/ We are not going to follow Italy

Please Register as a New User in order to reply to this topic.
ClimberEd 25 Mar 2020

(necessarily) 

There is an awful lot of shouting (forum shouting) that we are following Italy, what's happening in Italy is happening here etc.

With the idea that any other outcome is not possible. 

And any suggestion of a different outcome is met with 'that's not what is happening in Italy'

Can I suggest that this -may not- be the case. If you look at the worldometer there are very different numbers for different countries. (check out Japan for example). There will be other factors involved, some social, some connected with healthcare, and potentially some with genetics. Identification (asymptomatic, ill, or death) will also change the data.

So please stop blindly saying 'we are following Italy'. We may follow Italy's trajectory, but it is not a given..

Report
bouldery bits 25 Mar 2020
In reply to ClimberEd:

We are following Italy

Report
Toerag 25 Mar 2020
In reply to ClimberEd:

Looks like we are to me.....    http://nrg.cs.ucl.ac.uk/mjh/covid19/

It's plainly obvious that the only European nations not following Italy are the ones locking down or testing and tracing harder and faster than Italy did.  Expect to see an explosion of cases in a week's time when all the people who went out on the 'last night of the pubs' start showing symptoms.  Due to the lack of testing the country is running blind, and the general population aren't as worried as they would be if case numbers were higher. Because they're not as worried they were ignoring the social distancing rules. Until they start seeing deaths in their towns they think it doesn't affect them, and the fact that they're infective before being symptomatic, or aren't symptomatic at all plus it takes time to get tested (plus the infectivity of the virus) means the virus gets out of control before people realise it has.

Post edited at 07:15
Report
wintertree 25 Mar 2020
In reply to ClimberEd:

Oh dear.

Even the government are now saying we appear to be following Italy with a difference of about 14 days.

I’ve not seen anybody saying we are bound to continue along the same path.

I have seen people note why Italy is different to us to our advantage (more old people, more inter-generational households, more physical contact between friends) and  where there are similarities (generally slow and reactive not proactive response, not taking things seriously in the critical early part).

I’ll note various posters including myself were making the case that the UK needed to take more drastic action   if we wanted to avoid serious consequences.  Well, now we have the greatest ever curb on individual freedom ever seen in the UK and we are still lagging Italy by about 14 days.

> check out Japan for example

I guess you missed the various discussions about how the policy and reaction in the UK is much more like that in Italy than that in Japan?  

> Identification (asymptomatic, ill, or death) will also change the data

Again, this has been discussed.  The detection rate is not so good a comparator between countries but the death rate is.  Ours is currently 14 days lag from Italy’s.

Post edited at 07:26
Report
veteye 25 Mar 2020
In reply to ClimberEd:

Good.

I hope that you are right. 

It could be that we do follow them, but we may not. 

An unknown is how quickly we can adopt assays for antibody status of individuals. Are these all or nothing type tests, or do they give some idea of antibody titres/levels of response?

Rob

Report
ClimberEd 25 Mar 2020
In reply to bouldery bits:

> We are following Italy

At the moment. And we may. But, as I said, not necessarily. 

I don't think it's a good straw man to say that we WILL (sorry I don't know how to do italics) follow Italy.

Report
ClimberEd 25 Mar 2020
In reply to wintertree:

> Oh dear.

> I’ve not seen anybody saying we are bound to continue along the same path.

>

That's what everyone seems to say, in response to everything 'look what's happening in Italy'. What's happening in Italy is going to happen here. etc.

Report
wintertree 25 Mar 2020
In reply to ClimberEd:

> That's what everyone seems to say, in response to everything 'look what's happening in Italy'. What's happening in Italy is going to happen here. etc.

No.  What people are saying - exhaustively is “we are currently following Italy and our policy and actions are like Italy’s so we are likely to continue to follow Italy”

I talked about these reasons and your “but Japan” argument in my post.  You apparently chose to ignore all of this.  

Why don’t you try reading my post again?  I went in to a lot of explanations specifically for you.  You’ve just cherry picked one line and reverted to the stance from your OP.

Report
ian caton 25 Mar 2020
In reply to ClimberEd:

Doesn't matter if we do or don't academically. What matters is seeing what could happen and being prepared for it or trying from others experience to do better. 

It could of course be worse than Italy.

Interesting that the mayor of bergamo, the epicentre of the Italian outbreak, brought his daughters home from the UK because he thinks them safer there.

Report
Neil Williams 25 Mar 2020
In reply to wintertree:

> I’ll note various posters including myself were making the case that the UK needed to take more drastic action   if we wanted to avoid serious consequences.  Well, now we have the greatest ever curb on individual freedom ever seen in the UK and we are still lagging Italy by about 14 days.

What is really important to know is that those measures will have a delayed effect.  The incubation period is about a week, and the time to death typically 2-3 weeks.  This means that we are only just now seeing the effect of the first "please work from home" last Monday, and won't see the effect of the pub closures for a few days, nor the effect of the tighter lockdown for another week or so, and much longer on the death figures.

Report
wintertree 25 Mar 2020
In reply to Neil Williams:

> What is really important to know is that those measures will have a delayed effect. 

Agreed.  5-7 days seems reasonable.   I also expect the effect of Monday’s changes on detection rates to be masked by a massive rise in testing coming soon - 7 qPCR machines were moved by the Navy from Newcastle Uni to a testing centre in Milton Keynes a couple of days ago and are now set up for testing.  I imagine more are coming in from elsewhere.  The death rate takes over as the accurate reporter and that’s got more lag.

So, people, keep the faith - even if there’s an accelerating rate of detected cases for the next two weeks, lockdown is making things less bad than they would otherwise be.

Report
ClimberEd 25 Mar 2020
In reply to wintertree:

> > That's what everyone seems to say, in response to everything 'look what's happening in Italy'. What's happening in Italy is going to happen here. etc.

> No.  What people are saying - exhaustively is “we are currently following Italy and our policy and actions are like Italy’s so we are likely to continue to follow Italy”

It didn't chose to ignore the rest of your post. Forgive me for me not fully extrapolating - yes people are saying we are currently following Italy and our policy and actions are like Italy's so we are likely to continue to follow Italy. 

I'm say I don't believe this is necessarily the case. I'm not saying it is, or isn't correct, but I am saying that people who are presenting it as a fait accomplis are wrong. I believe there are too many interlinked factors to just say if we follow Italy's lock down time frame we will have the same out come as Italy

Report
StuPoo2 25 Mar 2020
In reply to wintertree:

> No.  What people are saying - exhaustively is “we are currently following Italy and our policy and actions are like Italy’s so we are likely to continue to follow Italy”

Morning wintertree!

Help me out ... i've not been keeping up (moving too fast )

Yesterday I was looking at a chart of Y: Reported infections per 1 million people vs X: # of days after total reported cases reaches 100.  i.e. objective was to baseline countries at the same starting point ... then look at their curve and asses how quickly the growth was thereafter.

Page 2:  https://www.jpmorgan.com/jpmpdf/1320748295335.pdf

In this chart ... after 100 cases/million Switzerland took off like a rocket, then Spain, then Italy, then Germany, then France ... then UK.  Agree 100% ... UK is ~14 days behind Italy ... but in this chart UK growth after 100/million hasn't been as aggressive as either Switzerland (which we really should be worried about), Spain, Italy, Germany or France.

