Coronavirus

[sparcs]

Gov. Newsome just spoke and said if their projections hold true they expected to push off peak until May. If Peak is in May then I’m not seeing the factory up and running by then.

California, Texas, and Georgia have fairly low per capita testing compared to other states who are most affected. I don't know if the data we get out of any of those states is something we can rely on to make predictions.

Suprisingly, Louisiana has almost as much per capita testing as New York.

 

[dfwatt]

So, how does Covid19 compare to other causes of death, per day ?

Covid vs. US Daily Average Cause of Death

That's a neat graphic. Thanks for posting. It's safe to say that the actual number of deaths associated with covid-19 is probably significantly higher than what the official stats are because of the tendency to underestimate the role that this virus has in triggering heart attacks. Up regulation of several interferons but most particularly interferon-gamma are directly implicated in the destabilization of chronic plaque structures in arterial vessels. It may take a lot of epidemiological digging to truly separate out what deaths are due to covid-19 and what deaths are due to mechanisms that are truly independent of viral infection. Particularly suggestive is the huge uptick in calls to 911 in New York City for heart attacks. Way more than during a typical week. While this has popularized the meme that a lot of people are dying with covid-19 rather than from covid-19, I think that mechanistic distinction is specious. These are folks that would have survived and lived at least for an uncertain but probably non-trivial interval longer, absent infection with covid-19. And you don't even want to get started on how much fudging and confusion there is in our statistics about so-called causes of death not just in this country but in Europe as well. Many deaths in the elderly are truly multifactorial and where a conspiracy of factors comes together to create a fatal situation.

 

[EVNow]

Makes one wonder if there may be a role for hyperbaric oxygen.

Yes, there is a clinical study going on ...

Hyperbaric Oxygen for COVID-19 Patients - Clinical trial in progress : COVID19

That seems unnecessarily speculative. Do you have some evidence that it would be beneficial? Hyperbaric oxygen treatment does not pressurize the lungs above the ambient pressure which is why ventilators are used for infections of the lungs that require positive pressure.

Apparently there is some reason to think it would be beneficial. See above.​

 

[bkp_duke]

Let's look at the specific examples you offered:
South Korea- South Korea has been used Chloroquine extensively in treating COVID-19. If South Korea numbers are especially low, it suggests the use of this anti-viral has contributed to those low numbers

Luxembourg- Sheesh, talk about cherry picking. Malaysia had a population of 31 Million in 2017. Luxembourg's population was about 600K in 2018.

Finland- More cherry picking. Finland had a population of about 5.5 million in 2018.

The reason I used Malaysia is because it has a reputation of using Chloroquine throughout the epidemic, moreso than any other country. In many ways it is the poster child for using the drug. I didn't choose Malaysia for its low numbers. Rather, I was curious to see its numbers because of its reputation for being such an early adapter of chloroquine use for treating the virus. For a counntry with over 30 million inhabitants, this is a remarkably low number of deaths.

You cannot take multifactoral outcomes, like death, which are determined by many risk factors and many different interventions, and attribute them to a single variable (like HCQ).

South Korea confounding factors:
1) much younger population
2) much higher usage of masks
3) much higher contact tracing

I could go on, but those were literally off the top of my head.

This is the problem with the lay person trying to interpret (really bad) scientific data. False attribution of an end point to a single input.

If it were easy, it wouldn't take gigantic studies to remove the noise from other variables and determine the real cause - effect relationship.

 

[EVNow]

And you don't even want to get started on how much fudging and confusion there is in our statistics about so-called causes of death not just in this country but in Europe as well. Many deaths in the elderly are truly multifactorial and where a conspiracy of factors comes together to create a fatal situation.

Yes - someone was joking on Reddit/Twitter that everything in Germany is coded as "heart stopped working".

 

[Frizer]

That seems unnecessarily speculative. Do you have some evidence that it would be beneficial? Hyperbaric oxygen treatment does not pressurize the lungs above the ambient pressure which is why ventilators are used for infections of the lungs that require positive pressure.

If the lung compliance is softer than typical in ARDS makes me wonder. There’s tons to learn still. And hyperbaric oxygen is available and relatively inexpensive.

 

[EVNow]

Gov. Newsome just spoke and said if their projections hold true they expected to push off peak until May. If Peak is in May then I’m not seeing the factory up and running by then.

I wonder whether Newsome has some other teams working on projections or has projections from yesterday. Before the update last night, IHME had CA peak in May beginning. Their update brings it to mid April.

