Coronavirus - worldwide spread and control

[Jonathan Edwards]

So honestly, the lockdown is easing, isn’t it, without actually officially easing. It’s not just about “essentials” anymore. Is this the right time to be doing this?

I suspect this is OK. Almost certainly there are aspects of 'lockdown' that are unnecessary. Click and collect in my experience works very well, with people tending to opt for 4 metre distancing if anything. I like it when staff wear gloves and masks. Most seem to wear gloves at least.

 

[Jonathan Edwards]

Seems to me countries you list as orders of magnitude better in their response are the outliers. China's figures are dubious at least; S Korea had the benefit of previous experience in dealing with SARS (1) and according to some accounts were lucky enough to have been 'game planning' an ideal response at a time when the WHO still considered the risk of a pandemic to be low. New Zealand of course has the benefits of being an isolated island with a tiny population and who had the time to witness what hadn't worked elsewhere.

Those that have done well may well be in the minority but I don't see that make4s any difference to UK being bottom of league (with a lot of others). The graph on excess deaths is relevant and not surprising to me,. Basically health care in the UK has come to halt.

You cannot possibly make China's figures other than excellent. There is no possible reason why they should have publicised the events in Wuhan and then covered up twenty times that. I have seen no evidence that Chinese figures are under-reported other than was always assumed from ascertainment problems during the peak. The claims of under-reporting are actually less than I had assumed.

South Korea had SARS but SARS spread elsewhere and was documented by all academic communities. Everyone knew the risk. We had the same benefit of previous experience. There was nothing lucky about their game planning if, as has been reported the UK also did the game planning and then didn't bother to implement the required actions.

Who cares what the WHO thought about 'risk of pandemic being low' whatever that means. It was barn door obvious to me in late January that flights from China to Europe should have been closed immediately. I was appalled to fin myself in the same passport queue as large numbers of Chinese on flights from Beijing wearing masks, presumably because they thought there was a risk of carrying virus, with no evidence that airport staff had any interest at all. I am an immunologist but my wife is a radiologist with no specialist training in infectious disease. It was just as obvious to her that this was negligent.

And of course the UK has the same advantage of New Zealand in being an island. And size is irrelevant. What matters is population density and there is not much difference between London and NZ cities I suspect. The difference is that they did shut down flights. They took the virus seriously rather than being happy to shake hands and have Spaniards over for football matches.

My knowledge of what to do in pandemics is largely based on Melvin Ramsey's excellent book on infectious diseases from the 1960s, stories from my virologist mother, and listening to the news. That is all you need to know that the UK government screwed up at pretty much every point. The key principle in medicine is that you take into account the worst reasonably likely outcome, not the most likely outcome. There is absolutely no way that policy has followed that golden rule.

If you follow back the root causes of all this they go back to axing service structures in the first decade of this century. Fever hospitals were closed. The PHLS was shut down. And so on.

 

[lunarainbows]

I suspect this is OK. Almost certainly there are aspects of 'lockdown' that are unnecessary. Click and collect in my experience works very well, with people tending to opt for 4 metre distancing if anything. I like it when staff wear gloves and masks. Most seem to wear gloves at least.

I’m not sure it’s ok, I can’t understand it at this moment in time, when it’s not essential. Especially if masks aren’t mandatory yet. If one person is infectious, and doesn’t wear masks and/or gloves, there is potential for it to spread. Especially in stores where the population it serves is larger. And when we don’t even have testing to know who is infectious or not or contact tracing either. Perhaps if there was extensive testing, tracing, masks were mandatory - I could understand it. (There are so many shops that would potentially open now for click & collect - potentially employing a lot more people and therefore more movement of people at this time).

It also gives a confusing message - so you go out once for exercise, once more for food and medicines, and then what? Another time for “non essential” items? So it’s not really essential things only anymore? Yet that’s the message we’ve been given. New Zealand is only opening takeaways & online ordering as the second phase of their lockdown. We are still apparently in the first phase, nowhere near close to coming out of lockdown and are already doing it. Essential shops for groceries and medicines I can understand; these (and Greggs), I can’t, not when things are already difficult in London & England & the U.K.

When the Chinese delegation of Red Cross arrived in Italy they were very critical of the amount of people still on public transport & going to work - we are doing the same (if not worse).

(Edited and to add sentences in first & last paragraph).