Can you bring me up to date on why we think the UK is so similar to Italy?  Or you are saying our detection rates are off and we are in fact on the Italy trajectory and will catch up once we start testing more?

Cheers

Report
wintertree 25 Mar 2020
In reply to StuPoo2:

> Can you bring me up to date on why we think the UK is so similar to Italy? 

I don’t think we are “so similar”.   The link Toerag posted is by far the best visualisation I have seen.  Our detection rates are slowing down now - just like Italy’s did a couple of weeks ago - social awareness, increasing control measures kicking in etc.

Our exponential growth curves still align well with a time lag of about 14 days.  It’s now better alignment for 15 days, so I’m the last week we’ve bought ourselves about a day - going off detection rates.

> Or you are saying our detection rates are off and we are in fact on the Italy trajectory and will catch up once we start testing more?

Detection rates are not the most useful as they under-detect by different amounts in different counties and the testing fraction can change.  Death rates are a better readout (other than old people being left unreported to die in care homes as in Italy) until healthcare is overwhelmed as in Italy, raising rates.

The UK will probably have a large rise in detected cases soon representing more testing not more infection.

It does look like the detection and death rates are lowering slightly for the UK for a 14/15 day lag compared to Italy, over the last 3 days.  If that continues to be the case that’s great.  Not nearly great enough. 

I’ve never said we are bound to continue following their trajectory, and have repeatedly said the only way we are likely to break out of their track is through taking more extreme measures and doing so sooner with respect to that lag.

> UK is ~14 days behind Italy ... but in this chart UK growth after 100/million hasn't been as aggressive as either Switzerland (which we really should be worried about), Spain, Italy, Germany or France.

You seem to be misreading the chart severely.  It is from 4 days ago before the UK passed 100/million so there is no “UK growth after 100/million”.

My own wishful thinking is that this is going to slow down as it runs in to all the people infected/immune from a mid January UK wave of infections spreading out from international students and businesspeople returning from Christmas and that puts herd immunity ahead of the infection.  Highly speculative, something that’s been discussed for a long time on UKC and now there’s the oxford study discussed elsewhere to give me hope.  It just doesn’t seem credible however

Post edited at 09:12
Report
wintertree 25 Mar 2020
In reply to ClimberEd:

> but I am saying that people who are presenting it as a fait accomplis are wrong.

Can you actually find and link a post where it’s presented as a “fair accomplishment”?  Several people have been going out of their way to interpret posts as that despite clear explanations otherwise.

The point is if we do not want to be like Italy we have to do other than Italy did.  Simple point, clearly made throughout and the point remains the same.  If it’d been heeded nationally and individually a few weeks ago we could have perhaps been more like Japan.

Report
StuPoo2 25 Mar 2020
In reply to wintertree:

Hey thanks for getting back to me winterfree.

> Our exponential growth curves still align well with a time lag of about 14 days.  It’s now better alignment for 15 days, so I’m the last week we’ve bought ourselves about a day - going off detection rates.

Why are you choosing T-14/15 rather fixing the starting at a fixed point , for example, 100 recorded deaths as a result of Covid-19?  I ask because an exponential curve can be made to look like another exponential ... if you slide it back along the scale until it fits.  This chart aims to remove the T-X days guess work and instead start the counting from a fixed baseline.

> Detection rates are not the most useful as they under-detect by different amounts in different counties and the testing fraction can change.  Death rates are a better readout (other than old people being left unreported to die in care homes as in Italy) until healthcare is overwhelmed as in Italy, raising rates.

Agree.  We could use the WHO sit reports data on deaths, baseline at 100 ... and chart for UK vs Italy.

https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200324-sitrep-64-covid-19.pdf?sfvrsn=703b2c40_2

> > UK is ~14 days behind Italy ... but in this chart UK growth after 100/million hasn't been as aggressive as either Switzerland (which we really should be worried about), Spain, Italy, Germany or France.

> You seem to be misreading the chart severely.  It is from 4 days ago before the UK passed 100/million so there is no “UK growth after 100/million”.

You'll need to help me with that one wintertree.  The chart is showing that after the UK reached the same baseline starting point as Italy (100 reported cases = day 0) the growth of infected per 1M is less steep than that of Italy i.e. People are getting infected much faster in Italy and/or Italy is detecting the infected much more effectively than the UK.   Help me out ... you are reading the chart differently and in fact the UK is on the same trajectory as Italy?  What am I getting wrong and what does the fact that the chart is 4 days old have to do with it?  

I'm not getting at you.  I'm only asking you to help me understand why you think the UK is on the Italian trajectory.

Thanks for keeping it civil!  Cheers.

Report
gravy 25 Mar 2020
In reply to Toerag:

  http://nrg.cs.ucl.ac.uk/mjh/covid19/

Makes you wish they used the graph colour consistently - same colour for each country on every graph...

Report
StuPoo2 25 Mar 2020
In reply to wintertree:

okay ... I've dug the numbers out from the WHO daily sit reports using deaths rather than infected as requested.

https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports

***Data***

date,UK,UK_death_million,Italy,italy_death_million
04/03/2020,,0.0000000000,80,0.0000013228
05/03/2020,,0.0000000000,107,0.0000017692
06/03/2020,,0.0000000000,148,0.0000024471
07/03/2020,,0.0000000000,197,0.0000032573
08/03/2020,,0.0000000000,243,0.0000040179
09/03/2020,,0.0000000000,366,0.0000060516
10/03/2020,,0.0000000000,463,0.0000076554
11/03/2020,,0.0000000000,631,0.0000104332
12/03/2020,,0.0000000000,827,0.0000136739
13/03/2020,,0.0000000000,1016,0.0000167989
14/03/2020,,0.0000000000,1268,0.0000209656
15/03/2020,,0.0000000000,1441,0.0000238261
16/03/2020,,0.0000000000,1809,0.0000299107
17/03/2020,,0.0000000000,2503,0.0000413856
18/03/2020,55,0.0000008278,2503,0.0000413856
19/03/2020,103,0.0000015503,2978,0.0000492394
20/03/2020,144,0.0000021674,3407,0.0000563327
21/03/2020,177,0.0000026641,4032,0.0000666667
22/03/2020,233,0.0000035069,4827,0.0000798115
23/03/2020,281,0.0000042294,5476,0.0000905423
24/03/2020,335,0.0000050421,6077,0.0001004795

If rather than using T-14/T-15 ... we instead use 0= 100 deaths.  UK = 19/03, Italy = 05/03.  (Which incidently is T-14 anyway.)

Rebaseline the data (get rid of the deaths/million .. irrelevant since population ~the same)

T,UK,Italy
0,103,107
1,144,148
2,177,197
3,233,243
4,281,366
5,335,463

By this measure ... Italy took off on different trajectory from UK - much steeper.

Help me out.  What are you seeing in this that I'm not.  How are you coming to the conclusion that the UK is on the same trajectory as Italy?

I appreciate you sticking with me on this one.

Cheers.

[1] I didn't go back on the UK after I reached 100.  Stopped there on purpose.

[2] I used google population of UK =66.44M and Italy = 60.48M respectively.

