This week's Test & Trace report shows cases falling sharply in the week to November 25th, thanks to lockdown 2.0.
There's also a sudden big improvement in contact tracing performance (which turns out to be smoke and mirrors), and some odd revisions to old pillar 1 testing data.
There's also a sudden big improvement in contact tracing performance (which turns out to be smoke and mirrors), and some odd revisions to old pillar 1 testing data.
Cases in England were falling sharply towards the end of November, with both the number and percentage of people testing positive falling by about a quarter compared to the previous week.
There's a small drop in the number of tests done, but it may well be due to reduced demand.
There's a small drop in the number of tests done, but it may well be due to reduced demand.
In case there's any doubt about the effectiveness of lockdowns, regional data shows clearly that all areas where cases were level or still rising when lockdown 2.0 began started falling in perfect unison about a week later.
The question now is whether that can be sustained.
The question now is whether that can be sustained.
Turnaround times for tests continue to improve, getting back to levels last seen in the summer, before the system descended into chaos as demand outstripped capacity.
They're still only managing to deliver 54% of "in person" test results within 24 hours though.
They're still only managing to deliver 54% of "in person" test results within 24 hours though.
Meanwhile I've spotted another oddity in the data - the number of people tested in Pillar 1 (NHS & PHE labs) in the week of October 29th - November 4th has been heavily revised every week since then.
Originally it was reported as 406,967.
It's now up to 463,682!
Originally it was reported as 406,967.
It's now up to 463,682!
It's not unusual for minor revisions to be made to old data, usually the following week due to reporting delays.
But not 10 or 20 thousand at a time, and not for three weeks in a row.
20,805 added in the November 19th report
22,213 added November 26th
13,697 added December 3rd
But not 10 or 20 thousand at a time, and not for three weeks in a row.
20,805 added in the November 19th report
22,213 added November 26th
13,697 added December 3rd
It might just be a coincidence, but that's the week when Pillar 1 lab capacity mysteriously doubled overnight, as the government tried to hits its target to have capacity for 500,000 tests a day by the end of October.
Did this effort cause issues?
Did this effort cause issues?
https://twitter.com/_johnbye/status/1323406605090697217?s=20
With cases falling, there was a sharp drop in referrals to contact tracers. Performance here is exactly the same as in recent weeks, reaching a reasonable 85% of cases, mostly within 24 hours.
And this week slightly more were called rather than just filling out a form online.
And this week slightly more were called rather than just filling out a form online.
The big change though comes in the percentage of contacts reached by the outsourced call center system run by Serco and chums.
This shot up from 59% to 71% in a single week, after no significant improvements for months.
Needless to say, there's more to this than meets the eye.
This shot up from 59% to 71% in a single week, after no significant improvements for months.
Needless to say, there's more to this than meets the eye.
The explanation comes in a note in the report. Previously some people were bombarded with phone calls because contact tracers had to try calling everybody in the house separately.
Now they just tell the person who tested positive to tell any children to self-isolate.
Now they just tell the person who tested positive to tell any children to self-isolate.
Matt Hancock claimed "continued improvements", but this is the only change in contact tracing performance.
It's sensible, and baffling that it took six months to do.
But underlying performance hasn't changed, as seen in contacts who aren't part of the case's household.
It's sensible, and baffling that it took six months to do.
But underlying performance hasn't changed, as seen in contacts who aren't part of the case's household.
Interestingly, the % of contacts with no contact details available has also dropped sharply.
Suggesting that lots of people refused to let contact tracers call children living with them, but they were still counted as identified contacts they had no details for!
Suggesting that lots of people refused to let contact tracers call children living with them, but they were still counted as identified contacts they had no details for!
So apart from this change to the way under age contacts are counted, there has been NO improvement at all in the performance of contact tracing during lockdown 2.0.
Which leaves the question, will cases start rising again as restrictions are lifted and we unlock for Christmas.
Which leaves the question, will cases start rising again as restrictions are lifted and we unlock for Christmas.
And here's my analysis of last week's report for comparison.
https://twitter.com/_johnbye/status/1332255356702117888?s=20
• • •
Missing some Tweet in this thread? You can try to
force a refresh
