So, coronavirus...

[Mahoney]

Unless you are testing at random you cannot know.

[Bimbowomxn]

Unless you are testing at random you cannot know.

I’m not looking for an exact number....

Just a rough idea.


More than or less than 1/2?

[Biffer]

This thread has fallen into a state where its mostly two idiots taking incredibly moronic positions on opposite sides of the debate.

[Muttonbird]

Mostly false positive I imagine

Hey rinky.....

Tell is how many of those would be False positives if the false positives rate was 1%?

22.

[Bimbowomxn]

This thread has fallen into a state where its mostly two idiots taking incredibly moronic positions on opposite sides of the debate.

The conversation around false positive is very important regarding policy though. There isn’t two sides to the facts around it,

[Tichtheid]

On false positives

https://www.bbc.co.uk/news/54270373

[CM11]

Head of our testing pointed out that they are seeing a 0.2% positive rate for serial testing so by definition the max number of false positives is 0.2% here. Not sure if same applies in the UK but the increase in overall positives simply can't be driven by false positives to the extent some are suggesting.

[Bimbowomxn]

On false positives

https://www.bbc.co.uk/news/54270373

This ignores the mass of testing on students and the young.

[Tichtheid]

On false positives

https://www.bbc.co.uk/news/54270373

This ignores the mass of testing on students and the young.

COVID-19 daily situation report COVID-19 admissions estimates in England

https://www.england.nhs.uk/statistics/w ... 01003.xlsx

[Bimbowomxn]

On false positives

https://www.bbc.co.uk/news/54270373

This ignores the mass of testing on students and the young.

COVID-19 daily situation report COVID-19 admissions estimates in England

https://www.england.nhs.uk/statistics/w ... 01003.xlsx

We have more students locked down as “positive , but with no symptoms “ at one campus than the total admission on a single day this week .

It’s also quite clearly not doubling every 7 days.

What point are,you trying to make in relation to 170-200,000 tests a day?

[Rinkals]

Unless you are testing at random you cannot know.

I’m not looking for an exact number....

Just a rough idea.


More than or less than 1/2?

Is Bimbo really saying that half of all test results are false positives?

My post that he has taken exception to was in response to his implication that Homer's graph was explained by the results being "mostly false positives".

If most (ie. more than half) of the testing results are unreliable, what is the point of testing at all?

[Insane_Homer]

BOJO this morning: ‘behave fearlessly but with common sense’

Is that advice or law? Somehow I don't think arguing 'common sense' and quoting the PM is going to get you out of a £10k fine.

[Tichtheid]

The point was expressed in the BBC link,

Dr Paul Birrell, a statistician at the Medical Research Council's Biostatistics Unit at the University of Cambridge, says: "The false positive rate is not well understood..

False positives are not having a significant impact on figures, also

Dr Birrell says that to be certain cases really are increasing, the daily case count "should always be considered alongside other information sources, such as the hospitalisations or deaths, or the community surveys run by the ONS or REACT".

The figures in the NHS link show increases in hospitalisation, there is another table easily looked for which show an increase in deaths

[Bimbowomxn]

Unless you are testing at random you cannot know.

I’m not looking for an exact number....

Just a rough idea.


More than or less than 1/2?

Is Bimbo really saying that half of all test results are false positives?

My post that he has taken exception to was in response to his implication that Homer's graph was explained by the results being "mostly false positives".

If most (ie. more than half) of the testing results are unreliable, what is the point of testing at all?

Well, that’s the question that’s being asked about the sheer scale and where testing is aimed in the UK. Test and trace is correct as a policy, testing for its own sake to say “we’ve made record tests” makes absolutely no sense at all.

The delay in reporting numbers seen this week show that folly even more clearly.

That this is dictating policy is much worse than folly.

[Bimbowomxn]

The point was expressed in the BBC link,

Dr Paul Birrell, a statistician at the Medical Research Council's Biostatistics Unit at the University of Cambridge, says: "The false positive rate is not well understood..