*Fixed my dodgy 2nd set of data.  Yeek.

Post edited at 10:05
Report
wintertree 25 Mar 2020
In reply to StuPoo2:

> Why are you choosing T-14/15 rather fixing the starting at a fixed point , for example, 100 recorded deaths as a result of Covid-19? 

The starting point used by the JP Morgan study you linked was 100 reported cases, not deaths.  This makes a massive difference to how piss poor their methodology is.

It's not "me" its the author of the document Toerag linked.  I find their approach useful.  To explain their methodology - it's not a "fixed point", and it's not "chosen".  It's the time offset that is "found" to give the best agreement between the curves.  The two things measured by doing this are how far "behind" one country another is and how "fast" the infection spreads in terms of the exponential rate.  

The first recorded cases are highly misleading as they can be dominated by imported cases which are - hopefully - quarantined and not by the local-local transmission, and by differences in how pro-active the countries are in testing travellers from infected regions and suspected cases.  The analogy I have use before is counting sporadic small fires vs studying the one that grows out of control into a giant moorland fire. 

> I ask because an exponential curve can be made to look like another exponential ... if you slide it back along the scale until it fits.

Not true.  This can only happen for exponential curves with the same growth rate.  If the growth rate is different, they will never look "similar".  In a log-y axis plot (as in the document Toerag linked), the exponential growth rate is the linear gradient.  

>  This chart aims to remove the T-X days guess work and instead start the counting from a fixed baseline.

I wouldn't like to speak about their "aim".  Hopefully I have now explained how "T-X days" is not "guesswork" but a method based approach.  Saying we are "14 days behind Italy on the curve" does not mean "we are 14 days behind Italy on the ground" but it does mean we have the same exponential growth rate (very bad news) and our detected cases are following theirs by 14 days (somewhat good news as it gave us more time we could have acted more in)

> Agree.  We could use the WHO sit reports data on deaths, baseline at 100 ... and chart for UK vs Italy.

Rather than baselining on 100 deaths, time shifting is more appropriate as I mentioned above; in either case the key parameters is the day-to-day growth as a ration, percentage, fraction or whatever.   In practice, deaths whilst still in the 100s are a better reporter of true infection rates and so there is good agreement between "T-X" days and baselining to 100 total deaths.  You can see this in the plot "Western Europe, Deaths Per Million Inhabitants" in the document Toerag linked.

The tentatively good news there is that in the last 3 days (time since the JP Morgan document you refer to), the UK is dropping below Italy in the "T-X" curve.  This suggests we are either lowering (exponential) rate of spread of infection and/or are providing better clinical care.  This is what happens if we do more or do it better than Italy and is really promising to see.   

> You'll need to help me with that one wintertree. 

About the JP Morgan document you linked, you said "but in this chart UK growth after 100/million hasn't been as aggressive as either Switzerland".  I pointed out that in the 4-day old data in that chart you linked the UK had not reached 100/million and so you must have misunderstood it as it's nonsensical to talk about "UK growth after 100/million" (your words) for a country that is (my words) "before 100/million".  What you are talking about does not exist in the plot.

> The chart is showing that after the UK reached the same baseline starting point as Italy (100 reported cases = day 0) the growth of infected per 1M is less steep than that of Italy i.e. People are getting infected much faster in Italy and/or Italy is detecting the infected much more effectively than the UK.   Help me out ... you are reading the chart differently and in fact the UK is on the same trajectory as Italy? 

The UK is so squashed down to the bottom of that chart it's hard to tell much about it, and comparing exponentials on a linear plot is really difficult.  If you trace the UK curve on a piece of thin paper and slide it along to the left it matches the Italian one quite well.  The massively misleading thing about the JP Morgan data is using "100 reported cases" as their baseline.  This is my "sporadic little fires vs giant wildfire analogy".  Their baseline is subject to an awful lot of randomness in small numbers before the freight train of exponential growth takes over.  

Consider two otherwise identical countries with 10x different testing (and therefore detection) rates both done randomly across the population.  The country testing more aggressively would explode sooner on the JP Morgan plot but both countries would actually be equally screwed.  The testing rate does not change the exponential rate of the curve however, it manifests as a left/right shift.  

> What am I getting wrong and what does the fact that the chart is 4 days old have to do with it?  

The specific statement you made and asked me to reply to is invalid.

> Thanks for keeping it civil! 

Any time.  It's been very interesting for me to see a lot of very smart people (some I am sure quite smarter than me) are having such difficulty seeing what I and other people from similar fields to me see in the data.  It really is very easy to be mislead, and quite a few people in the media preparing analysis are completely out of their depth in doing so and don't realise it.  

Edit: The key point to me is not to get drawn into endless navel gazing over numbers riddled with unknowns and situation dependant differences.  The simple point behind the Italy comparisons several poster have been making is that they were in real trouble and that the UK was taking fever protective measures and appeared to be going the same way, and should probably do something about that whilst it could.  We are now doing things - partial lockdown, preparing for a massive ramp in testing capability, preparing massive field hospital capacity.  I think we will now break the back of this virus sooner than Italy but by god until their U-turn this government took us closer to the brink than I can believe.  

Post edited at 10:18
Report
ClimberEd 25 Mar 2020
In reply to wintertree:

> >otherwise.

> The point is if we do not want to be like Italy we have to do other than Italy did.  

This is why you think I am not listening to you.

I don't agree. that 'if we do not want to be like Italy we have to do other than Italy did'

Report
wintertree 25 Mar 2020
In reply to ClimberEd:

> This is why you think I am not listening to you. 

No, it was the bit where you ignored 90% of what I wrote and went back to one point in isolation of what I wrote.

> I don't agree. that 'if we do not want to be like Italy we have to do other than Italy did'

Other than your poorly explained mention of “genetics” what do you think is going to make us fare better than Italy *without doing things differently*?

To repeat I am not and have never been blindly saying that we are following Italy.  I have been saying we need to work hard to not end up going down the same road as Italy.  You apparently also disagree with this?  Perhaps the warmer weather in England will save us compared to Italy...

Post edited at 10:29
Report
sammy5000 25 Mar 2020
In reply to ClimberEd:

Here is a fact if you dont think we are.

My sister is manager of a care home, on friday evening a work colleague of hers with whom she had been in close proximity of for a number of days, Fell seriously ill and was taken to hospital with suspected Meningitis.

It turns out it was covid.

So my sister had to be informed, my sister fell ill on monday evening with symptoms, it doesnt take a rocket scientist to know that she has it. Yet they will not test to see if she can go back to work.

She is so upset that her residents may have also caught it who are all vulnerable.

Her family are all now at home thats four under one roof I think we can safely say they are all infected!

The outcome of which so far we will not know.

My partner also works in the care sector for NYCC.

She was informed last night that one of her clients though not tested is also most likley infected this information was given to her by the local doctor.

So this is just me, how many others like me know of almost certain case's.

Give it two weeks you will see the full magnitude of what is unfolding!

Report
ClimberEd 25 Mar 2020
In reply to sammy5000:

> Here is a fact if you dont think we are.

> My sister is manager of a care home, on friday evening a work colleague of hers with whom she had been in close proximity of for a number of days, Fell seriously ill and was taken to hospital with suspected Meningitis.

