COVID19 Factors We Should Consider/Current Events - Page 60

# Thread: COVID19 Factors We Should Consider/Current Events

1. Member
Join Date
Jan 2014
Posts
95
Originally Posted by lazygun37
What do you think the model will predict for May 15?
Originally Posted by Mark Rippetoe
I don't know. Neither do they.
That's obviously not true -- they do know what the model predicts. Just because that particular plot stopped at April 30 doesn't mean there is something magic about that date. And I think we can both (hopefully) estimate pretty easily what that number would be, roughly...

But forget about those stupid models. Let me ask you an obvious question about the data that we have in hand right now. In the USA, right now, the death doubling time is around 3 days (Coronavirus Disease (COVID-19) – Statistics and Research - Our World in Data). To the best of my knowledge transmission rates aren't even slowing, let alone stopping. To put this into perspective, at 3 days doubling time, the numbers would be something like this over the next 3 weeks:

Today: 800 deaths
+3 days: 1600 deaths
+ 6 days: 3200 deaths
+ 9 days: 6400 deaths
+ 12 days: 12,800 deaths
+ 15 days: 25,600 deaths
+ 18 days: 51,200 deaths
+ 21 days: 102,400 deaths

That takes us just beyond Easter, so let's hope Trump is right in thinking that the virus is Catholic and will turn itself off out of respect for our traditions.

But I really want to hear your prediction: at what point in the next 3 weeks -- and why -- will this exponential increase will stop?

I wanted to extend a belated apology to DrT, because in re-reading bits of this thread, I realized I mixed him up with a different poster at one point. Specifically, in a response to him, I said the following:

Originally Posted by lazygun37
In a previous post, you wanted me to quote confidence intervals on quantities that were *clearly* dominated by systematic uncertainties, rather than random statistical uncertainties. I pointed this out at the time, and you claimed to understand the difference. Well, you clearly don't.
but the poster who'd asked about confidence intervals before that was somebody else. Sorry DrT.

2. Member
Join Date
Sep 2011
Location
Valley of the Sun
Posts
1,488
Not saying that Covid-19 is the flu or is no worse than the flu. But the media definitely lets some things pass without even reporting on them.

MAR 05, 2020
US Flu Cases Reach 32 Million, Pediatric Hospitalization Rates Hit Record High
US Flu Cases Reach 32 Million, Pediatric Hospitalization Rates Hit Record High

3. Originally Posted by lazygun37
That's obviously not true -- they do know what the model predicts. Just because that particular plot stopped at April 30 doesn't mean there is something magic about that date. And I think we can both (hopefully) estimate pretty easily what that number would be, roughly...
Right. The model predicts that by October 1 the entire mass of the known universe will have been converted to COVID19 virus particles.

But forget about those stupid models. Let me ask you an obvious question about the data that we have in hand right now. In the USA, right now, the death doubling time is around 3 days (Coronavirus Disease (COVID-19) – Statistics and Research - Our World in Data). To the best of my knowledge transmission rates aren't even slowing, let alone stopping. To put this into perspective, at 3 days doubling time, the numbers would be something like this over the next 3 weeks:

Today: 800 deaths
+3 days: 1600 deaths
+ 6 days: 3200 deaths
+ 9 days: 6400 deaths
+ 12 days: 12,800 deaths
+ 15 days: 25,600 deaths
+ 18 days: 51,200 deaths
+ 21 days: 102,400 deaths

That takes us just beyond Easter, so let's hope Trump is right in thinking that the virus is Catholic and will turn itself off out of respect for our traditions.

But I really want to hear your prediction: at what point in the next 3 weeks -- and why -- will this exponential increase will stop?
October 1, obviously.

4. Originally Posted by MWM
By way of a summary, here are, in my view, the most serious problems surrounding our governments' responses:

• There are absolutely no reliable statistics about the virus's case-fatality ratio. This is because, where extensive testing has been done, the vast majority of positive cases are either totally asymptomatic or present a very mild illness. The only sensible extrapolation is that huge numbers of people who have not been tested already have the virus either without knowing it at all, or without requiring any medical intervention whatsoever. The only closed environment which might feasibly provide a estimate for the mortality rate is the Diamond Princess cruise ship, where 1% of those who tested positive died, and that is in a totally unrepresentative sample in which the overwhelming majority of the population were elderly.

