gregf83 

It doesn't sound like there's an issue with the test, rather they were premature in claiming the presence of antibodies ensures immunity. It likely does but that has yet to be proven.

gregf83 

It's pretty close though. It correctly detected 88 out of 88 positive samples.

tds101 

Unfortunately I have no confidence in CityMD at this point. Besides, Quest Diagnostic is right in my neighborhood.

gregf83 

Very few of these tests have been verified with the normal rigor and the pressure to increase testing capacity may contribute to 'user error' or sample contamination. It sounds like either:

tds101 

1 I definitely had Covid 19.
2 It was definitely a false negative
3 I'm losing confidence in the current tests available, and this whole situation is too mismanaged to be 100% conclusive at this stage.

wgscott

It is hard to get the antibody tests to behave correctly. The differences between background and positive signal are quite small, and calibration is still a big issue. In principle, they should be as reliable as a pregnancy ELISA, but they aren't quite there yet, and a lot of technical user error can be present.

tds101 

I just plan on waiting until the end of June/early July to get retested. I don't trust the tests at all right now.

CycleryNorth81

The link you provide says "Wuhan now plans to test all 11 million residents after 6 new coronavirus cases were reported". So the Chinese government is going to spend all that time and energy because they don't have anything better to do. It's a hoax/joke that the Chinese government fell for, right?

CycleryNorth81

It is a conspiracy right? The communist leaders and all of the western leaders are in on this corona virus thing. Talk about world cooperation.

Trevtassie

If anything, your "Dear Leader" is desperately trying to undercount excess deaths, as are a few Governors. Problem is dead bodies don't lie. A quick look at excess deaths on the CDC website says a whole bunch more people than average died recently.... https://www.cdc.gov/nchs/nvss/vsrr/c...ess_deaths.htm Luckily most Americans have a degree of commonsense and self preservation and have been ignoring the "Dear Leaders" command to get back to work for the benefit of the countrty his billionaire mates, So they're responsible for keeping the death toll down. It's only the stupid shouty ones who think it's all some kind of media beat up/hoax/new world order plot and are out doing the bidding of their billionaire masters by protesting like good little sock puppets.

tds101 

Stay out of my thread,... I don't need the propaganda.

SHBR

Ok.

Good luck sorting out "the propaganda."

StanSeven

SHBR, leave this thread. That means no more posting here. If you want to make political comments, go to P&R

wgscott

I checked with my wife. She says the most reliable test right now is Roche, followed by A bb ot labs. My understanding is that these work essentially by detecting whether you have antibodies that bind to the virus spike protein. The problems seem to arise (when they do) from calibration, and the degree of signal you need to be considered a positive response. It also assumes that everyone who raises an antibody response does so to the spike protein, and I am not clear on how good an assumption that might be. There are also at least two different strains in NYC that have two different spike protein sequences.

tds101 

I need to take the Roche test,... The ****tt test didn't register correctly for me apparently.

wgscott

tds101 Did you see this?

Antibody tests for Covid-19 wrong half the time, CDC says

https://www.cnn.com/2020/05/26/healt...ong/index.html

I think this is primarily about false positives, rather than negatives, but it still makes me wonder...

Biker395 

Isn't that title grossly misleading? The title appears to be based on this statement from the CDC:

"For example, in a population where the prevalence is 5%, a test with 90% sensitivity and 95% specificity will yield a positive predictive value of 49%. In other words, less than half of those testing positive will truly have antibodies. Alternatively, the same test in a population with an antibody prevalence exceeding 52% will yield a positive predictive greater than 95%, meaning that less than one in 20 people testing positive will have a false positive test result."

That's just an example to illustrate how when you are trying to measure a small prevalence, the accuracy of the test can make the results rather meaningless.

https://www.cdc.gov/coronavirus/2019...uidelines.html

Not all the tests are as inaccurate as described in the CDC's example. The CDC recommends:

"Choosing a test with a very high specificity, perhaps 99.5% or greater, will yield a high positive predictive value in populations tested with prevalence >5%."

And if this is accuate, both <Costello's sidekick> and Roche have tests that appear to qualify:

https://www.evaluate.com/vantage/art...tibody-testing

CycleryNorth81

Questions about COVID-19 test accuracy raised across the testing spectrum

https://www.msn.com/en-us/health/hea...aZB?li=BBnb7Kz

livedarklions

That "prevalence >5%" is a very important caveat, though. Let's say you're testing a population of 1000 people with an actual prevalence of 2%. If the test is 99.5% accurate, you should get 20 correct positives and 5 false positives. It's a lot better than a coin flip, but there would still be 20% of the positive tests that might cause people to falsely think they have immunity. If the actual prevalence is 1%, that false positive rate increases to 33.3% (5 out of 15 positives being false).

Biker395 

Yep. The smaller the prevalence, the greater the accuracy the test must be to correctly estimate the prevalance.

Speaking of which ... are there any updated estimates for prevalence? I'm sure it varies widely by region, but I am wondering why they focused on the 5% number.

AdkMtnMonster

Base rate bias. I went over this in Post #71 . I also mentioned the better reliability of the Roche ab assay compared to other manufacturers in that post. And then I sort of got told off by the OP in Post #72 .

tds101 

Yeah,...ok,... It's amazing how I'm forced to prove to someone that I'd had the novel coronavirus, and that I'd end up in a dispute concerning the accuracy of the test that I'd taken. Do me a favor please, keep me out of your conversation. The sarcastic wink really helps BTW.

livedarklions

TBH, post 71 definitely fell into the tl/dr category. You took far too many words to explain something quite simple. BTW, I explained the very same thing many days ago in a different thread.

Anyway, I was responding to the numbers asserted in the report that biker quoted, so your reference back is really not helpful.

AdkMtnMonster

I don’t know why you’re so defensive about having Covid or not. Even if all of your tests were negative, you could still actually have Covid. It sounds like you had it for sure, and some other serious lingering effects. Your struggle to get healthy again is not left unnoticed. You don’t have to prove anything to anyone here. If your provider(s) diagnose(d) you as a Covid patient. A diagnosis code of U07.1 is Covid positive, but you’re probably already aware of that.

The little blue winky was because I didn’t really feel like I got told off, especially considering you edited that post to agree with needing the Roche test. Anyway, keep working on getting better. Here’s to health!

tds101