> It turns out it was covid.

> So my sister had to be informed, my sister fell ill on monday evening with symptoms, it doesnt take a rocket scientist to know that she has it. Yet they will not test to see if she can go back to work.

> She is so upset that her residents may have also caught it who are all vulnerable.

> Her family are all now at home thats four under one roof I think we can safely say they are all infected!

> The outcome of which so far we will not know.

> My partner also works in the care sector for NYCC.

> She was informed last night that one of her clients though not tested is also most likley infected this information was given to her by the local doctor.

> So this is just me, how many others like me know of almost certain case's.

> Give it two weeks you will see the full magnitude of what is unfolding!

Thanks for that fact, it is a nicely emotive scare story.  You have completely misunderstood my point. I haven't said that there isn't a huge health problem about to sweep through the population (or however you wish to describe it) I have just said I wish people would stop banging on about Italy as the poster child for what happens to a country if it gets hit by the virus and follows the same actions that Italy did. 

Edit: aside from the discussion, I should say I am sorry to hear about your sister and the residents in her care home. It is quite horrible how it is going to closely effect everyone.

Post edited at 10:49
Report
ClimberEd 25 Mar 2020
In reply to wintertree:

>

> Other than your poorly explained mention of “genetics” what do you think is going to make us fare better than Italy *without doing things differently*?

What do you mean poorly explained? Humans are not identical and different genetic make up will make them differently susceptible to different diseases. That is fairly well understood and agreed on.

Additionally, cultural behaviour will change the natural spread of the virus if there is no intervention. For example the Italians frequently live in multigenerational families, they are far more tactile and generally more sociable than the British. Who knows if this would make a difference or not, but I disagree with the assumption that we are the same and equally susceptible. 

Post edited at 10:52
Report
Neil Williams 25 Mar 2020
In reply to ClimberEd:

> Additionally, cultural behaviour will change the natural spread of the virus if there is no intervention. For example the Italians frequently live in multigenerational families, they are far more tactile and generally more sociable than the British. Who knows if this would make a difference or not, but I disagree with the assumption that we are the same and equally susceptible. 

They also mostly live in flats in their cities, we mostly don't other than London.  Most people in the UK can go out and return home (say for a run) without touching anything that anyone else has touched at all.  Most urban people in Italy, Germany, France etc can't - think doors, banisters, lifts, passing people on stairs etc.  It's only really the Netherlands that are suburban house dwellers to anything like the same extent as us.

Post edited at 10:48
Report
wintertree 25 Mar 2020
In reply to ClimberEd:

> Additionally, cultural behaviour will change the natural spread of the virus if there is no intervention.

You mean what we do will be different to what Italy is doing?  That has been my part of my point all along one one you disagreed with not two posts ago.

Although it’s remarkable that despite the big social differences we have been following such a similar track Italy with a lag of about 14 days.  The funny thing about lots of different factors is some are one way, some are the other and they tend to average out somewhat.  A phenomena called a “Fermi estimate”.  You focus on the ones that make it worse for Italy.  Yet our similar course to date suggests there must be others that make it worse for the UK; finding and controlling those should be a priority.

From my view you continue to argue against a stance I haven’t actually seen made on UKC, more against your misinterpretation of what has been said here and even lately Boris himself.

Post edited at 10:53
Report
ClimberEd 25 Mar 2020
In reply to wintertree:

> > Additionally, cultural behaviour will change the natural spread of the virus if there is no intervention.

> You mean what we do will be different to what Italy is doing?  That has been my part of my point all along one one you disagreed with not two posts ago.

arghhh! No! I mean pre virus, I mean cultural norms. I mean standard cultural behaviour pre virus intervention. 

Report
wintertree 25 Mar 2020
In reply to ClimberEd:

> arghhh! No! I mean pre virus, I mean cultural norms. I mean standard cultural behaviour pre virus intervention. 

Then how do you explain the exponential rate being so similar for both countries in the first month or so of local infection?    That’s unarguable now and has solid data (Toerag’s link) and has been recognised by sensible government

For that similarity to change going forwards, something different has to change, or there has to be a pre-existing societal difference that only “kicks in” about now.

It looks like the UK is going to take tougher control measures and do much more testing now so I expect the trajectories to diverge more.  Critical to this seems to be that the government got the message that we had to do more than Italy to avoid ending up in their situation.  

Post edited at 11:03
Report
Robert Durran 25 Mar 2020
In reply to ClimberEd:

>  I have just said I wish people would stop banging on about Italy as the poster child for what happens to a country if it gets hit by the virus and follows the same actions that Italy did. 

But all the evidence suggests that it most likely is. We definitely should be banging on about it!

Report
Robert Durran 25 Mar 2020
In reply to wintertree:

> It's been very interesting for me to see a lot of very smart people (some I am sure quite smarter than me) are having such difficulty seeing what I and other people from similar fields to me see in the data.  It really is very easy to be mislead, and quite a few people in the media preparing analysis are completely out of their depth in doing so and don't realise it.  

https://www.facebook.com/photo.php?fbid=10219301871391874&set=a.1147301637049&type=3&theater

Report
Richard Horn 25 Mar 2020
In reply to Toerag:

> Looks like we are to me.....    http://nrg.cs.ucl.ac.uk/mjh/covid19/

As an engineer I review a lot of study results like this - my main concern in drawing concrete solutions is that the author has shifted the plots sideways on each plot in order to overlay them but not by the same amount on each plot.

We know that different countries are recording the cases/deaths in different ways - Germany is not postmortem testing (i.e. if someone dies before a +ve CV test, then their death is not recorded as CV related), and France is not testing people who die outside of hospital, so both countries have artificially lower death rates. Britain is not testing as much so our cases rate is likely to be a significant under-estimate.

If I were "in charge" though I think the data is showing enough risk to justify the action the government is taking. Whether or not we will do an "Italy" is largely irrelevant. 

The picture will become much more complete when antibody testing is rolled out, until then we have to act to avoid the worst case scenarios...

Report
RomTheBear 25 Mar 2020
In reply to ClimberEd:

So far our trajectory is slightly worse than Italy. We’re much more similar to Spain right now. But we took action just a little bit Italy, relative to where we are, so maybe won’t be as bad.

Post edited at 11:33
Report
wintertree 25 Mar 2020
In reply to Richard Horn:

> my main concern in drawing concrete solutions is that the author has shifted the plots sideways on each plot in order to overlay them but not by the same amount on each plot.

That’s intentional and is one of the key points.  This shift does not change the *gradient* of the log-y plots, with the gradient being another key point.  Doing the shift makes a measurement (the X in T-X days) and along with a log y axis makes it very easy to compare by eye the key measurement embodied by gradient.

The amount each is shifted he is a measurement not an arbitrary quantity and is a holistic (using all dat) way of giving some sort of comparison between different countries with different local transmission start times.

I agree with the rest of your post.

Report
freeflyer 25 Mar 2020
In reply to ClimberEd:

Ok, let's take a different, non-numerical perspective. If we are not following Italy, how will that change things for you?

Will you feel better?

Will you behave differently?

