Read on Twitter
Israel's opening of schools is currently under a lot of scrutiny. They relaxed many COVID-related restrictions (including on schools) and are experiencing a large second wave. Some have suggested schools are to blame.
How strong is the evidence?
1/11 https://www.wsj.com/articles/israelis-fear-schools-reopened-too-soon-as-covid-19-cases-climb-11594760001?mod=searchresults&page=1&pos=1
How strong is the evidence?
1/11 https://www.wsj.com/articles/israelis-fear-schools-reopened-too-soon-as-covid-19-cases-climb-11594760001?mod=searchresults&page=1&pos=1
As this recent WSJ article reports, by July 13, a total of 1335 positive cases are known among students, and 691 among school staff.
What should we make of these numbers? It is important to put them in context: 2/11
What should we make of these numbers? It is important to put them in context: 2/11
Israel is a country of 8.8 million people, with more than 1 million students in grade-school.
Even in secondary education, it has a low ratio of teaching-staff to students of around 10:1. I have not found data on the primary school ratio, but presume it lower.
So we have: 3/11
Even in secondary education, it has a low ratio of teaching-staff to students of around 10:1. I have not found data on the primary school ratio, but presume it lower.
So we have: 3/11
1.1 million students
at least 110,000 teaching staff
with known case totals of
1335 and
691,
respectively.
These are case rates of
1 per 823, and
1 per at least 159, respectively.
4/11
at least 110,000 teaching staff
with known case totals of
1335 and
691,
respectively.
These are case rates of
1 per 823, and
1 per at least 159, respectively.
4/11
On the other hand, the total number of positive COVID tests in the general population (as of now) is 42000. Out of 8.8 million, this is a rate of
1 per 209.
In other words, the case rate for teaching staff is just 30% higher than the population. Moreover, we have ignored, 5/
1 per 209.
In other words, the case rate for teaching staff is just 30% higher than the population. Moreover, we have ignored, 5/
for example, that there are more total staff than teaching staff, and that children in the general population (nearly a third of the population) contribute much less to case rates. Either one of these is likely to contribute 30% or more.
Now, the WSJ article does say... 6/11
Now, the WSJ article does say... 6/11
that uncited "epidemiological investigations" showed that after school openings, infections spread from students to the broader population. But scientists quoted in the article are more cautious, pointing out that Israel lifted many restrictions at the same time, including... 7/
for large parties, weddings, etc., whose effects are difficult to disentangle from effects of schools. As a basic sanity check on the theory that schools are driving Israel's epidemic, it seems easiest to accept the theory if teachers have a much higher infection rate. 8/11
In fact, these crude calculations suggest that Israel's experience seems consistent with data from other countries which opened schools without finding higher occupational risk to teachers (or high infection rates in students).
https://twitter.com/ProfEmilyOster/status/1278011105034592256
9/11
https://twitter.com/ProfEmilyOster/status/1278011105034592256
9/11
There is an important lesson here anyways though.
It is no surprise that, with more 13% of Israeli society spending their days in grade school buildings, there are positive cases associated schools, and, of course, even some transmission should be happening in schools. 10/11
It is no surprise that, with more 13% of Israeli society spending their days in grade school buildings, there are positive cases associated schools, and, of course, even some transmission should be happening in schools. 10/11
But the question for opening schools (or keeping them open) should not be whether any cases have been associated with schools.
Instead the question should be whether schools are an especially risky environment - indeed, so risky, that the risks outweigh their benefits. 11/11
Instead the question should be whether schools are an especially risky environment - indeed, so risky, that the risks outweigh their benefits. 11/11
Addendum:
it is important not to equate older and younger students.
E.g., the occupational hazard data I cite above is from Sweden, where schools opened just for <16.
E.g in Israel they see children's infection risk begin "primarily at 15 years old": https://www.wsj.com/articles/israel-shuts-some-schools-as-coronavirus-cases-jump-after-reopening-11591203323
it is important not to equate older and younger students.
E.g., the occupational hazard data I cite above is from Sweden, where schools opened just for <16.
E.g in Israel they see children's infection risk begin "primarily at 15 years old": https://www.wsj.com/articles/israel-shuts-some-schools-as-coronavirus-cases-jump-after-reopening-11591203323
