Mega-Thread: Coronavirus Pandemic

[PhaelixWW]

Yeah i know. But what i was trying to mean is that IFR could be an order of magnitude smaller than CFR. That's why we need antibody testing.
We need data. We dont have said data.

I'm less worried about it, to be honest.

First, we're never going to do enough widespread antibody testing to be able to tell for certain anyway.

Second, the dead will be dead, regardless of data on who didn't die; locking down an IFR won't change that. I'll be satisfied if estimates are done after the fact, but I think the time and effort is better spent right now on helping those who are hospitalized. That's the here-and-now.

We can still tell relatively where we are on the curve by direct observation of viral testing.

I mean, sure, if somehow magically we can roll out billions of fast-result antibody testing in short order, then it'll be useful so that people can be more reasonably free to return to work while hopefully being immune, then that'd be great.

But I don't see that happening.

[Endus]

Current number of deaths on a linear scale is a bad metric for seeing how well a country handles it, as the number of current deaths also depends on how quickly it got started (depending on travel patterns and luck), whether the climate influences the spread, how densely people live, the demographics of the country, and whether the deaths are actually recorded, and a linear scale is bad for an exponential growth.

The graph makes it seems as if Brazil seem to have stabilized in number of deaths - but with 60 daily deaths one would expect a variance of +/-8 indicating that it may as well be luck.

I generally prefer Burn-Murdoch's graphs, which base every nation off a similar starting point in terms of number of deaths, to eliminate any bias for delay in the infection triggering, for that reason.

Here's FT's link to that; https://www.ft.com/coronavirus-latest

The main graph doesn't show Brazil, but it's in the later ones, and it's below the USA's curve.

And that's precisely why you should look at logarithmic graphs instead of linear graphs.

A later start to an exponential curve is hugely meaningful.

A logarithmic scale doesn't change anything with regards to a later start. It just exaggerates the lower end and underplays the higher end. Do you not understand what a logarithm is?

Your graph visually values the first 10 deaths due to COVID-19 equally with the last 9,000 deaths. They have the same space on the vertical axis. Logarithmic analyses do have value, but using them the way you're trying to seems just like a pure attempt at disinformation.

[Puupi]

New York is starting to bury its dead in parks.

https://twitter.com/MarkLevineNYC/st...56159896748032

[Acidbaron]

The ER doctors and staff will be, the specialized types might not be. They get paid extra for their specialization, so that would be a cost measure to have them do other stuff, if they're even qualified to do it, so that fits what you're saying, but it's not the majority of the discussion.


I mean, I quoted the part that was important, the doctor's opinion is that they don't need to do it NOW since they'll be needed at some point? Doesn't really matter.



I did read the article, and did not accuse you of not reading. Your conclusion is that money grubbers are endangering lives for profit, but the doctors complaints aren't about the risk...


Doctors are bitching that they're getting their total pay cut because they're working less hours when there's less work. You're siding with them for some reason?



How do you view this? The company has reduced hours, but the hours are still over the actual demand anyway because they want to be ready, but that's bad to you?



I didn't say that, you didn't read that, you're just interjecting your emotions into a pretty straightforward discussion. If a hospital worker is going to work 90 hours a week a month from now, is it unfair to cut them from 40 to 20 now when there's no work to be done? Is that a benefit? Or should you pay them the 40 hours now, and then pay them extra when it's busier? I'm curious how you think it SHOULD work?




Using an anesthesiologist as an example, they make triple what an ER doctor makes. While they are trained as a doctor, they cannot perform all the specific tasks that are usual for an ER doctor, any more than an ER doctor can perform all the tasks they would normally do in their specialized role. If there is no use for an anesthesiologist, so you're retasking them to assist ER doctors that WILL be in high demand, do you pay the ER doctors triple, or lower the Anesthesiologist to 1/3 rate or lower as he cannot perform the full task range?



You can keep saying it, but it does not refute that you're injecting emotion into the topic that is not needed.

The ER doctors WILL be need, but are not currently. In fact, right now in most places they are doing LESS work than normally. Do you reassign the assets now, or just ignore the situation to avoid the perception of being a meanie that loves money?

I am not injecting emotion into this but i am very cautious of the intentions of groups invested in the healthcare industry. As i stated before i will gladly be proven wrong.

As for the pay difference on that i can give you a good answer, i do stand by that people can be trained right now for the future situation where everyone will have to aid regardless of their function. This is what is happening in other Nations and that appears to be working. So just leaving them at home does for me seem more motivated out of economic interests rather than anything else. That is the link i make when i believe it will be costing lives, because deploying them and training them when needed won't work, because again unless i missed this there is no word of there being something like that in place.

Even more so people are apparently on their own initiative helping out.


Perhaps this comes from a difference of perspective as being European and being in the middle of it while America is more in the beginning phase or "the calm before the storm".

[Chairman Sheng-Ji Yang]

The 'treatments' are ridiculed because there's no evidence that they work, they have dangerous side effects with a narrow therapeutic window, and Trump is falsely selling the hype like a snake oil salesman.

More here: https://www.sciencedirect.com/scienc...858?via%3Dihub

Hydroxychloroquine, together with Lopinavir (antiretroviral HIV drug) and Remdesivir (antiviral Ebola drug) are currently being used in my country, being part of the treatment given to ICU patients.

[PhaelixWW]

A logarithmic scale doesn't change anything with regards to a later start. It just exaggerates the lower end and underplays the higher end. Do you not understand what a logarithm is?

Your graph visually values the first 10 deaths due to COVID-19 equally with the last 9,000 deaths. They have the same space on the vertical axis. Logarithmic analyses do have value, but using them the way you're trying to seems just like a pure attempt at disinformation.

Logarithmic graphs are the way you account for comparing exponential growth curves, especially with different start times.

Do you not understand what exponential means?

Perhaps I can let one of your fellow Canadians explain the situation for you, since you don't seem to understand.

Most of the graphs we see showing the total number of cases or deaths are linear graphs, but there are also logarithmic graphs. What’s the difference between the two and what do they tell us about the pandemic?

Linear graphs (below, left) look scary because they show the disease is growing exponentially. It’s like compound interest on an investment. Your investment grows more and more because you’re earning interest on the interest. Or, if something doubles each week, then after two weeks it is four times as large and after three weeks it is eight times as large and so on. So, linear graphs are accurate because that’s what happens in an epidemic.

But the same numbers plotted on a logarithmic graph can be informative because they can show how the rate of growth is changing and and whether or not it is exponential. And that's the best way to measure whether our efforts – for example, through social distancing – are helping to slow the spread.

[Acidbaron]

I think my main problem with the tracking of the outbreak is this idea that we're 2 weeks into anything. Started in January, we're now in April. Social distancing has been since March, with an extension through the end of April and probably after that. Likewise, deaths due to COVID are posted, but comparisons are being made in a vacuum. Normal death rate vs current death rate vs COVID death rates would be interesting to see. Of course, at this point politics are wrapped up in anything that is said, so sides are being drawn and information is being filtered through bias all around.

If we are speaking of the US, the response and public obedience on the call for physical distancing has been very fragmented, where in some places it is taken serious and in other cases people went ahead with it, knowing how much travel there is going on it is hard to say how much of it actually has an impact.

If you want to look at death rates you also need to include people who are being killed indirectly by COVID19 due to people unwilling to go to hospitals, i know it's a thing here people are delaying going to hospitals with other ailments.

Well the whole press briefings from the government in the US is one big political re election campaign, still too pre-occupied to attack the press who is "mean" , so it is no surprise that political bias plays a big role.

[PhaelixWW]

Hydroxychloroquine, together with Lopinavir (antiretroviral HIV drug) and Remdesivir (antiviral Ebola drug) are currently being used in my country, being part of the treatment given to ICU patients.

As far as I can tell, they're not being freely administered in Switzerland, merely tested, like most countries at this point.

[Puupi]

https://twitter.com/chapoisat/status...10685843554304

Seriously wtf. What a human garbage. Rot in hell.

