Dr Birx - HALF of Covid positive tests...are FALSE POSITIVES (and CDC now admits it)
Dr Birx - HALF (or more!) of #Covid tests that test positive are likely to be #FalsePositives.
US ADVISOR Dr Deborah Birx warned you can not assume the Covid test positive result is correct - it is likely to be wrong most of the time - a "false positive".
"If you have 1% of the population infected and you have a test that is only 99% specific, then if you find a positive, then 50% of time it will be a real positive and 50% of time it won't be! [ie, it will be a false positive]".
See full 1.5 hr video for context. Segment starts at 52m37s:
Update: CDC FINALLY admits it too (regarding antibody tests but same problem for PCR).
CDC know this well but did not admit it for months:
All these false positives are not obvious but here's why:
Specificity: probability that a test will be negative when the disease is absent.
Sensitivity: probability that a test result will be positive when the disease is present.
The key factor is PREVALENCE, eg, "If you have 1% of the population infected ".
Birx: “None of our tests are 100 percent sensitive and specific."
And Birx says so here:
A test with specificity of 99% is unrealistic, as in Birx example. But let's assume her numbers are realistic...
Check Dr Birx's numbers for yourself here with a test calculator that does the complicated probabilities math for you (Bayesian stats).
Here's a calculator and follow steps below.
Disease present (Sensitivity 99%): 1, 99 (although a more typical value as used in another version, the NPR calculator below, would be 90%, or 10, 90);
Disease absent (Specificity 99%): 1, 99 (Dr Birx's unrealistic/optimistic value for an almost perfect test);
Prevalence: 1% (Dr Birx'x example for the population prevalence).
Results (Birx example):
True positives: 50% (ie, 50% of those with a result of "positive" will be truly positive for virus markers - they "have" the virus)
False positives: 100 - 50% = 50% (ie, 50% will be FALSE positives - their positive result is false! They likely do NOT truly have the virus).
Tests are more likely to have only 80 - 85% specificity.
Reduce to a more realistic test specificity of 85% (15, 85), and sensitivity 90%, prevalence 1%...
Then you get:
True positives: 5.71 = 6%
False positives: 100 - 6 = 94%
Over 90% are wrong results!
Because the test was developed using a test-tube with 100% virus prevalence as the standard, then as prevalence INCREASES (more true cases in population), the test becomes MORE reliable (but not by much):
If California has, say, 5% prevalence, then above calc (85% specificity) becomes:
True positives: 24%
False positives: 100 - 24 = 76%
76% of positive tests would be FALSE and wrong - about three quarters are wrong.
Another calculator here (from NPR) - cross-check:
"If you used a test with 90% specificity in a population in which only 1% of the people have it...more than 90% of the positive results would be false positives."
Article + calc: https://lnkd.in/gVyKtUb
It says "antibody test" but it applies to any test including PCR swab.
A very easy read and example explanation:
Today’s riddle: A 99% accurate medical test is wrong half the time. Why?
More good videos explaining false positives, and antibody testing:
The Covid test is reliable for true negatives (eg, 99.5%) so if you don't have the (signature of) the virus, the test will reliably tell you that. But it extremely unreliable for positive tests if prevalence is low (eg, 1-5% of popn actually have it). The poor reliability and imperfect sensitivity and specificity of the PCR test (or any test) is a serious issue that undermines their entire narrative of "test test test!" and uses misleading results to justify questionable policy. This is astounding and is not understood or even discussed enough.
#Birx #Covid #FalsePositives #Coronavirus #Pandemic