UKC

It's the half-year anniversary of these posts; I originally started putting plots together for own understanding and to test some some claims made by others that I regarded as quite dubious; poster Veteye asked me to put them all in one place, and here we are.  

Last week I'd suggested we could be at the tipping point where stagnant and rising areas start driving top level growth in cases against the decay elsewhere; that seems to now be the case.  The Indian variant is featuring strongly in the news over the last few days in relation to these growth areas, and some alarming numbers have made the news, for example the BBC noting that "cases have tripled in a week" in one area.  It looks like much of the increase in numbers is coming from a small number of growth regions, driving by this new variant in those regions.  There's several links to SAGE documents and other presentations of the data on plotting #25.

None of this is good, and it's certainly not what any of us wanted to see, but I think there are some good reasons to keep an open mind about what could happen.

• This is very early days. The worst affected English UTLA has about 40 cases/100k, and the next worst about half that.   When the "Kent" variant was first publicly confirmed as being behind the last localised (and then nationwide) failure of control measures, we had 7 English UTLAs over 40/100k, and about 75 over 20/100k referring to plot 16 in thread #3 [1].  Compared to the current plot 16 I'll put in a later post, we were in a much worse situation when that hit - now all but three UTLAs are below 10/100k.  It's going to be a while yet before the absolute numbers pose significant problems - time to evaluate, think, change plans - something we are already seeing in the responses. 
• Late last year, the national response was blindsided by the Kent variant and it came to top level awareness only long after it had spread widely.  This time round, cases were so low and sequencing apparently so integrated and central to the monitoring efforts that the authorities have a good idea what is happening.  Understanding what's going on makes a big difference to how well it can be responded to - and we're seeing that already.
• Vaccines!  No vaccine is perfect against the old variants, and the chances are they'll be a bit less perfect still against this one with some reduction in their efficacy against transmission, symptomatic illness, hospitalisation and death.  But they're all starting from very high efficacies, and what I've read so far is encouraging.  
• The fundamental interconnectedness of everything - we have a whole bunch of issues at play in this situation - people returning to the UK to avoid near future quarantine requirements, inhomogeneous engagement with vaccination not uncorrelated to the travel, and a bunch of other factors.  An optimistic take on those is that there are some specific sub-populations that represent a greater risk of transmission, and that these are the ones energised by recent importation events.  It's easy for me to speculate from the comfort of my home; reading the SAGE documentation I think perhaps they're waiting to evidence this one way or the other, but they have the weight of responsibility on their shoulders. 
• Control Measures Work.  If there's one thing we've seen time and again, it's that when control measures are taken, they work.  There's a lot of effort being focused on the hotspots right now.   The data has long hinted IMO that the initially rapid doubling times seen when absolute case numbers are low will moderate over time.  The 5-day doubling times being reported now happened when cases were very low, people were returning to normal, almost all sense of urgency and caution was gone.  Cases have risen, I imagine local social media and old fashioned grape vines in the hotspot areas are abuzz, partial school closures have been reported, it's all over the traditional media, and there is a new option on the cards for people responding to all this - engage with vaccination.  
• Surge testing - there was a release of information to the BBC from "sources" a few days ago; there are different ways of interpreting it all, but one is that the infection got away from contact tracing, and that surge testing then chased it down.  This could generate a rise in case numbers that concentrates a more gradual rise over a longer period of time in to a short, sharp shock rise.  That feels compatible with the data to me, but perhaps I'm in to wishful thinking territory.

The paradox of Covid strikes here - most of the reasons I hope this is not going to turn in to a major setback will only apply if everyone responds to events as if they are a major set back.

All that being said, it will only take a continuation of the trends in the recent sequencing results and the UTLA level case numbers over the next week to sweep most of that optimism away.  I'm going to hang on to it for now.

Link to previous thread: https://www.ukclimbing.com/forums/off_belay/friday_night_covid_plotting_25-...

[1] https://www.ukclimbing.com/forums/off_belay/friday_night_covid_plotting__3-...

The four nations plots have something of a continuation from last week - Wales is still in decay, the clear growth in Scotland has continued, NI has a bit of growth, and England - well I'm not really sure what's going on with the data in England right now.  See the next post.  Last week the growth was only in the last couple of days so didn't really show in plot 9x, now there is little doubt that a significant shift to growth has happened other than in Wales.

There was some discussions last week about Glasgow south of the Clyde; the news today is linking this to the Indian variant.

Cases in England are all over the place in a way that doesn't resemble statistical noise.  They looks to have risen and now to be back in to decay over the last week.  This is the sort of thing we might expect if there was surge testing working its way through a significant but limited pool of live infections.  

There's also been a chance to the way cases are reported; Si dH linked to this on thread #25 - "by specimen date" is now going to be by the data the specimen was taken, not by when it was processed in the lab.  This is fantastic as it should significantly reduce the day-of-week anomaly ("weekend effect") in the data and greatly improve the ability of people to analyse the leading edge of the data consistently on any day of the week.  Assuming this change is made going forwards only it's going to introduce a disinclination in to the timeseries and bump up the case numbers for a couple of days in a way that reflects lag leaving the system, rather than higher daily infection numbers.  

We've also got very different mechanics at work - growth in some areas, decay in others and a fine balance between the two at the top level.  All these things make the top level cases numbers hard to interpret IMO.  

English hospital admissions are still falling and deaths are I think down in to small numbers statistical noise.

So, not much to take from these plots this week.  Next weeks are the ones to watch.

Plot 16 shows the change in cases/100k at the UTLA level baselined against a point 3 week before, the same set up as last week.  I've changed the threshold for annotating a UTLA significantly though as this week as a lot more are above their baselining level than last week.

Three areas are undergoing metric rises - I've put links in below to the dashboard plots for people to look at the raw data: 

• Bolton - https://coronavirus.data.gov.uk/details/cases?areaType=ltla&areaName=Bo...
• Bedford - https://coronavirus.data.gov.uk/details/cases?areaType=ltla&areaName=Be...
• Blackburn with Darwen - https://coronavirus.data.gov.uk/details/cases?areaType=ltla&areaName=Bl...

When we watched what turned out to be the "Kent" variant spread, individual UTLAs didn't rise this fast, and the effect spread week after week through neighbouring areas East of Ealing in particular.  It could be that there is no significant spread here because these are "flash in the pan" events caused by energising specific networks at high risk of transmitting the virus, or it could be because there hasn't been time for the geographic spread to occur yet, so fast is their rise.  Again, what this looks like in another week is going to tell us so much more, I hope.  In particular, what happens in the regions geographically bounding these three?

The two takes on plot 18 show the North West to have the most significant return to growth in cases.  Interestingly, there's a return to growth for hospitalisation in London, but looking at the raw data [1], I'm not convinced this isn't the result of measuring noise - as always the far right of the rate constant plots is provisional, so one to look at again next week.  One of the key indicators for the potential for severity of these outbreaks is the hospitalisation signal - how well do the vaccines break the link between cases and hospitalisations?  Assuming, of course, that the infected person has been vaccinated...  As a reminder, plot 18 censors data points from the rate constants when the underlying measure has a value of < 10/day, as by this point they're definitely measuring noise over anything meaningful.  

[1] https://coronavirus.data.gov.uk/details/healthcare?areaType=nhsregion&a...

Two demographic cases/day and rate constant plots, the first for England and the second for the 3 "outbreak" UTLAs annotated on this week's plot 16.  These use more filtering in the time axis of the rate constants to reduce the impact of noise on the outbreaks plot, where numbers are lower.  This moderates (by broadening and lowering peaks and troughs) the absolute numbers in the plots, compared to the usual version. 

This week it looks clear to me from the outbreaks plot that the age distribution of higher cases has shifted down compared to the November and January waves - this can be seen as the emerging purple area on the right hand side of the left colour map having moved down compared to the last two purple regions.  I don't know if this is:

• The effects of vaccination being towards the upper end of the scale - if so, great, as it shows the vaccines are reducing transmission of the new variant
• A sign the variant spreads differently with respect to age
• Related to the demographic of the inbound travellers who imported and seeded these cases.

The rate constants look quite determined in the outbreak UTLAs; if they don't moderate soon, there won't be enough cases left in the rest of England to decay and balance out the numbers, so top level growth in cases seems inevitable if this doesn't break, fast.

I was surprised by the island of orange in the upper right of the England plot - a definite growth in cases for the oldest ages.  I don't know if this comes from the outbreak regions but doesn't show in their plots (because the low numbers mess up rate constant measurement, but with  baseline from the rest of England this is stabilised) or from elsewhere.  I was going to do some plots to look in to it but ran out of time.  These are the English UTLAs that have seen more than a 10% growth in cases in people in the top 4 age bins over the last 7 days of demographic data: Bolton, Cheshire West and Chester, Dudley, Essex, Hampshire, Harrow, Kingston upon Hull, City of, Leicester, Sefton, Worcestershire

Grab bag plots.

The vaccine plot has changed form a bit.  I don't think I've done a very good job of explaining it:

• Dark blue are second doses given on a date.  Drawn above these in light blue or green are first doses.  They’re coloured green if that first dose has now been followed up by a second dose (according to a trivial model where people get their first and second doses in the same order) and blue where they have not.   You can read the lag between doses off this plot as the number of weeks between the right side of the green and the right side of the dark blue - just over 11 weeks.   You can visualise the “vaccine debt” of first doses that now need to be followed up as the light blue area to the right of the green - from this we can see that after another 5 weeks of giving second doses (“servicing the debt), there will be very few first doses due a follow up, meaning that most of that capacity can go to giving first doses. 
• With the recently stated intent to bring the gap down from 12 weeks to 8, it’ll be interesting to see if that comes at the expense of first doses due in the next 5 weeks or not.  LSRH on plotting #25 pointing to some diggings and analysis by Paul Mainwood on Twitter that suggest there may be significant stocks on hand to accelerate the process; if so, the sooner we service this debt, the sooner we switch back to giving half a million first doses a day to younger people, the importance of which as a transmission control measure goes up significantly under the more pessimistic side of the models over the India variant.

The variants plot.  As always the data behind this is riddled with more unknowns than knowns - at least in terms of what is categorically confirmed.  It's likely surge testing features in some part in the meteoric rise of the India variant.  I think if the ~5 day doubling time has continued by the next update to these numbers - expected on Thursday - well, let's not borrow trouble and worries before we get there.

The week-on-week take on the exponential rate plot for England is all over the place.  I don't know how much of the two very positive data points are to do with surge testing, and how much they are to do with the disinclination in the time series from the changing reporting practices over "sample date".  The final two points suggest we're back in to decay, although they might be a bit provisional - regardless, if the growth UTLAs don't break their trend very soon, clear growth will be coming to this plot.   The last two weeks of data on this plot just look "wrong" to me - you get a feel for what the timescale of variations is, what the noise model is, and it's all been thrown out of the window.  Waffle, not quantitative stuff, I know.  I think this reflects that lots of different things are going on at the moment and that the top level data is, as such, not very useful right now.

So, I'm going to hold the line until at least next Thursday when the variants data comes out.  It does seem like opening up indoors hospitality venues on Monday is really bad timing.  

