The Corona Virus

[Aedilred]

I suspect that the infection rate is a lot higher than known, although that is speculative. We won't know until many, many more tests have been done.

I'm fairly confident that I've had it. If so, it was mild and under normal circumstances I wouldn't have considered missing work (as it was, we were under lockdown by the time that I became symptomatic). I haven't been tested though and there is no prospect of being tested any time soon at current foreseeable testing rates, given my susceptibility and line of work. I know, and know of, several people in similar circumstances. I don't personally know or know of anyone who's been diagnosed with it.

That is of course entirely anecdotal but I think it's also pretty much a given that in most countries and the UK in particular the official tallies for both infection and deaths are understated, the first due to lack of testing, the second due to the counting method.

It also seems likely, given the nature of what we're counting in each instance, that infections are being understated by a much higher factor than deaths are. But that will vary by country.

It's difficult to make sense of the data on a global scale because each country is conducting different numbers of tests, the tests aren't all equally reliable, and each country also reports deaths differently. The more tests are done, the more accurate the infection rate is likely to be, not only because larger sample sizes are better, but because where testing is done on a smaller scale it's likely to be done on patients who are more obviously infected. But the tests also aren't being done in a vacuum: increased testing allows earlier interception of the infection and better management of healthcare resources which should in turn depress the mortality rate.

For instance, Germany's tested around 2% of its population and the infection rate among those tested is a little under 10%. The UK by contrast has tested around 0.6% of the population with an infection rate of around 35%. But the mortality rate among German confirmed infectees is to date about 3.5% and in the UK it's about 13.5%. That doesn't account for differences in how the two countreis count deaths, though.

The flipside is that increased testing also facilitates better containment of the virus. Which is good, because it means fewer deaths. But until a vaccine is ready, its spreading is also the only way to create immunity. And herd immunity is the only viable long-term strategy for dealing with the virus.

This is the looming problem that I don't think anyone, even in countries that have dealt with it pretty well so far, really has a clear idea of how to deal with: how we get out of this. Or more specifically: the goal to date has been to flatten the curve to minimise the short-to-medium-term impact, but what happens once the curve has been flattened? Even if the UK's ludicrous 35% figure is accurate across the whole of the population (it's not), that's still way too low for herd immunity.

Leaving the lockdown in place will keep the virus contained and prevent spread, but it's also just putting the problem on ice while creating a host of other problems which may well prove to be deadlier in some ways than the virus itself. A vaccine is still too far off to seriously consider leaving the lockdown in place until it's ready for use. But once you lift the lockdown, it'll immediately start spreading again and people will start to die. It's an incredibly difficult decision and I worry that nobody will make it, which could end up being the worst of all worlds.

[Tvtyrant]

I admire your optimism.

I told a coworker in early March that is isn't if but when you get it. He scoffed at me, told me we wouldn't lock down. I'm going for optimism because my family are all high risk (either elderly, morbidly obese, or both) and giving into hysteria is never good. Always assume you will survive and plan accordingly.

Here they're apparently only counting you as infected if you have to go into hospital, and also apparently they don't count you as recovered at all, so we have infected and deaths, and the uninfected. So it's possible that the number of people who've had the disease is much higher, but nobody knows.

Yeah but that issue is true of every virus, how many more or less of 40,000,000 was there really? The margin of error is always going to be there.

[Rockphed]

Yeah, that "theory" that people are just dying *with* the disease is absolute cobblers, you just have to look at the death rates as they are now compared to last year to realise that--you don't get a statistical increase like that without some underlying cause.

I saw a post somewhere arguing that the US death rate has been low for the last 6 or 7 weeks. Do you happen to know where to grab death rate data for the last few years?

I admire your optimism.

I told a coworker in early March that is isn't if but when you get it. He scoffed at me, told me we wouldn't lock down. I'm going for optimism because my family are all high risk (either elderly, morbidly obese, or both) and giving into hysteria is never good. Always assume you will survive and plan accordingly.

I am seeing things around the internet that suggest the infection rate in the US is 25 - 500 times the reported figure. I am seeing reports that New York did a test of people out and about and found that about 15 percent of them had antibodies for the virus, which either means their test had atrocious false-alarm-rate, or that we have vastly undercounted the number of infected.

