Daily numbers, graphs, analysis thread 11

[BigChocFrenzy]

Welcome to thread 11 of the daily updates

Resource links:

Slides & data UK govt pressers
NHS England stats including breakdown by Hospital Trust
ONS UK statistics for CV related deaths outside hospitals, released weekly each Tuesday
Financial Times Daily updates and graphs
HSJ Coronavirus updates
Worldometer UK page
Covidly.com to filter graphs using selected data filters ONS statistics for CV related deaths outside hospitals, released weekly each Tuesday
Plot COVID Graphs Our World in Data

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

Stats here for Italy - en.wikipedia.org/wiki/COVID-19_pandemic_in_Italy

Appears to show confirmed cases around 1,000 when they started to lift lockdown in early May, and a pretty steady trend downwards since, as well as a steady trend downwards in ICU cases. Local fluctuations of course - more than half of new infections are in Lombardy.

For Spain - en.wikipedia.org/wiki/COVID-19_pandemic_in_Spain

New cases seem to have been averaging around 180+ in early June as lockdown eased, and have dropped a little. Less data here though, no data on hospitals.

In France new cases seem to have been around 300/400 a day as lockdown started to ease and are still much the same. Overall hospitalisations are steadily declining.
geodes.santepubliquefrance.fr/#bbox=-1296621,5750062,3084442,1848309&c=indicator&f=0&i=covid_hospit.hosp&s=2020-06-26&t=a01&view=map2

Caveats to all of this is that I don't know how our testing regimes compare, so not sure if tests are really comparable, which is why hospitalisations are often more interesting .

Thanks Patricia, so probably most similar to Italy then...which hasn’t been disastrous - I’m looking for reasons to feel optimistic!

If I missed this, sorry, but can anyone cast some thoughts on USA with I think a much higher cases per 100,000 people and much lower deaths per 100,000 than the UK?

I am viewing data from https://vac-lshtm.shinyapps.io/ncovtracker/??_ga=2.131654671.1063228585.1593329295-775358824.1586505214 ( a bit slow but I understand it sjohn Hopkins data );

Reports of a possible local lockdown in the Leicester area due to rising numbers today www.thetimes.co.uk/edition/news/local-lockdown-fears-after-covid-spikes-in-leicester-zkg0s2q37

@midgebabe I haven't really looked at the US data in detail but there are vast numbers of journalists who say that Trump is presiding over a genocide or words to that effect.

This might be the case but it hasn't happened yet in that the badly hit areas (around NYC) brought their deaths right down, while despite claims that cases are exploding elsewhere, state death totals are stable or down at low levels

The last article I read said 'deaths lag cases by a month', which I think is an overstatement, but the idea is that this disaster will unfold real soon now.

Probably you'd need to look at ICU occupancy stats on a state level (i.e. trends) to see if there are lots of deaths upcoming.

Also it's going to be totally state dependent. Alaska is not going to have a pandemic , for example.

So you'd want to look at individual states, look at test rates per capita, who was tested (random Vs. Medical staff vs. sick people) and make conclusions from that. For example for the UK most of our infections occurred in March, so our CFR will be total bollocks because we weren't testing then.

Whereas the US is living with an ongoing level of infections that should be far below our March peak , but its happening continuously and they are testing 500k daily so this will be creating a lot of cases, whereas if we test 125k this won't because our infection rate is now very very small.

But anyway, it's possible they have many deaths to come from these recent infections whereas in the UK we can't have many deaths from current infections because they have been low for weeks.

That is if we've had 6 million infections and 60k deaths, then the number of 'pending' deaths is very small, since the current number of infections is just a tiny fraction of that 6 million.

Whereas it's possible that the number of people currently infected in the US is a more.significant percentage of the total.

@midgebabe I think the most likely explanation for UK cases being lower than US ones relative to deaths is that the outbreak in the UK came earlier when testing capacity wasn't there. In the time when we've been testing everyone with symptoms we've had much lower numbers.

