I hope everyone is okay! Me and a few friends rounded up 19 people in LA to get the antibody test, just out of curiosity. We all got the test from USC for about $95. This isn't exactly scientific but here were are results:

Good news:

I have about twenty friends in LA and antibody testing is pretty easy to get (for $95)

We tallied 19 friends have been tested.
18 positive, majority with no symptoms.
1 negative.

I am positive this has been around for a long time.



https://news.usc.edu/168810/usc-covid-19-antibody-researcher-answers-questions-about-testing-in-l-a-county/

See here. Two fake positives out of almost 400. Test is 90-95% accurate from USC.

It's probable it's more infections than we thought but far less deadly than we thought.

Genome testing has shown that it wasn't here before January.

you knew all 19 people? or just recruited some randomly?

Aggie95 said:

---

you knew all 19 people? or just recruited some randomly?

---

I know almost all of them.

could these friend all gotten it from you?

that's pretty incredible! (that many tested positive, not that you know that many people!)

I need to get the antibody test due to my condition. It would be good to know.

Sarduakar said:

---

could these friend all gotten it from you?

---

All of them? No. Haha

thanks for this. needed some good news on a day like today.

question: blood draw or finger stick?

i like that false positive rate you provided in the link. i hope that's real.

All for this being more widespread without negative consequences. The two main questions are some self-selection through a group of friends and the question of cross-reactivity with similar viruses.

Thanks for bringing some good news. I hope you and your significant other are doing much better!

I think the point of this was to brag about having that many friends.

This makes me want to get tested myself. I got sick (felt like a normal cold) in late February and also traveled in early March. I wonder if I got exposed then.

Holy crap!!

Where in Houston can I get the antibody test?

That does sound like great news.

Has Abbott even commented on trying to get testing throughout the state of Texas?

edit - didn't mean to reply to you beta, meant as a general question

BallerStaf2003 said:

---

I hope everyone is okay! Me and a few friends rounded up 19 people in LA to get the antibody test, just out of curiosity. We all got the test from USC for about $95. This isn't exactly scientific but here were are results:

Good news:

I have about twenty friends in LA and antibody testing is pretty easy to get (for $95)

We tallied 19 friends have been tested.
18 positive, majority with no symptoms.
1 negative.

I am positive this has been around for a long time.



https://news.usc.edu/168810/usc-covid-19-antibody-researcher-answers-questions-about-testing-in-l-a-county/

See here. Two fake positives out of almost 400. Test is 90-95% accurate from USC.

---

90-95% accurate for something that only has only happened 2% of the time is pretty terrible.

Quote:

---

Here's a simple way to look at it. Say you are running a test that gives five falsely positive results in a group of 100 people. That doesn't sound too bad. But consider this. If 5% of those 100 people were actually infected with the coronavirus, you should get five correct test results (true positives), along with the five false positive results.

---

https://www.npr.org/sections/health-shots/2020/04/15/834497497/antibody-tests-for-coronavirus-can-miss-the-mark

yes, we know. nothing is positive to you!!!!

Duncan Downer to the rescue! No good news allowed here

Duncan Idaho said:

---

BallerStaf2003 said:

---

I hope everyone is okay! Me and a few friends rounded up 19 people in LA to get the antibody test, just out of curiosity. We all got the test from USC for about $95. This isn't exactly scientific but here were are results:

Good news:

I have about twenty friends in LA and antibody testing is pretty easy to get (for $95)

We tallied 19 friends have been tested.
18 positive, majority with no symptoms.
1 negative.

I am positive this has been around for a long time.



https://news.usc.edu/168810/usc-covid-19-antibody-researcher-answers-questions-about-testing-in-l-a-county/

See here. Two fake positives out of almost 400. Test is 90-95% accurate from USC.

---

90-95% accurate for something that only has only happened 2% of the time is pretty terrible.

