Coronavirus

[Count Boso]

The Swedish epidemiologist Johan Giesecke is quoted as saying 'that “people aren’t stupid” and that if you tell them it’s important to keep their distance or wear masks or wash their hands, they will'.

Is he correct?

[And, yes, where is Belgium? (I don't mean geographically)]

[LMcD]

Might that be the time lag due to the bank holiday/weekend factor again? Hope we don't see an increase tomorrow but my bet is that, sadly, we will.

There's a definite dip in the figures most weekends, and the Bank Holiday on Friday could mean a noticeable rise in the next couple of days.

[johnb]

Belgium is highest of all, but isn't on this list.

The figures I have from Johns Hopkins University are slightly different to those shown in the chart - but it all depends on how the data is sourced and what is included, etc.

Having said that, using the Johns Hopkins data Belgium would come in at 10.1.

But it is a bit unfair on poor old Belgium because they include all care home deaths were Covid-19 is suspected but the UK (and some other European countries) only includes care home deaths where Covid-19 has been confirmed through testing (and we know what care home testing has been like in the UK).

[Dave2002]

Shouldn't we spare a thought for the poor police trying to enforce the new 'rules'? How the blazes will they tell if somebody is breaching them? And if they do nick somebody, it's a fundamental principle of our law that offences are meant to be clear, and any ambiguity must be read in the defendant's favour. Defence lawyers should have a ball!

If you can afford one of course...

If one watched Parliament today, another bright idea is the power scale for fines - if the police catch people and the "system" keeps up. Seems the intention is £100 for the first offence - £200 for the next, then £400, than £800, then £1600 and finally £3200. Or so BJ said.

I'm not sure that's workable, or if it might be open to serious legal challenge.

[johnb]

There's a definite dip in the figures most weekends, and the Bank Holiday on Friday could mean a noticeable rise in the next couple of days.

Added to that - today's figure of 210 is lower than what it should be because it includes a correction for double counting of deaths. It is 19 lower than the UK Hospital deaths and on average it is around 130 higher (though the difference varies wildly).

I wouldn't be surprised if the true current figure for deaths of people confirmed to have Covid-19 is around 400.

[Simon B]

Belgium is highest of all, but isn't on this list.

I have no brief nor wish to defend the UK govt or diminish the severity of the impact on the UK so far.

However, this graph seems to be in wide circulation currently. The omission of Belgium is a glaring example of bad statistics. It would be above the UK at 7.1/Mil on the same terms.

Elsewhere it's having various inferences drawn from it which it doesn't and can't in itself support. The overall per-nation toll of the pandemic could simultaneously have the exact inverse profile of that in this graph. A 7-day average across many nations whose epidemics started with skews much greater than 7 days is meaningless other than as a "where's the worst toll right now" indicator. Roll this window back a few weeks and the worst group would have their order interchanged. If it turns out to be a predictor of the longer term averages that will be barely more than coincidence.

*There seems to be good reason to believe that Belgium's figures look as bad as they do because of the particularly "generous" way they attribute deaths to CV-19 compared with others on the list.

The methods and conservatism or otherwise of the way the raw figures are arrived at by different nations is such that fine-grained comparisons are statistical delusions at this stage. All of that said, the UK is clearly not faring well.

Personally, I think the comparative excess death graph ensembles given here https://www.ft.com/content/a26fbf7e-...3-955839e06441 give a much more meaningful and nuanced insight overall, albeit at the price of inevitably greater timelags on the data.

[LeMartinPecheur]

There's a definite dip in the figures most weekends, and the Bank Holiday on Friday could mean a noticeable rise in the next couple of days.

LMcD: I've been analysing the daily Reported Cases figures from the .gov coronavirus website for 14 top-tier local authorities (mainly south-coast) for many weeks now. I can't identify any clear pattern such as you describe.

[johnb]

Personally, I think the comparative excess death graph ensembles given here https://www.ft.com/content/a26fbf7e-...3-955839e06441 give a much more meaningful and nuanced insight overall, albeit at the price of inevitably greater timelags on the data.

... and that suggests that the UK excess mortality is currently of the order of 60,000.

[burning dog]

So if I was to set off in my car at 00:01 and drive the 3 to 4-hour journey to my favourite west country beach, and having greeted the dawn there and enjoyed the day, heading back at around 20:00 to get home by midnight, that would be just fine. That is, indeed, just what the good people of Devon would welcome with open arms.

but the proposed trip from Cornwall to Berwick upon Tweed would be a tall order. I don't know what the official view on using a helicopter would be. I'm certain Boris and his experts considered all of this though.

Are you allowed a short drive so you can go for a walk in Wales from Hay-on-Wye? What about the other way round?

