Coronavirus

[AlanSubie4Life]

4X the total numbers, but remember we don't 100mil kids per year in the USA get influenza

Lol. There are 75 million kids in the US! Just how many do you think have had COVID???

Seems to be serious detachment from the data here.

It’s a serious disease, in children, still worse than most influenzas. That is what the data show.

And if someone had a kid die do you say it isn’t actually any worse than influenza, it is just more infectious?

 

[PACEMD]

remember we don't 100mil kids per year in the USA

Just the thought of 100mil kidsing is enough to hop me on the bandwagon of antikidsing........

 

[bkp_duke]

While I am all in for better vaccine efficacy, I would still say that the raw numbers are more important than the rates.

Would you rather have 100 kids die or 1000 kids die? I just don't care all that much about ratios.

I would rather have zero kids die. But there is utopian desire, and reality.

Compared to prior pandemics, COVID-19 was very VERY mild on the young. Things like prior influenza pandemics decimated that population (1918 Spanish flu being the most recent).

 

[PACEMD]

Would you rather have 100 kids die or 1000 kids die? I just don't care all that much about ratios.

So that ratio is either 10 or 0.1, if my covid math is on the button........

 

[Norbert]

Well, here's the problem.

IF your CFR is right (and I don't think it is, see below) - then that would be damning evidence for a really piss-poor efficacy on the vaccine. It's primary design goal - reduction of death - would have been a failure.

What I think is going on is that the data, particularly that over the last 18 months, is inaccurate. The cases reported is low, too low, because previously when people tested at home and had a positive they called their doc and/or went to the ER and that was a "reportable case", in addition to those tests found positive by labs, Dr. offices, and hospitals. Now? People aren't reporting it. My dad (3 shots) tested positive at home on a rapid test this week with COVID. He's 79, has URI symptoms and some malaise, but is otherwise improving (with strict instructions from his physician son to call 911 if he has any difficulty breathing). He didn't notify his doc or anything else. But he's also an ox and will probably die at about 99-105 (per the norm of his family)

So, your cases number is low, that is cutting the denominator, and thereby increasing the calculated rate.

It's some grey zone between CFR and IFR, in reality.



But - your own data, albeit somewhat incomplete (per above), shows that efficacy of the vaccine against generalized illness leaves something to be desired.

I don't quite follow the argument regarding the term "CFR".

By my understanding so far (and what it sounds like in Wikipedia using "diagnosed cases" and "confirmed cases"), we have used "CFR" as referring to the number of diagnosed cases. That is, if the test rate goes down, the CFR goes down as well. But, theoretically, not the IFR, so with less testing the discrepancy between CFR and IFR increases.

The discrepancy is undiagnosed asymptomatic cases plus undiagnosed symptomatic cases. CFR by this definition does not include any undiagnosed cases.

To me it would seem that the number you are looking for would be between CFR and IFR, whereas the number Alan is thinking about is in fact the CFR as defined.

At least that is how we used the terms here at the time when I was involved in such calculations, IIRC.

Or maybe you worded that in a way that I misunderstand.

 

[AlanSubie4Life]

I agree.......

A perfectly matched mRNA vaccine will not magically stop all infections. Would be likely about the same result as current bivalent vaccines!

Pretty clear that is the way it works now. Omicron is just not that immunogenic and can squeak through.

However it will of course improve outcomes due to the broad match of epitopes and the other immune system strengthening it provides. (Just like the current bivalent vaccine does, as it is quite well matched!!!)

It would be great to get the nasal vaccines that already exist but unfortunately likely a long road here.

But no need for perfect to be the enemy of the good.

 

[bkp_duke]

I don't quite follow the argument regarding the term "CFR".

By my understanding so far (and what it sounds like in Wikipedia using "diagnosed cases" and "confirmed cases"), we have used "CFR" as referring to the number of diagnosed cases. That is, if the test rate goes down, the CFR goes down as well. But, theoretically, not the IFR, so with less testing the discrepancy between CFR and IFR increases.

I think you have the logic a little backwards on this part.

Assumption - fatalities (the numerator) isn't changing. Those are going to be reported regardless.

If you are testing less, the denominator goes down, thereby increasing the calculated fraction. So CFR would go up with less testing (deaths remain the same, but those are spread out over fewer test cases).


And yes, to the rest of your post. IFR is basically a theoretical number that we can't ever measure. "Case" fatality rate comes down to how you define case, and there are sub-divisions there. I.e. "laboratory" CFR (i.e. you are including all lab tests for covid) or "symptomatic" CFR (i.e. only those "cases" with symptoms that present for medical care).

 

[PACEMD]

It would be great to get the nasal vaccines that already exist but unfortunately likely a long road here.

I'm waiting for the sugar cube vaccine. I still to this day remember 1960 Sabin Sunday when I had to shed my polio....