False positives are not having a significant impact on figures, also

Dr Birrell says that to be certain cases really are increasing, the daily case count "should always be considered alongside other information sources, such as the hospitalisations or deaths, or the community surveys run by the ONS or REACT".

The figures in the NHS link show increases in hospitalisation, there is another table easily looked for which show an increase in deaths

Check out professor Carl Heneghans work at Oxford.

Policy isn’t being made because of 100’s of hospital visits and tens of deaths.

[Tichtheid]

From a very quick search Prof. Heneghan seems to be choosing his words very carefully, he says there isn’t enough evidence to declare a second wave yet (my emphasis) and he thinks a lockdown over Christmas is a better idea than doing it now.

[Bimbowomxn]

From a very quick search Prof. Heneghan seems to be choosing his words very carefully, he says there isn’t enough evidence to declare a second wave yet (my emphasis) and he thinks a lockdown over Christmas is a better idea than doing it now.

I was thinking more about the issue of false positives and base rate fallacy....

There’s certainly no “second wave” yet, we aren’t even discussing that are we ?


We are locking down 11th Oct anyway I reckon.

[Raggs]

The point was expressed in the BBC link,

Dr Paul Birrell, a statistician at the Medical Research Council's Biostatistics Unit at the University of Cambridge, says: "The false positive rate is not well understood..

False positives are not having a significant impact on figures, also

Dr Birrell says that to be certain cases really are increasing, the daily case count "should always be considered alongside other information sources, such as the hospitalisations or deaths, or the community surveys run by the ONS or REACT".

The figures in the NHS link show increases in hospitalisation, there is another table easily looked for which show an increase in deaths

False hospitalizations no doubt...

[Sandstorm]

[tabascoboy]

The point was expressed in the BBC link,

Dr Paul Birrell, a statistician at the Medical Research Council's Biostatistics Unit at the University of Cambridge, says: "The false positive rate is not well understood..

False positives are not having a significant impact on figures, also

Dr Birrell says that to be certain cases really are increasing, the daily case count "should always be considered alongside other information sources, such as the hospitalisations or deaths, or the community surveys run by the ONS or REACT".

The figures in the NHS link show increases in hospitalisation, there is another table easily looked for which show an increase in deaths

False hospitalizations no doubt...

and false deaths, where are the bodies?

[Tichtheid]

I've read Prof Carl Heneghan's article in the Spectator from back in July on false positives, unfortunately it seems to be based on a hypothetical statistics exercise, rather than a real-life situation

Article here https://www.spectator.co.uk/article/how ... tives-/amp


There is a rebuttal in the Huff post here, and before anyone dismisses the Huffpost, I'm not interested, try and refute the material in the article instead.

https://www.huffingtonpost.co.uk/entry/ ... vsISQNuxsS

[Hal Jordan]

BOJO this morning: ‘behave fearlessly but with common sense’

Is that advice or law? Somehow I don't think arguing 'common sense' and quoting the PM is going to get you out of a £10k fine.

It’s a contradiction in terms.

Anyone else see that little shit Hancock refusing to answer a question by saying it was couched in "divisive language" and he "would not have it" and then just sitting down again? What. A. Cunt.

[Bimbowomxn]

I've read Prof Carl Heneghan's article in the Spectator from back in July on false positives, unfortunately it seems to be based on a hypothetical statistics exercise, rather than a real-life situation

Article here https://www.spectator.co.uk/article/how ... tives-/amp


There is a rebuttal in the Huff post here, and before anyone dismisses the Huffpost, I'm not interested, try and refute the material in the article instead.

https://www.huffingtonpost.co.uk/entry/ ... vsISQNuxsS

Heneghan is particularly interested in “false positives” – those people who test positive for Covid-19 but actually aren’t infected. Health secretary Matt Hancock has said the false positive rate (FPR) for coronavirus tests is “less than 1%”.