I think there are a number of things going on with the US data -

• bad/under/lack of reporting - different states do different things, so it really needs to be treated at a state level. Some states make that much harder by not releasing the data. Testing approaches and numbers can differ widely.
• different approaches to testing and if they started proper testing earlier in the pandemic than we did - if they did, they will have identified a lot more of the mild/asymptomatic cases than we did
• Different states are at different points in the cycle. North eastern states are more like the UK, whereas the southern states are much earlier in their progression, so we really won't know what's going on there for a few weeks in terms of progression of infection to death
• different patterns of demographic infection. New Jersey, for example, was like the UK in that around half their deaths are from care homes. Infection patterns in other states seem to reflect more that the states stayed open, or have reopened, so more infections are in the younger population, which may lead to fewer deaths in relation to infections. Florida is now saying the median age of those infected is in their 30s, down from 60s in March. If states can keep it out of care homes, that will make a huge difference.
• better treatments now that perhaps any of us had 3-4 months ago is likely to help, at least somewhat.

so lots of things really! It's much like looking a dozens of different countries really, rather than one.

I think the cases in the US have increased very rapidly, I think in states such as Florida they are almost doubling every three days at the moment. Hugely inconsistent rules/compliance with lockdown across states, exacerbated by the fact that the ‘sun belt’ - currently worst affected, is where everyone goes on holiday and no one is restricting this. They also have the ‘opposite summer effect’ ie it’s so hot people retreat indoors to A/C environments where we know virus can spread rapidly. I think that the death figures will soon catch up

Interesting theory put forward by Prof Francois Balloux:
I suspect it is far safer to get infected by an asymptomatic carrier, and ideally outdoors, as this is expected to lead to a lower infectious dose, which may a key determinant of symptom severity. It might be viewed as a form of 'variolation' (see below)
en.wikipedia.org/wiki/Variolati…“

The Florida pictures seems quite confusing in that on the one hand they are saying 13.5% positive , which is very high, but on the other it says that most tests are done by private labs and the number isn't counted , so I am not sure if the 13.5% number means anything.

www.clickorlando.com/news/local/2020/06/26/florida-reports-nearly-9000-new-cases-of-covid-19-shattering-single-day-record/

It looks like hospitalisations in South Florida have exploded

www.sun-sentinel.com/coronavirus/fl-ne-coronavirus-hidden-health-data-20200625-fxaqlfnlrbgm3ouzrcm55yhaxe-story.html

You would expect these recent hospitalisations to result in a death rate perhaps double the current rate in a week or two.

So say 500 weekly rather than 250 as now.

For comparison the UK is about 3x the population and had more than 1500 covid-19 deaths in the last week but one and less this week.

So it's not yet at our levels of death when we peaked at 10k+ in a week but who knows? I suppose it's a model of what happens if government takes little action, add in lots of holiday makers, air con, but take off the relatively more outdoors/drive my truck rather than take a bus nature of America ,it won't be comparable.

Thanks

We should treat Florida stats with caution from this coming week onwards

Can't tell whether "cooking the books" would happen now, even if it was indeed planned:

www.theguardian.com/us-news/2020/jun/26/florida-governor-ron-desantis-republican-coronavirus

Rebekah Jones, who says she was fired from her job in charge of the state’s official Covid-19 database in May for refusing to manipulate its figures
, claimed on social media to have evidence that
employees at Florida’s department of health “have been instructed this week to change the numbers
and begin slowly deleting deaths and cases so it looks like Florida is improving next week in the lead-up to July 4, like they’ve ‘made it over the hump’.”

“They’re only reporting all these cases now so they can restrict reporting next week to make everyone think it’s over"

Lower figures today.

901 cases

36 deaths.

Anyone have last Sunday's figures to compare?

Seems quite low numbers still considering they have declared an outbreak in Leicester.

That means cases outside of this centre should actually be even lower due to health even spread?

