Opening schools

The Quarantine Zone
341

There were two highly relevant articles on the CNN site tonight. Excerpts from the first one:

Fact Check: What role do kids play in spreading the coronavirus?

The former commissioner of the Food and Drug Administration Scott Gottlieb… wrote on Thursday that the “data clearly shows [children are] less likely to become infected and less likely to transmit infection.”

“But IMHO,” Gottlieb continued “we need to have humility on this question and recognize we don’t fully understand all the risks; and while kids are less vulnerable, less risk doesn’t mean no risk.”
…

In congressional testimony on June 30, Dr. Anthony Fauci, director of the National Institute for Allergy and Infectious Diseases, addressed questions around children and the coronavirus. “We don’t really know, exactly, what the efficiency of spread is” among children. The NIH, Fauci mentioned, is currently studying 2,000 families to understand the rate of infection for children and “how often they infect their families.”

During the testimony, Redfield also mentioned that the CDC is currently studying households to understand what role children play in passing the virus on. “We don’t know the impact that children have yet on the transmission cycle,” Redfield said.

Bottom line: The CDC doesn’t know yet, the NIH doesn’t know yet, and we here on the SDMB don’t know yet. It’s looking like schools will reopen anyway, and heaven help them. But read on.

342

On to the dangers to teachers:

One in four teachers at greater risk from coronavirus

Nearly 1.5 million teachers are at higher risk of serious illness if they contract coronavirus, according to an analysis released Friday evening.

These teachers and instructors, about 24% of the total, suffer from health conditions such as diabetes, heart disease or obesity, or are older than age 65, which make them more vulnerable, the Kaiser Family Foundation report found.

The share of teachers at high risk based on criteria identified by the Centers for Disease Control and Prevention is the same as for workers overall, Kaiser said. Schools face the challenge of high traffic and tight quarters, which could make social distancing difficult.

343

I just found this, and I can’t play with it now, but maybe somebody else will.

A professor here at CU has created a Google spreadsheet to model SARS-CoV-2 transmission in different environments. Things can be adjusted such as duration, room size, mask efficiency, etc. The provided example of a university classroom suggests that an infected instructor has a 4% chance of passing the virus to a student during a 50 minute class, but an infected student only has a 0.5% chance of infecting another person.

They parameters could be adjusted for primary school classes.

344

I should clarify, this is our elected state superintendent who’s the corrupt incompetent ladder-climber. Our local superintendent has been on the job for a little more than a month; we’ve had six superintendent turnovers in the past decade in my system. I’m the new president of my union local, and I’m in the process of establishing a productive relationship with our new local guy.

345

Ah. Our state level guy is appointed, not elected, and i actually don’t know what his title is. We just call him the head of TEA.

Six superintendents in ten years is more normal than not, I think. We are very very fortunate to have stability, and a solid guy. Previously, we had that sort of turnover, and the occasional arrest.

Good luck with the new gig. Thank you for your service. Sincerely.

346

I just have to point out that up to about two weeks ago pretty much nobody in Texas was being tested except people who were hospitalized. Of course the positive TEST RESULTS have gone up. Have the number of cases gone up? Nobody knows. Are the test results accurate? Nobody knows.

347

Don’t any of you supporting a rapid re-opening of schools take any pause in knowing that Donald Trump agrees with you?

348

We are testing more and the % positive is going up, quickly, not just the total cases. If we were broadening tests into populations less likely to have it, you’d expect that number to go down.

349

Less likely to have it, or less likely to be sick from it?

350

Studies show that missing fundamentals like math or reading makes it a lot harder to catch up at later ages.

Also, there’s no guarantee that young kids would do some edifying gap-year activity like an internship or summer classes. Their education needs to be managed by an adult. If one or both parents can’t afford to stop working and educate the child, then they just end up being babysat by Minecraft (or worse).

Fortunately our household finances have enough slack that one of us can homeschool if necessary. But it’s fundamentally unfair that a vocal minority is forcing less-fortunate families to compromise their childrens’ safety if they want an education.

Yes, this entirely possible. In fact in April-May this country went through an entire month when almost one September 11th worth of deaths were happening every single day. We collectively shrugged and said “grandma needs to be sacrificed for the economy”. Now that school days are rolling around, we’re collectively shrugging and saying “well, kids are resilient, and they need to learn that stuff happens.” Even if the “stuff” is getting killed by a virus that adults refused to manage.

351

Epidemiologist seem to concur that rising positivity rates paired with flat or increased testing are a strong indicator of growing incidence. And more incidence will be more people dangerously sick.

The goal ofde lining positivity rates has been presented as a key indicator since early May.

352

Catch up to whom? If we add a year, basically, and have a generation graduate HS at 19 instead of 18, it’s not about catching up anymore. I’d make that year voluntary, with the option to add it between each level of school.

353

Part 1, classrooms.

I got a chance to play with the estimation spreadsheet.

Results first. If the teacher is infected there is a 6% chance of infecting each student per class day. If a student is infected there is a 0.2% chance of infecting each other person day.

