Mormad said:
Wayland said:
And here we get to the problem of false positives. Other Coronaviruses setting off tests.
EDIT: Apparently there is another coronavirus from a German study that was setting of false positives 7% of the time.
All of this isn't an issue when you are using tests rationally. But when every false positive is used as a cause to shut down schools and businesses, you need to be diligent.
Political implications aside, what do you worry would have a bigger impact on society (not individual), false positives or false negatives? What is the current FP and FN rates of the most commonly used tests? Maybe Davie can help here with the analysis, but using those rates, if an asymptomatic individual tests positive one day, and negative the 2 subsequent days, what's the likelihood the person is actually positive?
I think i know that answer to the first, but I'm not sure of the other questions but i have been thinking about it. Thanks for your and Davie's thoughts about it.
I would have to qualify my answer with "It depends"
The lower the community spread the higher the impact of false positives is especially WHEN you take unreasonable actions and overreact to the point you are closing businesses and schools over a single test.
If someone is FN and asymptomatic, not worried.
It is the secondary actions of actions of the FP that is the problem. When we are in the quick growth stage of the pandemic, maybe we can accept a few FPs. The lower the overall actual REAL number of infections out there, the higher percent of the detected cases are FPs. And again, the problem isn't the FP, it is the gross overreaction to a positive PCR test.
If someone tests positive one day, and then even has only 1 negative test the next day, given the over-sensitivity of PCR testing, I would consider the person non-infectious. The ramp up is pretty quick once positive and for someone to test negative AFTER testing positive and still be infectious would be a pretty remote chance. And I would be comfortable with that assessment with either PCR or paper antigen tests.
The problem again with 'positive', for these arguments 'positive' means HAS COVID19 and is INFECTIOUS.
False positives could be lab error, detection of other corona virus, or detection of past infection (so technically positive but no longer infectious).
There is some value to determining that someone MAY have had COVID19 in the past, but we shouldn't be treating a past infection as an ACTIVE case when detected.
I would love for the media to ask questions on what the false positive rate actually is, or do some similar studies that they are doing in Europe and detecting 3-7% false positive rates when other coronaviruses are mixed into samples.
Which is why we need to be going to the 'less sensitive' tests.