COVID19

[kukushkin88]

ballpark figure...

given an R of 0.7-0.9 now and human nature being what it is (a proportion of the population will be out a lot more, more people will be returning to work even if behaving prudently)..what do you think is the likelihood of rowing back on lifting lockdown in a month or so, a so called second wave?

more likely than not? very likely? not likely?

Barring something that can’t reasonably be foreseen, there’ll need to be a second, probably more strict lockdown. It will be framed that it is the fault of people not following the rules, if you like Boris, you’ll agree that is where the blame should lie. It will be hard to objectively assess why it was actually needed because the data will be kept secret.

Oh for heaven’s sake man. Just let it go. The constant second guessing of policy is irritating beyond reason, as is your determination to find a metric that shows how badly we’re doing. Your echo chamber views and linking just "oushdiugdiyhoimsougfyiygi#&)#’skssyj(#*’a  

You

Simple maths is all that’s needed to evidence how badly we’re doing. I’m keen to find a metric that suggests we’re doing well or even OKish, it’s tough.

[Doobs]

He's just mentioned this figure of we've avoided 500,000 deaths again. Since when was that a thing? I thought the worst-case they were working towards was the 20,000 we've already passed?

Smells of trying to spin a failure into some sort of success.

20000 was the best case scenario.

500,000 was the modelled figure for if we did nothing

There’s quite a bit more to the 20k figure than that. Prof Ferguson said 7 to 20k best to worst case with our lockdown, it was discussed here at length. Powis and Vallance both said 20k would be a “good outcome”.

500k is what Ferguson told SAGE as ‘do nothing’ worst case, the number which is widely reported to have influenced the belated decision to partially lockdown.

There's a bit more to it than that.

When Prof Ferguson said those figures - that was a bit odd. That's what wa discussed on here before .... it was a very odd statement.

But the 20k figure came up originally in a daily briefing. It was in an answer to a question about what would be a good outcome and the answer was along the lines of if we can keep deaths below 20,000 that would be a very good outcome.

It's a matter of interpretation, but if I was told that if we can keep deaths below 20,000 it would be a good outcome - I would assume that this is not the expected value of deaths, and the expected value is higher.

Also, to be picky, 500k is what the Imperial model said the worst case scenario is - however that was modelled on the premise that no government would actually do nothing. No government ever has done nothing in response to a pandemic and no government ever would.

The government changed their strategy because it also modelled that the number of deaths if we used a modified version of the flu pandemic strategy that we have always used before it would result in 250,000  deaths.

Their were lower figures modelled than 20,000 and 500,000 was the highest figure mentioned but realistically the best and worst case scenarios shown by the modelling were more like 20,000 to 250,000.


NB: It's not clear if this modelling would actually mean all deaths or all hospital deaths - worth mentioning as confirmed cases in hospitals is the standard measure so it is possible to think it might be on that basis. But it is unclear so it could still be seen as absolutely all deaths.

It’s a stretch to think they meant hospital deaths only, I haven’t seen it suggested by anyone, at any stage. This is a decent summary of what was being variously said at around the time:

https://www.spectator.co.uk/article/uk-coronavirus-cases-slowing-key-adviser-reveals

That link reports to figures about deaths from when the official figures were the hospital deaths.

But I was pretty much just talking about the origin - the Imperial Report. That says things like, "...Our model estimates these changes by calculating backwards from the deaths observed over time to estimate transmission that occurred several weeks prior, allowing for the time lag between infection and death. ..."

The "deaths observed" would be hospital deaths.

It is still possible to extrapolate from hospital deaths to all deaths but it doesn't state this and deaths in hospital confirmed by a test is the international standard - but - it also does not state that's what the numbers mean. As I said, it's unclear.

The original Imperial report is here.

https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf

It seems pretty clear to me, they are talking about total deaths in the 81% of the population they thought would be infected.

In the (unlikely) absence of any control measures or spontaneous changes in individual behaviour, we would expect a peak in mortality (daily deaths) to occur after approximately 3 months (Figure 1A). In such scenarios, given an estimated R0 of 2.4, we predict 81% of the GB and US populations would be infected over the course of the epidemic. Epidemic timings are approximate given the limitations of surveillance data in both countries: The epidemic is predicted to be broader in the US than in GB and to peak slightly later. This is due to the larger geographic scale of the US, resulting in more distinct localised epidemics across states (Figure 1B) than seen across GB. The higher peak in mortality in GB is due to the smaller size of the country and its older population compared with the US. In total, in an unmitigated epidemic, we would predict approximately 510,000 deaths in GB and 2.2 million in the US, not accounting for the potential negative effects of health systems being overwhelmed on mortality.

