UKH

Post 1 - Four Nations

Plot 9x is the interesting one to me this week.  I’ve been sticking with an exponential rate constant and halving/doubling time plot even though we’re clearly well past the point there is sustained potential for exponential growth.  I still find it very useful as it gives us an indication of the capability for the virus to spread regardless of the absolute number of cases.  

There’s been a lot of growth in absolute numbers as is clear from linear plots of cases (e.g. as on the government dashboard) or from careful squinting at my log-y plots.  But, in terms of how the rise is happening, it's through a virus that can barely spread in terms of how many people one person is infecting.

The take home from plot 9x for me is that infection is barely capable of sustaining a spread from 1 person to 1 more person at the moment in any of the nations.  We have had rapid growth in case rates not because the virus is spreading aggressively, but because it is spreading weakly from a large baseline number of infections.  That large number has been arrived at by months of increasingly unaggressive spread interspersed with periods of lackadaisical decay.

That our journey from lower cases to higher cases has taken so long, with such unaggressive spread, suggests to me some positive view over aspects of our situation, and that’s the theme for this weeks posts.  It doesn't make the immediate situation over healthcare admissions levels any better, but it does give hope that things won't suddenly go exponentially out of control on rapid timescales, and there are clearly reasons for this.

The lack of rapid exponential growth means there is time to make decisions to control the situation, if the political will is there.  

Link to previous thread - https://www.ukhillwalking.com/forums/off_belay/friday_night_covid_plotting_48-...

Post 2 - England I

The latest burst of rising cases is coming to an end and that can be seen in 6s curving over towards flat.  It’ll show in the next post’s rate constant plots as well.

We've had a series of rising and falling periods over the last three months, each time the rise outweighs the fall and the baseline number gradually ratchets up.  It wouldn't take much growth to take it to beyond the level of the last wave.

Hospital admissions have been relatively low during this period of ratcheting cases, both from the vaccine program and more recently they reduced despite rising cases as the barycentre of cases shifted from the old to the young.  That demographic shift has started reversing recently with sustained growth coming to all ages for a few weeks; I think admissions are going to rise a fair bit yet compared to the last few phases of rising cases, due to the demographic shift in growth.  This will be to the clear discomfort of anyone whose been paying attention to the escalating news stories on the state of hospital admissions over the last 6 weeks; a problem far wider in cause than Covid, but none the less one that's only going to be made worse by it.  18 months ago I’d imagined that we’d see a crash program to invest in healthcare capacity on as many levels as possible, recognising the difficulty that lay ahead.  I try to keep these threads from becoming political slagging matches, so I’ll stop sharing my thoughts there.

Plot 9e suggests to me that the most recent rise in cases is now translating in to hospital admissions and deaths as indicated by all curves reaching similar doubling times, where-as the last spike in cases around 09-18 (which was when it really took off in schools) never had such follow through.  I think we may see faster doubling times for the hospital and death measures emerge over the next week, representing worse growth there from the demographic shift that’s hiding behind the top level cases curve.

Despite the increasingly large numbers, I find some good news encoded in what we’re seeing.  Despite progressively dropping restrictions and individuals increasingly dropping their transmission and risk control measures (e.g. the recent ONS survey), we're not seeing the potential for sustained growth in cases; since the Euros football ended, they've pogoed between decay and growth, taking around 12 weeks (circa 85 days) to double - this despite schools returning, the seasons changing, universities returning, nightclubs reopening and all the fading individual commitment to transmission control measures.

I find this comforting for two separate reasons. 

• To me, it speaks volumes about the state of immunity across the population and our journey out of the pandemic phase
• We have  everything in reserve to send infections and cases in to decay if we need to.  As I’ve observed for a couple of weeks, now feels like a good time to use some of the mildest measures to look for some decay in cases to inject some headroom in to the hospital system; as time goes on without that the definition of “mildest” starts to look less mild.
  • No sign of the government mandating any of this, but the messaging is getting clearer and clearer on taking basic measures to reduce transmission, with WFH getting the press time today.  I was pleasantly surprised to hear Javid mentioning opening windows in the recent briefing.  Would be nice if they were buying HEPA filters for households in fuel poverty who can't afford to be warm and ventilated in winter.

Post 3 - England II

Week-on-week method rate constant plot

• The rate constant here is doing a good impression of heading for decay, could be there in the next couple of days' data.  It's got the feel of another of these weather related wobbles to me, we'll see once all the weather data is in.   As small as the growth of this period was, it was applied to a large baseline.

Demographic rate constant plot

• The week-on-week demographic plot hints that changes in children are just lagging those in adults with the red markers looking a bit right-shifted from the black ones.  This would be good news for the shape of the decay in hospitalisations; it also suggests to me that on balance schools are not the driver of transmission any longer.

Plot D1.C - demographic cases

• The rise is remarkably uniform in all ages unlike the various past identifiable events over the last few months; a bit stronger in the older folks. 
• Needs a few more days data to see how the decay really emerges.  The demographic data is more lagged than the top level data.

Plot 18 - regional rate constants.

• Putting the South West to one side for a moment, hospitalisation rises are now landing in all regions from the recent rises; the recent turn to decay in cases does not look to be homogenous across the regions; hopefully all are in to decay very soon.
• The South West has had a stand out level of growth in cases. 
  • Lots of discussion on this towards the end of the last thread, many different factors around  the lab failure likely contribute to this growth - a depressed baseline from false negatives the is then corrected, people from the tail end of the period going for re-tests and their positives coming through later on in the timeline, more spread as a result of some failure to quarantine due to the false negatives.  Conversely, a sudden regional rise could be a signature of a more transmissive variant.  We’ll have to see if exponential growth falls in line with other regions over the next weeks or so.  There could now be a “false” decay component as the re-tests stop coming in.
  • On the lab front, Alan McNally has had more to say - quote below.  I got side-tracked for half an hour just reading his twitter feed.   Thoughtful stuff.  You won't be disappointed.  
    
    “In the long list of Covid disasters and scandals, this is pretty near the top,” said Alan McNally, a professor in microbial evolutionary genomics at the University of Birmingham, who helped set up the Lighthouse Covid testing lab at Milton Keynes.
    
    “You shouldn’t be relying on anecdotal reports to spot a problem of this size. That’s the unforgivable thing about this,” he added. “I don’t think it’s going too far to say that an absolute failure of quality in that lab is going to lead to very serious illnesses, maybe hospitalisations, and maybe worse.”
    
    https://www.theguardian.com/world/2021/oct/23/covid-testing-failures-at-uk-...

Comparison Plot

• This is one that last cropped up many moths ago; the various measures are smoothed quite a lot to make a comprehensible plot, and are all normalised to have the winter 20/21 wave peaks to the same height.  On a linear y-axis it makes it clear how much the situation has changed in the last year.
• What a difference a vaccine makes.  

Thanks!

Post 4 - Scotland I

Abridged content on Scotland this week as I’m using my Covid writing hour to broaden out in terms of interpretation and geography in later posts…

Cases have resumed their decay, having previously paused whilst England went in to growth - tentative support for a lot of this being related to the weather, perhaps.  Hospital occupancy continues to be very slow to decrease in Scotland.  Deaths are decreasing faster; this doesn’t seem very intuitive but I’ve not dug further in to it. 

Post 5 - Beyond Our Shores

Earlier I noted how cases have taken around 85 days to double in England.  A lot has been made in the press and elsewhere about the very high daily case rate (rate meaning cases per head of population) in England.  Last week I put some case fatality rate (CFR) plots in to make the point that whilst the situation with regards to deaths is still worse in England, it is much less so than cases suggest.    We’re having a lot more detected infection (cases) without either hospitalisation or death.    Difficult to infer the precise reasons for this between demographics and efficacy of testing.  The infections that occur without hospitalisation or death still invoke some immune response, and that seems to me to be a clear part of why we’re seeing the virus lack the ability to spread aggressively in the UK, with hints of this in things like the “shadows” in recent weeks’ D1.c plots.

We’ve also heard and seen that cases and deaths are lower in many nearby comparator nations.  Even accounting for likely differences in testing rates and demographics, it’s clear that the scale of infection has been around 2.5x lower in many of these nations over recent months.  So, it seems likely that less immunity has been evoked across their populations.

This is where there is clear if not precisely stated divergence in policy between nations.

With that in mind, I’ve not put together a plot of case rates across a reasonable set of comparator nations using data aggregated by and download from Ourworldindata; as flawed as cases rates are for comparisons, their change is very informative as local factors like testing efficiency mostly drop out of rate of change measurements, so exponential rate constants are more comparable in terms of understanding how aggressively the virus can spread.

Plot OWiD.Cases (Europe) shows the daily cases/million for the countries on a log-y axis (left) and measured rate constants and characteristic times (right).  

Plot OWiD.Europe (Deaths) shows the deaths per millions; the no rate constants as it’s all thankfully low numbers and so quite noisy; for now anyway

Some interpretation:

• The absolute numbers for cases were lower for the comparator nations so their rate constant plots are quite bumpy, but to my eye it looks like they’ve all been heading for growth since late September.
• Big changes abound in the cases data since last week; suddenly all the comparator nations have joined England in growth, and as of today’s data release, all of them have cases doubling faster than the UK (right plot). 
  • With the filtering OWiD do on the data and the way decay in some areas masks growth in others until growth is established everywhere, I doubt that any of the other nations have yet reached their peak doubling times
• This higher rate constant and faster doubling time can be seen in the case rates vs time plot (left plot) as the gradient on a log-y axis plot like this is proportional to the rate constant.
• Death rates are turning to rise across the comparator nations as well as the change in cases starts to feed through.

So…..

Despite the high vaccination levels across Europe and their higher adherence to transmission control measures than the UK, all the comparator nations are now seeing cases rise with far worse exponential rate constants (and so R values) than the UK.   If the current situation were to continue without change, cases in Germany would exceed those in the UK in under 3 weeks, but given the differences in CFR, locked in death rates could exceed ours much sooner.

This is why I’ve been cautioning strongly against drawing international comparisons on case rates over the last few few weeks - it is cherry picking one measure at one point in time, when the relationships between measures differs across nations and perhaps the biggest difference is (or, was….) a difference in timing rather than net total outcomes.  Now we start to see the risk of cherry picking being made clear by the bad direction the data is taking elsewhere.

Perhaps the case rates will break across Europe just as they have in England, perhaps more control measures will kick in etc.  Over the last 24 hours the German agencies have started some significant messaging through the press about a winter wave; they don’t seem to be giving of signs of wanting to stop it.   I hope poster “jimtitt” won’t mind me pasting their text from another thread  - this suggests that control measures will start kicking in soon in Germany so that the aggressive doubling times in cases will be moderated by control measures soon enough:

"However it should be noted that Germany has changed it's priorities, the preventative measures no longer kick in on simple case numbers but primarily on hospitalisation and IC occupation. As the relevant legislation is still in force some local areas are already reverting to more stringent measures  my area will probably go back to some extent end of next week particularly in the schools (and rely on the autumn break which starts on the 2nd Nov)". [1]

If we look at recent case rates,

• The UK has had the highest recent total cases, and is now seeing the lowest growth in exponential terms (which reflects the ability of the virus to spread far better than absolute growth).
• The next highest for recent historic cases is  is France, and they’re seeing the next lowest rate constant for now.
• Germany has had by far the least cases over the last 3 months, and is now seeing the fastest exponential rise.

This is very far from making a conclusive case around infection acquired immunity and the gaps in vaccine coverage, but it’s significant pause for thought.   Compare a ~85 day doubling time in the UK to a ~7 day doubling time in Germany, and despite what sounds like wider use of basic control measures in Germany, there are massive differences in how aggressive the virus can spread between the countries - with it spreading 12 times more aggressively in German 

It’s far from clear to me that the difference between the UK and the other nations lean in favour of one side or the other, especially as the recent and proximal-future emergence of more powerful therapeutics is a potential strong symmetry breaker in terms of outcomes, but I suggest the point that the situation is far from as one sided as it may have seemed to some over the past few weeks.

It pains me to have to state this, but given the level of various recent comments, I shall be explicit.  I do not want anybody to die from this virus, I do not have any jingoistic wish for the UK to do “better” than our friends and neighbours, and I no not know which approach is going to turn out to be less bad.  But I do believe strongly its important to try and maintain an informed perspective that recognises the complexity of the reality of our situation.    I don’t normally try and set bounds for these threads but keeping in mind how complex and dynamic the situation is, perhaps we could avoid simplistic pointing at UK case rates and doing a chicken little impression?   It’s not just that it’s not IMO supported by a detailed understanding of the situation, it’s that it detracts from the very real pressure points in our near future, for example over healthcare provision.

Post 6 - English Demographic Case Fatality Rates

I’m gong to return to these in the next post, so mainly just explaining them here.  It’s most a copy and paste from their appearance at the end of the last thread [1]

As always, interpretation of CFR is subject to the proviso that detected cases do not equal infections, and that the relationship between them can be flexible.  But I think the UK has been relatively consistent over time, and things should be even more comparable between different a get at one point in time.

These plots are made using the UK dashboard's demographic case and deaths data.  Cases data is deweekended.  CFRs are then calculated using 7-, 14-, 21- and 28- day lags from the cases and deaths data, and the result is filtered with a 7-day moving average.  The colour bands show the minimum and maximum CFR values across the various lags, and so give what I consider a reasonable bound for a "true" longitudinally determined CFR of each age based cohort.  When the number of deaths is very low, the trace is censored as it gets very noisy.

Two plots of the last year, both with logarithmic y-axes.  One the second one I've manually traced black lines back a year from the current CFRs.  This nicely shows how the vaccines have effectively made people ten years younger in terms of fatality rate.  Funny thing is, I don't feel ten years younger...

Also one extra plot I did for visualising the input data - it's a time series heat map of cases in a teal colour and deaths in red.  How dramatically hospitalisations and deaths could reduce if a relatively small number of cases were prevented; it was good to hear the health secretary tell people to open windows in his briefing a few days ago, but energy poverty and winter is the reality for too many people.  In-room HEPA filters with a bit of guidance on their location seem like an obvious step here.

[1] https://www.ukhillwalking.com/forums/off_belay/friday_night_covid_plotting_48-...

Post 7 - Risk

So, this post is out on a limb a bit compared to the normal stuff I do, and there's a chance I'm suffering from not recognising gaps in my skillset when making it, so I hope a couple of the regular readers and occasional posters from the actuarial and medical sides bring me up on any bad steps!  Read tentatively. 

• Last week there was some good discussion over the risk of hospitalisation for double vaccinated adults around their mid-life crises age [1].  The tentative conclusion was that, on average, an individual aged around 40 and double vaccinated has a similar risk of going to hospital from flu in a “typical” (no such thing) flu season as they do from Covid in the next 3 months, if rates don’t rise much more.  At an individual level, personal risk factors can make it better or worse, and those risk factors won't be the same for Covid and flu, so the amount an individual can conclude from this is limited, and they need to consider their personal risk factors. 

The CFR plots suggested another way of looking at the relative risk now posed by Covid.

The  plot below takes the CFR for time periods last October and this September, and shows the ONS all cause mortality as a black line.   My take is that all cause mortality can be considered as a fatality rate for life itself.    As a side puzzler, how does a bathtub curve live this emerge when each individual chemical part of a living being will have a half-life like decay curve?  Answers on a postcard to "Ageing Is Complicated And Emergent". 

My interpretation is that

• For over 65s, the lethality has reduced from much more than life itself to a comparable level.  Covid doesn’t replace all cause mortality so the total risk is that of both combined, or roughly double the pre-Covid risk.
• For younger adults, the risk last year was comparable to that of pre-Covid life itself - so again Covid roughly doubled the risk - but now the CFR is several times lower, which means the total fatality risk of pre-Covid life itself and Covid is not much higher than pre-Covid life itself, although Covid remains a significant part of the risk

So, we now have data to show that, for adults aged under 65, on average the risk of dying from a single Covid infection is no longer the most likely thing in your life to kill you within the next year.  That feels like a pretty major milestone to me.

There are two diametrically opposed and equally correct ways of looking at the risk of Covid for middle aged adults when it has the same fatality rate as pre-covid all cause mortality.

1. Covid is pretty unlikely to kill me (it is a small risk)
2. If I get Covid and I die in the next year, it's most probable that Covid kills me (it is a big risk) 

As we move out of the point we could overwhelm healthcare by a factor of 7x, individual risk becomes much more important as societal risk is no longer requiring the imposition of control measures.  Different people will have very different approaches to individual risk and none of them are right or wrong.  I like rock climbing, and I fall in category (2).  I've known a few BASE jumpers and I suspect they fall in category (1).  

For me, it's important that everyone making decisions over their exposure levels and their risk does so with an informed understanding of the risk, how they interpret that as applied to themselves is their own business.

I really wanted to provoke everyone in to thinking about their risks in the context of their own wider mortality to make sure people are taking a rational look at Covid; the discussion over the last few weeks has beaten about this bush quite a bit; the promoting to do the CFR plots over the last couple of weeks crystallised an approach to bringing it up.

We also have to consider the risks of infecting others, especially the older and the more vulnerable, and of overloading healthcare.  Covid undeniably remains an existential risk to many, and healthcare is not a happy topic of discussion lately.

For adults over 65, a detected Covid infection looks still to be more likely to kill them than their next year of life.  This is why I see the booster program as so important; it’s about  a lot more than just “topping up” the efficacy that’s starting to fade but about achieving yet more immunity to hopefully bring these shaded boxes on the lot down.

A big limitation to this is it compares a one-time Covid infection risk to an annual mortality risk from all (other) causes.  We don’t have a good picture of how often Covid re-infection is going to occur post double vaccination; presumably this will be less often in younger people than older, and seasonality is likely to feature heavily.  So, we have no idea of the annualised mortality risks of Covid in terms of how post vaccination repeat infections unfold.

Personal interpretation of this is fraught - there are many other risk factors evidenced in the data, and almost certainly more not yet well understood.  Critical point: The shading on the CFR zones here are what I consider the reasonable bounds for the cohort-wide CFR, not the bounds for individuals within that cohort which will be higher for some and lower for others.

My personal take is that for the first time since this started, Covid is no longer the stand-out high risk to my life, in that if I catch it, it’s no longer the most likely thing to kill me in the next year.  The threat it presents to me is still significant in terms of my survival probabilities however.  The good news is that I’m almost certain to live through the next year without Covid and perhaps only a bit less certain to live through it if I do get Covid.  Most of my risk factors are locked in, one that I can control is avoiding situations that pose a risk of high viral load. 

