Thread by @DavidSteadson

Sweden's first Covid press conference in 6 weeks today. When asked if they've estimated how many kids are going to be infected in schools, the Health Authority says they've not.

So I have.

Varsågod @Folkhalsomynd @Skolverket @socialstyrelsen @lenahallengren @Anna_Ekstrom

https://twitter.com/jorgenponder/status/1425796647695564804

I did some quick calculations based on R_e last week, but in this thread I'll look more directly at the real world consequences.

https://twitter.com/DavidSteadson/status/1422619487694622720?s=20

This paper looked at risk of hospitalisation, ICU admission, and mortality by age.

Note that *this is pre-delta*. Still "fog of war" data, but there are many indications outcomes for delta is worse, so numbers in this thread may be an underestimate.

medrxiv.org/content/10.110…

The age ranges aren't quite school age and are separated in 0-9 and 10-19 yr olds, but roughly -

Hospitalisation 0.1 to 0.2%
(1 in 1000 to 1 in 500)

ICU 0.01% to 0.02%
(1 in 10000 to 1 in 5000)

Death 0.0005% to 0.002%
(1 in 20000 to 1 in 50000)

Sweden's Health Authority data on Mortality and Intensive Care for this age group, assuming ~30% infected (their estimate) gives roughly the same figures

Death 0.001% to 0.003%
ICU 0.009% to 0.013%

folkhalsomyndigheten.se/smittskydd-ber…
>

Interestingly, the Swedish data has *higher* rates of mortality in the 0-9 yr group than the 10-19 yr group, opposite to Herrera-Esposito and de los Campos paper. Given the small numbers, likely just a result of normal variance.

Next to consider is MIS-C - Multisystem Inflammatory Syndrome in Children . As of April, the Swedish Child Rheumatology Register reported 195 cases. An August report put it at 234.

https://twitter.com/covid19skola/status/1379716614078279681?s=20

https://twitter.com/EbbaSvp/status/1425114495802880002?s=20

With the June estimate of 27.7% of kids having been infected we're looking at a prevalence of about 0.03%. Approximately the same figure is reported in the US.

MIS-C 0.03%
(1 in 3333)

jamanetwork.com/journals/jaman…

The final metric of interest is long-term illness resulting from infection.

ONS (UK) data gives an estimate of Long Covid prevalence of 2% to 10% after 4 months

Long Covid 2% to 10%
(1 in 50 to 1 in 10)

ons.gov.uk/peoplepopulati…

How does this translate to what we can expect in Sweden? As per my earlier thread, according to Folkhälsomyndigheten - the Swedish Public Health Authority - there's a minimum of about 1.4 million Swedish kids with no immunity against Sars-CoV-2.

folkhalsomyndigheten.se/smittskydd-ber…

Delta has an attack rate of ~70%, so we can expect at least 70% of kids to be infected in a fairly short period of time, a couple of months at most. Given an unmitigated school environment, this is probably a lower bound of what we can expect to happen.

https://twitter.com/DFisman/status/1425077617640755201?s=20

70% of 1.400.000 = 980.000 infections

They won't all be symptomatic, but if kids "stay home if sick" as they are supposed to, there is going to be an awful lot of absenteeism.

Deaths (0.0005% to 0.002%) = 5 to 20

Mortality fortunately should remain low, but this will be a likely doubling of children's Covid deaths in Sweden over the next few months. A wildcard is Delta, where data remains unclear.

ICU (0.01% to 0.02%) = 98 to 196

This may sound a small number, but that's roughly 3 kids a day being admitted to Paediatric ICU. In normal years less than 3 new kids a day are admitted to PICU for all other issues combined.

PICU will be under pressure.

openarchive.ki.se/xmlui/bitstrea…

Hospitalisations (0.1 to 0.2%) = 980 to 1960

That's between 15 and 30 kids *a day* being hospitalised by SARS-CoV-2, where they'll likely have to stay several days. The cumulative number in hospital at any one time will be very high.

Long Covid (2% to 10%) = 19600 to 98000

🚨This is the truly horrifying number that Swedish authorities and media just don't want to talk about.

Not a single mention of it in today's press conference about schools opening next week!🚨

So, in summary, this is approximately what the data suggests Sweden can expect for our children in Autumn from SARS-CoV-2 in schools.

* 980000 infected
* 5 to 20 dead
* 100 to 200 in ICU
* 1000 to 2000 hospitalised
* 20000 to 100000 still ill after 4 months (new)

@threadreaderapp unroll