I know I've banged on about inbound travel and managed isolation and quarantine before, but these events hammer it home.  It's clear that a traveller compelled to self-quarantine in a shared household is nowhere near a sufficient control measure - especially when we incentivise people to rush home from an area where control of cases has been lost, to beat a managed quarantine deadline, as that's exactly the kind of travel origin that's likely to be producing variants.  A lot of the publicly voiced concern over this variant is about immune escape.  My concern is about fatality rates - SARS-nCov-1 and MERS-nCov have fatality rates far worse than this virus, and the 1918 "Spanish Flu" likewise - and in younger people.  I don't know how evolutionarily accessible those fatality rates are to this virus, and I've not seen a good deconstruction of that question.  But, all things considered, I'd rather we didn't play Russian roulette with variants any more.  There're several compounds in the clinical trials pipeline at the moment that tackle the immune dysregulation contributing to these high fatality rates, and I'd really hoped we wouldn't loose control to any new variants until (or if!) we had good phase 3 data on the JAK inhibitors and nebulised interferon beta.   Let's hope the current events are a warning shot, and not an opening salvo.

Re Bolton and neighbouring areas. In previous waves, Covid appeared to leach into Wigan via the neighbourhoods of Wigan that border Bolton (Atherton and Tyldesley).  So far, this isn’t happening (Atherton has had between 0-2 cases in last 7 days).  Touch wood it’s being contained. 

And Happy Demiversary!  Your efforts are appreciated by me and many, many others. 

Thanks and happy 6 month anniversary! Your posts and the wider discussion on this thread have certainly vastly improved my understanding of what’s been going on, as well as being intellectually interesting.

All eyes on Thursday’s variants report. If it doubles again, that’s clearly bad news, though as the data will be over a week old it won’t yet reflect the behavioural impact of the recent media flurry (and won’t yet fully reflect the impact of surge testing).

Encouraging to see news photos of people queueing for vaccination in Bolton. 

I never understood why this was not the case; you had to mark the time and date the test was taken when submitting tests. This was clearly the way to get timely and accurate figures.

The only excuse I can think of is that 'we' weren't trusted to mark time and date correctly. That would have been a simple sanity check to rule out the small number of errors.

Yes, it always was a bit of a baffling choice. Perhaps back when the whole system was pretty overwhelmed, it was marginally easier to report by lab date and then it just stuck? Or perhaps reporting by specimen date would have shone too much of a light on how much the labs were struggling to keep up back then?

I hope they continue to work. I had by first one this morning and I haven't felt this bad since the day after my 21st! Don't want it to be in vain!

Thanks for posting the graphs and your interpretation of the data. Some of it's a bit over my head I'll admit, but it's far clearer than most I've looked at. 

It might also have been a way of trying to pretend things weren't so bad, given the original government approach of pretending it wasn't happening...

Thanks again. I would echo what others have said about the value and interest of the threads. As the last 2-3 weeks shows they are worthwhile even when the data starts to look a bit "boring"!

The 'live' (daily) not de-duplicated 28-day sequencing data for b.1.617.2 has risen again as of today to ~2320, from ~1650 last Thursday (when the reported gov.uk number was 1313). Can't really interpret anything from this but obviously still rising fairly quickly.

I'll be keeping a close eye on some of the hotspot areas over the next few days, like you say the key is whether they can be contained without significant spread to adjacent areas. I was pleased to note in the Sage minutes that they noted specifically the importance of taking measures in a broader area than where a specific outbreak had occurred...yet disappointingly to my knowledge this is not happening...effort is focused very much on the specific centres of infection. However, on the more positive side, looking at the first couple of days of 'incomplete' dashboard cases data for the three UTLAs you picked out (linked in your post at 21:23), which is usually close to complete, it does look like exponential growth has slackened if it's not an effect of the change in reporting mechanism.

My wife had her first Pfizer yesterday, phew! I have to wait for them to drop one more year. (Interestingly the centre is just a local church but she said they were giving out exclusively pfizers, I had mistakenly thought those were mostly done at mass centres due to the storage requirements.)

(There is already quite a bit of localised spread as you can see from the attached map from the north west. Each of the obvious outbreaks in Bolton, Blackburn and Formby were only significantly affecting 2-3 MSOAs if you looked at this a week ago.)

Hear hear!

Thank you for all the time and effort you've spent on this over the last six months Wintertree. I've found your analysis invaluable in understanding what is and isn't going on.

The ramping up of concern over the Indian variant is pretty worrying. In this context, and as a parent of teen-agers, I'm struggling to understand the government's decision to go ahead with making mask-wearing in schools optional from Monday. What kinds of upside do they anticipate from this that makes it worth the risk of increased school-based transmission?

I fear they are using schools as a test vehicle, to see what effect the removal of a measure has.

The problem with this (apart from the obvious ethical one*) is that we have a fast rising new variant, so distinguishing cause and effect won't be straightforward.

* the argument will be that the young are at low risk from covid. Teachers are at greater risk, but teachers have been invited for vaccination this week. That only leaves parents at risk...

Another heartfelt thank you from me, currently flaked out feeling a bit ropy from my second jab yesterday. Cheers! 

Thanks for the map.

I found it really interesting to think about what the map "should" look like if this was contained.  IIRC you've a background not unrelated to nuclear physics?  There-in lies one way of looking at this; bear with me...

In nuclear physics, force mediated by Pions (π± mesons) are hyper-local, decaying far more rapidly with r, the distance from source, than the 1/r^2 of the electromagnetic force.  This is because as well as the radial dilution with distance that the mediating pions share with photons due to the geometry of space, the pions have a finite and brief half-life in time; as they travel outwards from their source this corresponds to having in number over distance from source; this provides an additional exponential decay term to their potential and resultant mediated forces.  The magnitude of the potential from a point source decays proportional to exp(-kr)/r, where k is some constant - the Yukawa potential.  

Interestingly, this potential also arrises in chemistry where there is a point source flux of some compound A in to a volume of buffer B that undergoes strong forwards binding with A.  Under this, so long as buffer isn't exhausted, the concentration of free A develops into a static form with a radial decay with distance from the point of inbound flux, and the radial concentration profile is given by the Yakawa potential - there is a 1/r term from the radial dilution coming form the free diffusion of A over time and distance, and the exp(-kr) comes from the exponential-absorption-over-time mechanic of the forwards binding buffering mechanic translating in to a spatial dependancy though diffusion.  It's simple to show that the Yakawa potential satisfies the mass balance equation although it takes someone smarter than me to do the derivation without cribbing it.  One point to recall about diffusion; the rate of increase in the diffusion of a slug of stuff (measured by the mean square displacement of its component) increases linearly with time regardless of the number of dimensions it diffuses in; adding dimensions just increases the linear rate of diffusion.  

So, the virus is perhaps not dissimilar; if we imagine a small region (MSOA, say) with loss of control of cases through locally specific factors, it is going to act as a point source of cases.  These spread outwards radially through human interactions and travel, and to a crude model that's going to be a diffusion like process [*].  If we are lucky and the locally specific factors enabling growth do not diffuse, then the cases seeded elsewhere by diffusion will go in to exponential decay, and a "mass balance" is reached where, if the central rate of case generation remains constant, there is a decay in case rates with distance that is of the form exp(-kr)/r with distance from the epicentre.  Double cases at the centre, and cases double everywhere within this "influence function", as both the exponential decay rate and the diffusion mechanic scale linearly with the central value.   As long as various assumptions are met, the influence functions of multiple discrete sources superpose at other locations. 

[*] - the failure of lockdown to the Kent variant spread faster than a diffusive process; there, new centres of infection were being established and added to the entropic force driving the new infection north and west, presumably that goes some way to explaining it.  If however locally present are behind the failure of lockdown in epicentres and exported cases are driving the rise elsewhere, we might expect a slower time mechanic closer to that of diffusion.  

So, that gives me something of a theoretical basis for what to expect of elevated cases around hot-spots even if the other areas might lack the capability to sustain growth without an influx of cases from the hot spots.

It's somewhat encouraging that whilst the radial fall-off of colour on the map you posted looks linear with distance, it's got a pseudo-log colour map.

Rumour has it however that the mechanic of a pandemic at the societal level is a bit more complicated than nuclear physics, and that the ream of assumptions I made (some not listed above) are by no means defensible.  But I think it sets a criteria by which to start judging the "bleed through" from hotspots.  If it start to get anywhere like linear with distance => bad, and if neighbouring regions start to reach the same high levels => very bad.   When I put it like that, they sound like bloody obvious criteria...

I'm not sure that long ramble actually adds anything to the understanding, but I've a soft spot for the Yukawa potential in the life sciences.

I'm not really vey clear in the lags on the variants data, it would be nice to see it regularised by-sample-date.  You're right thought that it'll take longer for the media coverage and the last boots-on-the-ground measures this to loop round to numbers being released.  Still, I'm hoping to see headline behaviour drop on to a lower rate exponential over the next week.   

Indeed - hopefully it helps nip this in the bud now, and if it doesn't, given that absolute rates are still very low, a lot more people are going to have meaningful vaccine protection before the majority of people face the prospect of infection.

In reply to captain paranoia:

I assume it was related to complexity in the IT systems but really don't know.  Glad it's sorted though.  It would be nice to see a paragraph released detailing any artefact it introduces to the time series.

In reply to Andy Johnson:

In reply to flatlandrich:

 I had by first one this morning and I haven't felt this bad since the day after my 21st! Don't want it to be in vain!

Mine first dose was 9 days ago; the next day I had a revulsion at the idea of alcohol that I've not had since the morning after the morning after a rather excessive night out around 15 years ago.  9 days later and looking at the bottle of Port Charlotte is enough to turn my stomach.  Worrying.  At least it didn't put me off condensed milk. I've really gotten in to making my own custard lately - barely more hassle than powder, and far better.  I've been wondering about making a duck egg custard with 50/50 condensed milk and single cream.  That should fix me up. 

The ways of government decisions have remained largely a mystery to me throughout; I've found things easier to understand since late December, and I'd been hoping to see some measures retained on a local basis tomorrow in response to this - masks being once clear example that shouldn't have perverse consequences (unlike closing local pubs etc...).

In reply to various:

Thanks for the thanks.  There's been a lot of really good input from many others, both on local factors, other data sources, interpreting it all, digging in to details and review and checking of my outputs, so the thanks go far and wide.  

My thanks as well - unfortunately I suspect you will still be doing this when you get to the 1 year anniversary.

Whilst the pandemic was "bad", I'd look at lots of detail on these threads - trying to understand. When the pandemic was (rarely) "good", a quick skim would tell me whether I needed to look in more detail. At the moment we're sort of in between those with a worried but let's wait and see.

If I want to know anything about how the pandemic is going in the UK, then the first place I look is your series of threads because I know that if there's anything important/urgent going on, then either you or one of the other knowledgeable people who respond will have already discussed it or linked to decent sources. Much more reliable and informative than news sources. 

The only difficulty I see is the need for automation; tests were uniquely bar coded, so no data entry would be required. Having to enter the hand-written sample date would have required additional time and effort.

This might be my all time favourite covid post so far! 
 

I think even if the ‘model’ is seriously underdetermined and simplified it holds as having the right form. Therefore it’s at least a useful benchmark.. would be interestign to see a plot of ‘Yukawa constant vs. Time’ for the hotspot MSOAs (in the spirit of plot 18) but that might take quite some work!

Whilst obviously and indisputably true, it's great that people are coming forward, I wonder if the case for 'surge vaccination' i.e. opening up to all ages in hotspots might be a bit more on a knife edge.

To put it another way, if I asked you to breed a vaccine escape variant, what would you do?

Also happy half year and big thanks

This remains my go to source for covid information. Yours and a number of other contributer's efforts are very much appreciated. Maybe the fact that you (and others) spend your own time disemenating the knowledge and information in an easy to understand format is the good news that should be highlighted in that other thread

There doesn't seem to have been any noticeable change in the "Daily change in reported cases by specimen date" data. Today and yesterday, there were as usual in recent times only a very small number of cases added more than two days ago. Hopefully therefore any artefact will be in the noise.