[Dave2002]

Personally, I think the comparative excess death graph ensembles given here https://www.ft.com/content/a26fbf7e-...3-955839e06441 give a much more meaningful and nuanced insight overall, albeit at the price of inevitably greater timelags on the data.

The FT link you have provided is interesting, though parts of the article(s) are not quite as clear as I'd like.

What seems to be of interest though is that, unless I've totally misunderstood this - which is possible, the use of excess death data makes the whole picture look a lot worse.

In the press conferences it has been suggested that excess death data is more accurate (which it may be) but it has also been suggested that during the current crisis in the UK at least, that some of the excess deaths are not directly due to Covid 19. There have also been suggestions that there will be a lag in some excess deaths, for example due to cancer patients not getting adequate treatment for a period of maybe a month or a slightly more.

From the FT articles I now wonder whether this spin on events has been an attempt to obfuscate and minimise the real damage caused by the virus. The excess death measures do indeed show a much higher figure than generally reported for CV deaths. There could be several reasons for this. The simplest reason is simply that some deaths are due to CV, but not being recorded as such - an under reporting of CV related deaths.

This would probably not be deliberate, but a consequence of the administrative reporting procedures. Another possibility is that there is an increase in events, such as accidents, leading to deaths, and these are not at all CV-19 related. Yet another is that the measures taken against CV have diverted resources away from health services, thus not saving some people's lives, when they could have been treated. Diverted resources could include medical support, or ambulance drivers not being available to reach people having heart attacks.

I am sure there are many ways to look at this. Up to now I have at least partially accepted the arguments put forward by the chief medical office and other health experts that CV 19 deaths are not telling the whole story, and that CV-19 is not the only issue.

Now I'm wondering if, although there is undoubtedly likely to be some truth in that, that this line of argument has in fact been a smoke screen. The size of the death spikes in most countries which have experienced CV-19 seems to be so large that trying to argue that some of the excess deaths are due to other causes is not reasonable, as the discrepancies are so big.

[LMcD]

LMcD: I've been analysing the daily Reported Cases figures from the .gov coronavirus website for 14 top-tier local authorities (mainly south-coast) for many weeks now. I can't identify any clear pattern such as you describe.

I don't know what the status of The Health Foundation is, but they comment as follows: 'There is some day-to-day variation in estimates, particularly with lower counts on a Sunday and Monday, reflecting delays in the reporting of deaths over the weekend'

This seems to back up the following from BBC News: Many hospitals will not report deaths that happened over the weekend until the middle of the following week'

One of the health experts at a recent Downing Street briefing, commenting on the latest graph for coronavirus deaths, said 'The weekend effect is clearly visible'.

Whether or not there's a spike after this last weekend, let's hope the overall downward trend continues - it's the rolling 7-day average that I shall be keeping an eye on.

[Petrushka]

Excellent article in today's Guardian from Devi Sridhar. This should be required reading for Johnson.

https://www.theguardian.com/commentisfree/2020/may/12/tories-lockdown-social-distancing-testing-second-wave-coronavirus

[LMcD]

Excellent article in today's Guardian from Devi Sridhar. This should be required reading for Johnson.

https://www.theguardian.com/commenti...ve-coronavirus

Thank you - as you say, it should be required reading in certain quarters. Perhaps it could be reprinted in the Torygraph to increase BJ's chances of spotting it while taking a break from planning his schedule of peak-time TV appearances.

[Count Boso]

Are you allowed a short drive so you can go for a walk in Wales from Hay-on-Wye? What about the other way round?

No, not if you drive (but if you walk you may get round the law). The Welsh government has announced Keep out of Wales, and says that in Wales the Welsh law is the law. A problem for people whose work takes them across the border, but that's probably provable, whereas taking a walk would be a no-no.

"Earlier, the Welsh minister responsible for the coronavirus recovery, Jeremy Miles, said rules "do not permit people to get in their car and drive to destinations in Wales".

"And that also means people getting in their cars in England," he added, saying police in Wales "absolutely have the power to fine" people for making non-essential journeys.

https://www.bbc.co.uk/news/uk-wales-52614204

[Pulcinella]

Two gloomy links from our choir director.

The first is about the Concertgebouw Chorus (a Lebrecht article) and the second considers future music making (especially for choirs):

https://slippedisc.com/2020/05/concertgebouw-chorus-is-devastated-after-pre-covid-bach-passion/?fbclid=IwAR2SrXqZv15wO2KPsIpPLq0_IqS-Adojo2kaQ924KUAsiA9uiWb1PEQwoTA

https://www.pressherald.com/2020/05/09/choirs-may-have-to-remain-silent-long-after-society-reopens/?fbclid=IwAR0jFtOjQXQbF25iXF_hOItjramAb01TcyBTR0xLUvHW8e35Wt0E5wgl7Vc#