But Heneghan has argued that due to a bit of a fluke involving some slightly complicated statistics, the proportion of positive tests that are false in the UK could be as high as 50%.

The comparison of the two figures demonstrates the “Huff post” hasn’t got a fucking clue what Heneghans work is, was or what he published in July comes from.

It’s comparing a False positive rate with a result of applying it.

A false positive rate of 1% with a low incident of positive Results would mean as high as 50% or higher proportionately being wrong, that’s not a “fluke” it’s just Maths.

Heneghans work corrected the Covid official death figures by over 20% btw.

[Ali Cadoo]

I’m sure this must all come as great comfort to all those dead people.

[Bimbowomxn]

I’m sure this must all come as great comfort to all those dead people.

,

Wonderfully signalled .

[Ali Cadoo]

I’m sure this must all come as great comfort to all those dead people.

,

Wonderfully signalled .

How’s your ‘cunt signalling’ working out for you? Are you happy to have removed all doubt?

[Tichtheid]

If you click through to the British Medical Journal's site from the Huffpost article, they explain that the problem isn't with false positive results, which are minimal, it's actually with false negatives and we are all vastly underestimating the extent of the disease.

The pre-test probability is the crucial figure, and it depends on symptoms and prevalence of the disease in the patient's area as well as things like any travel etc.

If you are in a high probability group, ie around 90% you have a 99% chance of having Covid 19, but even after one negative test you still have a probability of 75% of having the disease, this falls to 47% even after two negative tests.

https://www.bmj.com/content/369/bmj.m1808

[.OverThere]

The point was expressed in the BBC link,

Dr Paul Birrell, a statistician at the Medical Research Council's Biostatistics Unit at the University of Cambridge, says: "The false positive rate is not well understood..

False positives are not having a significant impact on figures, also

Dr Birrell says that to be certain cases really are increasing, the daily case count "should always be considered alongside other information sources, such as the hospitalisations or deaths, or the community surveys run by the ONS or REACT".

The figures in the NHS link show increases in hospitalisation, there is another table easily looked for which show an increase in deaths

False hospitalizations no doubt...

How many were hospitalised in September for normal flu? How does this compare with the 10 year average, and how does it compare with covid hospitalisations? Similarly the comparisons for deaths?

[Ali Cadoo]

Guys - I’m reasonably confident that IF there were a way to accurately measure and record the impact of the ‘rona then some clever chap would have done it by now.

[Sandstorm]

Guys - I’m reasonably confident that IF there were a way to accurately measure and record the impact of the ‘rona then some clever chap would have done it by now.

I hope not. I have him on ignore.

[Ali Cadoo]

Guys - I’m reasonably confident that IF there were a way to accurately measure and record the impact of the ‘rona then some clever chap would have done it by now.

I hope not. I have him on ignore.

I believe that what we currently have are various different ways to measure the impact, all with an error margin of about 60%. However, suitably enough, this belief of mine also comes with a likely error margin of about 60%. It’s like statistical Russian dolls.

[Bimbowomxn]

If you click through to the British Medical Journal's site from the Huffpost article, they explain that the problem isn't with false positive results, which are minimal, it's actually with false negatives and we are all vastly underestimating the extent of the disease.

The pre-test probability is the crucial figure, and it depends on symptoms and prevalence of the disease in the patient's area as well as things like any travel etc.

If you are in a high probability group, ie around 90% you have a 99% chance of having Covid 19, but even after one negative test you still have a probability of 75% of having the disease, this falls to 47% even after two negative tests.

https://www.bmj.com/content/369/bmj.m1808

Yeah, Carl Heneghan is a clutz.

[Bimbowomxn]

I’m sure this must all come as great comfort to all those dead people.

,

Wonderfully signalled .

How’s your ‘cunt signalling’ working out for you? Are you happy to have removed all doubt?

No, I too am ever so caring about the dead people. I’ll ignore science I’m that concerned.