PS I'm intending to reply to your other post when my son has gone to bed and I can give it some more thought

Big day tomorrow - over 70% of adults should have had there first dose of a vaccine.

The uptake by age group is great - close to 90% in the over 45's - and of late there appears to be increased rate of first doses administerted - Is it conceivable that we can get >90% of adults vaccinated? If so that will be a massive public health success.

Yeah, probably: https://ourworldindata.org/grapher/covid-vaccine-willingness-and-people-vac...

I like that plot. It's one of the happier plots.

Wow. I'm half French (and vaccinated) and the scale of the difference between the two countries is, frankly (no pun intended), surprising.

I'm going to try and make a couple of hours later this week to sit down and have at it.  There's an MSOA shapefile from the ONS (last update 2011??), and there's a population estimate for 2019 as well as a relevant Covid download from the dashboard, so it should just be a case of re-jigging the code from my distance/time analysis last December to work with MSOA level data, not UTLA level data.  I figure the thing to do is to fit the Yukawa potential with a least squares fit to the cases/100k data, but weighted by the absolute population numbers at MSOA level.

In reply to captain paranoia:

Indeed; and these could be bounded by some basic sanity checks as you noted above.  Still, things that sound simple don't always turn out simple for big central IT projects...  It seems from the outside that the flow of information through testing has really improved in recent months.

In reply to Si dH:

Is that a copy of Krane being dusted off?  There's clearly a vast gulf between reality and a simple reaction/diffusion model on in homogenous and isotropic space, but it's nice to have some vague idea of what the mechanic might tend to under ideal conditions.

In reply to Dr.S at work:

It's been really reassuring to see that the uptake remains high as the age drops. Not just a public health success but a success of science and medicine over the tide of subversion that's been pushed this last year.  Good on the public for sticking with it.

In reply to Michael Hood:

I hope not; one way or another we have to break the back of this problem before next winter.   I I just think the consequences of having no practical choice but another significant lockdown are just too much - there's no more give left in a lot of the people and systems badly stressed during the last year, the consequences of going through it all again will be that much worse.  So, by necessity we just can't let it happen.  Thankfully the vaccination continues at pace, and more production capacity is coming online within our borders over the coming months (I'm particularly interested in the Novovax one), and adaptions for variants are happening.  As well as that, hopefully a couple of new therapeutics will be approved by winter that will reduce hospitalisations and reduce severity for those who still go to hospital.  

It's going to be a funny winter; the 2021/2022 flu could be interesting.  I've not kept up with the news on the antipodean forecast for our vaccine orders.

In reply to Longsufferingropeholder:

The survey results on vaccination attitudes in Western Europe really surprised me when I first saw them.  It's not great news for the UK that there's so much hesitancy near by.  

Reactor physics (and thermal hydraulics) rather than fundamental nuclear/atomics physics.  We spend a lot of time worrying about neutrons and their interactions but not so much about other stuff of similar size, so to be honest by knowledge of pions is minimal.  However I understand your analogy and there are definitely aspects of neutron physics that are very similar (and I think I can extend the analogy a bit).

[This post will be long so anyone not interested in a theoretical analogy between neutron physics and covid spread might want to look away now   I doubt that it will add any real understanding but maybe if I’m lucky there will be a piece of new insight for someone.]

I think your theory that covid might spread like a function of a Yakawa potential makes a lot of sense if you consider a source of positive cases in a hotspot that is surrounded by a homogenous land area with initial low case rate, uniform population density and propensity to generate new cases.  You would expect to see a map of covid cases that reduces as you extended out radially from the source and I can see that with a fixed size of source it might well drop off as you describe.  Of course the complication is that with time, even if the magnitude of the source is controlled to a steady level by local measures, it isn’t really a point and the geographic size of it increases/spreads.  The newly infected zones themselves act as sources for areas further out and this might compromise your model.  Either way, the resulting map would theoretically look like a series of gradually-expanding concentric rings of colour with each ring having a lower case rate as you move outwards.  In reality though, as we know it doesn’t look like that – a hotspot causes infections nearby, but we also know that some areas nearby get away fairly lightly, and that infections can spread long distances if, for example, an infected person jumps in their car to visit family across the country.  In practice the infection rate map looks very messy.

So how can neutron physics be of interest here?  Well in reactor design we have a similar problem.  It’s really important to have a good idea of neutron flux everywhere in the core in order to control both heat generation and reactivity.  However, we have to incorporate lots of different materials – fissile material like uranium, coolant, moderator, neutron poisons, structural materials – which all interact with neutrons in very different ways and which cannot be in a homogenous mass.  We deal with this in physics using what we call cross sections (I think you probably know what a cross section is but I’ll try to explain for anyone who doesn’t.)  Basically these are measures of the probability that a neutron passing by a particle of material undergoes an interaction of a given type (eg a fission, a scattering or an absorption changing the material isotope) with that particle.  We measure neutron flux in units of n/m^2/s and cross-sections in m^2; multiplying flux in a given bit of material with that material’s cross-section for reaction X tells you the number of reaction X occurring per second in that place.  To keep this analogy reasonably close to what we are talking about…If you create a single point source of neutrons that is surrounded by lots of heterogenous materials that are not initially sources themselves, then the reaction rate produced in a particular location of interest is a product of the flux that reaches them, <i>and</i> the material’s cross-section for that reaction.  The flux that reaches them will be a function of distance from the source (~1/r^2) but <i>also</i> what materials are physically in between the neutron source and the interactions neutrons undergo in them.  If there is a lot of shielding material in the way then the flux will drop off strongly like your Yakawa behaviour.  If it’s a vacuum then to a reasonable approximation the flux could just drop off ~1/r^2.  But, if there happens to be a lump of plutonium lying around that wasn’t previously critical, the fissions that then take place in the plute might actually result in an increased neutron flux on the other side of it such that net reduction from the source was < 1/r^2 – or, if the new stream of neutrons from your source was enough to turn the plute critical, it would create an entirely new source.

So what does this mean for covid spread?  Let’s equate a neutron to an infected person leaving the hotspot, a material type to a particular place in the surrounding area, a reaction rate to be some sort of interaction the infected person has, which could cause more infections (you could define lots of different reaction types, such as (A) causes a super-spreading event in a bar, (B) infects a family member in a house, (C) realizes the place is a dump and turns the car around, (D) falls off a cliff), and a cross-section to be the probability that a particular type of interaction occurs when an infected person enters that place.  So, an MSOA with lots of bars and restaurants would have a high cross-section for super spreading events. An MSOA full of dense multigenerational households would have a higher cross-section for household spread to family members.  An MSOA with lots of crags would have a (relatively) high cross-section for people falling off cliffs.  Importantly then, the resulting rate of infections in MSOAs surrounding the hotspot will be a function of their distance from the hotspot, <i>and</i> their particular propensity for different types of infection events (especially those events that produce lots of new infections) <i>and</i> the nature of the MSOAs in between them and the hotspot (an extreme example of this being two places that can see each other across the sea).  When we look at the infection map and try to think about whether something is surprising or concerning, I think we should consider these things, not only the distance.  Different factors will be important in different cases.

(Incidentally I’m sure it would be possible for a committed expert to create a Monte-carlo model of geographic covid spread using this approach, similar to how Monte-carlo models are used for reactor physics and other applications.  I have no idea if anyone is doing this.  I’m sure it would need a lot of CPUs and have to contain a lot of poor assumptions though.)

If I was to apply this to the current situation in Bolton…well obviously on average we expect rates to reduce with distance from the centre of Bolton.  I might expect covid spread out over the Lancashire moors to decay pretty fast in the way you described.  But I would not be surprised if the decay over more densely populated areas was slower.  I would also not be surprised to see hotspots develop in locations apparently separated from Bolton by areas of low prevalence but which have high propensity to events that will cause outbreaks of a few infected people come in.  For example (with reference to the map I shared earlier), the isolated small outbreak in Preston could easily have been caused by someone from Blackburn or Bolton rather than someone who lived locally.  And I would expect deprived or densely populated areas of Bolton and Blackburn to pass infections between each other faster than I would expect either of them to pass infections to the sparsely populated bits in between (or to Formby…how that ended up as a hotspot of the Indian variant I would be very interested to know…I would not have predicted it from my neutronics analogy!).

I do not know Bolton really at all.  If I had lots of time it would be interesting to know the demographics and nature of each of the different MSOAs and think about whether this approach could have predicted which MSOAs were likely to see cases rise first once the outbreak started in the area south west of the centre.  Having spent a lot of time staring at covid maps over the last 6 months, there is definitely quite a big randomness element.

I’m not sure how obvious or intuitive this all seems (possibly very).  I think if you have a go at applying the Yakawa potential theory you described, it is worth looking at your findings in light of the factors I described above.

I quite enjoyed writing this, but if anyone reads this far and managed it comprehend it, please give me a like so I can pretend I didn’t waste 2 hours of my life

Talking of winter, thanks to the vaccine rollout I suspect we’ll get through without a lockdown / significant restrictions, as long as vaccine resistant variants don’t take off. (key caveat - too early to say at the moment).

What concerns me for the longer term is high levels of vaccine hesitancy in some other countries, France being a prime example. Perfect breeding ground for variants, though they might get to herd immunity through vaccination and infection eventually... I hope vaccine hesitancy will reduce over time. 

You forgot to add [name of a place you don’t like] would have zero Covid because no one would go there because it’s a dump.

I had my first jab this week at my local vaccination centre where all 30 - 40 year olds in the area have been invited.

The uptake is massive, a combination of wanting protection, the desire for foreign holidays and also not wanting to be the odd one out amongst friends seem to be sustaining demand lower down the age demographic. Speaking to a few younger colleagues in their late teens and early twenties I am feeling very positive that uptake will be high across all age groups.

Just catching up on Indie Sage after an emotional BMC AGM weekend.

youtube.com/watch?v=QHIQa8KEEw8&

From 13 and a half minutes in they cover the variants situation quite well, including S-gene drop-out, which gives more recent comparative variant information than gene sequencing (where data is about 2 weeks old). Overall they show the main India variant is already almost certainly the main variant in London and the NW and very likely growing exponentially (approx doubling in a week) in the community.

One point that surprised me on the vaccination data is where they show very high take-ups for first dose (100% for one age group band over 70). Does anyone have any information on what happened with anti-vax sentiment as I expected percentages to be high but nothing like this high. It's good news given what we may face soon but 2nd dose data wasn't as good.

Yes, this was strongly suggested by the s gene data in latest variant report submitted to SAGE on Thursday. The total regional data doesn't allow a proper breakdown though, it's obviously much more prevalent in a small number of towns with high case loads than it is generally in a whole region at the moment.

Well, it's obviously not 100% in any large ish group. If you look at the data shared in last week's thread on Bolton you'll see that overall vaccination rates in vulnerable groups in the key parts of the town were still fairly low. Overall net take-up across the country has been great though, I guess the anti vaxxers' bark is louder than their bite.

Some data on vaccine acceptance here

https://ourworldindata.org/grapher/covid-vaccine-willingness?tab=table&...

I don't quite however follow as apparently the UK is vaccinated more than those who are willing

I think that's a great analogy to think about things with - the cross sections are a great way of grading the risk of an area, and this recognises the potential for an area to be "high gain" if excited by externally supplied infection; as cases become almost extinct in many areas, there is no longer any local measurement of, or feedback on R in those areas until a case lands.   Something several of your points hark to I think.

There's a divergence in that the cross section doesn't just "catch" external cases, but sends people out to get infected then brings them back but that I think doesn't affect the basis for the model.

As a basis for simulation I was thinking "I bet those clever reactor types have made some good finite element simulations of all this, applying that basis to 2D would give a refreshing break from all the Monte Carlo Covid modelling..."

