Corona Virus

[Ken]

Asymptomatic cases are hard to track, but I know the CDC, for instance, does not just report confirmed cases;

The article you referenced noted that the asymptomatic cases were not included for the flu:

An infection-fatality rate of roughly 0.6% is six times greater than the 0.1% estimate for seasonal influenza, which is based on CDC data. Though researchers point out the estimates are calculated in different ways and the flu estimate doesn’t take asymptomatic cases into account.

[Sour Masher]

Yeah, there was a guy I went to high school with who was in full on "End of the World" hysteria with Harold Camping back in 2011. He then would more or less post Bible verses and was harmless, but then he started in about the evils of gay marriage and whatnot. So I picked a fight and then he started deleting some of my comments, including one where I said "sorry dude, you're wrong about this one, just like you were wrong about the end of the world."

History is written by the winners! And everyone is a winner on FB!

[The Feral Slasher]

History is written by the winners! And everyone is a winner on FB!

Now i know why I am such a looser

[Sour Masher]

Now i know why I am such a looser

I don't know about that, but I do know you likely didn't win any spelling bees .

[The Feral Slasher]

I don't know about that, but I do know you likely didn't win any spelling bees .

irony is hard to explain on line....but good catch

[rhd]

Here's some very encouraging info:

https://abcnews.go.com/Health/wireStory/virus-antibodies-fade-fast-necessarily-protection-71901700

And:

https://www.sciencedaily.com/releases/2020/07/200716101536.htm

The 1st article says that altho the antibodies from a recovered CV-19 patient may last only a few months (the article gave an avg half-life of 73 days) the protection from reinfection may last far longer because the body's B-cells, which make antibodies, remember the pathogen and can make new antibodies upon reinfection. The 2nd article says that a recent Singapore study found that 1) 100% of patients tested that were infected w SARS (also a coronavirus) 17 years ago still have T-cells that are specific to the SARS virus; 2) these T-cells display cross-immunity against CV-19; and 3) 50% of patients that were never infected w CV-19 have T-cells specific to CV-19. It said that this immunity in uninfected patients could be because of prior exposure to other coronaviruses, such as common cold viruses. Elsewhere, I read that common cold viruses probably dont convey this immunity but perhaps other coronaviruses do. Also elsewhere, I learned that evidently if the pathogen-specific T-cells are present, they can make the B-cells that can make the antibodies to fight the pathogen. This is very good news re: not only the ability of many people to be protected from infection but also the development of effective vaccines and also the potential long-term nature of immunity from infection among those who have it.

[Sour Masher]

Even with increased testing, the CDC's newest study suggests the actual number of COVID cases is between 6-24 times higher than the reported cases, depending on the area. That would be more evidence reinforcing a lower mortality rate that we once feared, and move us closer to lower peaks of new cases going forward. It speaks to just how incredibly infectious covid is, though. Still, if the antibody data is true and people do keep immunity (or at least the majority of them do), the more people that have gotten it the less people can get it again. I don't think it likely that over 80 million have gotten it, based on the variation between their estimates in NYC and San Fran, but perhaps 40 million, as I've seen other studies suggest a 10x multiple as well.

If 40 million really have gotten it, and we assume 3/4 of them are now immune, at least for the next year or so, then close to 10% of the US population is now immune. That number is still far short of herd immunity, of course, and we will need a vaccine to get to that, if we ever do, but it does give some hope that the peaks we are seeing might decrease a little going forward, now that just about every state has been hit hard by this. Eventually, at least, the peak active cases have to stop setting new records, right, as more and more people get this? I got to hang my hat on some hope, as I don't have much hope in changing behavior in the US.

https://www.statnews.com/2020/07/21/...ovid-19-cases/

[Ken]

Even with increased testing, the CDC's newest study suggests the actual number of COVID cases is between 6-24 times higher than the reported cases, depending on the area. That would be more evidence reinforcing a lower mortality rate that we once feared, and move us closer to lower peaks of new cases going forward. It speaks to just how incredibly infectious covid is, though. Still, if the antibody data is true and people do keep immunity (or at least the majority of them do), the more people that have gotten it the less people can get it again. I don't think it likely that over 80 million have gotten it, based on the variation between their estimates in NYC and San Fran, but perhaps 40 million, as I've seen other studies suggest a 10x multiple as well.

