Daily numbers, graphs, analysis thread 7

[Barracker]

Welcome to thread 7 of the daily updates.

Resource links:
Worldometer UK page
Financial Times Daily updates and graphs
HSJ Coronavirus updates
Johns Hopkins Coronavirus Resource Centre
NHS England stats, including breakdown by Hospital Trust
Covidly.com to filter graphs using selected data filters
ONS statistics for CV related deaths outside hospitals, released weekly each Tuesday

Thank you to all contributors for their factual, data driven, and civil discussions.

UK Press Conf

Sir Ian Diamond, ONS Chief Statistician, says R0 has risen because of epidemic in care homes, and controlling it "a real challenge" over coming weeks

Raab says R0 is betwen 0.5 and 0.9

I think we will have coded areas like France. It's worth noting that Denmark's R went from 0.6 to 0.9 when schools "went back" but they didn't fully open.

Also, this is an interesting report apparently looked at by Gove at the COBR meeting about keeping schools at no more than 20% in order to keep R0 in check. They're currently 1.9%.

www.besa.org.uk/insights/policy-update-07-05-20/

Is there a clear explanation for how R is calculated given we do not as yet have much testing going on? That is, I know how R is calculated but I can’t see how we are using anything other than estimations and wondered how what (and how) estimates were made and then R calculated.

Is there a clear explanation for how R is calculated given we do not as yet have much testing going on?

Yes, would be good to know how R is calculated.... even if it’s a mean or median average.

Also, this is an interesting report apparently looked at by Gove at the COBR meeting about keeping schools at no more than 20% in order to keep R0 in check. They're currently 1.9%.

Sweden seem to be just about containing R0 with primary schools fully open.

It's worth noting that Denmark's R went from 0.6 to 0.9 when schools "went back" but they didn't fully open.

But is that causation or correlation. Even if a lot remained off limits in Denmark. easing restrictions even a little potentially means people are psychologically less guarded, and become a little
more free in their social interactions, nudging R0 up accordingly. That’s just a hypothesis.

Did Denmark open schools significantly before they relaxed other measures so they can definitively say the rise in R was caused by schools alone? My understanding was that hairdressers and some other businesses opened shortly after schools

The Danes are also saying there has been no acceleration in cases since lockdown was eased.

But is that causation or correlation. Even if a lot remained off limits in Denmark. easing restrictions even a little potentially means people are psychologically less guarded, and become a little
more free in their social interactions, nudging R0 up accordingly. That’s just a hypothesis.

Yes it could be all sorts of reasons. Simply more travel due to schools opening.

Nichola Sturgeon mentioned the rise of R0 in Denmark post schools opening in her speech; I googled and a few articles said the same and those rises. (Giving her reasons for not doing so soon.)

Im not sure you can compare as does the r0 take into account the number immune. As in sweden few cases and now surely most have been exposed. Slowing down spread.

So illogical to send in yr6. I would do yr 5.
Will be interesting what england do. If scotland say it will result in 58k? Extra deaths. To fully reopen.
If we can open 20% total thats about 18% more.

The Danes are also saying there has been no acceleration in cases since lockdown was eased

So are the Czech Republic and (I think) Austria

As BigChoc suggested the epidemic in care homes sounds grim, but one small silver lining here is that by their very nature they're fairly closed communities. It may be that they're contributing to a lot of positive tests, but hopefully it should be possible to put proper measures in place without the need for everyone else to exercise quite the same caution

Neuro but how much of that is that "school reopening causes R to rise" makes much more of a eye catching headline than "various easing of lockdown causes increase in Danish R value but no acceleration in cases" especially given our media's obsession with schools reopening. Most of the articles I've seen have schools in the headline but mention other businesses further down ( but make little reference to whether it was one or both that might have caused the rise I suspect because they don't know)

I'd hope Nicola S might have more firm evidence than newspaper reports but not found it myself

I don't know; I couldn't find any article in English that said more than "when schools reopened." Still below 1 though which is what they want.