[Peelee]

I am seeing things around the internet that suggest the infection rate in the US is 25 - 500 times the reported figure. I am seeing reports that New York did a test of people out and about and found that about 15 percent of them had antibodies for the virus, which either means their test had atrocious false-alarm-rate, or that we have vastly undercounted the number of infected.

Hold on, I need to grab my fainting couch.

[Kaptin Keen]

Here they're apparently only counting you as infected if you have to go into hospital, and also apparently they don't count you as recovered at all, so we have infected and deaths, and the uninfected. So it's possible that the number of people who've had the disease is much higher, but nobody knows.

Those danish numbers I keep quoting? The official number of infected is like ... 8000.

The real number, as I said a little while ago, is more like 200.000.

If you calculate death rate from the official number, it's just under 5%. But if you calculate it from the real number ..

[Tvtyrant]

Those danish numbers I keep quoting? The official number of infected is like ... 8000.

The real number, as I said a little while ago, is more like 200.000.

If you calculate death rate from the official number, it's just under 5%. But if you calculate it from the real number ..

South Korea had the highest percentage of testing and most effective response to Covid, its numbers come down to 2% of cases are fatal. The US has the most cases and deaths reported, it comes down to 5%. So Denmark's "fake" infection rate is accurate to all other infection rates fatalities, and your "real" infection rate gives us a .001% rate, totally out of whack with other countries. At this point you either have to admit you are wrong, or claim that there is a global conspiracy to set the numbers to a predetermined range of fatalities.

[Kaptin Keen]

South Korea had the highest percentage of testing and most effective response to Covid, its numbers come down to 2% of cases are fatal. The US has the most cases and deaths reported, it comes down to 5%. So Denmark's "fake" infection rate is accurate to all other infection rates fatalities, and your "real" infection rate gives us a .001% rate, totally out of whack with other countries. At this point you either have to admit you are wrong, or claim that there is a global conspiracy to set the numbers to a predetermined range of fatalities.

No, we just have much better dark figures than everyone else.

[veti]

No, we just have much better dark figures than everyone else.

Denmark has tested around 2% of its total population, which is about the average for a prosperous European country. Of those tests, about 7% have been positive, but of course the sampling will be weighted heavily towards "people with symptoms". Which to me suggests that, of Danes who presented themselves to health services for testing, less than 1 in 10 has/had the virus. In other words, there are still ten times as many Danes suffering from various strains of regular colds/flu, as from Covid-19.

Here in New Zealand, the government believes it has identified very nearly all the cases. The health system has not been overwhelmed (isolation is a wonderful thing at times like this), and the mortality rate is currently sitting at a little over 1%. It's likely to creep a little higher, but if we're lucky it may yet stay below 2%. Australia is looking similar, and several other (mostly small) countries are reporting comparable figures. The very low rate of detecting new infections (we're down to single figures per day, both here and in Oz) implies strongly that there really isn't the huge pool of undetected cases in the population that some theories seem to suggest/require.

[farothel]

Here in Belgium they did a study using around 4,000 samples from people donating blood (which is routinely tested on a large number of things anyway) and they came to approximately 3% of these tests having antibodies against the virus. This is already a month or so old, but the numbers will probably not have risen all that much.

This is in line with what has been reported in other European countries.

[Kaptin Keen]

In other words, there are still ten times as many Danes suffering from various strains of regular colds/flu, as from Covid-19.

Well - they're frankly not particularly similar. At least, chest pain is not a common symptom of the flu, nor is the dry cough.

Regardless - 'dark figures' are unreliable, I'll give you that.

Here in Belgium they did a study using around 4,000 samples from people donating blood (which is routinely tested on a large number of things anyway) and they came to approximately 3% of these tests having antibodies against the virus. This is already a month or so old, but the numbers will probably not have risen all that much.

This is in line with what has been reported in other European countries.

All blood donors are tested, obviously, here as well. Mortality rates among blood donors is 0,13%. Now, blood donors will generally be in good health, and younger than 65, so it tells us precious little about the risk groups. But 0,13% is pretty solid for the rest of us.

[Chen]

Overall fatality rate is kind of a meaningless number from any type of personal point of view anyways. Its so wildly skewed in different directions based on age and underlying health conditions that saying there's an overall death rate of 1% or 5% or whatever doesn't give any individual any good information on how at risk THEY are.