 

[Jonathan Edwards]

Sadly, it looks from the new piece in the guardian on the contact App https://www.theguardian.com/politic...-plans-an-nhs-app-and-an-army-of-health-staff

that there are still no reliable plans for contact tracing. I am pretty sure that the information from the app will be unusable - a muddle of flagging all sorts of possible contacts most of which being of no relevance with the added worry of loss of confidentiality. Contact tracing requires person to person analysis of each real case by someone intelligent. I think they will need about 30,000 such people. There is no sign that this is understood by the people making decisions.

 

[Snow Leopard]

What the real rate of infection is currently unknown, the official numbers are done for practical purpose, and are not in any sense representative. In Gangelt, Germany, past infection has been estimated to be 15 or even 20%. Death rate was - so far - 0.37% (with 15%).

Many of these claims of high "past infection" rates are bullshit based on underestimating the specificity of the test. The study in Ganglet, Germany is the case-in-point where they used a serological test that likely has around 90% specificity - which would mean past + current infection = 4%.

Sweden's death rate per tested cases is also high because the virus is passing through aged (and disability) care homes in an uncontrolled manner. I don't think that is something to be proud of as a society, I dare say this will severely damage Sweden's reputation as a socially responsible democracy.

 

[spinoza577]

@Snow Leopard Two different sources of covid-19 deaths in Sweden, the first (official) one may be the more accurate one,
but will be updated again (for the last days).
I don´t think that it is likely that the reported numbers in a country like Sweden are wrong.

I have a question: Who says, that exponential growth does empirically take place?

Is this the case with common colds?

(Some corona viruses are a possible pathogen for common colds. Admittedly this virus is not only a corona virus but also new.)



I have a second question: How reliable is the testing of acute cases?

Isn´t it nonsense to report numbers of positive cases without the number of tests done (or negative results).

 

[Jonathan Edwards]

I’m not sure it’s ok, I can’t understand it at this moment in time, when it’s not essential. Especially if masks aren’t mandatory yet. If one person is infectious, and doesn’t wear masks and/or gloves, there is potential for it to spread.

I very much agree with you that in some ways restrictions should be tightened rather than loosened. Everyone in a public space should be wearing a mask. Public transport should probably not be used. Etc..\

But my experience of click and collect has been a sort of parody of best practice. Everyone is so aware of the distancing that they double it. It is far safer than supermarket shopping. There is no real chance of anyone infecting another customer. You collect from an isolated outside point in a given time window so there are no queues of more than about three people anyway. In a way I see it as training people how we might return to a more normal way of life in a safe way and as a counter to the argument that rules are unnecessary. I don't go for the behavioural science nonsense about predicting rule fatigue but I do see rules that are overtly unnecessary as being a likely source of loss of trust and cheating.

Edit: Another point I would concede is that I am holed up in a rural area where most shops are out of town retail parks. I don't think click and collect would work well in London. Another illustration of the point that although strategy should be centralised implementation should be much more determined by local needs and resources.

 

[spinoza577]

...[Graphs]
I did see a reference to this in a sky news article which I posted on this thread earlier, that the excess deaths in other European countries (Spain, Italy, France), although they went high, they then came back to normal levels quite quickly. Whereas in the U.K. - 2 weeks later our excess deaths are still really high.

Why? What on Earth is going on? This is the Z score mortality measure. I don’t know exactly what that means and would appreciate some input. (Edit; I remember now when I looked it up, the number of standard deviations from the mean!). Anyway, why is it so high?

Thank you, very interesting.

Are the measures in the different parts of the UK the same?

(As all Non-England parts already went down.)


For a general assessment, of course, the graphs needed to be compared with other excess mortalities,
(as steeply achieved hights usually happen with the flue too),
and of course with a somewhat larger time frame.


PS: Parameter for an explanation for England might be?

• density of population
• transmitted virus loads
• carelessness at the begin of the epidemic
• genetics
• pollution

 

[Snow Leopard]

I have a question: Who says, that exponential growth does empirically take place?

I don't understand the question. Empirically, uncontrolled spread will lead to exponential growth, but if spread leads to death, people will modify their behaviour (even in the absence of government coercion) to minimise risk at some point. Which is what I have been saying since February.

