events between days 0-14 after vaccine dosing, then I am 100% with you -- that would be inappropriate and lead to substantial bias that would suppress the type of effects you are convinced are occurring -- but they didn't -- they do count all adverse events during that time period."
Except you DON'T know that, that's the whole point!
And we've had the Type II COVID-19 conversation before. Are you intentionally ignoring it, or not understanding that a disease can be caused by something other than a single source, and that source could itself be the therapy, and induce onset during the first 14 days. If we ignore that window, we're giving a free ride to potentially dangerous therapies. In particular...in this case where the spike protein is being used as part of the therapy, and evidence is increasingly suggesting that the spike protein has always been the primary difference between SARS-CoV-2 and other CoVs that cause far simpler symptoms.
But the entire point of this article is that your bullshit defense of making conclusions without properly collecting the data (like autopsies/virtopsies) would be entirely moot. There is no reason to be defensive of this solution EXCEPT to protect a centralized system that intends to manipulate the data.
It's telling that you had nothing at all to say about that aspect of distributed information technology. Whatever disagreements we might have...wouldn't it be better (meaning to the health of people, not authorities) to choose a system that bypasses those debates?
But since you're keen in temporal relationships, please comment on the WHO trials on hydroxychloroquine, Jeffrey. Let us know you apply your principles equally. I'll wait.
I am in favor of full data transparency and always have been
I am in favor of autopsies to be done
Yet, you're perfectly fine with the game of pretend where you compute mortality efficacy numbers without first observing that data (or mentioning that it might change the numbers dramatically if we call vaccine injuries due to spike protein "Type II COVID-19"). These scales aren't balanced. What's that you're doing with your thumb?
https://roundingtheearth.substack.com/p/gaming-measurement-of-vaccine-efficacy
I never said anything about mortality efficacy numbers. I was talking about efficacy vs infection.
As I said, all adverse events including death should of course be counted from treatment time.
And efficacy vs infection numbers should be presented for all time periods but it is fine with me to split out for times just after vaccination for infections that may have started before vaccination or before protection kicked in.
And if in clinical trial it is higher in that time frame they should explore what is going on.
Not sure what upsets you about this so
I'm not "upset". I'm "recentering the conversation around what is clearly important, which is finding out what the true risk-benefit looks like, and informing the public about what is important" instead of...dancing like a moth around the flame that is their narrative.
Any comment yet on the temporal relationship regarding HCQ treatment in the WHO trials? Seem like a simple way to establish good faith.
Because the spike protein is most likely the only part entirely designed as a bioweapon. Please see https://medium.com/@SherlockGNomes/an-engineered-origin-for-the-sars-cov2-genome-47914a6919ad
No they don't, they attribute deaths in this group to the "unvaccinated". This is an intention-to-treat issue but worse since they are not only ignoring deaths in the "newly vaccinated" group but attributing them to the opposite group
In the clinical trial they did not.
What data are you taking about?
I agree deaths after vaccination should of course be counted in days 0-14 or any time in any analysis when evaluating vaccines.
There arena lot of sloppy analyses being done on observational data that get misleading results because of unaccounted bias — always a risk with observational data.
Some of these analyses oversell the benefit of vaccines and/or downplay risks, but many other analyses are more carefully done and adjust for many key potential biases and yield more meaningful results.
In terms of deaths, if one wants to stratify by days since vaccination and present deaths in the first 14 days separate from those later, there is nothing wrong with that. But certainly the deaths occurring in the first 14 days should also be evaluated carefully, and certainly not looped in with “unvaccinated groups”.
The careful analyses I’ve seen tend to split out these tune periods, not ignore then and not lump in with unvaccinated, but also not lumped in with those after vaccination protection kicks in.
The best analysis I have seen is Mark Reeder's which accounts for inappropriate matching as well as exclusions from treatment (due to illness) as well as the temporal effect of the 5 weeks delay between the vaccinated and unvaccinated groups on a downward incidence curve. https://zenodo.org/record/5243901 If you have a different analysis or data set please link to it
He asks what data you're talking about. The funny thing is, I mentioned the problem Mark noticed to Jeffrey weeks ago. I wish I were kidding.
Because vaccine eff for these gene therapies is negative due to immunosuppressive effect first 14-21 days.
Then why did the clinical trials not demonstrate higher infection rates in days 14-21?
Do you mean days 0-21? Hache didn't say days 14-21, he said the first 14-21 days. Are there higher infection rates days 0-21?
And sorry I misread Hache as saying days 14-21 but either way the data don’t show any increase of infections in those days.
Of course the known common side effects of vaccines happen in that time especially in the 1-2 days immediately after vaccination.
Sure it does. Look at the danish study. Negative efficacy in that period of time. boriquagato.substack.com has written extensively on the topic.
Can you send me link to the danish observational study?
Wasn't the attack rate in all of the trials under 2% for each trial? I know it was for Pfizer's 6 month study. Would you really expect to have enough power to detect a difference over a 2 week period if the attack rate is under 2% over 6 months?
If you look at the Kaplan Meier curves from the clinical trial in the kinks I sent you see the placebo and vaccine lines are on top of each other for 10-14 days and the the vaccine line immediately flattens out while the placebo continues climbing at the same rate
*links