Coronavirus - Data Analysis Thread
Discussion
Another way to look at this:
1. ONS says 1 in 340 in England would test positive via PCR. That equates to 0.29% prevalence.
2. LFD testing, which is mainly of asymptomatic people is showing positivity of around 0.1%
So this does initially support the idea that 1 in 3 infected are asymptomatic.
But the implication of this is that there are around 170k people currently infected of which 57,000 do not have symptoms. That does not square with us finding only 2,000 cases per day.
There should be 113k symptomatic people out there. That would require there to be a lot of people with symptoms who have not come forward for tests.
Something is off in these numbers.
1. ONS says 1 in 340 in England would test positive via PCR. That equates to 0.29% prevalence.
2. LFD testing, which is mainly of asymptomatic people is showing positivity of around 0.1%
So this does initially support the idea that 1 in 3 infected are asymptomatic.
But the implication of this is that there are around 170k people currently infected of which 57,000 do not have symptoms. That does not square with us finding only 2,000 cases per day.
There should be 113k symptomatic people out there. That would require there to be a lot of people with symptoms who have not come forward for tests.
Something is off in these numbers.
Elysium said:
That would require there to be a lot of people with symptoms who have not come forward for tests.
Something is off in these numbers.
Sounds like it.Something is off in these numbers.
Unfortunately there are too many variables there:-
- Is it that a large proportion of the population have had enough of all the messages (or never listened in the first place) and when they get symptoms are either ignoring them (part of the "man up" problem society has) or are treating at home - (hopefully) isolating / self-medicating because they've not got it that badly.
- Is it that the statistics aren't lining up because of some unidentified intereference in one of the data-sets?
- Is it that the virus is becoming less potent with certain demographics?
- Is it that prior (asymptomatic?) exposure is giving positive test results despite not currently having the virus?
Quick update.
1. Tests and cases. Testing volumes are steady week on week, but they have not been able to stop cases dropping back to the long term trend of 18 day halving. 7 day average cases for the last three weeks are 5088, 3166 and now 2365:
2. Key metrics. 18 day halving seems 'nailed on' for cases per 100k tests and admissions. Deaths have fallen faster halving every 14 days:
3. Key metrics overlaid. This is a solid predictable fall.
Each time we lift restrictions without a bounce back the conclusion must be that those restrictions were not controlling the virus:
1. Tests and cases. Testing volumes are steady week on week, but they have not been able to stop cases dropping back to the long term trend of 18 day halving. 7 day average cases for the last three weeks are 5088, 3166 and now 2365:
2. Key metrics. 18 day halving seems 'nailed on' for cases per 100k tests and admissions. Deaths have fallen faster halving every 14 days:
3. Key metrics overlaid. This is a solid predictable fall.
Each time we lift restrictions without a bounce back the conclusion must be that those restrictions were not controlling the virus:
Thanks as always, Elysium 
I think I saw a DM article proclaiming cases rose this past day or so?
EDIT: Here you go:
https://www.dailymail.co.uk/news/article-9462731/B...
No mention of the number of tests in that article, obviously...
re: this bit:
Do you think it could be fair to say that reducing social interaction (by whatever method) when the virus is moving into its winter seasonal increase (as per most viruses) could indeed delay infections somewhat (and therefore reduce pressure on the NHS if required) but that once that curve has peaked and is dropping, restrictions make no difference and the country might as well just open up with it and crack on?
It would seem to be the case here, although I recall that other countries have seen a substantial bounceback after opening up again?? (I need to do some work so can't check Worldometers right now!)
I think I saw a DM article proclaiming cases rose this past day or so?
EDIT: Here you go:
https://www.dailymail.co.uk/news/article-9462731/B...
DM said:
Are cases starting to rise after England's lockdown easing on March 29? UK's daily Covid cases go UP by nearly a third in a week to 3,568 while deaths halve to 13
...
Statisticians added another 3,568 positive tests to the official tally, up 29% on last week's count of 2,762
...
Today’s spike in infections can’t be blamed on hardy drinkers choosing to brave the snow and hailstones to celebrate the ‘Glorious Twelth’ in beer gardens across the country.
Instead, experts say it is ‘entirely possible’ the U-turn in the figures could be down to the Easter bank holiday. It takes up to seven days for people to show symptoms of Covid and get their test results back.
...
Statisticians added another 3,568 positive tests to the official tally, up 29% on last week's count of 2,762
...
Today’s spike in infections can’t be blamed on hardy drinkers choosing to brave the snow and hailstones to celebrate the ‘Glorious Twelth’ in beer gardens across the country.