The numbers I posted show that with static/flat daily infection rates, even at five times what has been reported in the last few weeks, herd immunity will never be achieved in Sweden, which means the virus will not be eliminated and will instead continue to infect people, causing excess deaths compared to the Australia/NZ strategy of eliminating the virus and then reducing most restrictions (except quarantine on international travellers until an effective vaccine is found).

Any country, which merely "flattens the curve" of infection rates, will mean that the infection will continue to spread until a vaccine is found. While the infection spreads, even if the government does not enforce social distancing and other measures, the economy (and public health system) will suffer for years as many people will still modify their behaviour to avoid infections.

Also remember that there are common post-viral consequences, including Guillain Barre Syndrome, ME/CFS and lung damage, not just death.

(Some corona viruses are a possible pathogen for common flues. Admittedly this virus is not only a corona virus but also new.)

In English speaking countries, we call them colds or respiratory infections, only Influenza linked infections are called Flus.

I have a second question: How reliable is the testing of acute cases?

Accurate enough if you are testing a highly selected population showing key symptoms (fever, anosmia, lung abnormalities on a CT scan). The reliability is poor if testing a random community sample.

80% sensitivity, 99.5%+ specificity (RT-PCR)
80-95% sensitivity, 90% specificity (serology, aka tests for antibodies).

Isn´t it nonsense to report numbers of positive cases without the number of tests done (or negative results).

Yes, and those numbers are reported in most countries.

Here are is the stats of my state (roughly a sixth of the population of Sweden) in the last week, showing that elimination is a realistic goal:

 

[BurnA]

I think they will need about 30,000 such people. There is no sign that this is understood by the people making decisions.

Tomas Pueyo's latest article talks a lot about contact tracing.

 

[Amw66]

But the number of corona-deaths reflect only people who have died AND have been tested positive.

To estimate if they really died from corona or if they would have died anyway can only be decided from excess mortality within some time frame, say 6 months or a year or so. The current number of "corona-deaths" doesn´t tell much.

Therefore it is also no argument that there are more dead people who could have been tested positive, if they only had been tested.


Then the cases listed, without the number of tests done this number is completely clueless. If you test a lot of people you get a lot of positives. If you test only few people you get only a few positives.

Even for one country the number of cases does not appear to serve for a coherent comparison in the course of time, because the number of testing went up. There is bad communication and even thinking at work.


So I have to say that I disagree with your assessment, though I admit that the situation is unclear. I think the Swedish numbers are not of concern (and deaths are rapidly declining!), but the numbers of Italy, Spain, UK and US may be, but even this might be not clear.

In Spain there are now about 20.000 corona deaths, and we should expect about 600.000 deaths every year, I very roughly guess. This makes 200.000 deaths for every four months time.

What I am concerned about is,

• that there have happened other deaths (non-corona deaths), because people didn´t went into the hospitals (being afraid or whatever) or didn´t get the help they needed.
• that the measures themselves cause positively new illnesses (including in children, also from there parents, which will be a live long experience).
• that the social costs will affect the health care in the future negatively, causing illnesses and deaths in the future.

These points are part of the equation (and the first two points contribute to any excess mortality now). Therefore it is irresponsible not to communicate the preliminarity of the current known numbers. In my view Sweden behaves especially wise, though the situation might not be comparable with the one in Spain, for so far unknown reasons.

The FT article on ex essay deaths suggests Sweden only marginally above average rate.

Screen print of graphs bow

 

[Marco]

What matters is population density and there is not much difference between London and NZ cities I suspect.

Not really - more like Bournemouth :

https://en.wikipedia.org/wiki/List_of_English_districts_by_population_density

 

[spinoza577]

I don't understand the question. Empirically, uncontrolled spread will lead to exponential growth, ...

Are there empirical data that this happens with all infections?
E.g. for common colds?

Or do common colds after a while stop to spread (and may turn up later, usually in autumns)?


An example: Newtons law of gravity applies a certain mathematical relationship.

By no means does this law implicate that the empirical world does behave so.
Instead it needs to be tested. (Therefore it is an empirical science, not an a priori one.)

I.e., to say that any empirical matter will behave like any mathematical relationship says, is wrong.
Instead, laws needed to be tested, and indeed, today Newton´s law is replaced by Einstein´s theory.

So my question was, is this the case with infections, i.e. that they spread exponentially until everyone is infected?
Is this the case for common colds?
Is this the case for the established corona viruses?
Or does a saturation in a given time frame occur where the curve stops to go up steeply (not logically, but empirically)?

