One-*** efficacy questions

[D_W]

Absence of evidence is not evidence of absence. The whole point is, nobody knows what mutations will happen (other than that we know they will) and what the effect of those mutations might be - Covid came as a surprise to governments all round the world, after all. We also know that vaccines exert an 'evolutionary pressure' on viruses. Yes, we're all better prepared than we were a year ago, but variants can/could still cause us all a lot of trouble (I don't need 'evidence' to believe that!).

No evidence in a situation where evidence is very easy to observe. No concern. Vaccines are more effective than their initial claims. No variant evades moderna or Pfizer, despite the fact that even if one did, it would be easily controlled by a booster.

No evidence, no evidence of any need for fearmongering or unscientific policy making.

 

[Jonm]

'

No, I don't think we do.

This is a novel virus, decision making has to be done concurrently with data collection.

The difficulty with highly transmissible viruses is infection rates soar exponentially....slow decision making will always be too late.

I really do not follow you. I think that perhaps you need to re-read what I said.

I am talking about opening up the economy with rules based on the assumption that vaccines reduce transmission. As you say “The difficulty with highly transmissible viruses is infection rates soar exponentially”. We are talking about greatly reduced social distancing in confined spaces as in aeroplanes. The assumption being that those who have been vaccinated will not pass it on to those who have tested negative. As I said, seems a bit iffy to me. For all we know the vaccine may turn some people in to asymptomatic carriers.

As yet I have not heard anything from Whitty or Van Tam on transmission and vaccines, I tend trust them, as opposed to the buffoon in the middle.

 

[Jonm]

[
Exactly, so the problem isn't the slum dwellers themselves it's the failure if agencies worldwide to alleviate slum dwelling and all the associated ills.

Good, so we now agree that mutations originating in the one billion plus slum dwellers pose a risk to us.

 

[D_W]

100 positive tests out of 1.2 million vaccinated in Washington state. If coronavirus doesn't show up on a test, someone won't have enough viral load to pass it on.

I'd bet all 100 of those cases were transmitted from non vaccinated folks. Think about what that suggests. When most of an area is vaccinated, it disappears completely.

 

[TominDales]

Some may find this very interesting

This clip seems very dated in April 2021. Science works by people publishing their findings based on experimentation, they may be trying to prove or disprove a hypothesis and they do it by publishing so that other scientist can check their work and validate it or critique it. As time goes by a consensus emerges which we take to be a scientific truth. with Covid the consensus of known truths have been emerging over the past year, however at any one point in time the latest ideas are still being 'debated' and checked.

Bhakdi is arguing in this clip two points of view that have been proven over the past 9 months to be wrong.
Firstly the claim that the virus posed no more threat than influenza has not been born out by events. The virus has a higher transmissibility than Influenza A and a higher case fatality rate CFR (death rate). Even when he said this last August the data was showing R0 twice that of typical influenza and a CFR significantly higher. He also seemed to suggest that as the disease did not effect the young then life should go on as normal, this overlooked the possibility that the old would fill up the worlds hospitals and bring normal healthcare to a standstill, also that even the young don't like to see their grandparents dying in droves from a preventable event. As it turns out the disease is affecting a considerable number of the young as new variants emerge.
He also claimed that any Covid -19 vaccine would be pointless for two reasons, the virus will mutate and also our immune system will unlearn the virus. That was a very strong concern for the medical community 12 months ago. Many diseases do not yet have a vaccine and the profession worried that Covid19 may have been one of them. However trying to make a vaccine was not pointless, he was wrong in that assumption. That does not mean he is a bad scientist for expressing a contrary opinion, its important that these opinions get tested, but we have to be careful not to take on trust what random scientist say. Their arguments have to stack up. He should have known better as by last August early trials were showing promise, so he was either not following events or had got carried away with his views or his ego. The best scientist have to be humble and follow the data and facts as they emerge, he seems to have forgotten to do that. His old employer, the university of Mainz has distanced themselves from his views. Sad as a distinguished career and reputation has ended.

 

[TominDales]

Its confusing, my interpretation of these figures is;
If you are aged 80+ then the chance of hospitalization has fallen to 20% of what it was ie 80% reduction. and deaths have fallen by 85% from what they were. Given the the Case fatality rate (CFR) for 80+ bracket was 56% last summer ( I think its much lower now due to better treatment) then I suppose the CFM for 80+ years olds is now about 7.5% (15% of 50%) and for the average male its dropped from 4% to 0.6%. its very low people below the age of 40 in any case. These figures seem conservative as I'm read elsewhere that fatalities are virtually zero after 4 weeks, as recent deaths seem to be in people who caught covid within two weeks of the vaccine. There have been studies of care homes that are encouraging. I'll see if I can find them - just need to get a bit more gardening done tonight or I'll get shouted at ....

