|
Any and all updates regarding the COVID-19 will need a source provided. Please do your part in helping us to keep this thread maintainable and under control.
It is YOUR responsibility to fully read through the sources that you link, and you MUST provide a brief summary explaining what the source is about. Do not expect other people to do the work for you.
Conspiracy theories and fear mongering will absolutely not be tolerated in this thread. Expect harsh mod actions if you try to incite fear needlessly.
This is not a politics thread! You are allowed to post information regarding politics if it's related to the coronavirus, but do NOT discuss politics in here.
Added a disclaimer on page 662. Many need to post better. |
Austria653 Posts
On May 05 2020 02:00 warding wrote:Show nested quote +On May 05 2020 01:55 JimmiC wrote:
Stockholm only has a million people, and they have 1470 of the 2800 deaths in all of swedes. So it looks like you are off by a factor of 10. If Stockholm has half the deaths of the country, then on April 11th it's reasonable to assume it had 630 deaths. 630 / 100 000=0.63%. Definitely higher than 1.5% but not a factor of ten.
It's not that easy because you are missing all the deaths of the people that are infected on april 11 but not yet dead.
Might be a lot since it takes a while to die from this.
|
On May 05 2020 18:56 Garbels wrote:Show nested quote +On May 05 2020 02:00 warding wrote:On May 05 2020 01:55 JimmiC wrote:
Stockholm only has a million people, and they have 1470 of the 2800 deaths in all of swedes. So it looks like you are off by a factor of 10. If Stockholm has half the deaths of the country, then on April 11th it's reasonable to assume it had 630 deaths. 630 / 100 000=0.63%. Definitely higher than 1.5% but not a factor of ten. It's not that easy because you are missing all the deaths of the people that are infected on april 11 but not yet dead. Might be a lot since it takes a while to die from this.
Also all the cases that died from it but never got diagnosed. A lot of places have higher death rates than normal when checking that and removing Corvid deaths. Consider economy is slowed down it should be less deaths than normal, not more.
|
On May 05 2020 18:36 Gorsameth wrote:Show nested quote +On May 05 2020 10:22 Emnjay808 wrote:
What does “corona free” mean? Like antibodies or just complete absence of the virus in BC? I assume absence of virus. If lockdown reduces the R0 to less then 1 the virus will slowly burn itself out as less and less people get infected. It takes a long time if you have lots of infected but if your dealing with small numbers its something you could strive for.
I only wonder what the plan is from that point. Keeping the countries closed until a vaccine shows up? That's probably going to take many years. As I see it, the only possible way to manage this is to protect the people at risk and in general try to slow the spread of the virus so as to not overwhelm the health care system.
|
As has been discussed multiple times in this thread, the general plan is not (and has never been) to lockdown until vaccine. Originally, it was just flatten the curve to keep the healthcare systems from being overwhelmed, but that seems to have evolved into:
Reduce amount of infected people until you can individually track them again.
Individually track infections, and test and quarantine everyone the infected person was in contact with.
Which is clearly a possible way of solving this. I think the term used for it is "Hammer and Dance". Hammer down the infections, then open up quite far again. No one is arguing for infinite lockdowns, or lockdowns until vaccine.
|
It's incredible how many people still operate as though a longterm lockdown is or ever was even an option, threats of endless lockdowns are still routinely trotted out over here as support for reopening things.
|
On May 05 2020 09:00 Stratos_speAr wrote:
Also Warding's claims are based on faulty math and questionable sources. They will most likely be tossed in the bin with all of the other "this isn't really that bad!" claims that have been debunked over the last couple months.
Geez, I'm not making claims, I'm pointing to some pieces of evidence and asking whether my hunches are correct. I understand that nuanced and non-polarized discourse might sound like foreign concepts to americans but try not to categorize people as itsjustfluists if they try to look at things a little bit more optimistically.
As for the quality of the sources, there are dozens of studies on antibody prevalence in multiple geographies out there. We're a couple of weeks past the stage of having only those two CA studies suggesting a 50x to 80x undercount.