The next big question is how this risk is going to change after infection and re-infection, if it starts to reduce as a result of increased immunity then the problem fades away.  If the risk doest’ fade, or worsens due to accumulated health damage, the increased risk of Covid will accumulate over the years, if immunity builds and it fades away, it could become largely forgotten within years.  Only one way to find out - keep moving forwards through time at 1 day/day and see what happens.

[1] https://www.ukhillwalking.com/forums/off_belay/friday_night_covid_plotting_48-...

Thanks. I wonder if the lower case rates in some European countries are partly due to tests no longer being free there. I’m not close to the details but see, to recall that some countries now have paid for tests, at least for asymptomatic people. On the other hand, the recent requirement for a green pass for work in Italy could have led to more cases being detected due to the unvaccinated population having to get tests to be able to work. So there could be a few factors at work. I suspect though that the main factor is autumn. 

Your early comment about a less than 1 to 1 infection rate and how it could be curbed if the political will was there.

Unfortunately I don't think it is and I don't think the population would comply. A lot of people I meet think it's all over and media outlets like talk radio (yes I still enjoy shouting at the radio whilst driving) are actively encouraging people to ignore any new rules that may come in.

It's still out there and though the majority seem to be mild infections not all are. A guy I work with is on his 8th day in a medically induced coma on a ventilator despite being double jabbed and not having any known health problems. 

I think a lot of adults simply don't care anymore and just want life to carry on as normal.  Other people die - but not me, sort of thing.  Two or more years of Brexit, two years or more of Covid.  Time to move on.

Cracking set of posts this week.

I think the light at the end of the tunnel is starting to come into view and ”Freedom day” is looking more and more like an inspired public health move as time goes on.

Does Europe have enough infection induced immunity to avoid lockdowns this winter? Germany could certainly be in a tight spot. New Zealand will be in for a shocker.

There’s always been an unevidenced narrative that it’s only the weak / sickly / old that this virus takes.  

There used to be occasional media stories about young/fit/healthy people taken by it, they’ve faded away but the point remains.

I was very wary of this virus when it first came along because the cytokine storm lethality mechanism is driven by subversion of the host’s immune system, and so a strong immune system can potentially be a risk factor.  Vaccination improves this by reducing the probability of subversion.  This risk factor associated with the fit and healthy seemed stronger with original SARS and the Spanish flu, but has also been hinted at IMO from the start by the reporting over SARS victims.

There’s a series of papers coming out showing genetic factors associated with severe covid infection that are related to immune function and cytokines.  It seems likely that for some people, their card is marked in such a way that being a paragon of health could potentially work against them.  Of course, being unfit would work against them in other ways.

Perhaps I’m over- or mis-interesting the medical evidence here, but my take is that there is a Russian roulette aspect to infection.  Perhaps one say we’ll get personal genotyping derived personal risk scores for various diseases.  There’s a class of therapeutics called JAK inhibitors that have the potential to help with this lethality mechanism, a couple are now recommended in some circumstances and others are going in to trials.  

https://www.covid19treatmentguidelines.nih.gov/therapies/immunomodulators/k...

Edit:

I can’t take it any more.  Particularly Jeremy Vine.

You’re accumulating downvotes; I don’t think it’s time to move on but I do think a lot of people have, and their thinking is much as you say.  

Only problem is the “other people die” part… It’s not always someone else, and it shows how worryingly thin parts of our society are, and that’s not a good thing as we go in to generally uncertain times over climate, energy and resources.

If it works out in our net favour in the end - far from clear but a recognisable possibility that shouldn’t be denied - I’ll never be happy with our approach as the messaging has never been clear on the level of risk that remains, and if it had been I think we could have had much the same level of immune inducing infection with a lot less death.  We didn’t even try, if we had that would have been inspired - but it would also have required ministers to fully own the solution up front.  The lethargic roll out of the boosters is not “inspired”; if as various people think this has been a deliberate but not explicitly communicated strategy - and given previous SAGE models - the importance of this 3rd dose in reducing pressure on healthcare and deaths has been clear for months.  This just needs sorting, and given the smashing success of the first and second dose roll outs, it’s hard to understand. 

I hope so, because lockdown sucks.  I got the impression protests in parts of the continent against control measures have been a lot larger recently than in the UK, so the politics of it all will be fraught if it happens.

They may well be, but as I said last week it’s a much better outcome than what we’ve had, they delayed almost any death until everything was in their favour, so net total it’ll be a stand out good outcome I hope.  It’ll be toxic for the political leadership because others will use the deaths to paint their success as a failure, but Arden had the strength of character and belief in the science to do the right thing and then face the consequences despite knowing how it may be used to pillory her.  

Unfortunately I agree with what Bill and Dax said too. Although I agree with your original post that relatively small changes (compared to what has gone before) would bring the virus in to decline, I don't think the public are likely to make these changes. I think there is now an attitude amongst many people that we need to get on with things and all we can do to help is get vaccinated. There is also a large proportion of the population that is determined to believe that if they are vaccinated then they are no longer vulnerable because they don't see any other way out and they can't countenance being in restrictions for the rest of their lives. I think it would take something major to turn the dial on this; probably a rising wave of deaths and the Government introducing more closures together with more serious messaging again. Nothing in the current plan B will change perceptions.

An example - my mum travelled to a memorial service for an old university friend near Bath on Thursday. Currently as we know the area has some of the highest rates in the country. Most of the attendees would have been in their late 60s or early 70s (so hoping for a booster imminently but very few will have had it beforehand) and most of them university-educated. I asked her afterwards if there were any covid measures. Answer - none. The service was full with no additional spacing, and she said she took her mask but didn't wear it because not one other person was doing so. I think from this position, it's going to be extremely difficult to get the public back in the mindset of minimising transmission.

Third doses are vital. The roll out seems mixed. Everyone I know who’s eligible (not many) has got it, but that appears far from everyone’s experience.

Re your point about achieving immunity driven infection with lower deaths, how would you see this working in practice? The vulnerable could “shield” or something similar, but when they eventually emerge they still have the same risk profile i.e. net deaths would be deferred not reduced?

To clarify I meant looking forward when I said the UK was in a better position. Clearly NZ will have lower deaths/population than the UK. Having said that, NZ will likely see exponential growth and they’re going to have to manage that somehow. Lockdowns or other interventions look almost inevitable for them.

The difference here is simply the frequentist and Bayesian interpretations of the same underlying data. For personal risk, Bayes (likelihood modification) is going to look scary on a small risk, but it’s just base rate fallacy. The risk is low. (Not that I’m advocating ignoring it.. that’s a societal and political decision, this is pure personal risk analysis)

Its a big risk *factor* but a small absolute risk, if you will. Another way of saying life for middle aged adults in the uk isn’t very risky!

I feel a U-turn coming on… inevitable really given the level of protest, but I’d have given them until the end of the week.. 

https://www.theguardian.com/world/2021/oct/23/uk-government-paves-way-to-br...

Did you just call me a Bayesianist?  That's fighting talk, that is.  

I should have sent my post to you for review before posting it, I could have better stated:

• Covid is pretty unlikely to kill me (it is a small risk)
• If I get Covid and I die in the next year, it's most probable that Covid kills me (it is a big risk factor) 

So, it comes down to how much an individual in middle age is bothered about managing their personal risk factors; it doesn't matter much to survival chances within the year as you say, but it cultivates a habit that becomes more important with every year we age until it's an almost critical survival skill.  Especially as the decisions we make now to eliminate some of the largest (but still small) risk factors get baked in to our health going forwards and can have big payoffs down the line.

Well, boosters are the obvious step - for the more at risk, shielding more until the booster is out is a big difference, and getting them rolled out with minimal delay.

Risk profile is not static though - it gets worse (surprisingly quick as this drags on) as we age and - to date - with new variants, and it gets better with emerging therapeutics.  Difficult to know the balance there.

However,  life is all about deferring death, it's the only game in town.  Mustn't loose sight of that - being alive is different than being dead, so it's good to stay alive.  

In terms of staying alive, messaging around managing high viral load situations seems an obvious step - and one Si dH's example shows isn't getting through.  

Agree re NZ - hard to imagine them avoiding some significant restrictions, and I'll probably get rather angry at the way the Telegraph and Spectator are going to falsely present this as a failure of their policy.

+1

In reply to wintertree:

+1

Irt thread:

No opinion or judgement offered, just thought these would fit well here:

Austria warns of lockdown for unvaccinated https://p.dw.com/p/425Od

German COVID rate soars to highest level since May https://p.dw.com/p/425bE

That Austrian policy would sure make people think about getting a jab.

If you read into it they're pretty open about that being the strategy. A similar but much less draconian threat worked for France.

People in NZ lived much more normal lives on average during this pandemic. They didn't end up with a necessitated massive government financial support response, due to repeated imminent risk of hospital collapse.....that generated a mountain of debt. They had much less damage to their economy. The most at risk (and the groups after that) in their population are vaccinated (unless they chose not to) so per capita death rates worst case will likely be around a tenth of the UK and long term damage to the health service will be much less and long covid rates will be significantly down as well. There seems to be a perverse fascination in those wishing their future pandemic response ill when they have done so well.

Germany has scary growth right now but I'll give good odds they will deal with it. Their CFR indicates some big testing gaps when infection levels were low but it also likely shows better protection for school kids: for the sake of their (and their families) health, and also to cut educational disruption.

The UK is still flirting with system limits for our health service impacting our poorest the hardest. Our CFR is artificially low as we have allowed covid to rip through our schools. We have chosen to additionally disrupt the education of millions. On limited restrictions, the polls indicate a large majority want (and would support) a return to limited restrictions. People showed way more resilience with strict measures than the behavioural scientists predicted. The pessimism on display in our government and even from some on here needs better grounding: most will follow mandated restrictions. We don't know what the future will bring but we do know we in the UK are looking at a seriously corrosive effect on our health systems that are already punch drunk, daily deaths in the hundreds, long covid levels several times that level, and ongoing excessive school disruption. Our government covid public health actions give no reasons to celebrate beyond initial vaccination progress.

I think the idea nearly everyone in the UK (except the most vulnerable) has to catch covid is foolish. If R had been forced below 1 with limited restrictions a significantly larger proportion might have never have faced infection. We can't run alternative UK scenarios but again I'll give good odds that European countries that took a covid response closest to scientific advice will exit the pandemic with lower total infection levels, less deaths in the most vulnerable, less economic damage, less health service damage and less schooling damage.

Okay I’ll bite. What was the alternative to the current school situation. Yes we could have put in some additional mitigations (more ventilation ect), but the benefit to children from vaccination is marginal. They need to get immunity some way. Vaccination or infection - it doesn’t matter. It is better to give them immunity now so their future education isn’t disrupted rather than continue with this current sh@t show.
 

This is wishful and dangerous thinking. Every other pandemic we’ve ever faced has ended by everyone (or nearly everyone) being exposed and gaining immunity.

You might be right that there’s support for more measures, but if we use artificial means to suppress the R rate they need to be permanent measures. Otherwise Covid is back as soon as the measures are lifted - possibly with a more lethal variant. 

You’re also ignoring the single biggest measure we have in response to Covid. Test, trace and isolate…

The contracts aren’t going to run forever. You can see the cost and end dates for the contract in Sodexo’s press release.

https://uk.sodexo.com/media/news-room/sodexo-wins-30-per-cent-of-test-centr...

You’re essentially making the case for permanent masks, social distancing and permanent testing - not going to happen (thankfully).

I thought the conclusion on £35Bn test trace and isolate was that it is largely a waste of money. To describe it as "single biggest measure" is counterfactual when BMJ report it made no difference.

Lockdowns across the world show that avoiding people avoids the viruses they carry and have effects that can be seen over and over again as declining cases a week or two later. I don't see anything else with the exception of vaccination can be described as the "single biggest measure".

https://www.bmj.com/content/372/bmj.n663

BMJ but I think house of commons select committee had similar conclusion.

Countries that expanded existing medical and local public health professionals who did contact tracing for other infections rather than new call centres for contact tracing appear to have achieved more.

The Mrs got the booster 2 weeks ago and has been offered 2 more since.  (Good old NHS record keeping) but my business partners has been canceled to be re arranged at an indeterminate time because they are apparently short in our area. I know one person who has had 2 boosters , she said if she didn't need it they wouldn't have offered it. Leaves me wondering how many others have had multiple boosters.

Perhaps we should have run covid camps during the summer holidays? It would probably have been possible to isolate school age kids from their relatives until they had been infected and only let them home when they had fully recovered 

But the report doesn’t say whether it has a significant impact on the R rate, it says it wasn’t able to demonstrate that it changed the course of the pandemic. Slightly different things.

Put it another way if everyone with Covid could go about their business as normal (take the bus, go to the pub ect) would we be seeing significantly more cases right now - undoubtedly yes.

My point isn’t really about how much impact the interventions have, it’s that if you put in place the interventions (and bring the R rate below 1 by doing so - what Offwidth was proposing) then as soon as you remove them then Covid bounces back (potentially with vengeance).

Which means a wasteful TTI system, masks, social distancing, kids missing school ect.

Edit: For clarity - you can’t end a pandemic by using artificial means to suppress the R rate. That only postpones the inevitable. We can use artificial measures to control the timing of the spread. Whether it is wise to do so is a matter of debate, but the first point is just fact.

I’m not sure whether you’re saying that in jest or not, but it would have worked, even if it’s not the most practical solution.

I do think we need to think about the question “should kids and teachers with Covid be off school, provided their symptoms are sufficiently mild?”. 

With a few caveats (no public transport, no leaving the school site ect) would this be an unacceptable solution to the current problems with disruption to education?

When we stop testing for Covid, teachers will be in exactly the same position - why not aim to bring it in for the next summer term?

The history of Plague, Cholera and HIV tells us this is plain bullsh#t.

That's alright if you don't have an upper limit on how many infections we can cope with at once given our infrastructure. But when the infrastructure becomes overloaded (e.g. hospitals can't treat patients, schools shut because too many staff are off sick or caring for others who are sick) the whole thing falls apart and countless more damage is done than if infections had been controlled. Also, controlling infections buys time for more people to be vaccinated rather than infected reducing total harm.

Given where we are with NHS capacity, are you surprised that even Sajid Javid is trying to get people to reduce R, rather than celebrating our brilliant, inspired high infection rate at the start of the winter? According to some people, we wanted high infections in the spring and summer so we didn't have them in winter. Now it's autumn, the winter is looking absolutely grim, and apparently, high infections are still brilliant. What would it take for high infections to be seen a bad thing?

All those pandemics we've faced with the current technologies of the NHS, the modern economy, vaccinations...? What on earth are you talking about? You're just making it up!

So something that had no demonstrable effect on the pandemic is the "single biggest measure" according to you.
According to me lockdowns or vaccination, both of which have discernible effects, are respectively the biggest single measures in the short term and long term respectively.

Personally I think if you have some kind of lockdown* to take down cases then revert to what we have now we might roll along with R about 1 and 10 deaths per day rather than R about 1 and 100 deaths a day. I like the idea of saving the lives of 30,000 of my compatriots in the next year and think it's worth a try.

*harsher and earlier is briefer and less damaging

Generally I think postponing death is a good thing.

Cholera and HIV have thankfully never been pandemic. 

Their primary mechanisms for spread (unsafe water and bodily fluids respectively) mean they don’t behave anything like Covid does.

Plague (I assume you mean the Black Death pandemic) is an interesting one. Given estimates show that between 30-60% of Europe’s population died and the mortality rate for plague is 30-90% when untreated - I’m a bit sceptical of your assertion that it didn’t end by herd immunity.

Anyway Jim, you did this on the last thread - either explain why I’m wrong and put your own idea forward or downvote me an move on. 

Explain to me how a pandemic can end without herd immunity or elimination?

I’m open to ideas.

However what you’re arguing for is firstly impossible and secondly would pose real risks to all of us if we went down that route.

There are benefits of using interventions to slow the spread of Covid e.g.:

- more time for boosters

- better treatments

- reduce pressure on the health services

The downsides are:

- waning immunity

- risk of more lethal variants

The end point is always the same. We get population immunity by infection and vaccination. It’s the speed that can be changed.

Do we have data yet on whether the boosters are reducing infection rates and severe illness even further than the second dose did, or does it just punt you back up to the level you were at shortly after your second dose? 

I had my booster a few weeks ago now. The boss at work managed to get all staff and residents done on the same day, along with flu jabs. The team came in and did everyone in the facility so super convenient

Scarlet Fever, TB, typhus, typhoid, cholera and dysentery were post pandemic. They were endemic. They didn't go away despite exposure.

They were eliminated largely by artificial means such as sewage disposal, clean water, improved nutrition and improved living conditions.

We no longer have Victorian era child mortality rates of 30% due to artificial means.

Wrong. Wikipedia is but a short search away

https://en.wikipedia.org/wiki/Cholera_outbreaks_and_pandemics

https://en.wikipedia.org/wiki/Epidemiology_of_HIV/AIDS

That the WHO refer to the HIV pandemic as a 'global epidemic' seems to me to be a political distinction. By any epidemiological definition of pandemic it is one.

Exactly. A massive infrastructure change that effectively reduced transmission between infectious and susceptibles. We could have (and still could) chosen to do that with covid, better ventilation, different working environments, mask use, lifestyle and (structural?) economic changes, safer cities, plus vaccination and an effective contact tracing system. There has been a distinct lack of vision in our leadership through this pandemic.

You’re quibbling about terminology and not addressing my point (possibly because you can’t).

Covid is a highly infectious airborne respiratory virus. It shares none on the characteristics with any of the diseases that are being chucked at me.

I thought we’d agreed on this, but how do we end Covid without either 1) Herd immunity or 2) elimination.

Option 2 is definitely off the table now (hopefully the elimination fantasists have changed their minds now).

Option 1, given that vaccines aren’t 100% effective and not everyone is eligible, requires a mixture of vaccination and infection. I see nothing controversial in this statement.

You're forgetting option 3, endemic. Which is actually how most pandemics 'end', ie they don't.

We can choose whether we live with that at a high endemic level (e.g., HIV in Africa) or a managed low or negligible level locally (e.g., cholera, typhoid) 

I'd just like to see a plan for education, rather than crossing our fingers and pretending it can go back to normal.

A bit brutal for 5/6 year old to be removed from their relatives for 2/3 weeks, but giving 15/16 a taste on University life in the summer holidays opens up many positive possibilities.

