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

I forgot ONS was randomly selected. It's very strange, isn't it?

One possible take is it has looked at a bunch of fairly high risk professions, but then lumped the highest risk into "other occupations".

What I'm thinking is, what if "factory workers" are by far the most risky occupation, but it hasn't split them out? There are a lot of factory workers and many of them work under quite unreasonable conditions with few mitigations, or where they are permitted to wear masks, don't.

Or equally, it could be that "shop workers" were by far the most risky, and not split out. There are a lot of them and they could easily skew "other occupations" if they were high. Or some combination of a couple of very risky, but not split out professions?

It doesn't necessarily mean "Your occupation has no bearing on your level of risk."

It does say other professions, rather than other occupations. Usually the ONs break things down a bit more than this but the professions are a specific thing, generally. Need more detail!

I think it would be really interesting to see data for all workplace settings as was published in April.

Last time (April) we saw number of deaths per 100,000 I think, and care workers had an significantly elevated rate, but healthcare professionals did not. As this new data shows prevalence rather than infection rates it's difficult to compare but it would be interesting to reflect on why the rates for all these professions are now similar abs what's changed about social care for the better.

I look forward to seeing an updated report in the same vein - as a pp said, we are a good chunk of the way into autumn and it's an opportune time as we are just locking down again and that will skew the data.

It's interesting that you now need to say if you go to an educational setting and it's name when booking a test.

23287 cases
355 deaths
Early days, but hopefully cases are plateauing.

coronavirus infections may be 'stabilising' www.bbc.co.uk/news/health-54841375

Is there any possibility this is due to schools half term break last week?

Has this been linked?

Gp practices will be set to go with t evaccine roll out at the start of December (as long as the vaccine is ready.)

www.pulsetoday.co.uk/news/coronavirus/bma-announces-gp-covid-vaccination-des-revealed-by-pulse/

Is there any possibility this is due to schools half term break last week?

Possibly. But there's also a behavioural factor too, would people get tests or tested with those mild niggles if they were away?

School cases were also reduced but positivity is still rising.

If we accept that schools/uni are responsible for the ride in cases is it not possible that the pool of people still to be infected has reduced - a kind of herd immunity on a smaller scale?

Is there any possibility this is due to schools half term break last week?

The ONS data will not particularly be impacted by the closure of the schools, as almost all the positive infections would have had to catch it before, the last testing date was 30th, with 4 days incubation before positive in a test, that would've meant the Monday of half term for the majority of the country would've been the cut of date for infection, some areas possibly had longer half terms - it is a popular half term for that, but not all.

Certainly the Pillar 2 testing would have been influenced as Neurotrash says.

Zoe's estimated daily transmission rate has gone down from ~42,000 to 40,606 cases. It is the first meaningful drop we have seen since late July.

@Barbie222

Last time (April) we saw number of deaths per 100,000 I think, and care workers had an significantly elevated rate, but healthcare professionals did not. As this new data shows prevalence rather than infection rates it's difficult to compare but it would be interesting to reflect on why the rates for all these professions are now similar abs what's changed about social care for the better.

I look forward to seeing an updated report in the same vein - as a pp said, we are a good chunk of the way into autumn and it's an opportune time as we are just locking down again and that will skew the data.

What's happened is that schools are now open, but protection for teachers is very poor. So the risk for a teacher is now similar (or worse than) a care worker in an old people's home or a nurse on the front line.

This is not what you would ordinarily expect; you would think that a teacher, quite white collar profession and not concerned specifically with health, would be at less risk than a nurse. (After all, the nurses are actually looking after patients many of whom may have covid). You would think that teachers should be lower risk until you saw the classrooms with terrible ventilation and 30 kids crammed in together, in conditions that would not be allowed in the NHS.

The confidence intervals for these figures are very large, suggesting to me that a lot more research needs to be done in teasing out different factors.

It is certainly not possible to use these figures to say that schools are safe and a heck of a lot more research needs to be done.

@sirfredfredgeorge

Is there any possibility this is due to schools half term break last week?

The ONS data will not particularly be impacted by the closure of the schools, as almost all the positive infections would have had to catch it before, the last testing date was 30th, with 4 days incubation before positive in a test, that would've meant the Monday of half term for the majority of the country would've been the cut of date for infection, some areas possibly had longer half terms - it is a popular half term for that, but not all.

