Coronavirus

[SageBrush]

The Lancet study is a real eye-opener.

What do you mean -- it is great news for 'Merika.

Now trumpers can take their bleach without the hassle of those pesky pills.

 

[GFernandes]

CDC made some good calculations, but their quality seems to vary wildly, their recent death count was around 60,000 when everyone else was counting around 90,000.

Now that the "task force" is picking on them, it'll surely get worse. I'm glad we still have people like Bergstrom who don't give up "Calling Bullshit" from a scientific point of view.

So we stop assuming the 0,1% sIFR from CDC? Maybe their quality varied wildly there

 

[TheTalkingMule]

You are ignoring the fact that the number of deaths by exogenous factors are absolutely nothing compared to untreated diseases. And some the exogenous factors are increasing ( suicides ) not all of them are decreasing.

Stress, alcoholism, suicides, domestic abuse, untreated illness, traumatized impoverished children. All these clearly far far outpace the barely any people we've save with the lockdown, they would outpace those saved even if that figure were magically 200k+. There's no way to prove it or even any stats to point to yet, so whats the point in getting these people all fired up?

 

[Norbert]

But you have time to assume the IFR for influenza from the same source ( CDC ) ? Lets be consistent.
Making CFR calculations from Worldometers is ridiculous to say the least.

IFR for influenza? You must be confusing me with someone else.

"CFR" refers to "cases", actually diagnosed people (not theoretically expected to be infected), and that's what those worldometers numbers are (for the numbers fo deaths, you could subtract the number of "probable deaths" vs "confirmed deaths" if you want, but that doesn't change the result very much).

Your arguments are usually extremely biased, to put it politely, and I probably won't respond any further to that.

 

[jhm]

The 0.27% IFR from the new CDC report doesn’t jibe with NYC, or Diamond Princess.

I sure wish they had published their calculations, so we could dig deeper.

Curious. NY state is up to 1485 deaths per million. If 0.27% IFR were in effect here, that would imply more than 55% of the state population have already been infected. But there are 18,832 confirmed cases per million. So this would imply that only 3.4% of infected cases have been confirmed. 23.5% of tests are positive, which is seems pretty low if 55% have actually been infected and most of the testing is with symptomatic and at-risk persons. So yeah, this does not seem to jibe.

By comparison if you assume say 0.50% IFR, the same calculations yield that 30% of the population has been infected. This is closer to the 23.5% test positive rate. So this is still a bit of a stretch (How do you explain test positive rate lower than the infection rate? You have to argue you're testing the wrong people at the wrong time.), but not as much of a stretch as 0.27 IFR would imply.

Honestly the NY does not look good for the herd enthusiasts. It pretty much comes down to infect everyone in the shortest amount of time and hope that IFR is mercifully low.

 

[Norbert]

So we stop assuming the 0,1% sIFR from CDC? Maybe their quality varied wildly there

What sIFR and who is the "we" that were assuming that? (No need to answer that.)

When considering time to death lag, I am assuming an IFR of around 1% or slightly below or slightly higher for confirmed deaths, an IFR of around 1.3% when including probable deaths, and an IFR of 1.4% to 1.6% when including excess deaths. As you might know, I am basing that on the NYC numbers which currently are the only ones I know that I think can be considered actually informative because of their scale and supporting information.

 

[SageBrush]

Looks like it's a "game changer" in the reverse direction, although again with a modest odds ratio.

trump and his coterie of morons aside, a conservative take on the paper is that HCQ does not reduce Covid-19 related mortality in the hospitalized. I would not be quick to say it is worse, mostly because it is likely that sicker patients were treated with the drug.

An analysis that corrected for Covid-19 severity when the drug was started would lead to stronger conclusions.

Does that leave an opening for HCQ ? Very likely not since the less sick patients would have less reason to accept the risk of HCQ induced dysrythmias.

---
It is tempting to speculate why HCQ is not panning out despite promising pharmacology and in-vitro studies. The most obvious is that side effects have more than offset any Covid benefit. This is the part that physicians know all too well, and trumpers are blind to. The second part that I suspect is the underlying pharmacokinetics: either the drug does not have enough time to build up to a level to make a difference in vivo, or high doses lead to unacceptable adverse effects. This is another way of saying that the Lupus safety profile cannot be extended to Covid.

 

[Norbert]

Curious. NY state is up to 1485 deaths per million. If 0.27% IFR were in effect here, that would imply more than 55% of the state population have already been infected.