If we are only detecting half the actual number of cases this is probably quite widespread at the moment. I thought it was vulnerable relatives that were the problem rather than teachers or students.

Only about 4% of schools have boarding facilities.I was thinking empty university halls of residence, center parcs, holiday camps etc.

They need to put that in the next recruitment advert. 40 years of exposure to disease riddled teenagers does wonders for your immune system.

As others have noted the recent Cholera and Plague pandemics have only been stemmed by hygiene and other controls, HIV is only vaguely under control by social hygiene efforts.

The principles are the same. If you want to consider something more akin to SARS-CoV-2: TB -- airborne respiratory pathogen. Ever wondered how we manage that in the UK, while it's endemic in many countries?

It's probably not going to end in any kind of neat "we've reached herd immunity, hooray" way. Immunity wanes, the virus mutates, we keep fighting back with new vaccines and treatments, and it gradually blends into the background of endemic diseases.

We're already out of exponential growth, with the current set of conditions (which are unstable). That's great, but it doesn't mean we can all relax, because we don't have the infrastructure to cope with the levels of disease, including the very high levels of covid, in our society. There aren't enough hospital beds. There aren't enough carers. There aren't enough supply teachers. Etc, etc.

Your dream of waking up to find we've all had covid, it's over, and no one will ever ask you to wear a mask again is a fantasy and will not occur.

What is impossible? Having on-and-off restrictions which cause minimal disruption, to keep infections below the level where strain on infrastructure causes additional harm. Why is this impossible? What are the "real risks" of this?

Yes.

What's the comparison? Immunity from infections wanes too. If you need another booster, that's OK. If you've got to risk severe covid, that's bad. 

What's the comparison? With high infections, there's no risk or more lethal variants? We've got borders open to the rest of the world who can also incubate new variants, why do you think that the speed of spread in the UK is going to be the make-or-break for worse variants?

The end point is not the same if many more people get severe covid along the way, and non-covid harms are multiplied because infrastructure is overloaded. You are clinging to a false hope that we can all get covid and then it'll be over. Not going to happen. We've got to throw a lot of effort at vaccination, and control R when we've not done enough to prevent infrastructure overload, hopefully with measures that cause the minimum of disruption. If that means that the transition to the endemic phase is slower, so be it, if we can avoid the multiplying of harm by having too many cases at once (on top of flu, etc).

You are making an incredibly unconvincing case for high infections being great.

I wonder if there were any libertarians of the time, that claimed, not being allowed to spit in public, was a restriction of their human rights.

Beauty is in the eye of the beholder. Labels are in the mind of the wearer.

COVID is one of many airborne diseases.

"Many diseases can arise after exposure to airborne particles, including:

• the common cold, which can develop from a rhinovirus
• chickenpox, caused by the Varicella zoster virus
• mumps, caused by a paramyxovirus
• measles, caused by another paramyxovirus
• whooping cough, a bacterial infection caused by Bordetella pertussis
• COVID-19, caused by the SARS-CoV-2 virus
• aspergillosis, caused by the Aspergillus fungus
• tuberculosis (TB), caused by the bacterium Mycobacterium tuberculosis
• anthrax, a bacterial infection resulting from contact with Bacillus anthracis spores
• diphtheria, a bacterial infection caused by Corynebacterium diphtheriae
• meningitis, which can result from exposure to certain bacterial, viral, or fungal particles"

https://www.medicalnewstoday.com/articles/317632#common-airborne-diseases

On schools, the best alternative was to vaccinate kids in the summer but since we started late, to at least have stronger than average control measures in  schools (cf European countries who did vaccinate kids earlier). At the point JCVI decided to wait for data in the summer we should have had an urgent initiative on school ventilation measures. None of this was ever going to happen though, because it would have involved Boris and his cronies admitting they were wrong. Things are so bad this weekend that some officials under strict instruction to stay 'on message' with government are speaking publicly against it.

https://www.theguardian.com/politics/2021/oct/23/english-local-health-chief...

SAGE say we need some measures now, even Vallance is saying when things start to look bad you need to go harder and faster than you think (in direct contradiction to ministerial messaging)......yet according to the news this am the government are possibly planning for plan B but will wait until after COP so as not to 'spook the horses' too much (the words used by a government advisor speaking to the press this weekend...as reported on Marr).

Pretty much all the second section of you post is junk. Others already picked up why.

What’s your point?

How has this thread leapt back a year in time? We’re so close to the end and people are talking about putting significant permanent restrictions in place…

Either say you want permanent restrictions or propose a different solution?

No one has been able to put up an argument against the fact we have to reach population level immunity and that this will require infections.

The number of total infections will be broadly the same regardless of the speed we go at.

What actual permanent restrictions is anyone here proposing (outside the  dystopian fantasy in your head)?

I'd go for the government plan B until pressure in the NHS stops being crazy (likely in the spring months). None of that is expensive or especially limiting. I'd urge more investment in ventilation (which will help for flu) and statutory sick pay on top of that.

You said this: “I think the idea nearly everyone in the UK (except the most vulnerable) has to catch covid is foolish. If R had been forced below 1 with limited restrictions a significantly larger proportion might have never have faced infection.”

The idea that a significant number of people susceptible to catching Covid can exist in the population without being infected would require significant permanent restrictions to keep the R rate below 1.

Plan B wouldn’t allow for this, Plan C wouldn’t allow for this - there is no plausible scenario which allows for it.

I agree that we need to manage our healthcare capacity, but that will be all about managing the speed at which those susceptible to catching Covid catch it, not preventing them catching it.

The type of herd immunity you propose is not the accepted case in epidemiology at all. You are misunderstanding some of what has been said here and ignoring a bulk of other evidence. There really is no guarantee the total infections will be the same, be it reached quick or slow... that's a worst case scenario (assuming significant extra vaccine avoidance doesn't happen) that we should be in no hurry to reach.

What you claim is impossible but is exactly what we faced in the last two summers. Very low case numbers meaning the vulnerable didn't need to shield as strictly.

We still have millions in the UK who are vulnerable and largely shielding, due to a combination of very old age, illness or foolish choices on vaccination. The IFR for them might still be around 0.5%. that's 50,000 extra deaths per million and a bigger number with permanent lung disability. I for one hope we could keep them alive and well through the back of this pandemic as it includes all four parents and many good friends.

I get the impression you're reading more in to what I write than I mean.  Appologies if I'm getting the wrong end of the stick.

I've been consistently critical (aghast) at how the pandemic was handled in the UK, and it's clear I hope that I've been in support of more transmission control measures at almost all times.

Yes, I agree with you, as I hope is obvious.

I have from the start up and till now (and no doubt will have in the future) expressed my admiration for the NZ approach, as well as my disbelief at the idea that rushing towards naturally acquired immunity was appropriate for the UK - I have been clear that I think it's madness, economically damaging beyond the immediate health impacts and not the way to do this.  I still wish we'd tried to be more like NZ.  It's never too late to change.

Yes, I agree with you, as I hope is obvious.

I even went so far as to quote another poster resident there suggesting they expect more restrictions to kick in within a week or so.

This doesn't diminish my point that the potential for the virus to spread in many countries appears to be much higher that it is in the UK right now.   Having had high cases is not a one-sided bad thing as various people are portraying - there is an other side.  If we can't be honest about this, we can't have informed discussions.  Is the other side worth the cost?  I don't think so, not at all.  But I'm not going to pretend its not there.  

Yes, the situation is not good.  I thought I have been clear in my views on this, and a in wanting more control measures. 

I disagree here.  Our CFR is what it is - the relationship between our cases and fatality.  

I've been out of my way to make the point that CFR shows cases are a poor basis for international comparison, and they're not endowed with a fixed relation  to IFRs.  What you're saying is in effect that we have a lot more infection not causing serious health consequences than other nations.  To paint this as a one-sided bad thing is churlish.  It's clearly two sided.  I don't claim to know which side is less wrong, but I don't deny it.  

To be clear, this is happening more here than elsewhere in large part because JCVI's independent, medical decision is to  offer only one dose to school aged children, not two, offering significant health protection but little  protection against transmission.  This is apparently done against political pressure for more vaccination.  Do you take issue with JCVIs decision?

I agree with past comments from you that more could have been done over air handling in schools which would slow down the spread of infection. 

I don't know, they did when there was high messaging over scary death rates; now people have bought in to the vaccine as the way out (it isn't, not alone) things have moved on.  I'm all for the government mandating more restrictions mind.

I hope you're not inferring that I'm celebrating anything.  To quote myself from earlier:

It pains me to have to state this, but given the level of various recent comments, I shall be explicit.  I do not want anybody to die from this virus, I do not have any jingoistic wish for the UK to do “better” than our friends and neighbours, and I no not know which approach is going to turn out to be less bad.  But I do believe strongly its important to try and maintain an informed perspective that recognises the complexity of the reality of our situation. 

Other than my view you're being churlish over wilfully misinterpreting CFRs this is the only place I think I disagree with you.

I think elimination is beyond imagination now, and we're still getting new variants with likely higher R0 values and increased genetic distance from old variants and vaccines.  

Can you actually evidence any credible scientist making a case that elimination is now achievable?  If it isn't, it's going to end up in circulation.  Even NZ recognise this with their changing stance to allow the virus to spread.

I agree - and I think the decisions that lead to this being almost inevitable have already come to pass, and things I think comes to where we're diverging in our takes here.

With these threads I try to understand where we are, so that I have some glimpse of what the short term future holds.

Critical point: We are where we are.  I don't like how we have come to be where we are, I don't endorse how we have come to be here.  But, here we are.  We've paid for more naturally acquired immunity, and we've paid in blood and misery.   It makes a material difference going forwards.  Putting hints from the data about this  the table has been an entirely predicable  Holy Hand Grenade of Antioch for the thread.  Which is a shame.

I think that's more-or-less what the university system did (unintentionally, of course), several times over. 3 day doubling times with the alpha variant at my local institution, and some of the highest per-person rates of infection ever seen during the pandemic.  Nicely packaged away from parents, grant parents and so on, with teaching switched to on-line during the outbreaks. 

To be clear, no, I don't think sensing school kids to Covid camps is an appropriate recommendation.  Full, honest and open communication of risks to adults and improved active ventilation to better protect them within educational settings seem more proportionate and better than what we have done.

Edit in response to another post of yours:

Way back when, I suggested to someone promoting the Barrington nonsense that they and all like minded people could go and have a "focused infection" camp at somewhere like Center Parcs, rather than encouraging focused protection of the most vulnerable.  My reasoning was that it's much easier to isolate young people than vulnerable, with the later needing a lot of physical contact with carers.  I was saddened that the idea got no takers, it's almost as if they were all talk but no conviction in their beliefs.

So what special kind of herd immunity are you proposing (try not to post a link to a Guardian article or indie-SAGE when you respond)?

I’m proposing we vaccinate everyone and then depending on how vulnerable people are either give them booster or let them get infected periodically - bog standard herd immunity. Which is, by the looks of things, what’s being done.

Those that are vaccinated (and boosted) face a much lower risk of severe covid. Slowing the spread allows more vaccinations, and lowers the risk of severe covid for each individual.

But that's what restrictions to reduce R do, as well as reducing harm all told.

Reducing the level of infections so that more people are vaccinated rather than infected reduces the harm caused by covid. Reducing infection levels so they don't butt up against the limited capacity of our infrastructure reduces the non-covid harms

Having very high covid levels increases these harms, and I don't believe for a second that it'll be worth it in the long run because we'll make up one morning with the UK having reached herd immunity and the pandemic over. Immunity wanes, the virus mutates, and the real world scenario is *nothing like* the computer model where the red spots turn black after they've touched another red spot and the virus goes away once x% of the spots have turned black. That's not what's happening with covid on planet earth!

The only sensible way to look at the situation is to ask which policy scenario likely involves the least harm, all told. High infections fails the test, because of the relative risk of severe covid in a person who could otherwise have been vaccinated, and the indirect harms of infrastructure overload. 

Is there a part I can explain more clearly?

I agree with you and jimtitt over the many different ways past pandemics have ended.  I sometimes wonder if HERV-K is a sign of how HIV would eventually have ended if we hadn't invented medical technology.

Cholera is a great example I think because it shows how improved ways of living can strip the pandemic potential from a pathogen, moving to a model where we life with occasional outbreaks and mop ups.  The biggest reason to argue against that vs moving it to endemic status is that it's storing up trouble for a collapse of society - and we see this with cholera outbreaks being common killers in failed states and war zones etc.  Hopefully not a great concern to the residents of the UK mind.

Edit: Cholera also provides early examples of the power of data visualisation to epidemiology.

Disappointing isn't it, there is so much change for the better that's been signposted by this pandemic.

The Barrington sham were very keen on "focused protection" at a time when it was entirely inappropriate vs the scale of looming disaster; they've had the ear of our leadership and yet now we're at a time more targeted protection could be of great benefit, we see government stripping protections.  MPs (even cabinet members) at times basically agitating against their own government's advice on masks, on WFH.  

The structural economic changes are more important than ever IMO.

You need to re-evaluate this belief. Your whole view pivots on it, and it is false.

Straw man.

I think that depends on how you measure distance to "the end".  There are many different ways, and I think as we get closer, the going gets harder.

Very few people, and they're generally not getting broad agreement.  Yet, there is so much we can do that falls short of restrictions that has the potential to make big difference.

We may well have to have more restrictions this winter, and to be clear that's not because of Covid but because we have Covid, RSV and potentially Flu all threatening to generate stand-out bad levels of demand for healthcare, and because healthcare is not ready for that - and Covid is by no means the only factor on the healthcare side of the equation either.

Quite what it's going to like between Black Eye Friday and New Year's Eve I shudder to imagine.  

My niece caught it within 5 days of arriving at Warwick 

 >Nicely packaged away from parents, grant parents and so on, with teaching switched to on-line during the outbreaks. 

Although they were "allowed out" into the local community. 

I think I'm just frustrated at the lack of any coherent plan. At least there seems some honesty in the proposal, although I can't see parental buy in being particular high. 

With some planning there is loads we could have done, and much of it was suggested to Williamson nearly 18 months ago. Summo often quotes Sweden where primary and lower secondary schools have carried on pretty much as normal. We could have quickly moved to a system more similar to their normal, but that is very different to our normal. 

Where do I start Jon..

I’m of the view that vaccine uptake is largely saturated, apart from those needing boosters. You seem to think that there’s significant numbers to go. I agree we need to get on top of boosters, but who else is there to jab? They don’t want it.

I agree that Covid causes real harm to some people who catch it (as would anyone unless they’ve been living under a rock for the last 18 months), but regardless of how slowly or fast we get to steady state pandemic position these same harms will be inflected (therapeutics might moderate this to a certain extent).

You assert that slower is always better, but I strongly disagree. Every variant has been more lethal than the last. Maximising population immunity as quickly as our healthcare capacity allows (with a healthy buffer built in) de-risks the likelihood of us being slammed by a new variant.

We cannot overload our healthcare capacity, but at the end of the day we can’t shy away from the fact that sufficient population level immunity must be reached and this will likely entail a lot of infections.

Yes, this led to an absolute dumpster fire of town/gown relationships here.  Messaging from centra was - in no uncertain terms - "there is no evidence of transmission in to the community".   Frankly I expected better of academics than to claim "no evidence" rather than fully researching to the point they could credibly claim "evidence of no transmission", but then I remembered about reality.

Totally agree. Both the lack of a coherent plan and a lack of honest, coherent, on-target messaging for me.

Totally agree.

Someone held the USA up as a country doing "better" than the UK last week, based on cases.

To drive home the point we can't use cases as a comparison...

Plots of the UK and USA data from OWiD; cases/million have been in decay for almost two months solid in the USA, and are a couple of times lower than for the UK.

Death rates in the USA are however higher by about 3x, and have been rising for the last two weeks, despite falling cases.  This suggests to me that true infection has been rising in the USA for the last month or so, despite cases falling their the whole time.  This can be seen as an increasing CFR for the states.

It's always possible infection is falling and demographics are shifting, but this looks like a properly balmy situation.  I haven't been following the news there much, but this looks like they're more or less abandoning testing despite rising infections and deaths and the onset of winter.  Basically flying blind in to a storm without cases as a barometer of infections.  

By engaging with the points that

1. vaccines reduce harm compared to infections; and 

2. healthcare/infrastructure overload multiplies non-covid harms?

Boosters is a massive issue now that immunity is waning. Loads more can be done here. Could get more young folk jabbed with a bit of stick (passports) if the modelling suggests it would help. Can speed up the kids jabs. 

Your view that the vaccine uptake is saturated is false.

No they won't! Vaccination - including boosters - is preferable to infection! Why do you think you can keep ignoring this?

But you're not arguing for keeping a healthy buffer against healthcare overload! That would involve reducing R! When the next round of variants happen we don't know whether they'll be more or less deadly or to what degree  they'll resist the vaccines or infection acquired immunity. You seem to be clutching at straws to justify why high infections are brilliant, when it's entirely obvious that they're not.

So can we do this with the current level of infections and restrictions? The NHS is saying no! Might we be able to do it if infections were lower? Yes!

No one's denying that. There will be a lot of infections. I am disputing your assertion that high infections are great because herd immunity is almost upon us. They're not, because they entail unnecessary harm; and it isn't, because immunity wanes (whether vaccine or infection) and the virus mutates.

Your view is based upon false beliefs.

Edit:  Agree with a lot of what you're saying Jon.  I'm not posting this in disagreement with your point but as a jumping off point about why I see the booster program as so important. 

I think there's a far more fundamental issue than that, there are many people for whom waning is not yet much of an issue, but for whom Covid remains a stand-out mortality risk.  That's what I was building up to with my post on risk [1], for over 65s, assuming they get infected, Covid is the most likely thing to kill them in the next year, and the fatality rates are no longer insignificant.  A third dose of vaccine should pull that risk down.  

In reply to Girlymonkey:

I've skim read a couple of articles, looks to me like they're additional protection for people in the younger end of our booster range, and restoring lost protection at the upper end.  Bit early to be conclusive I think.

Not discussed here, but the compound-arm cohorts from Valneva and Nonovax trials are being offered two doses of Pfizer, to allow them to meet vaccine passport travel requirements as the trial ones are not yet approved.  Quadrupple whammy -  I hope someone is tracking them specifically and doing some serology assays on them after their Pfizer doses.

[1]  https://www.ukhillwalking.com/forums/off_belay/friday_night_covid_plotting_49-...

I think it goes far beyond what Jon "seem[s] to think" to hard evidence that the booster rollout is not keeping pace with the rate at which people are becoming eligible.

