Antibody tests often deliver false positives and shouldn’t be used as the basis for decision-making about returning to school or work, the Centers for Disease Control and Prevention said in a recent update.
Countdown to Tweet-contradiction being issued?
It’s the CDC. He undermines the CDC advice as a matter of policy these days.
Wait, you’re saying anti-body testing is complicated?
And more accurate tests will be available to anyone who wants one!
[/s]
Ways to keep a workplace safe:
These precautions are a minimum.
“antibody tests might have a “positive predictive value” of just 49%, the CDC said.”
As usual, every fresh chunk of “information” makes the situation more confusing than ever.
False positives will make planning more difficult, but I’ve said this before and never hear anyone addressing it, false negatives are far more dangerous.
Plus is anyone actually anti-body? I don’t see why we even need a test.
Not enough. You spend 8-9 hours in an enclosed space with an infected, no amount of social distancing and sanitizing will mitigate the risk you take. You simply can’t put non-PPE-armored workers in an enclosed space with an infected person and not expect infection. The respiratory secretion buildup still happens even with workers wearing masks, it just takes longer.
Better than nothing, but more likely tokenism to legalize sacrificing the proles.
Yes, anyone who thinks science is confusing should try being a scientist!
But in the case of this group of antibody tests, the unreliability is a long-standing problem. In a normal administration, the CDC and FDA would have been making this clearer all along.
Even if you have cracked the science, you have the ethics to figure out:
I find it hard to imagine that the downside of ‘immunity passports’ exceeds the downside of shutting down education for the next 2 years.
CDC Cautions Against Using Antibody Tests For Decisions On Schools, Workplaces
Dr. Birx, please pick up the White House courtesy phone.
“Courtesy”? With this White House? More like “Rude and Abusive Phone”, non?
'tis an oxymoron befitting the Trump worshiping morons.
The antibody tests are getting a bad rap. (As I read this, the cautions are aimed at policy decisions, not individual tests.) The challenge is dealing with something that has a 5% false positive rate where only 5% of the subjects are true positive. What this means if that if you test 100000 people, 5000 will come up infected and 5000 will come up false positives. But antibody tests are cheap. If you come up positive, you just do another test. (Question for Dr. Fauci: is there any correlation between two tests on the same person?) The challenge is if you use it to decide, if say, your workforce is immune. (The good news is we’ll soon get a live clinical trial of the immunity of having been exposed, at the expense of school kids, factory workers, and so forth.)
A bit more math: If I take my 5000 false positives from the above example and retest them, I will get 250 false positives. If I do a third test, I will get 12.5 false positives. This is a life and death thing, so I’d think we could come up with enough tests. (Oh, wait. Orange CoronaPOTUS is incurable and fatal to everybody in the same solar system.)
Why not put barriers around work stations? (I’d also use masks.) For a meat packing plant, this seems to be beneficial even without coronavirus; the wrong strain of E.coli or salmonella can ruin your whole day. Put intercoms in so that workers can continue to crack dirty jokes and swap football scores.
My impression is that for at least some of the false positives (cross-reaction to other antibodies) the result is reliably wrong. Some thing for some of the false negatives.
That depends on what you are testing for.
False positive is far more dangerous here, as it’s for antibody presence, because there is presumption of protection & folks let their guard down & do not take minimal protective efforts to not catch virus.
False negative here, would mean person continues to act susceptible, taking all precautions to minimize catchIng virus.
IOWs it’s better in this case to have a false negative.