Here's a new one from Slovenia:
2% to 4% infected (95% CI) suggesting an IFR of 0.12% to 0.24%.
https://www.reddit.com/r/COVID19/comments/gdufh8/preliminary_reports_first_national_serology_study/
On May 05 2020 18:56 Garbels wrote:Show nested quote +On May 05 2020 02:00 warding wrote:On May 05 2020 01:55 JimmiC wrote:
Stockholm only has a million people, and they have 1470 of the 2800 deaths in all of swedes. So it looks like you are off by a factor of 10. If Stockholm has half the deaths of the country, then on April 11th it's reasonable to assume it had 630 deaths. 630 / 100 000=0.63%. Definitely higher than 1.5% but not a factor of ten. It's not that easy because you are missing all the deaths of the people that are infected on april 11 but not yet dead. Might be a lot since it takes a while to die from this.
There is some lag in the deaths but I would assume there's also a lag in developing antibodies. I'm not sure which one is greater.
On May 05 2020 20:12 Simberto wrote:
Which is clearly a possible way of solving this. I think the term used for it is "Hammer and Dance". Hammer down the infections, then open up quite far again. No one is arguing for infinite lockdowns, or lockdowns until vaccine.
I don't think this characterizes the current debate. It's currently being debated whether or not we will have to lockdown again in the second wave, whether it's safe to open up beaches, whether it's safe to have children back in schools in the fall (or in summer camps?), whether it's safe to reopen the football season, and so on. There is ample debate on how draconian the measures of the 'dance' should be. And IMO not enough debate on the cost-benefit of each one of them and what the right cocktail might be.
|
|
|
On May 05 2020 20:12 Simberto wrote:
As has been discussed multiple times in this thread, the general plan is not (and has never been) to lockdown until vaccine. Originally, it was just flatten the curve to keep the healthcare systems from being overwhelmed, but that seems to have evolved into:
Reduce amount of infected people until you can individually track them again.
Individually track infections, and test and quarantine everyone the infected person was in contact with.
Which is clearly a possible way of solving this. I think the term used for it is "Hammer and Dance". Hammer down the infections, then open up quite far again. No one is arguing for infinite lockdowns, or lockdowns until vaccine.
On May 05 2020 20:44 farvacola wrote:
It's incredible how many people still operate as though a longterm lockdown is or ever was even an option, threats of endless lockdowns are still routinely trotted out over here as support for reopening things.
Maybe I didn't express my self clearly. I wasn't talking about endless lockdowns per se. I just wonder what will happen when you have eliminated the disease in one country since it will just come back again as soon as the country opens its borders if there isn't a large population with antibodies. If you keep the country in a permanent state of emergency, then you can maybe catch every new occurrence of the disease coming from the outside early enough - but then you have to keep those restrictions in place and it seems to me that they'd have to be very severe.
Isn't the Hammer and Dance strategy just another word for slowly letting people get infected but you "hammer" down the infection when it gets to dangerous levels? That would be another way of saying what I argue for, not a way of eliminating the virus.
|
On May 05 2020 22:27 warding wrote:Show nested quote +On May 05 2020 09:00 Stratos_speAr wrote:
Also Warding's claims are based on faulty math and questionable sources. They will most likely be tossed in the bin with all of the other "this isn't really that bad!" claims that have been debunked over the last couple months. Geez, I'm not making claims, I'm pointing to some pieces of evidence and asking whether my hunches are correct. I understand that nuanced and non-polarized discourse might sound like foreign concepts to americans but try not to categorize people as itsjustfluists if they try to look at things a little bit more optimistically.
It's not justfluism for being optimistic; it's justfluism for pushing ideas that don't even pass a smell test with a rhetoric that comes off as downright conspiratorial.
We had a pandemic fairly recently with death rates about in that range and a very high infectivity. It was called the swine flu, and it looked nothing like what we're seeing with the coronavirus. If you look at how this disease is playing out right now and assert that its numbers are probably more or less comparable to swine flu, then I'm not really sure what to say beyond offering concurrence with Stratos_speAr's statement quoted.
|
On May 05 2020 04:40 LegalLord wrote:
Inclined to view the 0.1% claims as absurd on their face, and likely based on cherry-picked data. Few reliable sources seem to support the optimistic conclusion.
How about this one, University of Oxford's Center for Evidence Based Medicine:
Taking account of historical experience, trends in the data, increased number of infections in the population at largest, and potential impact of misclassification of deaths gives a presumed estimate for the COVID-19 IFR somewhere between 0.1% and 0.41%.*
*Estimating CFR and IFR in the early stage of outbreaks is subject to considerable uncertainties, the estimates are likely to change as more data emerges.
https://www.cebm.net/covid-19/global-covid-19-case-fatality-rates/
Meanwhile, another study in Heinsberg, Germany, arrives at a 0.35% IFR, and it's author thinks it is likely to be lower given that they were conservative in their approach.
https://www.uni-bonn.de/news/111-2020?set_language=en
It's not that "it's just flu", it's still probably deadlier. It's still quite substancial, however, that most of us still have in our minds the numbers of 2 to 4% fatality rate and yet now it's likely to be somewhere between 0.1% to 0.5%.
EDIT: So none of these studies pass the smell test?
|
About Heinsberg:
This study got to this number by testing a bunch of people (919), having 15% infected, then continued to assume that the entire city of Gangelt also has 15% infected, then saw that 7 people died in the city and then computed that 0.35% death rate. Just hope nobody dies tomorrow, otherwise the lethality rises by 0.5%
It is up to you how you evaluate this. I know I wouldn't have passed any exam using this kind of statistic.
|
Also, note that Gangelt was definitively not representative of...anything. It is a very specific situation where a lot of the village went to some carnival festivity, and then were infected there. To generalize towards any larger population from any study about Gangelt is foolish.
|
I don't get the above posts. Gangelt is certainly not representative of the world, neither should we extrapolate it, but it's a data point. That's why I shared dozens of studies from dozens of geographies around the world.
Obviously if someone dies tomorrow in Gangelt then the IFR there goes up considerably. On the other hand, the study also tells you about the undercount in reported cases and that is more solid when it comes to extrapolating, at least to te rest of Germany. If you do that extrapolation, it seems to be that you arrive at a similar IFR for the whole country.
|
And if there was a retirement home in Gangelt, we would now talk about 10% IFR... Again: this study is based on SEVEN deaths. And from there they extrapolate all kinds of wild stuff.
Gangelt is a small village, one couple was infected, cuddled with everyone at a carnival, and 3 days later the entire village was locked down. Not the normal lockdown, but completely and utterly stand still, nobody moves locked down.
Projecting those numbers onto anything is absurd. But hey, in your first paragraph you go by "we shouldn't extrapolate" to "lets extrapolate from this" in your second paragraph.
And the fact that this study had about as many PR people hired as they had researchers and that they are led by a guy who since January is spouting "It's just a flu" on Twitter is just the icing on the reputable source cake.
|
On May 05 2020 23:38 warding wrote:Show nested quote +On May 05 2020 04:40 LegalLord wrote:
Inclined to view the 0.1% claims as absurd on their face, and likely based on cherry-picked data. Few reliable sources seem to support the optimistic conclusion. How about this one, University of Oxford's Center for Evidence Based Medicine: Show nested quote +Taking account of historical experience, trends in the data, increased number of infections in the population at largest, and potential impact of misclassification of deaths gives a presumed estimate for the COVID-19 IFR somewhere between 0.1% and 0.41%.*
*Estimating CFR and IFR in the early stage of outbreaks is subject to considerable uncertainties, the estimates are likely to change as more data emerges. https://www.cebm.net/covid-19/global-covid-19-case-fatality-rates/Meanwhile, another study in Heinsberg, Germany, arrives at a 0.35% IFR, and it's author thinks it is likely to be lower given that they were conservative in their approach. https://www.uni-bonn.de/news/111-2020?set_language=enIt's not that "it's just flu", it's still probably deadlier. It's still quite substancial, however, that most of us still have in our minds the numbers of 2 to 4% fatality rate and yet now it's likely to be somewhere between 0.1% to 0.5%. EDIT: So none of these studies pass the smell test?