You’re correct. 

You’re correct.

Boosters are a priority. Kids less so. Giving kids a single dose won’t do much to prevent transmission and they rarely suffer significant harm from Covid. We should absolutely be focusing on boosters.

In terms of young people who have lower uptake (18-30 ect) I don’t think bringing up vaccine uptake in that age bracket is going to make a whole lot of difference. Most of them will already have caught it (lots of socialising ect) and they don’t cause significant pressure on the healthcare system.

So no - apart from boosters, I think we’ve probably done all we can.

Of course it is! And I’m not ignoring it. However being double jabbed (and boosted of eligible) doesn’t prevent infection. Some (and I don’t claim to understand the mechanism behind this) might still be susceptible to catching Covid despite the jab. To get sufficient population immunity it is likely these individuals will need to have their immunity topped up by infection. In my mind there is no way around this.

This is my point around timing of cases. Yes, pushing infections will be a benefit if the NHS is overloaded. 

Immunity does wane (the evidence points that way anyway) - I agree.

Infection causes harm - I agree.

The virus mutates - I agree.

But I’m not sure how this really has an impact on anything. 

What belief? Population immunity?

Correct - not disputing this.

I meant apart from those who are after boosters, there isn’t significantly more people to go (unless we were to coerce them into getting jabbed).

I don’t, for example, think there are a bunch of 40 year olds who haven’t been jabbed that would now change their mind and decide they want it.

So allowing time for more boosters by reducing R is far preferable to allowing more infections. As for how much good jabbing the other groups will do, I don't have the expertise to comment, except to that ~2/3rd of hospitalisations are in the unjabbed so clearly hospitalisations could be reduced with more 1st and 2nd vaccines. I don't have any evidence about who might be persuaded with different techniques, so I have no reason to either believe or disbelieve your assertion that uptake is saturated. (This is another example of your post-hoc assumptions about what's going on, which you thought up to justify your pre-determined conclusion).

Who says? Why do they need infection not booster? Another post-hoc assumption based on thin air.

Really? You can't join those dots? It means that herd immunity is not around the corner. All allowing infections to rise over the winter will do is cause direct harm from covid and indirect harm from healthcare overload. Then we still won't have reached herd immunity, and we'll still be fighting to jab as many people as we can faster than they need a hospital bed.

My point is that your opinions are of very limited value whilst they are not in accordance with widely accepted facts.

1. Your opinion that test, trace and isolate is the "single biggest measure" is not in accordance with the observed impact of lockdowns or vaccination leading to a reduction in cases and better survival rates respectively. Each of these is a bigger single measure than test trace and isolate. 
2. Your opinion that "Covid is a highly infectious airborne respiratory virus. It shares none on the characteristics with any of the diseases that are being chucked at me." is not in accordance of the existence of other diseases that share the characteristics of being airborne (see the list), are viruses (see the viruses on list), transmit from person to person (see examples on the list that you can avoid by avoiding people) and have vaccines (see measles & TB on the list). I'm sure a specialist would be able to list far more common characteristics.

> How has this thread leapt back a year in time? We’re so close to the end and people are talking about putting significant permanent restrictions in place…

"We’re so close to the end" - this is just speculation.

Close to the end in terms of time - not if it becomes endemic as seems likely.

Close to the  end in terms of the death toll, cost to health and cost to the economy - I hope so.

We've had permanent restrictions of the right to sh#t in the water supply. We may have permanent public health restrictions related to Covid too.

Have you come across micromorts before?

https://en.m.wikipedia.org/wiki/Micromort

It looks like it has some similarities with the comparisons you're trying to make. The other thing it might do is make the time difference stand out more, in that you're comparing a 2-4 week window of exposure to covid with a 1-year mortality rate - covid is going to be the biggest risk factor over that window but then reduced for most of the rest of the year depending on reinfections.

I've only looked at it briefly but the only other thing I thought was:

I think you probably mean a detected infection, don't you - in that you're comparing CFRs not IFRs with the mortality rates. You could say a serious infection, in that probably the infections not picked up as cases tend probably to be less severe, but that's not certain to be the case so it's a bit lazy as a categorisation. 

You’re bizarrely cherry picking points to try to discredit me/ what I’m saying rather than addressing my points. I’m really struggling to understand what point you’re trying to make and why this whole exchange is taking place - what do you disagree with?

Either address what I’m saying or give up.

And just as a reminder what I’m saying is:

1. We must reach sufficient population level immunity and this can only be achieved by vaccination and infection.

2. We can change the speed at which we travel to the “end of Covid” but the number of infections required to do this is likely to be broadly the same regardless of the speed.

3. Interventions that reduce the R rate below 1 need to be permanent if you are to avoid these infections.

I noticed you had the common cold on your list of airborne diseases. How are we doing at curing the common cold again?

I hadn’t but it looks relevant and useful, thanks.

Absolutely; third bit last paragraph talked about this.  Annualising the covid risk is going to depend on the demographic mean time between reinfection to detectable levels post double vaccination.  I’m in effect taking a punt that that’s not going to be substantially less than a year.

You are entirely right and I should have been explicit on that, thanks.  I do indeed mean the risk associated with a detected covid infection.  I definitely should have sent this post to you and minimike before deploying it for review!

It’s a real mess to understand, as there’s a significant fraction of cases where the gating for detection is an LFT result not a symptomatic threshold. 

A tweet popped up on my feed today: there are more people in ICUs and in hospital with COVID today than 12 months ago there was a link to ons datasets at the bottom, which I didn't bother to check, so I can't say if the tweeter is bullshitting or not, but given we have 50k cases a day it seems at least feasible.

This is NOT, REPEAT NOT OVER.

Yes - I had noted it was there, I was thinking more that the way you would express it in micromorts ("X per day for 4 weeks; 28x over the infection window" versus "Y per day, ~28x per annum") versus  might  push it to the foreground since it would effectively be in the numbers you're being asked to compare rather than being part of the explanation of them - less easy for others to miss.

I suggest you take a look at plot 7.2e which is right at the top of the thread.

Pro-lockdown misinformation… that’s a first for me.

Far too simplistic! With higher population immunity, the virus spreads slower. There's not going to be day when we reach herd immunity, because the virus mutates and immunity wanes. Yes, we need good population immunity to prevent exponential growth. But no, we won't get to a point where covid goes away because we've all had it. The pool of infectible people is in constant churn, and people can get re-infected, and the virus mutates...

The end of covid is not going to happen because we reach herd immunity like in the computer models with the different coloured dots. Let go of that belief!  I hope that the virus mutates to less deadly variants and with those endemic in the population we can gradually forget about it. Or, we just have to cope with a pretty serious endemic disease, controlled by constantly updating vaccines and doing regular boosters. Given that this is new on top of flu and everything else that the NHS already struggles to cope with, we're going to need more capacity, and probably changes in behaviour to slow the spread of airborn disease (more WFH, wearing a mask when you've got sniffles, ventilation, blah blah).

"The end of covid" is not a thing. The virus has presented new challenges to the way we live which we have to adapt to. It's not the end of the world, many of the changes like WFH have side-benefits.

The pool of infectible people is constantly changing, the virus is constantly changing, there isn't this static number of infections needed after which we get to herd immunity and everything's OK. Interventions that reduce R to below 1 are needed at times when the country can't cope with the level of infections. Not permanently. While the restrictions are in place, and cases are in decay, people are getting immunity through vaccination (boosters, crucially), so this reduces the amount of harm suffered per increase in population immunity, all the while swimming against the tide of waning immunity.

Your whole understanding of the situation is bunk!

I agree, it’s bad form to produce a plot that can be accidentally misunderstood or wilfully misrepresented and for that I must fall in my sword!

In terms of your other point, it got me thinking more - one way we may be missing some vaccine efficacy is if it’s making more infections fall below the case detection threshold.  In extremis, if the severity of infection vs lethality looked like a Heaviside step function, the vaccine could be saving almost everyone, but the CFR would not change.  We know we’re nowhere near that extremis but there’s still scope there.  I’ve been holding out hope for a proper analysis of the ONS serology datasets to look at the time varying rate, of course it’s conflated with other measures.  Perhaps this is an area ZOE can contribute to.  

Soemthing missed by my current plots is that ITU occupancy is running much higher vs general occupancy compared to last winter, so the top level occupancy plots don’t convey what is becoming the more prominent limiting factor.  Which is ITU occupancy.   The Lissajous plots get this across but I’ve dropped them for now for unrelated reasons.

ITU occupancy is higher than this time last year for England (say), but this time last year cases were rising exponentially on short doubling times and generally continued to do so for two months; this year they’re not showing much inclination pandemic level exponential growth.  So whilst I don’t question the statistic, nor the human implications of it (none of them good), I sure as hell hope that Covid ITU occupancy doesn’t have two months of rise ahead of it like we did last year.

Nowhere near over, but hopefully nowhere near last year’s situation either.

Sorry - but you can't think about risk without considering factors beyond whether you are dead in 28 days.  How many serious diseases kill everybody that's going to die from them within 28 days of diagnosis?

Given that the headline daily death statistics do not capture outcomes like death after a long illness, serious reduction in quality of life for a significant time or long term damage to organs as a result of the virus or ICU treatment super detailed analysis of deaths as a proxy for risk is 'cherry picking' which makes Covid look less serious than it is and England less f*cked relative to other countries than it is.

I agree that England is getting less death per infection because, as well as the vaccinations an absolute f*ckton of people have been infected and have some immunity.   What isn't quantified is how many of those people who got infected and didn't die get long Covid or some other negative health consequence which will mess them up later in life.

The sensible thing is to focus on cases, keep them low, and quantify all the health outcomes before rushing to say its safe enough to let everybody catch it based only on how many people die within 28 days.   The statistic of interest should be what fraction cases are fully recovered within 28 days rather than how many cases are dead in 28 days.

And, as you say, there are better treatments and new vaccines like Valneva on the way.   There's absolutely no rush, the countries which hold off mass infection and keep case rates down will get through with less death and other health consequences.

No there’s no going to be a Big Bang end date. We won’t wake up one morning and suddenly theirs no Covid. However we will most likely transition to a phases where immunity is topped up by regular reinfection and boosters. Covid will put less pressure on the healthcare system and life will go broadly back to normal.

The computer model is probably more realistic than your view. I take real issue with the idea that Covid will mutate to become less deadly - do you have any evidence of this happening in the past.

Once we have sufficient population immunity that pressure won’t be on the NHS (it won’t be binary, but it will slowly fade). I’ve not seen any evidence that new novel diseases put any additional net pressure on healthcare once they become endemic. I.e. 20 years on from Spanish Flu there was no additional pressure on healthcare (ignore the war).

Yes some of the changes have been nice. I’m a lot fitter than I was 18 months ago, but I think we can agree that the overall societal impact has been horrendous. There will simply no need to put these measures in place once we have sufficient population level immunity.

How do you see boosters working? From your previous post you seemed to imply that you think it would prevent infection.

Best case scenario is that in most people it gives slightly better immunity than the original two jabs - that’s best case - it might just bring it up to previous levels, or you might never get back to the previous high point. The immune system is incredibly complicated and more vaccine isn’t always going to make things better. There’s a reason why when you were originally jabbed they didn’t put a shot in both arms and give you 4 doses.

My problem with your whole argument is that you imply Covid is some special disease, and not just like every other novel pathogen humanity has been exposed to.

Think about Spanish Flu. Why aren’t we still wearing masks and avoiding dance halls to protect us from that?

I think we actually agree on most things. Where we seem to disagree is everyone will have high risk of catching this covid variant (or an equivalently lethal mutation).

Most of the middle classes, who were being careful, avoided infection pre vaccination and the large majority of those shielding, not in care settings, have continued to avoid it throughout. That latter group is a few million. Then we have the unvaccinated who are scared enough to be careful. I see no reason why it's inevitable that nearly everyone will end up with high infection risk. Yes we will see higher levels than we might want but if case numbers drop a lot, a significant proportion might escape high infection risk (which would be good, as otherwise we still face of the order of a hundred thousand more deaths from the vulnerable and unvaccinated without infection immunity).

I linked to the Spanish Flu Samoa case earlier (Wikipedia): one island had carnage when isolation measures failed but the neighborhood island kept the flu away in that wave...when it eventually hit, the lethality had massively declined. I'm not saying this will happen with covid but it could. It could also get worse, with vaccine avoidance, in which case the countries with strong trust on clear messaging and established systems will fare much better (that's not going to be the UK). In addition I'm sure we agree that further medical advances will happen and that winter is no time for the NHS to be heading into high pandemic activity. So yes, we are where we are, but that includes a wobbly NHS with knock-on factors to other urgent health work, which could end up being really bad. So much so that I think this government is capable of making things worse again compared to our neighbours, despite higher population antibody levels that should give us an advantage.

We shall see.

Just checked our world in data.

21/10/2020

7328 hospital 659 icu

17/10/2021

7145 / 717

So 12 months on and hospitals are under the same pressure. 

It's not bizarre to cherrypick obvious bullsh#t that sticks out like a sore thumb.

I disagree with the two points I listed as being factually incorrect.

Herd immunity depends on R.
R depends on human behaviour - at its simplest, if you don't meet anybody you don't spread Covid.
Therefore vaccination and infection are not the only way herd immunity can also be achieved.
Therefore the word only is factually incorrect however much you dislike permanent anti-Covid measures.

There's another reason I disagree. It pre-supposes herd immunity is possible.

https://www.nature.com/articles/d41586-021-00728-2

We not only can, we have changed the speed of the spread of Cholera in the UK. The number of infections (zero or close to zero) is very different. The same might apply to Covid since herd immunity may be elude us.

Given that herd immunity may not be possible, permanent suppression of cases may be the best way to preserve average lifespan. Eg catch Covid ten times in your life and die early or catch Covid five times in your life and don't die so early.

Not as good as we are at eliminating measles & TB in the UK as major health threats.

How were we doing at getting to the moon before we got to the moon or cloning sheep before we cloned a sheep or flying helicopters on Mars before we flew a helicopter on Mars?

The whole 'we've not been able to do X in the past so we won't be able to do it in the future' argument is total bullsh*t.   

Nor have I, and nor do I.  You might notice that I also linked to a discussion over risk of hospitalisation vs flu as well.  Are you reading what I'm writing?  Because you almost invariably cherry pick parts of what I write to reply to. 

We have a separate measure of deaths where-by Covid-19 is mentioned on the death certificate.  It does not make much difference.

Are you suggesting the pathology of all serious diseases is similar?  That's a big claim.

You will note that I did not use this risk analysis to make any comparisons to other nations Tom.  You seem to be accusing me of doing something abso-bloody-lutely did not do.

I went out of my way not to do so, and only to look at relative risk of death between different ages within England, as a way of proving some thought and discussion over understanding what sort of order of magnitude the risk of death is for people in England if they catch Covid. It's not cherry picking for comparison "relative to other countries" because I did not make any such comparison, nor would I.  

I kind of trust people reading this to not be so terminally dumb that they think the only outcomes of catching Covid are "rapid, full recovery" and "death".  This was directly acknowledged by considering going to hospital - something that is clearly not good.  There is as you say also the long Covid risk.  If you put 10% of the effort in to producing a clear collection and interpretation of data on this, that you put in to misrepresenting posters and being ant-English we might have some clear, compelling data to discuss on that.  Nobody else has been minded to put much together, and my time on this is bounded.

Have you looked at my plots mid-thread for the USA?  They're keeping cases low - and falling  Do you want to be like the USA?  Hint:  Look at their death rates.  I do not want to be like the USA.  My point has been and remains that cases remain an extremely poor basis for international comparison (their comparative rates of change hold more meaning however).  I don't think this point is remotely controversial.

In the long run yes/ maybe.

In the future we might be able to custom build immunity to any disease. We might cure cancer. Time travel might be possible - although the lack of women/men from the future telling us how to cure Covid is telling.

However do I want to wait around for unknown amount of time for mankind’s scientific prowess to possibly do its thing - not really.

An outdated view at the end of the day.

The key point from the article you linked is:

“Long-term prospects for the pandemic probably include COVID-19 becoming an endemic disease, much like influenza.”

i.e. just like every novel pathogen before us, eventually it becomes endemic. I keep saying this, but Covid isn’t special - it’s just like all the novel pathogens we’ve faced before.

Think how Spanish Flu ended, think how other Flu pandemics ended. Covid will end in a similar way.

So you were opposed to restrictions while the current vaccines were being developed? 

Yes, but we'll need increased capacity in the healthcare system because it was already struggling without covid.

The computer model is just closer to what you believe. That does not make it any more realistic, just more comfortable for you.

I've no idea whether that will happen or not, that's why I said "hope". But my understanding is that the Spanish flu which was deadly when new is now just one of our seasonal flu viruses, presumably after many mutations. 

What? There is no comparable record! The world was a different place 100 years ago. Spanish Flu killed 100 million people! With covid, we've got modern healthcare including vaccines! What comparison are you trying to make here? 

Your just repeating the "we'll get to herd immunity and it'll all be alright" line. It depends what happens as the virus mutates, and as vaccines and healthcare respond to the ever-changing situation. Whether or not control measures will be needed in future depends on a whole load of stuff we can't predict.

This is more straw man rubbish. I'm making the assumption that boosters restore immunity to something similar to the 2nd jab, as shown for the Pfizer. It's bizarre to me the effort you're putting into talking up how great it is when everyone gets covid, and talking down what can be achieved by delivering boosters. 

Do you get paid for everyone who dies of covid or something? Or because the booster programme is going tits up, are you defending this by implying "it wasn't worth doing properly anyway"? What's going on with your pro-infection nonsense?

Covid is special at the moment because it's new. It won't be special when it isn't new, when we've adapted to it. With every disease we've encountered, we've made changes so that it's not causing undue trouble in society, such as requiring vaccination in childhood or for travel to certain countries, or having treatment available, or learning how not to transmit it.

Because it mutated into seasonal flu? And we have set up our modern healthcare system to deal, albeit inadequately, with that endemic disease, including by vaccination? 

Is your argument really, "look, Spanish flu is over, so covid will be soon too, let's just ignore it and then we can all go back to normal"?

Of course not - but coming up with a way to eliminate Covid in the UK is significantly more challenging than producing a vaccine.

The poster I was taking issue with was saying that we’ve eliminated TB (in humans) so why can’t we do the same for Covid. 

How long do you reckon that would take - 10 years, 20 years, longer?
 

We've almost had the full house on this thread. We just need an argument about the efficacy of masks and Tom to mention the f*cking tories.

