Jokes0 min ago
Bad Tests And False Positives
73 Answers
This is a rather long, but fascinating article
Written by a biomedical research scientist, he analyses whats wrong with the current testing procedure, why it's producing so many false positives (more than we've been told), and...yes...Matt Hancock has a lot to answer for. I've waded through about 3/4 of it. It's not easy reading, but worthwhile.
https:/
Written by a biomedical research scientist, he analyses whats wrong with the current testing procedure, why it's producing so many false positives (more than we've been told), and...yes...Matt Hancock has a lot to answer for. I've waded through about 3/4 of it. It's not easy reading, but worthwhile.
https:/
Answers
Best Answer
No best answer has yet been selected by pastafreak. Once a best answer has been selected, it will be shown here.
For more on marking an answer as the "Best Answer", please visit our FAQ.
// that is my definition of H I //
yes I had spotted that .....
"Herd immunity needs 60% here
and we are nowhere near that "
X: "no no- I dont take the definition of HI that everyone else does: I have my own. and it is in full swing!"
Let us call a thing a thing and not another thing (Bishop Butler)
[otherwise you can look at say a horse
and somene can say - oh look a cow!
and someone else say - no - its an orchid!)
yes I had spotted that .....
"Herd immunity needs 60% here
and we are nowhere near that "
X: "no no- I dont take the definition of HI that everyone else does: I have my own. and it is in full swing!"
Let us call a thing a thing and not another thing (Bishop Butler)
[otherwise you can look at say a horse
and somene can say - oh look a cow!
and someone else say - no - its an orchid!)
//…but it is not only the elderly who are succumbing to Covid 19.//
It is overwhelmingly so, danny. Over 90% of victims were over 65.
//This is also misleading because it ignores the fact that there has been a trend upwards for the last month.//
It is also misleading because it [478 new hospital admissions] does not put the matter into context. There are normally around six million hospital admissions annually in England alone. That’s about 120,000 a week. In the week to 4th October (the latest figures available) there were 2,549 Covid admissions. That’s about one in every fifty of the normal number. Bearing in mind that so-called “elective” surgery has been all but abandoned and that many other treatments which normally require a hospital admission have similarly been suspended, it really makes no sense to fret over such a small number of admissions.
It is overwhelmingly so, danny. Over 90% of victims were over 65.
//This is also misleading because it ignores the fact that there has been a trend upwards for the last month.//
It is also misleading because it [478 new hospital admissions] does not put the matter into context. There are normally around six million hospital admissions annually in England alone. That’s about 120,000 a week. In the week to 4th October (the latest figures available) there were 2,549 Covid admissions. That’s about one in every fifty of the normal number. Bearing in mind that so-called “elective” surgery has been all but abandoned and that many other treatments which normally require a hospital admission have similarly been suspended, it really makes no sense to fret over such a small number of admissions.
PP
//////Herd immunity needs 60% here
and we are nowhere near that "/////
In simple terms......how do we know that 60% of the population have Immunity, when arguably less than half of the population have been tested?
Yes, my definition does differ from the scientific one as mine seems to be more pragmatic.
//////Herd immunity needs 60% here
and we are nowhere near that "/////
In simple terms......how do we know that 60% of the population have Immunity, when arguably less than half of the population have been tested?
Yes, my definition does differ from the scientific one as mine seems to be more pragmatic.
I'm sorry, Naomi, that your only response to criticism is to accuse others of scaremongering. But all you do is throw the word around so much that it loses all meaning. There's nothing I've done or said during this thread or during this pandemic that can remotely amount to scaremongering.
Find something new to say.
Find something new to say.
well gold star for the fridge for Danny
only one ( exchange with Naomi above)
to see that an incidence ( cases per unit time (day/week) is seriously different to the prevalence - who has it on a certain day
the numbers are different and the implications are different
False positives -
People I am glad to see are edging towards the PPV the chance that if a disease tests positive, then you have the disease.
take a hundred a nice round number and ONE has it and there is a false positive rate of piffle - 4%
then the one will score and
four will score ( FP) - - making five
but only one out of the five has it (20%)
so here the chance of a positive test identifying a case is 20%
Not me - this is a characteristic of the test system.
even if the prevalence is 6% and the FP rate 4%
then the chance of a positive test identifying an infected case is 60%
this is not news
we knew this since testing began 1900.
well done readers of AB - this has escaped the notice of the editor of the BMJ and all her cohorts - despite quite a few people (not only me) telling her
- demurring - - no - telling her
only one ( exchange with Naomi above)
to see that an incidence ( cases per unit time (day/week) is seriously different to the prevalence - who has it on a certain day
the numbers are different and the implications are different
False positives -
People I am glad to see are edging towards the PPV the chance that if a disease tests positive, then you have the disease.
take a hundred a nice round number and ONE has it and there is a false positive rate of piffle - 4%
then the one will score and
four will score ( FP) - - making five
but only one out of the five has it (20%)
so here the chance of a positive test identifying a case is 20%
Not me - this is a characteristic of the test system.
even if the prevalence is 6% and the FP rate 4%
then the chance of a positive test identifying an infected case is 60%
this is not news
we knew this since testing began 1900.
well done readers of AB - this has escaped the notice of the editor of the BMJ and all her cohorts - despite quite a few people (not only me) telling her
- demurring - - no - telling her
// how do we know that 60% of the population have Immunity, when arguably less than half of the population have been tested?//
well we dont
a herd can be 100% immune and if it is untested no one will know but they will still be immune.
