Coronavirus

[Sugar Ape]

Wouldn’t surprise me if there was a drop in cases today if what Sturgeon says is happening in Scotland is replicated across the whole of the UK. 
 

 

[Sugar Ape]

3 minutes ago, Barrington Womble said:

Fuck, it's just rising daily isn't it? 

The app is one of the reasons I can’t be arsed with all the ‘cases are rising because of false positives‘ or ‘but we’re doing more testing’ brigade. 
 

It’s more than doubled on the app in a short space of time and that is just measuring how many people have got actual symptoms. Think you’ve been looking at it longer than I have but I understand it’s been pretty accurate throughout this?

[Strontium]

Isn't there a cold going round?

[Sugar Ape]

 

 

[Barrington Womble]

5 minutes ago, Sugar Ape said:

The app is one of the reasons I can’t be arsed with all the ‘cases are rising because of false positives‘ or ‘but we’re doing more testing’ brigade. 
 

It’s more than doubled on the app in a short space of time and that is just measuring how many people have got actual symptoms. Think you’ve been looking at it longer than I have but I understand it’s been pretty accurate throughout this?

I've been on it since the start. At first they were making the numbers purely based on symptoms (and some calculations based on how many were likely to have it or not), but since they've been able to be involved in the testing programme, they're able to create an improved data model to increase the accuracy. This is why you see in the chart you've posted that the cases for instance in the NW are 77-208 new cases per million as that's their margin of error. That's changed throughout the process depending how they've been able to access data and as their modelling has improved. I think that 4220 is based on the mid point of all of those numbers. As you can see it brings a large margin of error. 

[Sugar Ape]

6 minutes ago, Barrington Womble said:

I've been on it since the start. At first they were making the numbers purely based on symptoms (and some calculations based on how many were likely to have it or not), but since they've been able to be involved in the testing programme, they're able to create an improved data model to increase the accuracy. This is why you see in the chart you've posted that the cases for instance in the NW are 77-208 new cases per million as that's their margin of error. That's changed throughout the process depending how they've been able to access data and as their modelling has improved. I think that 4220 is based on the mid point of all of those numbers. As you can see it brings a large margin of error. 

Right got you, so unlikely to be rising just because there’s a ‘cold’ going around then. Considering the amount of positive results we get go up broadly inline with the rise on the app. 
 

Seems pretty safe to say that if people are getting close enough to be passing a cold on then they’re also close enough to be spreading Covid. 

[Barrington Womble]

Just now, Sugar Ape said:

Right got you, so unlikely to be rising just because there’s a ‘cold’ going around then. Considering the amount of positive results we get go up broadly inline with the rise on the app. 
 

Seems pretty safe to say that if people are getting close enough to be passing a cold on then they’re also close enough to be spreading Covid. 

Yeah, it's not rising because of a cold. That's the interesting thing I think, if there's a cold going around, how could covid not be on that same cycle of infection?

[Spy Bee]

15 minutes ago, Barrington Womble said:

Yeah, it's not rising because of a cold. That's the interesting thing I think, if there's a cold going around, how could covid not be on that same cycle of infection?

How do you know it's not?

[Barrington Womble]

10 minutes ago, Spy Bee said:

How do you know it's not?

Because the Zoe modelling is not based purely on symptoms, it's based from testinf. And I assume our testing can tell the difference between a cold and covid19. Unless you think our tests are that bad I can pop in with a hangover and be given a positive covid19 result? 

[Bjornebye]

 

 

 

[Spy Bee]

10 minutes ago, Barrington Womble said:

Because the Zoe modelling is not based purely on symptoms, it's based from testinf. And I assume our testing can tell the difference between a cold and covid19. Unless you think our tests are that bad I can pop in with a hangover and be given a positive covid19 result? 

So that's how you can analyse the spread of Covid. How can you compare that to the spread of a cold?

 

I presume flu is spreading faster, seen as it's killing ten times as many people as Covid is?

[Bjornebye]

10 minutes ago, Spy Bee said:

So that's how you can analyse the spread of Covid. How can you compare that to the spread of a cold?

 

I presume flu is spreading faster, seen as it's killing ten times as many people as Covid is?

Did you make the last line up? 

[Tony Moanero]

Spy Bee is still posting complete and utter bollocks in this thread?

[Spy Bee]

ONS figures lads.

[Spy Bee]

*flu and pneumonia, sorry.

[Sugar Ape]

12 minutes ago, Bjornebye said:

Did you make the last line up? 