I think we probably all agree the UK got it wrong in the past.

I think we probably all agree covid is going to become epidemic.

I think we all agree that some additional measures in the meantime are appropriate if necessary to make sure we keep a good margin to healthcare overload and minimise the number who get seriously ill because their booster hasn't arrived on time, and that because the booster programme has been slow and case rates have recently increased significantly, this has become more likely.

There is difference of opinion over whether we should already have, or should now, implement these measures. But I don't think (?) anyone is proposing that we keep those measures permanently.

Much of the rest is just noise and people wanting to paint things in their own terms.

Initially by changing and in many ways improving the way we lived? Particularly the poor and vulnerable. 

? 

Don't know, but considerably faster than for TB didn't it take until  the 1960's for significant medical interventions. 

Last post - I promise - banging my head against the wall here.

What evidence do you have that it won’t work like this? I’ll give you a clue - you have none. I have plenty of proof - results from previous pandemics and current modelling. Not perfect, but better than your evidence free assumptions.

This might blow your mind, but there’s no evidence that Spanish Flu mutated to become less deadly - we just have broad population immunity from it. It still kills people mind, people die of Flu, some of that will be Spanish Flu.

The point I’m making is that once truly endemic, healthcare won’t have to deal with Flu + Covid in the sense that it will create additional load on healthcare. It will be broadly the same net workload.

If the virus mutates we’re likely to be better off having natural immunity than the current vaccines as we’ll have been exposed to more viral proteins/ glycoproteins.

Once it’s truly endemic control measures won’t be needed.

I actually think your solution would cause significantly more net harm than what I’m advocating. It’s not nonsense. Most (sensible) posters are now agreeing with my position. Check our respective up and down votes.

This is where we start to branch into fantasy land. How do you learn not to transmit Covid? Work out a way not to exhale?

It didn’t mutate into seasonal flu - the opposite happened. We became immune to it. We have made no special accommodations for Spanish Flu since it became endemic. The same will likely happen with Covid.

Elimination by holding R<1 appears to be a pipe dream at this point, even before considering the less fortunate nations out there or the ever larger number of identified animal reservoirs.

The evidence to date is that new variants are both more transmissive and more lethal.  The lethality is masked somewhat by improved clinical care, improved therapeutics (dexamethasone in particular) and vaccination.  In terms of expecting that to reverse direction, consider the lethality of SARS-nCov-1 and MERS-nCov.  This virus seems to be on the lower bounds of its fraternity.   That suggests rather naively to me that it has plenty of headroom to expand in too.  Right from the very start I argued against Coel ("is it worth it?") against allowing cases to rise because I was concerned about future mutations; that concern has been proven right at least twice, and it hasn't gone away.  Controlling cases within the UK is no longer very relevant in terms of that risk, and it horrifies me that we've paired travel restrictions back so far - now to the point of LFT - as it's such a barrier to identifying importation of the next more transmissive, more lethal variant.  This is one of the reasons I have my "devil and the deep blue sea" dilemma, and why I want a dose of Valneva in particular in my arm.  Not unfortunately going to happen.

Analogies are often drawn between flu and Covid.  It's important to understand that the means by which the influenza viruses change over time and coronaviruses change over time are very, very different.

That and the lethality of other nCovs is why I think your "could" here is at best wistful thinking, and at worst borderline dangerous speculation more akin to what some of the denilaists have argued in the face of repeatedly being proved wrong.  I'd put my faith in the compounds going through the therapeutics pipeline a hundred times over before your "could".

Yes, I've said so enough times, or so I thought.  I'm starting to think it is literally impossible for me to post anything without itemising everything I've said a dozen times before however.  That to me is a classic sign that good faith is being lost in favour of point scoring in an attempt to make other points. 

There are other counterpoints however beyond the ineffable one of future variants.  Another is the ageing process - as time goes by, our individual risk rises measurably month by month as we age.  It's actually quite a big effect I think, but one I've avoided posting about because there's enough doom and gloom about frankly.  The more the spread is delayed, the older people without previous infection granted immunity will be.  At a population level the mean age doesn't change as individuals do, but the younger people growing older will have had infection with negligible health damage and immunogenicity, so there is a symmetry breaker in the ageing process.  

I can see multiple arguments to stall for time, and multiple risks associated with stalling for time.  

At the start of the pandemic it was so clear to me we needed to stall for time.  Now I haven no great clarity, but I do have a developing tetchiness over various posters pretending that this is purely one-sided.  I never took kindly to people advocating we rush in to this early on based on one-sided views, and I don't fairly see how I can now stand by and watch one-sided views put the other way.  For me - as we've established ample times - my natural leaning is highly cautious on this, but I don't deny the complexity or the number of unknowns.

Indeed, which is why there is a rationale for having had a lot of infection happen before winter.  Which means regardless of where one falls in judging the merits of this vs waiting, as we get closer to winter, we need to send cases in to decay out of measured caution, with the uncertainties over flu season looming as well.

Part of the point I have been trying to get across with the differences in where the UK is vs a lot of Europe is that I think it is - uniquely - within our power to send cases in to rapid decay with pretty light touch control measures.  Understanding and evaluating this possibility requires acknowledging the results of the last 18 months here, and recognising that as well as the obvious and clear harms of our approach (on which I have always been clear), there are benefits.  They may not have been worth the price, but if we're not prepared to capitalise on those benefits and use them to control cases when healthcare is in such a precarious situation, what was the point in all that extra suffering, death and economic damage?

I feel that what I'm trying to get across this week - provoking people in to thinking about the whole situation and how much power we - almost uniquely - have to lower cases without dramatic changes - is getting lost in noise...

Totally agree with that post.  Was typing something similar at the same time, is at the end of the post below yours.

In the absence of magic future technology, some will take this as implicit in any suggestions that we push for elimination, and I think that is part of the cross-purposes trench warfare going on.  It is my view that elimination within the UK has, since Alpha, and with current and reasonably expected future vaccines, only been a possibility if we have significant indefinite restrictions on life within the UK and on inbound travel.

In other news, the latest update to the data has English cases back in to decay; the last data point is often a bit provisional but the developing trend has been clear.

I watch this plot with keen interest.

Edit: How am I getting this many typos?  Why aren't I mowing the lawns?  Running out of good days to do so...

You don't have any "proof". There is none from previous pandemics, because the last one was a hundred years ago and killed millions and millions more people, and there were no vaccines, etc, etc. 

What comes up first on google for latest modelling?

https://www.imperial.ac.uk/news/231336/what-expected-covid-19-this-winter/

I don't know what will happen next winter, I haven't heard any predictions from any credible sources, presumably because it can't be predicted. I haven't made a single claim or prediction that you can falsify with your ridiculous ideas of what constitute "proof".

Where are you getting your bunk information - or do you just make it all up yourself?

https://www.washingtonpost.com/history/2020/09/01/1918-flu-pandemic-end/

“You can still find the genetic traces of the 1918 virus in the seasonal flus that circulate today,” says Taubenberger. “Every single human infection with influenza A in the past 102 years is derived from that one introduction of the 1918 flu.”

https://www.history.com/news/1918-flu-pandemic-never-ended

When do you predict that's going to be? I'm talking about needing controls to get through the winter, what are you talking about?

But that possible benefit is far outweighed by the known harms of infections and healthcare overload.

What? Check the moron buttons to find  the truth!? Most sensible posters agree with you!? The NHS bosses agree with me! (Or rather, I believe them because they know what they're talking about, and I think your opinions are totally misinformed and destructive to society). You refuse to even understand what my "solution" is, even though I've spelled it out:

Having on-and-off restrictions which cause minimal disruption, to keep infections below the level where strain on infrastructure causes additional harm. Why is this impossible? What are the "real risks" of this?

What? Don't go into the office with cold symptoms? Test yourself? WFH? Wear a mask under some circumstances? I can't understand what your difficulty is - what do you mean "fantasy land".

Unfortunately, your facts are incorrect, again. It did mutate, and we have made special accommodations, we vaccinate against the 'descendants' of Spanish flu.

It is probably correct to say that if measures are brought back now, compliance would be lower than previously. However political messaging and actual rules still matter. The political messaging has been “the vaccines will see us through” and most people seem happy yo go along with that because it’s what they want to hear. Dropping basic measures like masks and WFH and not introducing vaccine passports was a political choice but it also shifted the dial in public discourse. Stronger messaging and rules would help to shift the dial back somewhat - a lot of people do listen to what political leaders are saying and a lot of people do not generally break the law. These things matter. We might never get the same level of compliance again but every little helps. 

I think when considering *personal* risk, focusing on the risk of death for healthy people under about 50 or even 60 is kind of missing the point. The risk is pretty low on an individual level once vaccinated. What concerns me far more is Long Covid or simply being off sick in bed for a couple of weeks followed by several weeks of recovery. These risks are not insignificant for me as a healthy 40 year old.

It’s not missing the point, it’s making a different pair of points about how a particular risk has changed.

1. For older adults, vaccination reduces the mortality risk of covid a lot, but it remains a stand out large risk factor and a not insignificant absolute risk.  Hence disappointment over slow boosters
2. For younger adults, death is not a large risk and is a less significant risk factor post vaccination.

I’m surprised that showing how something has got better, and how in places it still falls short, without making any claims about other health impacts (which at this point are beyond denial to everyone) is getting such  a confused reception.

As I said to Tom, I credit people here with enough smarts to understand the bounds of what was said.  I’m starting to question my judgement.

You’re the second person to overlook that that post also covers hospitalisations, so no, not “focusing” on death, it’s looking at two different risks.  

Yes, long covid can be bad.  I’m all for a data driven discussion on that; it would be good to hear an analysis of that risk for working aged, double vaccinated adults in concrete terms beyond the ultra generic “symptoms after X unit of time”.  In 18 months of discussions I’ve not seen this; I’ve seen different definitions conflated to make different points and the reality of the risk is not clear to me.  Perhaps you can pull some stats together on this.

As for being laid up sick, is that risk any worse than say flu?  Seems like it probably is going off various anecdotes, but again, what does the data say?

It looks like we’re speeding up boosters now:

https://news.sky.com/story/covid-19-record-number-of-coronavirus-booster-ja...
 

I think the person in charge of the original rollout has been moved back to the vaccination project (can’t find a link though). Hopefully looking a bit better.

No, it is much easier.  What you do is have a completely draconian lockdown where almost nobody is allowed out their house and international borders are shut down for 2 or 3x the infection cycle.  Then you are very careful about how you raise the lockdown and deal with the small number of key workers who needed to be exempt.  

Keeping it eliminated after international travel restarts would be difficult.  But it seems technically feasible for an island to create a New Zealand like zero-Covid situation if it had the political will to do so.  China has got near zero and it isn't even an island.

The way to ‘beat’ Covid is vaccinate as much as possible - ramp up capacity to be able to both donate significant numbers of doses abroad and offer boosters here for everyone, introduce vaccine passports to encourage the hesitant to get vaccinated, plus measures like masking, combined with test and trace, to help keep on top of cases. All underpinned by robust public messaging. This might be required for the foreseeable but so what? It would protect people and healthcare and hopefully ensure lockdowns aren’t required again.   

I don’t disagree with what you say re deaths. Long Covid is a bit of a vague thing at the moment but anecdotally some people get all sorts of ongoing symptoms, from simply annoying to debilitating, for a period of weeks or months. So that’s certainly a concern. Hard to get precise stats at the moment as you say. As for illness, it might be no worse than a bout of flu but I don’t particularly want anything flu like either… It comes down to personal risk perception. The way I see it, my risk of death or injury is already significantly above average for someone my age due to doing a lot of climbing, particularly winter and alpine - no need to add unnecessary risks to the equation…

Only because there is still a considerable unvaccinated minority (plus under 12s, though with time I suspect vaccines will be rolled out to younger age groups). Hence we should be introducing vaccine passports. Italy have gone as far as requiring a green pass for work - as a result, some previously hesitant people have decided to get vaccinated.

You are assuming there is some kind of magic herd immunity threshold. The reality is that due to breakthrough infections (vs birth vaccination and prior infection), there will never be perfect herd immunity. However it’s possible to get to relatively low levels with vaccination (including boosters), some basic measures such as masks and WFH, plus T&T (which becomes more effective when cases are low).

Immunity will fade from both vaccination and infection. I’d rather get my immunity topped up via vaccination.

As for variants, what makes them more likely is large numbers of cases, so reducing cases is a good thing from that point of view. 

Yes, I agree that really good messaging could turn the dial. However unfortunately I don't think Johnson has that in him unless there is some sort of major event that he can point to (he would need to do another Downing St address like he did when announcing previous lockdowns if he is to get enough notice). And certainly noone takes any notice of what his cabinet says.

 A quick look at the dashboard confirms that this is correct. The difference is that a year ago the numbers in ICU were going up significantly faster than they are now. Hopefully it won’t get as bad again.

Agreed… though I do think they will turn to Plan B in a few weeks’ time anyway as they won’t have much choice. 

I think knowing what we know now, many people would have opted for that kind of draconian lockdown instead of continued semi-lockdowns… Whether we would have managed to keep the lid on Covid subsequently is another question. 

I assume you mean those who don’t want to be vaccinated as opposed to those who can’t be for medical reasons. I would posit that those who don’t want to be generally aren’t particularly scared of Covid.

If government were to collect data on negative outcomes beyond deaths and hospitalisations it would influence policy in ways they don't like.  So they don't collect the data.   The risk is they are creating a large cohort of sick people and the cost of that will be lost work days and increased health expenditure.

Should have dug my laptop out and run of a plot 22 update before posting that.

The last 12 days or so of data have seen hospital occupancy rising again after the fall that happened when the barycentre of cases shifted down into school ages.

But, at least so far, ITU levels are running lower than for comparable occupancy levels a couple of months ago.  As remarked on #36, the pre-vaccination lag from general admissions to ITU admissions was not visible when occupancy went in to decay, with no oval opening up to the left of the rising line as in previous waves.  

Very interesting to think about what this might mean.  Probably want another week or two of data to make sure it's not lag waiting to play out, but another tentative sign that the severity of infection associated with detected cases might be decreasing - certainly makes it look unlikely that its increasing.

Edit: I'm assuming here that access to ITU is demand driven and not supply limited.  Given some of what's been discussed here and elsewhere, I'm not convinced how clearly we'd be told if access to ITUs was becoming supply limited, likely from a staffing perspective - but should that come to pass we'd expect to start seeing a corresponding change in the deaths signal which doesn't seem to be the case so far.

That great to see.

Back to the modelling, some interesting observations on the front page of tomorrow’s Telegraph. Cases falling significantly by Xmas would be nice, though it feels rather optimistic. I recall that a couple of months ago you wondered about how long it would take to burn through - hard to answer that without knowing the % of breakthrough cases but perhaps the modellers have access to those data.
 

One thing find I odd is the idea that Plan B would simply delay deaths. I would have thought that vaccine passports would drive up the vaccination rate somewhat and thus deaths would be reduced. Perhaps not significantly so. Of course modelling is just that and it’s not always been accurate in the past but perhaps a glimmer of hope here. Also conscious this is just a summary on the Telegraph front page!

https://www.bbc.co.uk/news/blogs-the-papers-59032831

I'm really not wishful thinking. The three main scenarios I see without any restrictions are as follows.... and this ignores for now any other positive interventions, predicted or not, that might help us from a few months time... (fingers crossed we get good news, given what follows): first, covid starts to mutate to a less lethal virus, as flu likely did in 1918.. so we certainly messed up in the UK (I'd agree on balance from coronavirus history this is the lowest likelihood scenario, especially short term,  but we never had a coronavirus facing vaccines before); second, covid stays about the same infectivity/lethality (maybe most likely) ...multiple possibilities including too many that we could mess up the exit based on past history of poor government responses (but a chance our politicians might accidentally or callously sigh relief, despite more deaths etc), best case ten thousand more deaths over the winter and more than a million still vulnerable; third scenario, covid becomes more lethal/evades immunity ... so lockdowns ensue and we in the UK messed up big time, as the political and public confusion caused by the messaging and structural chaos will take too long to sort out, causing critical delays (and our state is more weakened than neighbours by the hit of overlong previous lockdowns, other poor public health decisions, and NHS stress partly due to previous stupid govenment delays). Not much good in any of that! Plan B style restrictions seem essential to me and as dumb as I think our government are I expect them to 'blink' and impose them in the next month. Even then the outlook doesn't look great.

I'd prefer we didn't guess, given the uncertainty, and just do the best we can now with what we face now....caution being a priority given the state of the NHS. Save thousands now a month (covid or otherwise) with some limited Plan B style restrictions and hope. If I'm wrong and things improve despite all indications to the contrary (especially unlikely looking in the NHS), Plan B will just accelerate that.

It's plain nuts to me we didn't implement plan B when SAGE asked for it based on health system concerns (not big modelling increases). They don't ask for trivial reasons and it's a perfect 'get-out-of-jail-free-card' for Boris. If anything the situation in the NHS is possiblity worse than predicted, I'm well connected and things look bad to me, partly hidden from the public by government pressure on negative news. The NHS won't fail, but in too many trusts it might have to 'turn off' too much other than covid, at the worst possible time. I still think Boris doesn't really believe his scientific advisors and took too much confidence from the unpredicted drop after the football, when big further case rises were foolishly predicted by SAGE modellers. We seem to me in the UK to be trying to snatch a defeat from the jaws of (an already too sad) potential victory based on a good deal of luck on vaccination. A weird irony is we now have A/B testing  in the Telegraph (etc) of the possibility of plan B after COP.

Shock Horror!  the Telegraph finds the most optimistic hint of news to support its regularly  flawed scientific position (in direct contradiction with publicised SAGE advice to implement Plan B).

It quotes John Edmunds and refers to specific modelling, so I’ll give it some credibility this time.

On a different point, some early studies on effectiveness of vaccination vs prior infection again Covid (about the same). https://www.theguardian.com/commentisfree/2021/oct/24/which-protects-you-mo...

I’m really not. You’re an elimination fantasist who expects that artificial interventions will keep the R rate below 1 so parts of the population don’t need immunity to Covid. This simply won’t be the case. Soon enough we’ll stop testing for Covid - that will really blow your mind when you can go to the pub and there’s no way of knowing who might  have Covid or not.

Yes it will fade and Tough. Even with a booster you might still be susceptible to infection. In which case it’s a top up by infection for you. 

We’re a drop in the ocean compared to the rest of the world when it comes to variants. I would strongly argue that natural infection (with prior vaccination to mitigate the health impacts of infection) is the best way we have of dealing with variants.