...Any country, which merely "flattens the curve" of infection rates, will mean that the infection will continue to spread until a vaccine is found. While the infection spreads, even if the government does not enforce social distancing and other measures, the economy (and public health system) will suffer for years as many people will still modify their behaviour to avoid infections.

(I disagree with your assessment in regard of our current pandemic.)

In English speaking countries, we call them colds or respiratory infections, only Influenza linked infections are called Flus.

Thank you, I have corrected it. (I even knew, but my ME brain gave "spanish flu" and then "other, non-influenza flus" ...)

 

[FMMM1]

Those that have done well may well be in the minority but I don't see that make4s any difference to UK being bottom of league (with a lot of others). The graph on excess deaths is relevant and not surprising to me,. Basically health care in the UK has come to halt.

You cannot possibly make China's figures other than excellent. There is no possible reason why they should have publicised the events in Wuhan and then covered up twenty times that. I have seen no evidence that Chinese figures are under-reported other than was always assumed from ascertainment problems during the peak. The claims of under-reporting are actually less than I had assumed.

South Korea had SARS but SARS spread elsewhere and was documented by all academic communities. Everyone knew the risk. We had the same benefit of previous experience. There was nothing lucky about their game planning if, as has been reported the UK also did the game planning and then didn't bother to implement the required actions.

Who cares what the WHO thought about 'risk of pandemic being low' whatever that means. It was barn door obvious to me in late January that flights from China to Europe should have been closed immediately. I was appalled to fin myself in the same passport queue as large numbers of Chinese on flights from Beijing wearing masks, presumably because they thought there was a risk of carrying virus, with no evidence that airport staff had any interest at all. I am an immunologist but my wife is a radiologist with no specialist training in infectious disease. It was just as obvious to her that this was negligent.

And of course the UK has the same advantage of New Zealand in being an island. And size is irrelevant. What matters is population density and there is not much difference between London and NZ cities I suspect. The difference is that they did shut down flights. They took the virus seriously rather than being happy to shake hands and have Spaniards over for football matches.

My knowledge of what to do in pandemics is largely based on Melvin Ramsey's excellent book on infectious diseases from the 1960s, stories from my virologist mother, and listening to the news. That is all you need to know that the UK government screwed up at pretty much every point. The key principle in medicine is that you take into account the worst reasonably likely outcome, not the most likely outcome. There is absolutely no way that policy has followed that golden rule.

If you follow back the root causes of all this they go back to axing service structures in the first decade of this century. Fever hospitals were closed. The PHLS was shut down. And so on.

I think part of the problem here may have been that the decision makers are clones (I didn't mistype); most of them seem to have studied "Philosophy, Politics and Economics (PPE) at Oxford" --- possibly Boris may not have --- did he just study classics? Also, they didn't have the right advisors; they needed people who had a background in pandemics/infectious diseases. It seems that there were plenty of good advisors e.g. here's the line up from the CovidReport
- Prof Anthony Costello
- Prof Allyson Pollock
- Dr Bharat Pankhania
- Dr Rosalyn Moran
https://twitter.com/hashtag/covidreport?lang=en

Boris/Hancock are responsible for who is in the room (i.e. advisors)/the decisions which are made ---

 

[spinoza577]

Accurate enough if you are testing a highly selected population showing key symptoms (fever, anosmia, lung abnormalities on a CT scan). The reliability is poor if testing a random community sample.

80% sensitivity, 99.5%+ specificity (RT-PCR)
80-95% sensitivity, 90% specificity (serology, aka tests for antibodies).



Yes, and those numbers are reported in most countries.View attachment 10789

This last one was my question, not how accurate the test itself is. The test is considered to be accurate, which probably was good luck (which the inventer Drosten had here).

So an accurate test is applied to a certain number of people, but for practical reasons (to stop spread, monitoring).

This makes that the testing is hardly representative.


In addition, TV channels, newspapers, and wikipedia just give the absolute number. In the reports of the German CDC it took a while until the total number of tests even was known! But now you don´t see the number of total tests easily, no, you have to look for it. The official site makes you see easily the absolute numbers, and only these are reported in press conferences (or a journalist asks for them).