It's useful as a statistical tool to determine overall deaths and such, but that's about it.

[Strigon]

Here in New Zealand, the government believes it has identified very nearly all the cases. The health system has not been overwhelmed (isolation is a wonderful thing at times like this), and the mortality rate is currently sitting at a little over 1%. It's likely to creep a little higher, but if we're lucky it may yet stay below 2%. Australia is looking similar, and several other (mostly small) countries are reporting comparable figures. The very low rate of detecting new infections (we're down to single figures per day, both here and in Oz) implies strongly that there really isn't the huge pool of undetected cases in the population that some theories seem to suggest/require.

If a high proportion of those infected are asymptomatic, though, they could still go undetected. It's quite possible the conditions over there simply aren't conducive to asymptomatic transmission, and that there are many other people out there who have the disease, but aren't coughing enough to reliablly spread it.

Of course, this is entirely speculative. Literally no direct evidence for this at all.

[Joran]

I saw a post somewhere arguing that the US death rate has been low for the last 6 or 7 weeks. Do you happen to know where to grab death rate data for the last few years?

For Europe, EURO MOMO has been tracking mortality rates. So-called "excess mortality" can be tracked; these are deaths outside what statisticians would expect for a given year.

Not all of the excess deaths will be directly caused by COVID-19; for instance, stroke/heart attack victims have mostly disappeared from emergency rooms, so instead they may be dying at home.

https://www.euromomo.eu/graphs-and-maps/

The Economist has a good article about excess mortality and why the COVID-19 death rates are probably undercounts.
https://www.economist.com/graphic-de...ross-countries

[NotASpiderSwarm]

I saw a post somewhere arguing that the US death rate has been low for the last 6 or 7 weeks. Do you happen to know where to grab death rate data for the last few years?

NYT ran a story about that around the 19th or so. I'm out of free articles so I can't track it down, and the site that summarized it for me is probably political enough that I shouldn't link it here, but short version: New York is looking at a 300% increase in deaths in March compared to the average for the last few years. Other countries were analyzed as well, including Spain, but not the US as a whole. If you've got a NYT sub, you should be able to find it with a bit of effort. The problem is, average death rate and current death rate are both stats that some countries/cities/states keep quiet for various reasons, so you can't get a full picture of, say, the entire US, just of areas that release the data.

Also, all the statistics are going to be messy, with a lot of conflating variables. Flu deaths are down. Domestic abuse deaths are up. Car wrecks vary by area, with fewer drunk driving accidents but more accidents due to people speeding because this is the first time they can get over 60 during their commute. The overall +- is probably fine for back-of-the-napkin calculations, but if you want accuracy, you need to break things out by cause before figuring out which are COVID.

[Rockphed]

NYT ran a story about that around the 19th or so. I'm out of free articles so I can't track it down, and the site that summarized it for me is probably political enough that I shouldn't link it here, but short version: New York is looking at a 300% increase in deaths in March compared to the average for the last few years. Other countries were analyzed as well, including Spain, but not the US as a whole. If you've got a NYT sub, you should be able to find it with a bit of effort. The problem is, average death rate and current death rate are both stats that some countries/cities/states keep quiet for various reasons, so you can't get a full picture of, say, the entire US, just of areas that release the data.

Also, all the statistics are going to be messy, with a lot of conflating variables. Flu deaths are down. Domestic abuse deaths are up. Car wrecks vary by area, with fewer drunk driving accidents but more accidents due to people speeding because this is the first time they can get over 60 during their commute. The overall +- is probably fine for back-of-the-napkin calculations, but if you want accuracy, you need to break things out by cause before figuring out which are COVID.

I found the post I had been thinking of, but they don't include a nice link to where I can get the data. Also, it is on a rather one-sidedly political site, so I won't link the post.

And thank you to Joran, though I don't see the actual data on that page, just their analysis of it. Although interestingly they show below average excess deaths until a couple weeks ago.

[Cikomyr2]

Sorry to veer this conversation, but WHO came out saying there's not evidence yet that someone who caught it can't catch it again.

I mean, I get that the absence of proof is not proof of the opposite, but on a scale of 1 to 10, how worrying is that? Shouldn't we have started to see an immunity in previously sick healthcare workers that have been at this since the start?

