It’s been an eventful day for data and evidence around children, schools and Covid-19. What does it all mean? Quick thread below

First off, a really important new report from @IndependentSage on reopening schools: https://twitter.com/IndependentSage/status/1263748718513307653

Good to see it repeat our call for “attention to the wider social and economic costs of keeping schools closed” and “intelligent, incremental reopening of schools… responsive to local contexts and informed by rigorous testing.”

Its modelling suggests the risk of Covid-19 could be halved – not eliminated – by postponing school reopening from 1 June to 15 June. But the risk is also reduced more by waiting even longer. So on 15 June there is an incentive to wait another 2 weeks, etc etc.

Our briefing made clear that there’s always an epidemiological benefit to waiting to reopen schools. If that was the only factor, we’d have an incentive to postpone it indefinitely.

https://www.childrenscommissioner.gov.uk/wp-content/uploads/2020/05/cco-we-dont-need-no-education.pdf

What matters is how this marginal benefit compares to the marginal social and educational costs of waiting, and at what point the latter exceeds the former. This remains a key unanswered question.

Lots of discussion during the live meeting on the need for better data, much of which will take time. How do we produce it? And how do we make decisions in the meanwhile?

No answers here, but as our briefing said, it means incremental changes with careful monitoring/evaluation.

The @IndependentSage modelling focuses on shifting the school open date. But bear in mind that 80% of schools are already open and teaching 230,000 children. The real question is when/how we expand pupil numbers. What are the options and the risks?

As it happens, today's new official SAGE modelling has looked at exactly that:

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/886994/s0257-sage-sub-group-modelling-behavioural-science-relaxing-school-closures-sage30.pdf

It suggests that effects on R are lower if younger children go back to school rather than older children. Another option is alternating attendance, but more work is needed to analyse this. But its findings are uncertain because the underlying scientific evidence isn’t certain.

What does emerge is that all the options considered have a much smaller effect on R than the overall level of social distancing in the wider community. So controlling that, while allowing schools to reopen, is likely to be much safer.

Finally, a new review by UCL finds that children are 56% less likely to get Covid-19 than adults. This doesn’t mean children can’t get it, just that they’re on average less susceptible than adults. https://www.ucl.ac.uk/news/2020/may/children-appear-half-likely-catch-covid-19-adults

This may mean children play less role in spreading Covid-19, but the study can’t answer this question. It says “further data are urgently needed”, especially from studies tracking secondary infections where the child was the source – which are rare.

So where does all this leave us? Does it change anything?
It actually reinforces all the main points in our briefing. Here are the 10 things that you still need to know:

1. There's good evidence that risks to children and the role of schools are limited
2. “Limited risk” never means “no risk”
3. More evidence is needed, but it will always be needed
4. This evidence isn’t conclusive and will probably never be

5. The epidemiological risk will never be zero (absent a vaccine), & always lower if we wait longer
6. The social costs of school closures are accumulating too
7. What matters is marginal epidemiological benefit vs marginal social cost, and when the latter exceeds the former

8. There are no exact answers ‘ex ante’ – i.e. in advance of trying something
9. We need to make decisions before we have conclusive evidence and while risks remain (no vaccine)

10. So the only way forward remains the same as we set out last week: incremental changes based on local factors, accompanied by TTI and careful monitoring, with rigorous evaluation to inform each next step.

/Thread. Happy weekends!