Chris - I've done a bit of google this lunchtime, here are two articles that we can probably trust. Looking at the raw data, its clearn that we should be multiplying the benefits of vaccination in reducing infection and then in reducing hospitalization or deaths as both the studies measure improvements in death rate and hospitalisation on those already infected. So in summary. The vaccines reduce the chance of hospitalision for the over 70s to 2.5% vs those not vaccinated and deaths in the 80+ aged group to 0.7% compared to those not vaccinated.

The first is government blog from 21 February - note its quite old, but it supports the Johnson/Whitty numbers. COVID-19: analysing first vaccine effectiveness in the UK - Public health matters
But intriguingly ends by quoting a public heath Scotland report showing vaccine effectiveness of 85% for Pfizer and 94% for AZ and that recent PHE data will be published to support the Scottish numbers - not seen it published yet.

The second article is a published report from PHE - I think this is where BJ and Whitty get there numbers: https://assets.publishing.service.g...e/971017/SP_PH__VE_report_20210317_CC_JLB.pdf this is quite comprehensive, probably using data gathered in January and February in care homes and the community so probably from data gathered 6 weeks ago. Both of these studies will be showing vaccination effectiveness against a mix of variants that were circulating at the time, the dominant one being the Kent variant estimated to be 96% of cases by the beginning of February.

If we look at the PHE study and do some maths.

First set of table are effectiveness of vaccine against infection as indicated by PCR tests average of 72% in 70+ years olds ie 28% got covid (70+ olds were the ones vaccinated at that time). Its safe to assume younger people will show a greater resistance to covid than this.

Of those that got covid 8.5% were hospitalised, so of those vaccinated 28%* 8.5% that is 2.5% went to hospital with COVID compared to 15% of unvaccinated group.
In numerical terms its was 9000 cases of unvaccinated people led to 1361 hospitalisations, wheres for the vaccinated only 2000 got covid and 172 were hospitalised. Again its safe to assume the younger age groups would be more protected.

Note the different numbers used in headlines, this is where statistics is hard for the lay person to comprehend from newspaper headlines. The vaccine reduced hospitalisations of those infected in this age group by about 40% BUT that is the IMPORTANT BUT, they were already less likely to be infected so you have to multiply the two benefits together to get the overall benefit, ie only about 2% of vaccinated group (9000 unvaccinated got covid cf 172 vaccinated ended up in hospital) vs to 15%, 2% vs 15% is an improvement of 8 fold. In absolute numbers: 172 went to hospital of those vaccinated but 1361 unvaccinated went to hospital. The statisticians don't do what I have just done as you cant be sure that 9000 people got exposed to Covid in both gourp of people - the vaccined vs the unvaccinated . But the overall effect is much more pronounced that the Johnson numbers indicated.

Death rates (CFR):
Here the data is for the over 80s (not comparing like with like as it was the over 70s in the previous data set, as it was the over 80s who had the vaccine and survived or died within 28 days whereas the over 70s hasn't had the vaccine for 28 days during the period of the study so this is very much a worse case scenario as younger aged groups are less likely to die,

Here the overall conclusion is it reduce deaths by 54% of those who caught covid. 13% died unvaccinated vs 6.6% vaccinated an improvement of 54% BUT again that is on a number that is already reduced by the vaccine so the true ratio is 8625 unvaccinated got covid of whom 1115 died whereas only 914 vaccinated got covid of whom 60 died. to my mind that is a 60 out of 8625 (who were exposed to coved) who died or 0.7% - again not statistically a valid thing to do but it kind of give the real world numbers. If you bear this in mind for younger aged groups this is very reassuring as the data is almost certainly for the Kent variant of covid.
Incidentally the Scottish PH number of 94% mentioned in the first article - the blog - may be explained by the data in the second article.
In this study 1115 unvaccinated died compared with 60 vaccinated that is only 6% (5.3% precisely) of vaccinated died ie the vaccine reduced deaths by 94%

 

[D_W]

https://www.doh.wa.gov/Newsroom/Art...ne-breakthrough-confirmed-in-Washington-state
I may have mentioned the washington state data here. Note the size of the group. Of course, there were probably more than 102 cases, but how many? They wouldn't have been hospitalized folks or even moderate - those folks would've been caught when they went to the doctor.

