Accuracy of Coronavirus Tests

[Messerve]

This is a question that has been bugging me, but how do we know that the current coronavirus tests are indicating a positive infection for COVID-19 and not also for the common cold or other virus in the same family? Initially they said it was hard to tell them apart because of their similar appearance, and we all know the common cold is, well, common. Can someone enlighten me on this?

 

[sfs]

This is a question that has been bugging me, but how do we know that the current coronavirus tests are indicating a positive infection for COVID-19 and not also for the common cold or other virus in the same family? Initially they said it was hard to tell them apart because of their similar appearance, and we all know the common cold is, well, common. Can someone enlighten me on this?

The tests are designed specifically to avoid confusion with other coronaviruses. I haven't seen any numbers, but I haven't heard of false positive tests being a problem. A much bigger problem is false negatives -- test results that say you aren't infected when you actually are. They seem to occur in something like 70% of tests. The problem isn't really the test, I think -- it's that the virus lives in the lungs, and swabbing the nose may not pick up much virus.

 

[Tanj]

This is a question that has been bugging me, but how do we know that the current coronavirus tests are indicating a positive infection for COVID-19 and not also for the common cold or other virus in the same family? Initially they said it was hard to tell them apart because of their similar appearance, and we all know the common cold is, well, common. Can someone enlighten me on this?

Further to @sfs comments, COVID19 has stretches of its genome which are unique to it. In much the same way we can tell two humans apart using their DNA despite the fact they are 99% identical, we can find the unique parts of the virus and look at those to determine that it's the virus we actually want to detect.

 

[Messerve]

Thanks to both of you! This makes sense then why it takes a long time to get test results back. I guess it's not really as simple as a chemical responding in a certain way, but they have to look at it closely. And that's a very good point about the swab potentially missing the virus in an infected person....

By the way, may I suggest a more marketable name? We should call this WURS (like 'worse') which stands for WUhan Respiratory Syndrome.

 

[JCFantasy23]

I havent been able to find an article about this but I've heard from two in the medical field the accuracy is about 70%

 

[Messerve]

The tests are designed specifically to avoid confusion with other coronaviruses. I haven't seen any numbers, but I haven't heard of false positive tests being a problem. A much bigger problem is false negatives -- test results that say you aren't infected when you actually are. They seem to occur in something like 70% of tests. The problem isn't really the test, I think -- it's that the virus lives in the lungs, and swabbing the nose may not pick up much virus.

Why don't they have people just cough into something...?

 

[sfs]

Why don't they have people just cough into something...?

Sputum samples are sometimes collected from patients with a cough. But I don't know anything about the logistics and safety issues attached to different kinds of samples.