Coronavirus - worldwide spread and control

Status
Not open for further replies.
anciendaze said:
It would be handy if you could provide a better reference than "someone on YouTube". YouTube even has flat-earthers and weirder people.
Click to expand...
Reference is irrelevant, these are obvious points already discussed on forums. In any case I do not recall his name, its some professor or something who has been advising on pandemic defence for a decade, in an interview. Canadian I think. The only reason I mentioned it is that this is being discussed, is obvious, but many are still not aware. When the obvious is ignored then society has a problem.
 
While trying to keep track of what's going on in Florida, I found an account that is unfortunately behind a paywall with a hefty cost. (I've been spending a great deal on subscriptions this year.) It gave Governor DeSantis credit for instituting a policy that patients hospitalized for COVID-19 had to test negative twice before they could be returned to long-term care. Why was this so at odds with my memory, which is admittedly fallible? The problem turned out to be timing.

He didn't recognize that we had an epidemic until March 1. It took a while after that before he caught on that most of the deaths were due to long-term care. He instituted a new policy on May 5. Even then he was refusing to name facilities which had problems. Here's a contemporary account.

We've had a lot of closing the barn door after the horse is gone. I expect to see more.
 
Jonathan Edwards said:
And why is the proportion of deaths to recorded cases so much lower in certain other countries? Presumably it is 1.5% in New Zealand because pretty much all cases got recorded and old people did not get the brunt of it. Why isn't the ration in the UK now somewhere near that?
Click to expand...

There is one major difference: New Zealand (and Australia until recently) had very little community transmission, which means there is a very different demographic distribution as to who was infected.

We cannot and should not assume that testing is somehow a magnitude of order lower than it should be without actually doing epidemiological modelling based on who is infected and try to estimate the various risk factors.

The mortality patterns we are seeing in the UK and elsewhere reflect the demographics of who is infected. A high rate of community transmission among the high-risk population can skew the numbers in the way that we are seeing.

Indeed for this (up to July) dataset, roughly 36% of the ~250,000 COVID cases in the UK were been in the 65+ age group, whereas this is only 18% of the overall population. This skew increases as age increases, with 21% of positive cases over the age of 80, who unfortunately have a very high risk of mortality. (With my short search, I could not find more granular data on UK population demographics to do a deeper analysis)

The Geneva Switzerland study showed that the 65+ demographic had ~40 times greater likelihood of mortality than the 50-64 age group.

While it is certain that testing rates have been lower than they should be, given the demographics of who is infected and subsequent mortality rates, it is quite unlikely that true infection rates are ten times the official number.
Perhaps the true "undertest" factor is in the ballpark of 1.5-3 times and this number, like the case mortality rate, will also decrease over time.

Note that while Italy and Spain have similar case mortality rates to the UK, other countries with different demographics of cases have much lower rates.
 

Attachments

Last edited:
Jonathan Edwards said:
I think it very unlikely that the method of attributing deaths will make a substantial difference. The proportion of people who die within a three month period (in general) is very small - maybe 0.3%. So just conceivably the figures might be 30% overestimates but that would not account for the sort of discrepancy we are seeing.
Click to expand...


Some are saying that the excess death rate is a more accurate way to look at the figures given recording problems. Deaths are recored as Covid-19 deaths if the person tested positive so no tests no record - which could have a big effect on reported numbers.

I wonder what type of varience would be expected in death rates in an outbreak given that different areas with different age/health profiles get hit and also issues such as care homes. If you look at the US figures https://www.worldometers.info/coronavirus/country/us/ the deaths per million in New Jersey/New York/Connecticut are very high (1779 per million in New Jersey compared to 667 for the UK). So other areas where there was a high infection rate also seem badly hit.
 
Jonathan Edwards said:
Something bothers me about the continuing figures for the UK, an indeed for all countries.

It seems reasonably certain that overall death rate from symptomatic Covid-19 infection is some where around 1%, or would be if all age groups were affected equally. I am being very approximate here.

Yet the proportion of recorded cases in the UK that die appears to be 15%. It is quite similar in some other European countries but in Portugal, for instance, is much much lower. I wonder why.

One thing might be that a high proportion of UK cases were in care homes, with high mortality. Another might be that case recording rates have been low. But one might expect by now for the high proportion of aged infirm cases to have dropped considerably and for case tracking to get within an order of magnitude of reality. If it isn't then case tracking is going to be useless.

But at the peak of the epidemic so far the proportion of deaths was 18% and it is still 11%. That is a major drop but even if the proportion of aged infirm cases had fallen to a half, which hopefully it has, that would account for it in toto. There seems to be little reason to think we are anywhere near recording the majority of symptomatic cases. Hopefully we are still only recording about one in ten, otherwise the death rate looks a lot more worrying than we have been led to believe.

And why is the proportion of deaths to recorded cases so much lower in certain other countries? Presumably it is 1.5% in New Zealand because pretty much all cases got recorded and old people did not get the brunt of it. Why isn't the ration in the UK now somewhere near that?
Click to expand...

Shouldn't try to respond to such a good analysis. If a very high proportion of UK cases were detected (by diagnostic test) then the % dying would presumably be much closer to that in Germany and other countries with developed health care systems. Suggests that we aren't testing enough people with symptoms; which also suggests we are not effectively quarantining people who are infectious and tracing and quarantining positive contacts.
 
Adrian said:
Some are saying that the excess death rate is a more accurate way to look at the figures given recording problems. Deaths are recored as Covid-19 deaths if the person tested positive so no tests no record - which could have a big effect on reported numbers.

