Friday Night Covid Plotting #31

New Topic
This topic has been archived, and won't accept reply postings.
 wintertree 19 Jun 2021

What a difference a week can make.  

  • The behaviour of cases is as mysterious as ever; the rapid exponential growth stopped just after last week;s update and the doubling time has slowed way down - see the English PCR cases plot. 
  • I’ve noted before how the fastest growth never seems to be sustained in an area for long, and indeed we saw the early outbreak areas turn a corner in to stagnation or decay of cases PDQ as localised responses span up.
  • We’ve also had a run of great weather.  Over the colder months, swings in temperature seemed to be behind swings of a similar scale in the exponential rate constant [1]; perhaps that’s part of it now.
  • It feels a bit to soon to me for the spread of infection to be bringing us in to the slackening part of the logistic curve of a pandemic - that is for herd immunity to be emerging, but I haven't run the numbers on it as a possibility.
  • Regardless, it’s great to see the rapid growth not sustained despite the worries over R0 of the new variant.   It's great because it buys time whilst the vaccine program continues to roll forwards.
  • When the month’s temperature data is out we can see if there’s a correlation there…. The correlation weakened a lot as restrictions kept changing after the return of schools, but we’re in a long period of quite static restrictions now.

Plot A has been re-jigged to keep it simple to look at; it now shows...

  • Red: Detailed prevalences (cases per 100k people per day) for regions to have exceeded 50 cases/100k/day
  • Orange and blue - simplified trajectories for the rising (orange) reigns not to have hit this level, and falling regions (blue).
    • Last week most regions falling over the period of this plot were growing over the most recent weekend less than half are left this week, they're also growing over the most recent days
  • Big picture - growth in the outbreak regions (by my arbitrary definition) remains stalled - actually in to decay slightly right now - so still no need to panic over this variant.  We know what to do, it still works, people are still willing to do it.  
    • Thank the gods and their little fishes for the vaccines and the roll out program, this year without them doesn't bare thinking about.

Plot E shows the actual daily numbers (not prevalances) for these areas stacked by colour, I've added a vertical scale par for 1000 cases/day. 

  • The turn to decay of the red "outbreak" areas is visible by eye, as is the decay in some of the larger case number orange areas.  
  • The 8 remaining blue UTLAs that have fallen over the whole period of the plot are, when measured over the last 7 days, now rising with a doubling time of ~50 days.  They're probably heading for a period of faster growth now.

Plot C - Demographic comparisons

  • I'm now doing this at the whole England level seeing as the new variant has more or less taken over everywhere; astounding how quickly that happened -being more transmissive and turning up at more or less the exact moment the key dropping of restrictions happened.
  • This is a probability distribution of the demographics for periods in January 2021 when cases were high with the Kent variant and little vaccination (including the 3-weeks post jab period for 1st jab to become maximally effective) and now with the India variant and lost f vaccination.
    • So it doesn't tell us about the absolute number of cases, just how they're distributed.  The area should add up to 100%.  Last weeks plots added up to 200% due to a daft mistake but that doesn't change interpretation.
  • Cases are really, really spiking in young adults and secondary aged children now.  This in no way proves that the variant is more transmissive in these ages.  The way I think of it is that in a parallel universe where we did exactly the same things as we have done this year, except vaccinate people, there would be an order of magnitude more older people infected and they'd outnumber the younger ones.  
    • So, in one sense perhaps we're running the pandemic much hotter now than at any point since the very beginning, but then a vast chunk of cases is cut out by vaccination.

Plot Cr - Ratio of demographics

  • This shows the ratio of the two demographic probability distributions.  I think it's a really beautiful plot in a data sense, thinking about the phenomenal amount of work that's gone in to reshaping the distribution of cases so.

Link to previous thread: https://www.ukhillwalking.com/forums/off_belay/friday_night_covid_plotting_30-...


2
OP wintertree 19 Jun 2021
In reply to wintertree:

All 4 home nations are now in clear growth of cases.  Scotland remains closest - only one doubling period now - from it's peak from the winter just passed.  Wales looks like it's heading for decay but the provisional window on the government dashboard (not used in my plots) has growth in it.  The most recent data point for England looks like growth might be speeding up; it feels to me like the trends are a bit more erratic than they used to be.  Not surprising with more "outbreak" mechanics in play at quite large scales


2
OP wintertree 19 Jun 2021
In reply to wintertree:

Scotland has generally seen the weakest decay and was the first of the home nations to turn to growth at a national scale, it's also the closest to hitting it's winter peak of cases, in one one more doubling time.  That doubling time as measured by my analysis has slacked off a lot, but looking at the cases curve In plot 6s I'm not convinced the slackening will last.  

A rising deaths signal is now clearly visible - with very small numbers; too small to measure a reasonable characteristic time for plot 9.

If the case rate is picking up again (in exponential terms), given that my plots end a few days in the past and given the latencies in policy change to infection rates and infection to detection as a case, it may well be too late to prevent Scotland from exceeding it's peak winter daily case rate.  Pre-vaccine, that would have been disastrous news for healthcare.

  • The Lissajous figure for Scotland is gone through in more detail in plotting #30.  It's a bit maths heavy to really interpret; the purple markers are the last wave and the blue ones this wave.  The more vertical space between the leading edge of the purple and blue curves, the better. 
  • Hospital admissions are currently running at ~half the level of the previous wave for the same case rate - and the gap between the waves is opening up in favour of this wave
  • Hospital occupancy is likewise at about ~half the level of last time, and looks to be opening up more.

So, as it stands, two doubling times in cases before things get as bad as last winter on the coalface of healthcare.  I think there's a few reasons - discussed on previous threads - to expect the conversion ratio from cases to hospitalisations to continue improving, for example ongoing vaccination and the details of the initial outbreak areas - whilst the importance of those is fading almost to irrelevance in cases, there is a lag through the healthcare pipeline.


2
OP wintertree 19 Jun 2021
In reply to wintertree:

All measures for England are now in growth with the emergence of a clear - but small - growing rate of deaths (less than one thousands of that at the peak from early 2021).

  • The doubling times for admissions have been muted compared to cases; the maths relating to this is discussed of the last couple of threads but what I think this shows is that the conversion ratio from cases to hospitalisations has been getting better during the recent growth in cases. 
  • The most recent doubling time for deaths looks alarming, but it's very small numbers so subject to a lot of statistical noise, and the high measurements are in the very provisional right hand edge of this plot.  

The Lissajous figure shows that hospital admissions are currently at ~66% of the level they were when cases were at this level last time around; in reality I think the situation is better than that as the details of the last wave pulled its purple curve down and right for a bit; the red line is my quick stab drawn "by eye" at the line I think the curve deviated from; we're more like half this level, as with Scotland.  Likewise, hospital occupancy is running at half this level. Again, there are several reasons to expect that the gap between these curves gets wider.  

Now, I'm watching the separation of purple and blue curves for Scotland very closely; hopefully they open up and then hopefully they open up for England.  This is the ongoing "decoupling" of cases and healthcare consequences.  It's not a complete decoupling, but it moves some policy lines in the sand significantly.


1
In reply to wintertree:

> Scotland has generally seen the weakest decay and was the first of the home nations to turn to growth at a national scale, it's also the closest to hitting it's winter peak of cases, in one one more doubling time.  That doubling time as measured by my analysis has slacked off a lot, but looking at the cases curve In plot 6s I'm not convinced the slackening will last.  

Scotland has different school term times from England, summer term is pretty much over.  It is young people where the cases are coming from at the moment and the Scottish Government will have factored in that the schools would be shutting for summer when considering whether they needed to take more action.

3
 Si dH 19 Jun 2021
In reply to tom_in_edinburgh:

Schools are not the source of most cases and haven't been for several weeks. School terms will have minor if any impact on the trajectory. The biggest case numbers since hospitality opened up and Delta spread more have been in 15-19, 20-24, 20-29 age groups. In fact according to WT's graph at the top there now more cases in people in their early 30s than in the 10-14 age group, which is where school spreading hits the most.

(Edit to made though, I do think Scotland's national rate will probably moderate itself earlier than England's anyway, because it's biggest population centres have all been hit early.)

Post edited at 21:55
OP wintertree 19 Jun 2021
In reply to wintertree:

Plot 16 - Growth in cases is near universal in England

Plots 18.1 & 18.2 -  Regional Rate Constants

  • I re-ranked this plot in terms of when there was a return to growth and set the earliest growth areas' traces to be red.  Plot 18.2 is an easier way of following the details for a specific regions.
  • There's a stand-out bifurcation between the North East and South West having rapid growth and everywhere else stagnating.  I haven't looked in to the details yet, I wonder if cases / 100k / day are lower in the two standout regions?  That's often when the highest growth rates are seen.  

Plot D1.c - Demographic Rate Constants

  • I've stopped doing this split by outbreak / non-outbreak as the new variant is everywhere now.
  • Growth remains strongest in young adults
  • We're seeing high rate growth in the most recent week or so for the oldest people (top right corer of the heat map).  This might tie back to some acrimonious discussions going on in the Pub forum over care staff and vaccine refusal, and might also tie back to the recent announcement that policy over vaccination in the care sector is moving from carrot to stick.
  • Given the extreme susceptibility of people in residential care to any illness, I think it'll be worth looking at the demographics on all measures in future weeks to avoid conflated signals.

Vaccine plot

  • We're past the peak of second doses coming due.  This increased the notional gap between doses slightly from ~10.5 weeks to ~11.3 weeks.
  • No sign of this notional gap dropping yet in response to the change to an 8-week gap; hopefully the improved system for rebooking will help with this - I've (gladly) shortened my gap!
  • First doses are not picking up the slack from falling second doses coming due; presumably issues over supply constraints on Pfizer and Moderna for younger adults vs the purported stock building up of AZ that will could shorten the gap for second doses coming due.
  • On the vaccine front, Novavax was recently approved in the USA [1].  I think this is a really exciting technology platform, and it doesn't engage in de-novo synthesis so perhaps it's far less likely to be involved in the side effects resulting from mis-expression associated with some other vaccines.
    • The UK has a production facility for this; it's to be bottled in Barnard Castle of all places.  
    • If any immunologists reading can explain to me in really simple words how this vaccine gets a good CD8+ response without engaging in synthesis I'd be most grateful.

[1] https://arstechnica.com/science/2021/06/we-have-another-highly-effective-co...


1
OP wintertree 19 Jun 2021
In reply to wintertree:

(Opens excuse book, rolls dice, turns to page 5...)

Once again proof reading was limited by the amount of time spent outdoors today...  Jr spotted an amazing butterfly on some yellow vetch.  If anyone can identify it for me, I'd be most grateful.

Looking at these photos on my laptop screen, I'm almost as blown away that they were taken on a handheld mobile phone; the technology hasn't half come along in the last decade.  


OP wintertree 19 Jun 2021
In reply to tom_in_edinburgh:

> and the Scottish Government will have factored in that the schools would be shutting for summer when considering whether they needed to take more action.

Are there any documented meetings of the SG's advisors that are published in the way SAGE's outputs are, or is it all closed doors stuff?

Projecting the data with momentum from today, there's two doubling times until healthcare hits a crisis; which with current exponential rate constants/doubling times is plenty of time to affect a change to policy if - against expectations - the decoupling doesn't continue to improve.

1
 Si dH 19 Jun 2021
In reply to wintertree:

Re:south west and north east. It seems to me that North Tyneside and Newcastle are now following a similar trajectory to Liverpool. There is also a localised but very significant outbreak in Durham. They're pretty big numbers.

Similarly in the south west you have rising rates in Bristol and a couple of outbreaks that have made Cornwall's rate rise a lot. Obviously Cornwall's rate had previously been very low. It's not a flash in the pan though, rates are still going up in the south west in the provisional region.

 Si dH 19 Jun 2021
In reply to wintertree:

> > and the Scottish Government will have factored in that the schools would be shutting for summer when considering whether they needed to take more action.

> Are there any documented meetings of the SG's advisors that are published in the way SAGE's outputs are, or is it all closed doors stuff?

> Projecting the data with momentum from today, there's two doubling times until healthcare hits a crisis; 

I'm no healthcare expert but Scottish hospital occupancy was presumably far less close to maximum overload than England's was in January? Peak case rates were certainly quite a lot lower because they didn't get much of the Kent variant before their lockdown? So getting back to January levels in Scotland wouldn't be as bad as it would in England. Unless Scotland's healthcare is poor.

Post edited at 22:11
 rlrs 19 Jun 2021
In reply to wintertree:

Common Blue, according to quick internet search.

OP wintertree 19 Jun 2021
In reply to Si dH:

> I'm no healthcare expert but Scottish hospital occupancy was presumably far less close to maximum overload than England's was in January? Peak case rates were certainly quite a lot lower because they didn't get much of the Kent variant before their lockdown? So getting back to January levels in Scotland wouldn't be as bad as it would in England. Unless Scotland's healthcare is poor.

That's a good point.

🏴󠁧󠁢󠁥󠁮󠁧󠁿 - England - max hospital occupancy in the last wave circa 33,000 out of 55.2 m = 0.06% of the population in hospital with Covid
🏴󠁧󠁢󠁳󠁣󠁴󠁿 Scotland - max hospital occupancy in the last wave circa 2,000 out of 5.4 m = 0.04% of the population in hospital with Covid

So, if the hospital provision is equal on a pro-rata basis between the nations, they were ~2/3rds as badly hit, so this gives them about 70% of a doubling time more headroom.

> Re:south west and north east. It seems to me that North Tyneside and Newcastle are now following a similar trajectory to Liverpool. There is also a localised but very significant outbreak in Durham. They're pretty big numbers.

Thanks for the comments.  It seems my peril sensitive sunglasses had been blocking out the Durham outbreak.  Looks like the university is trying to break it's (per-student) outbreak record from the last wave.  This being a university that extended term by a week aiming to facilitate the "Wider Student Experience" after the projected June 21st unlocking. [1].  I am sure that this monumentally incomprehensible decision was in no way motivated by the importance of the "hotel" income strand from the managed accommodation part of the business.

[1] https://www.palatinate.org.uk/university-extends-easter-term-by-a-week-for-...