Certainly the Pillar 2 testing would have been influenced as Neurotrash says.

The usual incubation period is 2-5 days (though up to 14 days is possible). So it is certainly possible that half term could have had an effect, although the tiers and the partial lockdown may also have an impact.

It will be interesting to see what happens with the figures when there is a partial lockdown but most pupils in schools are back. It is in effect an experiment, using children teachers and vulnerable families, (I would argue that this is unethical to a degree unusual in recent decades).

The usual incubation period is 2-5 days (though up to 14 days is possible)

I've not seen newer evidence that changes the original research
www.acpjournals.org/doi/10.7326/M20-1495
It's 100% false negative for PCR on day 1
And it's still 67% false negative for PCR on day 4

Whilst it is possible for 2 days, it's very much not common! There may be data and more accurate tests now, but I still think it requires a lot of infections to have occurred pre half term.

I think given how long it takes any measures to have a meaningful impact the chances any levelling are down to half term are minimal.

If one week off school made the difference then there is no way that that Melbourne would have needed to lock down for the length of time it did surely and why have Poland which shut most of its schools a while ago still seeing a rise in cases?

I really think if there is a flattening is that we are finally seeing a reduction from a tier system not one week off school, Liverpool, Wirral, st Helen’s, Sefton, Knowsley are all showing reductions and are the first to enter tier three, Knowsley in particular is interesting given no student population to screw it up. I honestly think any levelling is about tier system which is interesting as I had no faith tier three being effective but I actually think I was wrong!

I think given how long it takes any measures to have a meaningful impact the chances any levelling are down to half term are minimal.

I think levelling due to half term would not be down to fewer cases, but fewer tests. On the surveillance reports, it looked like numbers were down, but percentage of positives were up, which would point to that.
It might also explain why 2nd November (1st day back round here) seems to have had such a huge number of positive tests for England-maybe parents that were hoping the children would be better enough to go to school and so didn't test until they had to, or maybe schools sending children for tests.

Yes but ons and Zoe Lao suggest a levelling phe alone I would agree but all the main sources seem to be going in the same direction and only one is reliant on symptomatic testing?

I agree, although on the Zoe app it also says the total predictions are for people between 20 and 69, which obviously misses out the school age.

@herecomesthsun the data coming from Republic of Ireland is quite encouraging - they have had a similar lockdown to England for the last two and a half weeks (schools open, non essential retail and bars/restaurants closed) and have seen case rate drop in all but three counties, overall 14 day incidence rate fall from 278/100K to 228/100K and they believe R rate to be below 1. I’m not aware of any specific differences (and can’t find any online) between the school set up in ROI vs U.K., (except the wearing of masks which I believe will now be in English secondary schools too from skimming the thread) hopefully someone else will know.

Ireland approved a 375 million euro package to make schools safer, adapting classrooms, hiring more teachers and support workers etc.

"Is there any possibility this is due to schools half term break last week?"

In my LA cases had levelled and were starting to drop a week before half term. Here I think it was the introduction of tier 2 and the weather being so bad that people didn't go out anywhere even to their friends houses.

Does anyone have any (data based) thoughts on the mink mutation in Denmark?

On the surveillance reports, it looked like numbers were down, but percentage of positives were up, which would point to that.

That was my point earlier (badly made!)

We had a survey test through for my son and the test window was 3 weeks, 2nd Nov being the last possible date.

If everyone is healthy and you've booked a half term hol, you might choose to do it on your return than allow an asymptomatic test to ruin your break. Or if you rely on relatives for childcare and they're mostly staying with them during half term... (not always Gp's, but also GPs can be in their 40s.)

Similar with mild colds that you're not sure about.

Also, the NE's rates are interesting; flat line since they brought the tougher rules in on 18th sept. Shops were open, just no mixing with anyone. There's the tiniest of an upward slant near the end? The issue in the NE is that it's moving out to residential areas (so linking to children and schools) from the more student/ young professional rental areas. The tiers actually relaxed the rules on mixing.

The largest downturn is in the NW so it's likely that it's the extra measures pre lockdown?

There isn't a huge amount in it but Irish class sizes are smaller and their education spend anyway is one of the highest in the EU. Only the Netherlands manages to be worse than the UK for class sizes.