The way I see the relation between this kind of geographic model and the Yukawa potential is that the later may arise in terms of analytical solutions to the former in some specific circumstances.

Yes, my noddy model was assuming a fixed R >> 1 in the hotspots and a fixed R < 1 everywhere else.  In reality, R is not so trivially distributed.   

I've never been sold on the really high end modelling side of this - a coarse model is very powerful at getting a feel for the situation and the sensitivity to different parameters.  Using these models can help the people working on this to develop a rational feel for what's going on, and how things might respond to changes to different parameters.  Fitting different models with few free parameters (e.g. vaccine evasion vs more transmissible) to reality to see which produces the most reality-like data is a powerful technique, but...

A really detailed, fine grained model must contain so many assumptions and unknowns that it can I think always be over-fit to the available data and so has no predictive power, yet many people when confronted with the more clever, more detailed, more fine grained model will automatically assume it is better and more predictive than a coarse one.  I also don't think all fine grained modellers are very aware of the number of implicit free parameters that arise in the model by trying different things against the data until one works; and that this is not always accounted for in terms of determining the risk of over fitting.

I see the real value in the detailed models as a tool to evolve people's thinking and helping develop the lenses through which to examine the data.  I'm a bit of a stick in the mud about modelling though; and don't get me started on "AI" in the form of MLPs...

What your post really cuts to for me are the network effects in transmission; the mechanic of the virus is all about the networks, and the right changes to the networks could be very powerful control measures.

Surveys, ey?

In reply to Offwidth:

Perhaps it'll turn out the anti-vax sentiment was a very small number of people making an awful lot of noise - mostly on social media - where our oldest cohorts are least active.  I also suspect that people in those cohorts know far more people hospitalised or killed by the virus than most people pushing an anti-vax message.  

In reply to Misha:

I think that as things stand, everywhere has to get to board spectrum herd immunity eventually.   By "broad spectrum" I mean our immune systems know about many of the other proteins in this virus as well.  At which point, variants become just the normal way of life as with many other circulating viruses.

In the next couple of decades, we approach a key juncture for the species I think - and its coming is accelerated by the research response to Covid.  What do we want our future relationship with viruses to be?  Do we want to start ridding the species of them; perhaps starting with the early Mars colonies, or do we embrace them as a chaotic aspect of the evolutionary process that we tame and adapt as we have the plants and animals?  What would the medium and long term ramifications be of elimination?  Would it be a disaster, rendering immune systems ever less challenged and developed, so that the next zoonosis has species ending potential?  Or would it be a key step towards longevity?    Evolution is increasingly subverted by technology, yet over sufficiently long time scales viruses have almost certainly been critical to the realisation of much life as we know it.  Perhaps eliminating them from humanity signs our eventual death warrant as a species.  

What would I do?  

Culture some cells in the lab and passage the virus through the system as rapidly as I can, using immune titration assays to select the most evasive candidates for each successive generation, whilst adding a bit of mutagen to speed things up.  Might even try and recruit a few structural biochemists to give suggestions on what changes to make to the spike protein, although I think there might be some control in place for people ordering certain RNA sequences beyond a minimal length right now...  

Alright, not *you*, then. What would *one* do with no lab, only real world dials to tweak, given the rudimentary policy tools available to a local mayor and health trust?
Direct your available vaccine at an area with a high and growing infection rate, injecting as many infected, or soon to be infected, people as possible? And incite resistance to any form of lockdown? Yeah, probably the recipe I'd think of too.

(Let's just try not to think about Indian variant mk2)

The question on your Yakawa theory, is what does different levels of lockdown do to it?  I suspect when a region is fairly locked down the seeding is very geographically restricted - people are going to the corner shop on foot because they're WFH rather than doing a shop at out-of-town tesco on their way home from work. They're only meeting people in their locality and not doing big drives - climbers in the peak will be doing grit&lime with other peak climbers and not going to Wales / Cornwall and meeting mates from the lakes there. People are not going for a sunday drive in the next county and meal out, or a wedding halfway across the country. They're doing jobs in their locality if they're tradesmen. Come out of lockdown though, and all that seeding restriction goes out of the window. Instead of being full of Snowdonian people, Pete's Eats will be full of climbers from all over the UK.  People will be travelling all over the place for weddings and other family gatherings they've not been able to have during lockdown.  They'll be going for staycations in the tourist areas. Tradesmen will be going all over the place to do up second homes.  People currently WFH will be back in the office. What's currently contained won't be. Welcome to last autumn people .

The problem is that going down a vaccination route is too slow. We know that people see vaccination as a personal shield as soon as they've been jabbed and slacken their behaviour, so vaccination isn't going to contain any spread, nor give more people immunity than are infected. The only effective control measures in the short term will be physical restrictions.

The other complication is that 'R' will not only change with the physcial constraints (and a whole population R will inevitably slowly decline with time) but also,  with behavioural changes. Thus, we would hope that as cases rise personal experience would modify behaviour to reduce transmission (rather than either a stuborn lack of change or a fatalistic relaxation).

As of today, it probably blows it out of the water for two distinct reasons - one is a significant increase in long distance travel combined with socialisation as you give, and the other is that this analytic solution is firmly rooted in conservation laws, and so applies where the prevailing mechanic away from the point sources is exponential decay.  If we switch to a prevailing mechanic of R~1 or R>1 it no longer applies.

In reply to HardenClimber:

Indeed.  The data from the rises last summer/autumn suggested to me quite strongly that local R drops in response to rising local prevalence.  As well as behavioural, it could be progressive exhaustion of the reservoir of susceptible people (in terms of risk of catching and transmitting the virus) starting from the highest risk end.  

Yes, although the prevalence is still low (< 1%) in the bad areas, and hopefully the doubling times are going to be slackening off a bit, so the vaccine will kick in before many people would otherwise have caught the virus.

I hope the messaging over the lack of an immediate "personal shield" is particularly clear in the surge vaccination areas.  There were some really good numbers being reported on the BBC today with regards people going for surge vaccination in Bolton.  One stand out item in the government's description of measures is that they're going in to work places where people are unable to take the time off.  It's a damming indictment of inequality and worker's protections that this situation exists, but acknowledging and accommodating it is better perhaps than the alternative.

Perhaps the difference between what people say they'll hypothetically do and what they actually do? That or the numbers linked include the young, a group perhaps more exposed to campaigning internet bullshit but the vaccines are currently only going to older adults, a group certainly more exposed to death by covid.

jk

It's government cashing in on a bad news story to get people vaccinated. Better in the long term, assuming they can retain the physical restriction required to restrain spread until the vaccination effect kicks in. Some tricky balancing to be done!

Certainly looking forward to finding out whether the uptake remains high down the ages. There is a strong suspicion that the ~10% who might not want the vaccine are the same ~10% who get most of their news from social media, who typically aren't drawing their pension. But so far we've been surprised in the best way possible at every stage, so we can hope that continues.

Hmmm.

Just had a look in a coffee break...up to over 2900 today. Must have been some weekend lag to cause the jump from yesterday I think. There's a long way to go until Thursday's gov.uk update. Definitely looks like doubling time is under a week still, although surge testing is going on of course.

The map at the link below is now up to 8th May and shows more developments too. Several places now over 80% Indian variant, from 30-55% two weeks earlier. And quite a few new spots popping up.

https://covid19.sanger.ac.uk/lineages/raw?show=B.1.617.2&colorBy=p&...

I know a couple of my more pro-woo mates who've been busy repeating covid conspiracy nonsense this last year have none the less had their vaccines. I think the desire for normal life and given how high uptake seems, social conformity, will be quite the motivators.

jk

That's the hope, right? I mean, if they're easily influenced then a bit of peer-pressure should work its way effectively down the ages. And if that doesn't work well enough then the ones that are left probably have a big overlap with the set of people who would get it if they weren't allowed into the pub without it. The UK is still holding that card. Germany played it already with some success.

Hopefully it's not really that much more competitive than 'Kent', what we're seeing is largely the result of a mad rush home to beat quarantine followed by surge testing to clean up the ensuing mess. 

jk

Between pub and gym restrictions I think you'd mop up most of them

jk

If you had an optimistic hat on, the provisional window for Bolton is stabilising, Bedford is heading for decay and Sefton is in to decay.  

https://coronavirus.data.gov.uk/details/cases?areaType=ltla&areaName=Bo...
https://coronavirus.data.gov.uk/details/cases?areaType=ltla&areaName=Be...
https://coronavirus.data.gov.uk/details/cases?areaType=ltla&areaName=Se...

Hopefully they're going to undergo the same sort of flash-in-the-pan mechanic I'm optimistically reading in to the above areas; what happens with today's relaxation of control measures however is anyones guess.

Put the bulb selection box away and get a lava lamp for this week.

Yes, agree, I saw that same behaviour when I looked this afternoon. In Sefton (I'm not sure with others) surge testing started on 13th; cases actually peaked and started coming down before that. If I was being even more optimistic it shows the effectiveness of local measures before the national interest began - it wasn't well communicated, but there were reports of a gym/pool, a pub and a primary school all closing in Formby due to cases being traced there in the week leading up to the peak, which suggests some backwards contact tracing had been going on and action taken as a result.

Population numbers are estimates so the % won’t be spot, though you’d expect it to average out across 5-10 years. Whichever way you look at it, it’s very impressive in the older age groups. Those in their 70s are probably in the sweet spot of knowing they’re vulnerable but still being sufficiently mobile to be able to get to wherever the vaccines are being administered. 

quite a small sample size potentially, and its the share of those currently unvaccinated who would have a vaccine, not the whole adult population.

still, 35% for the USA!

Fair point. Interesting the percentage is dropping despite the unvaccinated pool reducing, presumably leaving the sceptical. 

indeed - ultimately the die hard anti-vaxxers could be all thats left and we will be down at 0% who would accept the vaccine.

Re: Bolton, this is from Bolton Evening News:

“Speaking in the House of Commons earlier, health secretary Matt Hancock said that there are now 19 people with coronavirus in Royal Bolton Hospital, and that most of them have not yet had the vaccine despite being eligible.”

This should be made headline news.

Something weird going on with the numbers. Total cases including LFTs seem to be steady or in slight decay. An artefact of the reporting methodology, the BH impact last week and surge testing? I’m reserving judgement until we get a few more days’ data but any decrease is a good thing.

Re Bolton, could be a combination of surge testing and behavioural change. Either way, promising but again let’s see the numbers for the next few days.

Loving the way the people running the local vaccination drive in Bolton have basically said we’ll find a reason to vaccinate you and stuff what the government says. Looking at the impressive queues, most people look under 50 and many are pretty young. I get that it may be better to prioritise second doses in areas of low prevalence but in Bolton  it’s a case of vaccinating whoever turns up and their dog.

Today’s unlock really isn’t helpful… a mooch round the going out area in central Brum around where I live showed a fair few people indoors. Not busy but I bet it will be at the weekend. Some people braving it outdoors but it’s pretty cold sitting drinking a cold drink. Had a couple of outside drinks just now - should have taken my belay jacket!

He also said earlier that 5 of them had had one shot of the vaccine and one person (can’t remember if they were additional to these 5) had had both but was frail. No info on when the 5 had it. If they had it more than 2-3 weeks prior to their likely infection date, that would be pretty concerning. 

It was yesterday:

https//www.bbc.co.uk/news/uk-england-manchester-57134500

PCR numbers look to be in decay too.  With the leading surge regions hopefully maxing its not that surprising perhaps - and now the cautionary note is that their decay from larger numbers is going to mask smaller but rising numbers in other locations from showing in the shape of the top level data.

