4 ms·

A company here in Canada tried to bring one to market, but I think it was completely disapproved because it was not precise and specific enough, which baffled m

A company here in Canada tried to bring one to market, but I think it was completely disapproved because it was not precise and specific enough, which baffled me since I don't really see the downside of it being available even if that's the case. Yes, accuracy is a nice thing to have, but if you aren't tested at all you are worse off. You can still follow precautions if you are negative on an inaccurate test, but we can't afford to have everyone take the precautions that those with positive tests should. If the test is highly sensitive but biased toward false positives, that would also be a useful form of test.

jacquesm 6y ago
That it will lead to a lot of follow up tests and people that are ill thinking they are not.
- birken 6y ago
  My understanding of the tests is they are very accurate at determining if somebody has a lot of virus, but not at detecting lower thresholds of virus. So compared to PCR tests, they have a lot of false negatives. The Abbott ID NOW COVID-19 rapid result machine that the white house uses has the same properties: > Based on our findings we could argue that the Abbott ID NOW detects samples with high viral load or possibly viable virus that could be of importance for transmission. But, the fact that it misses positive samples on patients being admitted to the hospital with clinical picture of COVID-19 makes this technology unacceptable in our clinical setting [1] So they will have very few false positives, but a lot of false negatives. However... being negative for having a lot of virus is actually meaningful, whether you have the virus or not, because the best assumption is that contagiousness is related to amount of virus in your system. So basically, the tests aren't perfectly accurate, but they do provide useful information, and they are instant. That is the key. A PCR test that is perfectly accurately is useless for contact tracing/isolation and peace of mind if it takes 14+ days to get the results. Think of a pregnancy test. If it comes back positive, you are very likely pregnant but should get it confirmed by a doctor. If it comes back negative, you likely aren't pregnant, but maybe it's too early to tell. Pregnancy tests are widely used even though they have these characteristics because people understand the quirks and the information they provide is useful and of course, comes instantly. 1: https://www.biorxiv.org/content/10.1101/2020.05.11.089896v1.full https://www.biorxiv.org/content/10.1101/2020.05.11.089896v1....
contravariant 6y ago
If it's not specific enough a significant portion of the population will have to go into quarantine for no obvious reason, and if it's no precise enough you either need to keep the restrictions or let a significant portion of the infected population move around with no precautions whatsoever. I'm not sure why you're not seeing any problems with this.
- phendrenad2 6y ago
  It doesn't have to lead to quarantine. Many people are running around acting as though the virus is gone. Some are staying indoors with a cold thinking they might have the virus. If they could self-test and make their own decision about the risk of changing their behavior, it would help immensely.
ttul 6y ago
Listen to the TWiV episode. The idea behind the cheap test is that it picks up COVID at maybe a billion times greater viral load than an RT-PCR test would. If a billion sounds like a lot, it’s not. The virus goes from undetectable to rampant within hours. The cheap test will miss out the very early stage of infection. But we are talking hours. Such a cheap test could allow kids to go to school or adults to go to work. It would be revolutionary.

jacquesm 6y ago

That it will lead to a lot of follow up tests and people that are ill thinking they are not.

birken 6y ago
My understanding of the tests is they are very accurate at determining if somebody has a lot of virus, but not at detecting lower thresholds of virus. So compared to PCR tests, they have a lot of false negatives. The Abbott ID NOW COVID-19 rapid result machine that the white house uses has the same properties: > Based on our findings we could argue that the Abbott ID NOW detects samples with high viral load or possibly viable virus that could be of importance for transmission. But, the fact that it misses positive samples on patients being admitted to the hospital with clinical picture of COVID-19 makes this technology unacceptable in our clinical setting [1] So they will have very few false positives, but a lot of false negatives. However... being negative for having a lot of virus is actually meaningful, whether you have the virus or not, because the best assumption is that contagiousness is related to amount of virus in your system. So basically, the tests aren't perfectly accurate, but they do provide useful information, and they are instant. That is the key. A PCR test that is perfectly accurately is useless for contact tracing/isolation and peace of mind if it takes 14+ days to get the results. Think of a pregnancy test. If it comes back positive, you are very likely pregnant but should get it confirmed by a doctor. If it comes back negative, you likely aren't pregnant, but maybe it's too early to tell. Pregnancy tests are widely used even though they have these characteristics because people understand the quirks and the information they provide is useful and of course, comes instantly. 1: https://www.biorxiv.org/content/10.1101/2020.05.11.089896v1.full https://www.biorxiv.org/content/10.1101/2020.05.11.089896v1....

contravariant 6y ago

If it's not specific enough a significant portion of the population will have to go into quarantine for no obvious reason, and if it's no precise enough you either need to keep the restrictions or let a significant portion of the infected population move around with no precautions whatsoever. I'm not sure why you're not seeing any problems with this.

phendrenad2 6y ago
It doesn't have to lead to quarantine. Many people are running around acting as though the virus is gone. Some are staying indoors with a cold thinking they might have the virus. If they could self-test and make their own decision about the risk of changing their behavior, it would help immensely.

ttul 6y ago

Listen to the TWiV episode. The idea behind the cheap test is that it picks up COVID at maybe a billion times greater viral load than an RT-PCR test would. If a billion sounds like a lot, it’s not. The virus goes from undetectable to rampant within hours. The cheap test will miss out the very early stage of infection. But we are talking hours. Such a cheap test could allow kids to go to school or adults to go to work. It would be revolutionary.