 

[PACEMD]

But no need for perfect to be the enemy of the good.

Exactly, I like to think of myself as the enemy of good!

 

[PACEMD]

My wife is sitting here giggling, against her will, and calling me a wiseass. I corrected her, you mean dumbass.....

Like a jackass but too lazy to get up off the couch. I don't even have to say........hold my beer........

 

[Norbert]

This is helpful. Thank you for posting it.



This is so circumstantial it can hardly be considered evidence. Although I agree that in all likelihood there was an animal source, the language used makes it feel like they have a hypothesis and are bending over backwards to reinforce it. It's along the lines of "we think he might have walked along the street the house is on the night of the murder, therefore he did it." Maybe there will be more, but so far it's not there.

I agree in a sense, however it understates the relevance of the new data (in so far as I understand it). In the absence of having conclusive data either way, you look for circumstantial evidence. And when the circumstantial evidence available gets as close to being consistent with a theory as you could hope for, it says something. If you don't have data about infected racoon dogs from that time, only samples from the market's surfaces, then the next best thing you can hope for (apparently) is samples with both virus and animal DNA appearing in the same single sample, at the exact same location. With racoon dogs being a viable candidate for animal transmission.

To me it sounds that is like finding out someone with a gun was at the exact location where someone was killed with a bullet, but maybe that is now overstating it. EDIT: Maybe better to say someone who had known access to a gun shop.

 

[AlanSubie4Life]

whereas the number Alan is thinking about is in fact the CFR as defined.

Underlying this, as you can tell from my posts, I am always driving at IFR (and more broadly disease burden). That is why I explicitly addressed that with ascertainment discussion.

IFR is very clearly bounded now. For this 0-17 group, it literally cannot be better than about 1 in 37500. (There is the complexity of multiple infections though and how you decide to count that.). The CFR is about 1/4 of that.

 

[Norbert]

I think you have the logic a little backwards on this part.

Assumption - fatalities (the numerator) isn't changing. Those are going to be reported regardless.

If you are testing less, the denominator goes down, thereby increasing the calculated fraction. So CFR would go up with less testing (deaths remain the same, but those are spread out over fewer test cases).

Sure, of course, it goes up. What goes down is the number of diagnosed cases, with less testing. Simple mistake writing the post.

And yes, to the rest of your post. IFR is basically a theoretical number that we can't ever measure. "Case" fatality rate comes down to how you define case, and there are sub-divisions there. I.e. "laboratory" CFR (i.e. you are including all lab tests for covid) or "symptomatic" CFR (i.e. only those "cases" with symptoms that present for medical care).

The fact that often only or mostly symptomatic cases are tested is generally part of the problem with CFR as I see the term used without qualifiers. I'd think that if you have access to tests that include asymptomatic cases in representative numbers, that could be used to estimate the IFR. Like tests in NY that attempted to test representative random people on the street.

 

[bkp_duke]

Sure, of course, it goes up. What goes down is the number of diagnosed cases, with less testing. Simple mistake writing the post.



The fact that often only or mostly symptomatic cases are tested is generally part of the problem with CFR as I see the term used without qualifiers. I'd think that if you have access to tests that include asymptomatic cases in representative numbers, that could be used to estimate the IFR. Like tests in NY that attempted to test representative random people on the street.

But in CFR - cases is not the numerator. It's fatalities. That number is effectively unchanged. If someone dies, we know how they died in the USA (i.e. unknown causes = autopsy, and that would have a viral panel that includes COVID).

That's what I was getting at, the numerator doesn't change in this scenario.



Anyway, there are calculations to estimate the IFR, but you are getting deep into the epidemiology weeds there.

 

[DrGriz]

I would rather have zero kids die. But there is utopian desire, and reality.

Compared to prior pandemics, COVID-19 was very VERY mild on the young. Things like prior influenza pandemics decimated that population (1918 Spanish flu being the most recent).

I submit that while there won't be zero kids dying, the question is, is there something we can do to lower the absolute number of kids dying, and should we do it? If vaccination brings down the absolute number, then why shouldn't we be doing it, especially since it kills more kids than flu (which we vaccinate against)? Even if the current vaccinations don't reduce transmission (just like flu vaccine doesn't), if fewer kids die, what's the downside?

The concern about myocarditis you correctly brought up is, from what I have read, almost completely mitigated by using (lower dose) Pfizer vs Moderna. So I feel we can have at it.

 

[Norbert]

But in CFR - cases is not the numerator. It's fatalities. That number is effectively unchanged. If someone dies, we know how they died in the USA (i.e. unknown causes = autopsy, and that would have a viral panel that includes COVID).

That's what I was getting at, the numerator doesn't change in this scenario.



Anyway, there are calculations to estimate the IFR, but you are getting deep into the epidemiology weeds there.