I wonder what type of varience would be expected in death rates in an outbreak given that different areas with different age/health profiles get hit and also issues such as care homes. If you look at the US figures https://www.worldometers.info/coronavirus/country/us/ the deaths per million in New Jersey/New York/Connecticut are very high (1779 per million in New Jersey compared to 667 for the UK). So other areas where there was a high infection rate also seem badly hit.
Click to expand...

Yes, I'd heard that excess deaths may be the best way to estimate the number of deaths due to coronavirus.

In Germany they identified that, after the controlled the initial outbreak, the majority(?) of new cases were occurring in the health care system and then focused resources on that. So the UK data seems pretty poor i.e. we don't have an indication whether those who are infected picked up the virus within the health care system.
 
Jaybee00 said:


Surprise!
Click to expand...


A very misleading headline. The article says the UK has world leading biomedical research and credits the research that showed dexamethasone is a useful treatment, and the vaccine development which it claims is ahead of the rest. So it's not about the control measures the UK government has used, or the fact that we have the highest per million death rate, it's just picking out a couple of useful or potentially useful bits of biomedical research.
 
A bit of a follow-up to my previous post:
mango said:
More "optimistic" speculations about herd immunity by the Swedish Public Health Agency...
Omni said:
Four out of ten Stockholmers may have viral immunity

According to the Swedish Public Health Agency, around 40 percent of Stockholmers may be immune to covid-19, reports DN.

Johan Carlson, Director General of the Swedish Public Health Agency, believes that between 17.5 and 20 percent have antibodies, and that people with T-cell immunity can be just as many. [...]
Click to expand...
Click to expand...
Some numbers from actual tests:
Karolinska University Hospital Google Translate said:
during weeks 25 - 29, more than 223,000 people in the Stockholm Region were tested. The proportion of positives was 16.7%
Click to expand...
Source: Karolinska Universitetssjukhuset: Den storskaliga antikroppstestningen i Region Stockholm fortsätter i oförminskad styrka med över 223 000 tester gjorda sedan det öppnades för allmänheten i mitten på juni

Google Translate, English

DN Google Translate said:
The company Werlab's antibody tests show that 14.5 percent of those tested in Stockholm in the last three weeks have antibodies against covid-19. [...]

During a three-week period, Werlab took samples of 83,000 people. Of these, 80,000 were made in the Stockholm Region and the rest in Gothenburg and Malmö. [...]

Those who have been tested do not constitute a representative sample of the population, they themselves have chosen to go and test themselves.

- There are a lot of healthcare staff, many in risk groups, but also anyone
Click to expand...
Source: DN: Werlabs testade 80.000 i Stockholm: 14,5 procent har antikroppar

Google Translate, English

In related news, the Swedish Public Health Agency recently published new guidelines on immunity for local and regional governments.
 
That percentage from the Karolinska University Hospital study is almost exactly what antibody tests in Florida show. We managed to achieve the worst of both worlds, a shutdown and a run-away pandemic that still falls far short of any kind of "herd immunity".

Today we found out about another 10,249 new cases, and 173 deaths. Since 38 of those deaths were in Orange County, I intend to be even more reclusive than usual.
 
mango said:
A bit of a follow-up to my previous post:

Some numbers from actual tests:

Source: Karolinska Universitetssjukhuset: Den storskaliga antikroppstestningen i Region Stockholm fortsätter i oförminskad styrka med över 223 000 tester gjorda sedan det öppnades för allmänheten i mitten på juni
Click to expand...

Assuming 0% false positives and 10% false negatives, this still means that the proportion of the population who has been infected must be between 4%-16.7%.
Each % of false positives on the test subtracts from the upper bound. Similarly, since people who are infected are more likely to be tested, this bias will reduce the true number from the upper estimate. The true proportion of prior infected individuals may well be be less than 10%.
 
There's a reason I used antibody tests above. People with symptoms or contact with an infected individual are typically given a nucleic acid test, so that test has a bias toward higher rates of infection. (It also has a delay of a week or more, particularly during a surge.) People getting the antibody test are concerned, but don't have strong medical reasons to be tested. At this point most people in Florida should be concerned.

The false negative rate for antibody tests could be 20%. There have been some efforts at random sampling, but they are too sporadic to estimate infection rates outside hot spots. I'll admit I don't know the positivity rate for the state of Florida, but I'm certain it is way too low to stop an epidemic.
 
Since we now understand that the virus is more airborne than previously thought, I'd think more attention would be paid to the quality and circulation of inside air in buildings. At the very least, HVAC systems should be set to bring in some outside air instead of just circulating all inside air. Maybe most business's HVACs are already set to do this, but it's a good thing to check. It would also help to have windows that open. I don't see as much emphasis on these issues as there should be.
 
IM's employer is now starting to "encourage" much of the work force back to work.

Apparently HR have stated that workers currently working from home can't be forced to come back in so they are being asked.

Difficult situation - some are fed up trying to work at home, some prefer it. However, a lot of them would prefer to be told they have to go back as they're worried if it might make a difference about the level of support they receive from the company i.e. they might be treated better if they came in because they were told they had to.

Some of them do quite specialized stuff where there are a handful of people who can perform certain mission critical jobs and nobody seems to be considering what would happen if all 4 or 5 of these people who hot desk in the same small area get sick at the same time. All it takes is one of them to become sick and because their shifts overlap....

If all of 'em were off for a week or two, there might not be a job to go back to.

Fun times. I'm sure lots of employees are facing the same situation around the world.
 
Status
Not open for further replies.
Back
Top Bottom