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 💐

ONS: Coronavirus Infection Survey Pilot, England, 25 June

With a v low infection rate, they can't test enough people to show clear trends

https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/england25june2020#covid-19-infection-survey-data

"When analysing data for the four most recent non-overlapping 14-day periods (Figure 1), these estimates suggest the percentage testing positive has clearly decreased over time since our first measurement on 26 April, and this downward trend has now flattened.

The confidence intervals overlap with the previous two time periods.

This suggests that the actual number of individuals testing positive in the period 8 June to 21 June could be higher or lower than in the two previous periods.

We therefore do not at this point have evidence that the current trend is anything other than flat."

With a v low infection rate, they can't test enough people to show clear trends

Indeed - their stats are based on 11 people testing positive out of 22,523 in 25 May-7 June, then 14 out of 24,256 in 8-21 June. Numbers so small that very few conclusions can be drawn about trends, just that numbers remain very small.

Cold air does seem to make it more easily spread. But maybe ski resprts due to the weather it stays in the environment.

So how bad winter is may play a part.
Uk is rainy so wash away frequently.
Playground equipment that isnt such a problem in jun may become worse in cold weather (ours is metal).

When does the ski season start?
I imagine uk people going to europe brought covid back before feb half term.
If those people have antibodies we have a date for when
Uk actually had cases.

The fact it's often the tourists getting infected then going home i imagine slows the outbreak at source. But i guess if you go for a week unfortunately that is time enough to start shedding it there (5d incubation).

Sutely ski slopes will not be open this year...

It's a huge industry. It'll be open. Starts end November.

Anecdotally, one of the London hospitals has started to seek,parental permission to test newly diagnosed children with type 1 diabetes for COVID antibodies. I have a type 1 child and belong to various groups on social media. It is considered that type 1s are born with a predisposition to developing the condition and that it is triggered by a virus. It has been clear in the groups that there have been more new members than usual in recent months. It will be interesting to see how this pans out over time.

On the issue of lung damage in asymptomatic Sars-Cov 2/COVID-19 cases - this comes from a recent Nature Medicine paper from China. The group looked at pcr-positive cases who had been tested (and isolated) as a result of being in contact with a known case. They found that a proportion of them, despite feeling completely well, had lungs with the ground- glass appearance when visualized by ct scan. This doesn’t mean that they necessarily have permanently damaged lungs, just that they did have this effect without any other obvious symptoms. It remains to be seen what the long term effects, if any, will be.
They also looked at how long the viral RNA was detectable for and interestingly they could detect it for longer in the asymptomatic group - annoyingly they called it viral shedding, but they did not actually look for replicative virus.
Re. antibodies- the same paper looked at antibody levels and while both asymptomatic and symptomatic cases (mild cases) produced antibodies, symptomatic cases produced higher levels and these antibodies were detected for longer than the asymptomatic group.
In terms of what is normal for an immune response to a typical virus- it can be difficult to generalise (my background is in viruses/immune response/vaccines)- but many viruses (like measles) induce antibody responses that can last for years/lifetime, and others/other coronaviruses the antibody response is much shorter- less than a year. But while there may not be circulating antibodies there could well be B-cells that can start producing those antibodies again upon reinfection. This would presumably mean the second infection would be less severe. This is quite likely the mechanism for the other ‘common cold’ coronaviruses.
Re T-cell responses, these are a prerequisite for B-cell responses, and also possibly have a direct antiviral effect ( there are several different types of T cells). Measuring T cell responses is much more complex than looking for antibodies and it’s very unlikely that a population wide T cell test could be rolled out.
There is also the innate response which kicks everything off, this is non-specific and provides some protection before the specific T and B cell response ramps up- but I’ll stop there!

annoyingly they called it viral shedding

Annoying because it's inaccurate?

I guess these kinds of terms of science might be subject to mistranslation (from Mandarin, I'm guessing?)

What would have been better?

(Viral shedding is a recent addition to my vocabulary.)

Thanks for your explanation above.

Oh also, I wondered, does this scarring count as a symptom and move those in whom it is noticed from the asymptomatic category to the symptomatic category?

Or is this different from a symptom because it is unnoticed until a scan (or for some other reason)?

Sorry for so many questions and at such a basic level.

I don't understand why everyone blames ski resorts. The worst affected towns and and city's in Italy are a two hour drive from the resorts.

Most likely tourists picked it up on the planes and airports in those big towns. Milan Bologna Verona and brought it back to other country's.

Sparrow biology is a gap in my education, it's fascinating thank you!

People hadn’t thought that air con systems might be a problem? Gah.
Re superspreaders. No one will know if they are a superspreader, but common hand hygiene and social distancing/mask wearing would surely vastly reduce the superspreading impact, wouldn’t it? And that alone would have a significant effect on infection rates.

The issue of viral shedding is important because it normally is used to mean they are shedding infectious material

However, if that is not the case, maybe "detectable viral RNA" might be better?