It clearly isn't a worst case as they specifically said that it doesn't account for the negative effects of health systems been overwhelmed, and the infection fatality rates aren't as high as others I have seen quoted.

I don't recall him talking about hospital deaths in the 7k to 20k figure either, as I remember saying something along the lines of that we were already close to 7k deaths at the time he said it.  It just seemed pretty odd at the time and looks even stranger with hindsight.  Why undermine your own results when there was little evcidence there were false at the time, and they are still satding the test of time as far as I see.

The Actuarial Profession's CMI total death figure was over 50k last week,and it now looks very likely that the right hand tail of the distribution is going to be longer than the left hand tail (ie more deaths after the peak than before".

[Jon MW]

...

 It seems, please correct me if I’m wrong, that you are attached to an idea that there is a big element of luck in how each country is being impacted.
...

Do you really think that there isn't?

The zoonotic jump to start wih comes from a random mutation that enabled the virus to leap species.

In the time before this new disease is identified there could be 0 people from your country in the affected area, or there could be a 1000.
There could be 0 or a 1000 travellers from somewhere else who then travel to your country shortly after.
At one time of the year there could be 0 or a 1000 in these categories so you get luck as to what time of year the virus first infects people; similarly it could change from year to year.

When it is identified the first thing countries do is test, track and trace arrivals from the affected area. But some will get through - that amount is essentially random. It might be 0% it might be 5% - it all makes a difference.

Outside of that test, track and trace program the virus is spreading anyway. The people infected might have a high number of people with natural immunity, it might have a high number of introverts or a high number of family orientated extroverts, it might have a high number of young fitness fanatics or a high number of obese pensioners (etc.)

There is nothing binary about these numbers it's a matter of weighting and probability - but more than anything else it's about randomness.

There are factors that come into play on top of that - like the size of your population, it's population density and the amount of international travel and entry points your country has - but all the numbers start with randomness.

And with exponential growth - random differences to the seed values make very large differences later.

Lots of randomness for sure. It would be difficult for a country that had no lockdown in place until there were 335 (the number was massively higher of course, on the lag) fatalities to cite the randomness and the unknown nature of early spread in its defence. The overwhelming correlations across all nations that are handling this well are comprehensive contact tracing (impossible for us) or comprehensive early lockdown, we are an international outlier amongst all comparable nations (US/Sweden/Belarus are obv a stretch in terms of being comparable) for choosing against the lockdown option.

How are we an outlier compared to France, Italy and Spain?

Awaits contortions of every possible fact and piece of data to show that the UK is the worst country in the world except the US of course

Or this:

https://www.bbc.co.uk/news/world-52103747

This still does the job, opportunity to act far earlier, acted significantly later, I think meets the criteria for 'outlier'. I'm happy to involve The Donald if we think comparisons are valid and reasonable.

Okay - so you were only talking about being an outlier in terms of when measures were taken. Not in the results.


EDIT: although that is only in lockdown measures. I haven't seen anywhere that compares travel restrictions and contact tracing like we were doing from January for a while (i.e. the other measures).

This is good, I'm having trouble fully interrogating it but it seems to have most things covered. Our hands were tied by lack of capacity for nearly everything we might have done effectively but whose fault do we reckon that was?

https://www.bsg.ox.ac.uk/research/research-projects/coronavirus-government-response-tracker

Let's give Boris, Trump and Bolsonaro a really generous interpretation of how unlucky they were at the start, then let's see if we can identify any commonalities in flawed response that mean they end up doing badly on all measures. Let's also give them (assuming they are still in power) economic growth until the end of 2022 as an extra measure to see how it plays out.

1. Total excess deaths as % increase of 5 year average all cause mortality in Europe + agreed reasonably comparable nations
2. % economic contraction in 2020 in G20.
3. % economic growth for US and Brazil by end 2022, % +- for the UK by end 2022, should be impossible wreck the economy that hard but I don't think it is for Boris.

I propose South Korea, Brazil and Japan as being suitable additions to the list for comparative excess mortality purposes. Perhaps the top ten, not already included, tourist destinations as well?

How are they in any way comparable?

If you want to compare countries I'd start with G8 for economics.
Top 10 tourist destinations as a proxy for international connectivity.
And filter out countries with either significantly different populations or significantly different population density.

All I’m trying to do is get the best possible list, South Korea seems fine imo. I’m happy for you to select all the additional countries though.

Is there a specific reason why we can’t extend to G20 for economics.
I think G20 for the economy and all of Europe for the % increase in all mortality, is a good starting point. Adjusting for population size (still debatable if that’s appropriate) is easy. The evidence is not entirely clear on pop density but NYC/London clearly have that in common.