[Endus]

Logarithmic graphs are the way you account for comparing exponential growth curves, especially with different start times.

Do you not understand what exponential means?

Perhaps I can let one of your fellow Canadians explain the situation for you, since you don't seem to understand.

That's literally not what your own source is saying.

And the growth is exponential. They're explaining that it can be useful to compare the rates logarithmically to evaluate how the rate of growth is changing, which is valid, but;

1> If your goal is to downplay the exponential growth, then you're being dishonest, and
2> This has fuck-all to do with varying start points, which is what you claimed you wanted to use a logarithmic approach to address. Logarithmic scales can't address that, leading to me questioning whether you know what logarithms even are.

Your own source does not support your misuse of logarithmic scales, and does not address the graphs I was using nor the way in which I was comparing them.

Edit: To make it more clear, you took issue with differing start dates, which is an issue on the X-axis of a graph. To address that, you made a case about logarithmic scales on the Y-axis, which does nothing to affect the X-axis. I freely acknowledged there's value to logarithmic scales, my point was that it does fuck-all to address what you were raising as a concern.

[Draco-Onis]

Below italy and spain. On par (or a little bit below) UK. Above Brazil?

Below the UK, The US is not on lock down 9 states are still open majority of red states have a lock down in name only with golf courses, churches, gun stores and some retail stores being exempt.

[Magical Mudcrab]

https://twitter.com/chapoisat/status...10685843554304

Seriously wtf. What a human garbage. Rot in hell.

Wow, what a lovely guy.

[PhaelixWW]

New CDC report.

CDC: Coronavirus Disease 2019 in Children — United States, February 12–April 2, 2020

Notably, up to 20% hospitalization rate for people under the age of 18:

Among 149,082 (99.6%) reported cases for which age was known, 2,572 (1.7%) were among children aged <18 years. Data were available for a small proportion of patients on many important variables, including symptoms (9.4%), underlying conditions (13%), and hospitalization status (33%). Among those with available information, 73% of pediatric patients had symptoms of fever, cough, or shortness of breath compared with 93% of adults aged 18–64 years during the same period; 5.7% of all pediatric patients, or 20% of those for whom hospitalization status was known, were hospitalized, lower than the percentages hospitalized among all adults aged 18–64 years (10%) or those with known hospitalization status (33%).

[Thepersona]

As far as I can tell, they're not being freely administered in Switzerland, merely tested, like most countries at this point.

When do you think we'll have some data on all of the treatments currently being trialed?
By the end of this month?

- - - Updated - - -

I'm less worried about it, to be honest.

First, we're never going to do enough widespread antibody testing to be able to tell for certain anyway.

Second, the dead will be dead, regardless of data on who didn't die; locking down an IFR won't change that. I'll be satisfied if estimates are done after the fact, but I think the time and effort is better spent right now on helping those who are hospitalized. That's the here-and-now.

We can still tell relatively where we are on the curve by direct observation of viral testing.

I mean, sure, if somehow magically we can roll out billions of fast-result antibody testing in short order, then it'll be useful so that people can be more reasonably free to return to work while hopefully being immune, then that'd be great.

But I don't see that happening.

I mean, it would help shape more effective policies (like a COVID passport being touted in Spain), get a massive pool of plasma donors for antibody treatment and to know the true R0 and IFR of this damned beast.

[Bryan Adam]

yeah i don't think civilization of the world could survive that and I'm not a doom sayer.. once the hospitals can keep up with new infections, some point after peek, shelter in place will be lifted and things opening up. The US is expected to peek as a country in 15 days, if that holds true some places should be able to resume "normal" in may.

According to highly-regarded news, it's likely we might see business returning in late summer. China is experiencing a second economy shockwave as they are still hurting from the global pandemic. The number of cases in United States seems to be slowing down a bit as it the daily cases grows roughly by 2,000 to 3,000 a day. I am hoping for a quick recovery once this settles down.

[Acidbaron]

According to highly-regarded news, it's likely we might see business returning in late summer. China is experiencing a second economy shockwave as they are still hurting from the global pandemic. The number of cases in United States seems to be slowing down a bit as it the daily cases grows roughly by 2,000 to 3,000 a day. I am hoping for a quick recovery once this settles down.

