Reveille said:

---

I hate to bring bad news but there is a lot holes in the Stanford study guys. But also the test they used as specificity of only 90%. This means 1 out 10 will be a false positive but that is when you close to 100% who are actually positive when you have lets use 10% infected the false positive rate is huge. (10%*90*=9.0) So what this means if test 100 people with this test and 10% are actually infected you would have 9 false positives so 19 out of 100 will test positive with the test they used.

In addition they recruited people on facebook which would also lend towards getting more positive tests as people with recent symptoms would be more likely to go get the test done, These are just a few of limitations. We can all hope that these numbers are real so we are closer to herd immunity but most likely they are not.

We all know the denominator is much bigger we just have no idea how much bigger it actually is. We all want it as large as possible.

---

Dr. Birx mentioned this in the press conference today when asked about it. She said based on the specificity of the test and the low prevalence of infection, up to 50% of the people who tested positive for the antibodies could be false positives.

I'm trying to find her exact statement, but she mentioned the importance of using these tests on health care workers. I assume because the risk of exposure is much higher it may be a more accurate way of determining rate of asymptomatic infection and recovery.

https://www.fiercebiotech.com/medtech/current-covid-19-antibody-tests-aren-t-accurate-enough-for-mass-screening-say-oxford

Thought this was pretty funny

k2aggie07 said:

---

Thought this was pretty funny

---

Not that odd. This is the crisis that epidemiologist have been dreaming of for most of their professional careers. It's natural that they don't want to be seen as having overestimated the threat. For some, they may not want to lose their new found time in the spotlight as well.

Also useful
https://www.medrxiv.org/content/10.1101/2020.04.15.20067066v1

Comes with a cool calculator!
https://larremorelab.github.io/covid-calculator1

Edit - and an even cooler calculator that lets you go one level deeper, looking at how the test was calibrated BEFORE the sampling

https://larremorelab.github.io/covid-calculator2

That's a fairly cynical view of people in a profession that strives to cure disease and keep people from dying.

I think his tweet is the opposite of what you are thinking.

I think he means: When crappy, biased studies come out with seemingly good news that this is way less deadly than we thought, I am in the odd position of hoping that study is true, but having to caution that it is bad or incomplete information

A solider dreams of battle, a police officer dreams of the chase, epidemiologists are no different. Why join that career field if you didn't want to fight against a pandemic? A global pandemic with an extremely contagious virus is like the big league for this profession. I have no doubt that they have good intentions, but it is human nature to want to insulate your self from criticism as well. It a good idea to always view the motivations of man with cynicism.

What Randy said. You took it backwards, he's cautioning people against good news, that he hopes is true. That tweet was about this study.

RandyAg98 said:

---

I think his tweet is the opposite of what you are thinking.

I think he means: When crappy, biased studies come out with seemingly good news that this is way less deadly than we thought, I am in the odd position of hoping that study is true, but having to caution that it is bad or incomplete information

---

Are all the studies from different places around the world indicating the same results also " crappy and biased"? The links to those studies have been posted in this thread if you want to read them.

I know what he said. I view it the same as a solider saying he hopes for peace while dreaming of battle.

Again those papers were PCR tests and patient symptom surveys. PCR surveys only give you a snapshot in time, they only show you if the person is currently infected. You then have to do follow-ups to see if those who were infected but not symptomatic ever became symptomatic, or ever were symptomatic.

It's not the same question as what these antibody studies are trying to answer, which is - were you ever exposed and did you recover?

We know, from studies like what you posted and from modeling of the outbreak that there must be a large cohort of undetected cases. No one is arguing that point, it's basically a certainty. But the key question is how large is that cohort? And what is it made up of - age, sex, blood type...?

You're acting as if people are saying - there's not a lot of asymptomatic carriers. No no, we're saying - how many are there? PCR testing can't tell us that.

Sage advice for TexAgs postings as well.

I wasn't commenting on the veracity of any studies. I am actually on the side that believes more people have been exposed and this is less deadly than we were told, and we should start opening things back up.

I was simply stating what I thought he was trying to say (from his point of view) with that tweet.

k2aggie07 said:

---

Again those papers were PCR tests and patient symptom surveys. PCR surveys only give you a snapshot in time, they only show you if the person is currently infected. You then have to do follow-ups to see if those who were infected but not symptomatic ever became symptomatic, or ever were symptomatic.

It's not the same question as what these antibody studies are trying to answer, which is - were you ever exposed and did you recover?

We know, from studies like what you posted and from modeling of the outbreak that there must be a large cohort of undetected cases. No one is arguing that point, it's basically a certainty. But the key question is how large is that cohort? And what is it made up of - age, sex, blood type...?

You're acting as if people are saying - there's not a lot of asymptomatic carriers. No no, we're saying - how many are there? PCR testing can't tell us that.

---

Does that mean all these studies are "crappy and biased", because that was the comment by the other poster I was responding to.

As far as the tweet, I'm always going to be skeptical of a group of people that were comparing this to the Spanish Flu pandemic, and are now criticizing information indicating that its far less severe.