It's really weird how the behaviour of that Lissajous plot has completely reversed when compared to previous waves. Anybody have any theories as to the cause?

To be clear - which I already thought I was - I applied "wishful thinking" only to your claim that there's a reasonable chance the virus could become less lethal (requiring a less lethal and more transmissive variant) and I gave a series of reasons why I am cautious against this idea, and why I don't think analogies with flu are so relevant here.  There’s also the link between viral load, transmissibility and the lethality mechanism to consider here…  

One other part of your post stands out as why I think we have such a problem...

Government should have followed sage!  Government shouldn't have followed SAGE!  As you'll recall I've been very unhappy with the modelling component from the start, and I remain so.  It's not setting the politicians up to succeed when we want them to listen to SAGE some times and ignore SAGE at other times.

If SAGE put half the effort in to understanding the data in real time they put in to modelling, they'd have spotted the football (It's to my shame I didn't look at the gender ratios until after the growth collapsed; the parallel was there from Scotland a few weeks before, and presumably contact tracing was giving some hints...?).

To be clear I think the government need to listen to both SAGE and NHS management over their serious and growing concerns over provision levels over winter; I also think that needs a much more holistic solution than just "less Covid" as even if we take Covid away completely now, it seems to me like RSV and/or Flu could take us back to the same worrying state, and that's before the pressures of the winter holiday period land.

But, when your and my opinion is that sometimes they need to listen to SAGE and sometimes specific SAGE outputs and the comparison vs reality are confusing and best ignored (although they're not, with the Telegraph and Spectator making maximum hay against control measures from modelling disparities which we have to recognise brings political pressure to bare), we have a problem.

I was hoping to get some comments on this last time around.

What it means is probably that the lag from hospital admissions to ITU admissions is dramatically reduced.  But that just proxies your question: what’s the cause of that?  

Why best wild guess is that we've got two cohorts going in to hospital -(1)  the double vaccinated who are generally less unwell and aren't going to ITU, and (2) unvaccinated and otherwise medically vulnerable who are going to ITU much sooner.  

Many of group 1 would previously have gradually degenerated in hospital until ITU admissions but now recover and leave; I think this new behaviour was slightly too soon to be associated with the new -mab therapeutics but I need to check the timeline there.

Why group 2 are going in to ITU much faster is a bit of an open question.  Selection effects perhaps, and also perhaps the likely increased lethality of delta over alpha although that doesn't seem like enough in itself.

Feels like a pretty big change to the behaviour of hospitalisations so I still hope to get some informed comment on here, but it's a bit buried in the noise this week; I'll put plot 22 in the headline plots next week...  Will be interesting to see where the leading edge is by then.

I'm sure the modelling work they are reporting on is genuine, but: firstly the point has been made several times in the last weeks that it's never been harder to reliably model, secondly models haven't done so well even when it was easier, and thirdly Vallance (of all people) is making cryptic pubic comments about when you need to restrict, going earlier and harder than you think is important. That's all ignoring any Telegraph spin... their journalists look like they have been sucking lemons all year when commenting on covid, having to admit we are still in trouble after the paper gave support to all sorts of unscientific nonsense, especially GBD style herd immunity that proved to be very wrong in 2020.

During the football bulge SAGE wasn't publicly calling for plan B, just some SAGE modellers were urging caution..... since then NHS pressures have worsened significantly and the SAGE position has changed.

In scientific terms it must be difficult advising govenment when much of Plan B should have been Plan A (especially masks, as it has zero economic cost) and when the health minister seems intent to go to war on multiple fronts with health staff in the middle of a crisis (threatening to sack senior managers and GPs who don't meet flawed targets, and coming soon woe betide staff who won't get vaccinated).

On the lissajous I'd be interested to hear more but the hospitalisation demographic has massively changed during this wave. The rumour mill indicates significant therapeutics planning is happening,.. but I've not heard any clear details yet.

whatever is happening in the SW isn’t going away.. but also isn’t spreading fast to other areas. Thoughts?

I think it's in to decay, looking at the regional cases data on the dashboard.   The provisional zone of an updated plot 18 agrees (for now...)

So far, no sign of any corresponding event in the downstream signals...

Couple of other plot updates - the decay in top-level cases is looking more convincing, although sometimes there's a drop related to testing behaviours and not changing infections around the end of school terms and half-terms, so keeping an open mind for another 7 days of data.

One ignominious update - the Covid occupancy in English hospitals has surpassed the last peak back in early September and is now the highest its been since early March 2021.

Good article on the latest LSHTM modelling. It doss seem on the optimistic side for the short to medium term. I suspect you’re right that the drop in the last couple of days is at least partly due to the end of the half term. 
 

https://www.bbc.co.uk/news/health-59039739

I hope I'm wrong but I think its a work of fiction. 

I'd make several points:

Firstly cases are not the main issue, hospitalisations and deaths are, and the very bad news is we have less leeway in the NHS than since the peak of the first waves. This is partly due to covid but mostly down to huge demand, low staffing, high sickness and isolation absence, and bed blocking as the care system is really struggling again.

As the article points out there are 5 million unvaccinated but that's not as bad as it sounds as that number is distributed to the younger end of the demographic and ONS antibody surveys indicate a good proportion likely have infection immunity. My concern about hidden populations is for those shielding, certainly a million plus: I don't expect them to drive numbers so much but it is sadly a group who will have a higher CFR than the average so the outlook is very worrying for them until case levels become low again. There will be a group of vaccine refuseniks shielding as well, likely tens of thousands and with an older demographic; I didn't initially understand the religious objections: some have a red line around use of vaccine test cultures with a foetal cell ancestry (particularly in some Catholic groups, in my view bordering on cults).

The big worry is behaviour... currently more people are starting to think about being more careful.. but as soon as cases start to drop (as I expect is already starting) the tendancy will be just go back to increased mixing. That's why I expected case oscillation back after the football peak drop and a more steady high hospitalisation level and ongoing 100+ deaths a day.... not that its any comfort being right on that. I think xmas is a bit of a red herring as people are mixing a lot with relatives compared to last autumn so that xmas hit should be much smaller. I see clear behaviour sub-cultures... I went to an indoor shopping mall last week...maybe a few percent mask use and little respect for social distancing... go to a indie cinema/bar most have masks and respect distancing. There is a gulf in mask use and distancing behaviour between Aldi or Lidl around here compared to Waitrose or Sainsburys

Thirdly all the advice about flu is it's coming to some extent and the combination of covid and flu is very likely more lethal than the sum of its parts

I've just been looking at a bit of dashboard data and it's interesting how variable the behaviour around the country has become again over the last few weeks in this latest rise. Forgetting the south west - I mean other places too. London is a known outlier at regional level but at utla level lots of other cities are behaving similarly. So I would suggest that perhaps the London cases over recent weeks are more typical of those in cities generally. I've attached a bunch of examples (alongside the graph for London.) 

Anyone else love to see the details of this latest treasury leak (on the economic cost of plan B being £18 billion for 5 months) fact checked by More or Less.? Of the order of two percent of GDP for working from home when you can and covid passports..... really????

https://www.theguardian.com/business/2021/oct/26/covid-plan-b-uk-economy-co...

What's certain, if pressures continue on the NHS such that backlogs continue to grow, that it will cost billions, ignoring the cost to business of people waiting to become treated to be well enough to work again.

https://ifs.org.uk/uploads/Green-Budget-2021-Pressures-on-the-NHS.pdf

I can imagine them handing £2bn to their mates for some blue/black covid passports with a Union Jack on them. The rest sounds unlikely.

Agree with most of your points. I wouldn’t call LSHTM modelling fiction but models have been wrong before (recently on the pessimistic side) and by their nature models can never be spot on. Indeed they stress that there are a lot of unknowns around behaviour and vaccine fade. Behaviour is what seems to have made the difference since Freedom Day (hate that term), hence actual outcomes have been better than modelled (but clearly still not good). As you say, behaviour will change with time in response to the news etc.

I do think cases are a useful indicator as they drive everything else. Clearly demographic and other make up of the cases is also key but 5k cases is generally going to be a lot better than 50k. If (big if!) the modelling proves to be about right, the NHS could get some much needed relief from Covid over the winter. What happens in spring and summer is still important but if we have to have X number of cases (VS’s argument), clearly it’s better to displace them into the warmer months.

I’m not convinced there are that many unvaccinated people shielding. My assumption is that the vast majority of the unvaccinated couldn’t care less about Covid risk. You are right about some religious groups but even the Pope said it’s fine… What we need is vaccine passports to encourage the unvaccinated to do the right thing.

Couple of hours in a cinema with masks vs shopping without masks - that’s an interesting one. I suspect the cinema risk is at least as high. I suspect that people aren’t present in shops for long enough in sufficiently close enough for that to be a major vector. Getting public transport to go shopping would be a greater risk. Bars and cinemas will be risky, masks or not, due to length of exposure, unless it’s nearly empty. Obviously masks make sense in all indoor environments. 

WFH does have an economic impact due to nothing being spent on commuting and a lot less on lunchtime sandwiches and corporate entertaining (whether formal or just going for a drink after work). I’m not convinced vaccine passports would have much of an economic impact. Reduction in unvaccinated people spending money in restaurants etc could be offset by vaccinated people being happier to go there. Either way, an economic impact of £18bn is  rather immaterial in the scheme of things. 

Thanks again for all your hard work. I honestly keep telling people that these treads are the best source of Covid information around. I'm just sorry I don't have enough knowledge or time to contribute.

Nice to see that cases seem to be going into decay but part of me feels we have been here before. I really do hope that this time we are reaching some sort of herd immunity and cases will go into sustained fall.

Anyway, as it's still looking a bit scary out there and I am eligible, due to a voluntary role, had my booster yesterday.

Keep up the good work

Very crudely, it’s been noticeable on the dashboard map. This time round,  less urban areas have led the way in case growth. The blue areas have been shrinking and are now limited to areas of very high incidence in at least one previous wave (Birmingham is an exception as I don’t think it’s had as high incidence as the other places but I may be wrong there - but we’re always in the middle on everything!). Of course engagement with testing could also be a factor.

The covid & flu combination is literally a double whammy.

"Covid-19: Risk of death more than doubled in people who also had flu, English data show"

https://www.bmj.com/content/370/bmj.m3720

It's really a stand-out difference just squinting at the plots you've posted.

When something approaches a tipping point, small differences in immunity levels (or, some equivalently, timing) can manifest as large changes around the tipping point....

... perhaps?  

In reply to Andy DB:

Thanks - it keeps me off the streets...  

I'm not sure it's "herd immunity" by the strict definition of driving R<1 for infections, but I think with rising immunity level, the daily rate of infections severe enough to get detected should go in to decay real soon now; quote how that plays out with a significant chunk of PCR testing being triggered by symptom-free LFT results...  That probably puts a floor on cases; but if this all pans out like that, hospital admissions should have dropped a lot by the time we get to that floor.

So far I'm resisting dusting off my optimist's cap because, as you say, we have been here before.

All eyes on tonight's cases number and the week-on-week plot...

AY.4.2 got its VUI badge

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/...

I don't see it as a bad thing if money that used to be spent on low tech services like transport and pubs/restaurants gets diverted to phones, apps, computers, network equipment and software.   Commuting is a waste of time and energy and expensive office buildings are a waste of corporate resources.   

More interestingly working from home and communicating via e-mail / instant messaging has an effect on power dynamics at work.  Having a physical presence / expensive suit / loud posh voice has much less effect on a Zoom call or e-mail exchange than face to face which isn't good for the careers of people who have made their way by bullshitting.

Interesting developments in all English measures over the last few days.  It's all in the noise for now,  but it's all off to the same direction.   Perhaps it's not just all good things that have to come to an end?  Will be very interesting to see where we are by the next thread update.

Just an updated rate constant plot for now; the third day in a row of week-on-week decay; small rate constant (long halving time), but applied to a very high number so significant reductions in cases and hopefully therefore in "locked in" hospitalisations.  Some risk of conflation from changing test behaviours over half-term.

Tentative musical interpretation....? -  youtube.com/watch?v=n4RjJKxsamQ&

As usual I'm not very optimistic - Covid doesn't seem to be going anywhere.

It seems we've been "on the threshold of an improvement" for the last few months with no sign of it happening.

Looking at rates constant plot, R hasn't gone less than one for a halving time or more since July.

This ties in with graphs posted by Si H above (& URL below) which don't show significant decay/growth in Aug/Sep/Oct (changes are less than a factor of two).

https://www.ukhillwalking.com/forums/off_belay/friday_night_covid_plotting_49-... 

Mind you, I never thought I'd see the fall of the Berlin Wall without war!

I’m trying the optimistic hat on for size; it’s got a one month returns policy.

I think it could have done in august for true infections, but the change of testing behaviours over schools returning masked this from cases, perhaps.

Undeniably.  Yet ongoing infection and vaccination continue to evoke immunity, and things are changing in the relationship between measures.

We’ll see.  If we don’t get a sustained decrease in hospitalisation as a result of current policy, it’s time to make some changes now.  

I've been looking for a better plot to use for comparing the situation across Europe.

This one shows the cases / million people / day on the X-axis and the rate of change of cases expressed as an exponential rate constant (left X-axis) and characteristic time (right Y-axis).

This is using Ourworldindata's GitHub download.  I use a 7-day moving average as a lot of the nations have a much larger day-of-week effect in their reporting than the UK.  Even so the data seems to be very noisy (partly down to currently low case rates as well).  The plot shows the most recent point in the data as an 'x' with the two previous weeks marked as trails.  Rate constant is measured by the change over a 14-day period (longer than my usual 7-day period to reduce the impact of noise on the measurement).  Edit: Colour is purely to help discriminate different lines and their labels, no deeper meaning.

The use of a 14-day window for the rate constant means the rate constant values on the plot are more lagged than reality.   

A picture tells a thousand words, as they say.

Oxford Vaccine Group boss claims it's unfair to 'bash the UK' over high Covid cases

The director of the Oxford Vaccine Group has claimed it is unfair to “bash the UK” over high Covid case numbers and compare it with the rest of Europe because of high levels of testing. 

Prof Sir Andrew Pollard, who helped create the Oxford/AstraZeneca vaccine, told MPs it was true the UK had high case rates but this was “very much related to the amount of testing”, reports PA. 

Daily cases in the UK have been averaging at more than 40,000 for over a week. 

He told the commons science and technology committee: 

If you look across western Europe, we have about 10 times more tests done each day than some other countries, this is per head of population.

So we really have to always adjust by looking at the data ... we do have a lot of transmission at the moment, but it’s not right to say that those rates are really telling us something that we can compare internationally.

He added:

If you make the adjustment of cases in relation to the rates of testing, and look at test positivity, currently Germany has the highest test positivity rate in Europe.

So I think when we look at these data it’s really important not to sort of bash the UK with a very high case rate, because actually it’s partly related to very high testing.

I’m not trying to deny that there’s not plenty of transmission, because there is, but it’s the comparisons that are problematic.

Timely crossing over of posts their kirsten.

I feel like we've heard some of those comments before in these threads.     

Various members of the Oxford group have had very interesting things to say about the developing immunity situation in the UK and where they think we'll be by the spring as well.

That latest plot (Europe) is startling.

Thank you again, certainly the best source of covid info around.

Well: “Coronavirus restrictions are to return to Belgium on Friday, just weeks after they were relaxed, amid rising cases and hospitalisations. The rules include wearing face masks in public places and mandatory masks for staff at bars, restaurants and fitness clubs. Covid passes will also be required to enter.”

My other “home” is down near the bottom of the European league table while all around it crash and burn, sure there’s no correlation to the fact that it’s still nigh on impossible to get a pcr test 🤷‍♀️

Great to see things moving in the right direction for a change! Some data issues around Welsh cases per the dashboard but not material. Whether this is a sustained fall or just another downhill section on the rollercoaster remains to be seen. We’ve been given false hope a couple of times before in the last 2-3 months. That said, any reduction is a good reduction. 

Germany 27 deaths   83 million people

UK 263 deaths  67 million people

Belgium already decided to put some restrictions back.  The EU countries are behaving like rational organisations with a feedback loop and their cases/deaths are not going to get out of hand because they will do something before that happens.  The Tories are open loop.

My feeling is the curve is not pure exponential any more because of the vaccination and the number of people who have already caught it.  It's more of an S curve and the UK is getting near the bit at the top where it flattens off.  Countries with low rates are further down the S so they can have higher rate constants.   Letting the case rate get so high you are limiting out and getting to the flatter part of the S is not smart.  

 In reply to tom_in_edinburgh:

There are none so blind as those who will not see, except perhaps those who also cherry pick.

Looks to me like you’ve cherry picked some data with weekend reporting effects in it working one way for one nation and the opposite way for the other, or some other glitch.  You haven’t given a source for your numbers (poor form, that) but if suggests a ~12x difference in death rate.

The actual rates are about 2.6x apart(see screenshots below). 

Incase I haven’t made myself sufficiently clear before, let me amplify my comment so there can be no further confusion.

Your endless use of cherry picked data is an absolute f*cking disgrace and is not better than the conduct of the covid denialists or other liars and trolls out there.

Every time you have a valid point at the core of your message you end up no better than a troll.

I’m not going to engage with the rest of your post because it’s built atop lies.

Tell that to the people who work on public transport and in hospitality.

It’s interesting that you apparently hate the Tories so much, because some of your statements are right out of the Tory playbook (this isn’t the only one - remember the debate about Irish GDP?). I’m not sure you actually hate the Tories as such. You just hate Westminster and the ‘English’ generally (whatever ‘English’ means to you - for the record, I live in England and consider myself British but not English because I wasn’t born in England).

You do make some good points and agree with you sometimes but your cherry picking of stats and attempt to twist everything in an anti-Tory/English narrative is distasteful and only weakens your case. Don’t get me wrong, like many people on this thread I’m no fan of the Tories and I’m not trying to defend them but they aren’t the explanation for everything under the sun.

I typed 'germany covid deaths today' into Google and copied what Google put at the top of the page.  That's it.   I don't spend hours on this stuff.

Google actually used the same source as you but they gave me a daily number where you used a 7 day rolling average.  It isn't exactly 'a f*cking disgrace' or 'liar and troll' level crime.

Funny that. Did you also mistype '217' as '27'?

To  be fair - the same chart does have '27' for the previous days data (25'th October) so I don't think it was a typo.

