Daily "cases"

Shotclog
Posts: 15
Joined: Mon Feb 15, 2021 11:37 am

Daily "cases"

Post by Shotclog »

For the last two weeks, possibly more, the daily "cases" reported by the Government have been consistently at or about 3,500-5,000 or so. No real increase, no real decrease.

How long before they wake up to the fact that this figure is well within even the most conservative estimate of the false positive rate for the PCR tests?

The pandemic is long over in Britain, but the Government and its cabal of Druids just can't bring themselves to admit it.

Splatt
Posts: 1436
Joined: Thu Oct 29, 2020 12:46 am

Re: Daily "cases"

Post by Splatt »

Shotclog wrote: Thu Apr 01, 2021 5:26 pm How long before they wake up to the fact that this figure is well within even the most conservative estimate of the false positive rate for the PCR tests?

The pandemic is long over in Britain, but the Government and its cabal of Druids just can't bring themselves to admit it.
Firstly its not. PCR is thought to be running in the region of 0.1% or so.
Also the majority of tests done in the UK currently are LFD not PCR.

Perfectly possible we've reached an equilibrium for the number of cases vs the degree of social interraction.

burke19
Posts: 49
Joined: Fri Dec 18, 2020 7:32 pm

Re: Daily "cases"

Post by burke19 »

Splatt wrote: Fri Apr 02, 2021 12:31 am
Shotclog wrote: Thu Apr 01, 2021 5:26 pm How long before they wake up to the fact that this figure is well within even the most conservative estimate of the false positive rate for the PCR tests?

The pandemic is long over in Britain, but the Government and its cabal of Druids just can't bring themselves to admit it.
Firstly its not. PCR is thought to be running in the region of 0.1% or so.
Also the majority of tests done in the UK currently are LFD not PCR.
What is "not"? Is it the pandemic that is "not" long over? Or is it that the positive tests are "not" in line with the error rate? As you know something about testing, and with around 3/4 of around 1.3 million daily tests in the UK being LFD, why not give the error rate of the LFD tests?
I know nothing about testing error rates other than to guess that the LFD error rate is higher than that of the gold standard PCR.

Shotclog
Posts: 15
Joined: Mon Feb 15, 2021 11:37 am

Re: Daily "cases"

Post by Shotclog »

Yes, I'm sorry Splatt, but I didn't understand your response.

You appear to give a figure of 0.1% for false positives for the PCR test-that is by some measure the lowest estimate I have ever seen. The last official figures I could find suggested it was more like 1-4%, but like everything else this Govt does it offered no certainty about that. If I have missed more recent research please can you point me to that?

As for the lateral flow tests, do you know what % of the total tests every day are LFTs? I don't, but again please let me know where that figure can be found and also the best evidence for the false positive rate for this test.

Total tests were running as high as 2 million a day, sometimes more. Assume 1.5m tests-if the false positive rate is 0.1% of that is 1500. More realistically, false positives are unlikely to be running at less than 1%, which with these figures would produce 15,000 false positives a day, many more than are currently being reported on a daily basis.

miahoneybee
Posts: 1288
Joined: Tue Sep 22, 2020 8:26 pm

Re: Daily "cases"

Post by miahoneybee »

They dont want it to be over shotclog..they know the figures show there is no reason to continue.. they must keep it all going to push the agenda behind all this..
More people are waking up to their lies and deceit...
I note bromley council have signs up in their parks..." this is a covid area keep 2m apart" ..what utter bollox..how do they know that..?have they tested the air for rona then? Could it be to do with the stand in the parks..or god forbid people enjoying the sunshine including children enjoying the play area. Oh wait Kent council have gated up their play area so the kids can't play and hired 5 ( yes 5) security guards ...
It's not about the rona it's about moving more freedoms and moving the goalposts so people will beg to take an experimental vaccine ( with a lovely new passport to get your freedoms) in the hope that we can get back to normal..
:roll: :roll: :oops: :cry:

Splatt
Posts: 1436
Joined: Thu Oct 29, 2020 12:46 am

Re: Daily "cases"

Post by Splatt »

Shotclog wrote: Wed Apr 07, 2021 5:21 pm Yes, I'm sorry Splatt, but I didn't understand your response.
Its not that complicated.

