Read on Twitter
There are lots of people in the UK staring hard at coronavirus case data, trying to see the effects of vaccination coming through.
Not a thing so far, which is causing some angst. (1/9) https://twitter.com/ChrisGiles_/status/1358016395276136450
Not a thing so far, which is causing some angst. (1/9) https://twitter.com/ChrisGiles_/status/1358016395276136450
But we have Israel's data, which not only *do* show a clear effect, but also give the UK some guidance on where to look.
First: the effects show up in hospitalisations a good 4 days before they do in cases - so we should look at that data first. (2/9) https://twitter.com/segal_eran/status/1356313705684869121?s=20
First: the effects show up in hospitalisations a good 4 days before they do in cases - so we should look at that data first. (2/9) https://twitter.com/segal_eran/status/1356313705684869121?s=20
Second, we see effects driven by *first* doses alone (far too few second doses had been given by the time the effects showed up). So we don't need to worry about the different schedules in the UK and Israel (3/9)
To get further we need assumptions. Using standard timings of ~12 days for 1st dose effect (BioNTech trials) plus ~8 days from infection to hospitalisation, we have to look 21 days back from an effect showing on 20 Jan = 31 Dec. By then Israel had vaccinated ~45% of 60+s. (4/9)
In the UK, we split our vaccination data by 80+ and below. According to NHS England then at 21 days before our latest data (11 Jan) England was not at 45% of 80+s.
In fact, only one NHS region was above 45%: North East and Yorkshire. (London was trailing down in the 30s.) (5/9)
In fact, only one NHS region was above 45%: North East and Yorkshire. (London was trailing down in the 30s.) (5/9)
So, if we look at the very oldest age-groups, we should be able to see a hospitalisation effect *just* starting to show in the North East and Yorkshire. And it should be absent in other regions - especially in London, which was a long way behind. We have this data. (6/9)
Here is the North East and Yorkshire, which should just be *just* starting to show an effect in the very oldest groups. (7/9)
And here is London, which should not be starting to show an effect at all. (8/9)
This is *not* evidence that the UK vaccination programme is working. These are miniscule shifts, and when you want to see patterns ... you see patterns.
But, if the UK programme *was* working, then so far, it should look pretty much like it actually does. (9/9)
But, if the UK programme *was* working, then so far, it should look pretty much like it actually does. (9/9)