And for New York City, it would mean that 72% would have to be infected even if you accept only "confirmed deaths".
If you include "probable deaths", 93% would have to be infected. If you dare to include excess deaths, more than 100% would have to be infected.

So obviously the IFR must be much larger than 0.27%.

 

[dfwatt]

trump and his coterie of morons aside, a conservative take on the paper is that HCQ does not reduce Covid-19 related mortality in the hospitalized. I would not be quick to say it is worse, mostly because it is likely that sicker patients were treated with the drug.

An analysis that corrected for Covid-19 severity when the drug was started would lead to stronger conclusions.

Does that leave an opening for HCQ ? Very likely not since the less sick patients would have less reason to accept the risk of HCQ induced dysrythmias.

---
It is tempting to speculate why HCQ is not panning out despite promising pharmacology and in-vitro studies. The most obvious is that side effects have more than offset any Covid benefit. This is the part that physicians know all too well, and trumpers are blind to. The second part that I suspect is the underlying pharmacokinetics: either the drug does not have enough time to build up to a level to make a difference in vivo, or high doses lead to unacceptable adverse effects. This is another way of saying that the Lupus safety profile cannot be extended to Covid.

You are right to raise the possibility that it's simply correlative as opposed to causal, but (as you say) this could be a researchable question if you have the way to dig into the data to see if chloroquine and hydroxychloroquine combos were used equally across hospitalized patients at various levels of severity. Any case, this should put the brakes on the wild enthusiasm by the right-wingers for hydroxychloroquine

 

[GFernandes]

What sIFR and who is the "we" that were assuming that? (No need to answer that.)

When considering time to death lag, I am assuming an IFR of around 1% or slightly below or slightly higher for confirmed deaths, an IFR of around 1.3% when including probable deaths, and an IFR of 1.4% to 1.6% when including excess deaths. As you might know, I am basing that on the NYC numbers which currently are the only ones I know that I think can be considered actually informative because of their scale and supporting information.

So you choose base your % from a city, instead of a whole country estimate that comprises all areas of the country. Isn't it normal that there are zones with higher % and lower % and that we should look at the average when comparing to, say, influenza? When we talk about influenza we talk about global averages, no?

 

[dfwatt]

Curious. NY state is up to 1485 deaths per million. If 0.27% IFR were in effect here, that would imply more than 55% of the state population have already been infected. But there are 18,832 confirmed cases per million. So this would imply that only 3.4% of infected cases have been confirmed. 23.5% of tests are positive, which is seems pretty low if 55% have actually been infected and most of the testing is with symptomatic and at-risk persons. So yeah, this does not seem to jibe.

By comparison if you assume say 0.50% IFR, the same calculations yield that 30% of the population has been infected. This is closer to the 23.5% test positive rate. So this is still a bit of a stretch (How do you explain test positive rate lower than the infection rate? You have to argue you're testing the wrong people at the wrong time.), but not as much of a stretch as 0.27 IFR would imply.

Honestly the NY does not look good for the herd enthusiasts. It pretty much comes down to infect everyone in the shortest amount of time and hope that IFR is mercifully low.

The New York statistics blow up all the right-wing obfuscation and pseudo statistics generated (like the joke Santa Clara study) to create the narrative that this is just the flu and that governments are infringing on our rights by asking us to observe any restrictions.

 

[Norbert]

So you choose base your % from a city, instead of a whole country estimate that comprises all areas of the country. Isn't it normal that there are zones with higher % and lower % and that we should look at the average when comparing to, say, influenza? When we talk about influenza we talk about global averages, no?

Obviously you didn't look at the Car Bergstrom thread that Alan referenced.

This graph likely uses "confirmed deaths" only, since "probable deaths" are usually not available.
As you can see, most fall in the 0.9% to 1.0% IFR range, and would be higher if other deaths were included.

 

[SageBrush]

An analysis that corrected for Covid-19 severity when the drug was started would lead to stronger conclusions.

My error -- the Lancet article does correct for disease severity.

 

[mikevbf]

Here is the weekly summary of NewYork-Presbyterian Hospital's update:

"Numbers: we are at 846 inpatients, 279 on ventilators (about 33%)

Columbia campus has most of these ventilated patients.

We are expanding our services. We are going to keep up the field hospital in case that we need it.

All 50 states now are working on reopening, and we need to watch that very carefully.