You appear to give a figure of 0.1% for false positives for the PCR test-that is by some measure the lowest estimate I have ever seen.
Its official estimates [1] also combined with tens of millions of results dating back a year.
Just take one look at the summer figures and positivity rates for a start in places with low prevalence.
The last official figures I could find suggested it was more like 1-4%,
There has never been any official figure putting it at 1-4%. Not one.
Where are you getting that "official" figure from?
Why does it not agree with reality?

As for the lateral flow tests, do you know what % of the total tests every day are LFTs?
I don't, but again please let me know where that figure can be found
Well if you bothered reading the actual official .gov coronavirus dashboard you'll find those figures have been published for months now and are updated on a daily basis.
And the ONS summaries published weekly.
and also the best evidence for the false positive rate for this test.
Again all documented officially on .gov documents on their own site and also in the ONS summaries.

Total tests were running as high as 2 million a day, sometimes more.
Again why are you guessing? Official figures are published daily.
Assume 1.5m tests-if the false positive rate is 0.1% of that is 1500.
For which test? We're including PCR and LFD which have totally different sensitivity and specificity. You cant just average them. The maths doesnt work.
More realistically, false positives are unlikely to be running at less than 1%,
Where on earth are you finding data to back up that statement.
Why is it "unlikely"? Where are you seeing papers to suggest that?
Why does that disagree totally with the UK and globally measures positivity rates for various tests in areas of low prevalence?

ONS official methodology calculate the specificity as "at least 99.92%". [1]
March 2021 Phase 4 LFD found a specificity of minimum 99.72% and most likely 99.9% for best data fit [2]


For LFD yep, its assessed at higher than that. For PCR, its not up at that level at all.
which with these figures would produce 15,000 false positives a day, many more than are currently being reported on a daily basis.
Great, except you're just inventing random numbers for false positive rates, ignoring false negative rates and as such have no valid assumptions to base those figures on.

You're also forgetting PCR self-screens by requiring symptoms thereby effectively increasing disease prevalence and reducing false positive risks.

So where are you getting your figures from? The things you didn't know about hint you've never browsed any official data portals at all for starters. All the data is out there.


[1] https://www.ons.gov.uk/peoplepopulation ... nformation
[2] https://assets.publishing.service.gov.u ... hase-4.pdf

Shotclog
Posts: 15
Joined: Mon Feb 15, 2021 11:37 am

Re: Daily "cases"

Post by Shotclog »

Calm down old chap. If you read my post it is pretty clear that I was not sure about any of the figures. I was not asserting that I knew better and you were wrong.

The 1-4% false positive figures comes from a FOI request to the ONS I read a while ago. It was put to them that these were the figures and could they confirm it. To say that their answer to this was obscure is putting it mildly-they certainly didn't answer in a way that I could understand if these were, or were not, the right estimates. Is "specificity" supposed to mean accuracy? Genuine question. For my part, I don't know why specificity is a better word for accuracy but then I'm a bit thick.

Ultimately, life is too short to get into a low-level spat about this on the internet. The short point that I was ham-fistedly trying to make was no more than that it is surprising, to me, that cases have now flatlined in a narrow parameter of about 2,500 to about 4,500 for days and weeks now. I suspect (but can't obviously prove) that the consistency reflects the false positives from testing between 1 and 2 million people. For my part, I suspect (but again can't prove) that almost nobody is really getting sick with Covid now (a few are of course, but then a few people fall down the stairs every day).

Splatt
Posts: 1436
Joined: Thu Oct 29, 2020 12:46 am

Re: Daily "cases"

Post by Splatt »

The 1-4% false positive figures comes from a FOI request to the ONS I read a while ago.
Thats not in a single ONS surveillance report nor their methodology documents.
You wouldnt need FOIA for the ONS anyway, thats kind of their whole point.
. Is "specificity" supposed to mean accuracy?
No. Specificity and Sensitivity help determine accuracy but they're not the same thing.

https://www.healthnewsreview.org/toolki ... ive-value/
that cases have now flatlined in a narrow parameter of about 2,500 to about 4,500 for days and weeks now
We are going to hit that point fairly soon, in the next month or two but its unlikely to be yet.
Ultimately there is always a background rate of infections for every level of social interaction. Its quite possible that at the current rates of social contact this is the natural equilibrium rate for this time of year.
As we head into summer Rt is going to drop and then we'll likely hit a false test floor. This minimum will go UP the more we switch to less accurate and huge numbers of lateral flow devices.

