Deterioration in R number

[underbank]

If a second lockdown happens, it would destroy the economy and with no economy you have no NHS...

The NHS was virtually destroyed back in March where some hospitals were reporting 40% of their staff were either off sick or self isolating due to symptoms. You can't run any kind oh health service with no staff! That's why the NHS closed its doors to all but covid patients. If infections rise again out of control, like back in Feb/Mar, the same will happen. Hence a lockdown WILL be needed again if covid infections look to be heading back to exponential growth.

 

[Bletchleyite]

The NHS was virtually destroyed back in March where some hospitals were reporting 40% of their staff were either off sick or self isolating due to symptoms. You can't run any kind oh health service with no staff! That's why the NHS closed its doors to all but covid patients. If infections rise again out of control, like back in Feb/Mar, the same will happen. Hence a lockdown WILL be needed again if covid infections look to be heading back to exponential growth.

Part of the issue then was the lack of testing - you pretty much had to assume that everyone with a minor case of the sniffles had COVID as you had no way to know otherwise. If we can keep testing capacity high so everyone with symptoms can be tested, this is much reduced. Though we will have to look at ways to reduce the 14-day contact isolation, which is difficult because tests aren't reliable on asymptomatic people, I read. Perhaps test at day 7 to get an idea?

 

[Sherlock49]

I am hopeful that everything will continue to go in the right direction - but I think it will be through sheer luck rather than science if we do avoid an increase in the next few weeks / months. I'm also dubious about how effective local lockdowns will be when the government has essentially sanctioned anyone with children travelling elsewhere in the country for childcare.

Part of the issue then was the lack of testing - you pretty much had to assume that everyone with a minor case of the sniffles had COVID as you had no way to know otherwise. If we can keep testing capacity high so everyone with symptoms can be tested, this is much reduced. Though we will have to look at ways to reduce the 14-day contact isolation, which is difficult because tests aren't reliable on asymptomatic people, I read. Perhaps test at day 7 to get an idea?

Testing is a lot easier if you can get to a testing site yourself but I was quite surprised how long it took to get us a test result recently with a home test kit - Day 1 symptoms appeared, it took till Day 3 for home testing kits to be available. Dispatched and delivered (by Amazon) on Day 4. Because it arrived after 4pm, collection couldn't be booked for the following day so it wasn't collected till Day 6. Results were supposed to be within 4 days but actually only came through 6 days later on Day 12. Negative result, but it didn't actually help anything as by the time it had arrived, the 14 day period was practically up anyway. For test and trace this would have been essentially useless.

 

[Bletchleyite]

Testing is a lot easier if you can get to a testing site yourself but I was quite surprised how long it took to get us a test result recently with a home test kit - Day 1 symptoms appeared, it took till Day 3 for home testing kits to be available. Dispatched and delivered (by Amazon) on Day 4. Because it arrived after 4pm, collection couldn't be booked for the following day so it wasn't collected till Day 6. Results were supposed to be within 4 days but actually only came through 6 days later on Day 12. Negative result, but it didn't actually help anything as by the time it had arrived, the 14 day period was practically up anyway. For test and trace this would have been essentially useless.

What I don't get is why they are setting up test sites at all when home kits would be much easier to do. If you go to a site you swab yourself anyway. So why not just get them posted out, then just chuck them in a postbox first class to return? (It is legal to send properly packed infectious disease specimens that way, and indeed that's what all the private testing labs are doing).

 

[northernchris]

Even if the R number is above 1, it doesn't matter if most people who contract it only experience mild symptoms. You also need to pay attention to the number of people in hospital with COVID 19, and the number of people who are seriously ill and require a ventilator.

This is true, and the stats are showing that both hospital inpatients and those on mechanical ventilators are decreasing each week

Strange isn't it? For months we have focused on the number of people dying whilst testing positive for the virus. Now that is going down, suddenly the R0 value is the important statistic.

The media do tend to latch on to the negatives, which in turn seems to fuel the more gullible in to thinking we're doomed and should remain in lockdown indefinitely.

