Airline Pilot Central Forums

Airline Pilot Central Forums (https://www.airlinepilotforums.com/)
-   COVID19 (https://www.airlinepilotforums.com/covid19/)
-   -   Faa recommends Covid testing for crew (https://www.airlinepilotforums.com/covid19/132960-faa-recommends-covid-testing-crew.html)

Minepza 03-05-2021 11:38 AM


Originally Posted by Excargodog (Post 3203235)
Except screening now that the disease is at low prevalence won’t give you meaningful information:

https://www.medmastery.com/guide/cov...vailable-tests

If I’m reading correctly, that’s talking about antibodies tests in that population. I don’t understand though, the infection tests we’ve been using are proven to be near 100% in detecting the virus. Of course it all depends on how well the test was administered but the test itself works. The chance of the test being wrong, if done properly is the exception not the norm. At the end of the day, why wouldn’t you want to know if the other crew member is infected?

flydrive 03-05-2021 11:39 AM


Originally Posted by highfarfast (Post 3203186)
It seems kinda late in the game for this kind of measure.

Just like every other BS policy that has been foisted upon us over the last year by our gutless management with the full support of our virtue signaling unions.

Excargodog 03-05-2021 11:52 AM


Originally Posted by Minepza (Post 3203248)
If I’m reading correctly, that’s talking about antibodies tests in that population. I don’t understand though, the infection tests we’ve been using are proven to be near 100% in detecting the virus. Of course it all depends on how well the test was administered but the test itself works. The chance of the test being wrong, if done properly is the exception not the norm. At the end of the day, why wouldn’t you want to know if the other crew member is infected?

You ARE incorrect. Both antibody AND PCR rests have less than 100% specificity and sensitivity ESPECIALLY for newly infected people, in fact, both say that multiple tests over a 2-3 day period may be necessary to ibtain a positive.

https://www.nejm.org/doi/full/10.1056/NEJMp2025631

and the readily available BINAX test is remarkably unsuited for screening:


https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e3.htm

Don’t get me wrong, you are ENTITLED to not understand this, but the government ought to know better.

But look at Bayes Theorem and plug in the current prevalence rate of new infections and do the math. Don’t take my word for it. It ain’t rocket science.

Regularguy 03-05-2021 12:06 PM

I don't get it?

Have we programmed so many people to simply say, "yes this must be good it comes down from the "experts," so let's do it."

Here's the answer which ALPA and all pilots should be saying loud and clear;

NO!

Enough is enough!!


This isn't that different than what the China Government is doing to foreign crews and others these days:
https://www.newsweek.com/fact-check-...-china-1573912

501D22G 03-05-2021 12:08 PM


Originally Posted by Excargodog (Post 3203217)
The problem isn’t difficulty setting up - the problem is the predictive value positive and negative of screening testing. It isn’t that great to begin with and as the prevalence decreases (which it certainly is doing currently) the predictive value gets even lower.

Google (or YouTube) Bayesian Law of conditional probability.

And EARLY in an infection ALL the tests have poor to negligible predictive value positive.

Could not agree more.

Early on, and late in the game, PPV is poor. In the middle, or near the peak of the "bell curve" if you will, PPV goes up while NPV actually decreases.

At this stage in the game, false positives would simply create a logistical nightmare if tests were used in this manner (whilst providing little practical benefit to sToPpInG tHe sPrEad)

You know no one is going to Google it and do actual reading... So here's a vidya

https://youtu.be/NSRK41UbTEU?t=173

galaxy flyer 03-05-2021 12:08 PM

The flight department I an associated with tests prior to every trip and provides at hone test on return. It’s theater, that’s all, but theater is what government is about.