But exactly the same tooltip produces a 7-day rolling average of 61 - which would make any serious analyzer of the data understand that the 27 has limited use for comparisons.

And if you were to type in 'uk covid deaths today'  the comparable figure is 38. Not 263. You get 263 by selecting data for the 26th October.

So not only has the sensible comparison metric of a 7-day rolling average figure, available with as much ease as the figures used, been avoided, data has been cherry picked across two different dates. This displays the weekend sampling biases, and nothing meaningful about comparison rates for the UK and Germany.

An equally valid (i.e. not at all valid) comparison would swap the sampling figures and come up with UK - 38 deaths, Germany 217, and conclude that Germany has a 5-fold higher death rate. This is also absurd.

See the two posts above.  You’re rumbled sunshine.  

It is the same kind of disgraceful abuse of data associated with some of the anti-control measure agitation we’ve seen from denialists and trolls.  I’ve been calling them out in very blunt terms for 18 months and I’m not going to make an exception for someone posting in the opposite direction. 

Excess deaths maybe: 

https://www.economist.com/graphic-detail/coronavirus-excess-deaths-tracker

Tens of thousands of the foolish religious, including right wing catholics and similar strict groups, isn't a big number, it's less than 0.1% of our population. I was put onto the problem after speaking to someone in their 80s who worried me: being clear he was very careful on covid but was never going to use a vaccine that involved such testing; as his church advised and with worldwide support on Facebook. I think its an underreported social media rabbit hole led by big US money and influence (that infamously ended up with one of their number in the US Supreme Court). What this person told me can't be blocked as covid misinformation as technically it's true and it links to US religious freedoms. You're right the pope disagrees but there is a clear push from the US right for a culture war in the catholic church as well. 

I broadly agree with your points on cinemas and bars: risk needs full consideration and that some of this is middle class signalling but you can judge risk by evidenced distancing, low numbers, ventilation, staff behaviour etc.  Plus I'd add some people care passionately about the arts and will be trying to keep struggling venues open, knowing the risk. You can guarentee in something like the Victoria centre walkways with of the order of a thousand people in an indoor public space that there will be customers with covid. Definitely a place to avoid.

Thanks again for all the work and improved visualisation though new graphs

I was initially distracted by the 1.4% Spanish CFR but soon realised I'd overlooked the benefits of the reductio ad absurdum style aspects of a moving average CFR when you started the new plotting to prove your point (which you already had, but that miscommunication ended up very useful for all). I still think the original statement without health warnings had the potential to confuse (as proven in the same thread by a clear misuse)  but now I've transitioned to think the benefits significantly outweigh risks. Pollard's quote linked above illustrates this....he is obviously wrong or misquoted, as 'incorrect' CFRs are not 'incorrect' (cf likely IFRs) enough for what he says (I'd say testing was a tenth of the UK in some western european countries at times with low cases, as the moving average CFR changes to show it, but it's not typical and simply isn't true when cases increase to concerning levels).

On cases: I do think they are dropping but I'd need to see the demographic...if it's only a big drop for younger groups it might not cut hospitalisations at all.

What wintertree didn't tell you about that graph is it's not quite as startling as it looks as the moving average CFR shows testing levels were too often too low in Europe until the deaths started to increase fast (ie infections were at higher levels than cases indicate). The nation comparison to me is what do those countries do when they approach our case levels, at a believable CFR re an expected an IFR, when testing is much more comprehensive. Even in the UK ONS data indicates we are missing half the infections by our testing.

If similar work helps illustrate complacency in Western Europe leading to better action that's a good thing.

On a different subject, the latest on the Wolverhampton PCR scandal

https://www.theguardian.com/world/2021/oct/27/covid-lab-returned-four-posit...

I didn't claim it was startling or not, I left people to think it over.

It's clear that for all nations on there except Belgium, there is time to arrest the rise in cases before things get to the kind of high levels seen in the UK.

I'm not sure you've thought that through.  

• Are you implying that the rising cases are due to sudden increase in the engagement with testing?  That to me is both grasping at straws and at odds with hints of signals in the deaths data or the instantaneous CFR data.
• Or are you implying cases aren't comparable because the instantaneous CFR shows detection of infection as cases is lower in many EU nations?  That is my estimate of what the data shows - and tallies with the comments from the Oxford group Kirsten posted - but it makes this plot more startling because the rapid increase in cases is likely also applying to a relatively higher level of infections vs the UK than this plot implies.

>  The nation comparison to me is what do those countries do when they approach our case levels

The point I've been trying to get across with all this...

1. is not about what other countries choose to do as they loose control of cases.  
   1. We have a pretty good idea that they'll respond with increased control measures to moderate the growth;
   2. I opened the subject with a quote from a poster resident in Germany to that effect, and as we've now heard Belgium is doing the same.  
   3. We've long since established that the UK is taking a different approach to controlling cases.  
2. is that the UK appears to be in a very, very different situation, with cases refusing to grow - and indeed apparently going in to decay - despite our rather "meh" approach to control measures, whilst many of our neighbours are showing potential for rapid exponential growth despite their apparently more keen use of control measures.

A reminder before the whole thread recycles; we are where we are.  We all need to understand that recognising this does not translate to endorsing how we got to where we are.  I am trying to understand where we are, and what it means, not to start a slagging match about which countries are doing "better" or "worse".

NHS not under stress at all… West Mids ambulance service looking et moving to highest ever alert level.

https://www.birminghammail.co.uk/news/midlands-news/catastrophic-consequenc...

NHS not under stress at all… West Mids ambulance service looking et moving to highest ever alert level.

https://www.birminghammail.co.uk/news/midlands-news/catastrophic-consequenc...

I know you didn't say that: I replied to the person who did. Why are you being defensive when I'm praising your new graphical innovations? 

I'm saying in a low case county with high moving average CFR (like Germany was), when infection levels start to grow the cases accelerate faster than expected as testing is forced to improve, as a result of the growth problem becoming obvious.  As you then detect a bigger proportion of infections the CFR drops to a more normal level. Hence your moving average CFR is a very useful indicator of testing levels, and a good warning of why we often can't fairly compare per capita case levels;  but is less useful as a covid mortality rate unless testing rates are very high. This means case growth in those countries isn't quite as exponential as the numbers indicate (it's still bad though). Fast exponential growth in hospitalisations and deaths is the key warning sign that something really bad is happening that might need stronger restrictions (maybe needed them a week before or earlier...... which is why it was a terrible idea places like Germany let their national testing rates drop).

I'm pretty clear what you are effectively predicting about the UK cf Germany and I'd agree it's a likely scenario but I prefer to think of predictions as fairly worthless until the data starts to match them over a long period. In long periods actions happen and they usually reduce the problem faster in places like Germany than they do in the UK as the Germans usually act faster on public health advice than we do. Whatever country we live in, when hospitalisations become a problem restrictions are a necessity and we are long past that point in the UK, as you have said.

Yes we are where we are and in the UK public health messaging terms that's still very bad. In movie terms if covid was a character it would be relentless, cold and merciless ....so like say Dirty Harry saying to our government: I lost count and can't remember if I have any bullets left in my gun, so do you feel lucky punk?

I'm clarifying that I don't support the interpretation you give, slightly defensively because of your very unusual choice of words that rather attaches it to me with the suggestion I'm deliberately omitting that interpretation, when I don't actually agree with that interpretation. ("What wintertree didn't tell you about that graph [...]").  

I also wasn't 100% sure exactly how much you were inferring I was "not telling", which is why I gave two bullet point interpretations in the hope we could narrow it down rather than move on to accusations of defensiveness.

I am not doing this to make predictions about Germany or any of the other nations on that plot.  As I have gone out of my way from my very first post on this to make clear, the expectation is that more responsive control measures will kick in in parts of Germany soon, for example.  I am not making any predictions about where case rates are going to go for these nations, lots of factors I have not the time nor the language skills to understand are clearly going to kick in and have dramatic effects - likely more control measures than we've seen in the UK for starters.

I am doing this to illustrate the massive differences that apparently exist in the present moment in the ability of the virus to spread rapidly between the UK and much of Europe.  It's an olive branch in the data that the dial is no longer set to "universally dismal" for the UK and that our situation is changing.  It's nice to look for and illustrate signs of cautious optimism in the data against the onslaught of bad news that is this pandemic.  

It's just as well I'm not predicting then.  I am making the point that the UK doesn't seem to have the potential for rapid exponential growth, where as other nations do; cases in Belgium have just doubled in the last 8 days, where-as it took over 50 days for the last doubling to occur in England.   This wasn't a low rate doubling either, but almost to the point the UK is at, and with significantly less clear water (in ratio terms) between their recent historic cases and fatalities than us.

Perhaps that's why I've gone out of my way to make the point it's illustrating the pratfalls in comparing cases and not commenting on "true" mortality rates.  Which, as I've said before, take an awful lot of carefully controlled analysis to get even half a handle on.  Well beyond these threads and not that relevant to understanding the data when it is detected cases that we have to go on.

To clarify, I meant that phrase as applied to the immune status of the population within the UK - where we are now determines how the future unfolds, how we got there is an issue of much contention (and much unhappiness from me, openly stated over the last 18 months) but it is of little relevance moving forwards.

I agree over the public health messaging.  I've had very blunt comments to say on a couple of threads that ended up in the "Politics" forum on the subject recently, and that's far from my only comments over the last 18 months.  I'm putting effort in to these plotting threads on, well, plotting however.   We have good informed comment on the grim and worsening realities across healthcare from many posters across multiple threads.  

Edit: You might check up the thread of this conversation to see which post and plot it is that BusyLizzie was replying to - it does't actually have any mention of CFR in it, and if there's some cross purpose talking going on, it might explain some of the confusion here, and perhaps my previous reply will make more sense...

I'm surprised there isn't a more dedicated thread emerging in the news media over the different issues affecting healthcare; various individual reports on here and elsewhere are painting a very desperate picture.

Let's hope a rabbit is pulled out of a hat with regards the Covid situation over the next couple of weeks.  This evening's data update was even more promising in that regards; I'm holding off much more comment until the next thread but I still hope that the winds of change are blowing.

But winter is always a pinch-point for healthcare, Covid or no Covid.  To have it in this much trouble now is deeply concerning.

I'd call DH's bluff as I know his weapon should be loaded with blanks.

"It's just as well I'm not predicting then.  I am making the point that the UK doesn't seem to have the potential for rapid exponential growth, where as other nations do; cases in Belgium have just doubled in the last 8 days, where-as it took over 50 days for the last doubling to occur in England."

That's effectively predicting. I hope you are right about the UK and wrong about Belgium but I'm not totally convinced you are in either case.

On Belgium, cases doubled from a lower level and per capita hospitalisations (and deaths) are noticably lower than ours with better funded and less stressed health system. Plus Belgium are acting quickly on this recent growth (unlike the UK) and I'm pretty sure they will do so again if things get worse.

https://ktar.com/story/4738470/covid-19-cases-spike-in-belgium-govt-poised-...

Jesus wept Offwidth.  You’re determined to read other than I’m writing.

• I am illuminating the situation now, not predicting.  
• You are extrapolating, not predicting.  
• I went out of my way when I introduced this to give an example of another nation expected to bring control measures in to limit this growth, and I’ve acknowledged that for Belgium too.  This is a caution against extrapolating much further and clearly shows I’m not predicting the data.  I have also explained this to you.  I’ve said it once before - it’s like you’re playing a point scoring game here and not discussing in good faith.  

> I hope you are right about the UK and wrong about Belgium but I'm not totally convinced you are in either case.

I am showing the present potential for the virus to spread aggressively or not in different locations.  That is what it is.  That suggests many places are going to have either more restrictions or more illness. (Edit: more going forwards that is, not more net total, ground we’ve also covered several times.)

No shit, Sherlock.  

Looks like someone is forgetting the lag between cases and other measures.   Usually a trait of the other side in this situation, that.  The rolling CFR in Belgium is significantly worse than that in the UK so if the demographics of their recent growth is neutral…

Indeed.  

Not that quickly when they’ve just about hit what you consider an unacceptable level for the UK before turning to more measures.

This adoption of control measures is something I have acknowledged up thread.  You seem to be making the case I’m wrong in predictions (that I’m not making) because other counties will introduce control measures (which I myself suggested in the very first post on this).

Do I have to start citing and quoting all the posts you’re apparently not reading?  Seems to me like you’re determined to either miss the point of this plot, or to bury it in noise.

I'm not rumbled, you are increasingly paranoid you see sock puppets and dark forces and get abusive with anyone who disagrees with you in just about every thread these days.  I'm just having a bit of fun, it doesn't matter enough to spend more than a couple of minutes on.   I used the data Google gave me at the time I typed the search.

I think your plots with the focus on exponential rate constants rather than absolute cases normalised for population are less useful in the post-vaccination scenario.  It's become an S curve.  The UK is running cases high enough to see the flatter bit at the top of the S so the rate constant doesn't look as bad even though the number of cases is terrible.  If other countries were nuts enough to run the cases at the level the English want to run they'd see the growth flatten off too.  But the absolute number of deaths, hospitalisations and longer term health problems would be sh*t and it would be a really stupid thing to do.

Yes you are. 

• You claimed a comparison of death rates
  • It was out by a factor of five or so
• You claimed typing some text in to Google gave you these numbers
  • Investigation by one poster (not me) showed that it does not give you a number, it gives you a plot
  • Investigation by another poster (not me) showed that the plot gives you daily and rolling numbers with equal effort (demolishing your hints you didn't have the time to do a better job) and that you chose the daily number despite the rolling average making it clear that was a poorly suited number.
  • Investigation by that poster also showed you picked numbers from different days so that the weekend bias pushed the UK number up and the comparator number down.

> you are increasingly paranoid you see sock puppets

Almost every sock puppet I have seen has.... turned out to be a .... sock puppet.

Like this time a new poster called "MikeBuddy" was the only person to support your disgraceful misuse of the YCS data, and I called them and their sequential pall "MikeDubby" out as sock puppets.  

You may recall this is the time you apparently forgot to sign out of your sock puppet and started talking about yourself in the their person and calling me "paranoid" for calling out the obvious sock puppet.

https://www.ukhillwalking.com/forums/off_belay/vaccine_side_effects_ctd-737935...

Calling me "paranoid" is a hallmark of one of the regular pop-up posters.  Are you trying to tell me something?  Because sometimes it feels like you're trying to tell me something.  And I don't mean telling me that I'm paranoid.

No, I'm going out of my way to be endlessly patient.  Your cherry picking of different numbers from different days to misrepresent the reality is inexcusable.  I have saved my insults to obvious pop up posters.  I have no bloody clue why I extend you the curtsey of thought out replies, I really don't.

• 60% of this week's plots are not rate constant plots.  Some "focus"
• The rate constant plot is what it is, and it is useful for understanding some things more than others.  It's has clear benefits in understanding the rate of change in a way that is invariant of the absolute number of cases, that has always been its purpose and that remains its purpose going forwards.  
  • If you understand the mechanic by which a pathogen spreads, the value of a plot looking at normalised rate of change has a clear use in understanding change
• You can find plenty of normalised cases/person plots elsewhere.  Turn of the 7-day filtering to cherry pick data to suit your agenda on them. I've no interest in duplicating effort elsewhere.

> It's become an S curve. 

• What is "it"?   A banana?
• Assuming you are talking about integrated total cases, that's not true, it always was a sigmoid like-curve, it's just that we're moving towards the far right of the S. 
• Only it's not really a sigmoid, as it's clear we're not going for elimination so periodic re-infection remains on the cards.

Let me summarise the thread so far.

Wintertree: “Here’s some data, 1+1, it looks like it equals 2, have a think about it”

Offwidth: “No, 1+1 = Rabbit, stop using logic against me”

Tom in Edinburgh : “1+1 = 3, and it’s all the fault of the English Tories that it doesn’t equal 2”

Anyway keep up the good work. It’s very appreciated by myself and presumably many silent viewers.

I missed a key part of your post.

We've already established - through review of your claims by two other posters - that it would have taken the exact same amount of time to produce the data without the weekend bias - less in fact as you wouldn't have to have sought out two different days to maximise the effect for two different countries.

Putting that aside - you're "just having a bit of fun" - please help me out with the interpretation of this Tom...

• You "are having" fun by deliberately posting misleading data that severalty misrepresents the situation to provoke replies (trolling)
• Something else - please elucidate. 

Because if it's trolling, might I suggest do one?

It's also interesting that you jump from me calling "rumbled" to a presumption of sock puppets.  I never said any such thing, and nor would it apply in any conceivable way to the posts under discussion.

I simply mean you're rumbled for posting a heavily biassed comparison that would require such incompetence to arrive at naturally it was almost certainly contrived.

Saying the UK lacks possibility of exponential growth and other countries don't is an effective prediction.  I'm not sure your as right as you think, but you are being honest to the data we can all see. I really hope you're right about the UK as more serious restrictions would be really bad for us. It smells like 'something's up', as the majority of the government front bench are suddenly wearing masks yesterday and Vallance was making philosophical statements last weekend about early and hard.

The reason I'm not totally convinced about the UK lacking the possibility of exponential growth is because we are still sitting at R effective at 1 with high case numbers and not back to normal yet in ONS behaviour and work surveys and behaviour is still shifting to less compliance and more people are going back to work and using public transport; all alongside the fact that we still have millions at higher risk of hospitalisation and death, through very old age, vulnerability and not being vaccinated (if not yet infected). 

So we disagree in some small aspects and that disagrement is normal in scientific discourse. I probably worry more about risk unknowns given our hospitals are at their limits. Such disagreement is also very obvious in the slightly different views of the honest covid experts in the media. Disagrement often generates deeper thinking and useful things can arise from that: in our case your new CFR plots. 

Belgium's breakout is massive in young school kids (1 in 4 infected), so their deaths won't be as high as cases and the current CFR  might indicate (also Belgium's method of measuring covid deaths is the most 'inclusive' in Europe, so their level always looks higher than comparative deaths elsewhere).  I've not heard anything especially worrying about Belgium hospitals and the lag time for that is only a week. Their hospital admissions per capita are half ours and if anything our increasing trend is worse and they have bigger capacity (Germany a quarter of ours and in slight decline).

We are where we are and we shall see.

It would be if I was claiming this predicts the future.  Let's take a forensic look at some of what I've said:

• "Despite progressively dropping restrictions and individuals increasingly dropping their transmission and risk control measures (e.g. the recent ONS survey), we're not seeing the potential for sustained growth in cases" - "we're not seeing" is anchoring this to the recent past 
• "I am showing the present potential for the virus to spread aggressively or not in different locations.  That is what it is. " - note the word present in there
• "This doesn't diminish my point that the potential for the virus to spread in many countries appears to be much higher that it is in the UK right now. - "right now" is limited to the present moment.