I am lacking words for this stupidity, whatever the appropriate result would be, higher or lower percentage of total infections now and in the past.
I want to be provided with the data as much reliable as possible in this difficult situation.
Given that the costs may be high in all regards (this disease now, this disease in the future, other disease now influenced by measures, other diseases developing now, a well enough supported health care system in the future), this should - naturally - also be important for making decisions.

 

[JaneL]

Coronavirus: Switzerland says young children can hug grandparents

Swiss authorities say it is now safe for children under the age of 10 to hug their grandparents, in a revision to official advice on coronavirus.
The health ministry's infectious diseases chief Daniel Koch said scientists had concluded that young children did not transmit the virus.

Dr Koch told a news conference this week that the original advice to keep distance between children and their grandparents was made when less was known about how the coronavirus was transmitted.
"Young children are not infected and do not transmit the virus," he said. "They just don't have the receptors to catch the disease."

https://www.bbc.co.uk/news/world-europe-52470838

 

[JaneL]

The FT article on ex essay deaths suggests Sweden only marginally above average rate.

The FT data on excess deaths is a few weeks old so is now out of date. The more recent data suggests Sweden is doing much worse than that unfortunately. If you scroll down to the map of z-score by country on the following webpage you’ll see what I mean:

https://www.euromomo.eu/graphs-and-maps#excess-mortality

The Euromomo website is also the source of the graph posted above by @lunarainbows showing that in the past couple of weeks the UK has zoomed ahead of Italy and Spain and now sadly has the highest rate of excess deaths in Europe.

Edit: typo

 

[Snow Leopard]

Are there empirical data that this happens with all infections?
E.g. for common colds?

I still don't understand why you are asking this question. No one is disputing that exponential spread until "herd immunity" is achieved is the simplistic case that doesn't account for complex human behaviour, as humans will change their behaviour to avoid getting sick and dying.

Or do common colds after a while stop to spread (and may turn up later, usually in autumns)?

With the advent of international travel, viruses don't really stop spreading, they just spread to different geographical areas. If they truly stopped, they would not come back in later seasons.
But over time, there can also be significant genetic drift (usually over the course of years, rather than months) due to selection factors - the virus evolves tries to reinfect those who had previously developed immunity, leading to a dance between the genetic structure of the virus, immune system defences and human behaviour (viruses that are mild are spread much more easily than viruses that kill, because most people will try to avoid dying)

It is notable to compare common human infections to infections of plants and trees. Plants don't have the luxury of moving to a different place to avoid infection (or environmental stress), so they have had to evolve many types of defences - which is why most plants have more genes than humans.

(I disagree with your assessment in regard of our current pandemic.)

Why do you disagree?
Let me put it this way, what you are seeing in Sweden right now is going to continue until a successful vaccine/treatment is found. This could be 2 years or more. Mathematically, herd immunity still won't have been achieved in 2 years unless far more people are infected each day, which in turn leads to higher deaths and chronic illnesses. More people will end up on disability pensions, costing the government more money. During that time, the Swedish economy will stagnate as people will be fearful as to when the pandemic will end.

We are playing out the alternative where I live and it will be far easier for our economy to recover quickly once the virus is eliminated.

In addition, TV channels, newspapers, and wikipedia just give the absolute number. In the reports of the German CDC it took a while until the total number of tests even was known! But now you don´t see the number of total tests easily, no, you have to look for it. The official site makes you see easily the absolute numbers, and only these are reported in press conferences (or a journalist asks for them).

As a general rule, science reporting in the media always leaves out essential details and fail to ask the most important questions (methodology). It is frustrating how badly the science of COVID-19 has been reported in the media, but it is not unexpected.

Test rates have been reported by most Nations since they started testing. Summary statistics are easily available to those who seek it:
https://www.worldometers.info/coronavirus/#countries

 

[Snow Leopard]

The FT article on ex essay deaths suggests Sweden only marginally above average rate.

Screen print of graphs bow

It's shocking how much worse things look since that graph was made. The key difference is other nations are using stronger measures to drop the active rate of infection, hence deaths are going to slow in other nations, but continue at a high rate in Sweden until Sweden has the worst death rate in Europe. Sweden may also be under-counting deaths due to lower testing rates.

In countries that are accurately detecting and reporting infections and related deaths, a death rate of <=5/million population is considered effective. Sweden currently has 244 deaths/million according to worldometers.info

 

[lycaena]