1
OP wintertree 19 Jun 2021
In reply to rlrs:

> Common Blue, according to quick internet search.

Thanks; I'm having a bit of a moment that the internet tells me this is the most common English butterfly and yet I don't recall ever having seen one before.

In reply to wintertree:

> Are there any documented meetings of the SG's advisors that are published in the way SAGE's outputs are, or is it all closed doors stuff?

AFAIK they take their science advice from SAGE and they have a couple of Profs from Edin Uni that consult to them, hopefully as a bullsh*t detector so they don't have to blindly trust everything that comes from the UK govt.

> Projecting the data with momentum from today, there's two doubling times until healthcare hits a crisis; which with current exponential rate constants/doubling times is plenty of time to affect a change to policy if - against expectations - the decoupling doesn't continue to improve.

All through this Sturgeon has been relatively cautious so I'm pretty sure she's watching the numbers and basing her actions on the advice and projections she's getting.

She will have more visibility on how many young people they expect to jag in the next few weeks and projections of the effects of the end of school and uni terms.

12
In reply to wintertree:

Common blue.

https://www.rspb.org.uk/birds-and-wildlife/wildlife-guides/other-garden-wil...

[edit: sorry; playing catch up...]

Post edited at 01:47
In reply to wintertree:

Some interesting ideas on hospitalisation/stays here

https://github.com/nicfreeman1209/covid-19/blob/main/conv-estimator/convolu...

Possible inspiration if you're still trying to unpick the rates and rises

Edited to better link

Post edited at 08:51
1
OP wintertree 20 Jun 2021
In reply to Longsufferingropeholder:

Thanks; I had go at a convolutional approach a few months ago, the big limit was the demographic bins in the public hospitalisation data being coarse and mis-matched to those in the cases data; demographics being central to the conversion rate.  Otherwise, it's such an under-constrained problem you can always get a fit, but I'm not sure I'd call it either a measurement or predictive.

Here's a really simple plot - it's the demographic case rates with each column (timepoint) normalised to sum to 1, so it shows us the relative distribution of cases rather than the numbers; like a heat map version of plot C.

The return of schools is very clear in this as a sudden opening up of a sump in the data, although the cadence of the LFDs causes almost a pulse strain; a bit of smoothing gives a clearer view of the average.

Other than schools, there's a sudden explosion in adults aged 20-25, as well I think as a descending envelop on adults in the range 45 to 65 as the bulk of cases shifts to younger ages; I've sketched the line I see "by eye" in red on the annotated plot. Some combination of ongoing vaccination, spread beyond the initial outbreak populations and perhaps a shift in risk perception with age as cases rise..

So, it looks to me like there's every reason to expect the hospitalisations to separate more from cases as the last couple of weeks growth in cases starts to work through the pipeline.

When I get a chance I'll apply the CDC relative risk weightings to the English demographic data to give a more quantified take on this - https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigations-discove...


1
 minimike 20 Jun 2021
In reply to wintertree:

Is there something in ‘soothed’ and ‘smithed’ I’m missing? Or just autocorrect...

OP wintertree 20 Jun 2021
In reply to minimike:

> Is there something in ‘soothed’ and ‘smithed’ I’m missing? Or just autocorrect...

Double autocorrect fail. Thats spectacular, even for me.  Next time I'll say "filtered" instead.

In reply to wintertree:

Something nice about 'data smithing'. Makes it sound like a proper trade. I might start using that. Off to 't data smithy for another day's graft.

OP wintertree 20 Jun 2021
In reply to Tom in Edinburgh from thread 30:

>> Beyond that, if I'm going to catch the virus there's bugger all I can do about it now having children in school/nursery, so what will be will be - accepting more risk elsewhere in my life just brings that point forwards. [...] In terms of personal health consequences for me, the risk was never that great and its lowered a lot by my first dose and by all the improvements learnt about and developed for covid care since the start.

> I think that is a dangerous way to look at it.   There's an implicit and false assumption you will only catch it once.

> If immunity following infection only lasts for 6 months to a year and if there's enough of this stuff in circulation that the prevalent strain changes every 3 to 6 months the end result of 'back to normal'  could be people who need to get out and about catching it every year or twice a year.   

> It's one thing to 'take it on the chin' once but if it keeps coming back every few months and punching you again until it eventually gets you that is quite different and you've got to be more careful about not getting punched.

With a child in school however, there is nothing I can do about that short of depriving Jr of a normal life and setting the state against me to break that risk of catching it.  

The situation you describe in your second and third paragraphs is the usual state of play for a bunch of other viruses that plague humans at a generally low level, and to my - limited - understanding of the literature the main differentiator between SARS-nCov-19 and some of these other viruses is that every protein within the former was unknown to the human immune system.  

Again as I understand it, whilst antibodies from infections are measured to fade over the kind of timescales you note, more durable memories of the antigens are being shown to persist in B memory cells.

One line of thought says that by hiding away indefinitely from the virus I deprive my body of the chance to generate broader spectrum immunity against other proteins (particularly the membrane proton and viral capsid protein?) which are undergoing less host adaption motivated variant than the spike protein, giving better protection against those variants.

Lots of uncomfortable choices lie in the year ahead over the way out of this.

I'll repeat that this is my lay persons view on the subject (not being an immunologist) and that I'm young enough that the health consequences are relatively low for me, especially after my first jab.  My opinion doesn't and shouldn't translate on to many people out there right now.

Edit:  In the past I have argued many times that this was a dangerous way of looking at it, even for individuals at very low personal risk, especially when used to justify ignoring precautions intended to slow the spread of the virus due to the effects on others.  Times are changing; with Jr back in school, rising prevalence, most restrictions dropped, meteoric rises in young adults, what I do or don't do isn't even pissing in to the wind, and the main sources of risk are out of my control other than reducing them by getting my vaccine.

Post edited at 22:22
 Misha 21 Jun 2021
In reply to wintertree:

Thank you as always. It’s a positive update in the scheme of things, though I’m still thinking “wait for it”. I wonder if this week was a sweet spot between the NW outbreaks stabilising while outbreaks elsewhere were gathering steam but not yet raging. If the position is similar in a week’s time, that would suggest that this wave is more or less under control with the current restrictions in place and the current vaccination level.

 minimike 21 Jun 2021
In reply to Longsufferingropeholder:

That’s going on my LinkedIn right now..

 Toerag 21 Jun 2021
In reply to wintertree:

> That's a good point.

> 🏴󠁧󠁢󠁥󠁮󠁧󠁿 - England - max hospital occupancy in the last wave circa 33,000 out of 55.2 m = 0.06% of the population in hospital with Covid

> 🏴󠁧󠁢󠁳󠁣󠁴󠁿 Scotland - max hospital occupancy in the last wave circa 2,000 out of 5.4 m = 0.04% of the population in hospital with Covid

That really goes to show just how low healthcare capacity is compared to the population!!

My noddy excel using UK stats show the increase in live cases rate has peaked on a 7 day count at 4.7% per day last week.  108k live cases now by my reckoning, January's peak was 742k.

In reply to Toerag, wintertree, minimike:

In other news, I was reminded today of that slightly heated discussion a while back about whether or not we'd ever see the uptake rates that Israel did. Any bets on that should be settled today or tomorrow.

Must get back to the data smithy. My data are getting noisy and need soothing

OP wintertree 21 Jun 2021
In reply to Toerag:

> That really goes to show just how low healthcare capacity is compared to the population!!

Indeed, and that's perhaps going to factor in to the coming winter, even if we hopefully make signifiant progress with Covid by then.  The BBC are running with the flu-mageddon concept today although they didn't get so far in to the details as the problems around identifying the likely future prominent strains in time for vaccine production.

>  108k live cases now by my reckoning, January's peak was 742k.

Presuming the India variant has a similar mean "live" time?  We might see some effect from the cooler weather coming on, perhaps.  I was very glad to see it rain today, some bits up here were starting to resemble the South East in terms of levels of parched yellow.  

In reply to Longsufferingropeholder:

> I was reminded today of that slightly heated discussion a while back about whether or not we'd ever see the uptake rates that Israel did.

It's just great, isn't it?  We've past 80% of adults now for the first dose.  First dose rates are picking up a bit as well.   

In reply to minimike:

> That’s going on my LinkedIn right now..

Is "data smithing" hitting data with blunt instruments until it's the shape a customer wants...?

 elsewhere 21 Jun 2021
In reply to wintertree:

> It's just great, isn't it?  We've past 80% of adults now for the first dose.  First dose rates are picking up a bit as well.   

https://ourworldindata.org/covid-vaccinations?country=OWID_WRL~GBR~USA

It's also a funny/tragic difference in a nation's psychology. Vaccination rates in the UK go up in straight lines like the Soviet centrally planned economy* compared to the US curving organically as demand tails off.

*except unlike the Soviet economy, NHS central planning worked 

In the US, 77% of Democrats are either fully or partially vaccinated but only 52% percent of Republicans. It's madness! 

In the UK, even the anti-vaxxers are vaccinated judging by 99% or 100% vaccination rates for older age groups.

We must be a very obedient bunch when it comes to meeting the goals of the central plan.

Post edited at 10:32
1
OP wintertree 21 Jun 2021
In reply to elsewhere:

> We must be a very obedient bunch when it comes to meeting the goals of the central plan.

I like to think of it as enlightened self interest...

I think the people of the UK have really exceeded a lot of expectations throughout this; in broad adherence to very disruptive control measures, in the efforts done to keep on top of their work despite the disruptions whilst also embracing home schooling, engaging with the vaccination without any carrot beyond the protection it offers and with very little stick.

 Offwidth 21 Jun 2021
In reply to elsewhere:

Those national percentages for the older age groups are almost certainly wrong as they will use ONS 2019 population numbers (which are lower than they are now). More or Less highlighted the issue two weeks ago. It always looked oddly high so it was nice to get an explanation.

Back to Cornwall it seems it was likely the G7.

https://inews.co.uk/opinion/comment/g7-cornwall-its-little-wonder-the-local...

Post edited at 14:06
1
 elsewhere 21 Jun 2021
In reply to Offwidth:

> Those national percentages for the older age groups are almost certainly wrong as they will use ONS 2019 population numbers (which are lower than they are now).

That's a pity. Any idea how big the corrections are? 

Post edited at 15:18
 Andy DB 21 Jun 2021
In reply to wintertree:

Thanks, wintertree as always. I have been another long time lurker who has found your analysis of the data really inciteful. 

 Offwidth 21 Jun 2021
In reply to elsewhere:

Just looked at ONS.... recent population change seems like a small difference (approx 1% growth every 2 years) but then you need to add on non-resident workers who are here for less than a year and new children of these and of longer stay non residents. I can't see it being more than a few percent out and then mainly at the younger end. So those stats for 70+ should be pretty close.

 elsewhere 21 Jun 2021
In reply to Offwidth:

Many thanks for that info.

In reply to Offwidth:

It's miles out, unfortunately. Resulted in some areas in scotland achieving well over 100% uptake in some age groups a few months ago. Not sure if the census numbers are easily accessible.

This shows the uptake by age group numbers you're after for England. Pretty sure S and W publish equivalent data somewhere. Either way, it answers the "how big are the corrections?" question.

https://coronavirus.data.gov.uk/details/vaccinations?areaType=nation%26area...

Post edited at 16:34
 chris_r 21 Jun 2021
In reply to elsewhere:

> That's a pity. Any idea how big the corrections are? 

ONS population figures (the 2019 Mid Year Estimate is used as the denominator for the % vaccinated) are really reliable, with errors concentrated at either end of life, under 5s and over 80s, as births and deaths are the major causes of change, and therefore inaccuracy.

The most recent ONS population projection for 2021 was based upon uplifting the 2018 Mid Year Estimate population data so even though it goes up every year from 2018 to 2021 its well short of the mark, even compared to the static Mid Year Estimate from 2019. Ignore it.

As for how much the population has changed between the 2019 and now big the error is, we cant really say without a data set to compare it against. And that gets messy.

OP wintertree 21 Jun 2021
In reply to chris_r:

> ONS population figures (the 2019 Mid Year Estimate is used as the denominator for the % vaccinated) are really reliable

Population measurements, estimates and errors were something I'd never thought about until about 12 months ago...  Until then I'd had this naive idea that there was some black box integrator in The System connected to the data sources for births, deaths, immigration and emigration.   Then I saw the disagreements over how many foreign workers left the country during Covid (ranging between 250k and 1m it seems) and realised things are a bit fuzzier round the edges than I'd expected!

> with errors concentrated at either end of life, under 5s and over 80s, as births and deaths are the major causes of change, and therefore inaccuracy.

It's worth noting that Covid has happened since the 2019 MYE and that the oldest ages are the ones to have had the largest fractional loss to Covid, so one can see why the 2019 data is tending to be an over-estimate in the highest age ranges.

OP wintertree 21 Jun 2021
In reply to Misha:

> Thank you as always. It’s a positive update in the scheme of things, though I’m still thinking “wait for it”. I wonder if this week was a sweet spot between the NW outbreaks stabilising while outbreaks elsewhere were gathering steam but not yet raging. If the position is similar in a week’s time, that would suggest that this wave is more or less under control with the current restrictions in place and the current vaccination level.

A lot of the other areas seem to be tipping over in to decay as well; this happened much faster than I was expecting, lots to think about there.  If it’s the weather - well, lots of glorious weather ahead.  If it’s the most densely interconnected groupings of non-immune running out of susceptible people, that’s the end of really explosive growth.  It’s going to take a few more weeks of moderate growth and decay at the UTLA level and I might be a believer that it’s settled down until the late autumn; what ever the reasons for it backing off it’s great as it gives the vaccination that much more time to race ahead.

Happy Solstice everyone.  

 Bobling 21 Jun 2021
In reply to wintertree:

Happy Solstice to you too Wintertree!

 leahgoodall 22 Jun 2021
In reply to elsewhere:

https://ourworldindata.org/explorers/coronavirus-data-explorer?zoomToSelect...

> It's also a funny/tragic difference in a nation's psychology. Vaccination rates in the UK go up in straight lines like the Soviet centrally planned economy* compared to the US curving organically as demand tails off.

