Thread by @dgurdasani1

ONS data on long COVID out today very concerning
-persistent symptoms *affecting daily life* affect a significant proportion of primary & 2ndary school age children
-clear impact on children's wellbeing 🧵
ons.gov.uk/peoplepopulati…

Let's look at the ONS results.

First, it shows that 1 in 100 primary school age children had at least one of 28 persistent symptoms post-COVID-19 *affecting every day life*. This is *not* 1 in 100 children infected. It's *1 in 100* children in the community. And pre-omicron.

Let's think about this again. 1% of all primary school age children in England estimated to have had long COVID symptoms *affecting daily life* for 12 weeks or more.

It's clear that a high proportion children reporting long COVID were significantly affected by it - in terms of their activity, ability to learn, and emotional well-being. About half reported full recovery, & another quarter partial recovery, with about 1 in 5 still not recovered

What about secondary school age children?
2.7% (1 in 37) children reported persistent symptoms post-COVID-19 for 12 wks or more impacting their day to day life. Again, this is *not* the proportion of children infected- it's the proportion of *all* children in this age group.

Now let's look at the 'control' comparisons, which many have hailed as the gold-standard (they're not).

Quick recap: Comparisons of 'any symptom' between PCR positive and those who haven't been PCR positive is a very flawed way of determining long COVID prevalence. I'm glad the ONS presented this as their secondary analysis rather than their primary analysis.

From a statistical perspective comparisons between 'cases & controls' looking at presence of any symptom will generally underestimate long COVID prevalence. Discussed here in detail, with numbers:

https://twitter.com/dgurdasani1/status/1487578265187405828?s=20&t=0rZfj2EU7yPz50kN8iASIw

As the ONS states, children who were not PCR positive (but some of whom did report history of COVID-19) were included in this 'control' group. Also remember testing wasn't being conducted in the UK until June '20 (so not in the 1st wave) & even after, many children weren't tested

This is very clear from the stark contrast between seroprevalence data from the ONS at the same time that showed ~40% seropositivity, while the ONS test data shows only 20% of primary school age children had a positive test. So a lot of 'non-positives' will have had infection.

This will also underestimate the signal in +ve cases because there will be children with long COVID in the 'non-positive group as well. So this along with using the 'any one symptom' approach, will both underestimate, as I've outlined in my earlier tweet.

So what did they find?

They found 1% difference in 'any symptom' between positives and non-positives for primary school age children. With ~46% 'controls' having 'any symptom' we can expect the signal to have been substantially diluted. 3+ symptoms picks up more of a signal and we see a 5% difference.

Remember that both of these will be underestimates, because of the approach used, as I outlined earlier. Even without accounting for several children with past infection and potentially long COVID in the -ve group. But even these differences are very worrying.

Even 5% of children showing 3+ symptoms, or 1% showing any symptom (both underestimates), given the millions of children exposed would leave huge numbers affected. Remember, that this is pre-omicron data, & during the omicron wave ~40% more children in this age gp were exposed.

Total exposure estimated at 85% in 5-11 yrs now. Millions of children in this age group have been infected. Looking at the components of the syndrome, it's deeply concerning that even just looking at one symptom, 5% of those infected had loss of smell & taste for 12 weeks or more

If long COVID wasn't biological or 'due to pandemic fatigue', why would this particular symptom that is so specific to this virus be so much more common in this group? Significant increase in cognitive disturbances as well - for 12 wks or more - 4.5% difference.

This could have significant impacts on children in school, even if it just lasts for 12 weeks. Of course it's very likely that for a significant number, this will have gone beyond. Long term follow up will be vitally important. But 12 wks of illness is a long time for a child.

What about secondary school age children?
Comparing with the non-positive group, we find that using 'any symptom' assessment, the diff between the groups is 8%. And using 3+ symptoms its 6%. That's huge, even considering these comparisons underestimate prevalence.

This is perhaps the most shocking of all- 16% of children with a +ve COVID-19 test reported loss of smell & taste for 12 weeks or more. 16%. Hardly anyone in the non-positive group. Tell me again that long COVID is non-existent in children.

Significant mental health impacts as well- just as we've seen with adults, children also have persistent neuro-psychiatric disease post-COVID. Given baseline data are not available here, we cannot make causal inference, but the pattern is worrying nevertheless.

In summary, the ONS data shows significant proportions with long-term symptoms post-COVID-19- even using approaches that underestimate prevalence. With significant impact on their well-being and day to day lives. It's time we stop ignoring this.

1 in 100 primary school age children and 1 in 37 secondary school age children reporting an illness that lasted 12 wks and impacted their day to day lives since March '20 is nothing short of a huge failure. And remember this is pre-omicron.

How much of this could we have prevented with mitigations in schools? With earlier vaccination of children? But we simply didn't prioritise this. And sadly there are now thousands of children who have continuing persistent symptoms for more than a year.

As adults got more and more vaccinated & boosted, rather than extending the same protections to children, we hung them out to dry - tolerating higher and higher levels of infection because they weren't translating to the levels of deaths we saw early in the pandemic.

In the process the least protected & most exposed were left behind. Our children.