From the very start I have been at pains to say I don't expect the growth to continue in high growth countries, as control measures will kick in.  Things can always change in the UK for the worse - god knows they have enough times in the past.

I am presenting my interpretation of the present moment.  I am going out off my way to use words like "present" and "right now".  I hope everyone reading understands that nothing is "set in rails" - if it was, we'd only have needed thread #1 and that would have been that.  

As I said a few week ago and you objected, I suspect many of our neighbours have been watching the trajectory of the pandemic through the UK very closely.  You suggested they would not as they'd be for more control measures.  

Edit: almost every plot I have put up can be extrapolated to give a baseline “if nothing changes” prediction for the next couple of weeks.  I have consciously avoided extrapolating almost any of the plots, ever, I think I’ve only done some once as a way of understanding the situation and explicitly said it was not meaningful as a prediction.  So, I’m curious as to why you’re so set on using this plot and interpretation as a prediction and no other plot? 

Well that's semantics. Maybe we can agree there is capacity for exponential growth in the UK right now but the current signs are the opposite in the case data. My system  concerns (that you  share) are much wider than  hospitalisations and deaths, as the NHS, care system, public health and social support is really struggling right now and always get worse in winter;  so we really need things to improve a lot in the covid situation, soon. The clear window of opportunity for exit I saw in the summer is being bungled by delays in vaccination, terrible messaging, seriously stupid attempts at burying NHS bad news (it always gets worse and becomes so obvious that it gets out) and lies about always following the science.

Other western European countries have had more controls but the specifics vary from country to country and the overall messaging seems very diferent My main point about them not following the UK was more about the combination of political messaging linked to an implied fully open system (in reality we still have various restrictions, city level, organisational level and many self imposed individually).  It must be weird and unsettling to be a leader in those countries and watching Boris 'tear up the rule books'. However one of those leaders right now needs to learn his teflon tricks fast (Luxembourg). 

Back in the UK from the covid Guardian live feed some more quotes from Vallance (who will have the best advice on  the same data we can see and some we can't .... and he has to be very careful with consistency with government political messaging).

"Nobody is really clear which direction this is going in, but they are clear about the two big variables that could change that.

One is waning immunity, so if immunity wanes faster than expected, you’ll see a bigger increase, and that’s why it’s so important to get booster shots going in the vulnerable and the elderly in particular.

The second is the behavioural change, how quickly we return to pre-pandemic behaviours. if you aggregate the models, most are saying ‘Actually, it looks fairly flat, don’t expect the very big peaks we’ve had in the past, it looks fairly flat, but at a very high level at the moment.’

So the high level remains a concern and from a high level you can go up quite quickly."

In a separate appearance on Sky News, he said:

"I think the vaccines have made an absolutely massive difference and now we also have antiviral medicines coming along. We’ve definitely got the armamentarium we need to tackle this now. Keeping immunity high is really, really important.

We’re a bit uncertain as to which direction the levels go at at the moment, so I think we’ve got a pretty difficult winter ahead of us."

Another couple of days of data updating the OWiD derived plot.  I've put a logarithmic x-axis on as this is a much fairer or more natural way of comparing what for many nations looks like a very exponential process.

I've not looked at detail beyond the UK data, but within our shores the rate constant has almost always behaved as if it has "inertia" - it can't change instantaneously given the complex and broad statistical mechanics it emerges from.

A reminder that there's a lot of day-of-week sampling effects for some of these nations, and low numbers both of which present as sources of noise to the rate constant measurements - and even to the case rate measurements (made with a 7-day rolling average) for places way over to the left.

I thought the different days point was a red herring as the other poster looked at the data from the following day and it all got confusing as to who was looking at what. However I might have misunderstood and can’t be bothered to double check. If data from different days were deliberately used, that’s a disgrace as you say. It all comes back to the point many people have made but Tom just doesn’t get because of his blinkered outlook. He makes some good points but then weakens his own position by twisting everything out of proportion or in this case misusing data.

It’s fair to say that the German death rate per 1m population has been significantly lower on the whole but it’s not been 10 times lower. Comparing deaths on a given day is clearly statistically meaningless. In fact comparing deaths over a week or a month is largely meaningless due to countries being at different stages of the pandemic. It’s necessary to look at longer timescales. Tom, I suspect you know all this and if you actually made sensible arguments based on sensible data, people would actually agree with you rather than arguing with you. I do agree that we should have Plan B type restrictions but I don’t agree with your misuse of data. 

No, it was completely accurate for the numbers Google gave me which were the latest numbers at the time I typed the query.

You want to use 7 day rolling averages.  OK, fair enough, probably a better approach, but it is a different criterion.  7 day average vs last available day.   

Because it did.

'Investigation'  Jesus f*ck.  Are you a cop or something?

Do you reckon Google gives the exact same response to everyone no matter what searches they have done previously, which search results they've clicked on previously, where they live, the day and time they do the search and a thousand other factors?  Maybe 10 or 20 years ago.

I honestly do not give enough of a f*ck to do that amount of work.   I am killing a couple of minutes, that's all.

I didn't pick anything.  You are making the fundamental mistake of conflating malice with laziness.

So now I have a sock puppet too?   I really don't.  I used third person ironically to underline the point I was making.  It was supposed to be slightly funny.

I'm talking about cases/day.  Earlier on, before there was a lot of vaccination exponential was a reasonable approximation, now for the UK it isn't:  when the absolute number of cases is large enough the growth flattens off, when it is relatively low you can get exponential growth. 

Now the flatter are at the top of the S is in play the exponential growth factor is less useful as a measure of how much trouble you are in.

You have cherry picked data that mis represented reality by a factor of about five times

You persist in lying that it would take more effort to use the rolling average than the daily number, as everyone who has tried your method has found out both numbers appear in the tool tip on the plot that one must use to access the numbers.

Your excuse/defence of laziness fails on two counts, one is that rolling average is available with equal effort and the other is that you’re willing to spend literally a hundred times the effort a fair comparison would have taken on defending your obvious and flawed cherry picking / misrepresentation.  You did not need more time, and you have shown that you do have the time.

I told you where I got the numbers.  Believe me or don't believe me, it's up to you.  

Hi Tom, I would never usually bother to check a number someone else gave in a thread on UKC. I did it for you in this case partly because it was so easy to replicate the search you said you used in Google, but moreover because you have consistently spouted fake news and mis-represented statistics over the past few months in your demented quest to paint everything under the sun as a damning indictment of the Tories, the English or England in general. As far as I'm concerned your posts are far more likely to be falsified than they are true, so I thought it was worth the quick check on this occasion. You should reflect on that. If all you want to do is troll the forum, fine. If you want to be taken seriously, I'm afraid you're losing the battle.

Can you show a screenshot of a Google page that shows the daily number and not the rolling average?

Several people have tried to reproduce this and have all failed.

Do you know why I don't believe you?  Because every time you make a "lazy" (your words) comparison, it always - always - lands with a bias in one direction.  That sort of thing doesn't consistently happen by accident.

Latest ONS infection report.

https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/con...

Reported as record infection levels for the UK on the BBC!

On a different topic: does anyone know what is going on in the Guardian age demographic case breakdowns as the overall decline across the demographic plots when weighed seems stronger than the drop in the total case moving average?

https://www.theguardian.com/world/2021/aug/20/covid-uk-coronavirus-cases-de...

Plus more on the demographic time bomb in Primary Care (why its a such a terrible time for the Health Secretary to pick a fight with GPs). Of course there was no pension help in the budget to try and alleviate one key factor incentivising GP early retirement.

https://www.theguardian.com/society/2021/oct/29/nhs-facing-mass-exodus-gps-...

In reply

Latest IndieSAGE weekly report:

Shorter on the data this week but with a good look at the Immensa scandal data. This includes separating out the 13 LAs most affected in the SW to show the awful high resulting impact on children and their parents age groups. Plus data on exceptionally high 'void' and 'no result' test outputs that are clearly showing testing is under really serious pressure across the UK right now.

youtube.com/watch?v=1vGdRA_7hu8&

Tell you what though, digging through the International data I get the impression the UK is still managing a stand-out high level of detection of infections as cases through testing compared to almost all of Europe and the USA.  It's not surprising that it's under pressure right now but we have to think as much about where we are going as where we are.

(None of which applies to Immensa) 

I've got some plots on this but I'm not going to share them as I can't face another round of misunderstanding, misinterpretations and tangental point making.   

Thanks for the pointers. Haven't time to watch right now... do you mean the voids and no-results particularly high at this point in time, or in the UK generally, or particularly in the south west?

Interestingly I noticed the Zoe study is also now showing a big rise in the south west so that is presumably a 'real' rise on people reporting symptoms, caused by rather the testing failures rather than mostly a result of testing issues themselves directly affecting the numbers. If so, it does give an indication of the importance the test and trace system must have had throughout the pandemic, despite all its problems.

Starts just over 10 minutes in and lasts a few minutes. Yes the testing pressures are across the UK and much worse than in the previous peak.

I agree we are doing very well in international terms on detecting and especially on sequencing. I should hope so, given the £37 billion.

Sad to hear you are not posting useful stuff for those reasons.

Cue a plot 18 update...

Most regions are now seeing hospitalisation turn to decay it seems; the South West is seeing the least benefit from this (I've coloured the curve for the SW red on the right side plots).

Very far from conclusive but tallies with your observation on Zoe.  

I was initially skeptical that the lab failure would have made a measurable difference in the data, but I think I was being dim and it took a while for the enormity of their failure to make it though to me.

I'm really sat on the fence about how much of the decay in cases and their rate constant is (a) real vs half term reduction in sampling and (b) sustainable vs half term reduction in transmission.  The decay or trend towards decay in hospital admissions for all the regions other than the SW in plot 18 has me just about believing that it's real, but as for sustainable....?

Oh f*ck off.   Do you reckon I screenshot every google search so as to 'prove' stuff to you?

The main biases in this forum at the moment are the focus on deaths/hospitalisations as the metric for harm caused by Covid which ignores longer term impacts and by treating them as zero makes England's strategy of running high cases look safer than it actually is.

The second bias which makes England look better than it is, is the use of rate constant plots rather than absolute numbers scaled for population.   England is running enough cases that exponential growth is constrained by vaccination and previous infection, it is getting to the top part of the S.  That makes the rate constants look not too bad even though the absolute numbers are sh*t.

No.

I don’t think you could screenshot it to prove it to me and to the other people who doubt your claim. I’ve never once seen Google give the daily number “at the top of the page”, it’s always a graph with the daily and 7-day rolling averages having equal precedence on the “tool tip” over the graph.

You can’t prove it because you’re making it up.

You cherry picked differently biassed data from different days and you can’t accept being called out on it by multiple posters. 

You’re rumbled. 

I’ve been at pains to point out how small positive rate constants mean big daily changes in every thread and the important and negative consequences that has.  I include absolute numbers plots in every thread.

My use of rate constants is about illustrating *change* not about playing international top trumps.  It’s about being able to follow change through the measures to better understand.

I’m sorry you don’t understand how the fractional change in cases, hospitalisations and deaths has a place of its own in the set of tools for understanding this data.  Call it a rate constant, a characteristic time or a fractional change, it’s another lens that shows some things better than others.

I’m not interested in making England looking better than it is.  I’ve said more times than I can count how unhappy I am with what has come to pass. I’m looking for insight and understanding.  If you can’t get that in to your head, then you are wasting your time here.  This is just coming across as petty minded anti-English bigotry.  

Indie SAGE implied it's likely a real decay in cases even though they showed the previous half term effect in their Friday stats.

That's bs Tom. Hospitalisations/ death are the main NHS concern short term. Plenty of us are pushing long term issues of concern including: new variants, long covid, the physical and mental degradation of NHS staff, the damage to our democracy of chumocracy contracts, or my biggest  concern (as per the puppy who doesn't understand what wintertree is saying) that everyone will catch covid and its lethality won't drop. The latter point implies anything up to a hundred thousand of the millions of vulnerable shielding will die at some point in the next few years... expert's talking about optimism on the TV in those terms need to think on that.

You sound like you're an engineer... you wouldn't be so sloppy or exaggerated in your job so why are you ruining some good points being like that here?

Latest Warwick modelling. My main takeaways from a cursory look is that immunity fade and behaviour are two key unknowns. Although I’d say immunity read is the real unknown…

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/...

Please, tell us more about bias

How the f*ck am I supposed to screenshot something from two days ago without a time machine.

I am concerned about all harms.  I think the metric of interest from the point of view of deciding whether it is safe to let it run wild should be 'complete recoveries after X days'.

The focus on death within 28 days and hospitalisations is also a Tory strategy for hiding other negative outcomes and making it easier for them to justify letting it run wild.  So is the 'this is about protecting the NHS' line.  I don't accept that and I have never accepted that, the goal of interventions is to protect the public from harm.

Yes. I am keen to learn more about this interesting idea.  Perhaps then they can teach us about irony.

That's something with a high ferrous metal content, right?

Flippancy aside thanks for your continued work collating and presenting the data.

Apparently there are 405,000 people who have had Covid symptoms for more than a year after infection and many of them are unable to return to work.

https://www.independent.co.uk/news/health/long-covid-symptoms-treatment-job...

So using the latest daily number which Google puts above the 7 day rolling average is 'lying', 'cherry picking' and 'misrepresenting'?

So, you admit you read it off the tool tip and chose to ignore the rolling average?  Progress.
 

• Lying is by omission (you don’t note one number is many times below the rolling average and the other over, a basic and critical piece of context to interpreting them)
• Cherry picking was by selecting different days for the two nations to give an effect highly unrelated to reality
• Misrepresenting - yes, by a factor of about 5

You have become impossible to tell apart from the trolls in the way you engage.

This is what you posted.
"Germany 27 deaths   83 million people

UK 263 deaths  67 million people"

You posted a comparison.

If you can find anywhere reputable where you sourced this comparison from I'd be suprised.
What you seem to be claiming is that just because the numbers are correct (in their individual context) the comparison inherits their rigor. However, Tom, *you* created the comparison, so it is your responsibility to make sure it is valid.

In two ways you have failed at this - by not using the rolling averages (poor practice, but might be acceptable in some cases, especially if admited) and by comparing data picked from different days (absolutely wrong - would have been mitigated if you'd used rolling averages...)

Finally - you've not yet admitted *that the comparison you posted is very, very wrong and misleading*. Instead you're doubling down on being argumentative. In data analysis, this is a sure sign of an unreliable source - why should I ever rely on something you say? Why should anyone?

Anyway - that's enough from me - I'm going climbing for a week on sunny limestone away from data and numbers.

Excellent!

It’s a day’s walking in interminable, wind driven rain here.  Mild for late October but not pleasant.  Just got to motivate everyone to get out in it and it’ll be fine…

Exactly...why on earth after such a blatant data error wouldn't anyone with good intentions just say "I got it wrong, I'm sorry".

A decent 'on the ground' summary of the lead-up to the Immensa debacle written by a local acquaintance.

https://www.opendemocracy.net/en/ournhs/i-raised-an-early-alarm-on-pcr-test...

Cheers David. The 'victim of mathematics' Twitter link inside that article has some good visualisations and commentary on the data so I thought that worth linking separately.

https://mobile.twitter.com/victimofmaths/status/1454094552269262852  (links mid-way so scroll up).

Good link.  That graph is fairly damning.

Yes, VoM does not normally speak so bluntly.

I was mulling over if there is a corporate manslaughter angle to this, if people have died downstream as a result of negligence.  Not much of a solicitor so I don’t have a clue.

Thanks David.

I was looking for daily cases, I found daily cases and I cut and pasted it.   

Your paranoia is getting as ridiculous as your persona of high inquisitor.

The data is from Google it is a 3 minute cut and paste job to make an order-of-magnitude comparison.   It got caught out by weekend effects which I didn't think about.

This is not my job.  I don't put that much time into it.

Still unable to evidence your claim that "I typed 'germany covid deaths today' into Google and copied what Google put at the top of the page." then?  Nobody else who tried has managed to find it at the top of the page.

It's not paranoia Tom to call out an utterly wrong and flawed comparison that's total bullshit. 

Your comparison used cherry picked data from different days to create a deeply misleading comparison.  You've been repeatedly called out by multiple posters over this.

Your comparison was flat out bullshit and you refuse to acknowledge this or apologise for posting deeply misleading data.  Instead you are going on a series of personal attacks on my character.  A classic sign that you're arguing for the sake of it and have no truth, no data, no evidence, nothing of worth at your back.

Calling out someone with a long history of posting cherry picked, biassed statistics in line with their consistent bias across many months is neither paranoid nor "high inquisitor".

Why aren't you personally insulting Si dH and Paul Baxter who have raised the same issues?  Are you just trolling me?  Now that is paranoid thinking.  Or is it?  You tell me.

There was no cherry picking of different days it was the most recent day available in both cases. The most recent day available happened to be a day different because it is two different countries.

I'm not reacting any more because if I did this would get far too hostile for something which is, basically, not that important.

You cherry picked unrepresentative data with a five times bias in the same direction as usual as aligned with your endless xenophobic bias.  I’ve seen a half dozen posters call you out over your increasingly delusional snti-English stance on a bunch of unrelated topics in the last month.

Perhaps you didn’t deliberately cherry pick, perhaps the issue is that you are completely incompetent at even the most basic of comparisons and only post data when your incompetence produces a result aligned with your pre-conceived bias.  This could also explain the consistently one sided outputs you produce.

Either way it’s misrepresenting the situation with cherry picked data.

I struggle to believe even that when another poster verified the same day of of data you used for one nation was available for the other just a few hours  later.  Especially as the daily OWiD data release happens shortly after the other poster showed a same-day comparator was available so you’d have both likely been looking at the same data set.

I regard pointing out blatantly misleading covid data as worthwhile and yours was wrong by a factor of five times.  Normally when someone gets rumbled the skulk off rather than dig in.  Perhaps you’ll make an effort to not cherry pick massively biassed data in the future, either through malice or through bias filtered incompetence?  Your claim not to be aware of day of week effects is risible given the really obvious toothcomb artefact visible in the plots headlining the Google results you claimed to use.  I’d expect any sort of engineer or scientist to immediately notice that and to proceed with a bit of caution about using raw data.

I repeat - why are you not responding to Si dH or calling them paranoid?  Are you just trolling me here?