South Korea - a country where 90% of their travellers are from China. Who have previously had experience of dealing with pandemics from China. Who get half the number of international travellers a year as Heathrow by itself, yeah sure go for it.

If you want to start with international travel and top 10 tourist destinations, maybe exclude population less than 20 million (that only loses Mexico). If we cut off any with population higher than 100 million that loses the US and China.

That leaves France, Spain, Italy, the UK, Germany, Thailand and Turkey.

Narrowing it down to G20 countries gives Germany, France, Italy, Turkey and the UK

Do you want to ignore population density? It's 280 per square km for the UK if you want a baseline, you could make a minimum and maximum cutoff around that if you do want to include it. But 5 might be a good number anyway.

[kukushkin88]

...

 It seems, please correct me if I’m wrong, that you are attached to an idea that there is a big element of luck in how each country is being impacted.
...

Do you really think that there isn't?

The zoonotic jump to start wih comes from a random mutation that enabled the virus to leap species.

In the time before this new disease is identified there could be 0 people from your country in the affected area, or there could be a 1000.
There could be 0 or a 1000 travellers from somewhere else who then travel to your country shortly after.
At one time of the year there could be 0 or a 1000 in these categories so you get luck as to what time of year the virus first infects people; similarly it could change from year to year.

When it is identified the first thing countries do is test, track and trace arrivals from the affected area. But some will get through - that amount is essentially random. It might be 0% it might be 5% - it all makes a difference.

Outside of that test, track and trace program the virus is spreading anyway. The people infected might have a high number of people with natural immunity, it might have a high number of introverts or a high number of family orientated extroverts, it might have a high number of young fitness fanatics or a high number of obese pensioners (etc.)

There is nothing binary about these numbers it's a matter of weighting and probability - but more than anything else it's about randomness.

There are factors that come into play on top of that - like the size of your population, it's population density and the amount of international travel and entry points your country has - but all the numbers start with randomness.

And with exponential growth - random differences to the seed values make very large differences later.

Lots of randomness for sure. It would be difficult for a country that had no lockdown in place until there were 335 (the number was massively higher of course, on the lag) fatalities to cite the randomness and the unknown nature of early spread in its defence. The overwhelming correlations across all nations that are handling this well are comprehensive contact tracing (impossible for us) or comprehensive early lockdown, we are an international outlier amongst all comparable nations (US/Sweden/Belarus are obv a stretch in terms of being comparable) for choosing against the lockdown option.

How are we an outlier compared to France, Italy and Spain?

Awaits contortions of every possible fact and piece of data to show that the UK is the worst country in the world except the US of course

Or this:

https://www.bbc.co.uk/news/world-52103747

This still does the job, opportunity to act far earlier, acted significantly later, I think meets the criteria for 'outlier'. I'm happy to involve The Donald if we think comparisons are valid and reasonable.

Okay - so you were only talking about being an outlier in terms of when measures were taken. Not in the results.


EDIT: although that is only in lockdown measures. I haven't seen anywhere that compares travel restrictions and contact tracing like we were doing from January for a while (i.e. the other measures).

This is good, I'm having trouble fully interrogating it but it seems to have most things covered. Our hands were tied by lack of capacity for nearly everything we might have done effectively but whose fault do we reckon that was?

https://www.bsg.ox.ac.uk/research/research-projects/coronavirus-government-response-tracker

Let's give Boris, Trump and Bolsonaro a really generous interpretation of how unlucky they were at the start, then let's see if we can identify any commonalities in flawed response that mean they end up doing badly on all measures. Let's also give them (assuming they are still in power) economic growth until the end of 2022 as an extra measure to see how it plays out.

1. Total excess deaths as % increase of 5 year average all cause mortality in Europe + agreed reasonably comparable nations
2. % economic contraction in 2020 in G20.
3. % economic growth for US and Brazil by end 2022, % +- for the UK by end 2022, should be impossible wreck the economy that hard but I don't think it is for Boris.

I propose South Korea, Brazil and Japan as being suitable additions to the list for comparative excess mortality purposes. Perhaps the top ten, not already included, tourist destinations as well?

How are they in any way comparable?

If you want to compare countries I'd start with G8 for economics.
Top 10 tourist destinations as a proxy for international connectivity.
And filter out countries with either significantly different populations or significantly different population density.

All I’m trying to do is get the best possible list, South Korea seems fine imo. I’m happy for you to select all the additional countries though.

Is there a specific reason why we can’t extend to G20 for economics.
I think G20 for the economy and all of Europe for the % increase in all mortality, is a good starting point. Adjusting for population size (still debatable if that’s appropriate) is easy. The evidence is not entirely clear on pop density but NYC/London clearly have that in common.