Yesterday’s local radio mentioned that five pubs in Newcastle were opening at midnight this morning to let people start as soon as the law changed.  I’m not really sure leadership telling people to act responsibility is going to make much difference.  Still, every day more people are vaccinated, and the hot spots are hopefully responding to everything.  Interesting to read Si dH’s local beta on more proactive control measures early on; seems to be working.

Today for the first time (I lead a sheltered life) I had a conversation with someone who said "I'm not anti-vax but ..." and proceeded to tell me why she didn't want to be vaccinated (" ok it's a tiny risk but it just might be me"). I was horrified, not at what I was hearing (I'm not that sheltered) but at who I was hearing it from - university lecturer, well able to understand the basics about risk. However, despite all that, she has had it, albeit not v happy about it. I think perhaps social preasure changed her mind.

There's a danger we are an echo chamber on here sometimes and I think we need to be careful to give people we meet like that the benefit of the doubt. It sounds like she had genuine concerns but decided to put them aside and get it done anyway. I think we should be grateful for people who do that. There are many people in younger age groups (I don't know the age of your contact so don't know how bad her understanding of risk is) for whom the personal risk balance now is genuinely not clear now if they are only offered AZ. People should be really understanding about that. Hopefully they will be able to get an alternative if they want it because it's not good for society to pressure people in to having something that helps others but is not for their own benefit, they need to do so willingly.

I'll be online at 7am tomorrow to book my jab (37yo) and I hope it's a Pfizer but if it turns out to be AZ only at the centre then I'll have it anyway. If I had a teenager who was being offered a jab in the US right now or maybe the UK in a few months, would I think the same way? Not sure. I definitely wouldn't let anyone vaccinate my 3yo until there are a few more years of data. We are all somewhere on the spectrum of risk appetite.

Until I started plotting this stuff it'd never really dawned on me that the population is subject to such wide estimates.  The population estimates I've used from the ONS all date to before the pandemic, so they're out of date, and the pandemic in particular will have effected significant reduction to the oldest age ranges.  Then there was the suggestion a while ago across headline media ofd an exodus of 1,000,000 foreign born residents during the pandemic although calmer heads (and not media headlines...) suggest it's around 250,000 [1].

I find it vaguely unsettling that there's such uncertainty over what the population is; yet somehow it all works out.  Much like I find it a bit unsettling that nobody really knows if the planet is getting heavier (constant micro-meteorite bombardment) or lighter (Hydrogen outgassing, a bit of Helium outgassing and space launches.).  

[1] https://blogs.lse.ac.uk/covid19/2021/03/22/the-data-suggesting-a-million-pe...

Thanks again for another update, it's been a very worrying week with regards to everything that has happened but I am just about remaining optimistic about it all. 

The recent news from Oxford that the vaccines are still effective against the Indian variant has definitely helped, and I am getting closer and closer to it being my turn now.

Big news this morning though with regards to Pfizer, I would have thought they would have been able to have worked this one out sooner but what do I know?

https://www.bbc.co.uk/news/world-europe-57152719

Don't wait!

You probably won't see this in time, but if you do, get on it now! It'll let you book a while before the published time. All you're waiting for is the big rush.

Haha I was on there at 630 but it didn't work until 0705

The spread of the latest variant (which the vaccine appears to offer protection against) has been quite dramatic. The introduction to UK and failed containment (at a time when we didn't know much apart from that it might be bad) perhaps underlines our failure to sort Track and Trace etc.

The vaccine rollout has been impressive (and some other counries are catching up).

The way b.1.617.2 has spread doesn't bode well for when (and I am sure it will be when) a vaccine escape strain emerges. I still wonder quite how much commitment there is to acting early on sequencing (and international epidemiology). I do wonder if it has been seen as a bit of a hobby for the curious - almost a distraction from things like evermore testing which can be delivered at a profit.

(on the other hand Taiwan looks as if it might be about to demonstrate the problems with putting too much weight on societal control as oposed to immunological control).

Yes you're right.

Taiwan has about a 25 million population and so far has 2500 odd cases and 14 reported deaths. I'd wait a bit if I were you before pronouncing they have got things wrong. Until now they have shown strong fast integrated action can save months in lockdown and nearly all the health impacts, most of the social and economic damage; plus aside from avoiding all the deaths, suffering and economic damage, we in the UK have, they don't have massive increases in non-covid hospital queues and around a million with long covid.

They have clearly got a lot right, and things have worked well for them so far and at present they are way ahead of UK (but COVID has shown itself to be a great leveler). It must be a worry to have nearly a third of all your total cases in the last 7 days, with a non-immune population. As you say, it is too early to say what happens (and they will make a much much better job of control than the uk ever did, or has the incliniation to), but they seem to have been very slow with vaccination. 

It just seems unwise to put all your effort into one approach... time will tell.

Current numbers are only so a big proportion as they were so successful before, despite being one of the countries most at risk initially. They have made mistakes in recent months so lets see where it ends up shall we. It looks almost impossible for any developed nation to beat the UK and US incompetence in handling the pandemic pre vaccination arrival.

https://www.theguardian.com/world/2021/may/17/how-did-covid-slip-through-ta...

Sadly, it sounds like they are repeating our mistakes; putting some minor restrictions in place, rather than a hard lockdown. 

Will no-one learn the 'hard and fast' lesson?

If there is one thing that I have become convinced by in this pandemic is that (in the absence of vaccines) CV will get to us all eventually, control measures buy time but wont save anyone in the long run if treatments/vaccines dont come along first... I am certainly not going to criticise any country that has thwarted the outbreak at its onset, but they would eventually be sitting ducks with a population ready for an outbreak as soon as the conditions were right. Its why I am somewhat surprised places like NZ are not immunising their populations more rapidly...

Is it a rectifiable failure though?  It seems that the outbreak got some steam up before it came on to the radar; given the voluntary nature of engagement with test and trace and social / employment factors, might it be wider society that set up the failure mechanism?  It makes me wonder about the merit of random sampling as a real time disease surveillance mechanisms. 

The way b.1.617.2 has spread doesn't bode well for when (and I am sure it will be when) a vaccine escape strain emerges. I still wonder quite how much commitment there is to acting early on sequencing (and international epidemiology). I do wonder if it has been seen as a bit of a hobby for the curious - almost a distraction from things like evermore testing which can be delivered at a profit.

I'd love to read more about the sequencing program; the technical scale of it is fantastic.  What I don't know is how a couple of thousand sequences a day can be - or are being - meaningfully analysed in real time to inform the response.  I can see how an algorithm can be set up to look for known variants, and for known motifs emerging in new variants, and for growing clusters of previously unknown variants but I can't visualise expert human input handling that many sequences daily in a coordinated way; I sort of assume there's some hybrid pipeline of automation and human review but I'm in the dark.

A couple of posters have alluded to two week lags in sequencing results; I keep meaning to ask them their source for this as the technical part should complete in a day (~9 hours sample to sequence for an Oxford NanoPore?).  

I often wonder how much weight we put on infection generated immunity in November 2020 and January 2021; the initial fall in both cases (before the Kent variant, and then the vaccines muddied the waters of understanding) seemed better.  But then we'd reconfigured society so much compared to our initial wave as well.

Counterpoint is that we don't have years of data on life post-covid. Not that that makes the choice any easier. We all get to choose between the long term outcome of either a vaccine invented last year or a disease invented a few months before. Place your bets.

"18 May

Revision to historical case data in England

The introduction of a new system disrupted the removal of cases where a positive rapid lateral flow test was followed by all negative polymerase chain reaction (PCR) tests taken within 3 days. These 4,776 cases have now been removed. Newly-reported cases for the UK and England were unaffected. Historical published date totals have not been changed.

Newly-reported cases at regional and local authority level within England are calculated as the daily change in the total number of cases. This means that for 18 May 2021, these show significantly lower numbers or zero, and do not reflect the actual number of new cases reported on that date. Details of the changes and numbers of newly reported cases for 18 May 2021 are available in csv (wide format) and csv (long format)."

Oops. Might explain some things.

I don’t know the exact lag but apparently it’s over a week. First of all you have spéculent date to test result date - that can still be two days. Another day to send a batch to say Sanger. Then processing there. As you say, you’d think it would be faster. 

You are right but it’s best to delay and use the time wisely to prepare…

There's been another exciting update too; vaccination data now broken down by area.

That's tonight's browser tab mountain decided then.

Good spot! Interesting numbers. Can’t see the 6k this weekend in Bolton (actually 2.4K) but numbers are up certainly significantly in the last few days. (Edit: looks like Sunday’s data are still provisional so that would explain some of the difference). Perhaps the NHS trust area is significantly larger than the council area.

Birmingham on a measly 50% first doses, the lowest out of the 7 West Mids local authorities. I get the relatively wealthy Solihull being top at just under the national average of 60% but I don’t think we’re more deprived or ethnically diverse than the other local authorities - yes the 4 Black Country ones are significantly ahead. Only Coventry is in the low 50s. Not good...

I would expect the big cities to have lower %s due to younger populations but 50% vs 70% is a huge difference and the variation between neighbouring, fairly similar local authorities is stark. Feel even more lucky to have had mine in mid March (or perhaps it was because the general uptake was lower than expected...).

France going ahead with reopening, with 13000 daily cases. Really? Surely that's not smart.

There's a similar discrepancy locally between Liverpool (52%) and Sefton (69%) / Wirral (72%). I think this probably is mostly if not all down to demographics with big city centres containing a much greater number of people in their 20s and 30s. They are mostly under the same health trust and have access to the same mass vaccination centre so I can't imagine it's some major difference in performance. It does indicate though how much variability there is around the country and therefore how the moderating effect of the vaccine on transmission over the next couple of months is bound to be very variable.

Edit to add, some bits of London are still in the 30%s...long way to go there.

Is ok though. It’s on the amber list, so no one will go there, because Boris said. Except his mates, who will be within the letter of the law. Oh and everyone else. But that’ll be their fault. Right?

I mean..... You can read the politics how you want. Indeed I'm sure someone will be along with a link to a guardian article shortly that explains how it's England's fault. But really, France? Really? Cases only 2 doublings away from the peak and you're reopening?! Wtaf. Do they know something we don't??

Yes you make an excellent point. Apologies, my pre-caffeination reply was driven mainly by 3 hours sleep so may have been filtered less than most!

Nice to have a book keeping error that, when resolved, makes the situation *better*.  Perhaps this helps explain my worry that "cases in England are all over the place in a way that doesn't resemble statistical noise.".  

So it looks like total cases (and by extension PCR cases) are back in to decay.  Several of the big outbreak areas look to be under control; their fall in numbers is perhaps masking what's going on in Blackburn with Darwen which probably hasn't yet, and maybe masking other growth areas.

If this is a sign of how importation events are going to go, I'm not best pleased about...

In reply to minimike:

We've seen very clearly that PCR testing and home quarantine under amber listing does not work to contain importation events.   So, having a bunch of amber listed destinations just seems futile.

Dont worry thats been dealt with. People have been asked not to go to amber listed destinations so its going to be fine.

Answers on a postcard why if it isnt considered a good idea they arent simply red listed.

Sent from amber-listed countries...?

The Royal Mail does a 72hr delivery delay quite reliably for me.. should be long enough!

BoJo is saying don’t go on holiday to amber listed but you can go for family emergencies. The way to deal with that would be to extend the list of reasonable excuses or whatever it’s called in the relevant regs. 

New cases up ~17% today compared to last Wednesday. Seems like more than noise to me.

Anecdata: nipped into three local independent shops today. All had customers who weren't wearing masks. In one of them the person serving wasn't wearing a mask, but they were last weekend. Very concerning.

