5 ms·

Beyond the reliability of the tests, which has also been questioned of those by Abbot Labs (the same tests administered to the President), the FDA states they s

Beyond the reliability of the tests, which has also been questioned of those by Abbot Labs (the same tests administered to the President), the FDA states they simply need further inquiry. Tests do not inject you with any form of matter or fluid. Why are we completely halting testing, even if it is only 1/3 correct? Just disclose that fact to the testee and/or use 3 tests for a statistically significant result. Does this reek of backdoor capitalism (corruption)?

anonuser123456 6y ago
Do you mean regulatory capture? I don't think so. More likely regulatory morass. FDA is still fighting the battle on thalidomide. But in the process, it's losing the war on many other fronts.
ZitchDog 6y ago
It’s more likely they don’t want testing rolling out too quickly. It would appear as a new wave of infection and spook the all important Stock Markets.
- alistproducer2 6y ago
  Ding, ding, ding we have a winner. This is the same FDA that stopped ethanol producers from making hand sanitizer, at the behest of the purell of course[0]. The feds have done practically nothing for the States making them compete each other for ppe and also confiscated shipments [1] while saying the national stock pile isn't for the States [2]. At this point I hope they just ignore the feds. We might be better off revisiting the articles of confederation at this point. Edit: adding sources 0: https://www.reuters.com/article/us-health-coronavirus-ethanol-sanitizer/u-s-calls-ingredients-in-some-ethanol-based-hand-sanitizers-unsafe-idUSKBN22C0LG https://www.reuters.com/article/us-health-coronavirus-ethano... 1: https://www.snopes.com/fact-check/trump-admin-seizing-ppe/ https://www.snopes.com/fact-check/trump-admin-seizing-ppe/ 2: https://www.politico.com/news/2020/04/03/strategic-national-stockpile-description-altered-after-kushners-remarks-163181 https://www.politico.com/news/2020/04/03/strategic-national-...
  - social_quotient 6y ago
    Any sources for this handy?
    - alistproducer2 6y ago
      Updated to add sources
      - social_quotient 6y ago
        Top notch - thx!
  - greglindahl 6y ago
    [0] includes text like: > In one case, the FDA said it had found significant levels of the carcinogen acetaldehyde in ethanol supplied by a company for use in hand sanitizer, according to a recent email exchange seen by Reuters. Is anyone surprised that using fuel ethanol might not be a good idea for hand sanitizer?
    - FireBeyond 6y ago
      They also stopped distilleries from doing so. Unless the argument there is "you can ingest alcohol made in this facility, but not put it on your skin"...
    - loeg 6y ago
      > Acetaldehyde occurs naturally in coffee, bread, and ripe fruit,[10] and is produced by plants. It is also produced by the partial oxidation of ethanol by the liver enzyme alcohol dehydrogenase Doesn't really sound like a problem in a topical solution like hand sanitizer given people ingest it regularly.
      - greglindahl 6y ago
        The quote said "significant levels" -- before dismissing it as "doesn't really sound like a problem", you might want to look a little harder.
- mrfusion 6y ago
  Any you can blame the media for jumping on positive increases and ignoring increased testing.
taeric 6y ago
I don't think you can just use three tests. Whatever caused the false positive is almost certainly a facet of the person you are testing, such that more testing of a single individual by the same test likely should be ignored. You could do independent tests, though, again it is complicated. The causal factor for the false positive is probably correlated between tests. Such that the math will be more complicated than just the inverse of any false positive rate. (Though, mayhap that is good enough.)
didibus 6y ago
Some statistician correct me if I'm wrong, but that doesn't work. You are assuming the likelihood of a test being wrong on the same individual is random every time it is taken. Which I don't think is something we know yet.
- tankenmate 6y ago
  statistically it depends on whether tests are independent; i.e. if the cause of the false positive is to do with the subject then using three tests won't help. if the cause of the false positive is entirely random then doing three tests will help improve the accuracy. if the correlation between false positive occurrence and the testing parameters are somewhat correlated then it may help but maybe not as much as you'd want.
- loeg 6y ago
  Even worse than that -- even if statistically independent, 3 instances of a 33% individual test chance of true positive does not mean 100% chance of a true positive result after 3 tests. Something like (1 - (1-33%)^3) gives you a 70% chance of true positive after 3 totally independent 1/3 change trials.
specialp 6y ago
Even the Abbott antibody test which is considered a good one is backed by a disturbingly small clinical study. The positive result pool had a sample size of around 75 people, and 1k negatives. I read the reagent insert for this. Then when labs are running this test, they have very poor correlation studies that are essentially useless. Everyone is trying to rush these tests out the door. I agree that they should start running them at scale, but all the data needs to be centralized and analyzed. They also need to bring in a lot of previously RNA tested positive COVID patients and see how many test positive. The problem with all these fragmented efforts with 100 different tests is that we never get the full picture or a good sample pool. I think the FDA does need to apply the brakes here a bit and get this going.

anonuser123456 6y ago

Do you mean regulatory capture? I don't think so. More likely regulatory morass. FDA is still fighting the battle on thalidomide. But in the process, it's losing the war on many other fronts.