[Kaptin Keen]

Sorry to veer this conversation, but WHO came out saying there's not evidence yet that someone who caught it can't catch it again.

I mean, I get that the absence of proof is not proof of the opposite, but on a scale of 1 to 10, how worrying is that? Shouldn't we have started to see an immunity in previously sick healthcare workers that have been at this since the start?

I'd say this has to do with misinformation. Not from the WHO, obviously, but ... other sources.

So .. say one source is being less than forthright with a lot of information - numbers of infected, dead, spread of the virus (we've contained it to Huwan Province), and they also lie about how many have recovered, don't actually test those officially counted as free of the virus.

Now you have a picture where, apparently, people who had it and beat it, seem to be getting it again.

Utter guesswork, of course - I claim no knowledge of this. But having one government behave 100% as expected to me sound more believable than having one virus behaving 100% not-as-expected.

[Strigon]

Sorry to veer this conversation, but WHO came out saying there's not evidence yet that someone who caught it can't catch it again.

I mean, I get that the absence of proof is not proof of the opposite, but on a scale of 1 to 10, how worrying is that? Shouldn't we have started to see an immunity in previously sick healthcare workers that have been at this since the start?

The issue is that we don't really know what immunity looks like.
We can look for antibodies, but there are different types of antibodies, and not all of them confer immunity. Of those that do, not all are permanent.

And keep in mind that we want to keep people from being infected. It would be unethical - to both healthcare workers and the patients they care for - to stop those who have already recovered from COVID from wearing PPE, just to see if they get infected again. Some locations have already begun reporting apparent reinfections, but the rate is low enough to be attributed to false results, either negative (they weren't yet fully cured) or positive (the test picked up an infection that wasn't there.) I suspect that, if there weren't some period of immunity, we'd see higher rates of reinfection - especially as people who believe themselves immune reintegrate into society, probably with less concern for the virus.

Essentially, though, it's just the WHO reminding everyone not to bet the farm on immunity, because it isn't guaranteed. If a country acts as though people who have recovered aren't at risk for getting sick again, or spreading the sickness, that country could be in for a world of hurt on the off chance that turns out to be incorrect.

Remember, this isn't just a matter of absence of evidence vs evidence of absence, it's a matter of trying to prove something to be impossible - a much more difficult task than proving something to be possible.

[jayem]

Sorry to veer this conversation, but WHO came out saying there's not evidence yet that someone who caught it can't catch it again.

I mean, I get that the absence of proof is not proof of the opposite, but on a scale of 1 to 10, how worrying is that? Shouldn't we have started to see an immunity in previously sick healthcare workers that have been at this since the start?

They have to be known to be sick (as in tested, not self diagnosed), known to be recovered, and then known to have 'ought' to have caught it again and then known not to have caught it again. So that's 6 weeks?

As people have been in dangerous situations and not caught it, we need probably a cohort of 100, not to expect a false negative.
While the people we need to be looking at here will be roughly those who were sick from the start of March. Non-medical staff shouldn't be back in risky positions (except deliberately for this purpose),medical staff are mostly going to catch it from the patients as hospitals struggle (I.E the epidemic should have been endemic two weeks before,mid Feb. So total tie is about 8 weeks).
That basically leaves us with Wuhan, or the one Italian Dr.
Wuhan, went into lookdown at 20 (recognised) deaths 23 Jan, which seems a good point to start our 6 weeks. So our hypothetical Doctor recovers and resumes work around the 23rd Feb (roughly just after the peak), and basically works the 2 weeks at the tail anyway (at which point there were only 5000 new cases and realistically most of them were caught before then, at which point its hardly surprising he didn't catch it again).

TLDR:Wuhan didn't last long enough to have two illnesses expected (I think the same could be said for Italy/Spain). Everywhere else hasn't lasted long enough to have two illnesses expected. So we shouldn't be surprised to see no cases of double illnesses [plus as noted, some false maybes are to be expected]
That will of course change if things go back anywhere, or end of lockdowns are badly handled.

[Chen]

A number of reports also seem to indicate that these people who have “caught” it again may not have actually been reinfected, but just virus was still detected in their system. As the WHO says theres no current evidence to show how immunity actually works. In the end the mechanisms for immunity are generally the same in vaccines or infection acquired immunity. You’re going to have SOME antibodies for covid if you get it. A subsequent infection (if possible) will almost certainly be less severe and be less likely to transmit to others. The question is how to quantify these factors.

