to explain why testing lots of asymptomatic people everyday is a bad idea

[chomalungma]

They want to do lots of testing on asymptomatic people everyday.
Aim is to see if people are positive, and if so, then those people have to self isolate.

Issue is: There will be many people who will have to self isolate who don't need to because of false positives

All to do with how specific a test is, how sensitive it is, and the prevalence of disease in the population you are screening.

Let's say the levels of disease are 1 in 10,000

So in 1 million people, 100 people will have the disease

And 999,900 people won't have the disease.

They get a really good test; It detects disease in 99% of the people who have it.

So in the 'ill' population, we have 99 positives and 1 negative

The test is also very specific. Say 99.5% specific. That means in the 999,900 people who don't have the disease, we will get 994,900 people who test negative and 5000 who test positive (False positives)

So we have 5099 positive tests.
And 99 of them are actual positives

So 98% of the positive results are false positves

(this is based on a level of 1 in 10,000 and those figures for specificity and sensitivity) Obviously a test needs to have really good specificity and sensitivity and it needs to be done on a population with a relatively high chance of having the disease.

yeah.right.

The OP is right. There was someone on radio 4 saying the same thing earlier today.

@TorysSuckRevokeArticle50

Can you please state what qualifications/experience you have that make you qualified to posset this opinion, also could you provide links to the peer reviewed research you have used to reach your conclusion?

Sorry, but I have to laugh at 'posset'. Laughs are thin on the ground at the moment, so I'm all the more grateful for this.

This was really well-explained - thank you, OP. I think a lot of people think that only false negatives are possible for all sorts of testing (possibly because of pregnancy tests?!). I've tried to explain this problem with mass cancer screening before - that if you end up with a lot of people having unnecessary invasive follow-up testing that could be harmful and relatively few genuine positives identified then you may have done more harm than good - and the person I was talking to just refused to believe that it could ever make sense, on a public health level, to not do any mass test that was available.

OP is entirely correct and yes I am qualified to confirm that!

Regarding false positives when screening for cancer, there was a very interesting piece on More or Less recently about a study looking at screening women over 40 for breast cancer. They talked about the rate of true false positives, as well as the cancers that would be detected and treated but would never have caused the woman any symptoms if they had not been found. Bit of a tangent from coronavirus but worth a listen! Actually all of More or Less is worth a listen and they have also done many interesting coronavirus related programmes.

@Michaelschofield

Why do we need to test people. So what if people are positive? Doesn’t mean the end of the world . 99% people will survive this.

This, too. I don't want a bloody test. If I had a positive test, it would simply mean that I would earn absolutely nothing again, as opposed to the tiny bit that I am starting to claw back after lockdown. I can't afford to do this. So unless they make Covid tests compulsory - and if they did this, I think I would emigrate - I wouldn't have one if it were offered to me.

I would, however, keep myself to myself if I actually had any of the symptoms of Covid (i.e. the genuine symptoms, not the "Oh, you had a migraine, therefore you must have Covid" symptoms which seem to pop up on here).

I am guessing false negative figures are harder to pin down as there will be a lot of a 'human error' element involved, so people not executing the test correctly etc ?

@chomalungma - totally agree (although you’re right - you can’t agree or disagree with maths, it just is...), but the worst thing is that the rapid turn around rests developed so far are neither sensitive nor good. So you’d have tens of thousands of false positives.

I’m all for monitoring random samples, but this moonshot idea is crap. The most worrying thing is that they felt able to announce such crap in public. Is anyone able to tell Bojo his ideas are bad? Who’s speaking truth to power here?

I’m embarrassed to admit it took me a long time - and a pen and paper - to get the false positive thing. But the penny has dropped. Wow!

I believe that false positives are the reason why smear tests are delayed. It prevents intrusive procedures being undertaken on healthy women

Where in the meantime has the assertion They want to do lots of testing on asymptomatic people everyday come from during the course of the day? Who wants to? Where was this published?

