Coronavirus in the United States - news and thoughts

Setting aside the many issues with those tests, as outlined in the article from NYT, that had a death rate at the top end of your range (0.5%). So why does your range move to 1/5 of that total?

Beyond that, if you use the current numbers with a 0.5% death rate, you would need an infection rate of New York City of 44.24%. Seems unlikely that almost half of all residents in New York have now been infected.

 

Because there are other studies, namely some of which I mentioned previously, that calculate the mortality rate as between 0.1-05%.

The number of infected does not seem unlikely at all when you consider that 80% of people are asymptomatic and 10-15% present with mild symptoms.

 

The one here opens Monday and the restrictions seem to be in line with what you post. Relying on social distancing, sanitizing stations, and sanitizing high touch items (chips, slot machine buttons etc), limiting players per table. Restaurants and bars will be opened with rearranged tables.

• Positions will be limited at each table to allow for appropriate social distancing.
  • Three players per blackjack / pai gow / carnival game table
    
  • Four players per roulette / poker table
  • Six players per craps table
• Guests will not be able to congregate behind players at a gaming table.
• Where possible, procedures have been adjusted to allow only dealers to touch cards.
• High touch gaming items (dice, chips, etc.) will be routinely disinfected, and cards will be refreshed more frequently.
• Dealers will provide hand sanitizer to every arriving player.

I dunno. Being in a large room with several hundred people doesn't sound inviting to me right now - will be interesting to see how busy they get.

 

Let me address these each in turn.

One reason why ventilators are problematic in this case is that they can result in pneumonia infections. That is a terrible intervention when you already have a lower respiratory issue.

Another part of this discussion is that physicians put people on ventilators when they determine it is necessary. Just putting people on ventilators caused an uptick in death rates. If you mean to suggest that only those who need to be on ventilators should be on them, then I can point you to the 80% of patients who dies on them. This remarkably high percent indicates that the ventilators are not helping them.

The bottom line is that you seem to be trying to delineate a fine point that doesn't exist. Instead, consider that hospitals began pulling away from them. In doing so, they saw death rates decrease back to that of the average cohort of those who needed hospitalization. Honestly, from a statistical perspective, or medical one, this is very telling: Pulling away from the ventilators decreased the death rate back to the normal (baseline). This is really all we need to determine that ventilators were not really helping. Let us be careful: Does this mean it will not help anybody? No, it does not. But it does suggest it should not be considered as SOP.

So, here, I would ask: How do you propose we prove the hypothesis?

Obviously, this is based on modeling. There is no other way. We would have to actually let it happen to prove it. But there are strong indicators:

- Suicide calls are up over 1000%
- Domestic abuse calls are up
- I suspect that we will see an increase in robbery (I concede I do not have the numbers for this yet)

I mean, this is hardly a shot in the dark.

You are correct that I linked you to the incorrect one. Here is the correct one:

https://www.thelancet.com/journals/langlo/article/PIIS2214-109X(19)30409-7/fulltext#

To be clear, I am not necessarily the biggest fan of models. Consider Dr. Neil Ferguson. His track record is abysmal. His code is undocumented. Still, he is the source of the 2.2M number dead in the US based on his modeling. He is "patient zero" for this number. He was also an outspoken advocate for Draconian lockdowns. Nevertheless, he was fired from his post (they kindly allowed him to "resign") after he was caught violating his own recommended mandates: He let his married lover enter his home during lockdown for a "social visit." A woman who lives at home with her husband and their two children.

Still, I encourage you to search his track record.

 

@ursidman Did you even bother to read the remainder of the conversation?

 

Which specific study has found a death rate of 0.1% in completed cases?

 

What the heck are you talking about? Whatever it is - yes, I've read the whole 554 pages of this thread

 

I think there are something like 7-10 of these studies now. It might have the LA one where they estimate the number of infected as 87X higher than reported. If you carry that out with the reported number of deaths, the mortality rate works out to 0.1%.

If you are asking me whether I think the mortality rate is 0.1%, then this is a separate issue. To be fair, we would have to discuss how infections and deaths stratify across different cohorts. FWIW, that's exactly how we discuss any disease, such as the flu.

 

You don't know the posts with which you disagree? lol

Your bacon on the number of deaths.

 

No, it decidedly does not without an idea of what those same patients mortality rate would have been without the ventilator. If 90% of them would have died without the ventilator, then it certainly helped them.

I'd be interested in seeing that data, however, they still likely strategically withdrew care. That still doesn't come close to saying that ventilators haven't saved lives, which was my initial claim.

And a variety of other forms of death are down (e.g., car accidents and murders). Where is this 1000% increase number coming from? I have seen estimates ranging from 40-800%, but haven't seen a single estimate that high.

Second, in terms of data that could provide this, you could look at excess non-Covid deaths.

Interesting line from that:

So if you take a country with extreme poverty to start and then don't provide much in the way of social protection, those results hold, but not in places with health and social protection. Maybe we should have some more of that then...

What does any of this have to do with anything other than an ad hominem to explain why you are willing to use projection models in certain circumstances but not in others?

 

They did not discover that, in fact. The LA estimate was 28x-55x, not 87x. And that is still setting aside the issues with the study, which are extensive and likely bias the results upward (the same authors had to "correct" their mess of a study in Santa Clara, but I have yet to see if they did the same to this study- which had severe, though less severe, issues than the Santa Clara study). Beyond that, trying to utilize that to find a death rate is obviously problematic because an uncompleted case would show up in the denominator but not the numerator.