 

[bkp_duke]

I suspect most of us have seen the video now. I love that Tesla is making progress on repurposing vehicle parts into a ventilator. It is also quite generous for Tesla to give these ventilators away. But as a shareholder, I have to ask what will this cost us? When Tesla is able to ramp up production, the materials and labor alone will cost quite a bit. One ventilator might cost as much as a Model 3 to make. So is there any way Tesla can be reimbursed for this?

Anyway, enjoy the video!

The goodwill and "like" factor for Tesla far outweighs the monetary costs for this usage of some parts.

As long as Tesla doesn't screw it up, however. That remains to be seen that they don't miss some important detail and it becomes a PR disaster like the cave stuff.

 

[Prunesquallor]

Gov. Newsome just spoke and said if their projections hold true they expected to push off peak until May. If Peak is in May then I’m not seeing the factory up and running by then.

I’d like to know what projections he’s using. As stated up thread, the IHME predictions show peak deaths in two weeks.

 

[dfwatt]

Yes - someone was joking on Reddit/Twitter that everything in Germany is coded as "heart stopped working".

Just to give you an example of how dizzyingly complicated this business of assigning a cause of death is, a lot of the patients that I would follow in our Alzheimer's service would be worked up and treated for various reversible conditions (often times depression or delirium), and then discharged with after care in assisted living or more likely a nursing homes and at some point would have an aspiration event. This is because Alzheimer's disease in its later stages tends to suppress the gag reflex so they would develop aspiration pneumonia. This was often times difficult to treat, and sometimes would lead to fatality.

Frequently this would be listed as the cause of death (pneumonia), but is that really accurate or is it in fact somewhat and maybe even profoundly misleading? Maybe Alzheimer's disease should be listed as a cause of death? Or the fact that they are so immunocompromised that they can't fight off even minimal infections? In that case perhaps immunosenescence should be listed as a cause of death? Given that immunosenescence is in fact a risk factor for Alzheimer's disease and may even contribute to its progression, you begin to see how much circularity and positive feedback there is in this whole process of mortality. It's not single factors it's conspiracies between factors and positive feedback between multiple undesirable age-related processes or as we call them aging phenotypes. But we like our sound bite over-simplification of a single cause just like we like to think that chronic diseases should have a single cause. Mostly they don't. All the classic diseases of Aging look more and more like they're hugely multifactorial and recursive.

 

[Prunesquallor]

I wonder whether Newsome has some other teams working on projections or has projections from yesterday. Before the update last night, IHME had CA peak in May beginning. Their update brings it to mid April.

Yep, the update compressed all peaks because of the Spain and Italy data. Newsome needs to hit F5.

 

[Papafox]

You cannot take multifactoral outcomes, like death, which are determined by many risk factors and many different interventions, and attribute them to a single variable (like HCQ).

South Korea confounding factors:
1) much younger population
2) much higher usage of masks
3) much higher contract tracing

I could go on, but those were literally off the top of my head.

This is the problem with the lay person trying to interpret (really bad) scientific data. False attribution of an end point to a single input.

If it were easy, it wouldn't take gigantic studies to remove the noise from other variables and determine the real cause - effect relationship.

Of course everything you say is relevant, but the conclusion you keep coming back to is that someone who is not a doctor cannot be taken seriously in this discussion. I disagree. If you keep looking at studies that are less than perfect but continue to show efficacy, and you compare countries that used chloroquine to those that mostly shunned it and the ones that used chloroquine or HCQ show consistently better results, then these are just more data points suggesting that the drugs can produce benefits.

Like I said earlier, I respect your opinions as a doctor. That doesn't mean that there's no room for discussion, however.

 

[bkp_duke]

Of course everything you say is relevant, but the conclusion you keep coming back to is that someone who is not a doctor cannot be taken seriously in this discussion. I disagree. If you keep looking at studies that are less than perfect but continue to show efficacy, and you compare countries that used chloroquine to those that mostly shunned it and the ones that used chloroquine or HCQ show consistently better results, then these are just more data points suggesting that the drugs can produce benefits.

Like I said earlier, I respect your opinions as a doctor. That doesn't mean that there's no room for discussion, however.

"Less than perfect" is a galactic understatement for the studies that support HCQ.

The term "polished turds" comes to mind. Shine it up all you want, in the end . . . it's still a turd.

 

[SMAlset]

Fauci is an extremely accomplished physician and Navarro is a Trump flunkey who knows exactly squat about medicine.

'Nuff said

I saw an article where Navaro said when asked about his qualifications that he was a “social scientist” and could read testing results just fine. Sorry without a solid medical background and understanding I don’t have any confidence in essentially an administrator level person interpreting lab results and understanding or anticipating reactions from different protocals. It’s one thing to run small sample tests and if appropriate test wider still but to play roulette with patients lives across a wide swath of ICU patients, like those who might live with life-long debilitating heart/lung conditions, with the attitude well maybe it could work or maybe it won’t but what’s to lose, is not the way medicine is practiced. As it is the WH wants to open up FDA approvals before being carefully vetted. Great for drug companies I guess.