I would be cautious with putting too much stock into this as there is such a thing as a week-end effect that when looking at European nations over the week-end and up to Monday there are fewer cases, while on Tuesday it goes up again.

[Mayhem]

I would be cautious with putting too much stock into this as there is such a thing as a week-end effect that when looking at European nations over the week-end and up to Monday there are fewer cases, while on Tuesday it goes up again.

They are also not testing enough, on a per capita basis they are behind Spain, France, Italy, Germany and and and...

[Acidbaron]

They are also not testing enough, on a per capita basis they are behind Spain, France, Italy, Germany and and and...

True, the real indicator of the disease progression will remain their hospitalizations what don't appear to be going down.

[PhaelixWW]

That's literally not what your own source is saying.

Oh, this ought to be good.

And the growth is exponential. They're explaining that it can be useful to compare the rates logarithmically to evaluate how the rate of growth is changing, which is valid

Ah, so you agree. Wonderful!

, but;

Whoops, oh boy.

1> If your goal is to downplay the exponential growth, then you're being dishonest

How on earth is it "downplaying" exponential growth? Logarithmic graphs are, by design, intended to make exponential growth easier to visualize properly. I mean, that's the whole point, and it's precisely why you're making the wrong inference from your choice of graphs.

When Should I Use Logarithmic Scales in My Charts and Graphs?

There are two main reasons to use logarithmic scales in charts and graphs. The first is to respond to skewness towards large values; i.e., cases in which one or a few points are much larger than the bulk of the data. The second is to show percent change or multiplicative factors.

The next example just describes rates of change. Suppose we had one widget in 1999 and doubled the number each year. The linear scale shows the absolute number of widgets over time while the logarithmic scale shows the rate of change of the number of widgets over time. The bottom chart of Figure 4 makes it much clearer that the rate of change or growth rate is constant.

Dr. Nicolas Bissantz in his blog, Me, Myself, and BI, would call the linear chart a panic chart. He says that “line charts are speed charts.” That is, they show the rate of change or slope of the number of widgets. A chart with a linear scale similar to the top chart of Figure 4 showing a quantity such as our national debt causes panic even if the rate of change is constant.

Logarithmic scales are extremely useful but are not understood by all. As in all presentations, designers must know their audiences.

People inherently have a hard time comparing two exponential graphs, even when the starting points are perfectly lined up. In the case of starting points that aren't lined up? Not a chance.

2> This has fuck-all to do with varying start points, which is what you claimed you wanted to use a logarithmic approach to address. Logarithmic scales can't address that, leading to me questioning whether you know what logarithms even are.

Lulz. See above. This is grade school math. I don't know why you're having such a hard time with this.

But hey, feel free to find a source that backs up your claim that logarithmic graphs are the wrong way to visualize exponential viral data. Then perhaps explain why pretty much every single data visualization includes a log option.

Edit: To make it more clear, you took issue with differing start dates, which is an issue on the X-axis of a graph. To address that, you made a case about logarithmic scales on the Y-axis, which does nothing to affect the X-axis. I freely acknowledged there's value to logarithmic scales, my point was that it does fuck-all to address what you were raising as a concern.

Ah, a partial retraction, goody.

You're still ignoring the fact that logarithmic graphs exist to make the data easier to visualize. And easier visualization is key when comparing things that aren't lined up perfectly, as in the case of different start dates.

Honestly, you're just continuing to argue about this because you don't like that your initial point wasn't sound.

- - - Updated - - -

When do you think we'll have some data on all of the treatments currently being trialed?
By the end of this month?

The timeline for COVID-19 is like 2-4 weeks. It's hard to imagine that we're going to get anything terribly meaningful before the end of April. We'll probably see bits and pieces come in, as the various studies being conducted had different start dates and different scopes. The data will only continue to get refined the longer it goes, of course.

[Captain N]

So the US is now over 10,000 deaths....

....idiots are still outside and still using the Obama Flu was worse garbage.