> *except unlike the Soviet economy, NHS central planning worked 

Except that the US has got falling cases and a much lower number of infections per capita, and we’ve got rising cases and a much higher number of infection par capita. Whatever they are doing is working better.

Post edited at 01:04
4
 Misha 22 Jun 2021
In reply to wintertree:

Yeah exactly. We’ll take any respite at this stage. As for winter, who knows. The vaccine situation should be as good as possible by around the end of September, with booster jabs on the way as well (assuming that’s considered beneficial - interesting point whether in the scheme of things it would be better to donate those doses elsewhere). So in principle we should be going into winter in a good position, perhaps with herd immunity from vaccination plus the impact of the latest wave. Hopefully anything over winter would be more of a ripple, concentrated in certain areas / population groups. Having said that, I don’t expect life to go back to ‘normal’ till next summer and not planning to ditch the face mask any time soon!

2
 Misha 22 Jun 2021
In reply to leahgoodall:

We’ve got a lot more Delta. It’s just not their turn yet.

 leahgoodall 22 Jun 2021
In reply to Misha:

> We’ve got a lot more Delta. It’s just not their turn yet.

Maybe. We’ll see. There is a lot of differences in the spread of delta between countries, not always easily explained.
 

2
 elsewhere 22 Jun 2021
In reply to leahgoodall:

> Except that the US has got falling cases and a much lower number of infections per capita, and we’ve got rising cases and a much higher number of infection par capita. Whatever they are doing is working better.

Not really on current data. On past data quite probably.

https://ourworldindata.org/explorers/coronavirus-data-explorer?zoomToSelect...

Post edited at 08:28
1
OP wintertree 22 Jun 2021
In reply to leahgoodall:

> Except that the US has got falling cases and a much lower number of infections per capita, and we’ve got rising cases and a much higher number of infection par capita. Whatever they are doing is working better.

I'm not sure the USA is "working better", not by a long way.  Not when looking at the endgame enabled with current vaccines - a push to mostly non-lethal endemic status.

On a 14-day lag, the US case fatality rate is about 10x that of the UK.  They are very far from the point where they can allow the virus to circulate freely without overwhelming healthcare amidst another wave of mass illness and high deaths.  

The success we are starting to see in the UK is what you suggest is us doing less well than the USA - the success is that cases are rising without hospitalisations rising so much, and with deaths barely rising at all - this means we are much closer to one way out of this mess.

I imagine the American summer and the old variants holding sway - for now - are key parts of why cases have been falling in the USA.  Low and slow vaccine uptake (in % terms) is perhaps going to cost them dear as they go in to autumn and winter. 

Post edited at 09:03
OP wintertree 22 Jun 2021
In reply to Misha:

> Having said that, I don’t expect life to go back to ‘normal’ till next summer and not planning to ditch the face mask any time soon!

It'll be interesting to see if the epidemiologists think face masks and the acrylic screens in stores etc are enough abate winter flu epidemics by themselves.  Even if we go in to winter with wide spread immunity (as you say, acquired either/both waya) against Covid masks are likely to be involved given the worries over the flu season.

 AJM 22 Jun 2021
In reply to wintertree:

I feel you might appreciate some of the discussion in here, as far as it relates to population estimates...

https://www.actuaries.org.uk/documents/c4-phantoms-never-die-living-unrelia...

 leahgoodall 22 Jun 2021
In reply to wintertree:

> > Except that the US has got falling cases and a much lower number of infections per capita, and we’ve got rising cases and a much higher number of infection par capita. Whatever they are doing is working better.

> I'm not sure the USA is "working better", not by a long way.  Not when looking at the endgame enabled with current vaccines - a push to mostly non-lethal endemic status.

> On a 14-day lag, the US case fatality rate is about 10x that of the UK.  They are very far from the point where they can allow the virus to circulate freely without overwhelming healthcare amidst another wave of mass illness and high deaths.  

It's of course very good to have a 10x lower fatality rate, but at the risk of stating the obvious here, if your cases are doubling every couple of weeks, this is quickly offset by the number of cases.

There will always be a reservoir of unvaccinated people, and people for whom the vaccine doesn't work, and that population is big enough to overwhelm healthcare if transmission goes exponential.

Post edited at 09:23
4
OP wintertree 22 Jun 2021
In reply to leahgoodall:

> It's of course very good to have a 10x lower fatality rate, but at the risk of stating the obvious here, if your cases are doubling every couple of weeks, this is quickly offset by the number of cases.

I disagree.  This is only true if cases are doubling uniformly across the demographic.  They are not - they are doubling more slowly in the more susceptible ages, and much faster in the younger adults.

You also miss entirely my point here about their CFR - it shows that their vaccination program is not moderating health outcomes to anything like the degree (by an order or magnitude) that ours is; in the senses that matter, they have a bigger mountain to climb than us.

> There will always be a reservoir of unvaccinated people,

Shrunk significantly by the ongoing live infection process. 

> and people for whom the vaccine doesn't work, and that population is big enough to overwhelm healthcare if transmission goes exponential.

Transmission isn’t really exponential, it’s a sigmoidal or logistic form.  It eventually tapers off; the vaccination has really started to push us away from the exponential-like phase I think.  In the meantime, whilst we can have transmission with health consequences moderated and death almost eliminated, there is a sense to me that endemic status might be achievable for us - a real way out.  The situation moves on but Rom’s take doesn’t.

For sure, the data in the UK is as complex confusing as ever, but there is change in the air and perhaps we are threading a needle towards the way out.  A time for caution, but not necessarily for keeping R<1 - that can always change with the data.

In reply to leahgoodall:

Do you have any opinion on the EC's contract with Astrazeneca? Just curious...

 leahgoodall 22 Jun 2021
In reply to wintertree:

> I disagree.  This is only true if cases are doubling uniformly across the demographic.  They are not - they are doubling more slowly in the more susceptible ages, and much faster in the younger adults.

Yes, and that is because transmission has stalled  in the older age groups. If cases went exponential in that group you’d still have quickly a big problem, even with lower hospitalisation rate.
Hence my point: transmission always wins if it’s too high.

> You also miss entirely my point here about their CFR - it shows that their vaccination program is not moderating health outcomes to anything like the degree (by an order or magnitude) that ours is; in the senses that matter, they have a bigger mountain to climb than us.

I did not miss your point. In fact acknowledged it. I am only pointing out that an order of magnitude less in hospitalisation could quickly be wiped out by exponential growth.

> Shrunk significantly by the ongoing live infection process. 

> Transmission isn’t really exponential, it’s a sigmoidal or logistic form.  It eventually tapers off; the vaccination has really started to push us away from the exponential-like phase I think.  In the meantime, whilst we can have transmission with health consequences moderated and death almost eliminated, there is a sense to me that endemic status might be achievable for us - a real way out.  The situation moves on but Rom’s take doesn’t.

Well if there is 10% of the population who is not vaccinated and/or for whom the vaccine does not prevent hospitalisation, if you had exponential growth in that population it would more than big enough to overwhelm healthcare.

> For sure, the data in the UK is as complex confusing as ever, but there is change in the air and perhaps we are threading a needle towards the way out. 

Time will tell, but for now I’d rather not make rubbish predictions. As for international comparisons, we’ll only be able to do that way down the line, and a good measure of success will be overall excess mortality over the period - maybe with some adjustment for level of NPIs.

6
OP wintertree 22 Jun 2021
In reply to AJM:

> I feel you might appreciate some of the discussion in here, as far as it relates to population estimates...

Wow - some fantastic plots in there.  I'm going to have to read the whole thing to understand what they really mean, but the big picture is engaging, thanks.

OP wintertree 22 Jun 2021
In reply to leahgoodall:

> Yes, and that is because transmission has stalled  in the older age groups. If cases went exponential in that group you’d still have quickly a big problem, even with lower hospitalisation rate.

Given the relatively weak restrictions in the UK and the massive growth in younger adults, and the very slow growth in the older age groups, why do you think transmission has stalled in older people?  On what basis do you think it would suddenly go high-rate postiive exponential?   

Obviously it's critical that the authorities continue to monitor and respond to changes in all the metrics, and I think that's exactly what we're seeing. 

In the mean time, I am going to look to the future, and I see many hopeful signs that it's emerging.  Perhaps I'm wrong, and we're going to go back to square one, but I believe I have a lot of reasons to be hopeful, and you have said and evidenced nothing against those, just some very generic statements weaving words around "exponential" where you continue to ignore various material changes in the situation, as reflected in the data, that have happened over recent months.  

> Well if there is 10% of the population who is not vaccinated and/or for whom the vaccine does not prevent hospitalisation, if you had exponential growth in that population it would more than big enough to overwhelm healthcare.

I flat out disagree.  It depends on which 10% of the population that is, and what their hospitalisation risk is - far from uniform across many dimensions - and on the speed of that logistic growth (we are running out of people without antibodies, and the number is decreasing by the day; exponential growth is a less appropriate model than it was); if it's low rate enough healthcare may never be saturated for a more vulnerable 10%; let alone the kind of status the UK appears to be heading for.  You seem to think the effects of vaccination are binary as well - hospitalisation or not is far from binary.  You can tell that now by the changing relationships between admissions and death/occupancy figures; people who have had one or two vaccine doses and are then hospitalised would appear to have a significantly improved outlook.

> as for international comparisons, we’ll only be able to do that way down the line

Funny, a couple of posts ago you said this to elsewhere:

> Except that the US has got falling cases and a much lower number of infections per capita, and we’ve got rising cases and a much higher number of infection par capita. Whatever they are doing is working better.

Can you explain to me how you get to make comparisons now but tell everyone else they can't?

> I did not miss your point. In fact acknowledged it. 

No, you completely missed my point.  On several levels.

 leahgoodall3 22 Jun 2021
In reply to wintertree:

> Given the relatively weak restrictions in the UK and the massive growth in younger adults, and the very slow growth in the older age groups, why do you think transmission has stalled in older people?

Seems obvious to me that it is due to vaccines, unless I am missing something else.

> On what basis do you think it would suddenly go high-rate positive exponential?   

They wouldn't, instead high prevalence rates amongst the younger, unvaccinated or partially vaccinated groups would spill out on the older groups and catch the small fraction that is still vulnerable. 

Thank to the ongoing vaccine rollout programme we can get R slightly above 1 and still keep admissions under control for now. That's a big plus.

But an increase of R to anything above 1.5 or 1.8 would - at this point - most likely result in substantial number of hospital admissions - despite the more favourable composition of the susceptible population, and despite optimistic assumptions about rate of hospitalisation amongst the vaccinated.

Post edited at 15:24
4
OP wintertree 22 Jun 2021
In reply to leahgoodall3:

Why are you posting from two different accounts on this thread simultaneously?

Getting your sock puppets in a muddle?

 MG 22 Jun 2021
In reply to wintertree:

Ah, Rom's back.  Is he simultaneously in Scotland, France and Cyprus still?

 leahgoodall3 22 Jun 2021
In reply to MG:

> Ah, Rom's back.  Is he simultaneously in Scotland, France and Cyprus still?

Still stuck on plague island - ahem sorry, Scotland - I'm afraid. Don't worry, not for long.

Post edited at 16:05
6
OP wintertree 22 Jun 2021
In reply to MG:

> Ah, Rom's back.  Is he simultaneously in Scotland, France and Cyprus still?

In a shocking twist, I think they're also back with a Covid denialism account.  Play both sides for maximum kicks.

 leahgoodall3 22 Jun 2021
In reply to wintertree:

> In a shocking twist, I think they're also back with a Covid denialism account.  Play both sides for maximum kicks.

Yeah sorry but that's not me. Would never propagate such bullshit btw.

Post edited at 16:15
5
OP wintertree 22 Jun 2021
In reply to leahgoodall3:

> Yeah sorry but that's not me. Would never propagate such bullshit btw.

Why would I believe a word you say when you hide dishonestly and disingenuously behind dozens (literally) of different UKC user accounts with many different names - sometimes male, sometimes female, sometimes insulting, and when you refuse to either disclose your full suite of identities let alone promise to stop?

It's possible that another account signed up at the same time as your latest batch of sock puppets, and posts on the same very restrictive sub-set of topics as you, but is actually a different person.  But how can I tell?

 leahgoodall3 22 Jun 2021
In reply to wintertree:

> Why would I believe a word you say when you hide dishonestly and disingenuously behind dozens (literally) of different UKC user accounts with many different names - sometimes male, sometimes female, sometimes insulting, and when you refuse to either disclose your full suite of identities let alone promise to stop?

Well you could look at my posting history and find out that I've spent quite a lot of energy fighting multiple covid deniers and other covidiots from the very start of the pandemic, in many cases beyond what could be considered a reasonable use of time.

But you will believe what you want, and hear only what you want to hear, and never question yourself, that's your style and fair enough.

Have a good day and good bye, I might pop up again at some point to poke a small hole in your absolute certainties.

9
OP wintertree 22 Jun 2021
In reply to leahgoodall3:

> Well you could look at my posting history 

You know full well that I can't look at your posting history, as you refuse to list which accounts are yours.  It's very dishonest and disingenuous of you to claim I could just look.   You've had a few dozen of them by know that've been rumbled, who knows how many others you have.  As your accounts are endlessly getting blocked it's even harder to see what you've written in the past over your dozens of identities.

I continue to believe it probable that you are behind the most recent misinformation troll account.  The timing and the forum activities are very correlated.   The word of a habitual liar willing to use multiple different identities to manipulate a single thread (not this one where you muddled your same-named sock puppets, but another one) from multiple perspectives - for what ever reason you are doing this - is worth nothing to me. 

> But you will believe what you want, and hear only what you want to hear, and never question yourself, that's your style and fair enough.

I disagree fundamentally.  

I spend a fair amount of time most weeks week looking at as much as I can and testing my interpretation and analysis against it to see if it holds water, giving me some confidence for my views.  Sometimes I'm pretty wide of the mark.  What I write is full of qualifiers intended to show where I am more - or usually - less certain in my interpretation. 

You - well I don't think you do.

> Have a good day and good bye, I might pop up again at some point to poke a small hole in your absolute certainties.