So, looking at the group as is- 1.2 million vaccinated people. A wild guess at the average vaccinated (effective duration), but let's assume it's 1 1/2 monhts.

two deaths. Both with underlying conditions and over 80.

Can you imagine how many people over age 80 with underlying conditions have been exposed to covid and haven't so much as even gotten a bit of infection at all? It has to be thousands and thousands.

over 2 months - 8 people have been hospitalized in that group (2/1 through the article date).

8

How many absolute health basket cases do you think there must be in that group? Certainly there are gobs of people on chemotherapy or with immune deficiencies.

I talked to a transplant patient this week - she emerged unscathed due to due care avoiding the virus and now she is vaccinated. She will go this week to confirm that she has antibodies (to make sure that her suppressed immune system still had some kind of protective response to the vaccine. It's likely that she will be fine, but she has to get confirmation - that seems sensible to me).

There is not enough thankfulness for this in my opinion. When I brought this up each time someone told me to wash my hands (that itself is hogwash - it has been well known from the start, but we're all still choking on lysol and novel sterilizing products - the CDC now estimates perhaps 1 in 10,000 covid cases has been contracted due to surface contact, and the same people who exhort you to wash your hands constantly think sitting in a "reduced capacity" restaurant is a reasonable measure.

Just griping as above- if we're going to follow science, please let's do it all the time and be thankful when it works. Not suspicious because it does.

 

[TominDales]

https://www.doh.wa.gov/Newsroom/Art...ne-breakthrough-confirmed-in-Washington-state
I may have mentioned the washington state data here. Note the size of the group. Of course, there were probably more than 102 cases, but how many? They wouldn't have been hospitalized folks or even moderate - those folks would've been caught when they went to the doctor.

So, looking at the group as is- 1.2 million vaccinated people. A wild guess at the average vaccinated (effective duration), but let's assume it's 1 1/2 monhts.

two deaths. Both with underlying conditions and over 80.

Can you imagine how many people over age 80 with underlying conditions have been exposed to covid and haven't so much as even gotten a bit of infection at all? It has to be thousands and thousands.

over 2 months - 8 people have been hospitalized in that group (2/1 through the article date).

Whats interesting about the DOH article, is how its written in plane English, with jargon explained to the layperson ie what breakthrough cases are etc. The PHE paper is much more full of scientific jargon and statistical language which makes it far less readable. I've often noticed how North American scientists usually explain their subject in more accessible ways than we do in the UK.

The two articles are not directly comparable, so expect different numbers between the two even thought the vaccines are likely to be equally effective. There are two big differences in analysis between the two studies, the DOH is taking a wider age cohort of people, they haven't said exactly what the age distribution is, whereas the UK one is aged 70+, secondly the DOH study is on people 2 weeks after the second dose of vaccine - fully vaccinated whereas the PHE report I sighted is 2 weeks after the first dose - I was directly answering Chris's question regarding the first dose. Looking at the DOH figures, I'd expect the UK to have similar results to them given what the PHE study is showing after 1 dose. This is a very high degree of protection. Another difference that may be significant is the PHE study is almost certainly dominated by the Kent variant of covid, the Washington cohort may not yet have experienced the Kent variant to such an extent at that time. I gather this variant is pretty widespread now in the USA unfortunately.

 

[D_W]

The washington state cohort would probably be mostly front line workers, plus nursing home residents, plus people with significant risk factors. Not sure about other states, but in my state, the whole general population will be eligible 4/19.

I did see an article that the kent variant has been in WA state since at least January of this year, so there's some chance that it's the dominant strain. If there is any resistance to the vaccines for any of the strains, it's not enough for them to not control them at this point (though the news outlets try every day to come up with a doomsday story of some sort - this mornings was "double mutation!!!" with lots of coulds, woulds and shoulds.

The relatively plain language is often a legislative requirement for releases here. It's nice, because you can see what you're looking at before deciding whether or not you want to read technical documents.

 

[Jonm]

But intriguingly ends by quoting a public heath Scotland report showing vaccine effectiveness of 85% for Pfizer and 94% for AZ and that recent PHE data will be published to support the Scottish numbers - not seen it published yet.

Thankyou for such informed comments. I attach a link to the BMJ article which I had found very re-assuring.
https://www.bmj.com/content/372/bmj.n523
In the “responses” section there is one comment by
23 February 2021
Graeme J Ackland
Professor
University of Edinburgh
Edinburgh

He said “...................the risk of hospitalisation from covid-19 fell by up to 85% (95% confidence interval 76 to 91) and 94% (95% CI 73 to 99), respectively." but you don't mention that after three weeks, AND after five weeks, the effectiveness is significantly lower.
So, while your article is technically correct, it is quite misleading.”

