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.

I would put essential worker in if I was a teacher.

Today's figures - 462 deaths with 26,860 positive tests

Last weeks figures for Saturday were 413 deaths with 24,957 +ve tests

Is the curve flattening at last?

@Tyzz

I would put essential worker in if I was a teacher.

Teachers are essential workers for testing purposes. I think they are listed as an example when you fill in form but full list is here www.gov.uk/guidance/coronavirus-covid-19-getting-tested#list-of-essential-workers-and-those-prioritised-for-testing-england-only

I didn't have any problems getting my school aged son a test recently (could have gone immediately, waited an hour to allow us to get there). Didn't have to chose essential worker and filled it in honestly. Suspect it's more system hiccups, local demand and tests being added to system during the day causing the availability to change but social media loves a conspiracy....

Teachers are essential workers.

Getting a test in Sept, for anyone, was like pulling hen's teeth...

Re todays numbers I don’t think we can overall say levelling no it’s an increase but trends beyond that seem interesting!

So today saw a number of ex tier one areas reporting record rises in cases over 3300 were added in the south west today, that is significantly more than London for example and only 300 less than the north west. The south east shows a similar pattern. Otherwise our previous hot spot of the north west is not just levelling but coming down.

My hypothesis would be

• tier one restrictions not enough
• tier three maybe ie we don’t need a full lockdown to get r below one and some businesses can operate.

Interested in terms of if anyone made a note on when other areas that were not Liverpool and Lancashire entered tier three? And in what order? It would be interesting to see if they also trend downwards in the next week or so following some impact of tier three, this is def the case in Liverpool now even before mass testing.

We were in tier 1 and we are 220 in 100000 now, we were 30 in 100000 not so long ago.

In fact, I was trying to fill the form in as honestly, as I could but at that point, there was no option for "school pupil", and you had to choose an option to go forward with the form. So I chose the closest one I could find, trying to get the system to work. it still took a day and a half to get the offer of a test, and the whole process took a week.

It's good to hear that the process has improved!

@Qasd places seemed to go into Tier 3 as they went purple on the map, at rates over 400/100k.

@Qasd

Re todays numbers I don’t think we can overall say levelling no it’s an increase but trends beyond that seem interesting!

So today saw a number of ex tier one areas reporting record rises in cases over 3300 were added in the south west today, that is significantly more than London for example and only 300 less than the north west. The south east shows a similar pattern. Otherwise our previous hot spot of the north west is not just levelling but coming down.

My hypothesis would be

• tier one restrictions not enough
• tier three maybe ie we don’t need a full lockdown to get r below one and some businesses can operate.

Interested in terms of if anyone made a note on when other areas that were not Liverpool and Lancashire entered tier three? And in what order? It would be interesting to see if they also trend downwards in the next week or so following some impact of tier three, this is def the case in Liverpool now even before mass testing.

You can see this from the legislation- it shows when each area was added it the footnotes www.legislation.gov.uk/uksi/2020/1105/schedule/2/2020-10-30

TBH I can't see Greater Manchester rates dropping significantly any time soon, we've been under so many various restrictions for months now and the rates have only gone up.
Students heading home for the Christmas break in week 3-9 December will cause a pretty-exodus testing spike, then a drop as those student cases redistribute round the country.

Thank you @Firefliess, that helped me to find the original research that the 8% figure comes from. It is here: www.medrxiv.org/content/10.1101/2020.11.01.20222315v1.full-text

It is a preprint that has not yet been peer reviewed, but has a pretty extensive and well-qualified author list.

Here is Figure 1 from it with the exact figures.

@herecomesthsun

In fact, I was trying to fill the form in as honestly, as I could but at that point, there was no option for "school pupil", and you had to choose an option to go forward with the form. So I chose the closest one I could find, trying to get the system to work. it still took a day and a half to get the offer of a test, and the whole process took a week.

It's good to hear that the process has improved!

Sorry, didn't mean to imply you did anything wrong (would have been tempted to do same) just more saying I didn't have to lie about my son being a child to get a test 2 weeks ago. I think there was a real issue with tests in September (colleagues struggled and they could tick essential worker looking at that list) but it's definitely improved around here.

