We're starting to get more useful data:

Originally Posted by: Brian Gaze 

Questions mount over Russia's low mortality rate
From the BBC news ticker:
"Questions are mounting over Russia's extraordinarily low reported mortality rate - only 2,305 deaths with 252,000 infected. Moscow's department of health says it does not include most deaths of Covid-19 patients in official statistics because they had other potentially lethal illnesses. St Petersburg has reported an unexplained spike in deaths from pneumonia - five-and-a-half times the usual number and 10 times the official number of Covid-19 deaths in the city.Meduza, an independent Russian website, says official statistics are doctored and Russians are effectively “not allowed to die from coronavirus”. Many of those who find official numbers hard to believe keep their own tally.

Medical workers have their own list of colleagues they say have been killed by coronavirus. Even Orthodox priests have launched their own list "because official statistics are incomplete". Earlier, the Financial Times said Russia’s Covid-19 death toll could be 70% higher than official figure, but officials in Moscow dismissed such reports as "fake".

You can add a zero on that % figure, and then double it would be nearer the mark.

Originally Posted by: Phil G 

On my Zoe Covid-19 app has people with symptoms has plateaued now at 243,500 people currently with symptoms. This has to be expected at some point as we had not changed strategy for two months. We have now though so I would expect a creep upwards from now. Deaths will follow and reach their minimum in about two weeks time then plateau. Then we will be watching closely to see what happens next.

Originally Posted by: fairweather 

I think PhilG, Crewe, Maunder, four, westv, bleur, hickster, Essan and Duane will be in attendance

As many are pointing out, this is a Britain First front. pic.twitter.com/xf1T33UO5y

— Mike Stuchbery 💀🍷 (@MikeStuchbery_) May 14, 2020

Originally Posted by: The Beast from the East 

roughly 10,000 people a day 

The findings will inform the government’s next steps as it considers whether it is safe enough to further ease restrictions on socialising, businesses and schools in the coming weeks. Experts suggest the current rates of infection remain “some way off” what would be needed to lift the lockdown.

The results are likely to fuel concerns about the potential of opening primary schools on 1 June to fuel transmission in the community, as no evidence was found of differences in the proportions testing positive between the age categories 2 to 19, 20 to 49, 50 to 69 and 70 years and over.

Originally Posted by: Brian Gaze 

I don't do twitter. Beast, as an aside Eurovision is on Saturday in case you aren't aware. Know you love it!

Originally Posted by: Phil G 

I think PhilG, Crewe, Maunder, four, westv, bleur, hickster, Essan and Duane will be in attendance

Originally Posted by: The Beast from the East 

Don't you mix with up with that lot!

Originally Posted by: westv 

I don't think its a new one, just repeats?! 

I did love it, in the days of Bucks fizz and when Terry Wogan presented it

Originally Posted by: The Beast from the East 

I think PhilG, Crewe, Maunder, four, westv, bleur, hickster, Essan and Duane will be in attendance

As many are pointing out, this is a Britain First front. pic.twitter.com/xf1T33UO5y

— Mike Stuchbery 💀🍷 (@MikeStuchbery_) May 14, 2020

Originally Posted by: The Beast from the East 

As expected, the media don't understand what 100% accuracy and what 99.8% specificity really mean for the antibody tests. The lower the prevalence of the decease in the population, the higher the chance that the antibody test will return a false positive, even with a specificity rate of 99.8%.

I made a simple table to illustrate this for testing 10,000 people. If the true prevalence of the decease is 0.5%, then the antibody test developed by Roche has almost 30% chance of returning  a false positive for any single person. This risk drops to less than 1% only if the prevalence is 20%.

Originally Posted by: xioni2 

No, the number of new cases will not plateau due to any lack of change of strategy, any more than the number of cases plateaud before lockdown.

If this is true, it shows that the R value is now about 1, and prior to this month it was less than 1. In other words, there has been a slight change in behaviour, with people mixing with others slightly more. You could expect a further creep upwards from now, due to the recent loosening of the rules, which will make the R figure creep above 1.

Originally Posted by: Darren S 

I think PhilG, Crewe, Maunder, four, westv, bleur, hickster, Essan and Duane will be in attendance

As many are pointing out, this is a Britain First front. pic.twitter.com/xf1T33UO5y

— Mike Stuchbery 💀🍷 (@MikeStuchbery_) May 14, 2020

Originally Posted by: The Beast from the East 

Of the 22,332 people who died since 31st March in England

26% - 5,873 had diabetes

18% - 4,048 had dementia

15% - 3,254 had chronic pulmonary disease.

14% - 3,214 deaths had chronic kidney disease as an underlying health issue.

Originally Posted by: Gavin D 

Of the 22,332 people who died since 31st March in England

26% - 5,873 had diabetes

18% - 4,048 had dementia

15% - 3,254 had chronic pulmonary disease.

14% - 3,214 deaths had chronic kidney disease as an underlying health issue.

Originally Posted by: Gavin D 

 Daft question - I thought the specificity meant it's accuracy in determining this strain of coronavirus from others?

I can't see your table as too small but to work out an idea of how many false positives it may cause wouldn't you also need to know or at least guess how much of the sample already have other coronavirus antibodies, or is it accepted that everyone will have them due to common cold etc?

I.e it could only produce a false positive if that person had a strain of coronavirus antibodies.

Originally Posted by: Hippydave 

Daft question - I thought the specificity meant it's accuracy in determining this strain of coronavirus from others?

Originally Posted by: Hippydave