tysker said:

---

Smokedraw01 said:

---

Do our COVID numbers match any of the other countries that have opened up their schools?

---

Depends on which numbers you want to point to:



chart from this article
https://mises.org/power-market/medias-jihad-against-swedens-no-lockdown-policy-ignores-key-facts

So far Texas Florida and California are doing pretty darn well even considering the recent spike or surge or explosion (choose your verb).

---

Again, I'm talking about similar numbers and have opened schools back up.

tysker said:

---

gvine07 said:

---

That's deaths, now look at cases

---

All cases aren't equal. Its intellectually disingenuous to compare cases especially as testing has gotten cheaper, better and faster over time.

---

...so should we totally ignore the number of active cases? That doesn't seem very smart.

I think the deaths per capita is a factor, but if we have 10 times as many cases per capita than those countries when they reopened schools, doesn't that seem intellectually disingenuous to pretend both are the same?

Again, it's nice to have the death rate. But maybe also make sure we don't run the hospitals over capacity.

gvine07 said:

---

tysker said:

---

gvine07 said:

---

That's deaths, now look at cases

---

All cases aren't equal. Its intellectually disingenuous to compare cases especially as testing has gotten cheaper, better and faster over time.

---

...so should we totally ignore the number of active cases? That doesn't seem very smart.

I think the deaths per capita is a factor, but if we have 10 times as many cases per capita than those countries when they reopened schools, doesn't that seem intellectually disingenuous to pretend both are the same?

Again, it's nice to have the death rate. But maybe also make sure we don't run the hospitals over capacity.

---

Active cases, at least how I understand it is being calculated, is a useless stat. It may tell where you were but gives you no guidance. New cases may be a decent measure but even those cant be looked at in isolation; the details matter especially given the number of superspreader events. Several Asian and European countries had schools open back in May/June and the UK is looking to open in Sept. You're going to tell us Europe and Asia were doing "better" than in the US during those months? Maybe it's as simple as education is of higher societal value in those countries than here in the US.

Keegan99 said:

---

"Case counts" are meaningless without correcting for a host of factors, including testing volume, availability, and screening criteria.

---

I get that they're not perfect. But meaningless? Would you argue that they shouldn't be used at all?

Data doesn't have to be perfect for it to be meaningful.

You can use them in a very narrow context, but not without putting them in full context with a host of other factors.

Comparing the numbers across jurisdictions or extended time frames is a huge red flag.

You can look at the "currently infected" plots on https://covid19-projections.com

Those graphs show summaries of estimates of the number of people actually infected at different points in time based on that model. There is substantial uncertainty (and that model only uses death data to estimate the parameters and variables, if I recall correctly). But some sort of estimate (like this one) is probably more reliable than any single raw summary statistic from very different data sources (such as # of confirmed cases when testing depth is very different).

just a couple of quick follow-ups:

Sweden probably had the highest prevalence when they opened education for older kids/young adults. I think mid-June was when Sweden re-opened universities (and schooling for the 16+ year olds, I think). At that point, the covid19-projections model estimates the currently infected to have been around 0.73% (confidence interval = .32% - 1.15%).

It is estimating that on Sept. 1, the US overall will be somewhere around 1.4% infected (confidence interval .79%-2.27%). But of course there is a lot of regional variation. the estimates for TX on Sept 1 are around 2.8% (between 1.07% - 4.64%)

While I do find the "currently infected" estimate to be useful... The best measurement of risk of getting infected when you are out in public, would probably:
1. Up-weight the newly infected people (because infectiousness decreases pretty sharply after one's immune system really kicks in), and
2. Consider the fraction of people who are walking around without knowing they have the virus.

In some places (e.g. S. Korea) the fraction of unwittingly infected is probably really low because they are finding a high fraction of their cases from contact tracing. So, a lot of folks who are testing positive were caught early in their infection. In most of the US, our testing lags and our high prevalence imply that we have a much higher risk of encountering some one who is infected but who doesn't know that they are infected.

Keegan99 said:

---

You can use them in a very narrow context, but not without putting them in full context with a host of other factors.

Comparing the numbers across jurisdictions or extended time frames is a huge red flag.

---

That's my WHOLE point... you can't just look at deaths per million and assume everything else is the same. You have to look at numerous other factors.

Pretending that an estimate of active cases is meaningless infers that it should not be used in decision making, and that would be a mistake.

Keegan99 said:

---

"Case counts" are meaningless without correcting for a host of factors, including testing volume, availability, and screening criteria.

---

I think even then they are probably meaningless. I think the best way to get an idea of how many cases there really are is to take the number of deaths as the numerator and the IFR (currently estimated by CDC at 0.65%, which is probably 2X too high at least) as the denominator. If you do that for NY, you get something like 4.7-11 million cases.

We should have done a random serological test nationwide months ago, so we'd have a better idea of the actual IFR. It is sort of circular without that, but it's par for the course. We've made some extremely bad policy based on bad data for months now.

How can anyone properly define active cases when so many of the population are asymptomatic? Its a guess at best and subject to manipulation.

gvine07 said:

---

Keegan99 said:

---

You can use them in a very narrow context, but not without putting them in full context with a host of other factors.