I don't know if it matters for the substance of your discussion, it's just that it made it more difficult for me to try follow the discussion (which is probably complicated anyway) when you were apparently using terms differently among the two of you.

 

[DrGriz]

I agree in a sense, however it understates the relevance of the new data (in so far as I understand it). In the absence of having conclusive data either way, you look for circumstantial evidence. And when the circumstantial evidence available gets as close to being consistent with a theory as you could hope for, it says something. If you don't have data about infected racoon dogs from that time, only samples from the market's surfaces, then the next best thing you can hope for (apparently) is samples with both virus and animal DNA appearing in the same single sample, at the exact same location. With racoon dogs being a viable candidate for animal transmission.

To me it sounds that is like finding out someone with a gun was at the exact location where someone was killed with a bullet, but maybe that is now overstating it. EDIT: Maybe better to say someone who had known access to a gun shop.

Well, from what I have read so far, they haven't found 'coon dog DNA along with Covid RNA. But they have found Human DNA. So...

 

[bkp_duke]

I submit that while there won't be zero kids dying, the question is, is there something we can do to lower the absolute number of kids dying, and should we do it? If vaccination brings down the absolute number, then why shouldn't we be doing it, especially since it kills more kids than flu (which we vaccinate against)? Even if the current vaccinations don't reduce transmission (just like flu vaccine doesn't), if fewer kids die, what's the downside?

The concern about myocarditis you correctly brought up is, from what I have read, almost completely mitigated by using (lower dose) Pfizer vs Moderna. So I feel we can have at it.

Theoretical discussion, I agree with you. No arguments there.

PRACTICAL discussion - if you haven't been a pediatrician and tried to convince parents about the necessity of a vaccine, it's hard to understand the difficulty associated with this. Even before COVID, even the most ardent pro-vaccine parents normally were hesitant with "new" vaccines. They wanted them out in the wild for a while, tested better, and more data, before they subjected their kids to them. To those that weren't of that persuasion but were not anti-vaxxers, it could be a really really hard sell.

So, it's tough for the pedi's out there to tell the parents they should have their kids take the vaccine, but it's not the best anymore because coverage is sub-optimal, and that in the boys there is a very small, but still possible, chance that they may have chest pain. Parents weigh that against the "well, they didn't get really sick from COVID when it was bad, they will probably do just fine now".

It's the same problem pediatricians have with the seasonal influenza vaccine. It's truly a CRAPPY vaccine. We don't have these problems with things like Tetanus, which only need boosters every 5-10 years.

Will people here (many that are not parents) argue against that mindset? Yep.
But it's not their kids, they don't get to make that decision for others.

 

[DrGriz]

Theoretical discussion, I agree with you. No arguments there.

PRACTICAL discussion - if you haven't been a pediatrician and tried to convince parents about the necessity of a vaccine, it's hard to understand the difficulty associated with this. Even before COVID, even the most ardent pro-vaccine parents normally were hesitant with "new" vaccines. They wanted them out in the wild for a while, tested better, and more data, before they subjected their kids to them. To those that weren't of that persuasion but were not anti-vaxxers, it could be a really really hard sell.

So, it's tough for the pedi's out there to tell the parents they should have their kids take the vaccine, but it's not the best anymore because coverage is sub-optimal, and that in the boys there is a very small, but still possible, chance that they may have chest pain. Parents weigh that against the "well, they didn't get really sick from COVID when it was bad, they will probably do just fine now".

It's the same problem pediatricians have with the seasonal influenza vaccine. It's truly a CRAPPY vaccine. We don't have these problems with things like Tetanus, which only need boosters every 5-10 years.

Will people here (many that are not parents) argue against that mindset? Yep.
But it's not their kids, they don't get to make that decision for others.

I won't argue your point there. It's an especially tough time for my FP partners (they are the de facto pediatricians here) to get parents to vaccinate their kids at all. That's not the same as not recommending it, however. And then, all you can do is all you can do.

 

[Norbert]

Well, from what I have read so far, they haven't found 'coon dog DNA along with Covid RNA. But they have found Human DNA. So...

Exactly that has changed:

Some of the new samples do have animal DNA/RNA and the virus in the same single sample:
From the Science article:

The analysis presented to the WHO panel this week now suggests some coronavirus-positive samples collected contained DNA or RNA from raccoon dogs, civets, and other mammals now known to be highly susceptible to SARS-CoV-2.

Your quotes are related to past papers from the chinese research group around Georg Gao, which is expected to publish new material. I actually thought your comment "Nothing to see here" was sarcasm.

The news that we are discussing now is based on new data that french researchers found in new samples posted by the chinese researchers which those didn't comment on yet. These samples were recently posted and then removed again. However they were examined by a group of western researchers who have not yet published a paper, nevertheless some of their findings were already reported by some media.