The ford plants had to close as there were some cases that they saw there. The CDC has come out and said that people to people transmission is more of an issue than surfaces.

Positivity rate is 3% of people who are tested.

This week we have announced a pilot project at lower manhattan hospital: we are allowing one visitor to come into the hospital for patients during the middle of the day with appropriate PPE, we have to screen them. Monitoring this very closely."

 

[Daniel in SD]

So you choose base your % from a city, instead of a whole country estimate that comprises all areas of the country. Isn't it normal that there are zones with higher % and lower % and that we should look at the average when comparing to, say, influenza? When we talk about influenza we talk about global averages, no?

You are confusing two different questions. How deadly is COVID-19 is if someone is infected and how many people will be infected.
If there were a good serological survey for the entire United States I believe it would show that COVID-19 is just as deadly everywhere. That's what the survey in Spain shows. How many people will get infected is an entirely different question.
There is much more regional variation in COVID-19 deaths than flu deaths because of mitigation efforts reducing the spread (and I suspect some things we haven't figured out yet).

 

[GFernandes]

I

Obviously you didn't look at the Car Bergstrom thread that Alan referenced.

This graph likely uses "confirmed deaths" only, since "probable deaths" are usually not available.
As you can see, most fall in the 0.9% to 1.0% IFR range, and would be higher if other deaths were included.

View attachment 543861

Isn't that from Imperial College? How amazing is that

 

[JRP3]

You are ignoring the fact that the number of deaths by exogenous factors are absolutely nothing compared to untreated diseases. And some the exogenous factors are increasing ( suicides ) not all of them are decreasing.

By significant numbers? Are cancer deaths suddenly shooting up?

Stress, alcoholism, suicides, domestic abuse, untreated illness, traumatized impoverished children. All these clearly far far outpace the barely any people we've save with the lockdown, they would outpace those saved even if that figure were magically 200k+. There's no way to prove it or even any stats to point to yet, so whats the point in getting these people all fired up?

Have any of those deaths suddenly shot up over the 2 months of "lockdown", which was not really a lockdown?

 

[dhanson865]

so I'm looking on amazon and you can't buy 3m masks of any kind I had seen in prior years. Nothing but cheap cloth masks and overpriced vanity masks.

Seriously I've tried a few brands and styles and the ones I made at home from Scott shop towels fit better and filter better. They just look like crap.

My wife is supposed to go back to work in a couple of weeks and it's looking like she is going to be wearing a homemade mask when she does.

In 2015 I bought "SAS Safety 8711 N95 Valved Particulate Respirator with Cushion Seal, 10-Pack" but you can't even find something that even looks like that for sale now, and yet we are "opening up".

 

[GFernandes]

By significant numbers? Are cancer deaths suddenly shooting up?



Have any of those deaths suddenly shot up over the 2 months of "lockdown", which was not really a lockdown?

This is just ignorant. You think cancer deaths are shooting up or cardiovascular are shooting up? Or that we are worsening the probabilities for these ( representing 70% of all deaths yearly ) for the ~2 months of lack of care/detection?

 

[Dr. J]

Looks like we may have a problem with the CDC.

There's clearly a problem with the CDC:
‘How Could the CDC Make That Mistake?’

It raises a lot of questions about who at the CDC has their thumb on the scale so to speak. This would need to be somebody in an administrative or executive position who believes that it is expedient to indulge President Trump's proclivities. Any ideas on that?

I think it's this guy: The new CDC director was once accused of research misconduct

He has a history.

One of the researchers who first reported Redfield’s misuse of data has been speaking out publicly after news of Redfield’s appointment emerged, Kaiser Health News reported. “Either he was egregiously sloppy with data or it was fabricated,” said former Air Force Lt. Col. Craig Hendrix, who is now based at Johns Hopkins University School of Medicine. “It was somewhere on that spectrum, both of which were serious and raised questions about his trustworthiness.”

Earlier, in the 1980s, Redfield called for mandatory HIV testing among military recruits and reporting of the results to health authorities, according to the science watchdog group CSPI.

He also called for segregating HIV-positive personnel in the military — policies that “most medical authorities at the time, including the CDC and the Surgeon General,” opposed, according to Sen. Patty Murray, the ranking Democrat on the health committee, who wrote a letter to President Donald Trump about the appointment. “This pattern of ethically and morally questionable behavior leads me to seriously question whether Dr. Redfield is qualified to be the federal government’s chief advocate and spokesperson for public health,” she wrote.