Last Sunday - 1,221 cases, deaths 43.

I don't think you can tell anything from the cases; the Leicester outbreak has been spread over the last few weeks so cases will have appeared in each day.

Good that both cases and deaths are down on past Sunday's figures.

And cases below 1000 again which is great.

I keep tentatively feeling positive about the downward trends.

@BigChocFrenzy wow re the Florida whistle blower..... please can someone ask Trump about this on National TV?!

Any knowledgeable about test result categories and the swabbing process?

Have just spotted one thread where several people have received an UNCLEAR result, asked to retest:
Unclear Result...can't order another test? www.mumsnet.com/Talk/coronavirus/3951714-unclear-result-can-t-order-another-test

But another thread (on difficulties of testing toddlers) says it's only possible to have COVID detected' or 'COVID not detected' as a result.

Which is correct, only one can be?!

I've been concerned for ages about conflicting versions I've read, as to how far up your nostril you have to stick the swab to be accurate, and whether that's possible for an untrained person to do. If the latter, how can home kits, or self swabbing at test centres, work??

Here's the toddler thread which contradicts the first thread re possible test results
Tests for toddlers www.mumsnet.com/Talk/coronavirus/3952001-Tests-for-toddlers

@BigChocFrenzy

We should treat Florida stats with caution from this coming week onwards

Can't tell whether "cooking the books" would happen now, even if it was indeed planned:

www.theguardian.com/us-news/2020/jun/26/florida-governor-ron-desantis-republican-coronavirus

Rebekah Jones, who says she was fired from her job in charge of the state’s official Covid-19 database in May for refusing to manipulate its figures
, claimed on social media to have evidence that
employees at Florida’s department of health “have been instructed this week to change the numbers
and begin slowly deleting deaths and cases so it looks like Florida is improving next week in the lead-up to July 4, like they’ve ‘made it over the hump’.”

“They’re only reporting all these cases now so they can restrict reporting next week to make everyone think it’s over"

So, is it this week's numbers that are fudged for Florida, or will it be next week's?

the Leicester outbreak has been spread over the last few weeks so cases will have appeared in each day

Thanks for making some sense of this, Patricia - elsewhere it's being made to sound like a sudden and colossal increase

I'm aware these threads have more informed folk on them, so please does anyone know if the current, daily new cases are available anywhere for Leicester? I've tried to find this, but all I'm getting is media hysteria ...

newjez She says the last week's figures were correct, to give a high point
but that the Florida governor wants to make it look like cases are falling, before the Independence Day celebrations next weekend

So she said figures up to now were correct, but that this coming week's figures would be altered downwards

However, imo her going public probably stopped the fiddling if it was indeed planned - they'd be idiots to continue now.

@puzzled - the link that @Derby posted upthread for me gives a website where you can download a CSV file with daily cases (just pillar 2 I think) by local authority. It's not very user friendly though - you need to create a pivot table to see the numbers in a sensible type of table. Numbers for Leicester are high, but no more so that a few other authorities. There look to be about a dozen with over 100 cases in the last fortnight.

Puzzled The data for Leicester I can find is not in a satisfactory form

Scroll down and put Leicester in search field for graph

www.theguardian.com/world/2020/jun/28/coronavirus-uk-are-cases-rising-or-falling-near-you

Uk dashboard gives total cases 1,052 for Leicester:

I wonder if the lack of recent figures for Leicester is due to delay accessing data from commercial test providers ?

news.sky.com/story/coronavirus-leicester-could-be-first-city-to-go-into-local-lockdown-amid-spike-in-covid-19-cases-12016655

There have been reports that the Leicester area had 658 new cases in the two weeks to 16 June.

Public Health England, however, says there have been 50 cases reported in the city in the 10 days to 26 June,
< which does not look like a spike >
taking the total to 1,046 since the start of the pandemic,

but that figure does not include the results of tests which are carried out by what PHE calls commercial partners.