Thread by @gnomeicide

Covid data in the UK again worrying. Here's the overall picture before looking more closely. tl;dr - we are on the edge of the precipice (1)

In a break from tradition I feel I must cover recorded cases first. We've gone over 100k reported cases in a day for the first time. 106,122 to be precise. The trajectory is bonkers (2)

643,216 cases over the last 7 days. 2nd day in teh 600k's. we were in the 500k's for 3 days, the 400ks for 4 days. In the 300k's for 21days. The recent rate of growth is extraordinary. (3)

Maybe its settling down at about a 7% rise a day but at this stage who knows? This is bonkers. We've seen nothing like this. (4)

Cases are currently doubling every 16 days, but we're stretching testing capacity to the limit and that could create some strange phenomena in the data. So we need to be a bit more creative to work out whats really happening (5)

Astonishingly despite the fact that cases have been trending up for such a ling time, deaths are still static. (6)

...but if you're happy with stasis at 112 deaths a day then I'm sorry but you're not ok. This is still awful (7)

But this is still the best news we have - deaths as a proportion of infections at a 15 day delay has been falling steadily since the booster program took hold in the UK. Vaccines save lives. But deaths aren't falling because infections have been rising. (8)

That figure of 0.232% is the lowest since Delta took hold. (9)

Because cases were rising while the proportion of cases becoming deaths was falling, and because the 15 day delay is a mid-point, its peak deaths but thats quite a variable figure, its hard to pick out any trends in the change daily deaths (10)

Hospital admissions data lags, with Scotland, Wales and Northern Ireland reporting on different schedules to England. So that data lags behind. Hospitalisations typically track cases - this doesn't look promising (11)

Deaths don't always track hospitalisations as closely. But again, its not promising (12)

We only have national hospitalisations data until the 18th. Data for England is available for the 19th and 20th and that shows a steep rise - this may be hospitalisations catching up with cases. We'll get UK Data soon enough, I'll update with another thread when we do (13)

Bluntly, we'll see in the next 2-3 days of national hospitalisation data whether or not hospitalisations are starting to follow the fast rise in cases. England data makes it look likely. We shall see. (14)

And I'm saying 2-3 days because we know how long it takes to get to peak infectivity from being infected, and the average time to die from that point, from which we can work out a proxy value for R from each data set (15)

....and if we look at trends for R through the whole pandemic we can line them up to work out what those delays are (16)

As you can see from cases (red), deaths (blue) and hospitalisations (green) the three calculations line up well. Look now a the recent trends in all 3 (17)

English data makes it look like hospitalisations are starting to go crazy. Cross your fingers thats a blip, that this time, for the first time, cases won't drag admissions and deaths up (18)

To me it still seems a hell of a roll of the dice, to gamble that they won't, to bet the house on Omicron being massively less hazardous than Delta. Booster vaccines are helping -enormously-, thats easily demonstrated (19)

...but to risk so many lives? I do not, and will never, understand. Take care folks, look after yourselves and each other. Because Boris Johnson is willing for your body to be one of those left to pile high. (fin)