Just updating the deviation graphs that I posted a few days ago: no major changes in trend, but good to see the cases in 75-79s continuing to head strongly downward under the influence of the vaccine, following the 80+: (short thread) 
(to be clear on method, these figures are the cases in each age group expressed as a % of the total, and then indexed to 100% on 25th Jan so that we can see the deviation in each group on the same scale) 2/n
here's the chart for each of the subgroups within the 80+, you can see the 85-89s falling furthest, but the 90+ now working to catch up. 3/n
one of the puzzles over the last few weeks has been trying to explain the behaviour of that 90+ group: they were the first to start being vaccinated, but it took a long while for this to become visible in the case data. 4/n
it could have something to do with this pattern which is from the Public Health Scotland data released yesterday (h/t @petesivey for spotting it) - 90+ started first but took a long time to complete. I don't know if England was similar, but if so it could help explain 5/n
alternatively, it may be that my theory that "lockdown works less well for the 80+" is just even more true for the 90+ because they are more reliant on care/support so can't lock down to the same extent. this would mean that we'd expect their case rates as a % of the total 6/n
to *rise* during lockdown. so the vaccine had a big hill to climb, to counter-act that effect, before it could start to show falling cases as a % of the total. it took a while to do it, and now we're starting to see the result. [or it could be a bit of both]. /end
• • •
Missing some Tweet in this thread? You can try to
force a refresh