• The published death figures are also unreliable in absolute terms because no causality is being established. Note that the media are (typically) careful to say that a certain number of people have died 'after testing positive for the virus.' We have no indication of whether the virus was the leading cause of death or whether there were other contributing factors. In short, we do not know whether these people died 'with' the virus, or 'of' the virus. In Britain, under the Emergency Coronavirus Bill passed at the beginning of this week, doctors have actually been relieved of any responsibility to perform an inquest into the death of a patient who tested positive for COVID-19. This means we simply do not know how lethal the virus is even in patients considered to be the most vulnerable to it. It is even feasible that no more people have died because of the virus than otherwise would have died from other concurrent factors.

• Even if all the deaths attributed to it are indeed 'excess' deaths, we must recognise the fact that more people die each day who have tested positive for other coronaviruses than are projected to die after testing positive for COVID-19 under the current worst-case estimates. COVID-19 is not the only coronavirus, and it so far follows the pattern of other coronaviruses we are already familiar with, in that it coexists in humans and animals, causes mostly very mild respiratory disease, and severe cases are confined almost exclusively to elderly patients and those with underlying lung and heart conditions. As of the present time there is no indication that this virus is materially different from any of the other coronaviruses which are already present in humans across the world.

• There is a long history of unfulfilled pandemic scares over the last few decades. In 1997, Bird flu was predicted to kill millions - it did not. In 1999, Mad Cow Disease and its human variant, vCJD, were predicted to kill half a million - in fact fewer than 200 died from it in Britain. In 2003, SARS was reported to have ‘a 25 per cent chance of killing tens of millions’ and to be ‘worse than AIDS’ - the final death toll was less than 800. in 2006, Bird flu was again predicted to ravage the world - it did not. In 2009, we were told that swine flu could kill 65,000 - it did not.

• There is no indication that the draconian measures implemented in Italy and China have slowed the spread of the virus or saved lives. In fact, if the recent Oxford modelling, indicating that the virus has been around for much longer and is much more widespread than originally thought, is correct, then these measures had no hope of working in the first place because far too many people are already infected. And, as mentioned above, the published infection and death rates are utterly unreliable indicators of the success of any measure designed to combat the virus because they are unreliable measures of the spread of the virus in the first place. We should also point to nations like Japan and South Korea, who have not enacted such measures and appear to have performed similarly well. Northern Italy and China are also tremendous outliers due to their dramatically higher levels of air pollution which increase the likelihood that residents will be symptomatic when infected with a respiratory virus.

• Our governments' reaction to the virus is not only disproportionate but profoundly dangerous. Large numbers of people have already lost their jobs. Many businesses have already closed with serious doubts that they will ever be able to re-open. Capital markets are performing worse than during the 2008 financial crisis. The consequences will be extreme and, crucially, interdependent. It is a false dichotomy to contrast economic consequences with patients' lives as many people are doing in defense of our governments' measures. We are frequently told that one of the greatest killers is poverty, and we are likely going to see a lot more of that once our economies have crashed through the floor. Not only that, but many people are apparently satisfied that these measures, however bad, are justified in order to protect the lives of the elderly. What they fail to appreciate is that thousands of elderly people die of various natural causes every day, but what keeps many other elderly people alive in the face of these natural causes is the social contact, social events, hobbies, exercise and training of which they are now deprived, shortening the lifespan of millions. On top of that are the dire effects on medical care: 'services to patients are reduced, operations cancelled, practices empty, and hospital personnel dwindling.' This is to say nothing of the political consequences, which so far have amounted to the handing over of many important liberties to the state under the guise of 'emergency powers,' with no restrictions on how those powers are to be used and for how long the state may maintain them. In Britain these include vastly greater powers of police detention for everyone, a lowering of standards and vital safeguards for those with mental illnesses or who rely on health and social care. Governments quite naturally distill fear into power, and many of the worst tyrannies of the last century came into being and sustained themselves on the basis of 'emergency powers' such as these, hastily rushed through nominally democratic channels, maintained long after the supposed need for them has passed, and justified through fear. Our representative legislatures, which are the manifestations of our shared traditions of liberty in Britain and the United States given concrete force, are supposed to prevent this distillation from going too far, but swept up in this panic they have abrogated their responsibilities so utterly that they can never again be viewed as a serious and important obstacle to the executive.

• Given all these points, the question I think we must ask is this: If we had not discovered that this was a novel virus and had not given it a name, would anyone have even noticed its existence in the first place? Would anyone have rushed to buy all the toilet paper they could carry? Would anyone have called for businesses to shut and for entire populations to be placed under house arrest? This is not the Black Death. People are not coming out with horrible buboes and dying in agony within a few days. This is a virus we needed technology to recognise. Even the numbers of deaths and serious illnesses being officially attributed to the virus are so small that, had we not formally identified it, it is perfectly possible to imagine no one would have noticed anything was wrong.
Excellent post. Confirms my bias, just what Lazygun wanted, so, yeah.