The fall in the death rate has wobbled a bit over the last couple of weeks and may be slowing (as you'd expect), although it could be partly a consequence of late reporting over bank holidays.

This is the only part I find concerning. There was a decent enough drop in the death rate until a couple of weeks ago, and since then its been hovering around 300-350 most weekdays. The number of new daily hospital admissions is also broadly similar and showing no decent level of decline

 

[Yew]

It's worth checking that we're comparing like with like here. My understanding is that there is not only empirical data, but a various number of models. Given that the famous model from Professor Ferguson vastly overestimated the deaths that Sweeden would see with it's less invasive restrictions, and we continue to see a drop in infections and deaths. I'm going to take a large pinch of salt with this article.

 

[DynamicSpirit]

The problem is the second the lockdown is released... we just get hit by the second wave.
There is no easy way out of this, and a harder lockdown does not actually get us anything unless you can somehow catch every single case - which we can't.

It kinda does, because if you can use a short but hard lockdown to get the numbers down to very low levels, then you can start using track-and-trace to keep the R-number below one, which therefore keeps the virus under control even with the lockdown being relaxed. But it's not feasible to use track-and-trace unless you've already got the virus down to low levels. (Of course, in the UK, it seems that we haven't yet got track-and-trace running properly anyway, but that's another story)

 

[yorkie]

The article from the BBC is interesting.

Coronavirus: Is R number balanced on knife edge?

There are concerns coronavirus might be starting to spread in parts of England again.

www.bbc.co.uk

Indeed; also see:

https://twitter.com/ProfKarolSikora/status/1268904884511952896

The German scientists who have been estimating their R value for months always stress as cases decrease local outbreaks have a bigger impact on the value causing fluctuations. A few days ago their R value was 1.2, yesterday it was 0.57. The infection number is so crucial now.

Also, we have more testing now!


People being asymptomatic is an issue.

Remember the fake news about Weston hospital? People falsely claimed it was caused by outbreaks in the community; the reality is...

Coronavirus: Weston hospital staff 'unknowingly spread virus'

Weston General Hospital temporarily closed in May due to an outbreak of coronavirus.

www.bbc.co.uk

He said: "Some staff who did not know they had the virus were wandering around without any symptoms and without realising it, were spreading it."


"It wasn't until they [staff who had the virus but were showing no symptoms] were found in a routine test everyone went, 'hang on there's a problem here'.

"The hospital realised that actually it was in danger of starting an outbreak rather than being the answer to solving one.

There had been speculation the outbreak was linked to crowds of people visiting the seaside town.

North Somerset Council also said there was no current evidence to show the virus had spread to the wider community from the hospital outbreak.

 

[kieron]

The "hammer" was the lockdown, both to get R down and to get the number of daily deaths/cases within what the NHS can cope with on an ongoing, long-term (potentially months or even years) basis.

There's another thread on here where people are talking about how the NHS is not coping with the level of infection the UK is experiencing at the moment. If a doctor recommends that someone should have chemotherapy, the NHS should be able to provide it for them, and in a safe environment. If there are too many seriously ill patients, and too few healthy staff, to ensure the the continued functioning of the health system in all relevant aspects, then this isn't a situation the NHS can cope with.

I think it would be a mistake to read an overall strategy into the government's actions.

 

[Puffing Devil]

As for a second lockdown, not a chance, we cannot afford it.

If a second lockdown happens, it would destroy the economy and with no economy you have no NHS...

If we are facing a second wave, we have a choice:

1) Lockdown, save lives.
2) Continue as is, more people die.

Which of your friends and relatives would you be happy to sacrifice for option 2?

 

[Sherlock49]

The media do tend to latch on to the negatives, which in turn seems to fuel the more gullible in to thinking we're doomed and should remain in lockdown indefinitely.

There's certainly some stories which are probably more negative than they need to be, but there's also plenty of 'positive' front pages about lockdown restrictions easing, pubs reopening and so on. The key problem you have is that we have one of the highest per capita rates of death (along with Sweden). Excess deaths are projected at over 60,000 - some of these will be because of medical needs which were unmet by the NHS and unrelated to Coronavirus but that's still something the government need to take responsibility for. The strategy from the government, whilst initially sound has become confused and unclear. The scientific messaging has got significantly worse over the last month (and as far as I can tell it's pure incompetence - the number of times I've had to explain why our death figures are in the 300's whilst Spain is reporting none is ridiculous - why have the government not made this clear already).

The government's job is to develop a clear strategy for dealing with the virus, execute it well and provide clear public communications. The media should be holding the government to account. This means reporting on failures with test and trace, hypocrisy from government ministers, death tolls being the highest in Europe amongst many other things. Whilst there are positive signs for the future, overall the handling of this by the government has been appalling and you can't spin that into a positive.

I remain cautious - I'm aware of the risks to me, if I catch symptomatic Covid, because of my age and health conditions there's around a 1% chance of death. 1 in 100 is not something I feel particularly comfortable with, so I do what I can to minimise my exposure. I don't think that's a particularly unreasonable approach to take. Those who are fit, young and healthy will and should assess the risks differently. Weeks ago, a story was put out by the media (citing unnamed sources) that they could release the lockdown significantly for younger people earlier and the more I think about it. The reaction wasn't great but the more I think about it, the more I agree that would have been the right approach to keep the economy moving whilst protecting public health.

On topic though, I don't think these figures really give much away. Everyone is in a bit of limbo for the next 3-4 weeks, whilst we see what effect (if any - positive or negative) results from the easing of the lockdown last week, and the further easing next week.

 

[chris11256]

I think the point going forward is that any future lockdowns would be heavily localised rather than a blanket nationwide lockdown. The last thing we need is to panic ourselves into complete lockdown over a single research paper. As the Germans have been experiencing, localised outbreaks of cases can have a massive impact on the R number giving the impression that it's worse than it actually is.

I suspect that as new cases get lower nationally, the R number gradually becomes less reliable unless you're able to get highly localised data.

 

[Bletchleyite]

People being asymptomatic is an issue.

It is, particularly in care and health settings.

It seems that at least some asymptomatic people have tested positive, though it can't be guaranteed that they will (so it's a bit like a coeliac test - positive means positive, negative tells you very little).

So it strikes me that we should be aiming to test people routinely in more and more settings, expanding this as quickly as feasible. Perhaps start with testing every customer-facing NHS and care-home member of staff weekly, and step it up from there to aim to daily. It seems it's asymptomatic and presymptomatic carriers who are the big issue, and if we can weed some of them out and get them isolated, great.

 

[Bletchleyite]

If we are facing a second wave, we have a choice:

1) Lockdown, save lives.
2) Continue as is, more people die.

Which of your friends and relatives would you be happy to sacrifice for option 2?

Not that simple. A second long lockdown would collapse the economy entirely (it is already teetering) - and we would cease to be able to afford to fund universal healthcare as a result. That could cause a number of deaths, over time, which would make 40K sound low, not to mention suicides and starvation.

 

[Qwerty133]

What I don't get is why they are setting up test sites at all when home kits would be much easier to do. If you go to a site you swab yourself anyway. So why not just get them posted out, then just chuck them in a postbox first class to return? (It is legal to send properly packed infectious disease specimens that way, and indeed that's what all the private testing labs are doing).

I would go to the other extreme, and only offer home kits to those who cannot get to a testing centre. Having the person go to a testing centre should mean they can be tested and the sample sent of to the lab within a matter of hours of symptoms appearing whereas home testing will inevitably take at least a couple of days to get the sample on its way to the lab.

 

[Qwerty133]

If we are facing a second wave, we have a choice:

1) Lockdown, save lives.
2) Continue as is, more people die.

Which of your friends and relatives would you be happy to sacrifice for option 2?

The choice is actually:
1) Lockdown, reduce covid related deaths, be unable to fund the NHS, cause millions of other preventable deaths
2) Continue as is, allow (mainly) elderly people to die of Covid, keep the NHS funded, prevent other preventable deaths

 

[Puffing Devil]

Not that simple. A second long lockdown would collapse the economy entirely (it is already teetering) - and we would cease to be able to afford to fund universal healthcare as a result. That could cause a number of deaths, over time, which would make 40K sound low, not to mention suicides and starvation.

Complete economic collapse? I doubt it. Do you have anything to support your forecast?

The choice is actually:
1) Lockdown, reduce covid related deaths, be unable to fund the NHS, cause millions of other preventable deaths
2) Continue as is, allow (mainly) elderly people to die of Covid, keep the NHS funded, prevent other preventable deaths

Millions of preventable deaths? You can, of course, substantiate that claim?

As others have already said, we can't operate an NHS with staff that are sick and facilities that are overwhelmed with Covid cases. Funding is secondary.

As for option 2, which of your friends and family are you prepared to sacrifice?

 

[Bantamzen]

If we are facing a second wave, we have a choice:

1) Lockdown, save lives.
2) Continue as is, more people die.

Which of your friends and relatives would you be happy to sacrifice for option 2?

Or perhaps

1. Carry on with lockdown, which is currently costing an estimated £100 billion per month and before long with such costs and a loss of tax revenue will have to result in deep cutbacks in all public services, including the NHS.
2. Start to move back to normality so that we can at least start to tackle the impending recession (expected to be the worst in 3 centuries) and make sure some revenue gets back to the essential services.

Now I know which one I would choose, so which one would you?

 

[Bletchleyite]

As for option 2, which of your friends and family are you prepared to sacrifice?

As you well know, we do not have that choice, so it is unhelpful and overemotive to bring it to the table.

 

[HSTEd]

It kinda does, because if you can use a short but hard lockdown to get the numbers down to very low levels, then you can start using track-and-trace to keep the R-number below one, which therefore keeps the virus under control even with the lockdown being relaxed. But it's not feasible to use track-and-trace unless you've already got the virus down to low levels. (Of course, in the UK, it seems that we haven't yet got track-and-trace running properly anyway, but that's another story)

That assumes track and trace is actually capable of permitting truly meaningful lockdown relaxation.
I am very doubtful that it is, in a virus where two thirds of cases are asymptomatic.

Millions of preventable deaths? You can, of course, substantiate that claim?

Ten thousand extra dead from dementia in a month?

28,000 cancelled cardiac operations.
Unless all those operations are completely pointless then the number of dead and loss QALY is mounting rapidly.

Or the huge drop in stroke, heart attacks and cancer patients presenting to the NHS.

As others have already said, we can't operate an NHS with staff that are sick and facilities that are overwhelmed with Covid cases. Funding is secondary.

Triage exists.

Ultimately the hospital can deny people treatment and discharge them if it does not have the facilities to treat them.
As for ICU facilities, given that the survival rate for ICU is absolutely appalling, a complete collapse in COVID ICU care is not going to substantially increase the death rate.

As for option 2, which of your friends and family are you prepared to sacrifice?

Thing about pandemics is that we don't chose.
The virus does.

 

[Starmill]

This has caused huge consternation in Greater Manchester.

 

[HSTEd]

The fall in the death rate has wobbled a bit over the last couple of weeks and may be slowing (as you'd expect), although it could be partly a consequence of late reporting over bank holidays.

Many countries have their case rate down to a few hundred a day, something that "track and trace" plus strict isolation of care homes etc can probably keep on top of.

The problem is that "track and trace" is just an article of faith for people who believe in the lockdown.
They believe it will work, not because there is much evidence of this, but because they need it to work for their preferred strategy to make any sense at all.

A lot of studies dont share this optimism.

And isolating care homes is apparently killing substantial numbers of people anyway.

 

[Mugby]

I'm not sure why so much emphasis is placed on the 'R' rate, it can hardly be described as reliable or even accurate.

As for a second lockdown, not a chance, we cannot afford it.

If a second lockdown happens, it would destroy the economy and with no economy you have no NHS...

I agree. The billions we've spent already will take years, if not decades, to recoup.

Frankly, I don't get the thing about a second wave, so many people have convinced themselves a second wave is on it's way - almost willing it to happen. I'm not saying there won't be, I'm saying I don't know, but so many people seem to be experts in viral science, suddenly.
If the second wave belief is predicated entirely on the Spanish Flu epidemic of over a hundred years ago, I'd say it's not exactly a good basis for comparison.

