Read on Twitter
I'm really struggling to work out what is *actually* going on with COVID-19 cases in the UK at the moment.
Changes in test numbers, test positivity rates and possibly reporting delays mean that the story very much depends on when you take the most recent 'complete' date to be.
Changes in test numbers, test positivity rates and possibly reporting delays mean that the story very much depends on when you take the most recent 'complete' date to be.
Depending in where you draw that line, cases could be still rising steeply (1st Jan), falling steeply (3rd Jan), the fall could have been arrested and cases are rising again (5th Jan) or that could just be a blip and they are really falling (any time after the 5th).
Once you add in geographical variation, the picture gets even more mixed. Is the national level wobbliness being driven by a collision of different trends at regional level? Suggestions that cases in London/SE are falling, but in the West Mids/NW they're rising. Maybe.
My best guess (and it really is a guess) is that this is a reasonable stab at what's happening. Cases falling in parts of the SE and South Wales, but shooting up in Liverpool and rising in the NW and much of the rest of the UK.
That's a pretty consistent story with @carlbaker's excellent MSOA level map (based on data cut a couple of days later), at least for England.
But the picture appears to be further confounded by differences in case trajectories by age.
The dip in case rates in the last few days, even if it's real, hasn't been seen in the over 80s and the 90+ age group now has the highest case rate of *any* age group.
The dip in case rates in the last few days, even if it's real, hasn't been seen in the over 80s and the 90+ age group now has the highest case rate of *any* age group.
We'd hope that the impact of the vaccination programme might start to make a dent in these numbers soon, but given the high fatality rate (almost one in 3 for the over 90s) this can't come soon enough.
So really, we'll just have to wait and see. 
Cases are a leading indicator, so what happens to them is followed a few weeks later in the hospital data and then the mortality statistics.
Cases are a leading indicator, so what happens to them is followed a few weeks later in the hospital data and then the mortality statistics.
The massive increase in cases in recent weeks will almost certainly mean that hospital admissions, bed occupancy, ICU occupancy and deaths will all continue to get worse, potentially much worse, for several weeks yet, even if cases do start falling fast.
I'm very interested if anyone has any clearer insight into what might really be going on with cases (and I don't just mean wild speculation baldly asserted as fact, thanks, there's altogether too much of that going on at the moment).
R code for these plots is here:
https://github.com/VictimOfMaths/COVID-19/blob/master/Heatmaps/COVIDLACaseData.R
and here: https://github.com/VictimOfMaths/COVID-19/blob/master/Heatmaps/COVIDPHECasesxAgev2.R
https://github.com/VictimOfMaths/COVID-19/blob/master/Heatmaps/COVIDLACaseData.R
and here: https://github.com/VictimOfMaths/COVID-19/blob/master/Heatmaps/COVIDPHECasesxAgev2.R