[detritus]

Hopefully all those posting "its just the flu" feel pretty silly about now. Mortality rates of those hospitalised are around 10%. We don't know the number of infected people because some obviously don't get the full treatment, but safe to say, its not nice, its far, far more lethal and contagious than normal flu. Stopping lockdown will have two effects - firstly, you will cause another spike in infections and deaths, and secondly, your health services will be swamped, meaning you will probably (if you are ill enough) die without help at home.

Did China lie about infections and death rates? Probably. Does blaming china for any of this seem logical or in any way helpful? No. Some countries obviously have got it right - have a look at New Zealand for example.

On the plus side, the air is cleaner and I'm being paid to sit at home.

[NotASpiderSwarm]

Another concern for the "It's not just a flu" bucket: We don't know the long-term effects. There's definitely lung damage in a significant chunk of cases, is that temporary or permanent? When it screws with your sense of taste, is that physical or neurological, and if it's neurological, what does that do long-term? Kids get it and don't die, but what happens to their long-term development? The oldest case of this disease is 6 months, we don't know what Shingles-like effect it might have in the future.

[Aedilred]

I don't doubt that the disease is dangerous, albeit "mortality rates of those hospitalised" is always going to be skewed because only the most severe cases will be hospitalised.

I remain however firmly in the "cure may be worse than the disease" camp. A few weeks back in this thread I suggested the restrictive measures then being put in place (less severe than now) could easily become the "new normal" if we weren't careful, and was pooh-poohed. Now we're being told by government ministers that these measures will be "the new normal" for the foreseeable future.

Entire sectors have been virtually wiped out. Not just hospitality and retail, which will inevitably recover eventually, but the performing arts, education, conservation, which are much more fragile and dependent on skilled personnel, and may not. And that's relatively minor stuff, considering. If I, as an adult, lose a year of my life to lockdown, I'll be pretty damn annoyed, but it's not the end of the world. For a young child to be deprived of socialisation for that period is the sort of thing that leads to lifelong developmental defects, and it's in danger of happening to a whole generation - which apart from anything else is going to be exacerbated by the long-term impact of the lockdown on the education system.

The consequences on household income and employment have already been catastrophic. Even if we ended lockdown now, we're heading for an economic calamity on the scale of the Depression, and while economic damage might sound more attractive than human death, it has a tendency to lead to that: the human cost of the Depression was horrendous even if you don't count WW2.

I just can't see how lockdown is sustainable if we are to emerge from this with society being in a remotely recognisable and indeed acceptable shape. But equally we can't end lockdown without more people dying, which makes leaving it in place the less immediately controversial option. So where do we go?

Of course I'm concerned about the virus. Not for my sake, but for others': my mother in particular, who has a "special vulnerability" and is consequently even more isolated than most. But that isn't what keeps me awake at night.

[NotASpiderSwarm]

That depends on what you mean by "reasonable" and "acceptable". I don't want to get too far into it, but suffice to say, there are a lot of parts of the economy that were in trouble well before the virus hit, with severe resistance to fixing those problems. Now the economy and society are going through a crucible. We as a people decide what to save and what not to. It's on us whether this virus causes a radical reimagining of the economy and how it serves us, or a collapse.

[Kaptin Keen]

Hopefully all those posting "its just the flu" feel pretty silly about now.

Not at all, no.

And I hope you realise that 'mortality rate of those hospitalized' doesn't tell you the lethality of the disease. It tells you, very specifically, how many of those committed, die. So, as an example, in a country with too few ICU's - where only the worst cases can get treatment - that number may be much, much higher.

We have like ... 2000 committed, 400 dead. That sounds pretty bad. More than half of those have been commited in the previous 5 years, once or more, for serious or life threatening disease. But that's a side issue.

We have 400 dead out of what really is something like 200.000 infected.

Your statistic shows that it hits a few people really hard. But it does not show anything about the real mortality rate of the disease.

And no. I do not feel silly.

[factotum]

We have 400 dead out of what really is something like 200.000 infected.

Your statistic shows that it hits a few people really hard. But it does not show anything about the real mortality rate of the disease.