Instead, experts say it is ‘entirely possible’ the U-turn in the figures could be down to the Easter bank holiday. It takes up to seven days for people to show symptoms of Covid and get their test results back.
No mention of the number of tests in that article, obviously...
re: this bit:
Elysium said:
3. Key metrics overlaid. This is a solid predictable fall.
Each time we lift restrictions without a bounce back the conclusion must be that those restrictions were not controlling the virus:
I believe you've previously suggested that the November(? October? whenever...) lockdown did have an effect, visible in that dip before Christmas and the subsequent bounce (which I presume looks similar to the Wales firebreak curve).Each time we lift restrictions without a bounce back the conclusion must be that those restrictions were not controlling the virus:
Do you think it could be fair to say that reducing social interaction (by whatever method) when the virus is moving into its winter seasonal increase (as per most viruses) could indeed delay infections somewhat (and therefore reduce pressure on the NHS
if required) but that once that curve has peaked and is dropping, restrictions make no difference and the country might as well just open up with it and crack on?It would seem to be the case here, although I recall that other countries have seen a substantial bounceback after opening up again?? (I need to do some work so can't check Worldometers right now!)
I also saw the Daily Mail article, which I think used reported cases. The piece itself actually mentions bank holiday reporting delays, but goes on the speculate that it ‘might’ be due to relaxation. If you look at cases by specimen date you realise it isn’t.
Lifting restrictions in lockdown 1 and 3 had no effect. Lifting lockdown 2 resulted in a sudden resurgence. So restrictions can work when the epidemic is growing, but there are times when it’s falls regardless of restrictions.
Lifting restrictions in lockdown 1 and 3 had no effect. Lifting lockdown 2 resulted in a sudden resurgence. So restrictions can work when the epidemic is growing, but there are times when it’s falls regardless of restrictions.
Update on the ONS deaths data published this morning:
1. All cause deaths in week 13 are below the 5 year minimum. In fact they are below the 10 year minimum. However, some caution is needed as we do not always have a bank holiday in that week. There were 400 deaths involving Coronavirus, but only 308 'due to' COVID. Which equates to 4% of the total weekly death toll:
2. These graphs compare deaths 'due to' COVID to excess deaths (deaths above the 5 year average) and statistically significant excess deaths (deaths above the average plus 2 standard deviations). The latter being the point at which we can say something 'unusual' is happening. Deaths have been at normal levels in the UK since the end of February. Which means that Coronavirus infections ceased to be a significant driver of mortality on or around the 5th February.
I think this now allows us to 'call' the end of the second wave and start to draw some comparisons:
1st Wave = 57,781 excess deaths of which 43,385 are statistically significant
2nd Wave = 48,547 excess deaths of which 21,655 are statistically significant
So the second wave is either 84% as big as the first or 50% as big depending on the comparator you use.
Worth noting also that since the second wave ended at the end of Feb there have been 4,380 less deaths than the 5 year average and 9,135 less than the 5 year maximum, which suggests that at least some of the second wave deaths are of people who were already near the end of their lives.
1. All cause deaths in week 13 are below the 5 year minimum. In fact they are below the 10 year minimum. However, some caution is needed as we do not always have a bank holiday in that week. There were 400 deaths involving Coronavirus, but only 308 'due to' COVID. Which equates to 4% of the total weekly death toll:
2. These graphs compare deaths 'due to' COVID to excess deaths (deaths above the 5 year average) and statistically significant excess deaths (deaths above the average plus 2 standard deviations). The latter being the point at which we can say something 'unusual' is happening. Deaths have been at normal levels in the UK since the end of February. Which means that Coronavirus infections ceased to be a significant driver of mortality on or around the 5th February.
I think this now allows us to 'call' the end of the second wave and start to draw some comparisons:
1st Wave = 57,781 excess deaths of which 43,385 are statistically significant
2nd Wave = 48,547 excess deaths of which 21,655 are statistically significant
So the second wave is either 84% as big as the first or 50% as big depending on the comparator you use.
Worth noting also that since the second wave ended at the end of Feb there have been 4,380 less deaths than the 5 year average and 9,135 less than the 5 year maximum, which suggests that at least some of the second wave deaths are of people who were already near the end of their lives.
Elysium said:
Worth noting also that since the second wave ended at the end of Feb there have been 4,380 less deaths than the 5 year average and 9,135 less than the 5 year maximum, which suggests that at least some of the second wave deaths are of people who were already near the end of their lives.