Might not be a bad translation; more likely the authors with ESL using the wrong phrase.
I've seen this in scientific papers in my field - and if I'd had to write papers in German I'd probably use the occasional wrong word too.

Asymptomatic people would rarely be given chest X rays atm ... unless they are part of a study !
So I suspect asymptomatic with ground glass lungs rarely know about it

I hope they are monitored to check how long it lasts and if it has any effects

I wondered if they asked those people to do fitness tests - some people never e.g. run up stairs, so they might not notice any reduction in performance that say a more sporty person would

People with damage to organs e.g. liver caused by other ailments have in the past not noticed until the disease became advanced & serious

We don’t have good data for how long infectious virus is shed for because those tests/ experiments require BL-3 conditions. But it’s an important question. The reviewers of the Nature Medicine paper should have picked up that inaccuracy. It would have been correct to say ‘viral RNA was detected’ and leave out the word ‘shedding’.
Technically a symptom is something the patient feels vs a clinical sign that is observed so it would be correct to still call those cases asymptomatic.

A question to the number crunchers on here.... is America as bad as it looks?

What is looking bad in the US is the rate at which new cases are arising. The daily count of new case is increasing which presumably will lead to a spike in deaths further down the line. You can see the daily cases graph on this page. [https://www.worldometers.info/coronavirus/country/us/] (And the explanation does not appear to be that they are doing more tests).

Link fail www.worldometers.info/coronavirus/country/us/

The situation in the US is very varied by state, with for example the North east seeing steady declines, but it's worrying that a number of states are not only seeing increased cases (as a percentage of tests, so indicating actual growth not just more testing) but also increased hospitalisation, so it's not a case of just identifying more mild cases; this graph shows how, for example, Arizona and Texas are seeing growth in hospitalisation, with several other states coming up too.

More positives but less hospitalisation wouldn't be so worrying, (and the Trump defence is much about more testing = more identified but not worrying cases) but that's not what's happening. Deaths are also starting to trend up.

No new deaths in Scotland today. First time that this has happened at the end of the week.

This is an interesting article about death rates in hospital having fallen. BBC News - Coronavirus death rate falling in hospitals
www.bbc.co.uk/news/health-53192532

You can click though to the actual article from the link here. They offer several possible explanations, though then state that the data aren't really explained by most of them. The most likely looking to me (and one that does seem to fit with the data) is that hospitals are no longer under such pressure so are admitting people at a lower threshold of illness. Which means that rates of infection may in fact be slowing slightly more than it appears from the hospital admission rates.

It could also be that at the most vulnerable are now being the most cautious. So the cases we are seeing in hospital are more likely to be less vulnerable people with a higher chance of survival.

That combined with less pressure on the system could well account for the fall.

I agree more vit d
Less likely other infections
Not existing hospital patients
Less virus about so smaller initial dose - from 2m away.... Vs march people at 0.
If they caught it from family we are not locked down and it's sunny so not hours and days inside with them
Bus drivers etc masks on them and customers and not as full
Over 70s have been in as have all people vulnerable
People generally arent as ill in summer as they arent cold.

Re ski resorts it's because it was all the returning school trips - would they have all been off round milan etc?
But also wasnt there an austrian ski town or something where a lot of infections came from.
It could also be that the mountains are where a lot of countries borders are

The areas in italy also are where there is high air pollution and - could be wrong, but covid didnt seem to have spread as much round the rest of italy even as it did to other eu capitals?

Yes that's true @Tom. The shielded people and anxious pensioners who've not set foot outside the house have been less likely to catch it since lockdown Though they discuss that possibly and say it doesn't quite fit because the average age of patients who die has in fact gone up, not down.

This suggests that the actual number of individuals testing positive in the period 8 June to 21 June could be higher or lower than in the two previous periods.

This is a bit shit:

27 - 10 May 33/11272
11-24 May 34/19314
25-7 June 11/22523
8-21 June 14/24256

So the actual numbers testing positive are tiny ,and the positive tests are 14 rather than say 12 as we'd expect from the previous period.

But as in Germany if you happened upon a highly infected area that would blow up your figures without necessarily meaning an overall increase......

I have just seen this any thoughts

The Office for National Statistics (ONS) said that the big fall in cases since April appeared to have come to an end, with new infections steady at more than 3,000 a day.

Separate figures published by the government concluded that cases were still falling at between 2 per cent and 4 per cent a day, underlining the uncertainty about the pandemic. Experts said that infections could now be declining in hospitals and care homes but remaining flat in ordinary homes and shops.

from www.thetimes.co.uk/edition/news/two-people-die-at-beaches-on-hottest-day-of-year-qtz8tkq03

I think the ONS needs to test bigger numbers- it's difficult to read anything into a sample in which only 10-15 people test positive out of 20k+, just one person can skew the results. I'm also becoming cynical about the Zoe figures following the latest article and the fact that they had been counting Ab tests as active cases. Tricky to get a real feel for what's happening with case numbers apart from admissions!