Covid : Infection rates v death rates and a possible second wave

[Bantamzen]

It's house parties and family gatherings that are being mostly implicated.

And factories, in the rush to blame young people this seems to have sunk under the radar.

 

[yorksrob]

It's house parties and family gatherings that are being mostly implicated.

Yes, I must admit, I'm a bit past it for house parties anyway.

There are certainly cases where they seem to be responsible for a few clusters - there have been a number of such incidents in the North East. That said, they won't be the only cause of the current rise, and the vast majority of pubs aren't seeing outbreaks.

There have definitely been a few. But nothing on the scale of work places, for example.

 

[Halish Railway]

2,919 new cases today. Nothing about the number of tests, so we still can’t calculate a positivity rate.

 

[Bletchleyite]

2,919 new cases today. Nothing about the number of tests, so we still can’t calculate a positivity rate.

175,687 today, but the 2919 is reported today so it won't be from those tests.

Coronavirus (COVID-19) in the UK

GOV.UK Coronavirus dashboard

coronavirus.data.gov.uk

 

[Steveoh]

175,687 today, but the 2919 is reported today so it won't be from those tests

That's the 2nd September figure, hover over the figure to see the date it's from.

 

[trebor79]

2,919 new cases today. Nothing about the number of tests, so we still can’t calculate a positivity rate.

175,687 today, but the 2919 is reported today so it won't be from those tests.

Coronavirus (COVID-19) in the UK

GOV.UK Coronavirus dashboard

coronavirus.data.gov.uk

The tests data hasn't been updated since last Wednesday (2nd September). Hospital admissions is also long out of date.
The cynic in me notes that they stopped updating the number of tests about this time of the huge increase in "cases"...

I do wonder, given the overwhelming negative light in which the government is seen, whether this missing data is a deliberate strategy.

 

[adc82140]

I'm curious to know why the positive results have been hovering around 2900 for most of this week, and not rising exponentially, as the narrative would have you believe.

 

[trebor79]

I'm curious to know why the positive results have been hovering around 2900 for most of this week, and not rising exponentially, as the narrative would have you believe.

Partly because of delays in reporting test results. In fact looking at the detailed data I can't see a single day above 2,500 by sample date...
I guess it's feasible there's lots of positive swabs waiting to be processed.
Or that the narrative is bollocks.

 

[bramling]

The problem there is that you are asking those groups of people to take all of the burden for the rest of society.
Not sure that is exactly a fair way of managing such a crisis.
Also remember some of those underlying conditions are fairly widespread (e.g. diabetes). I am not sure the government could afford to fully pay everyone who falls under those categories after they have spent so much on furlough etc!

Does that not intimate that perhaps it wasn’t the wisest of ideas to have spent billions furloughing non-vulnerable people, which has effectively put us not much further forwards, if we’re now in the position that we still need to support those who actually *are* vulnerable and now can’t afford to do so?

I can fully get the point about the burden falling on the elderly and vulnerable, however what’s the alternative? Likewise it has to be said the current elderly cohort have had a pretty generous hand of cards dealt to them thusfar in life, comparing particularly favourably to today’s younger cohort.

 

[Class 33]

So far, bit of a rise in the 7-day rolling average of deaths this week unfortunately. It's currently at 18.33. But hopefully some really low single digit figures for the remaining 4 days of the week to bring that average down a fair bit. The best we can expect now is an average of 8 to 9 by the end of the week.

Last week's rolling average was 7.57. And the week before that just slightly under 7.

But if we end the week with a rolling average of say 14, which is almost a 100% rise from last week, this isn't quite the same alarming scale of say an average of 600 per day deaths one week rising up to an average of 1150 deaths the next week.

 

[trebor79]

So far, bit of a rise in the 7-day rolling average of deaths this week unfortunately. It's currently at 18.33. But hopefully some really low single digit figures for the remaining 4 days of the week to bring that average down a fair bit. The best we can expect now is an average of 8 to 9 by the end of the week.

Last week's rolling average was 7.57. And the week before that just slightly under 7.

But if we end the week with a rolling average of say 14, which is almost a 100% rise from last week, this isn't quite the same alarming scale of say an average of 600 per day deaths one week rising up to an average of 1150 deaths the next week.

The 14 are spread over the past fortnight or so. You need to look at deaths by date of death, not by date reported to really understand what's going on. Highest daily deaths recently is 11, all other days single figures.

 

[adc82140]

At this low levels of death, there will be fluctuations as it's nothing more than background noise.

 

[Class 33]

The 14 are spread over the past fortnight or so. You need to look at deaths by date of death, not by date reported to really understand what's going on. Highest daily deaths recently is 11, all other days single figures.

Ah I see. I've been taking notice of the main figure on the Gov Coronavirus dashboard, and indeed this same number that gets mentioned as the number of deaths in the press and media. I expect most others have been too. This puts a different spin on it, that daily deaths aren't quite as high as made out to be. The deaths of 32 published on Tuesday - which the press and media made a bit of a fuss about the deaths going up a lot that day, well the true figure that day was only 8 then.

 

[LAX54]

It was stated today that Worldwide 900,000 had died with Covid, BUT...how many have died OF Covid ?

 

[Domh245]

Ah I see. I've been taking notice of the main figure on the Gov Coronavirus dashboard, and indeed this same number that gets mentioned as the number of deaths in the press and media. I expect most others have been too. This puts a different spin on it, that daily deaths aren't quite as high as made out to be. The deaths of 32 published on Tuesday - which the press and media made a bit of a fuss about the deaths going up a lot that day, well the true figure that day was only 8 then.