Apart from one day, PCR cases have remained in decay since the start of this thread.  But...  The provisional window on the data does not look at all promising at the UK level.  The main “outbreak” areas have peaked I think; now we’re seeing a broader rising baseline perhaps in many other areas.  The rise in that baseline is probably worse, masked (for now) by the decay setting in at the outbreak areas.

So I don’t think this is a great development but it’s still not panic stations; the rapid curtailment of exponential growth in the outbreak areas in response to control measures, combined with low vaccine uptake being a likely reason they were susceptible to outbreaks is kind of encouraging - we haven’t lost the ability to control this virus, just some of the will.  If we have to control it more, we can.

Jr’s school ask parents to wear masks for drop off and collection.  It’s basically pointless as it’s outdoors and not close contact, but it costs nothing to do as they ask, and it helps the more nervous people - and there’s a lot of nervous people after all the isolation of lockdowns - so it’s not a problem.  Since the last unlocking about half the parents have stopped wearing them.

At some point, allowing free circulation of the virus so long as the vaccine prevents severe illness is one exit strategy (likely the only one left to us now) - normalise the virus in to an endemic pest that updates immunity as it goes around and around changing as it goes, and takes its place in the necrotising pantheon as one of the many reasons people eventually die.  I think it’s way too soon for that step, but what I think doesn’t count for much in the world...  I’d be much happier if we waited until various immune modulating therapeutics were out of trials, as these will hopefully render the disease largely impotent if we do loose control of cases - even to an immune evading variant.

PCR cases seem to be plateauing again. Might be rising slightly once out of the provisional window. I wonder if we’re still seeing jitter from the methodology change, as well as the BH last week for the week on week comparators. A few more days should sort it out.

The dashboard map now runs to last Friday and is showing a lot of yellow in the more rural areas, as well as a few cities. Several other counties are only slightly above the ‘yellow’ threshold. So as you say, some areas are falling or holding steady, while others are growing, which means the headline numbers are hard to interpret.

From next week, we’re going to start seeing the impact of the 17th, though I don’t thing this will start feeding through in any significant numbers until the back end of the week. That will further confuse things as the headline numbers will reflect the impact of both B.1.172.2 and the 17th. Not helpful as ideally you’d want to use the headline numbers as a proxy for B.1.172.2 growth given the delays with sequencing (and of course not helpful as it turbo charges the spread).

Saw in the news that B.1.172.2 is now up to a touch shy of 3,000 cases, so it’s doubled again, albeit there may be some duplicates in the numbers if last week is anything to go by. See what tomorrow’s official update brings...

More food for thought. Blackburn vaccination rate for 50s is 95% and 6 our of 9 people in hospital have had one jab (doesn’t say how long ago and clearly small numbers aren’t good stats).

https://www.theguardian.com/world/2021/may/19/stop-blaming-vaccine-hesitanc...

Hancock said today the ‘majority’ of the 25 people in hospital in Bolton haven’t had the jab and ‘nearly 90%’ haven’t had both yet.
 

https://www.bbc.co.uk/news/uk-57172139

That suggests to me that 3 out of 25 have had both - perhaps the second one was recently but the first one must have been many weeks ago and as effective as it could get. Of course they may have been quite old - originally one person was reported to have had both and was described as frail. Also seem to recall that 12 out of the original 18 hadn’t had the vaccine - that’s a ‘majority’ but it’s only two thirds. Early days and low numbers but I do wonder about vaccine escape / efficacy. We could really do with more precise numbers, including when these patients had their jabs and their age range.

Nor is Bolton that far behind in the vaccination drive. Let’s take a slightly random date, 6 May - before the vaccination surge started and around the time of the infections which have been dropping through into positive cases in recent days. National first doses at 67%. Bolton 59%. Incidentally, Birmingham was only 48% and I imagine we will have had a fair few local people coming back from India. Did Bolton just get unlucky somehow?

It seems B.1.617.2 is burning through the c. 40% unvaccinated cohort but also sweeping up a few people who’ve had one or both jabs. I hesitate to extrapolate as the data are poor but it doesn’t look promising. At any rate, I don’t think we can say it’s solely due to Bolton’s low vaccination rate (which isn’t even that low) and it doesn’t bode well for other areas.

Could have only had it yesterday. In fact, given the surge vaccination and stuff, there a good chance they did.

theguardian.com/everything-is-always-somebody-elses-fault-and-getting-a-free-jab-that-would-have-prevented-this-is-too-much-to-ask

Absof***inglutely blame vaccine hesitancy says logic and common sense. 

What nobody's thinking about is how many vaccinated people were exposed and didn't get sick. Hint: not none. If the vaccination rate was 0% this would be a different game.

We've sort of covered this upthread. There are reasons why this was almost certain to show up in Bolton.

Apologies, this post is pure speculation because I don’t have time to dig out the trials data but.. Even if the vaccines are *only* 99.9% effective in preventing hospitalisation that makes the risk of that similar for vaccinated older people and unvaccinated under 40s.. so it’s not surprising that the hospitals have small numbers roughly evenly split between those groups..

This paragraph needs a really important caveat. You can't compare % who have had a jab as a measure of hesitancy. The offer isn't open to anywhere near everyone yet. When you look at how many have had a jab as a % of those eligible, those differences that look like 8% ish in your post are a lot higher.

It looked this way at the weekend but unfortunately the 1-2 days of provisional data are putting Bolton, Blackburn and Bedford back in rising territory. It's not a simple picture though, the weekend rates didn't look bad but then there was a big increase in Monday specimens. Tonight we should get fairly complete data for Tuesday and see whether Monday was just a surge/LFT spike but I suspect not.

Yesterday when Hancock announced the areas for additional surge testing, I had a look at the cases data for them - some of those are also in the early stages of an exponential rise.

The other point about Bolton (and I'm sure there is similar elsewhere) is that low vaccination rates amongst eligible groups were concentrated in the local areas just SW of the centre where the outbreak began and case rates are highest. We saw a map of this last week somewhere. The vaccination rates for the whole LA are not representative of that area at all.

The MP was making some points yesterday about lack of accessibility to a vaccine location for people in that area before the bus came along, and arguing that it's not mostly about hesitancy. If true it seemed a fair critique.

Having said all of that, I do think there is a huge slice of luck involved in where places suffer covid outbreaks from a low background and it can be difficult/divisive for people elsewhere to try to find something to blame in the local population, we need to be careful of that.

This is something I have s bit of trouble with. There's the social responsibility angle too. It's Bolton, not Antarctica. If people won't get the bus into town for their vaccine to do their tiny part in getting us out of this, after everything people have sacrificed so far, that doesn't deserve too much sympathy imo. Sure, there's the tiny few percent who can't for whatever reason, and I'm sure someone will be along with a link to a guardian article shortly to explain how it's England's fault, but for the most part, ffs. Deane and great lever are not remote, nor inaccessible, nor poorly serviced by transport links. I call bullshit.

Locally, earlier in the year people were refusing to go to the local vaccination centre (10-15 mins drive) as it was too far away, and bombarding the GP surgery with calls and complaints. The council had even put on a free bus service for the place.

You can call bullshit all you like but it doesn't make laziness or hesitantly true for most. Pretty much everything is worse in covid terms for those living in deprived areas.

Hancock's statements were ridiculous in scientific terms and I'd normally expect better than that from a minister. So saying "Absof***inglutely blame vaccine hesitancy says logic and common sense." is wrong for the same scientific reasons as Hancock; and linking common sense to logic is pretty dumb. The local view is very different:

https://inews.co.uk/opinion/bolton-indian-variant-problems-covid-vaccine-he...

My guess is the reasons behind the outbreak are a bit of everything and we won't know relative contributions for a while.

What we do know is the government management of travel from India was a disaster and the photos in 'inews' this morning of packed airport arrival queues from red, amber and green counties right next to each other is just nuts from an epidemiological perspective. Why haven't they just maximised numbers of immigration staff when airports and passengers have been complaining about low numbers and queues for weeks now??

https://inews.co.uk/news/heathrow-chaos-red-list-green-amber-arrivals-india...

Buses pose a not insignificant risk of covid transmission.

I wonder, once the dust settles, if there'll be a stab at estimating how much transmission happened on the busses.  Opening of ventilator windows on busses around me has been very poor.  

Another issue with travel is that it takes a lot of time, and not everyone can afford to take the time off work - recognised by the government who are now sending mobile vaccine units in to some work places.  

The whole madcap rush every morning to book a vaccine slot itself depends on people not being working at 7 am, or working in a job where they sit there hitting F5 on their computer for half an hour.  It's a bit like the designers of the system haven't considered people who work under significantly different conditions to themselves. 

So I agree that there's a social responsibility angle, but I'm not sure we've made it easy for people with lower wage and less secure employment to be responsible - continuing a trend that has perhaps contributed to some communities having much worse outcomes under the pandemic.  It may not just be some communities - the gender split on death is stark and I don't think it's yet decided how much of that is social and how much is intrinsic.

It would be really, really interesting to see a properly detailed, real breakdown of who is and isn't getting vaccinated. Not the guardian speculating and printing what its readers want to hear, but actual data. It's clearly being responded to but isn't obviously published.

Holy wowsers.  Some of that data just looks weird.  Still, it's not on the same high rate exponentials seen early on, and there's still some white between the outbreak areas on the MOSA map.

I was surprised that Bedford is only just getting added.  Having areas go in to surge testing this early on in their rise is a big change from last August.  There's a Monday spike in about half these regions as well - hopefully it is the surge testing.  It's getting very hard to get a feel for what the numbers mean lately.  There's still a falling baseline for hospitalisations I think from the previously falling cases, so that'll mask the initial upticks from hotspots; we know that's happening from news reports cited up thread.

Variants update with the de-duplicated numbers this afternoon/evening.  Still keeping a very cautiously optimistic hat on for now.  

In reply to Longsufferingropeholder:

I just edited that blanket comment as I felt it was unfair. Yes I'd agree some comments on the other side have been guilty of over generalisation. When we are talking individual behaviour and circumstances we need to be careful assuming everything is the same and any statistician will tell you what Hancock claimed is not statistically correct. In the end the virus arrived because of poor covid security for travel from India. The government were warned repeatedly of this but held out for weeks seemingly down to a PM trade visit. The photos in 'inews' today indicate arrivals infection control is a joke.

Perhaps that's in part because it's a damning indictment of inequality in Britain.  I can't now find the .gov.uk announcement I read on the measures to be taken in hotspot areas but they seemed to recognise specific issues along these lines.

I'll be double jabbed next week but won't going anywhere in a bus (or taxi) for a good while. Most people in deprived areas in our cities can't afford my luxury of alternative transport and the fitness to easily walk a few kilometres.

Mostly can't disagree with this. You could word it as "... arrived because of people traveling back from India into poor covid security". Each individual could have not done that, but the government (stupidly, for sure) didn't ban it. The line between individual and government responsibility is... somewhere. Seems incongruent to me that we would expect people to stay in their homes and give up their lives and livelihoods for a year (Edit: or quarantine in their home under fear of devastating legal penalty for 2 weeks) but traveling up to 2 miles* is an ask too far.

* This is a total guess. I don't know where the centres are but Bolton isn't a big place.

That’s the irony - the more you need the vaccine, the less likely you are to be happy taking a bus or taxi.  A reason why it’s been good to see mobile clinics in the hotspot areas.

Its okay, my peril sensitive sunglasses blocked that picture out so all is good in my bubble.  Otherwise I’d be firing off a long list of curse words both about the size of the red list queue, and about it not happening in a big army surplus tent somewhere else on the site.

Yeah the Heathrow pictures were baffling. I really hope that's a symptom of a reporter finding the photo that sells. If it's representative of a typical day then... bad things.

