Before continuing this conversation here ... I'm going to dig into this and perhaps you may wish to have a peak.
I am already on a promise of a peak from Boris, but I will have a peek.
Same time in recent Spectator column they say less than 1%. https://www.spectator.co.uk/article/covid-19-and-the-false-positive-trap
Yes, sounds good doesn't it? Only 1% false positives. Very good - if this is applied to the number of cases 'found'. Unfortunately, it applies to the whole number of tests. If you find 1% of all the tests register positive, they could all be false.
These tests are not sensitive enough for a low number of cases. It is like measuring inches with a ruler calibrated in feet.
yes. i do not get it. does not ring true. hence i am using my lockdown time to research it and re-learn much stuff from university.
Hancock’s numbers are wrong (though I suspect it is his terminology that was wrong).
But the numbers in Lockdown Sceptics article are also wrong. The numerator for the false positive rate is never ‘all the people you’ve tested’. The denominator (not the numerator) for the false positive rate is all the people tested who are actually negative. The numerator is those of them who falsely test positive.
If the test has a false positive rate of 0.8% (=a specificity of 99.2%), that means that out of 1,000 people tested who are actually negative, 8 will test positive, falsely.
If the infection rate is 11 in 10,000, how many of the infected people will test positive?
You haven’t said what the sensitivity of the test is but the false negative rate is usually quite high eg 9% (= a sensitivity of 91%) so out of 11 actually infected people, 1 will test negative falsely.
If 10,000 tests are done, this will be the outcome:
Tests done 10,000
Infection rate 0.11%
Number of infections 11
Sensitivity 91.00%
True positives 10
False negatives 1
Non-infections 9,989
Specificity 99.20%
True negatives 9,909
False positives 80
So in this example the false positives exceed the true positives by 8:1.
Undated document, but https://www.ouh.nhs.uk/working-for-us/staff/documents/staff-testing-privacy-statement.pdf from Oxford University Hospitals says
"The test is a swab of your nose and throat undertaken at one of the Trust testing sites. The test is not 100% accurate; at present it is thought to detect approximately 70% of coronavirus infections."
....
Same time in recent Spectator column they say less than 1%. https://www.spectator.co.uk/article/covid-19-and-the-false-positive-trap
These are consistent:
- "false positive less than 1% of those tested": <10 in 10,000
- "true positives detected: 70% of those infected". 11 in 10,000 infected, of which 7 or 8 detected.
False positives exceed true positives.