Comparing the numbers across jurisdictions or extended time frames is a huge red flag.

---

That's my WHOLE point... you can't just look at deaths per million and assume everything else is the same. You have to look at numerous other factors.

Pretending that an estimate of active cases is meaningless infers that it should not be used in decision making, and that would be a mistake.

---

I disagree. The number of real cases is way, way higher than any state or local estimate on active cases. An estimate of real cases would be more informative to the public, because it would drive home that for most people the impact on their lives is nil (i.e., most cases are mild or asymptomatic). What we have instead is mass psychosis where young people are the most afraid, which is ludicrous based on the actual risks. Our opinion leaders are so misinformed that we basically are living in a society that is having a mass psychotic break with reality.

tysker said:

---

How can anyone properly define active cases when so many of the population are asymptomatic? Its a guess at best and subject to manipulation.

---

It's an easy calculation to make. You take number of deaths (which is perhaps the one thing we really do know) divided by the infection fatality rate (which is an estimate, but CDC currently puts it at 0.65%, though I bet the real IFR is more like 0.1-0.2% meaning the number of actual cases in the US is insanely high).

Just because it's not exact doesn't mean it can't be useful.

Can we agree that we can use the estimates to identify trends?

Quote:

---

What we have instead is mass psychosis where young people are the most afraid, which is ludicrous based on the actual risks.

---

Totally off topic -- I've seen the fear of COVID versus age reports (which are backwards from what we'd expect). But I wonder if there's been fear of COVID versus social media platform/news outlet study?

tysker said:

---

Quote:

---

What we have instead is mass psychosis where young people are the most afraid, which is ludicrous based on the actual risks.

---

Totally off topic -- I've the fear of COVID versus age reports (which are backwards from what we'd expect). But I wonder if there's been fear of COVID versus social media platform/news outlet study?

---

That'd be interesting to know!

Quote:

---

Not sure if this has been posted, but Houston ISD had 7,541 students who did not log into HISD's online learning portal a single time last semester. Not one time. Yet, HISD is doing online learning again for the first six weeks of the Fall semester. Absolutely shameful.

---

I have a hard time believing that few kids never logged on in HISD. I suspect the number is significantly larger.

After yesterday's call with Round Rock ISD, we are opting to keep the kid at home. The at school option is stupid and doesn't satisfy the need of allowing for more social interaction.

What is stupid about it? (Legit question; I'm in Houston and not familiar with the plan. Not being an ass).

Are you going to homeschool or just doing the online learning through ISD?

I can't imagine a world where I would choose the later unless my kid was 6/7th grade and beyond. At that point if I had to do home teaching I would probably just leave it to the teachers. Under that its not worth it for the constraints of having to be on the teacher's schedule...

In classroom learning but not really. You go to one classroom, stay there all day, then go home. Every kid will have a laptop and will have to attend their classes from their desks. So they don't get to walk to any classroom. They stay in the same classroom all day long. They log into schoology for their class Virtually at school! It's stupid

https://www.businessinsider.com/almost-half-of-kids-counselors-got-coronavirus-at-georgia-camp-2020-7


This seems to go directly in the face of the current thought process when it comes to kids

That says nothing about the ability of kids to transmit.

I suggest anyone who considers that study to be important to go read the actual study itself and not a media article summarizing it.

https://www.cdc.gov/mmwr/volumes/69/wr/mm6931e1.htm?s_cid=mm6931e1_x

The report says the camp originally had 261 staff and 363 campers (624 total).

594 of them were GA residents.

Campers ranged in age from 6-19, median was 12.

Staff ranged in age from 14-59, median was 17

Test results were available for 344. So, what breakdown of that 344 were staff vs campers? How many of the 363 campers were tested?

What was the age breakdown of those tested? 51% of those aged 6-10 tested positive, but how many campers who were tested actually fell in that age range?

The report says 260 out of 344 tested positive, but how many of those were 14 and up? 20 and up? 59?

From a learning perspective, it's no different than her learning at home than her learning at that classroom.

Might be good for teachers but her home room teacher won't be anything to her but a baby sitter

All 100+ kids got it from patient zero?

I read it

terradactylexpress said:

---

I read it

---

It looks like, of the 260 that tested positive, about 146 of them were staff members (261 staff x 0.56 attack rate).

That leaves about 114 infections among campers - who ranged in age from 6-19 years. I wish they'd broken down the total numbers who tested positive by age.

if you open the PDF it has the breakdown

168 out of 346 campers = 49% attack rate

6 - 10 yo: 51/100 51%
11-17: 180/409 44%

terradactylexpress said:

---

168 out of 346 campers = 49% attack rate

---

Didn't open the pdf. I will when I get to a computer. But those numbers don't add up. Staff had an attack rate of 0.56. That means 146 out of 261 staffers tested positive. If 260 total were positive, that leaves 114 positive campers.

But your post says 168 campers tested positive.

This thread has a good breakdown:

To me it's appears to be more evidence that high school aged children are highly transmissible. The first confirmed infected person was a teenager staffer. The staffers had 4 days of interaction before the young kids got there. Young kids were there for a short period of time so exposure window was small. Seems likely the staffers got infected during the 4 day window then spread it to the kids.