 

[duncanp]

Apart from the R value, you also have to look at the K value, which is the variation in the number of people infected by one person with COVID-19.

Anyone who has done statistics will know this as the Standard Deviation.

To give a few examples:-

• If four people with COVID-19 each infect 5 other people, there are 20 infections, R = 5 and K = 0.
• If another 4 people infect 2, 4 6 and 8 people respectively, there are still 20 infections, R = 5 and K = 2.24
• If another 4 people infect 1,1,0 and 18 people respectively, there are still 20 infections, R = 5 and K = 7.52

The higher K is in relation to R, and particularly if K is higher than R, that suggests that there are a few "superspeaders" about.

So when look at the statistics, you have to examine them carefully and look at a whole range of data to get a complete picture, which can then influence policy decisions. (ie the R value for the North West may be higher than London, but is the K value in the North West lower than London, and what is the confidence interval for both the R value and the K value in both those regions?)

I think the key to lifting the lockdown successfully is having test, track and trace running successfully. I used to work in software testing, and I can understand the issues involved in getting the app rolled out on a nationwide basis. However, my experience suggests to me that it is better to delay the release of the app so that it is working properly, rather than release it early but with bugs and errors which lead to a lack of confidence in it.

I also think that having rapid access to testing, and rapid access to the results is another means by which we can learn to live with COVID-19 without having a second lockdown.

 

[NorthOxonian]

Apart from the R value, you also have to look at the K value, which is the variation in the number of people infected by one person with COVID-19.

Anyone who has done statistics will know this as the Standard Deviation.

To give a few examples:-

• If four people with COVID-19 each infect 5 other people, there are 20 infections, R = 5 and K = 0.
• If another 4 people infect 2, 4 6 and 8 people respectively, there are still 20 infections, R = 5 and K = 2.24
• If another 4 people infect 1,1,0 and 18 people respectively, there are still 20 infections, R = 5 and K = 7.52

The higher K is in relation to R, and particularly if K is higher than R, that suggests that there are a few "superspeaders" about.

So when look at the statistics, you have to examine them carefully and look at a whole range of data to get a complete picture, which can then influence policy decisions. (ie the R value for the North West may be higher than London, but is the K value in the North West lower than London, and what is the confidence interval for both the R value and the K value in both those regions?)

I think the key to lifting the lockdown successfully is having test, track and trace running successfully. I used to work in software testing, and I can understand the issues involved in getting the app rolled out on a nationwide basis. However, my experience suggests to me that it is better to delay the release of the app so that it is working properly, rather than release it early but with bugs and errors which lead to a lack of confidence in it.

I also think that having rapid access to testing, and rapid access to the results is another means by which we can learn to live with COVID-19 without having a second lockdown.

I wasn't aware of the standard deviation being described as K, despite actually having studied mathematical models for epidemics in my stats course at university. But it certainly is an important variable.

There's another important consideration. The R value, and most of our models, seem to be deterministic. They ignore any degree of randomness in how the disease is spread. That's a fair assumption when the disease is rife, and many people have it, but we're now reaching the point where infections in even quite large towns and cities are into single figures (Exeter and Gloucester are two iirc). If the small number of infected Exonians all spread the disease to more people than we expect, then the R value there could be 2 or 3, and we could see a small spike there. But if we get lucky, they might not spread the virus at all.

I think this partly explains the fluctuations in the R value. Imagine a region where 100 people were infected, and where the true R value was exactly 1 - by my reckoning the number of people that those 100 could infect is between 84 and 117*. Even if we had all the information about how many people were infected (which we don't), we wouldn't even be able to tell the R value to even one decimal place. As it happens, there are probably more than 100 people infected in all the regions, but not that much more...

* I went with a 95% confidence interval, and assumed that the number of transmissions per person was distributed as a Poisson distribution.

 

[HSTEd]

If the second wave belief is predicated entirely on the Spanish Flu epidemic of over a hundred years ago, I'd say it's not exactly a good basis for comparison.