Now the long bit where I explain things.

There are a few important caveats. This spreadsheet models airborne transmission, not droplet or contact transmission. It assumes a 6ft separation is maintained. The point is to get an order of magnitude estimate—the difference between a result of 0.2% and 0.6% is probably not reliable, but the difference between 0.2% and 4% is probably reliable and meaningful.

As advised, I adjusted the model quanta emissions and inhalation rate for the students by using body mass. I simply took 1/3 of the “university student” values. This is difficult in elementary school, as some of the kindergartners will be 35 pounds, and some of the 5th graders over 100 pounds. Anyway, that’s what I did.

I left the classroom size the same, and because my kid’s school got a new HVAC system three years ago, I bumped the air exchange rate up to 8 times per hour, which is the high estimate for in a classroom. That makes a tremendous difference. The low estimate for air exchange rate, 1.8 times per hour, produces an order of magnitude increase in likelihood of infection from the teacher. It goes from 6% to 17%. The student to student infection rate is a much more modest increase to 0.7%.

The number of students in the classroom does not change the probability of each person getting infected, but it does change the number of people expected to be infected each day. So, with 20 students in the class, and an infected teacher, 1.11 of the students is likely to get infected each day. With only 10 kids in the class, that drops to 0.55 likely to be infected each day.

No changes were made to adjust for children having an innate resistance to getting infected, or spreading infection. Just the change that children breathe less, so they expel and inhale fewer quanta. In the other direction, this is assuming that the students are breathing at resting respiration rate, and only briefly talking.

354

Part 2, campus.

The campus estimates are based on the population of the campus, and the community wide rates of transmission.

I adjusted down from a university campus to just a single elementary school with 300 students, and 50 faculty/staff.

I left the probability of a person in the community being infected at 0.3%, as the spreadsheet’s defaults reflect my local area.

The estimate is that over a 13 week semester, 3 students and 0 faculty/staff will become infected outside the school.

My takeaway from all of those estimates are that given current rates of community transfer in our area, it is unlikely that any teachers will come to school with Covid-19. However, if a teacher does come to school with Covid-19, then that will create an outbreak which spreads quickly. Probably 1-2 students per day will be infected.

Also, my county now has the plague to go with our Covid-19. Of course we get it every year or two, and unless you’re playing with prairie dogs, your unlikely to catch it.

355

There are too many unknowns to accurately predict how many children will contract COVID, spread it to more vulnerable kids or adults, or suffer/ from it. If there were, top public health officials and epidemiologists, who certainly have access to more data and can determine probability better than we do here, would be predicting, not saying, “We don’t really know, exactly, what the efficiency of spread is” [Fauci] and “We don’t know the impact that children have yet on the transmission cycle.” [Redfield] [cite in post 341]

We don’t know how many teachers will get sick or how many of those teachers will die. We know that 25% are at high risk for serious cases. [cite in post 342].

Everyone wants schools to reopen. No one wants children, teachers, or their contacts to suffer and die. Slice it, dice it any way you want to, that districts have such a high-stakes choice to make without the data on which to make it is the stuff of nightmares.

356

What I posted was an epidemiologists attempt to predict all of those things. One of the ways to do that is to create a tool that lets the unknowns be adjusted.

Often the unknowns are not completely unknown, so a range of values can be tested. For example, in the classroom setting the spreadsheet can adjust mask’s efficiency of emission and intake. It turns out that dropping the intake efficiency to 0 (kids can’t keep their masks on), barely raises the danger from an infected teacher. However, dropping the efficiency of mask emission dramatically increases the chances of other people being infected. If it’s possible to lecture while wearing an N95 mask (without an exhale valve), that would make things much safer in the presence of an infected teacher.

357

If health care professionals are having a difficult time obtaining N95 masks and other PPE, teachers have ZERO chance of getting them! If they manage to scrounge up a few, you can bet the teachers will be told to hang the masks up to dry overnight, so they can use them again (and again…).

~VOW

358

I am of mixed mind about that statement.

There should not be one-size-fits-all, and local factors, like rates of infection, and directionality of the numbers, and facts on the ground, all need to be taken into account.

But each district should not be having to invent their own wheels and building their own toolboxes, to some degree subject to arbitrary whims of local leaders and “experts”. In my dream world there would be a functional CDC working with a well led COVID-19 task force, to come up with guidance for different levels and directionality of metrics in local communities and having created a useful sets of approaches that would be acceptable for each level.

359

Do you believe that the latter supports the claim of the former? It does not.

360

In other news half of all people are below average.

Used in the way it is used above “greater risk” means as little.

Models are only as good as the assumptions they are based upon. There is no reason to have any confidence in the assumption used in this model.

There is a name for that cognitive error that I am not recalling, the automatic knee jerk conclusion that a statement is wrong because you generally disagree with the person who is stating it, and for that reason alone.

It would be a making a serious cognitive error. If my sync’ed to the atomic clock watch and my phone both say 6:00 I am not going believe it less because a clock I know is broken also says 6:00.