Meet the Other Phone. Only the apps you allow.

Meet the Other Phone.
Only the apps you allow.

Buy now

Please or to access all these features

Covid

Mumsnet doesn't verify the qualifications of users. If you have medical concerns, please consult a healthcare professional.

Daily numbers, graphs, analysis thread 10

966 replies

BigChocFrenzy · 08/06/2020 19:35

Welcome to thread 10 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

We welcome factual, data driven, and civil discussions from all contributors 💐

OP posts:
Thread gallery
90
wintertravel1980 · 12/06/2020 16:49

The r rate still high according to the guardian.

The absolute R rate can be a red herring. It is only one of many data points that feeds into the lockdown lifting strategy. In fact, R rates may go up as prevalence (absolute level of infection) falls.

The ONS results from country wide testing released earlier today and discussed a few pages back are quite encouraging.

cathyandclare · 12/06/2020 16:50

The lack of UK trips for shopping and entertainment is unsurprising when few shops and entertainment venues are open. However the infection rates are more encouraging this week. Hopefully things will gradually pick up and the infection rates will stay down.

wintertravel1980 · 12/06/2020 16:52

To clarify, we do want the R rate to be under one on the sustained basis but random spikes (especially in the regions with low prevalence like South West) should be taken in context.

itsgettingweird · 12/06/2020 17:00

I think the super spreader thing has been seen elsewhere. Back during height of pandemic I read about how 1 person in China infected another 78 in a restaurant With 200 people.

Derbygerbil · 12/06/2020 17:14

To clarify, we do want the R rate to be under one on the sustained basis but random spikes (especially in the regions with low prevalence like South West) should be taken in context.

Even then the South West is 0.8 to 1.1, so still probably below 1.

Derbygerbil · 12/06/2020 17:16

I’m not sure why people are so fixated on 500k since doing absolutely nothing at all was never an option.

Tell that to the “it’s no worse than the flu” lot who want us to return everything back to normal straight away!

EugeniaGrace · 12/06/2020 17:20

All this focus on r-rate in the U.K. has to be kept in perspective.

The best way to keep it consistent is to keep infected numbers high but plateauing or declining slowly so clusters don’t skew the data so much. For example if there Is a borough of 200,000 people, and one week 120 test positive and the next week 100 test positive the r is below 1, but still lots of cases.

However, if numbers are low so only 4 people test positive in the borough one week and the next week they run a study on the hospital and 8 new asymptomatic cases are found, then r is above 1 but numbers are still much lower than in the first scenario.

R is important to check exponential growth but it is better to have low case numbers overall and lots of testing.

wintertravel1980 · 12/06/2020 17:32

Yes, R rate is another example when the government tried to simplify the message and it has backfired.

Life if complex. Decisions are usually nuanced. We cannot develop the whole lockdown lifting strategy based on one, two or three parameters. We should also be careful with overlying on behavioural science. "Simple, specific, persuasive, personalised" messaging does work but it may have long-term unintended consequences.

Derbygerbil · 12/06/2020 17:59

I’m not sure why people are so fixated on 500k since doing absolutely nothing at all was never an option.

People needed to know 500k figure in order to do something. The exponential nature of the growth meant they couldn’t wait for it to dawn on people. To use an analogy, you need to know you’re driving towards a cliff edge in order to slam on the brakes.

itsgettingweird · 12/06/2020 18:06

Derby and apparently it's not just flu. It's a MILD (yes in capitals) version of a flu.
I think these people cannot understand the concept of 65k excess deaths - outside of flu season

ShootsFruitAndLeaves · 12/06/2020 18:17

It seems to me that the R numbers are a load of bollocks since infection rates have been falling fast for weeks and weeks. Has anyone looked at the data underpinning them?

Sunshinegirl82 · 12/06/2020 18:34

There are 14 models that feed into SAGE’s assessment of R apparently.

I wish the government had never introduced it. If you tell everyone that it is imperative that R stays below one and then there are lots of reports saying it is above 1 in certain areas people will panic and assume beach goers/school children/sunbathers have caused it to rise and a second wave is upon us.

Obviously it’s much more complicated than that but that wasn’t the message.

wintertravel1980 · 12/06/2020 18:37

Has anyone looked at the data underpinning them?

Unfortunately, the underlying data is not publicly available (or I haven't been able to find it).

alreadytaken · 12/06/2020 18:53

The number of positive tests has jumped today and the number of hospital admissions went up in England for the 3rd day in a row and back up in Wales too.

This was always going to be a marathon, not a sprint.

Sunshinegirl82 · 12/06/2020 19:15

The positive test data and the number of hospital admissions jumps about all over the place on a day by day basis. I’m not sure you can take anything meaningful from daily figures, the 7 day rolling average is probably more helpful.

wintertravel1980 · 12/06/2020 19:25

Here is some of R modelling for school opening:

assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/888759/S0313_SAGE33_Consensus_for_SPI-M_on_transition_strategies.pdf

The results do not appear entirely consistent. Looks like running estimates for R numbers may be more art than science.

