Covid-19 Vaccines

Pfizer vaccine success: better than we could have hoped for?

Early this morning USA time Pfizer, together with BioNtech, announced in a press briefing that their vaccine had successfully reduced the incidence of Covid-19 infection by 90%.  The results were announced in a press briefing though are yet to be published.  Does this mean that this (and hopefully other vaccines) are going to be the success we hoped for?

What was the study?

  • Like many other companies developing a vaccine, Pfizer started a major clinical trial of 43,538  participants, randomly divided into half that received the active vaccine and half that received a placebo
  • Participants from all ages from 12-85 were studied, the aim was that 45% would be over age 45
  • As far as I can tell none of the participants had a previous clinical history of Covid-19 infection and all were free of symptoms at the start
  • They had two doses of the vaccine, 21 days apart
  • The plan was that they would be followed up for up to 24 months to assess the risk of contracting Covid-19 and having side effects
  • The  researchers in the team who were collecting the follow up data did not know which group any participant was in, to avoid bias
  • Any of the participants who developed any symptoms suggestive of Covid-19 were then given a detailed assessment
  • Anyone who developed evidence of Covid-19 before 7 days had elapsed after the second dose were ignored
The detailed 146 page protocol for the study

What was the rationale for giving the results today?

  • Like any such trial, the researchers were only going to know if the vaccine was successful if the numbers of cases of infection were lower in the vaccine treated group compared to the placebo group
  • What had been agreed, based on statistical considerations, was that the first look at the results would take place after a total of 94 cases of Covid-19 had been documented across both groups of the participants combined.
  • A that stage, the code would be broken for those cases and an independent group would then be provided with the information as to how many of the cases had had the active vaccine and how many had the placebo.  
  • Prior to this analysis no-one in Pfizer would have had any idea how the numbers would turn out!

What did the result show?

  • Pfizer have not release the actual numbers but did say that the vaccine was “above 90%” effective
  • My simple maths would suggest that what this means is that probably 86 of the cases occurred in the placebo group and only 8 in the active vaccine group
  • There were no serious safety concerns reported by the independent data monitoring group
  • Thus, the headline result is of a substantial successful reduction in risk, which begins 28 days after the first vaccine dose

What is still to be found out?

  • It will take a bit longer to find out if there were any safety concerns that were not considered as ‘serious’
  • Normally a ‘serious’ side effect means one that requires admission to hospital
  • There are too few cases to know if the vaccine was successful in all age groups and in other sub-groups of the population, for example by ethnic background
  • We can only know the rate of infection up to the period of time the participants have been followed up.
  • We need to know how long the protection will last for, as immunity might wane over time
  • The virus may also change over time (mutate)
  • Thus as more information accumulates over the longer term follow up, these headline figures might change.
  • These preliminary findings will be being written up for a scientific journal and be scrutinised for their accuracy.  I have however read the detailed protocol for this study and I fully expect that the announcement in the press release will prove to be a robust conclusion of the results 

What happens now?

  • No vaccine can be administered in a public vaccination campaign without it being licensed by an appropriate national body
  • The Food and Drug administration in the USA will thus look at these results, probably as an emergency this week, and are likely to grant approval on the basis that the benefits considerably outweigh the risks
  • I trust that Pfizer and their partners will then, or simultaneously, ask the European Medicines Agency and (sadly since BREXIT) the UK’s MHRA to similarly grant a license
  • Once a license has been issued, public health  authorities are then free to begin mass vaccination of their target populations
  • There are only limited stocks of this vaccine worldwide.  I assume that most will be in the USA but UK and other countries will have some stocks I am sure

Was this result surprising?

  • No one could have predicted what the result would be
  • We knew that this and other vaccines in widespread trials did produce antibodies and that these antibodies could, in the laboratory, neutralise the virus
  • We did not know if this would translate into either:
    • A reduction in catching infection
    • A reduction in any infection being serious
  • Experts were hoping for say a reduction in infection rates of between 60-80%
  • Thus, these results are at the very optimistic end of the guesses

What about other vaccines in development?

  • The good news is that most other vaccines in development are attacking the same part of the Covid-19 virus: the ‘spikey bit’
  • The other vaccines have also shown that, like the Pfizer vaccine, they can produce antibodies that can neutralise the virus
  • There are differences in the way that these vaccines work and it is not certain that they will all be able to reproduce the results from the Pfizer vaccine 
  • But expect results from these vaccines to come out fairly rapidly as especially governments will want to be able to have access to sufficient quantities of any successful vaccine.

A quick comment on the anti-vacc lobby

  • I will be posting later this week what are the issues and challenges faced as a consequence of the anti-vaccination lobby
  • The Editor of the Lancet posted this weekend his concerns about fake news upending the success of the vaccine programme
  • Remember that there are 2 ways individuals are protected by a vaccine:
    • The vaccine that they receive
    • The herd immunity that is provided if enough other people are vaccinated 

If you would like to receive email notifications of new posts on this blog just click this link and enter your email at the bottom of the home page:

10 replies on “Pfizer vaccine success: better than we could have hoped for?”

Very Interesting – thanks, Alan. On the question of where initial supplies are going, Boris Johnson said in his briefing this evening that 30, (or was it 40), million doses had already been ordered for the UK – are we likely to lose out to the USA in a bidding war and not see them in a timely manner?


Dear Alan, Thank you for very helpful and insightful Covid blog. My concern with the Pfizer data (based on why you helpfully report) is that with only 94 cases amongst 43538 trial participants, the biggest uncertainty is that of exposure to virus (unless that was measured)…? With 99.8% likelihood of no virus exposure it seems very uncertain whether the two groups (placebo vs vaccine) may have experienced different virus exposure making the 90% benefit of the vaccine itself much more uncertain (since just a few per thousand more people arbitrarily exposed to virus in the placebo group would give the same outcome irrespective of vaccine efficacy)…?
I’d be interested in your comments on this please.
Many thanks


That’s an interesting point and indeed the rates of infection do seem quite low. Participants in these trials are often the more health aware and hence less exposed as you suggest. Plus we only have limited data on person time ie rates/1000 person months. Having said all that given the randomization there is no prior reason to believe a systematic bias. Statistically the probability of the difference being chance is quite small but again the stats have not been published and we don’t know what assumptions they made!


This is a variation on Levin’s question. The infection rate was unbelievably low (94 out of 44000 participants). It is known that Covid is far more serious in the elderly than in young people. Is it known whether the infection RATE is higher among the elderly? If it is, could this skew the trial which presumably saw participants at the lower end of the age scale?


The short answer is they haven’t released the data yet on age. I gave m6 estimates of infection rate in the placebo group and, remember the 94 cases were all symptomatic ones. It is possible their population was representative. The rate of infection is not higher in the elderly. With no social distancing there are no age differences in susceptibility, although with greater shielding in the elderly, actually infection rate is lower in the real world. The proportion of clinically detectable to asymptomatic cases is higher in the elderly


Of course, the big problem is that human challenge would not be an ethical way to test for efficacy. (I admit, I didn’t read the protocol so it is an assumption that they didn’t go down this route). So they have to test for antibody production and/or T-cell mediators. Tt won’t be until the bulk of the population has been vaccinated that we can see the true efficacy in vivo.


The Pfizer vaccine is not live but the genetic material (RNA) that when injected stimulates the body to produce antibodies to a key part of the virus (the spike). The Oxford Astra/Zeneca vaccine is a live but harmless to human virus


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s