[Raggs]

The point was expressed in the BBC link,



False positives are not having a significant impact on figures, also



The figures in the NHS link show increases in hospitalisation, there is another table easily looked for which show an increase in deaths

False hospitalizations no doubt...

How many were hospitalised in September for normal flu? How does this compare with the 10 year average, and how does it compare with covid hospitalisations? Similarly the comparisons for deaths?

No idea. Why not look it up? Sounds like an interesting question.

[Tichtheid]

If you click through to the British Medical Journal's site from the Huffpost article, they explain that the problem isn't with false positive results, which are minimal, it's actually with false negatives and we are all vastly underestimating the extent of the disease.

The pre-test probability is the crucial figure, and it depends on symptoms and prevalence of the disease in the patient's area as well as things like any travel etc.

If you are in a high probability group, ie around 90% you have a 99% chance of having Covid 19, but even after one negative test you still have a probability of 75% of having the disease, this falls to 47% even after two negative tests.

https://www.bmj.com/content/369/bmj.m1808

Yeah, Carl Heneghan is a clutz.

The paper in the BMJ is fully referenced and peer-reviewed and has a citing reference attached, but you are at liberty to point out the faults in it if you like.

[Bimbowomxn]

Published on the 12th of May.

[Ali Cadoo]

,

Wonderfully signalled .

How’s your ‘cunt signalling’ working out for you? Are you happy to have removed all doubt?

No, I too am ever so caring about the dead people. I’ll ignore science I’m that concerned.

You haven’t lost anyone to the virus, have you?

[Tichtheid]

Published on the 12th of May.

Nearly five months ago, yeah. If you read the replies to the article you will see that researchers have been engaging with the article from New Zealand, India and Norway as recently as late September.

Indeed, Henrik Vogt, GP and postdoctoral fellow, Centre for medical ethics, University of Oslo, Norway writes,

What has made PCR tests so tremendously useful in this pandemic (odes should probably be sung to their glory), is that they are based on gene technology and identify specific sequences of DNA or, in the case of SARS CoV-2, RNA molecules. Since RNA is extremely specific for a biological entity, this allows extreme specificity and few false positive tests and high positive predictive values - even in contexts with low prevalence.

and also

In a July article in The Spectator, Oxford professor of evidence-based medicine, Carl Heneghan, also refers to the BMJ practice (5). While the pointer used specificity 95% "for illustrative purposes", Heneghan refers to it as "a review" that "reported that the specificity of PCR tests could be as low as 95%". While Heneghan is also only trying to contribute to the understanding of tests, this shows that using a figure like 95% for illustrative purposes may be misunderstood as a reference to the facts, be misleading and give the impression that false positives are very common. Admittedly, Heneghan also writes in The Spectator, that "based on the latest data, the specificity may be as high as 99,9 per cent", but he also states he thinks this is too high.

The thing is, the higher the specificity the lower the number of false positives that will result, in fact at 98% numbers of false positives are vanishingly small

[Bimbowomxn]

Great,

The May article isn’t based upon mass testing, since the article that’s what happened.

Either way there’s a LOT of false positive and we’re seeing them described as “cases”.

That Heneghans work two / three months after the article was accepted enough to lower death statistics dramatically does mean he’s on the right track. It isn’t a “fluke” of statistics.

When we retest the positives to be “sure” we will be much better off.

[Tichtheid]

If there were a lot of false positives it would be down to a flaw in the tests, surely it would be roughly equal across the UK in terms of number of cases per X number of population?

There are huge discrepancies in the number of cases in different areas, as shown in the gov.uk data on the Huffpost article.

They say

...we know rising positive cases aren’t due to false positives for a couple of other reasons.

Positive test rates are going up as a percentage of total tests – this is not disputed.

But this can’t be because of an increase in false positives as the rate of false positives remains constant unless the actual method of testing changes, which it hasn’t.

Additionally, if false positives were causing the spike in numbers, it would be uniform across the UK and it isn’t.