I suspect we'll never manage to drop below 1500-2000 positives a day, maybe a little higher is Hancocks LFD-For-All idiocy pans out.

PCR is a little more complicated as tests rely on symptoms. If fewer people get symptoms fewer people get tested therefore fewer tests are performed and therefore fewer false positives are detected. In summer months people are less sick from anything so wont go for tests.
This will pick up again in Autumn causing a false "case increase" some time around the end of September.

thinksaboutit
Posts: 411
Joined: Wed Jan 06, 2021 11:38 am

Re: Daily "cases"

Post by thinksaboutit »

Splatt wrote: Fri Apr 09, 2021 11:09 am
The 1-4% false positive figures comes from a FOI request to the ONS I read a while ago.
Thats not in a single ONS surveillance report nor their methodology documents.
You wouldnt need FOIA for the ONS anyway, thats kind of their whole point.
. Is "specificity" supposed to mean accuracy?
No. Specificity and Sensitivity help determine accuracy but they're not the same thing.

https://www.healthnewsreview.org/toolki ... ive-value/
that cases have now flatlined in a narrow parameter of about 2,500 to about 4,500 for days and weeks now
We are going to hit that point fairly soon, in the next month or two but its unlikely to be yet.
Ultimately there is always a background rate of infections for every level of social interaction. Its quite possible that at the current rates of social contact this is the natural equilibrium rate for this time of year.
As we head into summer Rt is going to drop and then we'll likely hit a false test floor. This minimum will go UP the more we switch to less accurate and huge numbers of lateral flow devices.

I suspect we'll never manage to drop below 1500-2000 positives a day, maybe a little higher is Hancocks LFD-For-All idiocy pans out.

PCR is a little more complicated as tests rely on symptoms. If fewer people get symptoms fewer people get tested therefore fewer tests are performed and therefore fewer false positives are detected. In summer months people are less sick from anything so wont go for tests.
This will pick up again in Autumn causing a false "case increase" some time around the end of September.
The 1% to 4% FPR figure has been thrown around for a while by people on this forum, who seek to underplay the level of infection.
There is even one person (with a few cheerleaders) who stated a figure of more than 90% for quite some time.

Given changes in testing regimes and volumes, plotting cases against time, is likely to be quite misleading. As test volumes rise, so do positive results ( assuming a similar positivity rate).

There is also the opposite problem, in that many people won't take a test, even if symptomatic. This because they can't afford to not work, so don't want the result. It is entirely possible for the published cases to fall while the actual infections rise.

Watching hospital admissions if probably the best guide to the state of the pandemic.

thinksaboutit
Posts: 411
Joined: Wed Jan 06, 2021 11:38 am

Re: Daily "cases"

Post by thinksaboutit »

Shotclog wrote: Thu Apr 08, 2021 5:01 pm Calm down old chap. If you read my post it is pretty clear that I was not sure about any of the figures. I was not asserting that I knew better and you were wrong.

The 1-4% false positive figures comes from a FOI request to the ONS I read a while ago. It was put to them that these were the figures and could they confirm it. To say that their answer to this was obscure is putting it mildly-they certainly didn't answer in a way that I could understand if these were, or were not, the right estimates. Is "specificity" supposed to mean accuracy? Genuine question. For my part, I don't know why specificity is a better word for accuracy but then I'm a bit thick.

Ultimately, life is too short to get into a low-level spat about this on the internet. The short point that I was ham-fistedly trying to make was no more than that it is surprising, to me, that cases have now flatlined in a narrow parameter of about 2,500 to about 4,500 for days and weeks now. I suspect (but can't obviously prove) that the consistency reflects the false positives from testing between 1 and 2 million people. For my part, I suspect (but again can't prove) that almost nobody is really getting sick with Covid now (a few are of course, but then a few people fall down the stairs every day).
You need to understand the actual processes and the associated simple maths, rather than just sniff the air.

The published figures show the daily number of people being admitted to hospital with Covid. So no need to "suspect" just read the numbers!

Currently just above 200 people/day.

BTW specificity has a particular technical definition . If you type specificity and definition into google, you will get a head start.
While you are at it look up sensitivity.

Post Reply