PatriciaHolm · 12/06/2020 19:25

@Sunshinegirl82

The positive test data and the number of hospital admissions jumps about all over the place on a day by day basis. I’m not sure you can take anything meaningful from daily figures, the 7 day rolling average is probably more helpful.
Indeed - see this for England hospital admissions. There is a definite drop at weekends, this has happened the past 5 weeks so the wave is as much down to that as anything else at the moment. Straighter line is 7 day rolling average, which continues down. Latest data is Tuesday, so I expect admission might be fairly flat in terms of announced data for the next couple of days if it follows trends as they will be the data that relates to Weds and Thurs.
Daily numbers, graphs, analysis thread 10
whatsnext2 · 12/06/2020 19:31

Apologies if this has been seen before:

wintoncentre.maths.cam.ac.uk/coronavirus/covid-excess/

Deaths by local area, against excess deaths in care home, hospital, home and other.

PatriciaHolm · 12/06/2020 19:33

@wintertravel1980

Here is some of R modelling for school opening:

assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/888759/S0313_SAGE33_Consensus_for_SPI-M_on_transition_strategies.pdf

The results do not appear entirely consistent. Looks like running estimates for R numbers may be more art than science.

Very much, especially as you start dealing with small numbers of infections. This is from the Govt page on how the various methods work - epidemiological data such as hospital admissions, ICU admissions and deaths – it generally takes 2 to 3 weeks for changes in R to be reflected in these data sources, due to the time between infection and needing hospital care -contact pattern surveys that gather information on behaviour – these can be quicker (with a lag of around a week) but can be open to bias as they often rely on self-reported behaviour
  • household infection surveys where swabs are performed on individuals which can provide estimates of how many people are infected – longitudinal surveys (which sample the same people repeatedly) allow a more direct estimate of the growth in infection rates

I think from next week the Govt won't be publishing regional modelled R, and be publishing growth rates based on data with fewer assumptions, whatever that means....probably from the infection surveys.

Laniakea · 12/06/2020 20:03

People needed to know 500k figure in order to do something.

^ people are stupid*, I agree. The discussion needs to move on now, the situation & available knowledge have changed & people need to catch up quickly.

(*I don't think that should be the driving force of policy but I'm sure it is & it's almost too depressing to contemplate.)

BigChocFrenzy · 12/06/2020 21:42

R0 is important when there are a large number of infections

  • going above 1.0 with a million people infected is a crisis, because of exponential growth

However, if there are only few infections, then small outbreaks can cause R0 to go temporarily > 1, without cause for concern.

In Germany over the last few weeks, R0 has oscillated quite a bit between 0.7 and 1, sometimes briefly above 1, but it has not caused public or official concern,
because:
the numbers involved in the calculation are small, confirmed new cases are only in the low hundreds,
deaths with care homes included are only in double or single digits

The 4-day R0 here oscillates more, but the 7-day R0 was developed to smooth out minor blips.

==> btw, anyone know how many days the UK uses to calculate R0 ?

OP posts:
whenwillthemadnessend · 12/06/2020 22:43

Joining regular reader of thread

chomalungma · 12/06/2020 23:07

Out of the 19,933 participants' swab tests included in this analysis, 11 individuals in 8 households tested positive for COVID-19. As this is a household survey, our figures do not include people staying in hospitals, care homes or other institutional settings. In these settings, rates of COVID-19 infection are likely to be different

Wow...that is a small number. They could just not have tested 1 of those households and got very different figures.

Especially prone to the effect of who you test and how random you make it.

More or Less would have a field day with that.

BigChocFrenzy · 12/06/2020 23:09

(FT Paywall) "One metre or two? Economic pressure stokes distancing debate"

Deciding the balance - and persuading the public - is the job of political leaders in each country

https://www.ft.com/content/202a6f8a-7310-49ed-b285-3ea26fa963f3

Scientists and politicians at odds over physical distancing needed to limit transmission
.....
the judgment is complicated because distancing is just one of many factors that influence infection rates

  • and because persisting with two metre spacing would hamper the revival of social and business activities.
...... “Duration and distance both matter” ..... being at close quarters to somebody for a short time may not cause problems outdoors, where the virus will tend to disperse quickly in the air.

Conversely, even maintaining 2 metres distancing over an extended stay indoors may not be enough to prevent infection.

A paper on South Korean coronavirus cases this year found that people in 50-minute fitness dance classes with as few as five people became infected,
even though the 60 square metres room was theoretically large enough to allow two metre distancing.

“We have seen that the two metre rule doesn’t really work in crowded indoor environments”

< imo, % infected is key.
If v few people are infected, there will only be localised outbreaks and few enough to shut down promptly before they spread (hence why I feel going to the gym is a reasonable risk) >

Daily numbers, graphs, analysis thread 10
OP posts:
BigChocFrenzy · 12/06/2020 23:12

Those PHE figures confirm that there is v v low infection in the wider community
So a couple of infections more or less don't matter wrt this important finding

The remaining infections are probably concentrated in care homes, hospitals, probably other places of work where people are close together indoors for extended periods

OP posts:
Please create an account

To comment on this thread you need to create a Mumsnet account.

This thread is closed and is no longer accepting replies. Click here to start a new thread.

Swipe left for the next trending thread