Data and Analysis Thread, started Oct 29

[PatriciaHolm]

With a link to the previous header for all the great links to data -

www.mumsnet.com/Talk/coronavirus/4057030-Pure-data-thread-1-Daily-numbers-graphs-focused-analyses?

And with a polite plea to keep the focus on data and analysis if you please.

thanks all

@RedRedRobinBobbin

I don't think we know for definite whether the vaccine will create sterilising immunity or not but I don't think it's a definite that it won't.

The vaccine seems to perform better than expected so it may well be that the R value can be reduced by vaccination. We don't either way for definite yet as I understand it.

Either way the vaccine will improve things immeasurably.

Extract from a Lancet article which implies it is possible it will reduce R but shouldn’t be taken for granted.

“These observations suggest that we cannot assume COVID-19 vaccines, even if shown to be effective in reducing severity of disease, will reduce virus transmission to a comparable degree. The notion that COVID-19-vaccine-induced population immunity will allow a return to pre-COVID-19 “normalcy” might be based on illusory assumptions.“

www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31976-0/fulltext

@RedRedRobinBobbin

Yes, so we don't know the answer on sterilising immunity yet. Hopefully we'll get more information as time goes by.

@sirfredfredgeorge

TheSunIsStillShining You have no pubs/restaurants in your area?

We have 4 pubs. There has been no change in patterns of usage in the last month as all 3 are closed.

@Sunshinegirl82 I'd wait another week or so in general as nationwide schools went back at diff times. And by no means am I saying I am right, it's more that it is what I have observed, drawn a conclusion and now waiting for data to come in nationwide to see if it's true/plausible what I inferred or not.

@cantkeepawayforever
That seems logical, but doesn't change the fact that schools are drivers and would need to be drastically changed (many threads on ideas of how they should operate, not derailing this thread again with it). Instead it's the hill to die on. Despite evidence.

I'm really struggling with this schools logic @TheSunIsStillShining. You think children are spreading covid at school, yet can take 1-2 weeks off and all immediately start infecting each other / everyone else on their return - but the infectious times and the cases don't add up? And why would young children keep getting a higher proportion of negative test results than adults in every prevalence test regime globally if they are so infectious? It really isn't only the UK, you have to look at research globally and explain why you think every country is wrong on this.

[quote Sunshinegirl82]@TheSunIsStillShining

If the rise were solely down to schools I would have thought we would be seeing significant rises in the younger age ranges and I'm not sure that we are to be honest? This chart from RP131 on Twitter shows today's positive's by age compared to last week. It's up and down in various age ranges across the regions. I can't see that school age children are on the up more than any other age range particularly?[/quote]
There was a 50 fold increase in year 7- year 11 from the beginning of September up till October 23rd. Yes, this does appear greater than other age groups, over that time.

Data over half term and this week is a bit complicated because of the complexities of tier systems in some places and then lock down.

However, it does seem that figures for both primary and secondary aged children were very low before schools went back in September, and then shot up enormously.

As we have had schools going back with very few of the WHO recommended measures for schools in place, one strong possibility is that infections are spreading rapidly in schools because of the poor levels of provisions there.

In fact it would be hard to see how you could avoid this conclusion?

@herecomesthsun

Well, if you look at @MRex's post below yours that's how really. I'm not saying schools are definitely not a driver but when people say "the cases are stabilising because of half term" I'd say that, in my view, the data doesn't seem to support that as things stand.

RP131 on Twitter runs a daily chart of positives per 100k people by age range. Primary and pre school rates have been pretty consistent throughout, no significant spike.

The 10-19 age range shows a sharp rise but it rounds at almost the same rate as the 20-29 age group. Obviously children some of those in the 10-19 age group will be at college/university as will some of those in the 20-29 age group.

I can't see any impact on rates as a result of half term.

But I'm saying there are multiple factors - the tiers, the lockdown - as well as half term. There was Halloween when people normally get together, and I'm sure some people will have managed the odd firework party in there too.

Also, half terms are at slightly different times and durations in different schools and different party of the UK.

So it would be hard to extrapolate much from the past 2 weeks.

However, if you look at the data for the 6 or 7 weeks from the beginning of September, there is a large increase specifically in the 11 - 16 year olds.

And the most elegant explanation of this would be that this follows from the poor levels of virus mitigation in secondary schools.

@herecomesthsun - Primary now too! There is a date issue in your explanation. Cases everywhere and in all age groups have been growing since the beginning of August when the new variant from Spain started to really take off. Lots of people catch covid from household members as well as other connections, but the majority catch covid from a superspreader. If you had rampant infections in schools, it should be 80% of a class actually catching it, you'd know about the asymptomatic because they'd infect their parents. Not the one or two cases here and here that we see, but huge numbers.

Young people 14-18 have certainly had a big increase, though nothing like as high as the 18-24 group. So either they are still less likely to be infected despite having gone through puberty (possible, but lower probability), or it's to do with things they do differently. I do have concerns that there aren't enough mitigations for that age group and their teachers, though I suspect there's a lot more mingling out of school creating issues. I can't get concerned about younger kids in school because there just isn't the data there to support it.

If you'll forgive me the sarcasm, I also can't see the point in believing that kids are totally asymptomatic, don't get positive test results, don't infect parents nor other family and keep getting reinfected because half the class had covid in March, June and every week since the start of September... But believing it also matters that they're infected when all that's true.

It is so tedious having the same arguments ! when I am sure the figures have been posted on these threads. I'll see if I can find them again.