Is it "most" though? This is my point. I don't dispute that there are people who can't travel or can't afford to take the time. That this is more common in the deprived areas, also is fact. But is it 'most'? How many people could, with some, more, or maybe a lot of effort, get to the vaccination centres? What proportion is that? If "most" genuinely "can't", then I'm totally with you. But there's a spectrum of willingness and ability that isn't black and white, and how far up the grey we are, and in which demographics, is not so clear.

Either way the solution is to park up in a vaccine bus, so it's good to see that.

How did we get here again???

It's probably a bad day but there are warnings or reports (similar photos) going back to at least February.

https://travelweekly.co.uk/news/air/heathrow-warns-of-five-hour-queues-as-n...

https://www.mylondon.news/news/west-london-news/heathrow-airport-chaotic-sc...

https://www.theguardian.com/uk-news/2021/apr/29/heathrow-ministers-must-get...

The significant difference between areas is how big is the unvaccinated minority is.

In some areas the unvaccinated minority of over 50's might be 3%. In other areas it might be five times higher at 15% but most  (the remaining 85%) in that area are vaccinated.

Extrapolating from the majority in most areas to a minority in few areas is fairly pointless, it's a bit like extrapolating from me to an Olympic climber!

I agree - flood the area with vaccine buses, vaccinations at workplaces & places of worship etc, get GPs to make appointments, drop-in vaccinations at shopping centres and a free pint if you get a vaccination in a pub!

I think the only thing I would add is that I think many people in more vulnerable groups have felt scared of going anywhere they see as high risk over the last year, and may well see a busy bus ride as completely beyond a red line in risk terms. My next door neighbors (70s/80s) were originally going to have to travel to the mass centre a 20 minute drive away when rates were high, but I think they hung on a bit and were called by the gp. Unfortunately the bloke got covid before he was vaccinated...he was in hospital for around a month but is mostly recovered now I think.  I'm sure there are many like them.

Again, fully agree. Maybe I wasn't clear enough that I'm not denying these circumstances exist. There are also I'm sure many like my neighbours (70s one side and 60s the other side), who had the same dilemma and chose to do the 35 min drive each way (each, individually, not as couples, so 4 x ~1h round trips) to get the jab.
There are all sorts of circumstances that would prevent people, or make it much harder for people, to access amenities. But is that enough of a proportion to explain all of the difference in uptake? That's where I'm stuck. "Many", sure, ok. But enough to explain the differences we're seeing? hmmm.
Definitive numbers on can't vs won't vs just haven't would be amazing.

I bet the same people will happily jump on a bus to go on holiday given half the chance.

What's government going to do in 2 week's time when the numbers of areas in the early stages of exponential rise have increased massively due to unlocking? 'Surge test' the whole nation?

I'd be fairly surprised if they do anything nationally in two weeks. If by the 14th June the hospital numbers are either flat or only slightly up and if deaths are flat, I expect them to continue with their plan. If hospital numbers have risen a lot and there is an indication of vaccine escape I think they'll reverse one step in the roadmap, ie shut indoor hospitality and hope that's enough, with further restrictions once it's too late if they are wrong. If by 14th June hospitalisations rise significantly but there are very few deaths and the vaccine escape rate is within the expected bounds then I think they'll keep the status quo for a few weeks longer.

I appreciate your question might have been rhetorical.

Vaccinate the whole nation.
Well, ok, that'll be in 6 weeks time. But you get the idea. We're getting close now, and in the good-news-story-of-the-day, look to be speeding up.

In the lowest decile by income, nine times as many households (65%) are without a car/van compared to the highest decile by income (7% without car/van).

Vaccination rate also correlates with income so vaccination rate will correlate with car ownership.

It is unproven but not outrageous to suggest the relationship between vaccination rate and car ownership is causal and party explains low vaccine uptake in some areas.

https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfin...

There's loads of real facts on vaccine hesitancy too that could be used as a proxy for vaccination rates vs ethnicity, age, housing (rented vs owner), deprivation and income etc. 

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

I was wondering about feedback loops as well; in an area where fewer people can practically access vaccines, there will be fewer peer “role models” flying the vaccine flag - talking about it in person or on social media; so other people for whom it’s not such a priority have less direct exposure to the subject.  There’s been something of a community buzz in my circles around vaccination; I expect that helps engage some other people who otherwise would be more hesitant.  Take the buzz away...?

It's not outrageous at all. But it's the partly I'm pointing at. How big a part? And what % can't access a vaccination appointment some other way? Bolton's not big. The areas we're talking about are very close to the centre.
There's a considerable number of people in this unvaccinated group. It must be a much bigger number than the subset that can't access amenities. It just has to be. Otherwise we'd have other problems.

Rather than fewer (85% vs 97% vaccinated) positive role models amongst peers, five times as many (15% vs 3% unvaccinated) negative role models amongst peers of people who can't overcome the lack of transport, money for taxis, time off work or gumption that comfortably well off professionals working at home think are trivial.
 

Perhaps vaccination is like any project - 90% of the effort is for the first easy 90% then another 90% of the effort is for the final difficult 10%!

Just checked where stuff is in Bolton (Things have moved around a bit since I was there), and the car thing is kind of moot; the place I'd park for the Lever Chambers vaccination centre is basically in Deane anyway.

https://www.bbc.co.uk/news/uk-politics-57186059

Failures in England's test-and-trace system are partly responsible for a surge in the Indian variant in one of the worst affected parts of the country, a report seen by the BBC says.

Well there you go...

Sounds like T&T is having procedural issues and is working nowhere near well enough for either these variant importations or IMO the consequences of May 17th. 

My immediate thought after hearing the BBC leak based news story this morning is Matt Hancock will almost certainly have known about this latest test and trace failure when he was blaming spread of the Indian variant on (statistically unproven) vaccine hesitancy in Bolton. It looks like a classic 'dead cat strategy' political blame deflection.

Maybe I've missed the data somewhere else but I've not seen vaccinations by age group before.

https://www.travellingtabby.com/scotland-coronavirus-tracker/

First jabs seem to be progressing well to ages 18-29 (22%), 30-39 (32%) and 40-49 (70%).

If these 18-49 year olds were vaccinated earlier in JCVI priority groups 1-6* they would almost all have their second jabs by now???

NB a few question marks there.

*1-6 includes ages 18-49 who are vulnerable due to health plus those working in front line health & social care 

Low number of 2nd jabs suggests recent vaccination and that hints we will have good vaccination uptake for ages 18-49 

It will be interesting to observe if incredibly high 90-100% vaccine uptake percolates down though the age groups.

Other things I learnt from https://www.travellingtabby.com/scotland-coronavirus-tracker/ .

Pfizer - 1.9 million jags 
AZ - 2.9 million jags
Moderna - 18 thousand jags - current Moderna supply is insignificant on a national scale and explains why approval of Moderna made no difference

The highest hospital admissions is in ages 25-45.

...has made no difference yet, maybe?
When were the first deliveries? And what happens next?

Moderna was first deployed 7th April

https://news.sky.com/story/covid-19-first-dose-of-moderna-vaccine-given-in-...

Fingers crossed you are right to be more positive about Moderna supply!

I interpreted it more as a just coming on stream thing. I got jabbed with it the other day.

Every little helps. 

Totally agree with: “My guess is the reasons behind the outbreak are a bit of everything and we won't know relative contributions for a while.”

But the article chooses to miss out some key issues around attitudes/behaviours in some of the Muslim communities in the areas frequently hardest hit by Covid. There is a long list of Covid breeches including weddings and wedding receptions, funerals (Blackburn last year - 230 attendees and the Imam had Covid). Councillors travelling to Pakistan for weddings/funeral. When Cummings acted like he did he got totally slated, but these ‘leaders’ have got away with it, setting an appalling example. So totally agree about poverty, low wage economy etc but some of the ‘bit of everything’ is about behaviours which I do not think have been adequately called out for fear of appearing racist. 

The car issue is a red herring in my view. Whilst car ownership might be below average, there’s a very strong probability that someone without a car knows someone who does. Furthermore, there’s loads of places to get vaccinated and they opened up a ‘no appointment needed’ centre back in March. The football stadium is a stones throw from a train station. 

Yep. I don't buy that car ownership is the main barrier here. We're talking about Deane or great lever. They're right in town. Anyone who can get to the shops could get to the vaccination centre.

Edit: My earlier guess wasn't bad; Deane, Rumworth, Great Lever are all within a 2 mile radius of the (original) vaccination centre in town.

The latest data point for the Indian variant has landed right on the line for the 5.1 days doubling.  All the usual riders apply to this plot - lots of unknowns about the data, and the possibility for ongoing surge testing to be contributing to it.

We've had a couple of days of marginal growth week-on-week in the PCR cases at the England level in recent days, and a couple of days of marginal decay.

I don't know the lag between the "by specimen date"  PCR cases and the reporting date of sequencing data; there have been suggestions from a couple of posters of 1-2 weeks.   There's some unpleasantness in the provisional window on some of the outbreak areas, but it looks to have dropped below the initial sub-week doubling times the areas saw, so... I still hold out hope that this variant data will drop off the 5.1 day doubling time by next week as the assumed lag between PCR cases and sequencing counts works out, and as the surge testing catches up to new infections.   Somewhat wishful thinking, but we'll see.

There is some sign of hospitalisations turning to growth in the North West after cases in the regional rate constants plots, but it's very low numbers noisy.  The comments from authorities on the news are that most (all?) of the hospitalisations in an outbreak area were un-vaccinated individuals.  

The most egregious reported UK wedding breaches have been in the much smaller ultra orthodox community. I know of several weddings and funerals with christian services where rules were broken. An 80+ catholic explained to me how vaccines were tested on cells with a foetal line so he and his friends were not taking it. Why single out muslims? In my view it's all part of 'a bit of everything'. Behaviour specialists knew such things would happen and would be almost impossible to police and the consequences very difficult to track and trace. The solution is to keep such variants out.

The ‘no appointment needed’ hub is in Rumworth. The health experts have done loads and worth remembering lots of nhs/council staff will come from and live in the high risk areas. 

I was simply highlighting how some key religious groups in areas of high Covid have acted irresponsibly, especially key ‘leaders’, and this may be one of a number of contributory factors. Truth is, I’ve not heard of remotely the same proportion of similar breaches from Christian faiths, nor heard of councils travelling to high risk overseas destinations. I’ve written at length on this forum about how some religious groups feel so strongly about their faith etc that they appear to let this over-ride the rules and common sense. Fully agree about keeping the variants out, but it’s not made easy when many folk do not take personal responsibility. 

What I’d REALLY like to see if a breakdown of hospitalisation by variant.. doubt that data exists in public..?

It’s the sort of thing that can pop up in the SAGE minutes.  As usual, I’m behind on my reading there.

For info, I just saw something saying PHE are planning to release the next technical briefing on variants tomorrow. They're usually worth a read. I noticed they also designated a new VUI today that apparently has unusual mutations, in the Yorkshire/Humber area.

You are right that the 8% difference will be higher as a % of those eligible but that’s not really the point. The point is that overall Bolton wasn’t / isn’t that far behind in terms of % of people vaccinated. Also the article quotes 95% of over 50s being vaccinated which is about average. And yet it still kicked off. Now it could be that vaccination levels were particularly low in the MSOAs where the outbreak started. We don’t have that info.

What I’m saying is I’m concerned for other areas in light of this.

Clearly the government is most to blame, both for not redlisting India earlier (and then not bringing in redlisting overnight) and for not doing enough to combat vaccine hesitancy in some communities. 

