Rational thread on the risks of schools staying open

[WhatWillSantaBring]

I understand that there are a lot of strong feelings on this so hoping that we can limit this thread to facts and data (rather than emotions).

What (formal) evidence is there that teachers are at a greater risk than the general population?

What (formal) evidence is there that children are passing the infections to adults?

I've seen the data that educational institutes are associated with increased infection rates (though really unhelpful that the data isn't split between nursery/primary/secondary/tertiary) but that doesn't distinguish between the numbers of the pupils getting infected and the numbers of adults in those settings getting infected. In my head, if schools were responsible for the spread, you would expect to see (a) a higher incidence of C19 in teachers and (b) a higher incidence of C19 in the parents of school aged children. Is there data to back this up?

The only study I've found so far is one in Sweden (where schools stayed open throughout the spring) which gave a relative risk of between 0.7 and 1.1.

I'm really curious about this, and hoping that some epidemiologists are lurking and will post me some links to peer reviewed studies. (Disclaimer: I'm not a scientist but I can read and follow most scientific papers)

@Sonnenscheins - yes, fully agree. The risks of closing them are huge. Debating the risks on both sides of the open/close debate are things we do need to discuss, but first we need to understand the actual risk on the open side of the debate, because that seems to cause a lot of emotive/anecdata discussions, which I'm trying to avoid.

@WhatWillSantaBring there was a French cohort study that tracked transmission, here: www.medrxiv.org/content/10.1101/2020.06.25.20140178v2

I'm a social scientist, so can't comment on methods!

This is a good balanced thread. We do need this debate.

Are we seeing primary school children spreading the virus amongst themselves? Any big clusters that have tested positive in one class?

In our school (yes anecdotal but seen the same in other schools). One child tests positive after a parent has had symptoms and tested positive. Whole class isolates, nobody else gets it.

Hopefully studies are being done at the moment.

@Sonnenscheins
Second, it seems relatively simple to look at the prevalence of covid in parents of school aged children to see if that is much higher than those without children. If it’s not much higher then that would suggest that teachers are not at a much higher risk. Maybe it’s already been done?

It hasn't been done, or at least no such research has been published or flagged as being in progress.

The only thing would be the study in India of child transmission, article here:

www.princeton.edu/news/2020/09/30/largest-covid-19-contact-tracing-study-date-finds-children-key-spread-evidence

This showed that children can be superspreaders. But also that the average transmission rate in a household was well under 50%.

If we consider Sweden as a useful data set, we need to remember that upper secondary years were closed to students during the first wave. So their reassuring stats are not entirely comparable with the UK having all schools fully open.

Sweden did have ages 5-16 in school though, correct?

@Delatron We have had a similar experience in my DS’ class. The pupil caught it from his uncle. No one else in the class has become infected.

Prof Spiegelhalter on an episode of more or less a few weeks ago said that studies showed children under 12 had very low transmission rates. Sorry to be a bit vague but I listened awhile ago. I think the studies related to did they pass it on to household members but could be wrong.

With schools, it might be useful to think about what we DON'T know.

The way test and trace works, the index case for a cluster will be the first person to develop symptoms and test positive. Children are often asymptomatic, so if they caused a cluster, this wouldn't often be picked up.

So we just cannot know whether children at school are driving transmission.

What we do know is that children, with a few exceptions, are safe from this virus, but we want to prevent them spreading it. For those anecdotally pointing to children who have had it but not spread it to family members: this is true for most people. Covid has a low dispersion factor, meaning about 10% of people cause about 80% of transmission - superspreaders. So the big question is: can SOME children be superspreaders? With no mass testing of children, we just cannot know this.

I would go for face masks for everyone in schools and uni, all the time indoors, and a mass testing program for schools and unis. Swabs can be pooled for bubbles, making this feasible. We want kids to have an education, and we want to keep this virus under control. We can do both, if we get our shit together.

Pretty much every other developed country already had smaller class sizes than us and/or put in additional measures such as mask wearing in class.