I had expected the bmj article to be unbiased and objective but this professor has questioned that. Looks like he is a physicist/engineer/computer guy so I would expect him to be good at statistics.

 

[TominDales]

Thankyou for such informed comments. I attach a link to the BMJ article which I had found very re-assuring.
https://www.bmj.com/content/372/bmj.n523
In the “responses” section there is one comment by
23 February 2021
Graeme J Ackland

He said “...................the risk of hospitalisation from covid-19 fell by up to 85% (95% confidence interval 76 to 91) and 94% (95% CI 73 to 99), respectively." but you don't mention that after three weeks, AND after five weeks, the effectiveness is significantly lower.
So, while your article is technically correct, it is quite misleading.”

I had expected the bmj article to be unbiased and objective but this professor has questioned that. Looks like he is a physicist/engineer/computer guy so I would expect him to be good at statistics.

Thanks, the stats are not layed out in a way that is digestable to the public.

I think the issue of waning immunity was properly mentioned on BMJ artielce of 22nd Feb, the problem is not enough data to looks at it at that time. Prof Ackland is right to say that is an area to be investigated but his comments seem a bit harsh given that the lady did address this issue in the article and press conference - its says - half way down the page:

''Only four weeks of follow-up data were available for the Oxford AstraZeneca vaccine, but the longer period for the Pfizer BioNTech vaccine showed that that the impact on hospital admissions lessened slightly five and six weeks after vaccine administration, raising the question of whether this decline might be related to waning immunity.

“We haven’t specifically looked at waning immunity,” said Chris Robertson, professor of public health epidemiology at University of Strathclyde, and statistics lead for the health protection group at Public Health Scotland, pointing out that these were preliminary data. “It’s an important point, but we can’t, at the minute, say anything about that,” he added. “With our further analyses, when we’ve got a longer follow up then we will be able to provide more information.”

Clearly time is needed to see how things unfold, from what I've seen of more recent publications, they tend to be pointing in the right direction. Israel is a bit further ahead of the UK and their economy is unlocking successfully without a resurgence. On the other hand Chile also had a rapid vaccine roll out - Pfizer and Sinovac, but their unlocking was - in hind sight, rushed, and they got a resurgence of cases and have had to lock down again. i can see why Whitty et all are urging caution. But on the whole new data is looking more positive.

 

[Chris152]

'the impact on hospital admissions lessened slightly five and six weeks after vaccine administration, raising the question of whether this decline might be related to waning immunity' - if that's the case, it would suggest a really short peak in first-dose effectiveness, wouldn't it? Second doses are really ramping up on the daily rate, good news.

 

[Jonm]

Thank you for that response. I find getting at the figures for reduction in hospitalisation and deaths is really difficult. Looking at numbers of hospitalisations we need to know the numbers vaccinated Vs unvaccinated and exposure to the virus.

We have a situation in the older age groups where 95% have had the vaccine. It is possible that in the future we will have more or similar hospitalisation and deaths in the vaccinated groups than the unvaccinated. I can just see the posts on social media “more people die who have been vaccinated than unvaccinated”.

I have had the vaccine and am following the rules. As we open up I will be maintaining social distancing and continue to be careful. No aeroplanes, buses, trains etc in 2021. Will go to cafes/pubs and sit outside, inside perhaps occasionally if it has good social distancing. May change this if the data shows it is OK. But that is easy for me, not so easy for others.

 

[Spectric]

Absence of evidence is not evidence of absence. The whole point is, nobody knows what mutations will happen (other than that we know they will) and what the effect of those mutations might be - Covid came as a surprise to governments all round the world, after all. We also know that vaccines exert an 'evolutionary pressure' on viruses. Yes, we're all better prepared than we were a year ago, but variants can/could still cause us all a lot of trouble (I don't need 'evidence' to believe that!).

That is exactly how real safety is analysed when looking at potential hazzards and the risk factors, you either have conclusive data that supports the level of the hazzard and the risk of it happening in order to categorise it otherwise it has to be treated as potentially high until such times as you do. So we have to accept variants will happen because that is what viruses do and common sense tells us that without needing data which will come later.

 

[Chris152]

Interesting in the FT on second doses is the decision in France and Germany to recommend that younger people who have had a first dose of the O/AZ vaccine be given a different *** for their follow-up shot.

 

[TRITON]

I think that everyone who caught the virus, was hospitalized by it or came to suffer from long covid couldnt get a rats testicle for what the statistics say.

