Data and analysis thread, started 12 November

[NoGoodPunsLeft]

Previous thread here:

Data and Analysis Thread, started Oct 29 www.mumsnet.com/Talk/coronavirus/4064113-Data-and-Analysis-Thread-started-Oct-29

Regular lurker but I frequent poster, didn't want to lose the threads.

The cases page seems to have updated. However the cases download file from the about the data section is just empty rows when I just downloaded it, hopefully they will fix that soon

I think tests performed are always low on a Monday (the latest date on the board) - last Monday 160k, previous Monday 121k for pillar 2

Full cases files now...........full!

Yes, tests drop at weekends so Monday numbers are usually low, but it's Tuesday. I'd consider it being fewer tests, but we have mass testing in Liverpool to add in. Wait and see what the next few days bring...

Last time it was this low was 9/29.
I am cynical. but I think they are playing around with daily reported numbers, then adding on to specimen date numbers but rarely do people look at that.
@boys3
do you have an easy way to quickly cross check weekly cumulative numbers per report date vs per specimen date?

Please can anyone help me with a link where to look for where the cases are for each county?

Posted too soon, I mean daily cases for each county

I know today is Tuesday...but test numbers are reported 1 day in arrears, so the number reported today is for tests processed yesterday.

Pillar 2 was lower than this on 2/11
Does make you wonder where the Liverpool tests are being reported though?

@Gobacktothe90s

Posted too soon, I mean daily cases for each county

pick and chose:

www.worldometers.info/coronavirus/
ourworldindata.org/covid-cases

This is the latest update on how many tests have been done in Liverpool since the start of the pilot of anyone clever wants to try and work out which pillar they’ve put them in!

@Gobacktothe90s

Posted too soon, I mean daily cases for each county

If you go on the Government dashboard you can put in a postcode which brings up the MSOA data, but also the daily data for that county. You can download all the data from there too, but if you just want to look up one or two counties from a mobile then that works well.

sorry @GetAMoveOnTroodon, I read countRy.

RP131 on Twitter produces this graph daily showing how positive cases reported daily filter through to specimen date.

@GetAMoveOnTroodon the dashboard indicates that they are all lumped into pillar 2, which the pillar 2 positivity rates for which Liverpool has dropped dramatically seems to bear out.

SO wrong to put them all in Pillar 2! How to confuse and obfuscate.

@TheSunIsStillShining

Last time it was this low was 9/29. I am cynical. but I think they are playing around with daily reported numbers, then adding on to specimen date numbers but rarely do people look at that. *@boys3* do you have an easy way to quickly cross check weekly cumulative numbers per report date vs per specimen date?

@TheSunIsStillShining do you mean something like one of these?

Specimen and Reported cases (just England)

• Cumulative by day
• Cumulative by week (ending)
• Weekly totals

There is a graph for the lateral flow test on the dashboard

coronavirus.data.gov.uk/details/testing?areaType=nation&areaName=England

Then the UK level states

Testing and capacity, by test type

Virus testsAntibody tests

Number of confirmed positive, negative or void COVID-19 test results, by test type. This is a count of test results and may include multiple tests for an individual person. Virus tests test for the presence of COVID-19 and include all pillar 1 and 2 tests and any virus tests undertaken in pillar 4. Lateral flow device tests are included within pillar 2. Antibody serology tests test for COVID-19 antibodies and include pillar 3 tests and any antibody serology tests undertaken in pillar 4. Testing capacity is a projection based on reports from labs on how many lab-based tests they have capacity to carry out each day based on availability of staff and resources.

@boys3
You are truly amazing :) Thanks! It's exactly what I was looking for. Could you show the difference as a number?
Looking from sept, it does look like there is a bit of under reporting, but I can't decide if it's that bad or just normal redistribution. Albeit it can't be the latter as it is weekly and not daily and somehow reporting doesn't catch up. Does anyone else see this?

Just looking at some of the authorities with the highest current rates / more striking upward trajectories.

Swale (Kent) and West Lindsey (Lincolnshire) both just over 60 cases per 100,000 in the seven days to 11th October. In the 7 days to 15th November (and bearing in mind there will be further cases still to feed through) 618 and 533 cases per 100,000 respectively.

Then Thanet (also Kent) has gone in that same time period from 21 cases per 100,000 to 456 Dartford and Medway are also now at the 300 cases per 100,000 mark as compared with 50 and 34 in the week ending 11th October.

Some sort of East Coast thing - or East Coast and deprivation? Possibly as move north of the Thames and Essex, Suffolk and Norfolk all largely (bar just a handful of councils) below the English median. No sign of any significant Uni impact in Norwich.

Although deprivation does not explain Hastings - in 20 most deprived councils in England, but, at the moment at least, in the lowest 20 in terms of current cases per 100,000.

@TheSunIsStillShining I can (weekly figures attached) but the absolute cumulative number in England for both specimen and reported does match exactly at 1,211,951.

Just a minor thing but I noticed for the first time yesterday, that on the dashboard, my postcode brings up an incorrect MSOA (one several miles away), suggesting that if I tested positive it would come under a different MSOA? I only noticed because it said zero cases and I personally know 4 people in village who tested positive.

As I said, quite minor as only a few miles away, but if it does it for other postcodes that cross borough/unitary lines, it will give an incorrect overall picture.

@Danglingmod are you sure it is the wrong MSOA ? also don't forget it will suppress any numbers below three so the potential to mislead there too. Dependent on where you are MSOAs can span some reasonably large areas in the more rural districts.

That said postcodes could go astray - I know looking at Land Registry for house sales a very small number of postcodes that I 100% know are in the district in which I live showed on the Land Reg file as being somewhere else (including at least one in an LA that does not border mine???).

Some may know more but from the ONS MSOA files it looks as if no MSOA can span, in England, more than one lower tier local authority area. Happy to be corrected if that is not the case.

[quote boys3]@TheSunIsStillShining I can (weekly figures attached) but the absolute cumulative number in England for both specimen and reported does match exactly at 1,211,951.[/quote]
I would have thought they match. I was thinking more along the line of optics and narrative. It would be so easy to spot if they actually lied, that I don't think them that stupid (well...ermmm...)

Is there any data publicly available on Covid hospitalisations by age and/or gender?

I feel like there used to be, but I can't find it.

I'm pretty sure... It gives village names, and my postcode is listed as being in a village 8 miles away with a different second part of postcode (but same first part). My postcode is suburban, the area it puts me in is rural.

(And the zero didn't stay, it was upped later - our rates are VERY high now locally - but it was the zero that made me notice the wrong area.)