If 40 million really have gotten it, and we assume 3/4 of them are now immune, at least for the next year or so, then close to 10% of the US population is now immune. That number is still far short of herd immunity, of course, and we will need a vaccine to get to that, if we ever do, but it does give some hope that the peaks we are seeing might decrease a little going forward, now that just about every state has been hit hard by this. Eventually, at least, the peak active cases have to stop setting new records, right, as more and more people get this? I got to hang my hat on some hope, as I don't have much hope in changing behavior in the US.

https://www.statnews.com/2020/07/21/...ovid-19-cases/

Aren't there obvious boundaries on the high end that this surpasses? For example, in the US the positive test rate is around 8% (50mm tests, 4mm positive). Of the population who has chosen to get tested, 8% have tested positive.

You are saying there's a possibility that 12% of the US has had the virus? That seems highly improbable.

If a completely random sample of people were to be tested, and the true population had 12% who had been infected, you'd expect around 12% to come back positive. The fact that only 8% of those tests come back positive means that you would be far less likely to go in and get a test if you have it than if you do not. That's not logical IMO.

[Sour Masher]

Aren't there obvious boundaries on the high end that this surpasses? For example, in the US the positive test rate is around 8% (50mm tests, 4mm positive). Of the population who has chosen to get tested, 8% have tested positive.

You are saying there's a possibility that 12% of the US has had the virus? That seems highly improbable.

If a completely random sample of people were to be tested, and the true population had 12% who had been infected, you'd expect around 12% to come back positive. The fact that only 8% of those tests come back positive means that you would be far less likely to go in and get a test if you have it than if you do not. That's not logical IMO.

I think some of the data might be skewed by health care and other workers that take frequent tests (like baseball players now, only a much larger population). There are folks who do not think they have it that are required to test regularly. Does this skew things 4%? I don't know. Probably not. But I think it does skew things some. I know folks who have not gone into get tested, even though they should have, and I know folks who get tested all the time who don't really want to.

ETA: I do not know and can't find just how many places follow the CDC guidelines for frequent testing of health care workers and nursing home staff. I know some do, but I can't find national numbers. It is possible the repeat testing of certain workers for precautionary reasons is not statistically significant...I will keep looking for numbers.

[Ken]

I think some of the data might be skewed by health care and other workers that take frequent tests (like baseball players now, only a much larger population). There are folks who do not think they have it that are required to test regularly. Does this skew things 4%? I don't know. Probably not. But I think it does skew things some. I know folks who have not gone into get tested, even though they should have, and I know folks who get tested all the time who don't really want to.

That doesn't explain it though. It shouldn't ever skew the data PAST the actual positive percentage. It's like regression to the mean, given enough samples (and millions is high enough), it will only push the data toward the true percentage, never past.

Your high end on infection is 8% IMO and even that is dubious, it suggests we are a bunch of monkeys randomly getting tested rather than being more likely to be tested if we are positive. The true high end is likely closer to ~6-7% IMO, so your original discussion was off by at least a scale of 2x.

[Sour Masher]

That doesn't explain it though. It shouldn't ever skew the data PAST the actual positive percentage. It's like regression to the mean, given enough samples (and millions is high enough), it will only push the data toward the true percentage, never past.

Your high end on infection is 8% IMO and even that is dubious, it suggests we are a bunch of monkeys randomly getting tested rather than being more likely to be tested if we are positive. The true high end is likely closer to ~6-7% IMO, so your original discussion was off by at least a scale of 2x.

We don't have to be random monkies to get to the same place random monkey would get to. Sure it makes sense that those who believe they are infected are tested at a higher rate than those that are not. However, in the opposite direction, it is possible that there are a significant number of tests mandated for healthy individuals that even that out. You re probably right that 8% is the cap, but I would not be surprised if the true infection rate was every bit of that given how many people do not get tested, even when they have symptoms.