Report
Robert Durran 25 Mar 2020
In reply to Richard Horn:

> As an engineer I review a lot of study results like this - my main concern in drawing concrete solutions is that the author has shifted the plots sideways on each plot in order to overlay them but not by the same amount on each plot.

> We know that different countries are recording the cases/deaths in different ways - Germany is not postmortem testing (i.e. if someone dies before a +ve CV test, then their death is not recorded as CV related), and France is not testing people who die outside of hospital, so both countries have artificially lower death rates. Britain is not testing as much so our cases rate is likely to be a significant under-estimate.

It is a special (and slightly technical) property of exponential growth (and the log-y graph linked earlier being linear shows that the growth is exponential) that, if a country detects a fixed proportion of cases, or records a fixed proportion of covid deaths as being covid, then, if the true infection or death rates are following the same exponential growth, the graphs will fit if shifted sideways (wintertree alluded to this earlier in the thread). So it is entirely reasonable to shift the plots different amounts sideways to make them fit - the different shifts simply reflect different detection rates - and conclude that the true rates are similar.

Report
summo 25 Mar 2020
In reply to ClimberEd:

The UK as a whole might not follow Italy. But London might follow Lombardy, dense urban area, high footfall in shops and transport etc. With an incubation period of up to 14 days, plus (according to r4 more or less) the average time to Death, for those that die, being 2 weeks, the measures that started yesterday are a few weeks away from having a significant impact on fatalities, new infections should now of course fall. 

Report
Timmd 25 Mar 2020
In reply to ClimberEd:

Being alarmed at the prospect of following Italy doesn't seem like a bad thing if it means we do things to make sure we don't. 

Report
DancingOnRock 25 Mar 2020
In reply to ClimberEd:

Our deaths are dominated by London. 1/3 of them. Also London has a massive ability to work from home and are mainly young people. However, we have been cramming people into trains. But we have all been washing our hands like mad and are generally quite socially distant compared to the kissing Europeans. 
 

Italy deaths are dominated by old people in villages and care homes. 
 

Different demographics and behaviours. 

Report
stevieb 25 Mar 2020
In reply to wintertree:

> Other than your poorly explained mention of “genetics” what do you think is going to make us fare better than Italy *without doing things differently*?

I have no wish to underplay the risks, and I think our numbers will continue to rise scarily for the next couple of weeks, but there is one significant difference between the U.K. and Italy. 
That is, that even while the government dithered and delayed, a lot of the population and a lot of companies were making significant social distancing steps, especially the most vulnerable, precisely because we could see what was happening in Italy. 

Report
wintertree 25 Mar 2020
In reply to stevieb:

> That is, that even while the government dithered and delayed, a lot of the population and a lot of companies were making significant social distancing steps, especially the most vulnerable, precisely because we could see what was happening in Italy.

Totally agree - I pushed hard for my employer to react in advance of the government as did many others, and react they did.  Many individuals did too.  Those of us far from London will have that as further motivation.  I really hope to see this difference develop over the weeks ahead.

Post edited at 12:26
Report
ClimberEd 25 Mar 2020
In reply to freeflyer:

> Ok, let's take a different, non-numerical perspective. If we are not following Italy, how will that change things for you?

> Will you feel better?

> Will you behave differently?

It will make me feel better if there are less articles in the media (print, online, old and new) about doctors from Italy, people from Italy, the world and his wife saying that if we don't act fast we'll end up like Italy. 

Report
Robert Durran 25 Mar 2020
In reply to ClimberEd:

> It will make me feel better if there are less articles in the media (print, online, old and new) about doctors from Italy, people from Italy, the world and his wife saying that if we don't act fast we'll end up like Italy. 

Sorry, but that sounds as if you want to bury your head in the sand. I hope not, but I think there is a very good chance you might be looking back on that post in a few weeks and seeing it as somewhat foolish - either we shall be like Italy or the fear of being like Italy will have led us to take drastic action to avoid being so.

Report
ClimberEd 25 Mar 2020
In reply to Robert Durran:

nah, I'm well aware of the problems that we face, several friends work in the NHS, another in emergency planning for the government etc. blah blah.  I'm just bored of hearing about Italy, we're not Italy.

Report
Robert Durran 25 Mar 2020
In reply to ClimberEd:

>  We're not Italy.

But you're stichtplate and I claim my £5   ;-)

Report
mountain.martin 25 Mar 2020
In reply to ClimberEd:

> . blah blah.  I'm just bored of hearing about Italy, we're not Italy.

Yes but we are like Italy

> Oh no we are not

Oh yes we are

> Oh no we are not

He's behind you

Report
mountain.martin 25 Mar 2020
In reply to ClimberEd:

>   I'm just bored of hearing about Italy, 

Why did you start a thread about Italy then and contribute about 10 posts to the thread?

I guess none of us are busy key workers.

Report
Richard Horn 25 Mar 2020
In reply to Robert Durran:

Hi Robert, you miss my point, if you look at the cases graph, Spain is 5 days behind Italy, on the deaths chart the author has had to shift the plot by 8 days for Spain to overlay. This suggests the deaths/cases ratios are not the same, which in turn suggests we are not getting the full picture perhaps for either country. The overall message is pretty clear though...

Report
tom_in_edinburgh 25 Mar 2020
In reply to ClimberEd:

We *have been* following Italy, the graph is absolutely clear.

We still are not doing enough to get off that path.   London in particular probably has a disaster already locked in.  Maybe some other regions have shut down hard enough and fast enough.  But we don't have proper cordons between regions and folk with symptoms - including Prince Charles - are running from London and spreading it.

Report
wintertree 25 Mar 2020
In reply to Richard Horn:

> This suggests the deaths/cases ratios are not the same,

Or the infection/detection ratios 

> which in turn suggests we are not getting the full picture perhaps for either country.

Totally agree.

> The overall message is pretty clear though...

Yup, and the key point to me is that the day-on-day fractional increase - the key measure of spread - is unrelated to the x-axis shifts and relates only to the gradients, which are not changed by the shifts.  

Report
Robert Durran 25 Mar 2020
In reply to Richard Horn:

> Hi Robert, you miss my point, if you look at the cases graph, Spain is 5 days behind Italy, on the deaths chart the author has had to shift the plot by 8 days for Spain to overlay. This suggests the deaths/cases ratios are not the same.

Or the deaths/reported cases ratios are not the same; the shifts are a combination of the actual lag between the countries and the proportion of cases or deaths recorded. Either way, the graphs suggest that the doubling times for cases are similar and that the doubling times for deaths are similar.

Post edited at 13:45
Report
Ciro 25 Mar 2020
In reply to ClimberEd:

What is it that you think we will gain, by not considering ourselves on the same path as Italy unless we do something about it?

Report
Timmd 25 Mar 2020
In reply to Ciro:

> What is it that you think we will gain, by not considering ourselves on the same path as Italy unless we do something about it?

Exactly. 

Edit: It's a case of 'What happens if we hurry and take preventative measures which have a temporary economic hit but save lives?

Post edited at 14:22
Report
Baz P 25 Mar 2020
In reply to ClimberEd:

Italy aside, has anyone seen any reports from Russia as they seem to be conspicuous by their absence in tables?
 