The simple fact is, though, you just made that 200,000 number up. It has no basis in fact--your health services haven't tested all those people to confirm they have the disease, you're just pretty much hoping that lots of people who haven't been tested actually have Covid-19 so you can make these pronouncements about how not-dangerous it is. You even said you had it yourself, but since you weren't tested to confirm that, you could have just had regular seasonal flu--Covid-19 being around doesn't prevent other flu variants existing.

[Kaptin Keen]

The simple fact is, though, you just made that 200,000 number up.

Nopes. Not so.

That is the result of independent research by James Lind Institute, and a volunteer organisation called The Research Panel.

Not something I made up.

It is, however, based on volunteer contributions, but hundreds of thousands of them. Like all the other numbers brought up in this thread, these ones are problematic for certain reasons - mainly because they are unconfirmed.

But like all the other numbers in the thread, they are useful if used right. People who think they have covid-19 may be right or wrong - but they're almost certainly right that they have some form of flu.

Now, the health authorities have a pretty good idea of what flu infections are out there, and in what proportions. So for a given samle size - hundreds of thousands in this case - you can actually calculate how many have the right virus, and how many have some other infection.

There are still margins for error - obviously. These volunteer contributions are made over the internet. Thus, people who don't use the internet are not represented at all. And the largest risk group is largely non-internet-users: The elderly.

So it's by no means perfect. But it works for the population at large.

Edit: Oh - I forgot. The research is good enough that it's been taken over by SSI, which is the State Serum Institute. So the numbers I use are actually official now. Even if you won't find those numbers in the press. They invariably report the most sensational, ie. worst and most badly skewed, numbers.

[Eldan]

Another concern for the "It's not just a flu" bucket: We don't know the long-term effects. There's definitely lung damage in a significant chunk of cases, is that temporary or permanent? When it screws with your sense of taste, is that physical or neurological, and if it's neurological, what does that do long-term? Kids get it and don't die, but what happens to their long-term development? The oldest case of this disease is 6 months, we don't know what Shingles-like effect it might have in the future.

It seems to be neurological. First tests seems to show the loss of taste is not an effect on the receptors in the tongue, but either in the brain or the nerves between tongue and brain.

[veti]

Edit: Oh - I forgot. The research is good enough that it's been taken over by SSI, which is the State Serum Institute. So the numbers I use are actually official now. Even if you won't find those numbers in the press. They invariably report the most sensational, ie. worst and most badly skewed, numbers.

That's not what the SSI's own website says. More importantly, it's simply too widely inconsistent with the experience of other countries that should be comparable, like here in New Zealand. Either our government is completely wrong in its present assessment - and I really hope it's not, because if that's the case we're headed for disaster - or, your official figures - and those of every other western country - are a reasonable approximation to reality.

[Kaptin Keen]

That's not what the SSI's own website says. More importantly, it's simply too widely inconsistent with the experience of other countries that should be comparable, like here in New Zealand. Either our government is completely wrong in its present assessment - and I really hope it's not, because if that's the case we're headed for disaster - or, your official figures - and those of every other western country - are a reasonable approximation to reality.

I can honestly say I don't care what SSI chose to put on their website. I promise you it's true. I can send you a letter from SSI, thanking me for my contribution to the research. I'm not going to, I'm just saying - I could.

And like I said: None of the numbers are right!*

Zero percent of the numbers are correct. Not mine, either. But you have to take into consideration that
- Every time you have a nice, controlled environment (like the Diamond Princess, or the survey of all blood donors), the number are very, very non-threatening
- And every other survey - in non-controlled environments - have tonnes and tonnes of unknowns ... primarily all those who are infected, and never noticed, or never went to a doctor

And I'm frankly not trying to correct anyone, I don't know any better than the rest of you. But you all seem totally willing to ignore any number or statistic that isn't threatening.

Seriously. When they surveyed all blood donors, the mortality rate was 0,13%. Nice, closed, controlled environment.

Want to ignore that? Fine. Be my guest.

*Edit: I guess I should say all of the numbers are right, but only within their given, narrow context. There are no right numbers in a broader sense. None that I've seen. You can extrapolate, but if you want to extrapolate, you really need to do it on the basis of a closed, controlled environment - then adjust for expected inaccuracies (such as who is on a cruise boat, and who isn't, or who's a blood donor, and who isn't).