I don't know which studies were crappy and biased. The Stanford one isn't looking good, and the USC study is effectively a replication of it - but with a much better recruitment method. The others are not useful for answering the question, so they're not really relevant. Even if they're wonderful and perfect they can't tell us how many people have already been exposed. It's trying to measure sound with a ruler.

Quote:

---

As far as the tweet, I'm always going to be skeptical of a group of people that were comparing this to the Spanish Flu pandemic, and are now criticizing information indicating that its far less severe.

---

I mean, this is just ad hom again. Bergstrom's been a steady voice, and I don't recall him comparing it to the Spanish flu. In fact, I don't recall any papers showing a 2.5-3% IFR, which would be that severity.

Article about Santa Clara study; explaining it in common terms.

A Better Scientist

Quote:

---

Imagine we are testing 1000 people in an imaginary Santa Clara county.
Now imagine that 1% of the population has had COVID19. That would be around 10 people out of 1000. But, because people who were recently sick are more likely to volunteer for the study, maybe instead of 10, 20 people out of 1000 are positive. That's 2% of the sample.
The other 98% of the sample should have a negative test. But, we know that the false positive rate of this test is between 0.1 and 1.9%, which means you'll get another 1-19 people who test positive even if they never had the disease! Let's assume for now that we get 10 false positives. Now we have in total 30 positive tests out of 1000 people tested. That could lead you to think that 3% of the sample of 1000 people has had COVID19 and thus 3% of Santa Clara county has had COVID19. Even though the real rate in our imaginary Santa Clara example was only 1%!
In the real Santa Clara study, 50 out of 3300 tests were positive (1.5%). In principle, these could all be false positives!

---

Very short explanation is that the results from testing could not give any accurate results beyond setting an approximate upper limit for the percentage of those infected.

VaultingChemist said:

---

Article about Santa Clara study; explaining it in common terms.

A Better Scientist

Quote:

---

Imagine we are testing 1000 people in an imaginary Santa Clara county.
Now imagine that 1% of the population has had COVID19. That would be around 10 people out of 1000. But, because people who were recently sick are more likely to volunteer for the study, maybe instead of 10, 20 people out of 1000 are positive. That's 2% of the sample.
The other 98% of the sample should have a negative test. But, we know that the false positive rate of this test is between 0.1 and 1.9%, which means you'll get another 1-19 people who test positive even if they never had the disease! Let's assume for now that we get 10 false positives. Now we have in total 30 positive tests out of 1000 people tested. That could lead you to think that 3% of the sample of 1000 people has had COVID19 and thus 3% of Santa Clara county has had COVID19. Even though the real rate in our imaginary Santa Clara example was only 1%!
In the real Santa Clara study, 50 out of 3300 tests were positive (1.5%). In principle, these could all be false positives!

---

Very short explanation is that the results from testing could not give any accurate results beyond setting an approximate upper limit for the percentage of those infected.

---

This is what Trevor Bedford was trying to explain in his tweet thread some pages back -- that a small change in the sensitivity, even going from 99% to 98%, means that the entire "positive" population they found could be false positive and true positive rate is 0%!

Thats why clearly describing what the sensitivity of each of these tests along with potential cross-reactivity to other coronaviruses before even getting into the test results, is so critical in trying to determine what they truly mean. It also must be done right!

Nerd fight!

https://news.berkeley.edu/2020/04/24/study-challenges-reports-of-low-fatality-rate-for-covid-19/

Berkley estimates no lower than 0.5% IFR.

Paper:
https://www.medrxiv.org/content/10.1101/2020.04.15.20067074v2.full.pdf

k2aggie07 said:

---

Nerd fight!

https://news.berkeley.edu/2020/04/24/study-challenges-reports-of-low-fatality-rate-for-covid-19/

Berkley estimates no lower than 0.5% IFR.

---

I say we start taking bets here on TexAgs. The over/under is 0.5% IFR.

That's a really good O/U I think. I have no idea which way it is going to break.

I'm hoping that these numbers do pan out and the death rate ia smaller. However, it's largely irrelevant and we should be careful not to fall into this trap. I think some people are using this study as a basis for reopening. And of course there are people using other studies and reasoning to continue the oppression. Frankly, I think it's all irrelevant either way. We were told that the lockdown was to give the hospitals time to prepare.

Therefore, that should be the only major concern. Anything short of those is proof that we were lied to.

I say under.

The hospitals have been ready for a couple weeks here in Houston.

Yes. 100% agree. Any real forward thinking Governor would have been pushing for widespread mass anti-body testing ASAP. I thought so 3 weeks ago. Needs to be specific to this virus tho.

halfastros81 said:

---

Yes. 100% agree. Any real forward thinking Governor would have been pushing for widespread mass anti-body testing ASAP. I thought so 3 weeks ago. Needs to be specific to this virus tho.

---

I think the problem is that we do not have have a proven reliable antibody test yet. When we have a test that we can say has strong accuracy, I think we'll see that. That's more than likely the delay in pushing those tests. At least I'd hope we'd wait until we can do it right.