High School + COVID + Math = Hot-Zone Mess

Here's some basic math so you can get a grip on the back-to-school situation in New Jersey. For the past week, 426 cases of COVID-19 have been blossoming in The Garden State each day. On average. And depending on what source you check, the numbers can be higher than this. 

These positive tests are accumulating while bars and restaurants and gyms and schools are closed. 

The population of New Jersey is 8.82 million, but for the sake of making the math easy, we'll round the population up to nine million.

In comparison, Germany has a population of 83 million. Nearly ten times the population of New Jersey. Germany is also generating around 400-500 cases a day. Though Germany is 10x our size, they are "very concerned" with this number of cases per day. So while New Jersey may have this under control in comparison to Florida and Arizona, we do NOT have this actually under control.

Let's look at the math, see what will happen statistically if we send everyone back to school (if you don't want to look at the math, the Superintendent of South Brunswick High School has written a very compelling, non-mathematical argument of why it is too soon to open).

For the sake of easy arithmetic, we'll say that Jersey is germinating 3000 new positive tests each week. Obviously, there are far more than 3000 people in New Jersey that have COVID at any given time. Some people are asymptomatic and some have less severe symptoms. Some didn't get tested.

There may be ten times as many people with the virus as the testing indicates, but I can't even get into those numbers . . . they would be nuts.

COVID is transmissible before you have a fever, while you have a fever-- which can last from a couple of days to a couple of weeks-- and then it's still transmissible after the fever has broken. It is recommended that you quarantine for ten days once the fever naturally breaks. So a week's worth of COVID cases is probably contagious for around two weeks after that. The point is this: the five hundred or so cases that test positive each day in New Jersey don't just disappear a day later. They pile up.

Understand that all these numbers underestimate the actual prevalence of the virus. 

In a three week period, at the very least, 9000 people in New Jersey are going to test positive for COVID-19. Many more will actually be contagious. But we'll work with 9000 because that's approximately .1 percent of our population. 

One in every thousand New Jerseyans. 

Doesn't sound like much . . . until you put people in school.

Of course, far more than one in a thousand residents will be contagious in any three week period, but we'll use that very low number to illustrate my point.

New Jersey has lots of huge high schools. 


I teach in one of these schools: East Brunswick High School. We have over two thousand students (and that's just grades 10-12). We have over 200 adult employees. All crammed into an old, cobbled together building with crowded hallways, poor ventilation, and no central air-conditioning. 

If you had a school of exactly one student, using our simple mathematical model, there's a one in a thousand chance that your student has COVID (in any three week period).

But if you have 2000+ students in a high school, there is almost no chance that all people inside are NOT going to have the virus. One easy way to estimate this is to multiply 999/1000 times itself two thousand times. Then subtract that percentage from one hundred percent. You get 87%. That's the chance on any given day that some student is going to have COVID in a 2000 person high school. This isn't taking into account the teachers and janitors and guidance counselors and coaches and trainers and all the parents and child study teams and other humans that come into the building or work in the building. It's not taking into account the asymptomatic and mild cases. It's not taking into account sports, the possibility of playing teams from other towns. So any large school is probably going to have two or three or five or ten people with COVID in the building. Probably more.

Some schools might get lucky for a short period of time, but it won't last. It's statistically impossible. The virus will be present. It has to be. One in a thousand is a low estimate, but these high schools contain many thousands of people. Indoors, for long periods of time. So the virus will spread. That's what the virus does, even when schools are NOT open. Even in the summer.

So what happens when we open?

There's nothing like a school to harbor germs and spread sickness. In fact, schools are the germiest place on earth. Teaching is the germiest job.

Here's some research on this:

When Gerba and other University of Arizona researchers studied the desks, computers, and phones from various professions, teachers wrecked the curve.