If you're reading what I'm writing as "absolute certainties" on these threads then you are I think in a very small minority.  

 mcdif 22 Jun 2021
In reply to Misha:

>I don’t expect life to go back to ‘normal’ till next summer and not planning to ditch the face mask any time soon!

A lot of fully vaccinated people are still anxious about returning to a normal life. The first article of this briefing might help to put the risk into perspective:

https://www.nytimes.com/2021/04/19/briefing/european-soccer-league-mars-hel...

1
 mcdif 22 Jun 2021
In reply to wintertree:

> If it’s the weather - well, lots of glorious weather ahead.  

Covid seems to thrive in India in May just as well as it does in Russia in December! Can a fine spell of UK weather really have much of an impact?

OP wintertree 22 Jun 2021
In reply to thread:

Some updated Lissajous plots comparing cases vs various other measures.

There's a lot going on in these, but in short the further below the last wave (purple) the current wave (blue) is, the better.  

Following on from some comments from Si dH on the previous thread, I've added ITU occupancy.  There's less improvement in the cases/ITU occupancy between waves for all areas than there is in cases/total occupancy.  In the North West, ITU occupancy is now higher than it was in the last wave for the same number of cases.  It's not entirely fair to compare two waves in such a simple and direct way as this, because changing case trajectories also walk these curves around on temporary excursions, but it's concerning enough to put some big warning lights on.  

Lots of possible factors at play - some discussed before, and so many that I don't think it's worth going through them all as it just becomes a list of words without evidence against them - but it would be good to see the media pressing the government to address this worrying trend - for example with data on ITU admissions in the North West broken down by both age and vaccination status.  

It seems to be particularly extreme in the NW although ITU occupancy is higher - relatively speaking - than the last wave for the whole of England and for Scotland as well.

From the dashboard data, it's basically impossible to say where this is going.  My hunch is that lack of vaccination in older people is still a factor here, given the previous snippets of data on hospitalisations and vaccination status, but I'm out on a limb of hope and not strongly evidenced on this.  If the waves don't separate in another couple of weeks that's starting to look really concerning for allowing continued growth of cases IMO.


In reply to wintertree:

> I'm not sure the USA is "working better", not by a long way.  Not when looking at the endgame enabled with current vaccines - a push to mostly non-lethal endemic status.

The US is winning at the moment for two reasons:

1. They've not been using AZ, it is pretty much Pfizer and they have a ton of the stuff getting made in the USA so more effective vaccine and no supply constraints on vaccine.

2. They did the border control more competently than the Tories and didn't get seeded with the Indian variant when the UK did.   It is there now and eventually it will become dominant like it did here but they got a few months more jags in before that happens.

They were playing catch up because they started later than the UK but Pfizer and Biden fixed that.  Now the limitation is holdouts who won't take the vaccine and there are probably more of them in the US than in the UK.

8
 Dr.S at work 22 Jun 2021
In reply to tom_in_edinburgh:

re hold outs - tons more - their vaccination trajectory is really tailing off.

https://ourworldindata.org/covid-vaccinations

way to go Canada though!

 Misha 23 Jun 2021
In reply to wintertree:

Some thoughts:

1. I sort of agree with 'leahgoodall' that high case numbers are bad news as inevitably those cases would reach the 'vulnerable' population who are either unvaccinated or who are among the 10-20% of vaccinated people for whom the vaccines aren't effective at preventing hospitalisation (obviously only a small portion of this 10-20% would actually get hospitalised). After all, the modelling around a month ago suggested a really bad wave if Delta is 40%+ more transmissible, which it seems to be. I'd like to see an update of that modelling reflecting latest data on vaccine effectiveness and Delta hospitlisation rates. Cases are certainly rising but not as fast as feared, for now.

2. Hospitalisations seem to be stagnating but I'd like to see another couple of weeks of data.

3. I hope there was a glitch in today's vaccination data...

4. I don't get why people would post from more than one accounts (at most two, if people have a different device at work etc). Unless they are pure trolls.

1
In reply to Dr.S at work:

> re hold outs - tons more - their vaccination trajectory is really tailing off.

Plenty of people still support Trump even after the Capitol riot: there's a demographic that isn't going to listen to reason. 

The other factor is the infection rate in the US is fairly low at the moment because the Delta variant hasn't got established.   The UK has got far more infection and the more dangerous Delta variant which means a lot more incentive to get vaccinated.

Probably when Delta establishes itself in the US people will get scared and we will see US vaccine uptake rise.

Post edited at 05:18
1
 Si dH 23 Jun 2021
In reply to Misha:

> Some thoughts:

> 1. I sort of agree with 'leahgoodall' that high case numbers are bad news as inevitably those cases would reach the 'vulnerable' population who are either unvaccinated or who are among the 10-20% of vaccinated people for whom the vaccines aren't effective at preventing hospitalisation (obviously only a small portion of this 10-20% would actually get hospitalised). After all, the modelling around a month ago suggested a really bad wave if Delta is 40%+ more transmissible, which it seems to be. I'd like to see an update of that modelling reflecting latest data on vaccine effectiveness and Delta hospitlisation rates. Cases are certainly rising but not as fast as feared, for now.

> 2. Hospitalisations seem to be stagnating but I'd like to see another couple of weeks of data.

Agree with all the above. I'm feeling cautiously optimistic looking at the trajectory of case data for the last few days but it's too early to be sure. We don't have any idea yet what will happen in the areas where cases have moderated after restrictions are removed on 19th July.

(Strangely high reporting spike in Scotland yesterday, by the way - where I'd that come from?? I really hope it's not a genuine daily increase.)

> 3. I hope there was a glitch in today's vaccination data...

"22 June

No update to England vaccination

Following an IT issue reported to the NHS on 21 June, it has not been possible to update vaccination figures for England. UK totals only include updates from Northern Ireland, Scotland and Wales

The issue is now resolved, and the daily vaccination statistics will be updated on 23 June, as vaccinations are added to the digital system. There has been no impact on vaccinations taking place."

> 4. I don't get why people would post from more than one accounts (at most two, if people have a different device at work etc). Unless they are pure trolls.

The way I see it they are either:

- giving a genuine opinion, in which case they deserve a proper response, or

- trying to influence people in an underhand way, in which case the best way to prevent them doing so is to inform others by giving a proper response*, or

- just trying to get a rise out of people for a laugh, in which case you shouldn't rise to it because that's what they want - you should either ignore it or give a proper response, which they will find dull.

To be honest I didn't see anything in the thread above which warranted the heated stuff. It's sometimes best just to assume positive intent unless proven otherwise, then just ignore it.

* I'm talking about a forum like ukc here. On social media it's a bit different because of the way things get shared and get much greater views if they are responded to and shared more.

Post edited at 07:09
OP wintertree 23 Jun 2021
In reply to Misha:

> I sort of agree with 'leahgoodall' that high case numbers are bad news as inevitably those cases would reach the 'vulnerable' population who are either unvaccinated or who are among the 10-20% of vaccinated people for whom the vaccines aren't effective at preventing hospitalisation (obviously only a small portion of this 10-20% would actually get hospitalised).

Yes, their final contribution is a very different point to the one they started with and is relevant IMO. 

Stepping back to big picture stuff...  

Once we get to the point everyone has been offered the vaccine, there is a choice between maintaining controls indefinitely or allowing the virus to circulate and "fill in the gaps" in the vaccination.  

  • I don't think there's any support from anywhere for indefinite controls, so we are left with little practical choice.  We can't live in an indefinite limbo because neither any vaccination nor population engagement with it are 100% perfect.
  • If we choose to let the virus circulate, there is as I see it no benefit and significant risk to drawing this phase of the release out much longer than necessary.   The risks revolve around the continued divergence for variants from vaccines.

Obviously as cases rise we have to look at "bleed through" from the demographically "hot" areas of cases into vulnerable people in other ages.

  • If this is happening much it will become clearer as the case numbers further diverge between the high- and low- growth parts of the demographic - as they are doing with their demographically segregated rate constants.  
  • It's not obvious to me that drawing the process out a lot makes it safer for individuals in the higher risk categories when it comes to infection from the hot-growth categories.  It's not even clear that it's better in terms of healthcare overload to draw it out more.  Some models of human interaction, groupings and network effects will suggest that a shorter, hotter period for the young has better net isolation of cases from the old than a drawn out process.  I don't claim to have a clue either way on this, but I can see arguments both ways. 

As cases rise, policy has to be guided by the need to maintain universal healthcare, and at this point not to drive a large number of staff away from from the profession after the last 12 months.  We just can't run hospitals up to their red lines again.  But it's also increasingly hard to justify delaying the inevitable for too long.  I've said before I would rather wait until we have some of the promising therapeutics out of the trials pipeline - but that's not a certainty, and the risks of leaving the next phase too long are not insignificant.   

At some point, the choice becomes "indefinite restrictions" or "move forwards".  We'd been stalling for time for vaccines, we have those.   There's not much left to stall for now, the weather is as much in our favour as it can be, and that's a finite, short window.  The window in which to allow circulation of the virus before it diverges further from cross-immunity in particular from current vaccines is likely finite and shrinking by the day. 

>  Hospitalisations seem to be stagnating but I'd like to see another couple of weeks of data.

"Two more weeks" is the mantra of the last couple of months, isn't it?  I suspect in two weeks time people will be saying "we need another couple of weeks of data".  In it's own way it's reassuring that this is repeatedly the case.

> 4. I don't get why people would post from more than one accounts (at most two, if people have a different device at work etc).

I've seen a few really good reasons/posts from a second account for anonymity purposes.  They invariably state "This is not my normal account, because..."  In this case, they use one of the accounts to like/dislike other posts in a very obvious way, and that invariably evaporates with their account bans.  

As to why they've engaged with multiple accounts with different identities on a thread at once?  

> Unless they are pure trolls.

Imagine that.  In that case, they'd probably still be here under a different identity.  

1
OP wintertree 23 Jun 2021
In reply to Si dH:

> We don't have any idea yet what will happen in the areas where cases have moderated after restrictions are removed on 19th July.

I'd look towards some of the university outbreaks for a bit of insight - cases almost quadrupled down the road in a MOSA near us last week, daily rate of 1,800 per 100,000 reported a couple of days ago.  That's 1.8% of the MOSA testing positive on a day, I could well imagine 30% of the people there will have live infections at some point in the next week or so.   I don't think we'll see much correspondence in the hospitalisation signal - one to watch for in the next couple of weeks.

Most of these infected/infectious individuals are quartered away in student accommodation blocks and houses, so not sharing space with parents and grand parents; that'll be different come the inevitable outbreaks afterJuly 19th.  

The problem with growth rates like that is they leave no time to stop local outbreak growth with local restrictions before a potentially disastrous amount of local hospitalisation is locked in.  I'd want to stagger the next release geographically out of caution to maintain capacity and get a sense for how it's going to go, but others can make very good arguments against that on other grounds.

> To be honest I didn't see anything in the thread above which warranted the heated stuff

It sounds like you read the reply I wrote before consulting with my "should I bee an ass" flowchart I keep on my desk.

Using a sock puppet account to like your own posts is pretty tragic (the likes on their posts all evaporated with the account bans) and more to the point it's dishonest.  Intriguingly another pattern of single vote likes/dislikes has evaporated in the last half hour.

I think some sort of line in the sand on honest engagement with the forums is important to maintaining them as worthwhile.   When you have someone engaging through a suite of accounts - some created "in reserve" almost a decade ago, it's anything but honest.  When they use different "real" names for their profiles, it's less honest.  It's a form of lying in implying they are a genuine new poster - until actively challenged about it - and it shows no respect for the other people engaging honestly.  Having watched them drive a whole bunch of very different posters (myself included) round 30-post journeys in which their position subtly evolves to be the opposite of their starting point, I wonder if their mission in life is to derail certain discussions by filling them with noise.  It's a pretty specific set of topics on which they do this.  

OP wintertree 23 Jun 2021
In reply to Dr.S at work:

> re hold outs - tons more - their vaccination trajectory is really tailing off.

Indeed.  

It looks like it's perhaps starting to tail off at similarly low levels for a lot of our European neighbours.  In terms of risk of driving further vaccine evasion from the virus, that's disappointing and further raises the risks for everyone.   

> way to go Canada though!

Indeed.  I was also looking at there curve for NZ; they've started now but have a long way to go.  The consequences of loosing containment there are getting worse with the variants' increasing R0 values.  

 Offwidth 23 Jun 2021
In reply to AJM:

Cheers, its really interesting to get a look at how such estimates can go wrong.

Am I right in thinking that mattered a lot to actuaries in that data around 2011 as the 1919 and 1920 birth rates were so far out from normal assumptions that it impacted the estimates for those around 90 years old? That group of 'phantoms' are over a hundred now (not so many still alive) and the next set along  from the 1940's are smaller in number.

 jkarran 23 Jun 2021
In reply to leahgoodall:

> Well if there is 10% of the population who is not vaccinated and/or for whom the vaccine does not prevent hospitalisation, if you had exponential growth in that population it would more than big enough to overwhelm healthcare.

That 10% subgroup would have to be both largely physically contiguous* to facilitate transmission and largely at risk of serious complications for sustained exponential growth leading to healthcare overload. That or the transmissiblity will have to keep increasing to overcome the physical barriers normal life puts up between the few remaining unvaccinated. Or the harm done to the young who are currently at low risk but highly connected and unvaccinated will have to increase. Both are possible.

*or in localised clusters facilitating explosive contained outbreaks.

Schoolkids obviously form a contiguous mass of vaccinated people but they don't typically suffer severe complications. Care home residents might be a problem due to their high connectivity and lower vaccine efficacy in the old/sick but it's not looking to be the case at the moment.

It's a grand experiment we're living through but it does look like we might just have a way back toward 'normal' at the moment.

jk

 elsewhere 23 Jun 2021
In reply to wintertree:

> "Two more weeks" is the mantra of the last couple of months, isn't it?  I suspect in two weeks time people will be saying "we need another couple of weeks of data".  In it's own way it's reassuring that this is repeatedly the case.

It is very reassuring compared to the time when the "herd immunity by letting it rip" looked like a policy to literally decimate the elderly.

 AJM 23 Jun 2021
In reply to Offwidth:

I should start by saying I wasn't really involved with discussions on it around that time - I had only just qualified then and was working in a different area so less involved with assumption setting.

