1 in 13 bn risk once vaccinated. Boris hates those odds.

Reliability of the modelling / how mamy excess deaths are due to COVID-19 / how many are infected / Infection Fatality Rate / etc.
fon
Posts: 1172
Joined: Tue Dec 22, 2020 12:47 pm

1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by fon »

John Campbell's been looking at new data from the US cdc

https://www.youtube.com/watch?v=dr7AO2NvalI

which indicates they've had 5,800 cases up to now where fully vaccinated people have caught covid19 (beyond 14 days after the last dose), these are called breakthrough infections.

This is 1 in 13275, thus the fully vaccinated population would be around 77m, which I confirmed at ourworldindata. 5,800 is a suspiciously round number, which makes me think. I'm just saying. It might be right I suppose.

Thus the vaccines they've used seem effective. Of the 5.8k break throughs, 396 wound up in hospital (we have to assume they were mostly elderly since that is who has had the vaccine) and 74 of those died, which is ~ 1 in a million fully vaccinated people. Since 77m/74 =~ 1m (near enough for govt work.)

So if we think of the IFR as ~ 1 in 400 (factor of 0.0025) ,then 1 in a million is much better, esp. when one considers the fully vaccinated cohort is by and large quite elderly against whom the 1:400 ifr may be optimistic, 1 in 400 being the general ifr, including old and young people, the ifr for older people is worse so 1 in a million is great.

Note: 1 in a million is the risk a fully vaccinated person would die if they had a breakthrough infection. But the chance of a fully vaccinated person having breakthrough infection is 1 in 13,275.So those odds multiply up by Bayes Theorem iirc.

Which means: A fully vaccinated person of indeterminate age (but probably old), but not presently suffering from covid19 has a 1 in 13,275,000,000, i.e. a one in 13 billion chance of getting covid19 and then dying from it (13.3k * 1m)

As a vulnerable person over 60 with comorbidities.I like those odds, so I'm glad I took the vaccine. good look all.

If those odds play out over here, there will be next to zero older people dying of covid19, so even if a wave comes, it won't be in the vaccinated group. The govt's excuses for totalitarian rules are wearing very thin. And that's why BJ made his little outburst of anti-science on Tuesday. He's run out of road, so he had to boost his lockdown project. Janet Daley's rumbled him like millions of others:

https://www.telegraph.co.uk/news/2021/0 ... onspiracy/

jmc
Posts: 390
Joined: Mon Sep 21, 2020 9:01 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by jmc »

fon wrote: Sat Apr 17, 2021 7:20 pm John Campbell's been looking at new data from the US cdc

https://www.youtube.com/watch?v=dr7AO2NvalI

which indicates they've had 5,800 cases up to now where fully vaccinated people have caught covid19 (beyond 14 days after the last dose), these are called breakthrough infections.
One slight problem. The mathematics of sparse sampling. The sample period used is so short that the confidence interval is so large as to make any conclusions meaningless.

When you have an air carried infectious respiratory disease with such a low R0, such a high asymptomatic rate, such a low active infection true identification rate, and a currently serious depressed population dispersal rate due to invasive measures, its going to take a lot longer than 10 to 12 weeks to draw any statistically meaningful conclusions. With all the above caveats its going to be more like 12 to 18 months minimum of robust data before the confidence interval gets close to a value where you can draw strongly defensible conclusions from the data.

A quick look at the published literature on the clinical trials other vaccine candidates over the decades will illustrate this point quite clearly. It take an extended period of time to build up a reliable data set from which solid conclusion can be draw as to short term and long term efficacy.

You might see trends. Which may or may not prevail in the long term. Thats as strong as it gets with such a weak data set. A trend. And given that these vaccines have a low probability of long term efficacy due to the similar pathology of human corona viruses to influenza viruses even a very strong positive vaccine effect is unlikely to last more than one year. But only the long term data will answer this question. But those log scale efficacy rate roll off curves from the clinical trials dont bode well. Everything points to SARs CoV 2 vaccines having similar efficacy rates and protection period to the annual flu shot.

At least everything apart from the Science By Press Release that is.

Splatt
Posts: 1498
Joined: Thu Oct 29, 2020 12:46 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by Splatt »

The US doesn't use AZ - it uses vaccines with substantially higher efficacy.

thinksaboutit
Posts: 529
Joined: Wed Jan 06, 2021 11:38 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by thinksaboutit »

Splatt wrote: Sun Apr 18, 2021 5:48 pm The US doesn't use AZ - it uses vaccines with substantially higher efficacy.
Yes... we know you hate the AZ vaccine!