There are folks that are less likely to get tested WITH symptoms than some folks without them. Again, I don't know if those numbers even out, but anecdotally, I know of folks who will hide symptoms to go to work (I am not defending that behavior, of course), and I know folks who get tested regularly who have never tested positive, as a precaution for their work. And there are millions of people who still don't have access to tests, even if they wanted to take them.

Anyway, stop trying to take my hope away . Let's put the number at 6%, that is still about 5 times the number of cases, so it still is enough to lessen the peaks a little going forward. It seems clear to me that people are not going to stop coming back out and acting recklessly every time we beat this thing back. I am taking some hope that even with that fact, the peaks will grow a bit smaller (5-10% smaller each time) going forward.

ETA: Or maybe the peaks will simply stay the same or grow less than feared, once schools reopen. Who knows. I'm looking for a silver lining.

[Sour Masher]

Speaking of schools, the CDC director recently said he feels comfortable with his grandkids going back to school in the Fall, fwiw: https://www.msn.com/en-us/health/med...ol/ar-BB172I60

As the new school year nears, Redfield said he would "absolutely" be comfortable with his grandchildren heading back to their classrooms. Redfield added he only may have "some reservation" about his grandson with cystic fibrosis, "depending on how he could be accommodated in the school."

"I think it's really important to get our schools open," he said. "It's not public health versus opening the schools or the economy -- it's public health versus public health. I think there really are a number of negative public health consequences that have happened to our K-12 [students] by having schools closed."

"So it's so important now to work together with school districts to figure out how they can take our guidelines and operationalize them in a practical way and to do it in a way that is safe," he continued."

[Ken]

We don't have to be random monkies to get to the same place random monkey would get to. Sure it makes sense that those who believe they are infected are tested at a higher rate than those that are not. However, in the opposite direction, it is possible that there are a significant number of tests mandated for healthy individuals that even that out.

How would that even it out? Sorry from a data perspective this discussion is bugging me. (Although I enjoy discussing data any time -- not "upset" in case tone not clear...)

Lets assume that we have a set of individuals that are more likely to have coronavirus because they have symptoms. They get tested.

Then lets assume we have another set of individuals that gets "mandated". The "mandated" individuals should follow the general population numbers (unless you are suggesting they are somehow less likely to have it - if so... why).

So ((symptomatic - i.e. higher than population % * symptomatic count) + (mandated - i.e. population % * mandated count)) / (test count) can never be < population %. It can only be higher.

[Sour Masher]

How would that even it out? Sorry from a data perspective this discussion is bugging me. (Although I enjoy discussing data any time -- not "upset" in case tone not clear...)

Lets assume that we have a set of individuals that are more likely to have coronavirus because they have symptoms. They get tested.

Then lets assume we have another set of individuals that gets "mandated". The "mandated" individuals should follow the general population numbers (unless you are suggesting they are somehow less likely to have it - if so... why).

So ((symptomatic - i.e. higher than population % * symptomatic count) + (mandated - i.e. population % * mandated count)) / (test count) can never be < population %. It can only be higher.

I defer to you on data stuff, so I take no offensive if you punch holes in this logic. But here goes:

1. I am saying that someone who is tested multiple times that is asymptomatic skews the data. Does it not? If someone who does not have covid and does not think they have covid gets tested several times, that is several negatives for one person. Yes, eventually, assuming they get sick at the same rate as others, they will test positive at the same rate as other populations, but that might be after dozens of negative test results. Would not the infection rate per test, not per person, be far, far lower if everyone in the US got tested each week, regardless of symptoms?

2. I am also saying that there are people who are symptomatic who refuse testing, because they want to keep working don't trust testing, etc. That behavior is also different from what you are assuming, right?

Both those things skew the data in the opposite direction of the skewing that happens because most people get tested because they fear they might have covid or been exposed to covid, no?

[Sour Masher]

My thinking is that there is a difference between infections per test and infections per person. We could theoretically get to a point of 100% infection and 10% of tests being positive, if everyone who is infected gets tested 10 times and each has only 1 positive test. That is why I think that workers that get tested frequently could skew the data. But again, if I am seeing this wrong, it won't be the first time I've goofed in my logic.