It's impossible to pin down info. Today Fareed Zakaria interviewed...

[LAS14]

.. an expert who said the rapid tests aren't perfect because 3 out of 100 people will test negative when they are positive. But I've heard forever that the rapid tests are only 85% reliable.

Why the discrepancy?

Can anyone reconcile these "facts?"

tia
las

[Silent3]

1. I think he just make a factor-of-ten error

Because 30% is the false-negative rate I've heard from everywhere else.

I have yet to hear what the false-positive rate is.

[Ms. Toad]

3. False positive is lower - but non-zero

Just ask DeWine, who got out of a visit with the president with a false-positive.

[soothsayer]

2. Others will test positive when they're negative?

[LAS14]

4. I don't think so. Another way to state what I've heard is....

... 85% reliable and 97% reliable. That's also the language experts have been using.

[Silent3]

5. I really wish people would banish the word "reliability" when talking about test accuracy

Without false-negative and false-positive being clearly distinguished from each other, "reliability" is a damn near useless concept.

[PoindexterOglethorpe]

9. Yes. That's a false positive.

Different tests have different numbers of false negatives and false positives. I'm not sure the information on the precise reliability of any of the tests is actually available, although I have not tried to search for it on line.

[genxlib]

6. It changes

Depending on where the patient is in the cycle. Higher false negatives before symptoms but that error rate falls after symptoms set in. That is why it isn’t really suitable for screening.

[LAS14]

8. Thanks! Interesting. nt

[BootinUp]

7. The reliability is just calculated differently I assume

And covers other possible problems.

[GeorgeGist]

10. I think what he's trying to say is ...

If you test 100 people who are known to be positive
3 will test negative with the rapid test.

Those are called 'false negatives'.