Quote:

---

Here's a simple way to look at it. Say you are running a test that gives five falsely positive results in a group of 100 people. That doesn't sound too bad. But consider this. If 5% of those 100 people were actually infected with the coronavirus, you should get five correct test results (true positives), along with the five false positive results.

---

https://www.npr.org/sections/health-shots/2020/04/15/834497497/antibody-tests-for-coronavirus-can-miss-the-mark

---

So 18 false positives out of 19? That would suggest it has a 5.26% accuracy rate? You're saying it's a possible coincidence that there were that many false positives in this case?

Quote:

---

The company also tested 371 COVID-19-negative patients, with only two false positives.

---

this specificity looks okay, though

can't tell the sensitivity. but that's not the meat of the finding.

Great news!

This could be really great news. Do we know the rate of cross-reactivity with other coronaviruses?

If this only shows us that 18/19 of you and your friends have had 1 of the 4 strains of common cold/coronaviruses over the last year or two, it's not nearly as exciting. I would love for it to be COVID-19 specific!

Mateo84 said:

---

This could be really great news. Do we know the rate of cross-reactivity with other coronaviruses?

If this only shows us that 18/19 of you and your friends have had 1 of the 4 strains of common cold/coronaviruses over the last year or two, it's not nearly as exciting. I would love for it to be COVID-19 specific!

---

Exactly. If there's any cross reactivity at all then this is a completely useless test. And actually ends being damaging, since it would give people false hope and create extra confusion when what we need is clarity and direction.

Until we get an idea of sensitivity/specificity with no cross reactivity involving other strains this will be a "wake me when it's been proven" scenario, pending more information.

If the specificty is high enough, who cares about sensitivity in an 18/19 sample? The only thing sensitivity would say is that maybe that 19th person was actually positive when the test said otherwise. This sounds like a damn good f1 score to me.

The test would have to be dramatically wrong, and USC would need to have not done their own validation of the test, to trigger a number of false positives to change the conclusion that "a ton of people have had this asymptomatically".

WoMD said:

---

Mateo84 said:

---

This could be really great news. Do we know the rate of cross-reactivity with other coronaviruses?

If this only shows us that 18/19 of you and your friends have had 1 of the 4 strains of common cold/coronaviruses over the last year or two, it's not nearly as exciting. I would love for it to be COVID-19 specific!

---

Exactly. If there's any cross reactivity at all then this is a completely useless test. And actually ends being damaging, since it would give people false hope and create extra confusion when what we need is clarity and direction.

Until we get an idea of sensitivity/specificity with no cross reactivity involving other strains this will be a "wake me when it's been proven" scenario, pending more information.

---

Huh? They test the assay against ~300 pre covid samples and get a false positive rate of 1%...how is this useless? Did they specifically exclude anyone that had had a cold in the prior year when validating the test? This seems to be conspiracy level skepticism.

It's useless if it's coming back positive for coronavirus but not COVID19

https://en.m.wikipedia.org/wiki/Sensitivity_and_specificity

This is what I'm wanting to know. Is it specific to COVID 19? Or would other coronaviruses flag as a positive?

cone said:

---

It's useless if it's coming back positive for coronavirus but not COVID19

---

True....but then why do it at all if it provides no answers.

cone said:

---

It's useless if it's coming back positive for coronavirus but not COVID19

---

That's the reason they test it against a large population of pre covid samples..

In this case, the false positive rate was 3 out of 300+ sample. Specificity should be ~99% (no idea what the confidence interval is).

okay i see what you mean, you'd expect a large enough population to also have had the common cold and pop a false positive

400 people should be a large enough population if specifically tested negative for COVID via PCR

cone said:

---

okay i see what you mean, you'd expect a large enough population to also have had the common cold and pop a false positive

400 people should be a large enough population if specifically tested negative for COVID via PCR

---

I don't think these things are perfect yet, probably far from it. That said, the results of all of these, with the exception of the Stanford study, would still be very meaningful even with a high rate of false positives.