4 ms·

Thus the need for cheaper, quicker tests, even if the accuracy is t that great. TWIV Podcast episode #640 covers this. $1 per day test that quickly gives a resu

Thus the need for cheaper, quicker tests, even if the accuracy is t that great. TWIV Podcast episode #640 covers this. $1 per day test that quickly gives a result, but perhaps is only 50% accurate.

watwut 6y ago
The tests are quick. The delay is because labs are overwhelmed.
stonogo 6y ago
If that's the test you want, you can literally flip a coin.
- dap 6y ago
  A coin flip won’t be nearly 50% accurate.
  - lostlogin 6y ago
    ELI5 - If the test has 50% accuracy and is producing equal false positives and false negatives, is the coin producing inferior results?
    - ls612 6y ago
      No because Pr(Infected) << Pr(Not Infected) even in the places with the most spread right now.
    - dap 6y ago
      Suppose you test a population of 100 people in which 10 are infected. The coin flip test reports 50 people infected on average. Among those positive tests, at least 40 have to be to be false positives Because there are only 10 infected to begin with. That’s a false positive rate at least 80%. Among negatives, we would expect 5 to be false (since 10 people are infected and each of them have a coin flip chance of testing negative) or 10%. I’m not sure that you can have a test that’s 50% accurate and errs equally in both directions. At least, I’m not sure what it would mean.
- aquadrop 6y ago
  That's exactly was is talked about in the podcast by the doctor. It's not random 50% accuracy, it catches all the people it needs to catch, because accuracy depends on amount of viruses they shed. Accuracy on people who are transmitting decease is more than enough.
acqq 6y ago
My post with the summary, 7 days ago: https://news.ycombinator.com/item?id=23881080 https://news.ycombinator.com/item?id=23881080 "Apparently even the "less sensitive" tests are in fact sensitive enough to catch these who produce a lot of virus. Those who produce a lot of virus have many orders of magnitude more virus RNA present in their samples than those who are in the phase of infection where they aren't so dangerously infective." Thinking more about it, the most important point I'd add now is: We should learn not to think about "the test" but about (at least) two different kinds of tests for different purposes: 1) fast tests to reduce the chance of virus spreading. Anybody isolated before managing to spread the virus is better than not testing, so it doesn't have to be too precise. We just have to understand its limits. 2) more reliable tests for clinical purposes, whenever they are needed. Interestingly, even in such scenarios antibody tests could find their use and allow cheaper or faster results, something like this: https://www.clinisciences.com/en/read/newsletter-26/sars-cov-2-covid-19-diagnosis-by-2264.html https://www.clinisciences.com/en/read/newsletter-26/sars-cov... E.g. if somebody has symptoms and IgM (or IgM and IgG), that could be good enough of information to decide how somebody is to be treated. One more example where I'd guess nobody needs to wait for a PCR result.
lern_too_spel 6y ago
The current positive rate in the US is 8.5%. A test that always returns negative would be more than 90% accurate but completely useless.
Stevvo 6y ago
I've lost count of the amount of times I've been tested, and it never took more than 15 minutes. There are plenty of quick and cheap tests out there.
- HaloZero 6y ago
  I think quick is more about getting the results back quicker. But in terms of taking the test, it takes time to get to a testing site. A take home test would be a huge boon.
- jader201 6y ago
  Curious, if you can explain without going into any sensitive details: 1) Why are you getting tested so often? 2) Where/how are are you getting tested and getting quick results?
  - xenospn 6y ago
    Assuming they're a healthcare worker. My friend who works for a plastic surgeon gets tested all the time.
  - oneplane 6y ago
    I'm not the person you asked, but I also get tested often and quickly due to some work on physical access controls. Often being: once a week. Not considered often compared to physical contact workers, but considered often compared to work-from-home workers. Heck, even considered often compared to the entire continent of North America. Quickly being: drive-in test or walk-in test that is done in 5 minutes, results within 3 days, usually within 24h, either by SMS text message or phone call (depending on the test center). Test center appointments are generally divided based on proximity, there are 4 test centers nearby for me (4 kilometres to max 30 kilometres -- sadly the one closest is usually queued up full for 1 or 2 days). So far my experience has been that they are very efficient; a lot of open space, about 6 people physically there at the most, testing room is about the size of a shipping container, one person doing the tests, one person observing/checking/entering data. The most efficient one was actually in a parking lot next to the hospital that also has a lab where they do mass testing. There is space for about 30 cars, but they regulate the appointments in a way that even when some people come early only 4 cars are there. Half the spaces are blocked/crossed out, entry and exit are made using ramps and zigzags (if that makes sense) so they are very accessible but also naturally prompt people to keep their distance (aside from the stickers/billboards all over the place). Because it's a single-direction flow for both cars and people and the people flow exit ends where the car parking flow starts you never run in to anyone getting in or out.