The info in Figure 1 of the LSHTM study seems to suggest that adults living with children (1) have got a slightly higher chance for catching COVID (~8% for children over 12) but (2) are at a much lower risk of dying from COVID. In other words, parents and caregivers are more likely to have a mild version of COVID than the general population in the same age group.

I remember one of the older German studies indicating that close contact with children may be protective - the hypothesis was that parents get an opportunity to build additional immunity from frequent exposure to common colds. It may be one of the explanations for (2) although, as usual, correlation does not mean causation and there may be other factors at play (e.g. differences in life style between parents and people without children in the same age group).

Thanks @MotherofDragonite. Good to have a look at the actual paper. Looks like a solid piece of research and they've controlled for all the obvious factors with the exception of number of people in the household. They did control for number of adults but not total number of people (I'm guessing that doing so would have been tricky as there isn't such a large number of households with 3+ people none of whom are children) And as others have pointed out up thread, living with anyone is going to increase your risk relative to living alone or with fewer others. Comparing parents with other adults is really tricky in general because so many factors are strongly associated with being parents (health, income, relationship status, etc) You do get "threshold studies" that compare just either side of a significant boundary - so would be interesting in this case to compare parents of 17-18 year olds living at home, with parents of 18-19 year olds living at home and working from home (the closest proxy you could get to having school aged children studying from home). But research is rarely perfect. I think the study does suggest that any increased risk through living with children isn't huge. And if we closed all the schools it still wouldn't be zero unless we locked all children in their rooms and let them see noone, which would be obviously wrong (and impractical)

Haven't RTFT, but why are there no hospital admissions numbers since 10th November?

@Freeekedout

Haven't RTFT, but why are there no hospital admissions numbers since 10th November?

Because that's the last data all four UK nations reported; there is data for 11th for England and 13th for Wales/NI, but only 10th for Scotland so the overall number is from the 10th.

Thanks @MotherOfDragonite and @Firefliess, that's interesting. I had forgotten about the suggestion that the host of other recent coronavirus infections was possibly protective. I do think that not accounting for different sized households makes a big difference though, because larger households have consistently been shown to be a larger risk factor (www.medrxiv.org/content/10.1101/2020.10.13.20212126v1.full). Though equally, in other research the larger size of household isn't necessarily separated from deprivation factors.

Ahhh that makes sense. Thank you!

The info in Figure 1 of the LSHTM study seems to suggest that adults living with children (1) have got a slightly higher chance for catching COVID (~8% for children over 12) but (2) are at a much lower risk of dying from COVID. In other words, parents and caregivers are more likely to have a mild version of COVID than the general population in the same age group

That's a possible conclusion, but also remember people are very unlikely to die in this age group without additional factors, isn't it quite likely that those same people are less likely to have children? So the groups aren't really fully comparable.

*GetAMoveOnTroodon
The BMJ aren’t holding back today! Covid-19: politicisation, “corruption,” and suppression of science

The author of the article in your link (and the executive editor of the BMJ) doesn't try to hide his political bias;

Twitter-Kamran Abbasi
"With a nod to the demographics, the same people who gave us Thatcher, people who voted her in out of greed and self interest, have now given us Johnson and Brexit. You can fool many of the people many times over since nothing predicts behaviour like behaviour."
10:24 AM · Dec 13, 2019

P2 7 day positivity update, usual format (so light green 2% or less, light blue up to 5%, etc until the deep red for over 15%; DoT (direction of travel) only flagged green or red if week on week movement at least 2 percentage points, most recent 7 day period goes up to last Thursday, no 7 day periods overlap).

Just to stress the Liverpool figure is not a typo.

Upper tier LAs attached, and districts will follow on the next post.

and the district council areas

Wow, Liverpool!
By the way @boys3, I find it hard when my phone is darkened to see the difference between the two orange colours. Would it be possible to upgrade red to purple, then have red, then orange please?

(For future versions I mean)

Remember on the last thread's closing discussion we were talking about small MSOA and schools driving the numbers. Half term holiday effect was up until now. Both on a council and MSOA level we are on the rise rapidly :(
This is not scientific enough, I know and it pains me, but there is still quite an obvious causation, not just correlation. imo.