Among your accusations are "sugar pill" and "Placebo"... Yet you've never shown the study with 0% efficacy.

thinksaboutit
Posts: 529
Joined: Wed Jan 06, 2021 11:38 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by thinksaboutit »

fon wrote: Sat Apr 17, 2021 7:20 pm John Campbell's been looking at new data from the US cdc

https://www.youtube.com/watch?v=dr7AO2NvalI

which indicates they've had 5,800 cases up to now where fully vaccinated people have caught covid19 (beyond 14 days after the last dose), these are called breakthrough infections.

This is 1 in 13275, thus the fully vaccinated population would be around 77m, which I confirmed at ourworldindata. 5,800 is a suspiciously round number, which makes me think. I'm just saying. It might be right I suppose.

Thus the vaccines they've used seem effective. Of the 5.8k break throughs, 396 wound up in hospital (we have to assume they were mostly elderly since that is who has had the vaccine) and 74 of those died, which is ~ 1 in a million fully vaccinated people. Since 77m/74 =~ 1m (near enough for govt work.)

So if we think of the IFR as ~ 1 in 400 (factor of 0.0025) ,then 1 in a million is much better, esp. when one considers the fully vaccinated cohort is by and large quite elderly against whom the 1:400 ifr may be optimistic, 1 in 400 being the general ifr, including old and young people, the ifr for older people is worse so 1 in a million is great.

Note: 1 in a million is the risk a fully vaccinated person would die if they had a breakthrough infection. But the chance of a fully vaccinated person having breakthrough infection is 1 in 13,275.So those odds multiply up by Bayes Theorem iirc.

Which means: A fully vaccinated person of indeterminate age (but probably old), but not presently suffering from covid19 has a 1 in 13,275,000,000, i.e. a one in 13 billion chance of getting covid19 and then dying from it (13.3k * 1m)

As a vulnerable person over 60 with comorbidities.I like those odds, so I'm glad I took the vaccine. good look all.

If those odds play out over here, there will be next to zero older people dying of covid19, so even if a wave comes, it won't be in the vaccinated group. The govt's excuses for totalitarian rules are wearing very thin. And that's why BJ made his little outburst of anti-science on Tuesday. He's run out of road, so he had to boost his lockdown project. Janet Daley's rumbled him like millions of others:

https://www.telegraph.co.uk/news/2021/0 ... onspiracy/
Something very wrong with the reasoning here....

You conclude 1 in 13.275 billion chance of a vaccinated person dying from covid, based on data where 74 out of 77Million actually died. (In the original data)

Something is off here, don't you think! By a factor 4 orders of magnitude.

So question note1 :, where you say

"Note: 1 in a million is the risk a fully vaccinated person would die if they had a breakthrough infection. "

This should be 1 in a million is the risk a fully vaccinated person would die. Full stop!!

Chance of a vaccinated person dying form a breakthrough infection is 74 /5800 ( it's right there in the original data)

fon
Posts: 1172
Joined: Tue Dec 22, 2020 12:47 pm

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by fon »

thinksaboutit wrote: Mon Apr 19, 2021 11:57 am
Something very wrong with the reasoning here....
Yes, it was That was nonsense. The chance of a fully vaccinated person catching covid19 is 5800 in 77m, or 1 in 13,275. And then having caught covid19, the chance of him dying is 74 in 5800, or 1 in 78.
I screwed up, it's not 1 in 13bn, it is 78*13275 = 1 in 1045453.

One in a million. Still good odds. Must have been tired!


[/quote]

Splatt
Posts: 1498
Joined: Thu Oct 29, 2020 12:46 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by Splatt »

thinksaboutit wrote: Mon Apr 19, 2021 11:43 am
Among your accusations are "sugar pill" and "Placebo"... Yet you've never shown the study with 0% efficacy.
The south African study showed 0% efficacy within huge confidence intervals. At most 10% but with huge intervals.
The SA variant outbreak in a AZ vaccinated London care home last week also shows it really isnt doing anything to stop infection or transmission.

The estimated against non variant in the UK by KCL, UCL, OX all put it at 63-65% for reduction of infection. Those are also the figures Imperial and Warwick use in their models provided to SAGE.