Teachers had six times more germs in their workspace than accountants, the second-place finisher, with slightly cleaner desks but five-and-a-half times more germs on their phones, nearly twice as many germs on their computer mice and nearly 27 times more germs on their computer keyboards than the other professions studied.

The reason for all the germs is, of course, the reason why the teachers are there in the first place.

"Kids' desktops are really bad, too," Gerba said. "Probably the dirtiest object in a classroom is a kid's desktop."


During a typical school year, I get sick a couple of times. A cold or two, perhaps bronchitis, a stomach bug, occasionally strep, and the one year I didn't get the flu shot, I got H1N1. I used to think this was normal for adults, but now I realize it's not. 

Since the lockdown, I have not been sick at all. Not even a sniffle. The last time I was sick was February. I had an awful cough for two weeks and a fever. It may have been COVID, though I tested negative for antibodies (that test isn't supposed to be accurate). The teacher across the hall from me had COVID . . . so who knows? The point is, when you are in a huge school, there's stuff going around all the time. It's a petri dish. 

I used to think this was a perk of the job. My immune system is so strong! It's dealt with everything! 

Now I think it's a bargaining chip. We are going to be on the front line of this pandemic and we've been on the front line of general sickness and we should be compensated for it, with money and health benefits. I never really considered this until now. Many of us work in hot, crowded, poorly ventilated buildings, and-- unlike the meatpackers that have been sacrificed during this pandemic-- we have a union. 

It's going to be quite a clash.

I understand that it's hard to wrap your head around this because it's so statistical. We all really want things to go back to normal, the economy to open, schools to open, etc. It sucks. 

But 1 in every 550 people has died in New Jersey. That's significant. Many of these people were old and/or sick, but not all of them. We've had over 182,000 cases and nearly 16,000 deaths. So nearly nine percent of the people that tested positive died. That's an insanely high rate of death. Yes, many of them were in nursing homes, but lately, according to recent hospitalization data in the New York Times:

Adults aged 18 to 49 now account for more hospitalized cases than people aged 50 to 64 or those 65 and older.

Again, this is all happening with schools closed, bars closed, restaurants closed, gyms closed.

If you haven't felt the pernicious power of this virus, you are lucky. You are probably also fairly well off economically, you probably have the ability to work from home, you probably don't live in a multi-generational house or apartment, you might not have a lot of underlying health conditions, and you probably don't work in an essential service, as a grocery store employer or meatpacker or a nurse. 

You might not know people in those situations. 

So to understand the situation, you need to study the numbers.

You also need to understand that schools are the social-class blender of many towns. Kids from mansions and kids from apartments mingle. Kids who spent the summer in quarantine hang out with kids who worked all summer. Kids with their own bedroom in their own large suburban house come to learn with kids who live in crowded multi-generation households. And kids from different towns go to battle against each other on the pitch or court or field . . . the COVID permutations of high school sporting events are incalculable.

While kids probably won't die from COVID, they will pass it around. Especially high school kids. And then other people will die. COVID is not as dangerous as Ebola, so it's hard to put it in perspective. Death rates lag behind infection rates, so once again, you've got to look at the data.

The final arbiter is that many more people are dying than "normal." This is with schools closed. People who think this is an overreaction need to understand these numbers. The death toll is the final statistic. The bodies are piling up. And we're probably undercounting. 

Here are some numbers about "excess deaths" from the white paper I linked to.  Not only are there many many more deaths than usual this year-- and these deaths are directly attributable to COVID-- but there are also extra deaths above and beyond the COVID deaths. 

Results: There were approximately 781,000 total deaths in the United States from March 1 to May 30, 2020, representing 122,300 more deaths than would typically be expected at that time of year. There were 95,235 reported deaths officially attributed to COVID-19 from March 1 to May 30, 2020. The number of excess all-cause deaths was 28% higher than the official tally of COVID-19–reported deaths during that period.

Conclusions and Relevance: Excess deaths provide an estimate of the full COVID-19 burden and indicate that official tallies likely undercount deaths due to the virus. 