ZitchDog 6y ago

It’s more likely they don’t want testing rolling out too quickly. It would appear as a new wave of infection and spook the all important Stock Markets.

alistproducer2 6y ago
Ding, ding, ding we have a winner. This is the same FDA that stopped ethanol producers from making hand sanitizer, at the behest of the purell of course[0]. The feds have done practically nothing for the States making them compete each other for ppe and also confiscated shipments [1] while saying the national stock pile isn't for the States [2]. At this point I hope they just ignore the feds. We might be better off revisiting the articles of confederation at this point. Edit: adding sources 0: https://www.reuters.com/article/us-health-coronavirus-ethanol-sanitizer/u-s-calls-ingredients-in-some-ethanol-based-hand-sanitizers-unsafe-idUSKBN22C0LG https://www.reuters.com/article/us-health-coronavirus-ethano... 1: https://www.snopes.com/fact-check/trump-admin-seizing-ppe/ https://www.snopes.com/fact-check/trump-admin-seizing-ppe/ 2: https://www.politico.com/news/2020/04/03/strategic-national-stockpile-description-altered-after-kushners-remarks-163181 https://www.politico.com/news/2020/04/03/strategic-national-...
- social_quotient 6y ago
  Any sources for this handy?
  - alistproducer2 6y ago
    Updated to add sources
    - social_quotient 6y ago
      Top notch - thx!
- greglindahl 6y ago
  [0] includes text like: > In one case, the FDA said it had found significant levels of the carcinogen acetaldehyde in ethanol supplied by a company for use in hand sanitizer, according to a recent email exchange seen by Reuters. Is anyone surprised that using fuel ethanol might not be a good idea for hand sanitizer?
  - FireBeyond 6y ago
    They also stopped distilleries from doing so. Unless the argument there is "you can ingest alcohol made in this facility, but not put it on your skin"...
  - loeg 6y ago
    > Acetaldehyde occurs naturally in coffee, bread, and ripe fruit,[10] and is produced by plants. It is also produced by the partial oxidation of ethanol by the liver enzyme alcohol dehydrogenase Doesn't really sound like a problem in a topical solution like hand sanitizer given people ingest it regularly.
    - greglindahl 6y ago
      The quote said "significant levels" -- before dismissing it as "doesn't really sound like a problem", you might want to look a little harder.
mrfusion 6y ago
Any you can blame the media for jumping on positive increases and ignoring increased testing.

taeric 6y ago

I don't think you can just use three tests. Whatever caused the false positive is almost certainly a facet of the person you are testing, such that more testing of a single individual by the same test likely should be ignored. You could do independent tests, though, again it is complicated. The causal factor for the false positive is probably correlated between tests. Such that the math will be more complicated than just the inverse of any false positive rate. (Though, mayhap that is good enough.)

didibus 6y ago

Some statistician correct me if I'm wrong, but that doesn't work. You are assuming the likelihood of a test being wrong on the same individual is random every time it is taken. Which I don't think is something we know yet.

tankenmate 6y ago
statistically it depends on whether tests are independent; i.e. if the cause of the false positive is to do with the subject then using three tests won't help. if the cause of the false positive is entirely random then doing three tests will help improve the accuracy. if the correlation between false positive occurrence and the testing parameters are somewhat correlated then it may help but maybe not as much as you'd want.
loeg 6y ago
Even worse than that -- even if statistically independent, 3 instances of a 33% individual test chance of true positive does not mean 100% chance of a true positive result after 3 tests. Something like (1 - (1-33%)^3) gives you a 70% chance of true positive after 3 totally independent 1/3 change trials.

specialp 6y ago

Even the Abbott antibody test which is considered a good one is backed by a disturbingly small clinical study. The positive result pool had a sample size of around 75 people, and 1k negatives. I read the reagent insert for this. Then when labs are running this test, they have very poor correlation studies that are essentially useless. Everyone is trying to rush these tests out the door. I agree that they should start running them at scale, but all the data needs to be centralized and analyzed. They also need to bring in a lot of previously RNA tested positive COVID patients and see how many test positive. The problem with all these fragmented efforts with 100 different tests is that we never get the full picture or a good sample pool. I think the FDA does need to apply the brakes here a bit and get this going.