Having said that - it will have been important, because mortality rates around that age are moderately important (I say moderately because they are quite high by age 90, and after a certain point as long as the answer is still "high" the expectation of life from that point onwards is fairly small anyway. At a high level, in recent years/decades mortality improvements have tended to focus on slightly younger ages i.e. we have significantly improved the chances of people getting to 90 or 95 but made far less impact on their chances of getting from there to 100 or 105). But on the other hand, it's one specific event which (once the lives get above a certain age and the mortality gap starts to become significant) sticks out like a sore thumb on the right graph - although obviously knowing which graph to plot would benefit considerably from hindsight!

All in all, I would have guessed that it's moderately important, but less so than the cohort effect ( https://www.cambridge.org/core/journals/british-actuarial-journal/article/a... ) in terms of the impact it has on pricing and reserving. Key questions on how long you expect it to persist (both in terms of how many birth cohorts it extended to and how late into life it persisted, neither of which would have been clear at the start) can have quite significant impacts.

 elsewhere 23 Jun 2021

Changing the subject and as usual looking at the Scottish data...

First jabs (currently Pfizer & Moderna, ages 18-40) fairly static. Pfizer & Moderna usage is consistent with a steady supply. 

Outer Hebrides has 1st jab vaccination rates of 83% & 80% for ages 18-29 & 30-39 which is astounding. That must be close to completed/saturated.

Across Scotland, 2nd jabs declining, AZ usage declining and 40-49 not yet getting 2nd jabs. 
However, 2nd (AZ?) jabs almost completed/saturated at 80% for 40-49 in Outer Hebrides. I am not sure why AZ usage has not been maintained by moving on to 2nd jabs for 40-49 across Scotland. 

AZ usage no longer exceeds Pfizer & Moderna combined.
 

Post edited at 12:27
 Offwidth 23 Jun 2021
In reply to AJM:

I've beeb aware of the cohort effect for some time and thinking about it's wider implications. Looking at the Whitehall studies and health trends for specific worker groups it was a key reason the actuaries helping the trade unions were saying longevity was overestimated in many schemes before the UK increase in longevity slowed and stopped. The miner's scheme showed this at its most stark... idiots like Ralfe in the Telegraph were saying taking on the scheme would cost the taxpayer a fortune, when in fact it's so much the opposite that it looks like the government will reduce its percentage take to avoid political embarrassment in the red wall seats. USS looks a complete mess on current valuation standards but will go the same way....the decrease in autonomy and increased stress levels I see in Universities are not consistent with a long and healthy retirement........my generation of academics and the generation preceding really did 'have it easy' and long life on average was to be expected.

I guess I would normally apologise for taking the thread off at a tangent but I see covid as being significant in making a lot of professions change, some changes for good (flexible working) and some for the worse. In terms of academics I wonder how much their health has been damaged, by how many had to put a pause on their main motivation and promotion opportunity (research efforts) to make time for the extra teaching work, on online media and communication. Going back to 'normal' may be difficult under management that want their new media output cake and to still eat the benefits of research at old levels. I'm certainly glad I retired as this hit and don't have to face a more problematic future.

Post edited at 12:46
 JHiley 23 Jun 2021
In reply to wintertree:

> It seems to be particularly extreme in the NW although ITU occupancy is higher - relatively speaking - than the last wave for the whole of England and for Scotland as well.

Where are you getting the ITU data? Looking at Hospitalisations and mechanical ventilation numbers I'm not seeing that effect. Last time we had (rising) case numbers at current levels was at the end of September when the UK 7 day average in MV beds was about 328.4 vs 214 now while hospitalisations were 2531.3 compared with 1294.1 now.

Is it possible the data has been "smithed" a bit too much?

 AJM 23 Jun 2021
In reply to Offwidth:

I've not seen anything looking at the cohort effect specifically at a more granular level (broken down by socioeconomics). A lot of stuff on socioeconomic impacts more widely, obviously, but I don't recall anything specifically looking at cohort effect & scope/strength of. Be interested to see it, if you have a link?

I know some of the pension scheme consultancies do a lot of socioeconomic profiling of scheme memberships to try to do their mortality analyses. I've seen a bunch of presentations around that which have always been quite interesting. A lot of schemes are dominated by higher socioeconomic groups once you weight by pension amount just because the execs get so much more £ pension, even if by member count they look quite different.

I would have assumed that with Ralphe the main difference was investment returns (that was the item you've discussed previously) - I would have generally assumed the scheme liabilities would be more sensitive to assumed investment returns than longevity unless the changes to the latter were quite substantial. Did he argue about mortality as well?

Edit: p.s. agree your thoughts on covid driven changes at the bottom

Post edited at 13:14
OP wintertree 23 Jun 2021
In reply to JHiley:

> Is it possible the data has been "smithed" a bit too much?

I think the problem is I'm not a good word smith, rather than the data.  I think I muddled some terms in my previous message which doesn't help.

I'm (trying to) note a difference in the ratio of hospital occupancy then and now to the "MV bed" (ITU) occupancy then and now.  In the absence of many deaths, occupancy seems to be the key measure to keep under control. 

Using the dashboard data and 7 day moving averages between 18th June 2021 ("now") and 28th September 2020 ("then") chosen for similar daily cases in a rising phase 

  • Hospital occupancy:
    • old: 8762
    • now: 1140
    • ratio old:new =  7.7: 1
  • ITU occupancy:
    • old: 757
    • new: 200 = 3.8:1

So, for rising phases with similar daily case rates, we've seen a 7.7× drop in the total Covid patient occupancy of hospitals in England, but only a 3.8× drop in ITU occupancy.  This could be partly explained by the lags from admission to ITU admission and the different exponential rates for infections between last summer and now, but still...  There's more going on here I feel than I understand.  Perhaps it's more pre-emptive used of ITUs?  Perhaps it's a selective effect where people who are not protected from infection by their vaccines have worse immune systems and so are more likely to end up in ITU?  There's a sheaf of other possibilities.

Hope that clears it up?

OP wintertree 23 Jun 2021
In reply to Šljiva:

Ask and ye shall receive?  Some very interesting quotes from the advisor in that article on the coming months depending on how you interpret them.

 Toerag 23 Jun 2021
In reply to wintertree:

Following a comment on social media along the lines of 'How many are in hospital though?' on a thread about Jersey's current rise in cases, I looked at the English stats - assuming it takes a week to become a hospital admission, it would appear that someone testing positive* for covid has about a 1 in 17 chance of being hospitalised at present. That's not great odds.  Would be interesting to see if / how that rate has changed during the course of the pandemic but I can't be arsed to do the data mining & smithing...

*I'm assuming the hospitalisations are coming from people who tested positive a week before, and not people that didn't get a test but eventually presented themself to hospital when they were really sick.

 Toerag 23 Jun 2021
In reply to wintertree:

>  So, for rising phases with similar daily case rates, we've seen a 7.7× drop in the total Covid patient occupancy of hospitals in England, but only a 3.8× drop in ITU occupancy.  This could be partly explained by the lags from admission to ITU admission and the different exponential rates for infections between last summer and now, but still...  There's more going on here I feel than I understand.  Perhaps it's more pre-emptive used of ITUs?  Perhaps it's a selective effect where people who are not protected from infection by their vaccines have worse immune systems and so are more likely to end up in ITU?  There's a sheaf of other possibilities.

I'd suggest vaccination is preventing a load of 'minor' hospitalisations in people like 50yr olds perhaps?  We need more demographic admissions data to know really.  I don't think the lag will have changed, I've not heard of any studies noting a change in lag due to new variants.

 jonny taylor 23 Jun 2021
In reply to wintertree:

> Ask and ye shall receive?  Some very interesting quotes from the advisor in that article on the coming months depending on how you interpret them.

I don't know which ones stood out to you, but one that stood out to me was:

> infection rates [...] were concentrated in unvaccinated groups, including school-age children. "Unfortunately that picture will continue for a few weeks more until we get past the third wave, which will unfold in the next couple of months".

Any ideas on how I should unpack that statement? Do you think I'm right to read that as: the third wave will end once basically all the unvaccinated, including school children, have had it?

As you know, I'm feeling frustrated that there are some big underlying motivations and objectives being left unspoken about what is coming out of the government(s) at the moment. A final "rip" may well be the right - and only - way forward at this point, but it's quite a change of policy to not-be-announcing. I'm feeling pretty disconcerted this week at the mismatch between explicitly-stated advice and what the underlying policy is increasingly seeming to be aiming for...

 Offwidth 23 Jun 2021
In reply to AJM:

The stuff I had was mostly ten years ago or further back and I've no idea if it's in one of my filing cabinets in the garage gathering dust or got thrown away. I wish I kept better index records now but my life as an academic, doing a lot of international work as well as being on a union NEC and driven to climb as much as I was (to stay sane?) was a bit of a rollercoaster and rather incompatible with perfect record keeping.  I came away with clear impressions....including lots of unexpected complexity and a lot of respect for the actuaries that worked with the unions for explaining that and helping inform where to campaign. Plus some fabulous work from driven academics, whom wintertree reminds me of.

Yes Ralfe argued bullshit on numerous factors, mortality included. Early mortality was a particular factor for long term miners. The simple truth was the combined prudence on individual factors on final salary scheme under the valuation methodology added up to an imprudent effect on those pensions, that became critical due to various new factors following the 2008 crash. That same methodology failed to spot the miners scheme surplus, that work similar to that I saw with USS predicted. The final salary schemes were simply way more healthy than the methodology indicated under anything other than black swan threats. It's becoming increasingly counter-factual for USS as time goes on.... assets have doubled since 2011 yet the most recent valuation mysteriously implies new liabilities have outstripped that by a large margin. How in a fairly stable sector???

 AJM 23 Jun 2021
In reply to Offwidth:

Ah, that's unfortunate, but it also sounds like your record keeping is akin to mine!

> That same methodology failed to spot the miners scheme surplus

I would pin the blame on the assumptions rather than the methodology, although that may be a question of terminology. I'm not sure the method is structurally incapable of predicting a surplus - it's just that what happened was better than the assumptions used at the time. The valuation is obliged to be done on a prudent basis, so obviously you would expect a surplus more times than not, but whether the "unpadded" (before prudence) assumptions were unreasonable given what they knew at the time or appear unreasonable more with the benefit of hindsight is probably the key question. I'm not sure I know the answer. But in general that's why I think proper sensitivity analysis around the assumptions is useful in this sort of work - all models are wrong, but they can still tell you things that are useful.

OP wintertree 23 Jun 2021
In reply to Toerag:

> I'd suggest vaccination is preventing a load of 'minor' hospitalisations in people like 50yr olds perhaps? 

That's the tempting answer isn't it; hard to tell from data alone.

> .  I don't think the lag will have changed

Ah, sorry, I wasn't very clear there.  When using a fixed level of cases/day to pick comparator periods as JHiley and I did, with a fixed lag between case and hospitalisation and between hospitalisation and ICU, the relative proportion of the measures will still change between waves if the underlying case rate changes - the slower the cases are rising, the closer one is to a steady-state where the lags disappear; the faster cases are rising; so because cases are rising more slowly now than the last wave, even if nothing else had changed it's not strictly speaking a fair comparison between hospitalisation levels etc.  But as the lag from hospitalisation to ITU admission is small (~2 days from memory of various documents) this really can't explain the 2x difference in the ratio (ITU occupancy / hospital occupancy) between waves.

> I've not heard of any studies noting a change in lag due to new variants.

I've not seen the hypothesis tested in either direction.  As I've noted before, in a rising exponential phase the effect of changes to the hospitalisation rate and the lag can't be separated from the daily numbers alone.

OP wintertree 23 Jun 2021
In reply to jonny taylor:

> I don't know which ones stood out to you, but one that stood out to me was: 

infection rates [...] were concentrated in unvaccinated groups, including school-age children. "Unfortunately that picture will continue for a few weeks more until we get past the third wave, which will unfold in the next couple of months".

> Any ideas on how I should unpack that statement? Do you think I'm right to read that as: the third wave will end once basically all the unvaccinated, including school children, have had it?

To me it sounds like he's expecting this wave to play out to a state where gaps in immunity (from gaps in the vaccination program) - and in particular in the younger - are filled up to the herd immunity thresholds (for the India variant) as a result of the wave.

Which, when you consider the approval of the Pfizer for older children in the UK and the "leaks" that JCVI are unlikely to recommend extending the vaccination program in to that cohort in general, makes one wonder about exactly what sort of detailed conversations might have happened that aren't in the minutes.   

Edit:  I don't mean to give the impression I'm at odds with the decisions one might infer, but without published studies and analysis, I'm left inferring what I think the plan is and then putting my faith in the powers that be.  That comes naturally, obviously.

> As you know, I'm feeling frustrated that there are some big underlying motivations and objectives being left unspoken about what is coming out of the government(s) at the moment. A final "rip" may well be the right - and only - way forward at this point, but it's quite a change of policy to not-be-announcing. I'm feeling pretty disconcerted this week at the mismatch between explicitly-stated advice and what the underlying policy is increasingly seeming to be aiming for...

Yes, I've got the same panel of "information disconnect" lights flashing as last March, but without the corresponding "panic" lights that blew themselves up last year.  Strongly, re: quote you picked out.

One of the key questions to me is when the requirement to isolate upon a positive Covid test will be dropped - this has pretty major ramifications in a lot of ways, not just directly related to Covid, but for a lot of employees who are very jittery about returning to the workplace.

If we're not both barking up the wrong tree entirely, it seems to me that there's a bit of a needle to be threaded to get this right during the summer, and not have it happen in the winter when the weather is against us and the potential bad flu season lands.  

Post edited at 16:16
 jkarran 23 Jun 2021
In reply to jonny taylor:

> As you know, I'm feeling frustrated that there are some big underlying motivations and objectives being left unspoken about what is coming out of the government(s) at the moment. A final "rip" may well be the right - and only - way forward at this point, but it's quite a change of policy to not-be-announcing. I'm feeling pretty disconcerted this week at the mismatch between explicitly-stated advice and what the underlying policy is increasingly seeming to be aiming for...