So the odds in the original post are miles off when you factor in that vs the far far more effective vaccines used in the US.

thinksaboutit
Posts: 529
Joined: Wed Jan 06, 2021 11:38 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by thinksaboutit »

Splatt wrote: Mon Apr 19, 2021 6:40 pm
thinksaboutit wrote: Mon Apr 19, 2021 11:43 am
Among your accusations are "sugar pill" and "Placebo"... Yet you've never shown the study with 0% efficacy.
The south African study showed 0% efficacy within huge confidence intervals. At most 10% but with huge intervals.
The SA variant outbreak in a AZ vaccinated London care home last week also shows it really isnt doing anything to stop infection or transmission.

The estimated against non variant in the UK by KCL, UCL, OX all put it at 63-65% for reduction of infection. Those are also the figures Imperial and Warwick use in their models provided to SAGE.

So the odds in the original post are miles off when you factor in that vs the far far more effective vaccines used in the US.
So 10% observed efficacy in the poor SA trial. Yes the CI, gives a wide range, but that doesn't give you the right to select the low end as a supporting "fact", to your assertion.

Anyway, your zero assertions have not been in relation to the SA variant. Just generalised nonsense.

In the mean time, real world evidence seems to be showing the AZ vaccine to be just as good.

Splatt
Posts: 1498
Joined: Thu Oct 29, 2020 12:46 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by Splatt »

thinksaboutit wrote: Tue Apr 20, 2021 9:03 am
So 10% observed efficacy in the poor SA trial. Yes the CI, gives a wide range, but that doesn't give you the right to select the low end as a supporting "fact", to your assertion.
Except i didnt. That was mid end.
The low end was MINUS 49%. The high end 30%.
In the mean time, real world evidence seems to be showing the AZ vaccine to be just as good.
Are you getting your facts from CBBC?

KCL have given 70%.
UCL study 65%
IC and Warwick are using 63 and 65% in their modelling.
Oxford themselves have given optimistic 70%
Or the Lancet preprint that found 50% PCR reduction ( https://papers.ssrn.com/sol3/papers.cfm ... id=3777268 )


That is against wild type not variants and compared to 90%+ for Pfizer and Moderna.

The SA trial showed 0% with huge CIs.
Other studies there have shown zero neutralisation using serum which backs that up.

Where on earth are you finding AZ is 90% against preventing all infection of covid? What is your source ?
Why are you ignoring everyone else that finds 63-70% maximum?

Absolutely nobody is claiming AZ has anywhere near the efficacy of Pfizer or Moderna in reducing infections.

thinksaboutit
Posts: 529
Joined: Wed Jan 06, 2021 11:38 am

Re: 1 in 13 bn risk once vaccinated. Boris hates those odds.

Post by thinksaboutit »

Splatt wrote: Tue Apr 20, 2021 4:44 pm
thinksaboutit wrote: Tue Apr 20, 2021 9:03 am
So 10% observed efficacy in the poor SA trial. Yes the CI, gives a wide range, but that doesn't give you the right to select the low end as a supporting "fact", to your assertion.
Except i didnt. That was mid end.
The low end was MINUS 49%. The high end 30%.
In the mean time, real world evidence seems to be showing the AZ vaccine to be just as good.
Are you getting your facts from CBBC?

KCL have given 70%.
UCL study 65%
IC and Warwick are using 63 and 65% in their modelling.
Oxford themselves have given optimistic 70%
Or the Lancet preprint that found 50% PCR reduction ( https://papers.ssrn.com/sol3/papers.cfm ... id=3777268 )


That is against wild type not variants and compared to 90%+ for Pfizer and Moderna.

The SA trial showed 0% with huge CIs.
Other studies there have shown zero neutralisation using serum which backs that up.

Where on earth are you finding AZ is 90% against preventing all infection of covid? What is your source ?
Why are you ignoring everyone else that finds 63-70% maximum?

Absolutely nobody is claiming AZ has anywhere near the efficacy of Pfizer or Moderna in reducing infections.
I don't need a source for 90%, since I have never made a claim about its efficacy.

I simply challenge your serious bias against the AZ vaccine.

You regularly write "placebo", "sugar pill" for AZ, but then quote figures which disprove it... massively.

None of the efficacy numbers you write are ZERO.

You should admit, to yourself at least, that you misrepresent the AZ vaccine on a regular basis.

Your own words prove demonstrate this.

Post Reply