It’s Flu and pneumonia deaths. Fullfact says the claim is a load of shite. 
 

https://fullfact.org/health/flu-covid-deaths/
 

Several newspapers, magazines and broadcasters—including the Sun, the Mirror, the Mail, the Week, the Spectator, the Times and talkRADIO—as well assome social media accounts, have claimed that influenza (flu), or “influenza and pneumonia” is now causing more deaths than Covid-19 in the UK.

This is a misunderstanding of figures from the Office for National Statistics (ONS), which counts deaths in England and Wales. 
 

These figures show the number of deaths where influenza, pneumonia or Covid-19 are mentioned on the death certificate, not those where they were listed as the underlying cause of death. This makes it misleading to say that these are cases where people necessarily died “of” the disease, or were  “killed by” it, because in many cases influenza and pneumonia are only mentioned as factors in a death where the underlying cause was something else.

 

We don’t know exactly how many people are dying of flu at the moment—meaning that “flu” is the underlying cause—but it is almost certainly far fewer than these reports suggest.

 

With pneumonia included, the total number of people dying of pneumonia or flu is probably higher than the number dying of Covid-19 at the moment, but not by the margin that these headlines suggest.  

How deaths are categorised

Death certificates give doctors space to record several different diseases or conditions that contributed to someone’s death, but only one can be recorded as the underlying cause.  The underlying cause in this context means “the disease or injury which initiated the train of morbid events leading directly to death”. 

 

For example, someone might develop cancer, which leads to other conditions, from which they ultimately die. All of these conditions might be mentioned by a doctor on the person’s death certificate, but the cancer would be listed as the single underlying cause. It would be right to say that this person “died of” or “died from” cancer, and it would be misleading to say that they died of or from something else.

 

In its bulletin on the number of deathsbeing registered in England and Wales each week, the ONS reports the number of “deaths involving COVID-19”, which means the number of death certificates that mention Covid-19 somewhere, whether or not it was the underlying cause.

 

The ONS also reports in the bulletin the number of “deaths in England and Wales [that] mentioned ‘Influenza and Pneumonia’”. Again, they didn’t have to be the underlying cause of death. 

 

This matters, because when Covid-19 is mentioned on a death certificate, it is much more likely to be the underlying cause of someone’s death than when pneumonia or influenza is. 

 

Up to the end of June 2020 in England and Wales, Covid-19 was the underlying cause of death on about 93% of the death certificates that mentioned it. In 2015, 2016 and 2017, however, influenza or pneumonia were listed as the underlying cause of death on just 28% of the death certificates in England that mentioned either of them.

“Or”, or “and”, or “and/or”?

Another problem is that the reporting often summarised these deaths involving “influenza and pneumonia” as being deaths from “flu”, which is incorrect.  

 

The phrase “influenza and pneumonia” used by the ONS may have caused confusion here, because it sounds like it refers to death certificates that mention both, and so could be simplified to just “flu”. In fact, it refers to any mention that belongs in the “influenza and pneumonia” category.

 

This means that a death counted in the “Influenza and pneumonia” category could be someone who died after having pneumonia, or after having flu, or after having both.

 

Pneumonia is usually caused by a bacterial infection, but it can also be caused by a virus such as flu or the one that causes Covid-19. It’s worth noting that many people who die with Covid-19 have pneumonia as well, so the ONS explains that “deaths where both were mentioned have been counted only in the COVID-19 category”.

What does this mean for flu?

As we have seen, most of the people with influenza or pneumonia mentioned somewhere on their death certificate probably did not die with either as the underlying cause. And among those who did, the underlying cause was especially unlikely to be flu.

 

Flu is an important factor in many deaths. Estimates from the FluMOMO model suggest that there have been between 4,000 and 22,000 deaths associated with flu in England in each of the past winter flu seasons.

 

However, it is not often considered to be the underlying cause of death by doctors. 

If we look at the underlying cause of death data for 2019, we can see that 1,213 people died specifically from flu, which is just 4.6% of the 26,342 who died from either influenza or pneumonia.

 

In other words, on average, about 23 people died each week last year and had flu identified as the underlying cause of death (although this is likely an undercount and the disease may have been an important factor in many other deaths). 

Flu levels are unusually low right now, even for the summer. And the number of death certificates mentioning influenza or pneumonia is currently below the five-year average, as the ONS graph reveals. 

So what are the latest figures?

In the week ending 7 August, which is the latest reported by the ONS at the time of writing, 152 death certificates mentioned Covid-19.

 

If the proportion actually caused by Covid-19 is the same as it was in June (a more recent figure isn’t available), then about 93% of these deaths will have Covid-19 as their underlying cause, meaning about 141 of them. 