We do seem to have an odd approach to risk here. We won't vaccinate children because the reward to them is small and it would involve a small but non-zero risk of some suffering adverse reactions to the vaccine. Yet we need them to develop some herd immunity which as you note is unstated by government for obvious reasons so we let them catch and spread covid instead of accessing a vaccine that is likely safer. I guess it's more palatable for covid to kill/maim say 5 children for the greater good than the vaccine program to maybe kill/maim 1 to the same end.

While this seems bonkers it probably does make utilitarian sense due to the risk of fuelling wider vaccine aversion through the creation of a few highly emotive stories of tragedy.

jk

 jonny taylor 23 Jun 2021
In reply to wintertree:

> One of the key questions to me is when the requirement to isolate upon a positive Covid test will be dropped - this has pretty major ramifications in a lot of ways, not just directly related to Covid, but for a lot of employees who are very jittery about returning to the workplace.

Definitely. In terms of *contacts* self-isolating (which I know is not what you said): I have no desire whatsoever to self-isolate all the time on behalf of my employer. In terms of *positive tests* isolating (if people are even being tested any more...), I rather hope that a positive change from the pandemic will be people being more willing to stay home if sick - and employers being supportive of that. Although in many jobs of course that's easy to say now, but harder in practice...

> If we're not both barking up the wrong tree entirely, it seems to me that there's a bit of a needle to be threaded to get this right during the summer, and not have it happen in the winter when the weather is against us and the potential bad flu season lands.  

Yesterday's scot gov review of physical distancing (para 85) refers to "uncertainty around how much voluntary physical distancing will persist". They mention it in an economic context, but I daresay that's also a big error bar in somebody's projections of population immunity evolving between now and the winter.

But if the last two days numbers are anything to go by, if they are going to miss threading the needle it's not going to be in the direction of "too slow". I told myself I would wait until todays Scotland numbers were out; now that they are, all I can say is I *really* hope the govt advisers knew this was coming, and that this curve fits in with their plan...

Post edited at 17:03
 Si dH 23 Jun 2021
In reply to Si dH:

> (Strangely high reporting spike in Scotland yesterday, by the way - where I'd that come from?? I really hope it's not a genuine daily increase.)

Even higher today. Scotland's highest reported number of daily cases throughout the entire pandemic. Did they test everyone going to Hampden Park or something?

If this isn't an isolated outbreak or the product of a testing anomaly there is a large problem.

https://coronavirus.data.gov.uk/details/cases?areaType=nation&areaName=...

Post edited at 17:22
 minimike 23 Jun 2021
In reply to jonny taylor:

Agreed. This really feels like the end of the localised pre-third wave and the beginning of it proper (ie national). 
 

anecdotally from a GP friend at a large practice her patients no longer care and are unwilling to be tested even with symptoms. But they still want a f2f appointment obvs!

 jonny taylor 23 Jun 2021
In reply to Si dH:

> Did they test everyone going to Hampden Park or something?

Interesting idea - although the positive rate has also soared, which would seem to imply it's an outbreak rather than increased asymptomatic testing?

 Ramblin dave 23 Jun 2021
In reply to jonny taylor:

Yeah. A related thing is customer-facing businesses having to shut down entirely because of a positive test.

It's difficult to know what to make of it when the government's policy basically seems to be "don't bother trying to control case numbers" but the burden of self-isolation means that it's still a significant issue for an individual to catch the thing, regardless of personal risk.

 JHiley 23 Jun 2021
In reply to wintertree:

Ah OK, I see what you're getting at. Thanks again for toiling in the data smithy.

OP wintertree 23 Jun 2021
In reply to jonny taylor & Si dH:

> > Did they test everyone going to Hampden Park or something?

> Interesting idea - although the positive rate has also soared, which would seem to imply it's an outbreak rather than increased asymptomatic testing?

A lot of this “OMFG spike” as I shall call it is landing on the most recent two days of data for Scotland. In England where LFD and PCR data can be accessed separately, PCR data normally starts landing a bit further back. It seems reasonable to infer that some of this is LFD results landing from mass testing.  Or not….?  Shame the data isn’t broken down for Scotland.

Jonny - indeed I should have addressed both forms of isolation separately; the other Dr Wintertree is likely to be in a similar situation to the one you raise. I suspect this won’t get public air time until after July 19th even if things go well until then.  A good point Ramblin dave raises related to this too.

Post edited at 17:51
 elsewhere 23 Jun 2021

Scotland first jabs for ages 30-39 stuck at 76% for three consecutive days including today so that looks completed/saturated. I expect it will creep up another 2% over the next fortnight like ages 40-49 did after "getting stuck" but nothing significant.

Assuming uptake/completion/saturation percentage tails off in a smooth curve with age this suggests 1st jabs for ages 18-29 will be completed/saturated at about 67% in about 10-14 days then creep up by 2% over the following fortnight.

Post edited at 18:17
 Å ljiva 23 Jun 2021
In reply to jonny taylor:> > Any ideas on how I should unpack that statement?

ha, I wouldn’t have let that one through, it doesn’t make much sense. How can we get past something which will will take a couple of months to unfold in a few weeks?!

 jonny taylor 23 Jun 2021
In reply to Šljiva:

> which will will take a couple of months to unfold in a few weeks

I refer once again to the last two days of case numbers

In reply to elsewhere:

If you're still trying to get to the bottom of population estimates and how they affect uptake figures, Mainwood has tackled it: https://mobile.twitter.com/PaulMainwood/status/1407617961406484480

The take away is that the ONS and NIMS estimates diverge as you go down the age range, so headline uptake percentage figures are likely to be way less reliable in the younger ages.

Graphic here: https://mobile.twitter.com/PaulMainwood/status/1407256437286375425

Post edited at 18:39
 Offwidth 23 Jun 2021
In reply to Longsufferingropeholder:

Cheers for that but am I missing any detail as it just seems to give the overall effects?

Today's cases are worrying.

https://www.theguardian.com/world/2021/jun/23/sharp-rise-in-uk-covid-cases-...

2
In reply to Offwidth:

There's loads more if you manage to pick through the tangled web of Twitter. Just click on his name and read all his stuff and you'll stumble on it

 Offwidth 23 Jun 2021
In reply to Longsufferingropeholder:

I've read lots of his excellent tweets. I've just not seen any detailed analysis of population data.

 elsewhere 23 Jun 2021
In reply to Longsufferingropeholder:

Not bothered if denominators are a bit off,  it doesn't change how people decide whether or not to get vaccinated so no impact on real world completion/saturation. It doesn't change whether or not real people registered with a GP get an invite for vaccination either.

Post edited at 20:16
In reply to elsewhere:

But it massively changes the numbers you're pointing at in these threads. Potentially by 20%

 Si dH 23 Jun 2021
In reply to Longsufferingropeholder:

Are you, in fact, Paul Mainwood?

In reply to Si dH:

Lol. I'm not, no. I can't stand Twitter. He was definitely a climber in the past though, so quite probably has an account on here.

 elsewhere 23 Jun 2021
In reply to Longsufferingropeholder:

20% discrepancy in numbers I quote?

You reckon age 30-39 first jabs in Scotland completed/saturated closer to 56% or 96% rather than 76% that I think?

OP wintertree 23 Jun 2021
In reply to Si dH:

> Are you, in fact, Paul Mainwood?

We've all been thinking it...  

In reply to elsewhere:

Neither. I'm not offering any better guess. But the denominator being potentially off by 20% is a big deal.

In reply to wintertree:

Just a fan of his work. Combined with these threads it's just about all I could want to know.

But..... WintertreeMainwood...... Both into covid stats...... Coincidence??? 🙂

Post edited at 21:04
 elsewhere 23 Jun 2021
In reply to Longsufferingropeholder:

Potentially up to 20% off includes potentially 0.000000% off. It's not quite but almost meaningless unless you suggest a guess, correction or estimate of where something is in a 40% range.

You reckon age 30-39 first jabs in Scotland completed/saturated 20% earlier or 20% later in the 6 month vaccination programme? Maybe 30 days ago in May or in 30 days from now in July rather than in the last few days?

OP wintertree 23 Jun 2021
In reply to Longsufferingropeholder:

> But..... Wintertree. Mainwood...... Both into covid stats...... Coincidence??? 🙂

Synchronicity?  The chromium plated megaphone of destiny?

You can knock it,
You can rock it,
You can go to Timbuktu,
But you'll never find a wintertree on twitter!
You may see an Anaconda, or Giraffe and Kangaroo,
But you'll never see a wintertree on twitter!

 The New NickB 23 Jun 2021
In reply to leahgoodall:

> Except that the US has got falling cases and a much lower number of infections per capita, and we’ve got rising cases and a much higher number of infection par capita. Whatever they are doing is working better.

Hardly testing and pretending that cases are low, deaths tell another story.

OP wintertree 23 Jun 2021
In reply to The New NickB:

> Hardly testing and pretending that cases are low, deaths tell another story.

It’s hard to know how much is down to under-testing and how much is down to significant social stratification meaning those who are dying are from separate strata to those getting tested.   To me, the whole covid situation has shone a spotlight on several major weak points of the modern world, the most prominent of which by far is unsustainable levels of inequality.  Gross inequality when socially normalised undermines so much.

 Misha 23 Jun 2021
In reply to wintertree:

I'd be more cautious about 'freedom day' (I know that's not quite what you mean), for two reasons.

Firstly, more cases will lead to more hospitalisations. Nick Triggle summarised it nicely today: "During the winter, around 10% of positive cases ended up in hospital 10 days later. At the moment this looks to be around 3%, showing the impact vaccines are having. But if cases keep rising as they are, it's possible we will see in excess of 1,000 admissions a day later in the summer - what the NHS would see for all types of respiratory illness in the middle of a bad winter. The hope is the wall of immunity built up by the vaccination programme will start to flatten the wave soon." I don't think we want to see 1,000 a day again, even if deaths remain relatively low. If 3% is correct, that's 30k cases, which we will easily get at this rate - it's only one more doubling time, give or take (almost 15k by date reported for Monday as of today).

Secondly, runaway cases in a highly vaccinated population means vaccine resistant variants are more likely to emerge.

The other point is that we have already removed the vast majority of restrictions. I appreciate that not everyone is in the same boat but, from my point of view, most of the things which are still restricted are 'nice to haves' and personally I can already do pretty much everything I might want to do (the main impact on me is WFH, face masks and general SD but none of that particularly bothers me for the next few months / year). I get that some people want to go to gigs etc - well, tough, will have to wait another year. More importantly, some people still aren't back to work but, as a taxpayer, I don't mind paying for their furlough in order to reduce the risk to the wider economy (plus the economy is heating up and there are job shortages in some sectors, so if someone really wants to find work they should be able to do that, albeit in a different role / sector).

As things stand today, I just don't think it's worth the risk of opening up fully on the 19th. No doubt the data will evolve and we would have a better view in 2-3 weeks' time (when the facts change, I change my mind).

I agree that a wider unlock may be reasonable once materially everyone who wants it has had a second dose + 2-3 weeks. That won't be till the around September. Even then, I think some restrictions may well need to remain - WFH, face masks, SD indoors and limited international travel (perhaps less limited than it is now) would be high on my list.

There is perhaps a case for letting cases burn through the unvaccinated (noting that there will be some people who can't be vaccinated for medical reasons, though that is presumably a pretty small %) and doing that before we run into the winter flu season. However I don't think that's wise or fair until everyone who is willing has been able to get a second jab + 2-3 weeks.

1
 Misha 23 Jun 2021
In reply to wintertree:

We have another outbreak at Birmingham Uni here (impacting 2 MSOAs in student areas). The worst hit MSOA has a rate of over 1,000! Those students will be socialising in pubs etc alongside the non-student population, as well as fanning out across the country back to their parents in the next week or so...

 Misha 24 Jun 2021
In reply to jkarran:

Also by not vaccinating children we save about 15m double doses which could be used by another country to vaccinate adults (in practice it would be less as not all children would have it anyway).

 Misha 24 Jun 2021
In reply to jonny taylor:

> Definitely. In terms of *contacts* self-isolating (which I know is not what you said): I have no desire whatsoever to self-isolate all the time on behalf of my employer.

True but I also have no desire whatsoever to be in an office with colleagues who might have Covid but aren't self isolating because they don't need to. Which goes to your point that hopefully it will no longer be socially acceptable to come in to the office with a cold etc. But not everyone is socially responsible.

 Misha 24 Jun 2021
In reply to wintertree:

> It’s hard to know how much is down to under-testing

Under-testing is certainly an issue and has been throughout. Question is whether the % of people unwilling to get tested has increased. I guess it's very hard to find out (surveys?). There could be an element of this as (1) more places are open now and some people wont want to 'miss out' on whatever it is they want to be doing ; and, more significantly, (2) most people are back to work now and some of these people can't afford to self isolate, whereas when they were on furlough it didn't really matter. Still, I'd be surprised if people not wanting to get tested is a significantly larger issue now as a % of actual cases. The other consideration is that Delta seems to have a wider range of symptoms but scientific opinion seems to be split on that so far.

 Misha 24 Jun 2021
In reply to Ramblin dave:

> It's difficult to know what to make of it when the government's policy basically seems to be "don't bother trying to control case numbers" but the burden of self-isolation means that it's still a significant issue for an individual to catch the thing, regardless of personal risk.

The thing with personal risk is we just don't know what the long term effects of even mild infection might be. It's probably ok as long as you don't get 'long Covid' straight away but we just don't know if Covid can somehow lie dormant for years and then come back with a vengeance like some viruses do. A virologist might tell me I need not be concerned and I'm not overly concerned by this but at the end of the day I just don't want to catch the thing at all (despite being a healthy double jabbed 40 year old).

 Misha 24 Jun 2021
In reply to wintertree:

I can confirm that Longsufferingropeholder is not called Paul Mainwood. Whether he is the same person as the person posting as Paul Mainwood on Twitter, who appears to have a crocodile as a profile picture, is another question...

2
Roadrunner6 24 Jun 2021
In reply to tom_in_edinburgh:

The weather and restaurant scene here is very different. It's just not hard in the summer to stay outside, and everything is bigger so less congested for the most part. That's not always the case of course but generally it is. In the NE USA we just move outside from now until labor day.