Report
Robert Durran 25 Mar 2020
In reply to tom_in_edinburgh:

> We still are not doing enough to get off that path.   London in particular probably has a disaster already locked in.  Maybe some other regions have shut down hard enough and fast enough.  But we don't have proper cordons between regions and folk with symptoms - including Prince Charles - are running from London and spreading it.

I reckon it's all just another Westminster conspiracy.

Post edited at 14:06
Report
jkarran 25 Mar 2020
In reply to ClimberEd:

> This is why you think I am not listening to you. I don't agree. that 'if we do not want to be like Italy we have to do other than Italy did'

Why not?

There was a lot of this said a week or two back and it sounded like the usual English exceptionalism, I didn't agree then and I'm struggling to now.

jk

Report
jkarran 25 Mar 2020
In reply to summo:

> The UK as a whole might not follow Italy. But London might follow Lombardy, dense urban area, high footfall in shops and transport etc. With an incubation period of up to 14 days, plus (according to r4 more or less) the average time to Death, for those that die, being 2 weeks, the measures that started yesterday are a few weeks away from having a significant impact on fatalities, new infections should now of course fall. 

...If they're sufficient.

It may well be in a dense urban environment like London or New York that is very heavily dependant upon public transport that it's impossible to fight/treat the disease without the transport network open but that the environment is sufficiently contaminated it's impossible to contain the growth with it open. In which case different measures are needed, whether they're practically achievable or palatable...

jk

Report
freeflyer 25 Mar 2020
In reply to ClimberEd:

I understand. So you think there's a chance that the UK experience will be different - better or worse? Should we not act, or are you just a bit fed up with all the CV coverage?

Report
cb294 25 Mar 2020
In reply to ClimberEd:

>   I'm just bored of hearing about Italy, we're not Italy.

Of course you are not Italy.

If you were, you would have won the football world cup more than once (and would not have needed the help of some dodgy linesman ...).

CB

Report
StuPoo2 26 Mar 2020
In reply to wintertree:

My apologies for the delay in getting back to you wintertree ... home school to run. 

Last questions:  

#1 The charts in http://nrg.cs.ucl.ac.uk/mjh/covid19/.  I see on most charts multiple dotted lines at various gradients.  The country lines appear to show initially a steeper gradient before curving round towards a more shallow one.  Am I reading them right to understand that the exponential growth rates are deceasing (for whatever reason)?

I appreciate you sticking with me - thank you.

Report
wintertree 26 Mar 2020
In reply to StuPoo2:

> I see on most charts multiple dotted lines at various gradients.  The country lines appear to show initially a steeper gradient before curving round towards a more shallow one.  Am I reading them right to understand that the exponential growth rates are deceasing (for whatever reason)?

Correct - keeping in mind it’s either the exponential rate in *detected cases* or *recorded deaths* that is decreasing, They have to curve over at some point until they aren’t exponential any longer, as a country runs out of uninfected / non-immune people.  Hopefully they start dropping under the initial curve sooner as control measures kick in - with a delay from when they start happening to when they show an effect.  Without control measures it goes from an exponential to a “logistic sigmoid” over time - https://en.m.wikipedia.org/wiki/Logistic_function - control measures may just reduce the exponential rate (gradient in the plot you linked), or may force it sigmoidal sooner, may lower the final line and may introduce all sorts of complex structure.

Post edited at 11:43
Report
StuPoo2 26 Mar 2020
In reply to wintertree:

Got you.

Can't help but think that a big part of the problem is the inability for us humans to naturally understand the exponential:  https://www.albartlett.org/presentations/arithmetic_population_energy.html

Anyhoo ... thanks again.  Been a good experience on UKC forums.

Peace out!

Report
wintertree 26 Mar 2020
In reply to StuPoo2:

> Can't help but think that a big part of the problem is the inability for us humans to naturally understand the exponential

If they did, car finance and living on credit would never have become commonplace.

I think the media have a lot to answer for here as well - many of them didn’t understand either and so they did not communicate the situation at all well; I still see that in some of the data in the press now.  To my mind the media should have a duty to understand what they report on.  I’ve been particularly unimpressed by the BBC “expert”.

Two powerful visualisations on an exponential are putting 1 grain of rice on a chessboard square, then 2 on the next, then 4 and so on.  How much rice do you need?  It’s only 64 numbers, doubling each time; a simple paper exercise. Speaking of paper, how many times do you need to fold a piece of paper before it’s a km thick?  About 24!

Post edited at 12:04
Report
Toerag 26 Mar 2020
In reply to StuPoo2:

Yes. On the log scale graphs exponential growth is a straight line. The rate of that growth is denoted by the steepness of the line. It is still exponential growth though. So, most countries do 35% per day to start with - doubling every 2-3 days. Countries with moderate controls (social distancing, partial lockdowns like UK and most of Europe) grow at 22% - doubling every 5 days. Countries with hard lockdowns or strict rules who caught it early (South Korea, Japan) see lower rates still, but it's still exponential, just over a much longer timescale. Growth builds on growth, like compound interest.

Maths experts - can you tell me if the 35% / 22% 'slope rate' is a compound one? I.e. if it's 33% does that mean 33% of today's value each day, or 33% of the previous days value?

What no-one seems to be doing is working out what rate we need to keep below to prevent the NHS being overwhelmed. There's some ideas of ultimate case numbers the NHS can handle being bandied about (NHS have xxx ICU beds, x% of case will require them, therefore the total number of cases = yyy), but nothing on when we'll hit those numbers with what growth rates. For example, South Korea's growth rate is relatively low, but it's still growth and at some point either the health system gets overwhelmed or herd immunity kicks in. Which comes first for the UK at what rates?

Post edited at 12:25
Report
Toerag 26 Mar 2020
In reply to Richard Horn:

> Hi Robert, you miss my point, if you look at the cases graph, Spain is 5 days behind Italy, on the deaths chart the author has had to shift the plot by 8 days for Spain to overlay. This suggests the deaths/cases ratios are not the same, which in turn suggests we are not getting the full picture perhaps for either country. The overall message is pretty clear though...


The author alludes to that in his commentary on the US death rate per million graph:-

"Confirmed cases in the US have been growing very fast recently. The US testing procedures were very slow to get started, but are reported to have scaled up quickly. How much of the rapid increase in case data is due to this scale-up? This graph shows the US death rate, with Italy and France for comparison. When looking at confirmed cases, the US is roughly 15 days behind Italy, whereas when looking at death rates, the US is roughly 19 days behind Italy. This probably indicates that a small part of the recent increases is indeed due to improved testing, but the difference is nowhere near as large as Germany or Luxembourg, so the underlying infection rate must also be high. If this gap widens in the future, then this would be evidence that testing is getting ahead of the epidemic."

Italy is only testing hospital cases too. This is one of the problems with the whole thing, no two countries are testing the same sorts of people, or counting the deaths the same. Here in Guernsey they're only testing people who are hospitalised or have travel history or contact history with travellers due to a lack of ability to test. They've only identified one 'community seeded' case, but that's because they're simply not testing for it .  However, as long as a country's method remains constant you can compare the graph shapes.  We've now just got a tester on island, so I'm expecting our test and confirmed case volumes to rocket over the next few days.

Post edited at 12:21
Report
Robert Durran 26 Mar 2020
In reply to wintertree:

> Speaking of paper, how many times do you need to fold a piece of paper before it’s a km thick?  About 24!