Early antibody testing suggests COVID-19 infections in L.A. County greatly exceed documented cases

 

Absolutely not. This is entirely incorrect.

First, an 80% death rate with ventilators compared with 40-50% is decidedly against the use of them as SOP. The same cohort was in both groups. That is really a definitive answer.

Second, your argument was not that it may have helped a single person, rather, that we ought to use it as SOP. The data are unequivocal in this regard. If you have other data, then present them. Otherwise, what is your point? That ventilators may have helped someone? Nobody says otherwise. You framed your argument as "appropriate care." This type of care is SOP. That ventilators proved deleterious as SOP is clear.

So what increase in suicide rates is acceptable to you?

What is your point? Nobody was talking about social programs here. This was one of the papers on which they based the model. The idea to be taken from this is hardly news: Higher unemployment leads to higher suicide rates.

If you're going to make an argument for another recompense bill, then that is a separate issue altogether.

I thought I made it quite clear: I am not the biggest fan of modeling. More to the point, I am not a fan of someone who has an abysmal track record with modeling. Even more, I am particularly not a fan of someone who refuses to share his/her code for said models. This should hardly be a hot take.

Moreover, his track record would be comical if it were not a travesty.

Finally, it was hardly about an ad hominem. I provided you with context. One could argue that if he doesn't heed his own warnings, why should anybody else? Ultimately, this is not to say measures should not be taken. The point was once you sum it all together (his track record, his undocumented code, his firing, etc), he has proven untrustworthy. So too has his code.

I could go on with the inherent difficulties with virtual modeling for drug development, but that's a bit tangential.

All of this is to say what I said previously: Take the models as you will. I do not take any one model as more or less worth than the other necessarily. Check your sources.

Finally, if you disagree with something, then why don't you provide the data for your disagreement.

 

What part of I think did not you not catch? Do you think I just memorized these? I posted up like four of these weeks ago. Maybe it was the NYC study. I told you there are something like 7-10 now. And honestly, how freaking different is 28-55X? Honestly, this is comical. Myriad of studies are demonstrating the same trend: The infection rates are much higher than reported.

Trying to make an argument out of an "uncompleted" case is a making a much larger point than is there in reality. When the number of estimated cases is really orders of magnitude higher, then we need not nitpick. Also, there's a reason why people are bringing down the number of deaths across the country. There's a reason why CDC is advising hospitals to do so.

 

Look again:

The comparison is non-covid vs. covid both needing a ventilator according to a Doctor. The question we need to answer is Covid with a Doctor saying they need a ventilator getting one and not getting one. Totally different question.

Not based upon the studies you posted, none of which answered that question.

It isn't about acceptability. But if you are going to make the claim, be able to back it up. We don't know the increase, but you stated the number. What is the basis for that number?

That is not at all what that study said. In fact, the primary lift was from other causes. However, the important part is that the results are not what you are using as a basis for this argument in conditions like developed countries with social welfare programs that are well-established. You can't just bring those results into a totally different context,especially when the results themselves don't work in any context similar to the one we are discussing.

Not really separate if the argument is that we could wipe out that excess mortality, as shown in that paper.

What does the fact that he violated his own rules or that he was let go from a government post for doing so have to do with his capability of modeling? If you have an issue with the model, focus on that and I am happy to discuss statistical issues in detail. The rest is clearly ad hominem.

 

Not if you drive down the numbers infected before reopening and commit to the development of tests. Because with fewer cases, there are fewer people to test. But, when leaders are incompetent and do not plan, everything seems impossible.

 

Well that helps some. I disagreed because reducing contact among people reduces the risk of infection. I was remembering the super spreaders responsible for scores of infections and deaths from attending parties, funerals and other social events such as what occurred in Albany Ga and other places. If Albany had been in quarantine then that spike in cases and deaths wouldn’t have happened to the same degree.

 

It is pretty freaking different. An overestimation by 58-300% is pretty big, especially when discussing something like death rates. If your range was off by 200%, that would be a pretty big deal.

Most people agree that the undercounting is an issue. But the death rate on the confirmed cases is almost 6%. You have a long way to go to get to 0.1%.

 

To use your words: prove it.

Also, you need to illustrate how it stratifies across age groups.

 

No, it is not. They discussed backing patients away from ventilators and observing better outcomes.

From another source. Honestly, just search it. It's not hard to find.

Honestly, I have no clue what the hell you are talking about. This was a model of recession. This is not a recession. The only idea was that they took this to model what could happen with prolonged shutdowns. I have no idea what the hell you are babbling on about here.

I just told you:

- HIV
- Foot and mouth disease
- Flu
- Covid

ALL WAY OFF.

We can't really discuss his code because he won't document it.

2.2M is a far cry from where we are.

You need to stop telling me what I'm saying. I gave you the context. I then told you why I gave the context. The rest is up to you (Hint: If he didn't think endangering another human knowingly and three others unsuspectingly, then why should you take him seriously?). Decide for yourself if you think his model was worthy. Go search his previous models. Come back and discuss the worthiness of his models. I have already had this conversation with my bewildered colleagues, whom like me, could not understand why he was taken so seriously. For example, I would love for you tell me how his advocation for slaughtering an entire livestock held up.

 

All completed with antibody tests with as many false positives as real positives