(hey shades of schrodingers cat in all this)
the figure comes from the Ro - shush - no "I dont believe in R numbers" that is like saying "I dont believe in pi" (22/7, that one)
Measles - Ro of around 12 - you would need lots and lots immune before 10 / 12 hits dont score an infection - and the herd immunity reqd is 90%
Covid Ro not greater than 2 ever - yes ! you only need to cut the successful hit rate by fifty % before it fails to be passed on
and so you can see the actual herd immunity needed is closely related to the Ro in that population ( as measured 5 y later)
and the projected herd immunity needed is closely related to any estimation of Ro now
and it is around 60%
You probably would not have had ANY lectures on this at all during your course ....
[DIfft argt for diphtheria where the immunised population actually wipes out the infection - the immunised bacterium is more efficient as it doesnt have the energy requiring lysogenic virus]
well we dont
a herd can be 100% immune and if it is untested no one will know but they will still be immune.
(hey shades of schrodingers cat in all this)
the figure comes from the Ro - shush - no "I dont believe in R numbers" that is like saying "I dont believe in pi" (22/7, that one)
Measles - Ro of around 12 - you would need lots and lots immune before 10 / 12 hits dont score an infection - and the herd immunity reqd is 90%
Covid Ro not greater than 2 ever - yes ! you only need to cut the successful hit rate by fifty % before it fails to be passed on
and so you can see the actual herd immunity needed is closely related to the Ro in that population ( as measured 5 y later)
and the projected herd immunity needed is closely related to any estimation of Ro now
and it is around 60%
You probably would not have had ANY lectures on this at all during your course ....
[DIfft argt for diphtheria where the immunised population actually wipes out the infection - the immunised bacterium is more efficient as it doesnt have the energy requiring lysogenic virus]
-- answer removed --
Sqad isn't this a function of the random testing of volunteers program..also the Covid symptoms app, not the nhs one?
So you look at how many tests have been done? (not tests people got through the get a test if you have symptoms route because that would be silly)
out of those how many were positive?
et viola a ballpark number for HI, assuming that you can only get it once.
So you look at how many tests have been done? (not tests people got through the get a test if you have symptoms route because that would be silly)
out of those how many were positive?
et viola a ballpark number for HI, assuming that you can only get it once.
/////a herd can be 100% immune and if it is untested no one will know but they will still be immune./////
Lol......I give in.
You are almost correct re lectures as in 1953 we were all told to buy this little book..Statistics For Medical Students,cost 25 shillings and it taught us all one thing.........we had thrown away 25 shillings.
I will prefer my definition of Herd Immunity.?........it makes more sense.
Lol......I give in.
You are almost correct re lectures as in 1953 we were all told to buy this little book..Statistics For Medical Students,cost 25 shillings and it taught us all one thing.........we had thrown away 25 shillings.
I will prefer my definition of Herd Immunity.?........it makes more sense.
// PP Congratulations for a post that was 100% understandable.//
thx but no thx
the great sqad has checked out on
you cant be immune unless you have been tested
( er yes you can - all the recovered viet patients are immune and untested ) - yes yes we know infections confers immunity because we know the reinfection rate is virtually zero
thx but no thx
the great sqad has checked out on
you cant be immune unless you have been tested
( er yes you can - all the recovered viet patients are immune and untested ) - yes yes we know infections confers immunity because we know the reinfection rate is virtually zero
ye woofie
stats for med st really does seem a waster of money
if you take all possible tests and their positive rate (6%) ( and it is not falling) then the lower bound has to be " greater than 6%"
and if 6% is not growing particularly fast then it is "above 6% but not much"
a very large generation of doctors of great eminence and erudition say ' oh numbaz - I dont do them'
stats for med st really does seem a waster of money
if you take all possible tests and their positive rate (6%) ( and it is not falling) then the lower bound has to be " greater than 6%"
and if 6% is not growing particularly fast then it is "above 6% but not much"
a very large generation of doctors of great eminence and erudition say ' oh numbaz - I dont do them'
I like you danny, a lot, but naomi is right. There is a hell of a lot of scaremongering going on and we need to get back to normal as soon as humanly possible before it's too late and even more people lose jobs through all the closures going on all around us. I'm a pensioner and I can worry about and take care of myself, this has gone on long enough now and I can't believe we may be on the way to another lockdown and yet more closures. Young people need jobs and to provide for their families and life has to continue. Enough is enough now. Sadly, Boris is still listening to all the scaremongers.
