Opening schools

Thank you.

She provided a Google Drive link to her data. I am not going to go through it. It is possible that only 2 weeks of data is there, and if so I can’t say if that is a linking error or other concern. Certainly an error may have been made, but I don’t understand how you arrive at your conclusion the case rates in schools is higher than that of the community in the worst performing cohort. No doubt if this is true, it is a major error and the NYT and Atlantic will soon be retracting this article - so I suggest you address your concerns to them or the author.

If, on the other hand, you are looking at the line of best fit for community cases, and noting that for staff in the worst cohort it is slightly above this line, for secondary students it is quite a bit lower, and for elementary students much lower… I do not see that means her conclusions are faulty.

I’m not asking you to go through it. I’m saying that I took the sum of column C (known student cases) and divided by the sum of column G (in person enrollment) and came up with a far higher infection rate than her worst community cohort. Just for shits and grins, I did it again with all students, even those who aren’t attending and the number was still higher than the worst cohort. No cherry picking, not fudging, no making up of data. The article (op-ed actually) was crap. Might be an accidental mistake. Might be an agenda. Was definitely wrong.

I read the article, as I have a subscription. FWIW, the author is an economist, not an epidemiologist, virologist, public health official, or a journalist who’s specialized in the field of infectious disease. She writes books on pregnancy and parenting. I’m not saying she’s wrong on any of this, merely that she’s no expert.

I’m going to retract my statement that it is more dangerous to go to school than to be in the worst community cohort, as it can no longer be backed up.

I’m not going to give the article a pass, as the numbers I used were directly from the article and were incorrect there. Her worse cohort over the four weeks was 28 per 100,000, which is 0.028% as I noted above. If that were correct, my statements would all have been accurate. In actuality there were 48,365 new cases in that four week period in New York (the whole state, not just the worst cohort), out of a population of 19,450,000, which is the equivalent to 249 per 100,000, which is 0.24%. Since that is almost an order of magnitude larger than her numbers, it changes things drastically.

So, her opinion piece is still shit, just different shit.

The problem we’re having around here isn’t that COVID-19 is spreading through schools, but that as a random kid or staff member brings it in the resulting quarantines take out enough staff so the schools have to close. The district decided it wasn’t practical to keep schools open while the community spread is so high that infected staff and students are inevitable. 9 out of 56 schools were closed for those reasons when the whole district closed down.

I dug into the numbers even more as I got a bit curious. Not only were her community numbers wrong, her student data wasn’t made up of the four weeks that she claimed either. I have no idea where it came from, so if someone else wants to post a link to the original source, I’ll happily dive deeper.

Since her analysis was so flawed and since this is an important topic, I decided to do a quick back of the envelope analysis myself.

Category Value
Population of New York1: 19,453,561
Ages 5-172: 2,990,091
Others: 16,463,470
Total Covid Cases3: 150,118
Ages 5-174: 14,701
Others: 135,417
Known Case Rate: 0.77167%
Ages 5-17: 0.49166%
Others: 0.82253%

1https://datacommons.org/place/geoId/36
2https://www.health.ny.gov/statistics/vital_statistics/2017/table01.htm
3https://www.worldometers.info/coronavirus/usa/new-york/ (09/01/2020-11/19/2020)
4https://schoolcovidreportcard.health.ny.gov/#/summary (Lab reported: 09/01/2020-11/19/2020)

Notes: If I was doing this in an official capacity, the primary call out would be that 09/01/2020 is not the ideal start date, as the virus had no time to spread through the schools by that time, so we don’t know the rate in which it ramped up. Unfortunately, I can’t find all of the necessary data with a date such as 10/01/2020. I also couldn’t find the trend data for lab reported cases of 5-17 year olds or I would have analyzed that as well.

In summary, in the state of New York between the dates of 09/01/2020 and 11/19/2020, the lab reported case rate among 5-17 year olds was only 60% of the case rates for all others. If the ramp-up time noted above doesn’t play a part (a big if), it was safer to be student-aged for the past 11 weeks than not.

These two studies are tough to read for those of us who want it to be safe for kids to go back to school.

The first is out of Brazil, and “is the largest population-based study of prevalence of antibodies to SARS-CoV-2.” The results are summarized in this table, which find seroprevalence roughly equivalent in children and adults, with about 1.5-3.0% of those tested having a positive result.

The second study from the CDC takes a prospective look at transmission within households, and the findings “suggest that transmission of SARS-CoV-2 within households is high, occurs quickly, and can originate from both children and adults.” The secondary infection rates were very similar regardless of the age of the index case.

While I want it to be safe for kids (hell, my wife is a teacher, so I selfishly want schools to be safe), this doesn’t shock me much. Most of the “kids aren’t carriers like adults” arguments have been based on flawed studies (and there are a metric shit-ton of those around COVID-19). Many of the studies didn’t even perform tests on the children in COVID-positive households unless they were symptomatic and some were simply assuming “first person with symptoms in the household” were the spreaders for their research. Both approaches are hugely flawed and are nearly worthless if we want to learn the relationship between children and the spread of this disease.

To address this problem the state is easing the quarantine restrictions. Now instead of entire cohorts being put on quarantine, only close contacts of an infected staff member will be quarantined.

I want schools to be able to open safely, but it isn’t clear to me that this easing of restrictions is based on patterns of contagion instead of just clearing of obstacles to staying open. The two statements “we never saw spread into the wider cohort outside of close contacts” and “sending too many people home was hard, so we’re going to send fewer people home” have very different implications for safely keeping schools open.

This has been the dialogue so far leading up to this:
Governor: Here are some guidelines for safely opening schools.
Schools: We’re following those guidelines, and being forced to close schools.
Governor: You should really try to stay open.
Schools: !!!

We’ve only ever quarrentined close contacts (have to keep rigid seating charts) and we still have schools closing with not enough staff to stay open. If we quarrentined cohorts, it would be impossible.

We are currently quarantining all grade 7-12 students for three weeks. Teachers still come to school but teach online from their classrooms. It’s not ideal, but it’s working okay so far. We can’t really show any videos because of lag problems, but we can give the students the link to watch stuff online on their own. Our classes are 2h and 45m long—one in the morning and one in the afternoon. At least nobody has to wear a mask online, so there’s that.

Didn’t Fauci just do a 180 about opening schools this past weekend?

Well, it’s not really up to him but what do you mean?

I mean he was for closing schools in the beginning. Now he’s not. People followed whatever he said. My cousins in Belgium all went back to school. No problem.

I don’t think so. He said, “Close the bars and keep the schools open.” That sentence has two parts. “Keep the bars open and open the schools” isn’t what he said, but that’s how a lot of folks seem to be hearing him.

The point is, you close the bars and get this shit under control, and that lets you have schools be open. Letting everything go into a literal death spiral isn’t the way.

Right. Or be open and say “economic activity has priority over non economic activity, like family gatherings, people dying in nursing homes, and schools. Your kids have to stay home and your mother with dementia must spend her last months alone and confused so that we can keep bars open”.

I’m saying in March Fauci said we should close the schools. When Trump said get the kids back into schools, people like Pelosi and Cuomo said he didn’t care about kids ad nauseum.

In March/April, we were dealing with vast uncertainty: we had very few tests, little PPE, no idea about how or why or when it spread. Under no circumstances should what made sense in March be guiding our steps now.