That CEBM page doesn't support your hypothesis though. Their estimates based on China and on Italy data split it all out by age ranges, but overall CFR is far higher than what you mention. Moreover, they say in the opening paragraph that because there is so much heterogeneity (read: uncertainty) in the data, that an overall "worldwide" estimate cannot be made (read: any extrapolation is guesswork, not statistics).
|
mahrgell I understood your point regarding the small sample of death, did you get my point about extrapolating based on undercounts?
EDIT: Also, I'm curious. Because a guy had an opinion in January and in April he still has the same opinion, reinforced by a study he did, that makes him unreputable because... that opinion is different from yours? Do you suppose he's secretly financed by Big Corona or something?
Acrofales I'm not sure I follow. In that page they say that about CFR because there is indeed heterogeneity in the data - some countries do a lot of testing, some countries do little testing and some countries data may be... uhm... dubious. My point was about IFR which they do attempt to point towards an estimation. They state that the purpose of that page includes setting out a best estimate of IFR. They say that an IFR of 0.41% is likely an overestimation. Then they state that, taking into a account a bunch of factors, they presume IFR is between 0.1% and 0.41%. How does this not support my point?
|
What’s the utility in bending over backwards to estimate IFR, a measure that’s necessarily less accurate than CFRs given its inclusion of asymptomatic and undiagnosed cases? The fact that the CFR is already difficult to calculate leans against use of an IFR estimate, particularly for any non-statistical purpose. Surely, you aren’t leaning on an IFR estimate for the hell of it.
|
On May 06 2020 02:25 warding wrote:
mahrgell I understood your point regarding the small sample of death, did you get my point about extrapolating based on undercounts?
EDIT: Also, I'm curious. Because a guy had an opinion in January and in April he still has the same opinion, reinforced by a study he did, that makes him unreputable because... that opinion is different from yours? Do you suppose he's secretly financed by Big Corona or something?
Acrofales I'm not sure I follow. In that page they say that about CFR because there is indeed heterogeneity in the data - some countries do a lot of testing, some countries do little testing and some countries data may be... uhm... dubious. My point was about IFR which they do attempt to point towards an estimation. They state that the purpose of that page includes setting out a best estimate of IFR. They say that an IFR of 0.41% is likely an overestimation. Then they state that, taking into a account a bunch of factors, they presume IFR is between 0.1% and 0.41%. How does this not support my point?
His estimations on the undercount in Germany are done by using his IFR...
So if you doubt his IFR computations, using this value any further is not creating more accurate data.
And not sure why you are surprised by me expecting confirmation bias (and weird interpretations aligning with his previous stated opinions) for someone with his track record.
That would be like a study by the AfD showing that immigrants are the root of all evil, the Left finding out that UBI is the ultimate solution to all problems, market liberals showing "research" that lower taxes are the only way to get anything done or Portuguese showing that only the referee decides a footie game.
|
On May 06 2020 02:25 warding wrote:
mahrgell I understood your point regarding the small sample of death, did you get my point about extrapolating based on undercounts?
EDIT: Also, I'm curious. Because a guy had an opinion in January and in April he still has the same opinion, reinforced by a study he did, that makes him unreputable because... that opinion is different from yours? Do you suppose he's secretly financed by Big Corona or something?
Acrofales I'm not sure I follow. In that page they say that about CFR because there is indeed heterogeneity in the data - some countries do a lot of testing, some countries do little testing and some countries data may be... uhm... dubious. My point was about IFR which they do attempt to point towards an estimation. They state that the purpose of that page includes setting out a best estimate of IFR. They say that an IFR of 0.41% is likely an overestimation. Then they state that, taking into a account a bunch of factors, they presume IFR is between 0.1% and 0.41%. How does this not support my point?