The caveat (as with most data relating to this) is that the accurate daily figures are subject to change. You get a more accurate representation, but you need to wait longer to get it. It's why death wise, I prefer the ONS weekly numbers (themselves already better than the Gov ones as it's deaths where COVID was a contributory factor, not the 28 day rule)

 

[takno]

The caveat (as with most data relating to this) is that the accurate daily figures are subject to change. You get a more accurate representation, but you need to wait longer to get it. It's why death wise, I prefer the ONS weekly numbers (themselves already better than the Gov ones as it's deaths where COVID was a contributory factor, not the 28 day rule)

The ONS numbers are where Covid was mentioned on the death certificate, so at best it's cases where Covid could conceivably have been a factor. In fact writing "definitely not Covid" would count as a Covid death

 

[adc82140]

The ONS numbers are where Covid was mentioned on the death certificate, so at best it's cases where Covid could conceivably have been a factor. In fact writing "definitely not Covid" would count as a Covid death

Also you have the death certificates that will give viral pneumonia as a cause of death (on average 80 people die of this a day) but will then go on to list a whole load of viruses that may or may not have caused it including SARS COV19 or whatever it's called. It would be remiss of a doctor not to.

 

[Domh245]

In fact writing "definitely not Covid" would count as a Covid death

Whilst an amusing image, it's actually done by the codes used on the certificate, not just "ctrl+f COVID"

 

[Dent]

So far, bit of a rise in the 7-day rolling average of deaths this week unfortunately. It's currently at 18.33. But hopefully some really low single digit figures for the remaining 4 days of the week to bring that average down a fair bit. The best we can expect now is an average of 8 to 9 by the end of the week.

Last week's rolling average was 7.57. And the week before that just slightly under 7.

But if we end the week with a rolling average of say 14, which is almost a 100% rise from last week, this isn't quite the same alarming scale of say an average of 600 per day deaths one week rising up to an average of 1150 deaths the next week.

First of all, how did you arrive at 18.33? By my calculation the current 7-day average is (14+8+32+3+2+12+10)/7=11.57.

Secondly, I'm a little confused about the "the remaining 4 days of the week" given that today is Thursday, meaning that your "week" starts on a Tuesday. Not most people's definition of a week.

Also, the current 7-day rolling is artificially high due to the 32 from Tuesday, which was high due to delayed reporting (as evident by looking at the date of death graph), so not really representative of the true picture. It actually looks like todays 14 may also have been artificially high due to delayed reporting, so again not representative of the true trend.

 

[Peter Mugridge]

I'm a little confused about the "the remaining 4 days of the week" given that today is Thursday, meaning that your "week" starts on a Tuesday. Not most people's definition of a week.

Tuesday is the start of the reporting week for the virus figures.

 

[Class 33]

Tuesday is the start of the reporting week for the virus figures.

Indeed, that's how I got the current average of 18.33, and hence including today we still have 4 remaining days of this week to get that average down much lower hopefully.

Otherwise, away from this I've always considered Saturday to be the start of the week.

 

[Dent]

Indeed, that's how I got the current average of 18.33, and hence including today we still have 4 remaining days of this week to get that average down much lower hopefully.

Otherwise, away from this I've always considered Saturday to be the start of the week.

I'm not sure what this "reporting week" is. The figures are reported daily, and the average is a 7-day rolling average.

In any case, your calcululation of the current 7-day rolling average was wrong. (14+8+32+3+2+12+10)/7 is 11.57, not 18.33.

 

[adc82140]

The "R" value is slightly above 1 and the ONS have said that cases have risen. Cue the hysteria from the Karens. However these figures lag by a couple of weeks, so surely the hospital admissions should be going up by now, and they are not.

 

[DB]

The "R" value is slightly above 1 and the ONS have said that cases have risen. Cue the hysteria from the Karens. However these figures lag by a couple of weeks, so surely the hospital admissions should be going up by now, and they are not.

Ah, but it's all young people at the moment and before you know it they will all be infecting their grannies...

(just predicting the likely response from the Karens!)

 

[Halish Railway]

3,539 new cases today, yikes.

On the positive side some of the testing statistics have been updated; 227,465 tests on Thursday.

 

[Domh245]

3,539 new cases today, yikes.

On the positive side some of the testing statistics have been updated; 227,465 tests on Thursday.

There does seem to be a bit more of a delay to today's results, at least for England who make up the bulk of cases- around 200 of them are from over a week ago, although 90% of them are from Monday-Wednesday

On a less positive note (ha!) however, the positive rate is definitely going up


As an aside, the 227465 figure is misleading as that includes all pillars - however positive cases are only published for pillars 1 and 2 (hospital/NHS and community testing respectively). Pillars 3 and 4 are antibody and surveillance testing, which results are not published for, although granted there's only ~20,000 or so per day.

 

[Bletchleyite]

I think this is now probably coming from school reopening. Which raises a big question about universities which are next...

 

[adc82140]

Schools have only been open for just over a week. Most of these results are from midweek. Still too soon. In addition hardly any are from school aged kids. It's the 18-50s. Social distancing has broken down, that's what. And you know what I'm going to say: masks are to blame, in that they give a false sense of security.

 

[bramling]

I think this is now probably coming from school reopening. Which raises a big question about universities which are next...

Did Scotland see any increase after their schools re-opened? As I understand it they went back earlier so that could provide an insight?

It still seems a bit early for it to be schools, some have barely been back a week.

One does wonder if a month of packed restaurants and cafes might have played a part, though one might have expected that to show through sooner.

We've been out and about these last two days and can already overhear Karen types advocating lockdown.

 

[adc82140]

We've been out and about these last two days and can already overhear Karen types advocating lockdown.

Wait till the end of October when furlough becomes potential unemployment. They'll soon change their tune.