It’s referring to the Moonshoot plan I think. Boris has said he wants to implement a mass testing programme, processing millions of tests a day. And some scientists are opposing it. For the reasons the OP has explained.

YANBU

I think the problem we face with COVID-19 is actually the other way around: false negatives will be more problematic than false positives.

Unlike OPs example. the current gold standard COVID-19 test is not particularly sensitive, but very specific. Which means I'm less concerned about the false positive rate (which is in the single % figures) than the false negative rate (which we think is up to 30%).

Given the ease of disease spread, relying on a test with such a high false negative rate means many COVID-19 cases won't be identified but will potentially carry on behaving as if they don't have the disease, thus spreading it to others.

It becomes even more complex when you factor in pre-test probability: the likelihood of the disease being present before you even do the test. Some risk factors (type of work, recent contact with a known case, prevalence of the disease in the community) will increase the pre-test probability of disease. In most diseases, symptoms would also contribute to the pre-test probability, but given how variable COVID-19s symptoms and severity are it is much harder to say how likely you think the disease is based on what you know about a patient and therefore what weight you give to a positive or negative test result. This could mean loads of mildly symptomatic people with false negatives assuming (understandably, but wrongly) that they don't have COVID-19.

False positives are inconvenient but acceptable and affect a relatively small proportion of positives with this test, false negatives are much more problematic and mean this approach probably won't work.

(I'm a research scientist and medical doctor).

That's still an awful lot of people able to go about their business who might not otherwise be able to.

In a lockdown situation.

The NHS Trust that I work for is well aware of this issue. All elective patients are tested prior to admission and any that are positive are re-tested. This massively reduces the chance of a final false positive result. Presumably this would also be done if testing is rolled out to the wider population.

If you test everyday even after a positive result, false positives will be corrected by the next day’s test (depending on the false positive rate, some would be the day after). It’d be crazy to stop testing after a positive result for all the reasons mentioned here.

For more rapid tests, you’d do more than one test in a row or more than one test at a time I suppose. The exact process would depend on how the test results tend to be incorrect.

On the basis that they seem to be struggling to swab and process even 200,000 people I think the daily test is several months, if not years, away!

The Madcat family have just joined the ONS study that takes blood and swabs us to estimate transmission. [If we'd had coronavirus it would have been back in February/March when you needed to be hospitalised to be diagnosed]. Quarantining for a fortnight would be annoying, but easily doable for us (though DC would be miffed at missing face to face contact with friends at school).

I know some families were anxious about sending their DC back to school. In a tiny way I think they find our willingness to pop a stick down our throats and around our nostrils when asked, strangely reassuring. Likewise the teachers and sports coaches.

If daily testing were available, and the authorities were mindful of the risk of 'false positives", surely they would prioritise the re-testing of anomalies? Or at least commission a study to see if there was a reason why some people are asymptomatic?

Snap @Reallybadidea 😊

I like this thread for the qualification announcements

But I agree re false negatives

But can someone science-based talk a bit about how many people are curtailed currently under our blunt instruments and whether this is more or less than false negatives - op?

Cost-benefit-Analysis hasn't driven the decision making At.All.

I am like Celan. Just trying to keep my head down & hoping that 1984 isn't in my future.

J0rr1$$ B0n$0n doesnt want to understand, he's fixated on his moonshot/money shot idea
he's more trump than trump is and now he thinks he's elonmuskrat

Re false negatives - aren't they largely false negatives because the individual isn't shedding enough virus to be collected by the swab? So they may not even be infectious anyway.

I believe that David Spiegelhalter has written and spoken often about how, if there is mass testing in conditions of low prevalence, that harm from false negatives are much less than the harms caused by false positives. I get confused by the math, but basically the tests are more likely to generate false positives than false negatives. Spiegelhalter is a Cambridge Uni statistician, probably is a topic covered on Radio4's More or Less, too.