 

Over the same period there were 1,013deaths registered with influenza or pneumonia, but not Covid-19, mentioned on the death certificate. If 28% of these had influenza or pneumonia as their underlying cause, which was the rate from 2015 to 2017, then this will make a total of 284 deaths from either cause. 

 

If we’re looking just at flu and 4.6% of those people who died had flu as the underlying cause identified on the death certificate, that would make just 13 people "killed by flu” in the same week. Though, as mentioned, this is likely an undercount.

 

These numbers are rough estimates. But they show that while pneumonia is probably killing more people than Covid-19 at the moment, there isn’t evidence that flu itself is killing more people than Covid-19.

 

 

 

 

[Spy Bee]

8 minutes ago, Sugar Ape said:

These figures show the number of deaths where influenza, pneumonia or Covid-19 are mentioned on the death certificate, not those where they were listed as the underlying cause of death. This makes it misleading to say that these are cases where people necessarily died “of” the disease, or were  “killed by” it, because in many cases influenza and pneumonia are only mentioned as factors in a death where the underlying cause was something else.

 

We don’t know exactly how many people are dying of flu at the moment—meaning that “flu” is the underlying cause—but it is almost certainly far fewer than these reports suggest.

 

If you are going to apply that to flu, you would obviously need to apply it to Covid also. 

 

The ONS states that 12.6% of people had flu/pneumonia when they died, and 1.1% of people had Covid for the last recorded and published week, which I believe was a fortnight ago.

 

I don't see how this differs? If we had a running total of people dying with flu and pneumonia every day, then the numbers would be a lot higher than those dying with Covid.

 

EDIT to clarify... I mean, it would be mad to think that nobody was dying with Covid, and that Covid was always the cause of death when somebody with Covid died. The figures count all causes of death up to 28 days after a positive test.

[Sugar Ape]

10 minutes ago, Spy Bee said:

If you are going to apply that to flu, you would obviously need to apply it to Covid also. 

 

The ONS states that 12.6% of people had flu/pneumonia when they died, and 1.1% of people had Covid for the last recorded and published week, which I believe was a fortnight ago.

 

I don't see how this differs? If we had a running total of people dying with flu and pneumonia every day, then the numbers would be a lot higher than those dying with Covid.

 

EDIT to clarify... I mean, it would be mad to think that nobody was dying with Covid, and that Covid was always the cause of death when somebody with Covid died. The figures count all causes of death up to 28 days after a positive test.

The ONS don’t go off the 28 day rule though do they, just the death certificate if it mentions Covid? And because Covid is more deadly than flu a lot more of the mentions of Covid on the death certificate are the actual cause of death than flu.
 

I mean it says deaths caused by flu only and not pneumonia or something else was just 4.6% of all mentions on a death certificate for the period they looked at. All seems pretty clear to me the way the article is laid out. 
 

The claim that flu is killing ten times the amount of people than Covid is clearly false anyway. 

[Barrington Womble]

1 hour ago, Spy Bee said:

So that's how you can analyse the spread of Covid. How can you compare that to the spread of a cold?

 

I presume flu is spreading faster, seen as it's killing ten times as many people as Covid is?

I am not analysing anything, Kings College London are. 

[Spy Bee]

Okay, might have skipped over the detail a bit on that one. But back to my original point, shouldn't more people be dying with flu, than Covid?

[TheHowieLama]

1 minute ago, Spy Bee said:

Okay, might have skipped over the detail a bit on that one. But back to my original point, shouldn't more people be dying with flu, than Covid?

No.

 

ONS also shows 52k in their Covid numbers - significantly more than the Hopkins numbers.

The data from Public Health on flu death rates is back a few pages from when Stronts first tried this tactic.

Average deaths from flu for the last 5 years is like 17k or something along those lines.

[Spy Bee]

I haven't got time to get into the detail just now. I'll have a look tomorrow. I must be misinterpreting something.

[Colonel Bumcunt]

Cases in UK down 300 on last Monday.

 

I think that's encouraging, though give it another week before judging.  Schools not long been back, that's the litmus test.  

[lifetime fan]

15 minutes ago, Colonel Bumcunt said:

Cases in UK down 300 on last Monday.

 

I think that's encouraging, though give it another week before judging.  Schools not long been back, that's the litmus test.  

It’s the universities starting back and freshers week that will cause the problem rather than primary schools. 

[Bruce Spanner]

1 minute ago, Colonel Bumcunt said:

Cases in UK down 300 on last Monday.

 

I think that's encouraging, though give it another week before judging.  Schools not long been back, that's the litmus test.  

Limited testing leading to lesser cases being identified.