Week 12 NHS Test and Trace data summary. 13/8/20 to 19/08/20.
Mixed news this week - good for case numbers, less so for NHS T&T performance.
Summary in picture, detail in thread 👇
gov.uk/government/pub…
Mixed news this week - good for case numbers, less so for NHS T&T performance.
Summary in picture, detail in thread 👇
gov.uk/government/pub…
KEY POINT.
8% fall in cases compared to week 11, on the back of a 27% increase the week before.
This includes
-24% fall in pillar 1 (hospital cases/care home outbreaks)
-6% fall in pillar 2 (community cases)
8% fall in cases compared to week 11, on the back of a 27% increase the week before.
This includes
-24% fall in pillar 1 (hospital cases/care home outbreaks)
-6% fall in pillar 2 (community cases)
Alongside
-2% overall rise in testing (3% fall in pillar 1, 4% rise in pillar 2)
-Fall in percentage of tests positive in both pillars 1 and 2 (0.46% from 0.59%, and 1.94% to 1.76%)
-2% overall rise in testing (3% fall in pillar 1, 4% rise in pillar 2)
-Fall in percentage of tests positive in both pillars 1 and 2 (0.46% from 0.59%, and 1.94% to 1.76%)
This is interesting in that there’s a general trend in increasing cases since wk 6, but no consistent wk by wk pattern and certainly no exponential growth.
Also interesting to see fall in percentage of ppl testing positive, which may a bit be due to more community testing. But intrigued that number of pillar 1 tests have fallen as well as fall in % positive. This is probably good news.
looking at gov.uk daily data:
today have 1,522 cases in UK incl. 1,350 in Eng (highest in Eng since 08/05).
Worth noting there are periodic daily spikes in cases (e.g. 1,284 on 14/8). But on 6 out of past 7 days, cases in week to 27/08 are more than week before.
today have 1,522 cases in UK incl. 1,350 in Eng (highest in Eng since 08/05).
Worth noting there are periodic daily spikes in cases (e.g. 1,284 on 14/8). But on 6 out of past 7 days, cases in week to 27/08 are more than week before.
Cases still trending upwards but definitely not (yet) exponential. And neatly mirroring NHS T&T data (as you'd hope), meaning likely to see small increase in cases for wk 13 (wk ending 26/08).
coronavirus.data.gov.uk/cases
coronavirus.data.gov.uk/cases
And as with last few weeks, still no sign of increased hospital admissions or deaths over past weeks. Likely reflecting younger people being infected with less severe infections.
Why this week’s drop in cases & large increase the week before?
-I’m unsure about impact of holidays. More community testing this week than before so ppl are still getting tested, but perhaps less circulating virus (or at least, continued low rates, as per ONS)?
-I’m unsure about impact of holidays. More community testing this week than before so ppl are still getting tested, but perhaps less circulating virus (or at least, continued low rates, as per ONS)?
-People might be getting tested late because of waiting to come home from hols, meaning more false negs. But if this were the case, it should show up on ONS data.
-Whatever is happening, with relatively low circulating virus there is likely to be a fair bit of wk by wk variation.
-Whatever is happening, with relatively low circulating virus there is likely to be a fair bit of wk by wk variation.
Drop in pillar 1 cases also interesting and is good news.
As before. Read this thread on Florida – it is so important to continue to track this, young people can go on to infect those who are much more vulnerable.
As before. Read this thread on Florida – it is so important to continue to track this, young people can go on to infect those who are much more vulnerable.
Here's the current epi curve with major policy announcements annotated.
Now NHS T&T.
Numbers transferred is 7,941 from just 6,115 cases. This 1,826 difference basically makes up for the 1,853 cases that weren’t transferred last wk.
Numbers transferred is 7,941 from just 6,115 cases. This 1,826 difference basically makes up for the 1,853 cases that weren’t transferred last wk.
We know that there was a delay in uploading 681 cases (reported last week), but it seems that the other 1,200 or so were also significantly delayed.
Hopefully this is now all rectified and will be back to 'normal' next week.
Hopefully this is now all rectified and will be back to 'normal' next week.
Proportion of people reached was just 73%.
It hasn’t been below 79% since week 3. Maybe this is to do with the delayed case transfers (unexpected workload? information incomplete?)
Either way, it’s a big drop in performance and worrying to see as cases trend upwards.
It hasn’t been below 79% since week 3. Maybe this is to do with the delayed case transfers (unexpected workload? information incomplete?)
Either way, it’s a big drop in performance and worrying to see as cases trend upwards.
And continued and worrying decline in number of people providing details of contacts. I find it hard to believe this is to do with people not having contacts.
Is it lack of trust in T&T?
Not knowing details of contacts?
Are these more likely if cases are younger?
Or is it real?
Is it lack of trust in T&T?
Not knowing details of contacts?
Are these more likely if cases are younger?
Or is it real?
Big jump in contacts because of more people transferred to system, but contacts per case for non-complex has fallen for third week in a row and is now 2.7, lowest since wk 3 (had got as high as 3.5 in wk 9).
Is this also because of changed case demographic?
or less trust in system?
or is it genuine as people are away on holidays with family and not socialising in groups?
Contacts per case for complex roughly unchanged.
or less trust in system?
or is it genuine as people are away on holidays with family and not socialising in groups?
Contacts per case for complex roughly unchanged.
(also, T&T are now really usefully reporting median numbers of contacts per case - this is great to see. I can't replicate these numbers as would need individual level data, hence I calculate means)
One bit of good news, % of contacts reached has increased from 72% to 76% (including increasing from 59% to 62% for non-complex cases).
As mentioned before, some important issues around regional variation that need to be teased apart.
And finally, the test time is really very poor and getting worse.
Some time to reach case from test will be driven by the issue of last weeks' cases not being put onto the system, but even once on system, time to being contacted is getting worse (now <70% w/i 24hrs)
Some time to reach case from test will be driven by the issue of last weeks' cases not being put onto the system, but even once on system, time to being contacted is getting worse (now <70% w/i 24hrs)
So with a week before schools go back, it's better news for case numbers (but gov.uk data suggest still generally trending up), and worse news for NHS T&T performance.
I'm still unsure about impact of holidays on case numbers, contacts, demographics, etc.
I'm still unsure about impact of holidays on case numbers, contacts, demographics, etc.
It may all change once schools are back.
And whatever happens, fingers crossed will see improvements in NHS T&T perforance as more local govs are involved (& with launch of NIHP??!)
And hopefully will see this improvement as schools return and before flu starts.
And whatever happens, fingers crossed will see improvements in NHS T&T perforance as more local govs are involved (& with launch of NIHP??!)
And hopefully will see this improvement as schools return and before flu starts.