There's going to be a vaccine soon, and then things will go back to some sort of new (and hopefully more vigilant) normal. But this has exposed some serious problems in our infrastructure and preparedness. Most of our public schools are hot, crowded, poorly-ventilated places where large numbers of humans congregate without much thought to hygiene and the spread of sickness. 

Elementary schools may have a shot to open because you can keep the numbers very small. The Daily did a good podcast on how other countries (with the virus under control) have had some success in elementary schools. 

We may be able to send small pods of kids back to high school. Perhaps special education students and others that need school the most, but trying to parade several thousand bodies through a typical high school-- even on a rotating schedule, even with masks-- is going to perpetuate and accelerate the spread of COVID. No question about that.

I know people don't want to hear this. I'm not happy with my math. It's inconvenient and awful. But that's the story, right now. If we want schools to open, we're going to have to get the case count way, way down.

Israel tried to reopen schools on a large scale and this probably fueled a new outbreak. 

Sweden has kept schools open, and they have the highest death rate (12%) of any European country and several teachers have died of COVID.

I don't envy the administrators and politicians that have to make these decisions, but if you look at the simplest of math, while underestimating the amount of COVID in the population, there is still only one conclusion:

The fall is going to be a hot-zone mess.

12 comments:

  1. The push back to your initial discussion that you are going to get would be about testing rates. Including some information regarding positivity rates and the trend lines will help bolster your argument.

    ReplyDelete
  2. so the idea that because we are testing more, we are "discovering" more people with the virus?

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. Sweden is still being held up as the model by a number of people. I've been told that this piece is the basis.

    https://jbhandleyblog.com/home/2020/6/28/secondwave?fbclid=IwAR1QTfGSueqdhQJ6cQDaaz3JkWWxYJ94hC5gW4Gi2dDaSMZyKUVt0P_0BoA

    I'd be interested if someone could logically explain why it's bunk.

    July 29, 2020 at 1:00 PM

    ReplyDelete
  5. the mayo clinic does:

    https://newsnetwork.mayoclinic.org/discussion/herd-immunity-and-covid-19-what-you-need-to-know/

    sweden has a 12% death rate and a 6% antibody rate. people are doing 30 percent of what they once did: bars, schools, etc. so it's self-imposed social distancing.

    that post is interesting, as maybe so many more people have had exposure that we are closer to herd immunity than we think. we'll have to see in a couple weeks, if the death rates follow suit with the high infection rates in Arizona and Florida and texas.

    but it's disconcerting that jersey is still pumping out so many cases a day, during a semi-lockdown.

    ReplyDelete
  6. i did more reading on this-- it really depends what kind of model you use. it's really complicated math with a lot of variables at each stage. i think it's the kind of stuff that can really vary. while the case count, hospitalizations, and the death rate and fairly solid measures.

    i really have no clue what to think.

    ReplyDelete
  7. Correct. People will push back saying that of course the numerator is getting larger, but so is the denominator since we are testing more and more. It is true that we are finally getting testing to adequate levels so there may be som truth in that statement. That is why it is key to look at the positivity rate on a rolling 7 day average. If it is falling despite an increased stick count of infections then increased testing and/or prophylactic measures are likely the cause. If it is rising, you tell the Trumpy “it’s because of our big, beautiful testing program” people to fuck off.

    ReplyDelete
  8. i hope we've reached some kind of herd immunity. college sports and college campuses will be the first litmus test.

    ReplyDelete
  9. The model Trump uses is Melania.

    ReplyDelete
  10. If we wanted to do this sensibly, we would test EVERYBODY in a short time frame, identify the sick and quarantine them. Of course, we can barely test ANYBODY so that isn't possible.

    ReplyDelete
  11. right, then we could truly get a grip on things and lower the case count to nearly nothing. then open up restaurants and bars and gyms and such and see how we do. then open schools.

    ReplyDelete