.........and then only another 18 times to reach the moon.

I do wonder whether there is an inbuilt "biological" reason why we find exponential growth hard to grasp. Is it something we simply were not exposed to affecting our lives while evolving in small groups on the African savannah? Rather like we often massively overestimate risk - if you hear gossip of a man eating lion amongst your small savannah living band, it means a very real and immediate danger, and hence the overreaction when people hear about a paedophile at the other end of the country on the news. Though of course, in the case of Covid-19, the distant danger, through exponential growth, very rapidly does become local!

Report
RomTheBear 26 Mar 2020
In reply to Robert Durran:

> .........and then only another 18 times to reach the moon.

> I do wonder whether there is an inbuilt "biological" reason why we find exponential growth hard to grasp. Is it something we simply were not exposed to affecting our lives while evolving in small groups on the African savannah? Rather like we often massively overestimate risk - if you hear gossip of a man eating lion amongst your small savannah living band, it means a very real and immediate danger, and hence the overreaction when people hear about a paedophile at the other end of the country on the news. Though of course, in the case of Covid-19, the distant danger, through exponential growth, very rapidly does become local.

We probably have evolved to understand the constant acceleration of a lion coming toward us, not that of a lion with increasing acceleration.

Post edited at 21:50
Report
sammy5000 29 Mar 2020
In reply to ClimberEd:

oh shit! no scare mongering stories what are todays totals?

17 deaths in lancaster yesterday our nearest a and e.

two ambulances in our street last night.

it aint rocket science.

Report
tom_in_edinburgh 29 Mar 2020

Any doctors/biologists like to comment on chloroquinine?   

https://www.mediterranee-infection.com/wp-content/uploads/2020/03/Hydroxychloroquine_final_DOI_IJAA.pdf

Post edited at 05:51
Report
elsewhere 29 Mar 2020
In reply to Toerag:

33% exponential growth is 33% regardless of scale or when on the curve.

10 cases one day , 13 the next

1000 cases one day, 1330 the next

100 thousand cases one day, 133 thousand cases the next.

In reality it's not 33% every day as changes in testing and behaviour have an impact but the general direction is clear.

The log(confirmed cases) Vs date graphs are near enough straight lines so you can extrapolate to a peak when exponential growth stops because a large proportion of the population have already been exposed.

UK confirmed cases grow by about 10 in last 12 days

17,000* cases 28 March

170,000 cases 9 April

1.7 million cases 21 April

17 million cases 4 May 

170 million cases 16 May to emphasise exponential growth CANNOT continue

Looks like growth will saturate mid/late April when a large proportion of the UK population will have been exposed.

*many unconfirmed cases for every confirmed case

https://www.worldometers.info/coronavirus/country/uk/

Post edited at 07:23
Report
Blunderbuss 29 Mar 2020
In reply to elsewhere:

Actual growth (not those who have just been tested) will slow due to social distancing measures being implemented and mostly adhered to....there is obviously a time lag for them to take effect.

A ramp up in testing may see the recorded growth continue to be higher than it has been due to the simple fact that more people are being tested.

TBH the only real metric we can measure with any reliability is the total deaths as these are fairly unambiguous.

Report
Ardo 29 Mar 2020
In reply to Blunderbuss:

> TBH the only real metric we can measure with any reliability is the total deaths as these are fairly unambiguous.

Real question, as I've not found clarity on this point: are the stats quoting deaths caused by covid-19, as opposed to all deaths where people test positive?

Report
Blunderbuss 29 Mar 2020
In reply to Ardo:

I think it is deaths where COVID19 was a contributory factor...as long as the measurement is consistent it gives us a good idea of the impact of any measures we are taking.

Report
Michael Hood 29 Mar 2020
In reply to Blunderbuss:

Except that Germany and France (and presumably most places) are using different criteria to record CV deaths.

However, as long as each country is consistent in its methodology over time, then on a logarithmic graph the gradient will still show how well (sic) we're doing compared to other countries.

The time shifting (X days behind Italy etc) just makes the comparison easier to see.

Same applies to detected cases, just that changes in methodology (i.e. more widespread testing) are more likely and will make it a bit more difficult to interpret.

Report
Dr.S at work 29 Mar 2020
In reply to Ardo:

The way they are reported suggests that it’s patients who die that have been tested + for Coronavirus, not that they died of Coronavirus.

That means notionally you could die of a heart attach having just been told you have tested +ve for corona virus and be recorded as a Coronavirus death.

it also means people who die of corona virus but do not test positive may not be counted - tests are invariably not 100% accurate.

At this stage there are probably not the resources to pick out which deaths are the ‘just caused by’ corona ones.

Report
Michael Hood 29 Mar 2020
In reply to Ardo:

Similar to what Blunderbuss has said, as long as each country's measuring is consistent then they can be compared.

Remember that what is being compared is not the amounts, it's the rate of increase in the amounts (sometimes expressed as the doubling time).

If country A always tests 10% of all actual cases out there, but country B always tests 50%, then (even if you don't know that the numbers are 10% and 50%) their rates of increase can be validly compared (even if their actual amounts cannot be validly compared).

Report
Offwidth 29 Mar 2020
In reply to Michael Hood:

Except that never happens.  Typically in Europe all contacts are tested in the initial stage, and later, when the system is swamped, testing is mainly only those in the health system  (patients and staff). Testing robustness varies a lot from country to country (shit initially in the US and good in the far east). The most accurate numbers will be deaths but even then those with underlying conditions might go different ways in different countries on the labelling of comorbility (hence complains from Italian medics about German assignments of causes of death)

In the meantime we are still about 14 days behind Italy and I'm still hoping at some point our numbers start to look better. The Medical Director on the government  BBC news announcement yesterday said he is sure everyone in the UK will be following advice. It's an idiotic thing to say. Lots of people will always do really stupid things. As a coronavirus example mobile data from Paris indicates about a million left the city before it was locked down.

https://www.independent.co.uk/news/world/europe/coronavirus-paris-france-lockdown-one-million-orange-phone-data-a9429896.html

Report
Roadrunner6 29 Mar 2020
In reply to tom_in_edinburgh:

Not an MD, the French paper was deeply flawed, and had a similar overall mortality.

so far the trials suggest it may help a bit but no great change. Other actual anti-virals look more hopeful.

Report
HansStuttgart 29 Mar 2020
In reply to Offwidth:

> The most accurate numbers will be deaths but even then those with underlying conditions might go different ways in different countries on the labelling of comorbility (hence complains from Italian medics about German assignments of causes of death)

The official German numbers (rki.de) for deaths include both people who die because of the coronavirus and people who die while having coronavirus and where the cause of death is unclear.

Hypothetical cases of people who die at home during the night without showing any signs of corona symptoms before are not counted. But the virus seems to act relatively slow and therefore most serious cases should be noted.

The other sign that the cases/death ratio is very different is that the ratio between the number of people in intensive care and the amount of cases is also much lower than in other countries.

Report
cb294 29 Mar 2020
In reply to HansStuttgart:

Statistics do play a role, but cannot explain the entire difference. Germany had some isolated cases early, but the massive increase was largely driven by cases brought in by skiers, i.e. a resonably young and fit section of the general population. Where it hits nursing homes, e.g. in my wife's hometown, death rates are higher.