Minepza 03-05-2021 01:59 PM


Originally Posted by Excargodog (Post 3203251)
You ARE incorrect. Both antibody AND PCR rests have less than 100% specificity and sensitivity ESPECIALLY for newly infected people, in fact, both say that multiple tests over a 2-3 day period may be necessary to ibtain a positive.

https://www.nejm.org/doi/full/10.1056/NEJMp2025631

and the readily available BINAX test is remarkably unsuited for screening:


https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e3.htm

Don’t get me wrong, you are ENTITLED to not understand this, but the government ought to know better.

But look at Bayes Theorem and plug in the current prevalence rate of new infections and do the math. Don’t take my word for it. It ain’t rocket science.


From Harvard health below it says positive tests are almost always correct. Negative tests may be less definitive. I guess what I’m asking is if it’s known that the accuracy is near 100% meaning 85% or more, why assume that none of it works. Let’s be realistic, the chance of you driving up to a test site and knowing if you have it or not is very good. Everyone I know including me that has tested can agree with this.


https://www.health.harvard.edu/disea...he-coronavirus


How reliable are the tests for COVID-19?

Two types of diagnostic tests are currently available in the US. PCR tests detect viral RNA. Antigen tests, also called rapid diagnostic tests, detect specific proteins on the surface of the coronavirus. Antigen test results may come back in as little as 15 to 45 minutes; you may wait several days or longer for PCR test results.

The accuracy of any diagnostic test depends on many factors, including whether the sample was collected properly. For PCR tests, which are typically analyzed in a laboratory, test results may be affected by the conditions in which the test was shipped to the laboratory.

Results may also be affected by the timing of the test. For example, if you are tested on the day you were infected, your test result is almost guaranteed to come back negative, because there are not yet enough viral particles in your nose or saliva to detect. The chance of getting a false negative test result decreases if you are tested a few days after you were infected, or a few days after you develop symptoms.

Generally speaking, if a test result comes back positive, it is almost certain that the person is infected.

A negative test result is less definite. There is a higher chance of false negatives with antigen tests. If you have a negative result on an antigen test, your doctor may order a PCR test to confirm the result.

If you experience COVID-like symptoms and get a negative PCR test result, there is no reason to repeat the test unless your symptoms get worse. If your symptoms do worsen, call your doctor or local or state healthcare department for guidance on further testing. You should also self-isolate at home. Wear a mask when interacting with members of your household. And practice physical distancing.

Duckdude 03-05-2021 02:15 PM


Originally Posted by CX500T (Post 3203112)
Here's an issue. I'm a commuter. I get a positive test on sign in.

Where the heck do I stay the next days?

What if its a false positive.

PcR will show positive 90-120 days post covid.

Sent from my SM-G965U1 using Tapatalk

I see your point and agree your scenario is concerning. My understanding is false positives are pretty rare, false negatives not so much.

So let’s say that sniffle you had in the morning wasn’t allergies like you thought, but a super mild case of COVID and the pre-departure test catches it. In your view, the concern is where to stay, how to get home, etc. What if you didn’t have the test, and you unknowingly gave it to your F/O, purser, three hotel van drivers, and one bar tender? So is it better for you to have to figure out how to get home, or for those six people to get sick?

Oh yeah, you’re certainly right about a false positive - unlikely, but it would suck if it happened.

Minepza 03-05-2021 02:34 PM


Originally Posted by Duckdude (Post 3203302)
I What if you didn’t have the test, and you unknowingly gave it to your F/O, purser, three hotel van drivers, and one bar tender? So is it better for you to have to figure out how to get home, or for those six people to get sick?

.

Right. If you happen to test right before there was detectable amounts of the virus then there’s nothing anyone can do. But at least if you really do have it and are contagious, you’d at least know that before you infect other coworkers.

ugleeual 03-05-2021 02:38 PM

China style anal swabs at KCM?


All times are GMT -8. The time now is 06:23 PM.


User Alert System provided by Advanced User Tagging v3.3.0 (Lite) - vBulletin Mods & Addons Copyright © 2024 DragonByte Technologies Ltd.
Website Copyright ©2000 - 2017 MH Sub I, LLC dba Internet Brands