Indeed. I know I've definitely posted it all before. They're in the surveillance graphs, if it's age distribution you want. Or you can look at the age heat maps to see that not all areas are identical, but the youngest are much lower risk. You can look at the government map of cases to see that the rise started in August.

inews.co.uk/news/politics/scientists-urged-government-to-close-secondary-schools-744795?ito=twitter_share_article-top

1. From the above article:

"Scientists advising the government urged ministers to close secondary schools for the month-long lockdown in England, i understands.

Shutting schools for older children was one of the measures scientists said would help bring R below 1 due to high infection rates among teens, along with closing pubs, restaurants and non-essential shops.

But the proposal was blocked because keeping education open throughout the second lockdown is a red line for the Prime Minister.

Official figures show that coronavirus cases in the year 7 to year 11 age group in England increased 50 times between 1 September and 23 October, from 40 to 2,010, despite schools operating a system of bubbles for classes or year groups."

1. There were also figures in October showing sharp rises in primary school children, in whom ONS graphs at one point showed the sharpest gradient of increase. (Sorry, please have a look at that yourself, but it is there) See also the ONS figures attached.

1. It's really interesting that people choose to ignore this - because it's inconvenient?

Yes, here is a chart the surveillance data below that I have posted a number of times before. I agree that the situation is very different in different parts of the country - but overall the figures very much support what I wrote.

This is why both SAGE and Independent SAGE were concerned about schools remaining open, and why the WHO thinks there should be far more stringent measures in schools, especially secondary schools. Are you aware of that?

Ah yes, the other point was the holidays. Just because there was an effect of people returning from holiday, it doesn't negate the effect from schools, I'm afraid.

Why would return from holiday cause a 50 fold increase in cases in secondary school children over the next 7 weeks? Surely that was the return to schools?

@herecomesthsun

Demonstrating that cases went up in school age children does not establish that schools are a driver of infection. It's the old correlation is not causation argument.

If cases are high in the community, they will be high in schools. We know that cases started rising in late August, before schools went back.

Certainly primary schools appear to have had a very limited impact on spread. There is more of a question mark over secondary schools (particularly older children).

Ah but the chart I attached with ONS data showed that from 1st September - 23rd October the rate of increase was far higher in secondary schools than anywhere else in the community.

That would strongly support transmission in secondary schools.

Here are the figures again.

@Piggywaspushed

I have a thread I actually started and it is constantly dropping off my Threads I'm On because we revisit monthly. I might start putting the odd . to regenerate it. Hadn't thought of this genius really before until people started casting aspersions on anyfucker. Top tip.

Does no-one use 'Threads I'm Watching'? Just make sure your settings are that threads get added to this list when you post. You can easily 'watch' a thread at the top without posting on it.

Then go to 'Threads I'm Watching' and you can 'bump and reset' to update where you're watching.
Nothing drops off unless you have over 100 threads on there. no matter how long ago they were started/posted.
Anyway. Sorry.

As regards Primary, the ONS graphs suggest primary schools heading towards a similar incidence to secondary schools on October 23 in Figure 5 here www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/23october2020#age-analysis-of-the-number-of-people-in-england-who-had-covid-19

(I'm sorry, I couldn't get the graph to copy into this post)

This was just as the infection rate was taking off, just before half term. There was
-a sudden apparent dip in secondary school incidence,
-a plateau in year 12- age 24 and
-primary school incidence here appeared to be trending to equal to or greater than the other age groups in the community. These figures were hard to interpret and the various tiers and lockdowns then followed sharply.

I think the figures strongly suggest transmission in secondary and I would at the very least reserve judgement with primary.

If we had WHO recommendations in place in UK schools it might be very different.

I don't see how there can be any real debate around infection rates in schools -- even the Sage documents are very clear about the significant impact on R rate of closing/opening schools (particularly secondary schools and colleges).

Interesting to see that the rates among primary school age children in the ONS survey have seen steeper rises recently than other age groups. Does anyone have any ideas about why that might be?

I think the ONS survey is the most reliable source of data for that age group as we know that children often present with symptoms other than the 'big three' that qualify for testing at the moment.

By the way, does anyone have any idea of when the preliminary data from the Bristol schools research is likely to be released?

@MRex
Let's say I come around to your argument pov. Then the question arises: what drove down the case numbers in this specific MSOA when nothing else changed? For this specific instance i would safely dare say causation rather than correlation, but am really open to other explanations.

So with mass outbreaks of freshers 18-19y. Then you have a level students turning 18 in sept so at least 1/6 of that year already 18. The likelihood of them not being superspreaders too is erm ...
Kids have siblings so if we allow all secondary to infect each other and their parents then they have also dramatically increased the chance of it spreading into primary. And obviously via secondary parents into care homes and hospitals and primary schools and offices.
We know that as few as 1/10 were symptomatic in one university. So replicate that to the numbers of cases in thousands at secondary to find how many were actually infected.
Anyway not allowing teachers to wear masks is a mistake as they obviously can infect each other and need to be in a small space with a TA and can infect the children.
Obviously children arent going to be vaxxed any time soon. So infecting the children is part of their strategy but if the infectionis going so fast they have to keep shutting the economy diwn due to the 0.5 of r that is schools then surely it is easier to sort that. If over 5 say are contagious then there would be millions leaving virus around in shops where under 11 dont need to wear masks and on PT. And it's the sheer volume of them.

@MRex why would young children keep getting a higher proportion of negative test results than adults in every prevalence test regime globally if they are so infectious? Have another look at the latest ONS figures, you got that backwards, sorry.