Thus the need for cheaper, quicker tests, even if the accuracy is t that great. TWIV Podcast episode #640 covers this. $1 per day test that quickly gives a result, but perhaps is only 50% accurate.

watwut 6y ago

The tests are quick. The delay is because labs are overwhelmed.

stonogo 6y ago

If that's the test you want, you can literally flip a coin.

dap 6y ago
A coin flip won’t be nearly 50% accurate.
- lostlogin 6y ago
  ELI5 - If the test has 50% accuracy and is producing equal false positives and false negatives, is the coin producing inferior results?
  - ls612 6y ago
    No because Pr(Infected) << Pr(Not Infected) even in the places with the most spread right now.
  - dap 6y ago
    Suppose you test a population of 100 people in which 10 are infected. The coin flip test reports 50 people infected on average. Among those positive tests, at least 40 have to be to be false positives Because there are only 10 infected to begin with. That’s a false positive rate at least 80%. Among negatives, we would expect 5 to be false (since 10 people are infected and each of them have a coin flip chance of testing negative) or 10%. I’m not sure that you can have a test that’s 50% accurate and errs equally in both directions. At least, I’m not sure what it would mean.
aquadrop 6y ago
That's exactly was is talked about in the podcast by the doctor. It's not random 50% accuracy, it catches all the people it needs to catch, because accuracy depends on amount of viruses they shed. Accuracy on people who are transmitting decease is more than enough.

acqq 6y ago

My post with the summary, 7 days ago: https://news.ycombinator.com/item?id=23881080 https://news.ycombinator.com/item?id=23881080 "Apparently even the "less sensitive" tests are in fact sensitive enough to catch these who produce a lot of virus. Those who produce a lot of virus have many orders of magnitude more virus RNA present in their samples than those who are in the phase of infection where they aren't so dangerously infective." Thinking more about it, the most important point I'd add now is: We should learn not to think about "the test" but about (at least) two different kinds of tests for different purposes: 1) fast tests to reduce the chance of virus spreading. Anybody isolated before managing to spread the virus is better than not testing, so it doesn't have to be too precise. We just have to understand its limits. 2) more reliable tests for clinical purposes, whenever they are needed. Interestingly, even in such scenarios antibody tests could find their use and allow cheaper or faster results, something like this: https://www.clinisciences.com/en/read/newsletter-26/sars-cov-2-covid-19-diagnosis-by-2264.html https://www.clinisciences.com/en/read/newsletter-26/sars-cov... E.g. if somebody has symptoms and IgM (or IgM and IgG), that could be good enough of information to decide how somebody is to be treated. One more example where I'd guess nobody needs to wait for a PCR result.

lern_too_spel 6y ago

The current positive rate in the US is 8.5%. A test that always returns negative would be more than 90% accurate but completely useless.

Stevvo 6y ago

I've lost count of the amount of times I've been tested, and it never took more than 15 minutes. There are plenty of quick and cheap tests out there.

HaloZero 6y ago
I think quick is more about getting the results back quicker. But in terms of taking the test, it takes time to get to a testing site. A take home test would be a huge boon.
jader201 6y ago
Curious, if you can explain without going into any sensitive details: 1) Why are you getting tested so often? 2) Where/how are are you getting tested and getting quick results?
- xenospn 6y ago
  Assuming they're a healthcare worker. My friend who works for a plastic surgeon gets tested all the time.
- oneplane 6y ago
  I'm not the person you asked, but I also get tested often and quickly due to some work on physical access controls. Often being: once a week. Not considered often compared to physical contact workers, but considered often compared to work-from-home workers. Heck, even considered often compared to the entire continent of North America. Quickly being: drive-in test or walk-in test that is done in 5 minutes, results within 3 days, usually within 24h, either by SMS text message or phone call (depending on the test center). Test center appointments are generally divided based on proximity, there are 4 test centers nearby for me (4 kilometres to max 30 kilometres -- sadly the one closest is usually queued up full for 1 or 2 days). So far my experience has been that they are very efficient; a lot of open space, about 6 people physically there at the most, testing room is about the size of a shipping container, one person doing the tests, one person observing/checking/entering data. The most efficient one was actually in a parking lot next to the hospital that also has a lab where they do mass testing. There is space for about 30 cars, but they regulate the appointments in a way that even when some people come early only 4 cars are there. Half the spaces are blocked/crossed out, entry and exit are made using ramps and zigzags (if that makes sense) so they are very accessible but also naturally prompt people to keep their distance (aside from the stickers/billboards all over the place). Because it's a single-direction flow for both cars and people and the people flow exit ends where the car parking flow starts you never run in to anyone getting in or out.