First, I should declare my interest. I want my kids to go back to school. The younger one is OK: she’s four, and happy with her own company. She’s unilaterally decided to reject the staid pre-lockdown convention that people ought to wear pants; but apart from that, she’s much as she ever was, and can play happily with her Sylvanians for hours without adult intervention. But the six-year-old misses his friends; he misses running around. His parents are just not as good at, or interested in, prolonged games of tag as are other six-year-olds. He’s climbing the walls a bit and needs constant attention. It’s hard on him and it’s hard on us.
These are very minor problems, obviously, compared to tens of thousands of dead from a frightening disease. But in the Chivers household they’re very salient, and it would be dishonest of me to pretend they’re not a factor.
Like what you’re reading? Get the free UnHerd daily email
Already registered? Sign in
So. I wanted to look at the elements that go into a decision to reopen (or not reopen) schools; what the evidence says; and what the trade-offs involved are.
First: there’s been a lot of attention lately on a strange syndrome that appears to be associated with Covid-19 in children. It’s an autoimmune condition, similar to a condition known as Kawasaki syndrome; the body’s defences overreact and cause severe inflammation, especially of the coronary arteries. It was first spotted in Bergamo, Italy, and Prof Liz Whittaker of the Royal College of Paediatrics and Child Health (RCPCH) told a briefing that Bergamo would normally expect to see about five cases of Kawasaki syndrome in a year, while they saw 10 of this similar condition “in five or six weeks”. Since then, about 75 or 100 have been seen in the UK.
It’s got a lot of press attention and Matt Hancock, the health secretary, has said he is “worried” about it. But — without wanting to minimise the seriousness of it — it is very rare. Whittaker’s colleague Prof Russell Viner told the same briefing that even including the Kawasaki-like cases, children are at extremely low risk. “The deaths tell us all we need to know,” he said. “You can count [children’s deaths from Covid-19] on two hands, out of [a total of] around 30,000.” There have also, he says, been fewer than 500 hospital admissions for under-18s in the UK, compared to hundreds of thousands for adults.
That doesn’t make the deaths that did happen any less awful, but it does mean that each child is not at significantly greater risk than they are going to school anyway, from accidents or other illnesses. As Viner said elsewhere, about 160 or 170 children die in car crashes each year: “Those deaths are tragic but also rare. How do we manage those? We take steps to prevent these deaths. We buy child seats, we use seat belts, we sometimes buy better cars. But we don’t stop driving. This syndrome is much, much more rare than car accidents involving children.”
Every paediatric specialist and epidemiologist I’m aware of says that the risk to children is minuscule. But, obviously, that’s not the only factor involved. We want to know the risk of children spreading the disease; and, equally importantly, we want to know the costs of children not going back to school.
I’m not going to pretend to be able to give you a definitive answer. But I’m going to go through the factors to give you a sense of why it’s so complicated.
First: we need to know whether kids have the disease in great numbers. Dr Sanjay Patel, a paediatric infectious disease specialist at the University of Southampton, pointed me to a study that came out in Spain last week, which tested 60,000 people for Covid-19 antibodies. Distressingly it found that only 5% of Spaniards had had the disease — suggesting that, if those numbers are accurate, despite Spain’s awful outbreak, they are still nowhere near herd immunity.
More hearteningly, though, it found that only about 1% of babies, 2% of one-to-four-year-olds, and 3% of five-to-nine-year-olds had it. Earlier studies in Iceland and Italy found almost no children infected at all.
The economist Emily Oster, who like me is very interested in all of this because she wants to send her kids to camp, also points to an interesting study which found that in the Netherlands, 6.5% of people who turned up at their GP with flu-like symptoms tested positive for Covid-19 — except for young people. The under-20s actually did have the flu.
That said, the Office for National Statistics released some data recently, looking at live virus rather than antibodies, which found children were no less likely to have the disease than adults. Other work finds that children have viral loads comparable to that of adults.
So children may have the disease less than adults, or they may not. It’s not clear. If they do, it’s probably only somewhat less — half as much, or so — not orders of magnitude less.
But that’s still not the only question. The next question is — if children do get infected, how likely are they to pass it on?
There are reasons to be hopeful. An un-peer-reviewed paper in Australia, released on Monday, found very little transmission from children in schools. And there’s been some attention on some more Icelandic research which I am extremely keen to believe: a company running the testing and tracing in that country has so far failed to find a single example of a child passing on the disease.
Patel thinks the situation is unlikely to be that extreme — “it’s not that children cannot be infectious; that just doesn’t make sense and isn’t likely to be true”. But there are, he thinks, good observational reasons to think that children are less likely to spread it.
First, he points out that there haven’t been any outbreaks in schools and nurseries among the children of key workers, or in countries such as Iceland where the schools stayed open. In Denmark, the spread of the disease accelerated briefly after schools were reopened, but quickly fell back to where it had been. Patel points out that all these schools have maintained social distancing, which may have kept the disease from spreading, but “you can’t expect totally robust social distancing in nurseries and primary schools”.
And besides, to some extent, if it’s working, it doesn’t really matter whether it’s the social distancing or something innate to children that’s stopping the disease from spreading — although if it’s the latter, it may mean we can end social distancing in schools a bit sooner, which is good because those pics of children playing two metres away from each other break my heart.
So, we don’t really know whether kids get the disease at the same level as adults, and we don’t really know if they pass it on as readily, although there are encouraging hints on both fronts. Dr Aubrey Cunnington, a paediatric infectious disease specialist at Imperial College London, is extremely cautious: “The bottom line is we really just don’t know whether children are going to spread this virus to each other, and to their families and the community, once we reopen schools.”
But that’s still only half of the story. The other half is — what are the costs to keeping children out of school? If there were none then it would be simple: just keep them at home. Obviously, it’s not simple.
There are economic costs — there are nearly nine million children in UK schools, meaning that millions of parents will have to take time off work to look after them. And those economic costs are real and damaging to people’s lives. But they’ve been addressed elsewhere, so let’s look at the direct costs to the children.
Sam Freedman, CEO of the education nonprofit Education Partnerships Group and former executive director of Teach First, breaks down the impacts for children in three ways. They are: direct educational cost; impacts on socialisation; and risk of abuse at home. “People who are on the ‘keep everyone safe until September’ side of things are underestimating the negative impact on kids,” he says.
If children don’t go back until the start of the new academic year in September, they’ll have missed around 22 weeks of school since March. “People tend to think it’s only a few more weeks, it’s how much difference can it make, but what I’m worried about is the loss of learning over that period,” Freedman says. You can detect “learning loss” — kids forgetting what they’ve been taught — over the six-week summer break, so a near half-year gap would be presumably much worse. And, as with so many things, it will be unequal — “some kids will be fully homes schooled and having a great education, for most it’ll be worse than being at school, and for some it’ll be nothing”.
For younger children, like mine, the actual stuff you’re learning in school might not seem so important — they can colour in pictures of Pokémon just as happily at home. But the socialisation side is even more paramount. “It’s so important at that age,” says Freedman. “My daughter is four; if she doesn’t go back to school until September, she’ll have gone 15% of her life without seeing a child her own age.” Cunnington agrees: “socialisation at that age is key, and you can’t do it with your parents.”
And of course there’s the abuse factor. I don’t know how common it is, but, Freedman warns, “it’s probably a lot more prevalent than most of the middle-class people arguing on Twitter realise”.
So, getting children back to school is important. But at the moment, the plan is to start with three school years — Reception (age four to five), Year One (five to six) and Year Six (age 10 to 11) — from 1 June. Then the impact will be assessed, and assuming no major change in the spread of the disease, other years will be brought back.
But by then it’ll be only four or five weeks until the summer holiday. And presumably children won’t be going back full time — one school I’m aware of is planning half the children in Monday and Tuesday, a deep clean on Wednesday, and the other half in Thursday or Friday. Something like that will have to be implemented everywhere to allow social distancing. “Half a day for four weeks is not a lot in the scheme of things,” says Freedman.
Other countries such as Italy, which have summer holidays from June, have simply said they will restart after the summer break, and Freedman wonders why the government isn’t bringing forward our summer holiday; it’s not as if anyone has any foreign holidays booked. But perhaps there are good logistical reasons for it, and it would be better than nothing — there may be some repetition of what’s already been taught this year that can reduce learning loss; it would be good for socialisation; it would give parents and kids some time apart and thus hopefully reduce abuse.
Cunnington adds that some contact with teachers means that those teachers can provide children work for the rest of the week, and also that from a safeguarding point of view even contact just once or twice a week would be a useful way of checking up on at-risk children: literally just asking them if everything is OK at home, or in other ways seeing if they’re not doing well, could be very useful.
The main thing, though, is that none of this is once-and-for-all. Freedman is watching other countries with interest — not just Denmark and Norway and Iceland, where the outbreaks were so limited that they do not make a good comparison, but other hard-hit states like Belgium, the Netherlands and France. “Our debate is very parochial,” he says. “Every country is having this debate. Watching the impact is what I’m interested in. If we get to the end of May, and France or Holland seem to be seeing a significant increase in infection rate, that’s a strong argument for us being cautious. If they aren’t, then it isn’t.”
Patel, likewise, says that we can be cautious — “it’s not just once we’ve made the decision we can’t go back. It’s doing it tentatively, with clear testing, tracing, and isolation in place”. He appeals to a sense of collectivism – parents are understandably worried about the risks to their children, he says, but the harm to other children by keeping them off school could be even greater.
“We have to be entirely honest with the public, that we don’t have the answers yet but doing nothing is not an option,” he says. “There is a very small risk for the child, grandparents can’t do drop-off, parents can’t aggregate at school gates: it’s not a green light to do anything, but if we do it sensibly then we can get our country back to some new sense of normality, and reduce the harm from the educational and wellbeing health perspective.”
As I said at the top, I’m keen to see my children back at school, for selfish reasons, for work reasons, and for their sake. I entirely understand those — I know several — who go the other way: the IFS recently found that 50% of parents wouldn’t send their children in, even if schools reopened.
But if it’s done carefully, and with acknowledgement of the huge uncertainties and a close eye on the experiences of other countries, it could be done safely, and could reduce the real damage that gets done. I hope we manage it. I want my kids back at school.