Well, I'd say they quite explicitly DON'T make an estimate of the IFR. They say they could, and that that would probably be an overestimate, and then spend the subsequent paragraphs hedging that, after spending the previous paragraphs hedging that as well. They also explicitly mention that this is a somewhat useless number, as the risk to elderly is considerably higher than any "overall" IFR would indicate, so you'd have to stratify it by age.
I don't see much justification for a conclusion like this:
Am I wrong to think that after this first peak in Europe and the US, we're going to realize the virus is less deadly than we first feared and that future outbreaks in the same geographies are likely to be way smaller?
based on anything I read there. The problems plagueing Italy and Spain had very little to do with Covid's direct mortality rate among the poppulation at large. It hit hard, because it hospitalized many people and elderly in particular. I'd agree with you "the same geographies" could be hit less hard if it had gone through all the elderly, and those remaining are immune, but that seems extremely unlikely. So a renewed outbreak would find its way into elderly homes that escaped unscathed the first time around (which are still in the majority, because despite all the scare, the virus has not infected anywhere near enough people for their to be any semblance of herd immunity under normal behavour). If we keep up social distancing even when deescalating and keep the transmission rate low, and people get infected slowly enough that hospitals can cope, then hopefully we can mostly go about things almost normally. But if the transmission rate goes back up then we are right back where we started. And the policies of deescalation around Europe are aimed exactly at trying t find that "new normal" with a low transmission rate. IFR doesn't really have much to do with that. We've seen CFR spike and should really avoid that again.
|
Estimating IFR (infection fatality rate) in a situation where the healthcare system is not beeing overrun is not that difficult though it does come with considerable uncertainty.
You take the best estimate for casualties at a certain date which is the excess death rate. This almost per definition excludes all the people that would have died anyway if there was no virus.
Then you compare this with the best estimate for the number of cases ~15-20 days before that date.
This is more difficult and does come with a lot of uncertainty as far from every case is detected. It depends on how much testing is beeing done amongst others and its different for every single country but it seems fair to at least 10 fold the number of detected cases for most western nations.
Less deaths due to less economic activity and traffic and more deaths because people dont dare to go to the hospital seem to be so small compared to the deaths of the virus that it will have minimal effect on the calculated IFR and factors like this can be ignored.
The most accurate and complete data set we have is the data set from the diamond princess which shows a 2% death rate with the best medical care which was available at the start of the pandemic and in a situation where the healthcare system is not overrun.
https://www.statista.com/statistics/1099517/japan-coronavirus-patients-diamond-princess/
This data set might be skewed towards elderly people but i have no accurate info on that.
There are factors which should lower the IFR over time,hospitals get better at treating patients and to put it bluntly,eventually the virus will run out of vulnerable people to infect.
There is also a factor which could raise the ifr over time and that is long term effects of the virus infection the extend of which at this point is virually impossible to predict but there definitely is evidence for long term damage in some of the cases which will have an effect on the life expectancy for those patients.
In the end the most "interesting" and imo relevant figure is the impact of the virus on the average life expectancy of the whole population,which at this point is virtually impossible to predict.
Asume for example that the whole population will get infected at one point and that eventually 10%+ of the population will die from the virus when they get older (this might sound far fetched but in many regions the virus has accounted for more then 30% of all deaths during the past 2 months). Then you have a 10% ifr which sounds horrible. But if it on average "only" lowers life expectancy with 1-2 years then it isnt really worse then things like air polution or the flu and far less worse then smoking. (smoking lowers life expectancy with 7-8 years).
In the long run i think the impact on life expectancy will be small,a vaccine and better threatment methods will come over time.
@first post on next page. I will slightly adjust,considering a situation in which the healthcare system is not beeing overrun.
|
|
|
|
|
|
EPT