The US is looking increasingly divided though so it's hard to talk about it as a single country, those states with 40-50% (some are in the 30's) vaccinated will suffer as these new strains take hold and especially in the fall. The blue states which will likely be at 70-80% vaccinated by the fall will very likely not suffer the same. In MA we're at (69.6%) 70% with at least one shot (60% have had both), and that's still ticking up about 0.5% a week, and also we'll start vaccinating younger kids soon and push that to 80%.

It looks like we'll hit regional herd immunity very soon, sadly I think many states will have to suffer another bad winter to increase vaccination numbers.

Post edited at 02:50
Roadrunner6 24 Jun 2021
In reply to The New NickB:

> Hardly testing and pretending that cases are low, deaths tell another story.

You keep saying this as though it's true. It's not regionally. 

We are seeing 3-4 deaths a day in a state of 7 million people.

The US is increasingly splitting in its response. But yes we can have confidence. We've vaccinated 70% of the population with at least one shot of a vaccine with 95% effectiveness. Why would we not see significant decreases in deaths and cases?

Post edited at 02:49
2
Roadrunner6 24 Jun 2021
In reply to elsewhere:

> In the US, 77% of Democrats are either fully or partially vaccinated but only 52% percent of Republicans. It's madness! 

Possibly madness but it's entirely predictable. Even now Trump is on TV saying don't vaccinate your kids. They've systematically attacked and chipped away at public confidence in science for decades and we now have this and the climate situation.

My wife has patients who are basically dying of lung cancer who won't get the vaccine because they won't put chemicals in their body which may have long term impacts.

We can't really stop regional travel though sadly so I think all we'll do is focus on regional herd immunity, many of us won't travel the same and we'll have to go through another winter for those states.

All we can hope is at least the elderly and vulnerable are vaccinated in those states but I'm not sure that is true. 

 BusyLizzie 24 Jun 2021
In reply to wintertree:

It's all a bit nerve-wracking at the moment isn't it. Many thanks, wintertree and all (well, most) on here for a consistent source of objectivity and perspective.

 Si dH 24 Jun 2021
In reply to Roadrunner6:

Where do you get your estimates from for how many people need to be immune to reach herd immunity, and what is it's basis? I've heard with the latest variants it's supposedly getting on for 90% but I think that just makes a simple assumption that the whole population acts the same (in practice you get more benefit from vaccinating a bus driver than a recluse.)

 Si dH 24 Jun 2021
In reply to Misha:

> Under-testing is certainly an issue and has been throughout. Question is whether the % of people unwilling to get tested has increased. I guess it's very hard to find out (surveys?). 

I don't think it's really a case of being unwilling to get tested. In my personal experience in my family (between us we've probably had 6-7 PCR tests with symptoms I think) it has nearly always been a judgement call as to whether our symptoms warranted it. I think there has only been one black/white case where one of us was coughing an awful lot. Many people will have very mild symptoms that they decide don't really tick the boxes to get a test - especially now the primary symptoms of the virus have apparently evolved a bit. When that happens, people inevitably weigh the possible need for a test against the downsides of doing so - an acceptance that you should probably all self isolate and the physical/emotional downsides of doing the test (significant for kids but not adults). Now that many people are double vaccinated and there is a general perception that high cases are less of a problem, I think many people with a heavy cold or an occasional cough who aren't sure whether they really tick the symptoms box, who would have got tested before will no longer do so. This will be a much greater number of people than those who would tick a survey box saying "I won't get a test if I have covid symptoms".

Post edited at 08:05
 VSMaxMax 24 Jun 2021
In reply to Misha:

> I'd be more cautious about 'freedom day' (I know that's not quite what you mean), for two reasons.

> Firstly, more cases will lead to more hospitalisations. Nick Triggle summarised it nicely today: "During the winter, around 10% of positive cases ended up in hospital 10 days later. At the moment this looks to be around 3%, showing the impact vaccines are having. But if cases keep rising as they are, it's possible we will see in excess of 1,000 admissions a day later in the summer - what the NHS would see for all types of respiratory illness in the middle of a bad winter. The hope is the wall of immunity built up by the vaccination programme will start to flatten the wave soon." I don't think we want to see 1,000 a day again, even if deaths remain relatively low. If 3% is correct, that's 30k cases, which we will easily get at this rate - it's only one more doubling time, give or take (almost 15k by date reported for Monday as of today).

^ this.

Also worth noting that the growth in cases has been firmly in the young for the past month and a half, but is now starting to spill out to the old, where it will inevitably go fetch the unvaccinated/vulnerable.

This could push case hospitalisation ratio up and create a rather sneaky wave of admissions, as the silent wave of infections suddenly starts hitting shore by moving through more vulnerable populations.

3
In reply to Misha:

> As things stand today, I just don't think it's worth the risk of opening up fully on the 19th. No doubt the data will evolve and we would have a better view in 2-3 weeks' time (when the facts change, I change my mind).

This is where the more detailed modelling becomes useful. Waiting longer is actually worse since it would push the 'exit wave' back into winter. All seemed to show 4 weeks was the optimal delay.

OP wintertree 24 Jun 2021
In reply to Misha:

>  don't think we want to see 1,000 a day again, even if deaths remain relatively low.

Soon enough, we may not have much choice left over how many people are going to hospital, like it or not.  The only choice we get is over how quickly this happens, and there are risks to going too fast and different risks to going too slow.  It's not a happy situation, but each variant to take over in the last year has reduced the range of choices further.

The key measures are occupancy rather than daily admissions and it looks like ITU occupancy is the measure that will lead the way.

> There is perhaps a case for letting cases burn through the unvaccinated (noting that there will be some people who can't be vaccinated for medical reasons, though that is presumably a pretty small %) and doing that before we run into the winter flu season.

Is there any alternative left?  Indefinite restrictions I suppose.

> There is perhaps a case for letting cases burn through the unvaccinated

If we can't reach the herd immunity threshold for the India variant through vaccination, what other choices remain but this and perpetual restrictions that tighten with each new, more transmissive variant?

> I agree that a wider unlock may be reasonable once materially everyone who wants it has had a second dose + 2-3 weeks. That won't be till the around September.

In practice I’m much less concerned about the next round of unlocking than the traction being gained by the political lobbying from the travel industry.  

  • With the scale of growth of cases (edit: in ages 15-25 where vaccination is less complete) (cases which are probably becoming more detached from infections, as various posters have noted) in the young end, I think there's going to be a lot of naturally acquired immunity there pretty soon.
  • This is the same travel industry that imported the first few thousand cases in to the UK in early 2020 by keeping with business as normal during the early months, and the same industry that brought so many India variant cases in recently having lobbied extensively against risk control measures that might put travellers off.  It’s fair to say I think they have a fundamental conflict of interest here and they are only to happy to externalise major costs - human and financial.  I suppose they're well practices at externalising costs when it comes to climate damage.

As BusyLizzie said: t's all a bit nerve-wracking at the moment isn't it

In reply to VSMaxMax:

> Also worth noting that the growth in cases has been firmly in the young for the past month and a half, but is now starting to spill out to the old, where it will inevitably go fetch the unvaccinated/vulnerable.

You only have to look at the publicly accessible demographic cases data to see that cases in all ages have been in growth for 6 weeks.  To my eye, there's no way to test a hypothesis of "spill out" from that data.  You seem to be trying to put a fixed narrative on to a fluid situation.

Post edited at 09:09
 Ramblin dave 24 Jun 2021
In reply to wintertree:

> > There is perhaps a case for letting cases burn through the unvaccinated (noting that there will be some people who can't be vaccinated for medical reasons, though that is presumably a pretty small %) and doing that before we run into the winter flu season.

> Is there any alternative left?  Indefinite restrictions I suppose.

> > There is perhaps a case for letting cases burn through the unvaccinated

> If we can't reach the herd immunity threshold for the India variant through vaccination, what other choices remain but this and perpetual restrictions that tighten with each new, more transmissive variant?

I do feel like the argument is a bit academic though - there doesn't currently seem to be much interest in "once every adult has been offered two doses of the vaccine". Cases are increasing and the talk is about how soon we're going to relax restrictions and let them increase faster. We're firmly on course to let the cases "burn through the unvaccinated" now, and that means that anyone young enough not to have been offered a second (or even first) dose yet had better assume the position...

 The New NickB 24 Jun 2021
In reply to Roadrunner6:

> You keep saying this as though it's true. It's not regionally. 

It is nationally.

 VSMaxMax2 24 Jun 2021
In reply to wintertree:

> You only have to look at the publicly accessible demographic cases data to see that cases in all ages have been in growth for 6 weeks.

Yes, from a situation where it was mostly the young.

>To my eye, there's no way to test a hypothesis of "spill out" from that data.  You seem to be trying to put a fixed narrative on to a fluid situation.

https://twitter.com/VictimOfMaths/status/1406941599779725321/photo/1

1
 Si dH 24 Jun 2021
In reply to Ramblin dave:

The government claims everyone will get the offer of a first dose before 19/07. Obviously a second dose would be better but I don't think they would be popular for delaying the date for the sake of people in their 30s or below getting a second dose, so they won't do it. The only thing to cause a delay will be if hospital data in the next few weeks suggests we are at risk of overload.

Moreover I think a delay to September before allowing some further rebound in cases would carry quite a big risk not only because of autumn/winter but also because of the unknown duration of immunity from the vaccine. The most vulnerable people had their second doses quite a long time ago now and we don't want any significant number of them to run out of good protection while there is a sizeable wave of infection occurring. Unfortunately, the first sign we get of this will be if significant numbers of people in their 80s and 90s start needing hospital and dying - so for many, it will be too late to react. It doesn't seem like a booster programme is going to happen soon enough that we can be confident that everyone will retain their immunity throughout. In conclusion...if it were up to me, I would only delay until everyone had their second dose if I was confident that doing do would preclude a large exit wave from occurring. I don't think we yet have enough information about immunity or r0 with Delta to be able to have that confidence though.

Post edited at 10:09
OP wintertree 24 Jun 2021
In reply to VSMaxMax2:

> Yes, from a situation where it was mostly the young.

Hello Rom.   One account per post now?  Seriously?  

> >To my eye, there's no way to test a hypothesis of "spill out" from that data.  You seem to be trying to put a fixed narrative on to a fluid situation.

You may have missed it, but I too have been looking at cases demographically - absolute numbers, probability distributions and exponential rate constants.

I refer you to my plot D1.c showing that broadly speaking all age ranges turned to growth at the same time, and growth has been higher in an exponential sense in the young, consistently.  But all ages have been growing with no clear leader/follower pattern.   

The lower rate growth in older people is now getting to the point where their data markers on my and Colin's plots change colour on my pseudo-log-z and his log-z colour maps, and become more visible.  This isn't a sudden "spilling over", it's just the lower rate exponential growth in the old reaching some arbitrary threshold.

Tell you what - when I said "to my eye, there's no way to test a hypothesis of "spill out" from that data. " I was wrong - with enough controls, perhaps that hypothesis can be tested:

  • For sure, if there were zero cases in those aged under 30, day, growth in higher ages would be lower, because transmission is a network effect and the network is only partially stratified along age boundaries.  
  • Growth anywhere contributes to growth everywhere.  
  • With different exponential rate constants in different bands, the amount of influence from the band with higher absolute cases on another band with lower absolute cases should increase with time as the case rates diverge in the bands 
  • This means that - all other things remaining unchanging - the rate constants would increase in the older bands over time, not as a result of in-band changes to the transmission rate but from out-of-band importations.  
    • I don't see any support for that in the data.  None.  

I think you might at this point Rom have to accept that you don't really understand the nitty gritty details of what you're talking about, haven't thought it through in sufficient detail and are little more than a source of noise.

 Offwidth 24 Jun 2021
In reply to wintertree:

The moderators should maybe write some script that auto banned first time posters on the likes of covid threads. Would probably save time.

 VSMaxMax2 24 Jun 2021
In reply to wintertree:

> Hello Rom.   One account per post now?  Seriously?  

> You may have missed it, but I too have been looking at cases demographically - absolute numbers, probability distributions and exponential rate constants.

> I refer you to my plot D1.c showing that broadly speaking all age ranges turned to growth at the same time, and growth has been higher in an exponential sense in the young, consistently.  But all ages have been growing with no clear leader/follower pattern.   

> The lower rate growth in older people is now getting to the point where their data markers on my and Colin's plots change colour on my pseudo-log-z and his log-z colour maps, and become more visible.  This isn't a sudden "spilling over", it's just the lower rate exponential growth in the old reaching some arbitrary threshold.

> Tell you what - when I said "to my eye, there's no way to test a hypothesis of "spill out" from that data. " I was wrong - with enough controls, perhaps that hypothesis can be tested:

> For sure, if there were zero cases in those aged under 30, day, growth in higher ages would be lower, because transmission is a network effect and the network is only partially stratified along age boundaries.  

> Growth anywhere contributes to growth everywhere.  

> With different exponential rate constants in different bands, the amount of influence from the band with higher absolute cases on another band with lower absolute cases should increase with time as the case rates diverge in the bands 

> This means that - all other things remaining unchanging - the rate constants would increase in the older bands over time, not as a result of in-band changes to the transmission rate but from out-of-band importations.  

> I don't see any support for that in the data.  None.  

> I think you might at this point Rom have to accept that you don't really understand the nitty gritty details of what you're talking about, haven't thought it through in sufficient detail and are little more than a source of noise.

Sure, so maybe it isn't "spilling out" in the sense that infection are not necessarily being transmitted from the old to the young specifically.
But whatever the cause or dynamic is, the result is the same: infection level is growing into the older age group which is bad news.

Post edited at 10:56
2
 Misha 24 Jun 2021
In reply to Si dH:

Good points.

 Misha 24 Jun 2021
In reply to Longsufferingropeholder:

Yes that’s a fair point, though as the situation evolves, so will the modelling... It’s certainly a tricky judgement call as the situation is far less clear cut that it was a year ago. 

Roadrunner6 24 Jun 2021
In reply to The New NickB:

> It is nationally.

No it’s not. The US is increasingly splitting on Red blue lines in their response. 
 

nationally means little in such a massive country managing covid at the state level.

2
Roadrunner6 24 Jun 2021
In reply to Si dH:

We don’t know do we? It’s changing all the time and it’s not an exact line. The Mass Dept of public health has said a few times mid June was when we’d reach it. 
 

I think they are going with about 80% of adults vaccinated. 

even when we say we’ve hit herd immunity with other diseases we can still see transmission, it’s just unlikely.

 Misha 24 Jun 2021
In reply to wintertree:

My key point is we should probably wait till everyone who wants to be jabbed has had two jabs + 2-3 weeks. Then let it burn through but the population without any immunity would be far smaller, so the impact on healthcare would be lower. Fair point that a winter wave may be worse but I’m not entirely convinced as it would be smaller overall - however I’d be very interested to see revised modelling in a couple of weeks’ time. I get that it’s concentration over a period of time rather than just total numbers. “Follow the science”, I’m just a bloke with a keyboard...

Re admissions, agree but you can see that occupancy is still rising, though less so than at the equivalent point last year. So I’m using it as a sort of proxy but it’s a crude measure.

Well there is another option - mandatory vaccination, or rather quasi mandatory by making it mandatory for a lot of places / vaccinations through vaccine passports (unless medically exempt). I would be fine with that as a last resort but obviously it’s not great on a number of levels. Better than another wave or perpetual restrictions though... The real issue is practical implementation difficulties probably make it not very feasible.

As for travel, ironically at the moment we are more of a risk to most other countries than the other way round but I agree it makes sense to retain significant restrictions, probably till next summer. 

1
 Misha 24 Jun 2021
In reply to Ramblin dave:

Well, the unvaccinated population can opt to avoid high risk situations. Personal choice and suspect most younger people willl just crack on with life. I would question going to the wall now if I weren’t double jabbed and if cases rise much higher locally, I may well refrain for a few weeks anyway... see how things go. 

1
 Misha 24 Jun 2021
In reply to wintertree:

I think it’s fair to say that the current wave originated in the younger age groups and is spilling out to an extent but significantly constrained by the vaccination drive. The question is what impact that would have on healthcare metrics.

By the way, what does Rom mean - is that a name?

 Misha 24 Jun 2021
In reply to VSMaxMax2:

You’re just annoying now. 

 MG 24 Jun 2021
In reply to Misha:

> By the way, what does Rom mean - is that a name?

It's the original name of the poster who has myriad identities, currently VSmaxmax122424

 Ramblin dave 24 Jun 2021
In reply to Misha:

I'm not sure how well you can opt to avoid high risk situations if you work in a pub or a shop or a call centre, though? Particularly if the government go ahead and remove all social distancing requirements in a few week, and if daily cases are running into the hundreds of thousands.

I mean, it might still be the least worst option available in the grand scheme of things, but it should be recognized that it's not just going to be a few mad antivaxxers getting immunity the hard way, it's going to be large numbers of young people taking a fairly hefty one for the team, again.

Post edited at 12:09
 Bottom Clinger 24 Jun 2021
In reply to Misha:

Some incoherent ramblings:  Bolton is worth a look. Cases have been  dropping for the past 4 weeks yet they still have below average vaccine take up (even with the extra ‘push’). Hospitalisations and people on ventilation appear to be plateauing. Deaths do seem to have gone up though, but seems reasonable to assume they will start coming down fairly soon. So even with current  restrictions there does seem to be some slack. I’m guessing there must be a lot of immunity acquired from infection?

BTW, I think the current Delta wave originated in adults coming back from India, but quickly spread into younger age groups. 

The idea that most of us can do most things apart from ‘the nice extras’ doesn’t wash with me: most folk I know really are sick to the back teeth of not being able to do normal stuff, especially stuff to with family and children and young people. And these ‘nice extras’ are for many people what makes life worth living. In many ways, the ‘nice stuff’ is more important then the ‘must do stuff ‘, if that makes sense.

The govt has promised all along ‘vaccines are our way out’ so at some point they need to put their money wher their mouth is. People are fully aware that there is a high probability that a new variant will pop up and that the whole cycle starts again. If we don’t watch it, a new cycle will begin before we are out of the old cycle.

I guess/hope in all the modelling then the Human Factor is factored in - not sure how much more of this people can take and they certainly won’t police things. 

My concern is ‘will enough people in the Usual Suspect areas get jabbed?’ I doubt they will by Freedom Day 2, and my gut feeling is that the number of un-jabbed and no second jab could be large in these areas.  

Note: I see Salford has now ‘plateaued’, Bolton and Stockport going down. 

 The New NickB 24 Jun 2021
In reply to Bottom Clinger:

I’m not sure Bolton take up is below average, given the different measures at National an local level. It is very hard to tell though.

Post edited at 12:36
 Toccata 24 Jun 2021
In reply to Bottom Clinger:

There seems to be huge pressure from the travel industry to open up on the 19th and it looks like the Government is putting out public opinion 'feelers' to see how the press reacts. One wonders what the 'science' is behind the decision-making process when the Prime Minister (today) says

"If you look at it, we've got more than 60% of our population have now had two jabs, I think 83% have had one jab, we're really getting through it now".

As I make it 46.5% and 65% (of the population) one wonders whether this is deliberate misinformation or whether it betrays a fundamental lack of understanding of population health management.

 Toerag 24 Jun 2021
In reply to Misha:

> By the way, what does Rom mean - is that a name?

The poster's old account was called 'Rom the bear' iirc.

 Toerag 24 Jun 2021
In reply to Toccata:

> "If you look at it, we've got more than 60% of our population have now had two jabs, I think 83% have had one jab, we're really getting through it now".

> As I make it 46.5% and 65% (of the population) one wonders whether this is deliberate misinformation or whether it betrays a fundamental lack of understanding of population health management.

They're invariably referring to the adult population.

 Toerag 24 Jun 2021
In reply to wintertree:

Some info on other jurisdiction's approaches:-

Jersey are saying double-jabbed people don't have to self-isolate if they're a contact.

https://jerseyeveningpost.com/news/2021/06/23/hundreds-to-be-freed-from-iso...

Guernsey are allowing double-jabbers in from the UK common travel area with no testing whatsoever from 1st July. Foolhardy in many people's eyes as we're completely unrestricted on-island at present, but the authorities believe enough islanders have been fully or partially vaccinated to prevent another lockdown. The reality is that self-isolation of contacts is going to be problematic - Jersey's stats show ~1000 people having to self-isolate through being contacts of 60-odd cases but they aren't allowing large gatherings. Apply this to Guernsey where we still have nightclubs open and festivals happening and there could well be super-spreader events.  Even if people aren't made ill the numbers of self-isolators will trouble the economy and society.

Post edited at 12:57
 Bottom Clinger 24 Jun 2021
In reply to The New NickB:

> I’m not sure Bolton take up is below average, given the different measures at National an local level. It is very hard to tell though.

Nationally: 82.5% 1st, 60.3% 2nd

Bolton: 75.2 and 54.6%

 Toccata 24 Jun 2021
In reply to Toerag:

That's my point. The simplest explanation is that the adult population numbers sound better. But is it a misunderstanding that the problem is solved once all adults are vaccinated? I'm pretty sure that every minister must be aware under 18s contribute to spread of infection. So what's to be achieved by conveying numbers that imply 2/3rds are vaccinated when actually it's probably closer to a third (that are three weeks post second vacc)? What bothers me is that this is yet another political fudge building to take yet another ill-advised course of action.

 Bottom Clinger 24 Jun 2021
In reply to Toerag:

More like they are accidentally on purpose missing out the word ‘adult’. 

 Bottom Clinger 24 Jun 2021
In reply to Toccata:

I’ve started to try and take a very long term view of Covid. I reckon it will take the import and export of a number of PITA variants over a number of years, and mini lockdowns and death spikes, for the govt and public to realise that global travel needs a massive rethink. 

 Offwidth 24 Jun 2021
In reply to Bottom Clinger:

Good article from Reicher (from SAGE) on covid response and behavioural psychology.

https://www.theguardian.com/commentisfree/2021/jun/24/psychology-uk-covid-r...

 Bottom Clinger 24 Jun 2021
In reply to Offwidth:

Thanks for that. A good read. When I get chance, I’d like to delve deeper into the detail on this stuff. 

OP wintertree 24 Jun 2021
In reply to Ramblin dave:

> I mean, it might still be the least worst option available in the grand scheme of things, but it should be recognized that it's not just going to be a few mad antivaxxers getting immunity the hard way, it's going to be large numbers of young people taking a fairly hefty one for the team, again.

Seconded.  

The respectful and mature way to approach this is with clear, consistent and honest messaging as well as guidance and support decided and publicised well in advance that recognises the consequences of the process.

In reply to Toccata:

> "If you look at it, we've got more than 60% of our population have now had two jabs, I think 83% have had one jab, we're really getting through it now".

This is pretty much what the Captain Picard facepalm meme was invented for.   

> one wonders 

One does indeed.

In reply to Misha:

> I think it’s fair to say that the current wave originated in the younger age groups and is spilling out to an extent but significantly constrained by the vaccination drive. The question is what impact that would have on healthcare metrics.

I'm not sure it is fair to claim that.  Have a look at the version of plot D1.c I've scribbled a straight edge on below. It looks to me like the tip-over to growth occurred at all ages at about the same time, but that the growth is seriously moderated in older people by the vaccination process.  The very oldest people are delayed a bit, but they are most likely in managed environments more protection cast around them.


 Maggot 24 Jun 2021
In reply to Toerag:

I now know two people who have been double-vaxxed weeks ago that have very recently tested positive

 TomD89 24 Jun 2021
In reply to Maggot:

> I now know two people who have been double-vaxxed weeks ago that have very recently tested positive

OK but what does them testing positive actually translate to? Are they seriously ill? Slightly ill? No symptoms at all?

Roadrunner6 24 Jun 2021
In reply to Maggot:

> I now know two people who have been double-vaxxed weeks ago that have very recently tested positive

That’s what we expect. 

 The New NickB 24 Jun 2021
In reply to Bottom Clinger:

> Nationally: 82.5% 1st, 60.3% 2nd

> Bolton: 75.2 and 54.6%

If you look at the regional averages, not a single region is a high as the National average. It’s because local and regional numbers are based on a different dataset to estimate adult population than the national numbers. The national numbers slightly underestimate populations, whereas locally and regionally they tend to overestimate population, particularly in areas with higher education institutions. 

It is covered in this episode of More or Less https://www.bbc.co.uk/programmes/p09kxjnc

Post edited at 16:35
 Bottom Clinger 24 Jun 2021
In reply to The New NickB:

Yes,  I know that and it wasn’t a good comparison. However,  neighbouring Wigan has rates of 79.6 and 58.9, not massively better but significant especially given Bolton’s extra push. I think the way cases are coming down in Bolton has to show infection acquired immunity kicking in (they’ve had high rates for large periods of time). 

 Si dH 24 Jun 2021
In reply to Bottom Clinger:

> Yes,  I know that and it wasn’t a good comparison. However,  neighbouring Wigan has rates of 79.6 and 58.9, not massively better but significant especially given Bolton’s extra push. I think the way cases are coming down in Bolton has to show infection acquired immunity kicking in (they’ve had high rates for large periods of time). 

I agree - at least in the sub populations who are most exposed to the virus. It's a good thing. There's a big question though as to what happens there when you open up completely because more people will fall into the highly exposed group.

 The New NickB 24 Jun 2021
In reply to Roadrunner6:

> No it’s not. The US is increasingly splitting on Red blue lines in their response. 

> nationally means little in such a massive country managing covid at the state level.

I totally get that responses and outcomes are different, state by state. At an individual level, the state response is going to be more important to you than the national average response, more so than say the Greater Manchester response for myself, I totally get that. However, that does not change the reality of very low levels of testing. When cases are around a sixth per capita compared to the UK, but deaths are around five times as high per capita. Something does not compute. 

 Misha 24 Jun 2021
In reply to Ramblin dave:

Yes you're right - was thinking about people cutting down on socialising indoors but that's not much use if you work in a pub in the first place. I guess people can opt to wear masks and I hope employers will not put in place policies requiring employees NOT to wear masks.

Your other point is precisely what I'm wary of - feeding to the wolves anyone who hasn't yet had two jabs (even if they want to get vaccinated). It just doesn't feel right and it won't be good for the NHS either, albeit we need updated modelling to properly understand the healthcare impact.

 Misha 24 Jun 2021
In reply to Bottom Clinger:

> Bolton is worth a look. Cases have been  dropping for the past 4 weeks yet they still have below average vaccine take up (even with the extra ‘push’). Hospitalisations and people on ventilation appear to be plateauing. Deaths do seem to have gone up though, but seems reasonable to assume they will start coming down fairly soon. So even with current  restrictions there does seem to be some slack.

Cases in the worst hit MSOAs have indeed started to come back down without local hospitals being overwhelmed as far as I know. However that's with current restrictions + local messaging + surge testing etc. Take away all of that and what would happen? That's the big question.

> BTW, I think the current Delta wave originated in adults coming back from India, but quickly spread into younger age groups. 

Good point.

> The idea that most of us can do most things apart from ‘the nice extras’ doesn’t wash with me: most folk I know really are sick to the back teeth of not being able to do normal stuff

I get that this is an issue for some people (I don't know about 'most') but we also have to bear in mind the risk of having to have restrictions reimposed and the fact that NHS staff deserve a break - I bet they're sick to the back teeth of dealing with Covid cases and now massive waiting lists.

> The govt has promised all along ‘vaccines are our way out’ so at some point they need to put their money wher their mouth is.

I don't disagree but we aren't there yet and won't be there till around September. The vaccination program is far from complete.

> I guess/hope in all the modelling then the Human Factor is factored in - not sure how much more of this people can take and they certainly won’t police things.

People have been saying this since March last year... 

> My concern is ‘will enough people in the Usual Suspect areas get jabbed?’ I doubt they will by Freedom Day 2, and my gut feeling is that the number of un-jabbed and no second jab could be large in these areas.  

And indeed in many other areas.


New Topic
This topic has been archived, and won't accept reply postings.
Loading Notifications...