 

[jhm]

Datagraver has a new interactive tool for generating Covid19 charts. I've shared many of their charts but now you can set the assumptions. For example, to predict when the US will cross 100 DPM, I can select just the countries I wish to compare with and start at days since 10 DPM or 29 DPM to align it most closely to my prediction problem. It also shows the death growth rate comparisons so I note that Netherlands and Switzerland are actually the two closest comps in terms of recent death growth rates. Following Netherlands, the US still looks on track to cross 100 DPM on April 14, but following Switzerland this could be a day or two later. So play with it yourself.

Covid-19.

 

[keydiver]

I'll help you: overall 1/4 hospitalized end up on a vent and ~ 80% of those on a vent die.

A friend of mines wife is a nurse. She said that COVID patients that are classified as a "Stage 3" go on a ventilator, and 85% of those end up dying. Not good odds.

 

[dhrivnak]

The goodwill and "like" factor for Tesla far outweighs the monetary costs for this usage of some parts.

As long as Tesla doesn't screw it up, however. That remains to be seen that they don't miss some important detail and it becomes a PR disaster like the cave stuff.

I agree, except not sure the cave stuff would have even been noticed if it were not for the inappropriate tweets. The sub ended up being a very minor issue in the overall problem.

 

[Daniel in SD]

Of course everything you say is relevant, but the conclusion you keep coming back to is that someone who is not a doctor cannot be taken seriously in this discussion. I disagree. If you keep looking at studies that are less than perfect but continue to show efficacy, and you compare countries that used chloroquine to those that mostly shunned it and the ones that used chloroquine or HCQ show consistently better results, then these are just more data points suggesting that the drugs can produce benefits.

Like I said earlier, I respect your opinions as a doctor. That doesn't mean that there's no room for discussion, however.

And ignore the studies that don't show efficacy...
Doctors aren't claiming it doesn't work, that's why they're studying it!
You, however, are claiming it works and are simply looking for data points that support your view. Someone claiming the opposite could do exactly the same thing. You're going to need to do a much more sophisticated analysis than looking at plots.

 

[dfwatt]

I saw an article where Navaro said when asked about his qualifications that he was a “social scientist” and could read testing results just fine. Sorry without a solid medical background and understanding I don’t have any confidence in essentially an administrator level person interpreting lab results and understanding or anticipating reactions from different protocals. It’s one thing to run small sample tests and if appropriate test wider still but to play roulette with patients lives across a wide swath of ICU patients, like those who might live with life-long debilitating heart/lung conditions, with the attitude well maybe it could work or maybe it won’t but what’s to lose, is not the way medicine is practiced. As it is the WH wants to open up FDA approvals before being carefully vetted. Great for drug companies I guess.

I think the problem is that Navarro really isn't a scientist, or at least he does not seem to understand what are core scientific virtues and requirements. I do believe that good scientists can read outside their area but they should maintain core scientific virtues of basic skepticism ("show me the money . . .err, data!), the need to exclude the typically many and competing alternative explanations before drawing a conclusion, respect for double blind placebo-controlled trials when you are evaluating clinical treatments, and most of all, humility in the face of Nature . . . simply because it's way more complicated than we know and perhaps way more complicated than we'll ever know. Navarro displays none of those core scientific virtues. This is not surprising because anyone who believed in those things could not possibly work for the incompetent sociopath running the current Administration. So his claim in that sense is specious.

 

[ZeApelido]

Of course everything you say is relevant, but the conclusion you keep coming back to is that someone who is not a doctor cannot be taken seriously in this discussion. I disagree. If you keep looking at studies that are less than perfect but continue to show efficacy, and you compare countries that used chloroquine to those that mostly shunned it and the ones that used chloroquine or HCQ show consistently better results, then these are just more data points suggesting that the drugs can produce benefits.

Like I said earlier, I respect your opinions as a doctor. That doesn't mean that there's no room for discussion, however.

He didn't you had to be a doctor, I think he was implying you should likely have expertise in interpreting data to work with such noisy, multidimensional data. (btw something many medical doctors cannot do) otherwise you end up making the basic common mistakes seen in data analysis.

In specific to your Malaysian example, it's not that your data does not have any merit, but that this sort of analysis is usually called "cherry picking" because with complex data someone could just find another country that shows the opposite effect, simply due to noise in the data. The analysis needs to be deeper or else its too exhausting to discuss.