South Korea - a country where 90% of their travellers are from China. Who have previously had experience of dealing with pandemics from China. Who get half the number of international travellers a year as Heathrow by itself, yeah sure go for it.

If you want to start with international travel and top 10 tourist destinations, maybe exclude population less than 20 million (that only loses Mexico). If we cut off any with population higher than 100 million that loses the US and China.

That leaves France, Spain, Italy, the UK, Germany, Thailand and Turkey.

Narrowing it down to G20 countries gives Germany, France, Italy, Turkey and the UK

Do you want to ignore population density? It's 280 per square km for the UK if you want a baseline, you could make a minimum and maximum cutoff around that if you do want to include it. But 5 might be a good number anyway.

OK, let’s go with that 5 for now. We can make the broader comparisons as well, acknowledging that the whole G20/whole Europe comparison might be a less meaningful/valid comparison.

[Doobs]

...

 It seems, please correct me if I’m wrong, that you are attached to an idea that there is a big element of luck in how each country is being impacted.
...

Do you really think that there isn't?

The zoonotic jump to start wih comes from a random mutation that enabled the virus to leap species.

In the time before this new disease is identified there could be 0 people from your country in the affected area, or there could be a 1000.
There could be 0 or a 1000 travellers from somewhere else who then travel to your country shortly after.
At one time of the year there could be 0 or a 1000 in these categories so you get luck as to what time of year the virus first infects people; similarly it could change from year to year.

When it is identified the first thing countries do is test, track and trace arrivals from the affected area. But some will get through - that amount is essentially random. It might be 0% it might be 5% - it all makes a difference.

Outside of that test, track and trace program the virus is spreading anyway. The people infected might have a high number of people with natural immunity, it might have a high number of introverts or a high number of family orientated extroverts, it might have a high number of young fitness fanatics or a high number of obese pensioners (etc.)

There is nothing binary about these numbers it's a matter of weighting and probability - but more than anything else it's about randomness.

There are factors that come into play on top of that - like the size of your population, it's population density and the amount of international travel and entry points your country has - but all the numbers start with randomness.

And with exponential growth - random differences to the seed values make very large differences later.

Lots of randomness for sure. It would be difficult for a country that had no lockdown in place until there were 335 (the number was massively higher of course, on the lag) fatalities to cite the randomness and the unknown nature of early spread in its defence. The overwhelming correlations across all nations that are handling this well are comprehensive contact tracing (impossible for us) or comprehensive early lockdown, we are an international outlier amongst all comparable nations (US/Sweden/Belarus are obv a stretch in terms of being comparable) for choosing against the lockdown option.

How are we an outlier compared to France, Italy and Spain?

Awaits contortions of every possible fact and piece of data to show that the UK is the worst country in the world except the US of course

Or this:

https://www.bbc.co.uk/news/world-52103747

This still does the job, opportunity to act far earlier, acted significantly later, I think meets the criteria for 'outlier'. I'm happy to involve The Donald if we think comparisons are valid and reasonable.

Okay - so you were only talking about being an outlier in terms of when measures were taken. Not in the results.


EDIT: although that is only in lockdown measures. I haven't seen anywhere that compares travel restrictions and contact tracing like we were doing from January for a while (i.e. the other measures).

This is good, I'm having trouble fully interrogating it but it seems to have most things covered. Our hands were tied by lack of capacity for nearly everything we might have done effectively but whose fault do we reckon that was?

https://www.bsg.ox.ac.uk/research/research-projects/coronavirus-government-response-tracker

Let's give Boris, Trump and Bolsonaro a really generous interpretation of how unlucky they were at the start, then let's see if we can identify any commonalities in flawed response that mean they end up doing badly on all measures. Let's also give them (assuming they are still in power) economic growth until the end of 2022 as an extra measure to see how it plays out.

1. Total excess deaths as % increase of 5 year average all cause mortality in Europe + agreed reasonably comparable nations
2. % economic contraction in 2020 in G20.
3. % economic growth for US and Brazil by end 2022, % +- for the UK by end 2022, should be impossible wreck the economy that hard but I don't think it is for Boris.

I propose South Korea, Brazil and Japan as being suitable additions to the list for comparative excess mortality purposes. Perhaps the top ten, not already included, tourist destinations as well?

How are they in any way comparable?

If you want to compare countries I'd start with G8 for economics.
Top 10 tourist destinations as a proxy for international connectivity.
And filter out countries with either significantly different populations or significantly different population density.