The statistics say using unguarded machinery is bloody dangerous, and yet...

 

[TominDales]

'the impact on hospital admissions lessened slightly five and six weeks after vaccine administration, raising the question of whether this decline might be related to waning immunity' - if that's the case, it would suggest a really short peak in first-dose effectiveness, wouldn't it? Second doses are really ramping up on the daily rate, good news.

There is very low statistical confidence in the six week observation, so I think its best not to draw any conclusions until more data has emerged. So far the 1st and 2nd dose data it looking to be better than earlier data.

 

[Jonm]

Interesting in the FT on second doses is the decision in France and Germany to recommend that younger people who have had a first dose of the O/AZ vaccine be given a different *** for their follow-up shot.

There is a lot of politics in this.
Phizer vaccine, German.
AZ at start, Germany said not enough evidence it protects elderly, only give to young. Macron said quasi effective.
EU behind in vaccine rollout due to delay and commit to pay for vaccines at an early stage so got a “best endeavours” contract with AZ. Lots of criticism of EU procurement so leaders fling some mud and spread the blame.
Germany then finds AZ link to blood clots so stops use then changes to only for elderly. France does similar.

The problem here is that the EU countries are now having to deal with the Kent mutation, cases and deaths are rising. I can see the logic of not giving AZ to people vulnerable to these very rare clots but the way Germany and France are behaving is undermining the confidence people have in the AZ vaccine.

Politician logic, let us undermine confidence in AZ so public will think that lack of doses is no big loss and not blame us. Also, good for German business!

On the news there was a French woman being wheeled in to hospital with Covid, she had refused AZ vaccine and was waiting for Phizer, she still thought she had made the correct decision.

The important point is that AZ is cheap and can be stored at normal fridge temperatures. It is therefore suitable for use in third world countries. Phizer is five times the price to buy and has to be kept at -70 deg C, transported in special liquid nitrogen filled flasks in batches of 1000 doses. Not easy to transport and administer in areas with dirt roads, high heat, intermittent electricity etc. A local hospital here in uk lost 1000 doses due to mis handling.

It would be a disaster if AZ vaccine is not rolled out worldwide due to these continuous attacks by some EU Leaders. Vaccines like Phizer will not get to many places in the world which AZ can reach.

I have had first dose of the AZ vaccine and will take the second dose immediately I am offered it.

 

[TominDales]

Interesting in the FT on second doses is the decision in France and Germany to recommend that younger people who have had a first dose of the O/AZ vaccine be given a different *** for their follow-up shot.

That's interesting as its a big departure from data driven work and clearly a reaction in France to the high vaccine skepticism. On the whole France has been very cautious about rolling out the vaccine. France is the most vaccine sceptic country in the EU if not the world. Here is an article explaining the reasons.
https://www.cnbc.com/2021/01/13/fra...e-most-vaccine-skeptical-nation-on-earth.html, France had a number of medical scandals in the 1970s/80 the worst being the blood scandal where France delayed changing its blood transfusion regime and prescribing of drugs fro HIV, as it wanted to develop an all French test and drug response. That scandal caused a massive loss of trust in French public heath. Also there have been other scandals where the French elite enacted policies that served them and not the citizens. So there is even less trust in the political and professional classes in France than in the UK and most developed countries.
We had our blood transfusion scandal, but that was due to ignorance, a bit of incompetence and a slowness on behalf of the NHS to respond to the data, it was not due to any deliberate scheme as in France.
I don't know Germany so well, but suspect they too have some vaccine hesitancy. The UK is up there on the vaccine positive side, largely because however under-resourced or inefficient the NHS is, we trust it and its staff and they tend to build that trust by being open with their data.

 

[TominDales]

I think that everyone who caught the virus, was hospitalized by it or came to suffer from long covid couldnt get a rats testicle for what the statistics say.

The statistics say using unguarded machinery is bloody dangerous, and yet...

That's true for that cohort, but the purpose of releasing good data in a timely manor that is statistically robust is all about building and maintaining public confidence our public heath system. Having independent doctors, academics etc access to proper studies and publishing the results - good or bad, although inconvenient in the short term is the best way to maintaining public trust.

The AZ vaccine in holding up well in the Uk for a number of reasons, not least a very consistent message or cautious confidence from out leaders, and timely open and transparent updates and measured reactions to new information. That is where statistics plays an important role.

You make a good case in pointing out that we should not get carried away with every data point. Its important that we follow this story, we check the those in charge are making logical and self consistent statements, its that that keeps our confidence in their decisions. But reacting to every development on a long evolving story is, as implied in your comment, a bit OTT.