5. Member
Join Date
Jan 2019
Posts
558
Originally Posted by Yngvi
Your article says mechanical ventilators are not available. It does not say the virus has put them to use, creating a shortage. It sounds like they did not have any of the machines in the first place.
Here is a fellow who thinks Papa Giovanni XXIII Hospital had no ventilators "in the first place". He also thinks his ideas about medical equipment usage are more sound than the entire medical establishment.

Shortages exist right now. HCWs in the US are reusing disposable N95s, rated for a maximum of 8 hours, for days at a time - putting themselves at great risk. In order to make this dire situation legal, the CDC was forced to change its guidelines, weeks ago. PPE-intensive procedures, like elective surgery, are cancelled indefinitely. Yet, because there is still a week's worth of N95s in the supply closet (under lock and key, in most hospitals), you blissfully declare there is no shortage.

Unlike toilet paper, actual usage of PPE has increased dramatically. Many more staff have to follow contact and droplet precautions, more of the time.

So, it is currently irrational panic based on unproven projections that has overwhelmed the system, instead of the actual virus?
PPE predictions are for the next 1 or 2 weeks out. These predictions come from different data, and from different people, than the epidemiological predictions which have been (rightfully) maligned.

6. The largest electrical contractor in the state of Utah just laid off 200 electricians. The state is considering a 30-week shelter-in-place ordinance.

7. Member
Join Date
Mar 2008
Posts
475
The other day, an ER doctor at Harborview here in Seattle said on the news that over the past 2 weeks, they’ve seen a leveling off and some decline in the number of Covid-19 cases requiring hospitalization. When they cut away from that, there was no mention of what was just said and they went right back to the hysteria-inducing modeling.

8. Member
Join Date
Jan 2014
Posts
95
Originally Posted by Mark Rippetoe
Right. The model predicts that by October 1 the entire mass of the known universe will have been converted to COVID19 virus particles.

October 1, obviously.
Rip -- you know the model doesn't predict that. It predicts exponential increase only in the early stages, which we're in.

But could you please seriously answer my question? The current ~800 deaths and 3 day doubling time is observed. It's what's happening right now.

So when do you think the exponential growth will stop? Obviously before April 15 right? But can you give an indication of when and why in that time frame?

I'm honestly not asking this as a rhetorical question. Given where the numbers are now and the trend we're on, I find it genuinely difficult to understand why somebody like you -- who believes in maths and science -- doesn't see a significant problem. There is a disconnect here between what you see and what I see, and I want to understand what that is.

9. Member
Join Date
Jan 2011
Posts
1,124
Ok Professor, this is getting tiresome.
Originally Posted by lazygun37
The article I linked shows that, among the 2449 US cases with known ages studied
1. the percentage of cases requiring hospitalization was between 20.7% and 31.4%
2. the percentage of cases requiring ICU care was between 4.9% and 11.5%
3. the percentage of cases who ended up dying was was between 1.8% and 3.4%
Do you know the difference between a point estimate and a confidence interval? Because the author(s) of the paper don’t. They used the point estimate as the lower bound and completely made shit up for the upper. They took the same numerator and used a different denominator and called that the upper bound. Either you don’t know the difference either, or you didn’t read the footnote under the table you just copy pasted.
Originally Posted by lazygun37
This is almost certainly true…
The need for a ventilator will usually be
I haven't been able to find the exact numbers
I think I found one very small study
I also recall another Chinese study
which presumably means
But both of these suggest
reasonable first approximation
Aren’t you trying to pass yourself off as a numbers guy? For someone accusing others of “not engaging with the numbers”, you sure as hell like to not use real ones (or use them incorrectly).

Originally Posted by lazygun37
They would probably prefer to receive ventilator treatment, if they are given a choice...
Another assumption on your part not based in fact, Professor. Did you know that a couple of the people from the old folks' home in Seattle that had that cluster of deaths early on had DNRs? I got that from a JAMA podcast (Dr. John Lynch, MD, MPH).
Originally Posted by lazygun37
But none of you seem to actually want to do even the slightest amount of home work to actually show that your views are backed up by actual data.
My main view (as I have said over and over again), is the data is shit.

10. Member
Join Date
Jan 2019
Posts
558
Source? 30 weeks is ridiculous.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•