Report
HansStuttgart 29 Mar 2020
In reply to cb294:

I actually thought this as well, together with the idea that in Italy more families live together with the grandparents. But, it is rainy and snowy here today and there was not much else to do, so I looked up some numbers. The demographic distribution of cases is here:

https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Situationsberichte/2020-03-28-en.pdf?__blob=publicationFile

Those can be compared to the population pyramid. The result is that the population is pretty much uniformly impacted in terms of age. The largest deviation is that men of 80+ are more affected and persons younger than 15 are less affected.

So the current fatality rate is actually a good representation for the German population. This rate will go up in the coming weeks because the number of cases will level of whereas the number of fatalities will continue to rise for a while longer. And then at a later stage we have antibody testing and then the rate will drop down again once we detect the hidden cases.

Post edited at 16:01
Report
cb294 29 Mar 2020
In reply to HansStuttgart:

Good point, patient population structure does not seem to be the reason, then.

I still doubt that the difference in death rates to, say, Italy, is just driven by testing rate or other statistical effects alone, and certainly not quality of the medical system. There must be some other factor at work. Otherwise, there would not be be a stable time lag relative to Italy of 12ish days for cases but 21 days for deaths.

Whatever that factor is, it needs to be identified before it catches us out when or if it changes (e.g. in my hypothesis, cases shifting from skiers to the general population).

CB

Report
HansStuttgart 29 Mar 2020
In reply to cb294:

> I still doubt that the difference in death rates to, say, Italy, is just driven by testing rate or other statistical effects alone, and certainly not quality of the medical system. There must be some other factor at work. Otherwise, there would not be be a stable time lag relative to Italy of 12ish days for cases but 21 days for deaths.

Why not? To get the fatality rates to match, Italy must have been only testing 10% of the cases compared to Germany. With a doubling time of 3 days, this factor of 10 shows up as 9-10 days delay.

Report
Robert Durran 30 Mar 2020
HansStuttgart 30 Mar 2020
In reply to Robert Durran:

very nice!

Report
jkarran 30 Mar 2020
In reply to HansStuttgart:

> Hypothetical cases of people who die at home during the night without showing any signs of corona symptoms before are not counted. But the virus seems to act relatively slow and therefore most serious cases should be noted.

Presumably it's not so much those caught out, dying rapidly at home that are likely missing, it's those who may not be medical priorities or willing to engage with the state for one reason or another choosing to stay at home so they're not all alone in a swamped hospital for their final days. It's not been put that starkly to the public here yet but we're being softened up to pre-make this choice already.

jk

Report
HansStuttgart 30 Mar 2020
In reply to jkarran:

> Presumably it's not so much those caught out, dying rapidly at home that are likely missing, it's those who may not be medical priorities or willing to engage with the state for one reason or another choosing to stay at home so they're not all alone in a swamped hospital for their final days. It's not been put that starkly to the public here yet but we're being softened up to pre-make this choice already.

Hi Jk,

I was specifically answering a point about German statistics. I don't know enough about the situation in other lands to comment about the situation there.

In Germany most of these are not an issue. We are very far away from swamping the hospitals. And the capacity is so large that everybody with serious symptoms is a priority. To give you some idea, there is public data available about half of the ICU units (14000). From these currently 6700 are free, 1100 are used for Covid patients.

Hans

Report
Dr.S at work 30 Mar 2020
In reply to HansStuttgart:

And interestingly still building extra capacity

Report
Bobling 30 Mar 2020
In reply to ClimberEd:

I've been wondering for the last week what effect the decision not to close London off before last weekend will have.  The flight to second homes etc in almost every corner of the country will surely have an effect on the pattern of the outbreak here.  Just a thought!

Report
wintertree 01 Apr 2020
In reply to ClimberEd:

A week since your OP and we’re still really close to the trajectory of Italy, but we’ve bought ourself an additional day, now lagging by 15 days not 14.  That might be back to 14 when today’s data is assimilated into the analysis.

http://nrg.cs.ucl.ac.uk/mjh/covid19/

I’ve still seen nobody blindly say we are bound to follow their trajectory, and a week later you’ve not actually referenced anyone as having said that.

Report
ClimberEd 01 Apr 2020
In reply to wintertree:

Following Italy was splashed across the media many times. Italy is the curve upon which our progress is measured and the 'curve of concern'. I believe this was overhyped.   I don't need to reference it because you ask me to, you know full well that it was what was being said/discussed.

Each country is different, we are different. 

And you and I are never going to agree. 

Report
wintertree 01 Apr 2020
In reply to ClimberEd:

> Following Italy was splashed across the media many times.

Until this post you’d not actually qualified your concern as being with regards the media, rather than the many relevant discussions on UKC at the time.

> I believe this was overhyped.

Or a key part of getting the government to start taking Italian style measures.  


> I don't need to reference it because you ask me to, you know full well that it was what was being said/discussed.

What gives you the ability to assert what I do or don’t know?  I maintain that the news stories I read and the discussions on UKC were sufficiently nuanced that the consistent message was not “the UK will blindly follow Italy”, rather that the point was made things could rapidly spiral out of control if we didn’t rapidly enact tough measures, and that the head in the sand approach wasn’t likely to work out any better for us.  I suspect you can’t reference any UKC or main stream media stories to back up your position because things were more nuanced than that.

> And you and I are never going to agree. 

No, but the curves continue to.

Post edited at 21:41
Report
SenzuBean 05:48 Sun
In reply to ClimberEd:

> Following Italy was splashed across the media many times. Italy is the curve upon which our progress is measured and the 'curve of concern'. I believe this was overhyped.   I don't need to reference it because you ask me to, you know full well that it was what was being said/discussed.

> Each country is different, we are different. 

> And you and I are never going to agree. 

Has your opinion changed now that the UK is recording 700 deaths a day?

https://www.worldometers.info/coronavirus/country/uk/

Post edited at 05:48
Report
ClimberEd 07:20 Sun
In reply to SenzuBean:

nope. Each country has it's own profile. We are not Italy. (Italy is not France, France is not Spain ad infinitum.) 

Report
BnB 07:36 Sun
In reply to ClimberEd:

This is very interesting

https://giftarticle.ft.com/giftarticle/actions/redeem/f7de3a3e-ff0c-49e2-8c4c-f1665cc78d53

Apparently the divergent hospitalisation policies of Lombardy and Veneto have led to a significant contrast in casualties. Basically, increased hospitalisation accelerates the death rate. 

Report
Michael Hood 10:51 Sun
In reply to BnB:

I presume that once a hospital becomes overloaded, and/or the medical staff have insufficient PPE, they become centres of infection.

Two weeks later, higher number of deaths.

But up until that point (not overloaded, enough PPE) I can't see how hospitalisation would make things worse.

Problem is, most places are reaching (or will reach) overload. Maybe there is a tipping point where you stop admitting people for the good of society as a whole. That will be tough on those individuals denied medical intervention that might stop them dying.

Edit: https://go.nature.com/2yqz47x is pretty uncomfortable reading.

Post edited at 11:06
Report
Offwidth 11:47 Mon

Please Register as a New User in order to reply to this topic.