My point is I imagine in areas with higher R0 (eg 0.9) fewer restrictions may be lifted, maybe Scotland, maybe pockets of the U.K.? Might increase travel to those pockets where things are lifted though.

Im not sure you can compare as does the r0 take into account the number immune. As in sweden few cases and now surely most have been exposed. Slowing down spread.

I don’t think it’s credible to believe that Sweden somehow is close to achieving “herd immunity” with a death rate that’s less than 8 times that of NYC and less than 18 times that of Bergamo (both of which have conducted various antibody tests that show neither are close to achieving herd immunity themselves at 20-25% and 35% respectively).

I found it interesting in today’s presser that London apparently currently has the lowest R0. Could this mean that the immunity in the population is higher than expected?

Indeed if Sweden have somehow achieved herd immunity, it means we can probably just all get back to normal... as if they have, we definitely will have, as will Italy, Spain and Belgium!

Office for National Statistics (ONS) @ONS

Using #COVID19 death registrations and 2011 Census records,
we’ve been able to show that some ethnic groups have a higher risk of #COVID19 related death compared with those of White ethnicity

http://ow.ly/GSNg50zzmDA

In our provisional analysis we’ve reported the risk of dying from #COVID19 across ethnic groups,
adjusting for factors that could also affect this risk:

Age

Other characteristics (such as socio-economic, geographical, and health)

http://ow.ly/5sCZ30qDVhC

After adjusting for age,
the #COVID19 mortality rate for Black males is 4.2 times higher than for White males

http://ow.ly/5v9330qDVi8

Similarly, the rate of death by #COVID19 for Black females is
4.3 times higher than for White females

http://ow.ly/5J8p30qDViL

Both males and females of Bangladeshi and Pakistani, Indian, and Mixed ethnicities also had statistically significant raised risk of death by #COVID19 compared with those of White ethnicity

ow.ly/u4Vm30qDVjo

After adjusting for socio-demographic characteristics – including deprivation –
and self-reported health and disability at the 2011 Census,
males of Black ethnicity still have a #COVID19 mortality risk that is 1.9 times higher than those of White ethnicity

ow.ly/4BKb30qDVkp

Using the same characteristics,
males in the Bangladeshi and Pakistani ethnic group have a #COVID19 mortality risk that is
1.8 times higher than those of White ethnicity.

For females of this ethnicity, the mortality risk is 1.6 times higher in comparison

ow.ly/Qy8230qDVlo

Meanwhile, the #COVID19 mortality risk for females of Black ethnicity is also 1.9 times higher than for those of White ethnicity

ow.ly/b8F730qDVmh

Our analysis shows that
part of the difference between ethnic groups in #COVID19 mortality is due to socio-economic disadvantage and other circumstances,
but a remaining part of the difference has not yet been explained

ow.ly/SBPP30qDVmL

.

I found it interesting in today’s presser that London apparently currently has the lowest R0. Could this mean that the immunity in the population is higher than expected?

Possibly... My hypothesis is that there may be pockets of higher infection rates within London. Those that are key workers, and mix with other key workers will probably have been more infected over time, so now as a loose “social group” they are less likely to infect each other as they will have developed a non-negligible level of “group” immunity (20-30% maybe), driving down R0. However, equally, there will be very many Londoners who will have isolated and will have barely mixed since before lockdown. They will lower the average infection rate across the city perhaps to the 10% that is reported.

@BigChocFrenzy

Thank you. Interesting analysis.

One salient factor that doesn’t look like it’s been considered is the relative location of white and black males once adjusted for socio-economic characteristics. My understanding is that a less affluent white male will typically be more likely to live in post-industrial and seaside towns, as opposed to a less affluent black male who will be more likely to live in major cities and metropolitan conurbations .... the same major cities where there have been more significant Covid outbreaks. I’m not sure how easy it would be to adjust for this, but instinctively I feels like it could be an important factor, thus reducing the ethnic bias.

Same applies to females obviously...

Isn't R0 calculated from the number of new cases over a certain number of days ?

It doesn't measure "immunity" as such

A hospital consultant friend of mine shared this: