Coronavirus outbreak

[mjr]

I can disprove it:

Absolutely not, so ignore all the stats in this thread, apart from this one.......

100% of the statisticians presenting endless data on this thread about the level required to achieve herd immunity are wrong.

So even me, who agreed with you (and the WHO) that it's not yet known, is wrong?

 

[mjr]

Is there an agreed % of people needed to have had this before immunity is reached?

Not yet.

 

[mjr]

I'm amazed why you all care so much?

The easing of lockdown will happen as and when Bozzer says so, in England anyway, whatever anyone on here thinks or the stats of various Internet sources is immaterial, why you're ars*d to quote/post all this is beyond me.

That's rather surprising U-turn from someone who asked "does it matter what rules are in place?" and suggested that it's for each person to make their own decision about what's OK and what's unwise to do. I think this discussion is basically loads of people trying to figure out what they think will be wise when and under what conditions. Surely you should be supporting them trying to figure it out for themselves?

 

[Deleted User]

What U turn?
I'll be out at the 1st opportunity, which is down to the Government, others can please themselves.

 

[roubaixtuesday]

Herd immunity calculation
'R' Number For an infectious disease, such as COVID-19, View attachment 578760 is the basic reproduction number of the disease: the average number of people an infected person goes on to infect, given that everyone in the population is susceptible to the disease with no restrictions. For COVID-19 B1.1.7 variant estimates vary: let's go for View attachment 578761 =3.
NPIs (restrictions on behaviour etc) can reduce the 'R' - so for any time and dominant variant R = Rt. NB This Rt (as I shall use it) does not account for the number of people who are no longer susceptible to the disease: that's accounted for by the other calculations qv.
Only a proportion of 1-1/Rt of the population need to be immune to achieve herd immunity.

Effectiveness If a vaccine has an effectiveness of x%, then this means that x% of people who receive it are immune: not susceptible to the disease.
But let's take care to be clear what we mean by effectiveness.
#1) Immune so the virus does not cause illness: of course this means the subject does not transmit virus.
#2) Immune from serious illness. Catches C19 but asymptomatic or slight illness but the subject CAN transmit virus (transmissability will vary with subject, and this matters - I've treated it as factored into the 80% figure below).
Aiui the quoted efficacy figures from the RCT (Ph3) trials were for #2: Pfizer 94% and 70% for Ox-Az. Since these trials far more data are available and aiui effectiveness averages 89% (#2 version).
For herd immunity considerations, the severity of the illness 'doesn't matter'. So (leap of faith) let's assume for 'type #1 immunity' (which takes those people out of the transmission chain) x = 80% effectiveness. 20% of those vaccinated can still catch C19 and if they do they will be able to transmit the virus (assumed still at Rt). Note that effectiveness will be less until 7 days after the second dose, but I'm ignoring that for now - by mid-summer 32M will have had their second dose.

To achieve herd immunity we therefore require that P% (the proportion) of people who are vaccinated times the effectiveness (x%) is at least 1-1/Rt of the population. In other words,
Substituting Rt = 2.5 (slight restrictions) gives P needs to be greater than 0.6/x
For a vaccine that is 80% effective (#1) with no restrictions we therefore need to vaccinate (first dose) at least 73% of the population (50M) to achieve herd immunity (see Note below).
I estimate we will achieve that (4M a week) by 21 June.

Recall at its maximum in UK since May, Rt was at 1.6 on 2 Oct (90% confidence, upper bound) - this was a time when a few areas were in Tier 3, but for most, restrictions were 'light'. We might assume that by July the restrictions will be lighter still.
If we used Rt = 2 then the numbers needed to be vaccinated for herd immunity is 42M (63%).

Note: There is uncertainty about several (all?) of the figures I've chosen above, but to make some sense I have tried to choose 'reasonable ones'. Choosing Rt = 2.5 is certainly pessimistic but allows for a more transmissible dominant variant. Choosing 80% effectiveness is optimistic (but as the data come in we'll be able to narrow the uncertainty on that).
Conclusion: With these figures, to get to herd immunity in the summer we may need to retain some restrictions or Rt will push the herd immunity percentage up to a figure we (UK) can't reach.
@matticus - think you need to review your use of the expression "exponential", let alone "pretty".

I think that's all very reasonable.

Couple of comments:

1) It's not clear to what extent COVID is seasonal, but most people seem to think at least somewhat. So a further wave may come in the Autumn rather than summer as a result.

2) The %immune figures are more challenging to reach if we don't vaccinate children, or not all of them (I don't know what plans are, trials ongoing, ethics unclear to me, what will takeup be?).

The modelling done by SPI-M seems to indicate it's very likely there will be a further wave at some point as a result of these, when all restrictions are lifted.

 

[Craig the cyclist]

I have posted nothing that disagrees with that, and have clarified it on a number of occasions to boot.

You're arguing with yourself here.

You did though. You then went on to clarify which is very helpful, but initially you said the 75% figure which appeared to be a FACT. Although you did clarify, and have done a couple of times since, after I picked you up on it.

Everyone needs to stop doing home analysis on this stuff and referencing bodies and publications as though those stats have actually come from those bodies or publications. The figures are produced officially come with a ton of background data which you don't have access to.

Look, all I am saying is please beware, one thing we all know is that a supposition on the internet can become a fact on Facebook in the blink of an eye, and out of context analysis does no-one any favours at all. I am not having a dig at anyone, just asking for some thought.

 

[roubaixtuesday]

You did though

We disagree on this. Let's drop it.

Everyone needs to stop doing home analysis on this stuff

I think people like @Ajax Bay working through back of the fag packet calculations is entirely laudable - it builds understanding. If they claim they know better than official bodies - sure, call them out - but nobody is doing that here. At least as far as I've seen.

I am not having a dig at anyone

Lol.

 

[classic33]

You did though. You then went on to clarify which is very helpful, but initially you said the 75% figure which appeared to be a FACT. Although you did clarify, and have done a couple of times since, after I picked you up on it.

Everyone needs to stop doing home analysis on this stuff and referencing bodies and publications as though those stats have actually come from those bodies or publications. The figures are produced officially come with a ton of background data which you don't have access to.

Look, all I am saying is please beware, one thing we all know is that a supposition on the internet can become a fact on Facebook in the blink of an eye, and out of context analysis does no-one any favours at all. I am not having a dig at anyone, just asking for some thought.

The second is reliant on the first for it's existence, is it not.

 

[Ajax Bay]

Fair play to you for persevering with this but is there a straightforward answer to this for us thicko's reading?
Is there an agreed % of people needed to have had this before immunity is reached?

When you say 'immunity is reached' I'll take it you mean 'herd immunity'.
That'll be a 'nobut'.
Will be at least 60%. Might be more than 100% (ie we can never get there). I've offered some figures in my post above.
And then there is the international dimension. I expect if the UK does achieve 'herd immunity' that the challenging task will be to somehow control importation of infection (in particular VoC) by a combo of restriction on travel and travellers, and T&T measures focused on outbreaks. Side note: For British Isles integrity we need to get a few million doses over to Ireland.

 

[Deleted User]

When you say 'immunity is reached' I'll take it you mean 'herd immunity'.
That'll be a 'nobut'.
Will be at least 60%. Might be more than 100% (ie we can never get there). I've offered some figures in my post above.
And then there is the international dimension. I expect if the UK does achieve 'herd immunity' that the challenging task will be to somehow control importation of infection (in particular VoC) by a combo of restriction on travel and travellers, and T&T measures focused on outbreaks. Side note: For British Isles integrity we need to get a few million doses over to Ireland.

It was herd, apologies and thanks.

 

[Ajax Bay]

Comments:
1) It's not clear to what extent COVID is seasonal, but most people seem to think at least somewhat. So a further wave may come in the Autumn rather than summer as a result.
2) The %immune figures are more challenging to reach if we don't vaccinate children, or not all of them (I don't know what plans are, trials ongoing, ethics unclear to me, what will takeup be?).
The modelling done by SPI-M seems to indicate it's very likely there will be a further wave at some point as a result of these, when all restrictions are lifted.

Agree (1) but I would spin the seasonal effect differently. There is some evidence that 'R' for C19 is less in the summer (here's hoping). It was between 0.7 and 0.9 all through June and July last year but we have the B1.1.7 variant now which is ?40% more transmissible. If it is then this depresses Rt (I acknowledge that there'll be upward pressure on Rt from progressive relaxation of restrictions). This would allow herd immunity a little earlier. We could expect prevalence to fall to levels similar to last summer (July/August).
By the autumn, when Rt seasonally increases, a high percentage of the population is vaccinated and herd immunity is maintained because of the reduced number susceptible.
When there is a third wave, it will be a wavelet. Because of high levels of vaccination in the more vulnerable (JCVI 1-9 ie over 50s plus plus - 95% plus take up and 90+% effectiveness #2), the numbers with severe (needing hospitalisation) illness will be much reduced (thank goodness) and the NHS will be able to continue its services including all the elective work without adverse C19 effect.
2) Pfizer can be used to vaccinate down to 16, aiui: the others' lower age limit is 18 at present. There are no data to offer evidence of safety or efficacy in ages below those. There are about 12.3M under 16s. If uptake averages 90+% in adults (surveys suggest this intent) the vaccination programme will reach 49M 'over 16' arms (by mid June on my maths). This may be enough for herd immunity (73% of total population).
Trials of all 3 vaccines currently in use in UK are ongoing: Pfizer started trials with teens down to 12 in October and Moderna started theirs in December. Trials with Ox-AZ have started on a small group (?300) of children in UK aged 6-17 to see if the AstraZeneca gives that age cohort a good level of immunity to Covid-19 (and checks for safety). Janssen(J&J) started trials on 12-17s last year and are set to continue that trial right down to tinies.
3) SPI-M-O supporting paper (11 Feb) to easing restrictions decision (announced 22 Feb) - the option chosen was modelled as the green line scenario (5a) (below)(shading shows uncertainty).

 

[roubaixtuesday]

Agree (1) but I would spin the seasonal effect differently. There is some evidence that 'R' for C19 is less in the summer (here's hoping). If it is then this depresses Rt (I acknowledge that there'll be upward pressure on Rt from progressive relaxation of restrictions). This would allow herd immunity a little earlier. We could expect prevalence to fall to levels similar to last summer (July/August).
By the autumn, when Rt seasonally increases, a high percentage of the population is vaccinated and herd immunity is maintained.
When there is a third wave, it will be a wavelet. Because of high levels of vaccination in the more vulnerable (JCVI 1-9 ie over 50s plus plus - 95% plus take up and 90+% effectiveness #2), the numbers with severe (needing hospitalisation) illness will be much reduced (thank goodness) and the NHS will be able to continue its services including all the elective work without adverse C19 effect.
2) Pfizer can be used to vaccinate down to 16, aiui: the others' lower age limit is 18 at present. There are no data to offer evidence of safety or efficacy in ages below those. There are about 12.3M under 16s. If uptake averages 90+% in adults (surveys suggest this intent) the vaccination programme will reach 49M 'over 16' arms (by mid June on my maths). This may be enough for herd immunity (73% of total population).
Trials of all 3 vaccines currently in use in UK are ongoing: Pfizer started trials with teens down to 12 in October and Moderna started theirs in December. Trials with Ox-AZ have started on a small group (?300) of children in UK aged 6-17 to see if the AstraZeneca gives that age cohort a good level of immunity to Covid-19 (and checks for safety). Janssen(J&J) started trials on 12-17s last year and are set to continue that trial right down to tinies.
3) SPI-M-O supporting paper (11 Feb) to easing restrictions decision (announced 22 Feb) - the option chosen was modelled as the green line scenario (5a) (below)(shading shows uncertainty).
View attachment 578805

I think that's all very fair; whether a wave or a wavelet is very hard to know, reflected in the huge uncertainty bounds on that plot.

Hope for a wavelet, plan for a wave would be my view.

 

[lazybloke]

<snip>
2) Pfizer can be used to vaccinate down to 16, aiui: the others' lower age limit is 18 at present. There are no data to offer evidence of safety or efficacy in ages below those. There are about 12.3M under 16s. If uptake averages 90+% in adults (surveys suggest this intent) the vaccination programme will reach 49M 'over 16' arms (by mid June on my maths). This may be enough for herd immunity (73% of total population).
<snip>

Herd immunity is presumably most effective when the unvaccinated are evenly dispersed within a population group.
But children congregate in large groups; maybe 1000 or 2000 in each school; 30 in close contact in each class (6 hours a day, 5 days a week).
Infection rates might be dropping now, but what will happen this autumn if children still haven't been vaccinated?

I guess masks and social distancing are here to stay.


And for anyone thinking "but children don't get ill with covid", surely best to get a jab into them (one clinical trials complete satisfactorily), so that they do at least have some protection against possible nasty future variants.

 

[Ajax Bay]

. . listen to The World Health Organisation?
"The proportion of the population that must be vaccinated against COVID-19 to begin inducing herd immunity is not known. The percentage of people who need to be immune in order to achieve herd immunity varies with each disease."

Expanded and actual quote: "The percentage of people who need to be immune in order to achieve herd immunity varies with each disease. For example, herd immunity against measles requires about 95% of a population to be vaccinated ['R' for measles is between 12 and 18 - mega-transmissible]. The remaining 5% will be protected by the fact that measles will not spread among those who are vaccinated. For polio, the threshold is about 80%. The proportion of the population that must be vaccinated against COVID-19 to begin inducing herd immunity is not known."
Whilst the upper bound for R0 estimates for the current dominant C19 variant in UK (B1.1.7) is 4.5, the difference with measles is that the two-dose MMR vaccine is 97% effective against measles (and 88% effective against mumps), preventing both catching the disease and transmitting it.

 

[mjr]

I think that's all very fair; whether a wave or a wavelet is very hard to know, reflected in the huge uncertainty bounds on that plot.

Hope for a wavelet, plan for a wave would be my view.

It is unlikely to be a wavelet because we will not start from zero and it won't go negative (resurrections?). The term is being misused.

We've had three waves already, too.