And, realistically, what would be the cost of funded quarantine, compared with the cost spent so far, e.g. the £38Bn on test & trace? I imagine the perceived problem would be capacity.

I’m not sure where the BBC got the data from but they have an MSOA level uptake map in the article below.  The correlation with the MSOA level cases map isn’t perfect but the striking similarity is one of the reasons I’m not going straight to the red light bulb over these outbreaks:

https://www.bbc.co.uk/news/health-57094274

Re quarantine, who knows - some may have stayed at home but their households didn’t because they weren’t required to...

Surely not with all those business travel hotels standing empty...

Thanks. Not an exact correlation but it’s there. Saying that, all but one area were over 65% for over 40s, which is still not bad (just not as good as elsewhere). The other question is which age groups are causing the spike in cases. I’ve seen suggestions it’s younger people and the article alludes to this as well (it also mentions a uni and a sixth form college in the focal areas). If it’s mostly under 40s, vaccination rates aren’t going to explain it. However I’ve not done any analysis to confirm this.

Red light bulb - well, depends how you see it. Wait till it gets into areas with significantly lower vaccination rates (Birmingham 50%, parts of London also low). Early days yet but I just can’t see how it’s going to be possible to keep a lid on it. Main hope is most people won’t get seriously ill due to being vaccinated or young. Hope being the operative word.

I agree the doubling times might be artificially high for a few reasons but I also struggle to see how that kind of doubling rate can be sustained thus far without greater transmissibility. The jury’s out as to how much greater.

Case data all over the place. An artefact of the methodology change, bad news or a bit of both? Mon and Tue reported cases here in Brum were the highest since mid April. Hello B.1.617.2? 

Not generous enough with donations, perhaps...?

Making quarantine free would be counterproductive. Don't f***ing travel is the solution. Making travel more accessible gives more opportunities for cases to slip whatever net is in place. 

https://coronavirus.data.gov.uk/details/vaccinations?areaType=ltla%26areaNa...

Look at the last few days on this chart. Could have stumbled on a great way to address vaccine hesitancy here...

It's been seeded right in the places, demographics, chunks of society, households where we've seen the fastest growth and most stubborn decay and lowest engagement with control measures and vaccination all the way through the pandemic. It really was a bullseye. It's had the best start possible. Very hard to decouple its natural characteristics from that. It will find other pockets where it gets an easy ride too, for sure, but the national picture doesn't say 'shit the bed' just yet. I'm with wintertree on the choice of bulb.

I have been analysing the English data by age group and cities and I have big concerns in certain conurbations. The uptake rates in London, Manchester and Birmingham are very low.

eg. in Manchester only 70% of those aged 40+ have taken up the jab and the uptake in all cohorts is <90% e,g 80+ is only 87%

There is an awful lot of vulnerablity in these 3 cities.

Nationally the figure for 40+ is better at 84% but uptake increases week on week have virtually stalled in those aged 50+ and are coming to standstill for those in their 40s. This is to be expected as the time since each cohort was offered it becomes longer.

There is a definite and significant decrease by age band. Nationally 95% of those aged 75+ have had it but this falls to 85% for those in their 50s and I predcit it won't get above 75% for people in their 40s. It'll drop even further for those aged <40.

It is concerning as if this Indian variant has a significantly increased R it obviously makes herd immunity levels harder to reach.

I looked at Israels programme and they have double jabbed 87% of their adult population. There is no way on earth we are going to hit this level IMO.

From the weekly reports located here. I've used them this week to look at uptake at a LA level in our core cities. See my other post

https://www.england.nhs.uk/statistics/statistical-work-areas/covid-19-vacci...

Thanks for the nhs link.

I have been monitoring the local daily spread around MSOAs in the north west this week, I plan to see if there is much correlation in Bolton with vaccination uptake on that map come the weekend.

I thought I had seen headlines suggesting that national uptake for over 50s (and possibly over 45s?) was over 90%. Does quoting figures for over 40s bias your conclusions a bit? Some of them will have booked but not been jabbed yet, or could still be looking for a booking in more remote areas.

According to our world in data Israel is around 60%, not 87%, see graph? Not convinced it's relevant but I think we'll definitely overtake that figure.

That's by total population. They have a young population compared to us 29% aged <18 (compared to 21% in the UK) so it works out at 87%, actually got the 87% double jabbed wrong that's the total 1st jabbed with double jabbed slightly less.

English rates as of the 16th May:

40-44  45-49  50-54  55-59  60-64  65-69  70-74  75-79  80+

64.2%  75.3%  83.7%  86.4%  88.8%  91.6%  94.1%  95.2%  95.0%

Rolled up for those aged 50+ is is 89.8% but you can see the obvious drop by age band once you get below 70.

I am having to take an extra day off next week as a driver. Distant vaccination centre and non driving wife. I can take the wage hit, but others are not so fortunate. 

I do think there seems to be great disparity in availability of vaccination centres. I guess maybe this is stating the obvious but it seems very stark. I booked my appointment the other day for this Saturday... the centre is a 5-10 minute walk from my house but, more to the point, there were three other options all within a 15 minute drive. Obviously I appreciate I'm lucky in this respect. I know there were fewer options around here a few months ago, it has obviously ramped up a lot and I was really surprised to see the options now available as I'd expected to have to drive to St Helens or something.

There is a lot of info on LG Inform

You can get MSOA vaccination data here:

https://reports.esd.org.uk/reports/8167?pat=LA&pa=E08000032%3AMLSOA&...

And all sorts of other data, with customised searches.

(Sorry if this is another rabbit warren to explore!!)

Fully get that, but people living in Bolton are close to a vaccination centre, especially people living in high risk areas/communities. 

For the vaccinations they will be pulling the data from.... https://www.england.nhs.uk/statistics/statistical-work-areas/covid-19-vacci...

there are reports to be fund on 'everything'  https://reports.esd.org.uk/search

Which part of Manchester are you referring to? (LA of Manchester or the sub region of Greater Manchester?). From The Guardian:

“Bolton’s vaccination rate was in line with the national average, with 88.9% of the 40+ population receiving their first dose compared with 89.8% England-wide. However, there are variations in vaccination rates within the local authority. In Lever Edge, 84.7% of those aged 40 and over are vaccinated. The area with the second highest case rate, Rumworth South, reported a vaccination rate of 79.4%.”

Edit: just been checking.  The general vaccination centre is on Ashburner Street, 1.2 miles from Rumworth Street. The no appointment centre is on Ginlow Lane, 0.5 miles from Rumworth Street. The local politicians criticising lack of access should shut up and support their staff. 

Some stuff you might find useful/interesting here:

https://www.boltonjsna.org.uk/coronavirus - note the table of cases/age group

https://www.boltonjsna.org.uk/deprivation

https://www.boltonjsna.org.uk/ethnicity

This. Even before the new one opened, according to the national list there were 6 vaccination sites in Bolton. Six. You have to try pretty hard to be in Bolton and more than a mile from one of them. It's one of the better served places from what I can tell from the maps and lists, so I find it hard to believe the reason it's below average is only or even mostly because people can't get to the centres. My mention of this got a lot of criticism further up the thread though, so.... shrug.

The LA area as defined on the NHS data sheet.

The Guardian is wrong as those % are for over 50s not over 40s.

Very concentrated in younger ages, which is promising, but I'd like to see that same graph extended backwards so we could compare the profile early in the autumn wave.

I'll add these to my qualitative search for correlation with areas of highest covid spread. Initial impressions are the correlation is very weak.

I was also interested to see in that link how badly Bolton was hit during the first wave - higher deaths and hospitalisations than the second. I previously had the impression that all of the north west (like where I live) was less badly hit then.

Bury was all over the news in the first wave if I remember right.

Ah, thanks for pointing that out and doesn’t surprise me those Guardian figures are wrong. 

Thanks both for the links - lots to dig in to there.  I suspect another rabbit hole is what the size of the error bars (and so % vaccinated calcs) should be at the MSOA level...  

It's very tempting to draw the obvious conclusions here, although that could be masking other issues.

I promised myself I wouldn’t do this due to the resulting head spin but here goes:

Bolton LA ranked 64th most deprived LA, Covid death rate of 257, 65.5% first jab, 37.2% second

Wigan LA 67th, 285, 69.7%, 39.8%

Manchester LA 6th, 183, 45.5%, 23.5%

Looking at those three areas, the poorest area has had by far the lowest death rate and by far the lowest vaccination rate. Could easily be due to demographics (younger population living near a city centre, poverty meaning fewer older people). But still counter intuitive. 

The most common defining factor appeared to be if the town began with the letter B.  Bury, Blackpool, Blackburn and Burnley were headline areas. Soon followed by Bolton and Bradford. 

The current top four are Bolton, Blackburn, Bedford and Burnley. Might be more effective to change their names

I was thinking more of the situation where you impose a restriction, as in the India case. If you allow people to come back before some deadline, then free quarantine means they return and do quarantine properly.

Alternatives are instant imposition of travel restrictions, with paid quarantine, or an NZ solution of allowing citizens to return to free quarantine, provided they left prior to covid, otherwise they pay.

So, here is some advice submitted to SAGE on 11/05 about the Indian variant that was only published today. When you get to the end it might become clear why it hasn't been published until a few after the latest unlocking... however, some of what they predicted then has already demonstrably not come to pass...so, draw your own conclusions.

https://www.gov.uk/government/publications/juniper-potential-community-tran...

Reads like a pdf version of this thread. Not the first time that's happened.

The very last plot looks like it was lifted from this thread! WT, are you SURE you’re not a member of SAGE?!

Dual axis - nice.  I have neither the eminence of position nor - I strongly suspect - the temperament for SAGE, although I have noticed that various plots on this thread seem to foreshadow the appearance of a similar plot on the dashboard; I half wonder if there’s a silent reader or two on the thread, but I don’t flatter myself...  

Mind you, re: their plot I’d note that it’s not a “doubling time” - I split label the two sides of y=0 as “doubling time” and “halving time” and I also add a units of [days]...  Although my new week-on-week plot still needs sorting all the way...

Edit: I really appreciate their plots having the far more tangible/translatable characteristic time axis as well as the R axis.

In reply to Si dH:

The report makes the key point that waiting for definitive proof of a bad thing means waiting until it’s too late if its true.  Hopefully it won’t bite us this time, but I’d be happier if there was more sign of that message translating in too policy...

Whether it’s due to people previously hesitant or people previously not eligible is another question (they seem to be happy to jab anything which moves and rightly so).

What you say is not incorrect but I just think it doesn’t explain everything. I think the picture is more complex than what you suggest. Spread seems to be among younger people, who mostly aren’t eligible yet. Makes sense as they’re the ones going out etc. I think it’s roaming in the wild now and will soon be ripping through the unvaccinated (and some of the vaccinated) across all demographics and communities.

Thanks, that’s a treasure trove of info. I don’t know Bolton at all so had a look at Birmingham. Overall, only 50% have had first jabs, so ripe for a third wave. MSOA rates are lower in more inner city areas. That will partly be due to younger populations (the lowest MSOA is the Brum Uni ‘student ghetto’ area) but the analysis by age group shows a similar distribution. It seems that vaccine hesitancy in poorer areas with higher ethnic minority populations is definitely an issue. Interestingly, the highest uptake areas aren’t the  most affluent ones, more like fairly average really. Could be due to having more older people or just better organised local vaccine infrastructure. The other observations is that all the stats are below average. Not good. 

Thanks.. one party and this could be a hotspot too. Still taking a hit. 
Mind I would happily walk the three hours myself.

What you say about spread mainly in younger people:  if I was the gov’t I’d say ‘anyone book a jab ASAP’. With so many adults with immunity, may as well go for the 18’s upwards now and stop hanging about.