But you then end up with no comparitors, because you can think of a reason to exclude everyone.  France, Spain, USA and Italy all have much lower population densities, but are the big cities really much different? CDG airport is similar to Heathrow in size.  Put Madrid and Barcelona together and you have something bigger than Heathrow and so on.  I don't think population density matters that much, as pretty much everywhere in the UK has Covid now.  I suspect nearly all reasonably sized population centres will have Covid cases in all these other countries too.

South Korea seems absolutely fine.  They may have less tourists, but they also got less of a notice period.  Taiwan has a similar population density, but were caught early, and have stronger links to China.  Hong Kong has much higher population density amd has more travel from China etc.  Japan has a lower official population density, but Tokyo is huge and the mountains mean that the population is really a lot more concentrated than other places.  I am not suggesting all these countries are great comparitors just examples to show we aren't amongst the best.

I don't think it matters,because we clearly haven't done that well. Shit happens and all that, just don't rock up and tell everyone that we made examplar preparations etc.  

I don't think you can use stats to show we are worse than Italy, Spain, USA or France just yet, as we are still a long way from supressing it and counting methods differ etc.  We are currently in a similar ballpark to all of them.  

[Marky147]

Can we start it from a new post, though

[Jon MW]

The other reason to use comparators is to narrow it down to a sensible list.

Comparing 5 countries which have plenty of things in common means there's not too much to look at and there's not too much to complain about them not being valid comparisons.

[Jon MW]

...
But you then end up with no comparitors, because you can think of a reason to exclude everyone.  ...

Which circles back to the argument that it's only of limited use to compare to other countries at all.

[kukushkin88]

The other reason to use comparators is to narrow it down to a sensible list.

Comparing 5 countries which have plenty of things in common means there's not too much to look at and there's not too much to complain about them not being valid comparisons.

Assuming my treatment restarts (I’m pretty sure it will), I will have an abundance of time in the months ahead. We shouldn’t let time needed to collate the data be a limiting factor in our study. I terms of how valid a measure it is, I would be sure it’s a better measure than nation specific modelled scenarios.

[Doobs]

...
But you then end up with no comparitors, because you can think of a reason to exclude everyone.  ...

Which circles back to the argument that it's only of limited use to compare to other countries at all.

I was suggesting that all this "we have bigger airports, more travel, we are more crowded etc" feels spurious.  A lot of these places are sufficiently similar that you can make useful comparisons, and learn a lot from those comparisons.  

We do this stuff all the time for all kinds of things like economic growth, education, healthcare, quality of life and so on.  

[Doobs]

Who is planning to go to a garden centre on Wednesday?

https://www.independent.co.uk/extras/indybest/house-garden/garden-centres-open-coronavirus-near-me-lockdown-uk-england-a9508251.html

Suspect it is going to be carnage on day 1 and this weekend.  Think of all the at-risk customers they have. 

Currently planning on a weekday next week for a hit and run tomato plant and growbag run and definitely need a plant or two to replace the things I got rid of in winter.  A lot of stuff can wait and can't see me doing too many visits.

[nirvana]

i have read that

it says

"You can exercise outside as often as you wish and from Wednesday 13 May, you can also sit and rest outside – exercise or recreation can be alone, with members of your household, or with one other person from outside your household, while keeping two metres apart at all times."

and

"It is still not permitted to leave your house to visit friends and family in their home. The government is looking at how to facilitate greater contact with close family or friends, and will explain how this can be done safely in the coming weeks."

From Wednesday I am allowed (thinking in my case as a non resident parent) to visit one member of another household in a park. If in that setting I socially distance from them, sitting or standing two metres away, why not two or three members of another but in the same household?

Yes, why not, I did this last Friday with my daughter and grandchildren in the playing field out back - way ahead of the curve

[Ledders]

Who is planning to go to a garden centre on Wednesday?

https://www.independent.co.uk/extras/indybest/house-garden/garden-centres-open-coronavirus-near-me-lockdown-uk-england-a9508251.html

Suspect it is going to be carnage on day 1 and this weekend.  Think of all the at-risk customers they have. 

Currently planning on a weekday next week for a hit and run tomato plant and growbag run and definitely need a plant or two to replace the things I got rid of in winter.  A lot of stuff can wait and can't see me doing too many visits.

I would say you have a fair chance of success turning up 45mins before closing on Thurs/Fri.

Everyone will go early doors and unlike supermarkets, do you need to be concerned about shelves being empty?

[Cf]

Sweet. Pubs etc open no later than July 4th according to the PM.

[Cf]

Very first question and they've not answered it. They've waffled around it a bit and common sense might say the answer is obvious but they've not actually answered the specific question asked.