It's not true that just about every other developed nation has smaller class sizes, at least not is secondary schooling.

Do Swedish pupils wear masks in lessons?

@WhatWillSantaBring i think i understand what you are aiming for from the data now. What would be nice from teacher data would be for it to be compared against similar groups. Perviously BAME, age, pre-existing factors were all relevant. So knowing 'teachers' as a whole were the same average risk as 'population as a whole' is ok - but not if the profession is heavily weighted to low risk groups. I suppose i'd want to know is a 50 year old diabetic teacher more likely to get ill than a 50 year old diabetic in 'all other occupations' and then broken down to things like 'healthcare, retail, legal, beauty'
I guess we wont know this stuff for years.

@WhatIsGinLiqueurAnyway the question of whether some children can be superspreaders, is answered in the India study I posted an article about.

The answer is most definitely 'yes'.

I'm not sure whether anyone has looked at the ONS figures on this, but obviously all published government papers are subject to spin and aren't peer reviewed.

www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/articles/whichoccupationshavethehighestpotentialexposuretothecoronaviruscovid19/2020-05-11

It indicates education workers are at lower risk, especially secondary

This article is interesting. There was mistakes made but still no less safety measures than here.
www.nytimes.com/2020/08/04/world/middleeast/coronavirus-israel-schools-reopen.html

Good question OP and I was hoping we'd see some evidence by now. But the tiny pieces of evidence we do have suggest that what the government says is true. Primary aged children do not contract/spread the virus at anything like the rate of adults, but between ages of 11-18 this becomes the same rate as adults (as you'd expect).

Lots of people say how do we know, since primary aged children generally don't show symptoms so won't be tested - but ONS community surveillance shows exactly the same pattern so I think we can be clear it isn't just asymptomatic children.

There's very little evidence of transmission between primary aged children. My experience is same as other posters - 4 cases in primary of 500 pupils, no instance of transmission from any case, no teacher cases. There have been few high school cases identified in my area either (rate of c350 per 100,000), afaik closure of bubbles at similar level to primary.

@FourTeaFallOut

Pretty much every other developed country already had smaller class sizes than us and/or put in additional measures such as mask wearing in class.

I wonder if thats including private schools or options classes or sixth form? I’ve never seen a class below 25 in mainstream core subjects.

Hmm, it might be the case but if the class size of sixth forms are bringing down that average that would be even better for us in terms of managing covid in the more mobile and spready older children/ young adults.

I’m thinking of my dds classes in sixth form. One was 2, the other two were 30. Depended on what courses people took to whether they have the small classes.

Do Swedish pupils wear masks in lessons?

They’re not big on masks generally, so I doubt it.

Are these studies actually testing everyone though? If up to 80% are symptomatic, and I would assume that it would be even more than that in children with the older ages having the symptomatic cases, then surely we don't know whether little Johnny whose uncle had it spread it to his class mates. All we know if that his classmates never became symptomatic.

Same goes with the graphs of total cases. If testing is unavailable unless you have symptoms and we assume a vast majority of infected children would be asymptomatic, it means very little. Only after they finally start testing the whole cohort will we know what the situation in schools is.

@Bedraggledmumoftwo No we don’t know if he spread it to his classmates who are asymptomatic. However, more testing is going on in my DC’s school than most as they have their own testing machine and are testing the teachers and boarders (day students are not being tested at the moment). There have been three isolated cases in three different classes which have not spread throughout the school. In each case transmission to the student was identified outside the school.

Vulnerable teachers...overweight and over 50 for example, or diabetic.....personally it would be good to understand the impact of setting those teachers into furlough in areas with high virus levels

Ideally, yes these teachers shouldn't be exposed to greater/ as yet unknown risk, but then who would teach their classes? It just wouldn't work.

I don't know the source of that figure of "19" but in over 25 years of teaching in a variety of secondaries this has never been my experience. There are rural primaries with only a few in a primary class I suppose. My own range from 22-33. Sometimes lower ability classes are put in small groups for support.