It's not.
The second wave of the Spanish flu epidemic was a somewhat different virus - which is why it appears it was able to reinfect everyone.
The second wave here is caused simply by the fact that very few people have been infected and are immune, and some proportion of the population is going to be infected for the forseable future.

The second restrictions are released or are no longer effective, the virus will start spreading again and we are back where we started.
This is why a lot of lockdown boosters put their faith in magical "track and trace" systems.

Otherwise we will have to repeatedly lockdown over and over until a mass vaccination programme can be completed.

The NHS was virtually destroyed back in March where some hospitals were reporting 40% of their staff were either off sick or self isolating due to symptoms. You can't run any kind oh health service with no staff! That's why the NHS closed its doors to all but covid patients. If infections rise again out of control, like back in Feb/Mar, the same will happen. Hence a lockdown WILL be needed again if covid infections look to be heading back to exponential growth.

In an all in mitigation strategy critical medical staff would not be self isolating.
If possible they would be transferred to treating COVID patients, who would receive any and all resoures that were not required otherwise.

The vast majority of those self isolating staff were not infected after all.

 

[Puffing Devil]

Ten thousand extra dead from dementia in a month?

You have a source for this claim?

28,000 cancelled cardiac operations.
Unless all those operations are completely pointless then the number of dead and loss QALY is mounting rapidly.

Again - you have a source? You can say how many critical to supporting life?

Or the huge drop in stroke, heart attacks and cancer patients presenting to the NHS.

Or suspected stroke, heart attack and cancer patients presenting?

Triage exists.

Ultimately the hospital can deny people treatment and discharge them if it does not have the facilities to treat them.
As for ICU facilities, given that the survival rate for ICU is absolutely appalling, a complete collapse in COVID ICU care is not going to substantially increase the death rate.

By discharging them, do you mean sending them home to die when they could have had a chance of treatment?

Thing about pandemics is that we don't chose.
The virus does.

We do choose. We lockdown, minimise contact, contain the virus. Fewer people catch it, fewer people die.

Do it fast, well and early and there's a good outcome. I'm looking at New Zealand here.

Dither, come out of lockdown too soon and we're left with a killer virus circulating in the country. Either we lockdown properly to halt the spread or we accept that people will die, including those that are treating them and we would rely normally for other care.

 

[Puffing Devil]

Or perhaps

1. Carry on with lockdown, which is currently costing an estimated £100 billion per month and before long with such costs and a loss of tax revenue will have to result in deep cutbacks in all public services, including the NHS.
2. Start to move back to normality so that we can at least start to tackle the impending recession (expected to be the worst in 3 centuries) and make sure some revenue gets back to the essential services.

Now I know which one I would choose, so which one would you?

Where do you get your £100 billion a month from?

I value human life - what would you say a life is worth? When you start to talk about economics you need to remember the lives that will be lost without controlling the virus.

 

[Puffing Devil]

As you well know, we do not have that choice, so it is unhelpful and overemotive to bring it to the table.

We do have a choice. Lock down and protect, or stay out and spread. Without a vaccine and with a circulating virus your nearest and dearest remain at risk. It's that simple.

 

[NorthOxonian]

We do have a choice. Lock down and protect, or stay out and spread. Without a vaccine and with a circulating virus your nearest and dearest remain at risk. It's that simple.

Do you not accept that lockdown could have severe health implications in and of itself? We're about to enter a massive recession, which will be worse the longer lockdown continues - do you think the NHS will be able to survive that?

The horrible fact of the matter is that a lot of people are likely to die - the question is whether we'd rather have them die from this virus, or because the NHS no longer exists (and in that world the economy and society have been trashed too). Neither is palatable but one is less awful than the other.

And before you ask "who would you sacrifice?", I've already lost a loved one during the course of the virus. But we need a sense of perspective, we could lose everything if we're not careful here. There's a risk that we could end up with tens of millions of people seeing out their lives in misery thanks to economic collapse.

Mind, this doesn't really have much to do with the small fluctuations seen in the R number.