Whilst that's possible, I think it's not the significant driver.Far more likely is that seasonal flu just hasn't happened this year. Old folks haven't been mixing, and haven't been covered in much toddler snot so flu deaths are negligible this year.
Interesting Youtube update from Dr John Campbell today. It seems that the blood clotting issue has been know about for some years, I wonder if taking aspirin would reduce the risk. https://youtu.be/WuyAtvwP2H4
W124Bob said:
Interesting Youtube update from Dr John Campbell today. It seems that the blood clotting issue has been know about for some years, I wonder if taking aspirin would reduce the risk. https://youtu.be/WuyAtvwP2H4
My takeaway wasn’t aspirin. It was administer the vaccine correctly. Very interesting. Courtesy of the Daily Mail:
https://www.dailymail.co.uk/news/article-9474997/C...
"Public Health England (PHE) data shows that more than half of all people in the country are living in areas that are almost entirely free of coronavirus"
Good job we can't meet anyone indoors for weeks
https://www.dailymail.co.uk/news/article-9474997/C...
"Public Health England (PHE) data shows that more than half of all people in the country are living in areas that are almost entirely free of coronavirus"
Good job we can't meet anyone indoors for weeks
RSTurboPaul said:
Courtesy of the Daily Mail:
https://www.dailymail.co.uk/news/article-9474997/C...
"Public Health England (PHE) data shows that more than half of all people in the country are living in areas that are almost entirely free of coronavirus"
Good job we can't meet anyone indoors for weeks
And your analysis of the data is...?https://www.dailymail.co.uk/news/article-9474997/C...
"Public Health England (PHE) data shows that more than half of all people in the country are living in areas that are almost entirely free of coronavirus"
Good job we can't meet anyone indoors for weeks
spikeyhead said:
RSTurboPaul said:
Courtesy of the Daily Mail:
https://www.dailymail.co.uk/news/article-9474997/C...
"Public Health England (PHE) data shows that more than half of all people in the country are living in areas that are almost entirely free of coronavirus"
Good job we can't meet anyone indoors for weeks
And your analysis of the data is...?https://www.dailymail.co.uk/news/article-9474997/C...
"Public Health England (PHE) data shows that more than half of all people in the country are living in areas that are almost entirely free of coronavirus"
Good job we can't meet anyone indoors for weeks
Smollet said:
havoc said:
You just posted a Daily Mail article. At least have the decency to be embarassed!
Are you blaming the messenger because you don’t like the message? 
"The Daily Mail has been noted for its unreliability and widely criticised for its printing of sensationalist and inaccurate scare stories of science and medical research,"
Hardly the best viewed newspaper.
vaud said:
Smollet said:
havoc said:
You just posted a Daily Mail article. At least have the decency to be embarassed!
Are you blaming the messenger because you don’t like the message? "The Daily Mail has been noted for its unreliability and widely criticised for its printing of sensationalist and inaccurate scare stories of science and medical research,"
Hardly the best viewed newspaper.
Smollet said:
I get that but this isn’t an opinion they’ve dreamt up but what PHE have published unless I’m missing a parrot or something
First time they released all the restrictions they got it in the neck afterwards when we went into the second spike and it all shot skywards, killing tens of thousands more.Second time they relaxed things (Christmas) they had to backtrack because we saw another jump upwards.
...now they're being cautious and the Daily Fail is moaning that there's hardly any new cases.
Forgive me, but the word "hypocrisy" is suddenly popping to mind...
Finally, I would note that regional variations in lockdown levels didn't work very well last year, so exactly HOW are they going to do it at a far smaller UTLA level?
The reality is that the government is damned here no matter what they do, in addition to which the "couldn't care less" brigade have been ignoring many of the lockdown restrictions anyway, and were the ones that swamped the pubs at the weekend. Those who are rather more law (guidance?) abiding continue to be inconvenienced in the name of not making the same mistake a third time.
Personally, I'm not sure what to believe we should do anymore, as most of the statistics and press releases are coming from a government that time and time again has proven corrupt, self-serving and thoroughly incompetent. The vaccine 'win' is down to a gamble made last summer that the AZ vaccine WOULD be signed-off, and is the only sensible thing they've done...although of course it was a gamble with taxpayers money to yet another big corporate that's probably sponsoring several ministers.
vaud said:
Terminator X said:
^^ not spreading in the pubs though, our local has had not a single case of covid the entire time.
TX.
Not sure you be so assured given it can be asymptomatic? Plus those who get mild symptoms, write it off as a cold and never get tested.TX.
Gassing Station | News, Politics & Economics | Top of Page | What's New | My Stuff