select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to press release from the ZOE app on vaccine effectiveness over time

A press release from the ZOE COVID Study suggests that vaccine protection provided by two doses of the Pfizer-BioNTech and Oxford-AstraZeneca COVID-19 vaccines is waning in the first vaccinated.


Dr Peter English, Retired Consultant in Communicable Disease Control, Former Editor of Vaccines in Practice, Immediate past Chair of the BMA Public Health Medicine Committee, said:

“The press release is disappointing. As far as I can tell, it refers only to effectiveness against (any) infection. It doesn’t go into detail in the press release itself – you’d have to read and understand the papers it links to – but I assume “infection” means “confirmed infection”, ie people who report having tested positive. 

“There are various ways in which vaccines can or might be effective. They might prevent infection altogether (and thereby onward transmission); they can reduce the duration, and level, of infectiousness in people who, despite vaccination, become infected; and they can prevent minor symptomatic infection (cough/cold/headache), more serious flu-like illness not requiring hospitalisation, illness severe enough to require hospitalisation, illness severe enough to require critical care (ICU admission), and death.

“There is a world of difference between efficacy against, on the one hand, any infection and on the other hand, illness severe enough to require hospitalisation, critical care, or to cause death.

“I am disappointed that the press release failed to tease out these differences, to explain more clearly what it means by “infection risk reduction”, and to comment on effectiveness against different outcomes.” 


Dr Simon Clarke, Associate Professor in Cellular Microbiology, University of Reading, said:

“The ZOE app is a useful tool in monitoring people’s symptoms and that is what it relies on for its data, but on its own it cannot accurately monitor the number of infections in the population, as some have claimed.

“The app is dependent on a self-selecting cohort, so even though the sample size is large, it does not necessarily use a study group which is reflective of the UK population.

“Therefore these data are likely to be under-reporting infections as anyone with the virus but without symptoms, won’t be reporting this to the app.  It seems likely that the ZOE app would therefore lead to over-reporting of the protection given by the vaccines.

“It is also hard to compare infection rates between May and July in the UK and claim the difference is due to waning effects of vaccination, when wider community infections were more than eight times higher on 31 July (when there were 26144 new daily cases) than on 26 May (when there were 3067 new cases).

“The claim that immunity levels will hit around 50% by Christmas is not based on any robust analysis of data, and seems more like a finger in the air prediction. Immunity is a complex process and we cannot assume people’s immunity will fade at a uniform rate over time.

“However other, more robust data from other studies shows that while double vaccinated individuals are well protected against infection, and even better protected against serious disease, their level of immunity differs between individuals, and does dissipate over time. This is a reminder that we cannot rely on vaccines alone to prevent the spread of Covid. Lessons from countries like Israel, where the majority of the population were vaccinated early in 2021, shows that a new wave of infections, driven by new more infectious variants, can still drive up infection rates quickly.”


Dr Alexander Edwards, Associate Professor in Biomedical Technology, University of Reading, said:

“This study follows from the recent preprint reporting vaccine protection, but with slightly different methodology±, and many of the conclusions- and caveats and cautions about interpretation are the same. One major problem with this ‘real-world’ study is that many things are changing at the same time, making it a little harder to interpret the results than more formal clinical trials. For example, during the period studied, there may be non-random changes that affect infection rates- such as changes in the profile of people who are/aren’t vaccinated,  in exposure and levels of infection in the population, and to behaviours, which could all affect vaccine effectiveness. We can still hope that the apparent waning in protection won’t continue.

“Again, as noted before, we are mostly concerned about preventing severe disease, which may be more effectively prevented by vaccination than simply getting infected at all (without measuring how severe).

“However, there are very important conclusions from these studies- alongside similar reports from other parts of the world. Vaccination does not (unsurprisingly) make people invulnerable, and does not prevent all infections. Variants have real and significant impact on public health, and a lot of people are still tragically dying in the UK from this nasty virus. The vaccines we have are remarkably safe and effective, and still remain far better than other vaccines that give massive benefits. We must pro-actively plan our public health strategy to account for imperfect protection, and for the possibility of falling protection over time.”

± see

‘Impact of Delta on viral burden and vaccine effectiveness against new SARS-CoV-2 infections in the UK’ by Koen B. Pouwels et al. is a preprint of work carried out by the University of Oxford, ONS and DHSC.


Prof Paul Hunter, Professor in Medicine, The Norwich School of Medicine, University of East Anglia, said:

“This latest analysis from the ZOE app study provides additional and valuable information on the effectiveness of the vaccines currently in use in the UK. In general the reported findings from the ZOE study are consistent with those from other recent reports such as the ONS Oxford University study. These findings can be summarised as that the Pfizer vaccine offers improved effectiveness over the AZ vaccine and for both vaccines there is a demonstratable decline in effectives over a very few months. These observations are not surprising and were predictable from several months back

“One issue is how much of this decline is vaccine effectiveness is down to waning immunity and how much is due to the greater prevalence of the Delta variant which is known to be more resistant to prior immunity from vaccination.

“What is not clear from the ZOE press release is whether this decline in effectiveness is based on symptomatic illness or whether they have identified declining effectiveness against severe disease. As we have pointed out previously COVID can be thought of as two phases of infection The first phase is infection of the nose and throat (a mucosal infection) and we know that immunity to mucosal infections are rather short lived whether from vaccine or natural infection. Most cases of symptomatic covid are now these relatively mild mucosal infections. But it is these mucosal infections that are most likely to be associated with further transmission. The second phase a viral pneumonia which is more severe and is the illness that puts people into hospital. Immunity to these more severe (systemic) illnesses, are much more long lasting.

“This study adds further evidence that the effectiveness of vaccines against mild disease wanes after relatively few months and that this waning immunity is likely to be associated with reduced ability to reduce transmission. But as yet there is no strong evidence that immunity to severe disease wanes substantially over the same time scale or that vaccines are less effective at reducing the risk of severe disease from the delta as opposed to the alpha variants.”


Prof Ian Jones, Professor of Virology, University of Reading, said:

“Waning immunity has been a concern since the start of the epidemic, based on data from the commonly circulating coronaviruses. To date however, the studies that have followed vaccination have been a bit more sanguine, suggesting the fall off in antibody titre may be slower than first supposed. This latest study confirms that a decline is happening, but it is not yet clear what this means for disease severity, the key aspect of protection afforded by the vaccines. Sterilizing immunity is not induced by any of the vaccines so the fact that infection still occurs is not surprising but the immunity that is generated, which includes the non-neutralizing antibodies and T-cells that are less tested for, ensures that disease is minimised. The worst-case scenario suggested is certainly possible, but a better-case scenario would be that, even at 50% protection from infection, protection from disease remains robust and hospital numbers remain manageable. The need for boosters still needs to be balanced with global vaccine distribution to populations where even a first shot will lower virus circulation, and with it the chance of future variants.”   


Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, Respiratory Sciences, University of Leicester, said:

“These findings from the ZOE app reflect what other studies have already started to show:

“We know already from lab studies that the delta variant can escape vaccine and natural protection when antibody levels start to fall below a certain level – to cause infection:

“This is not so surprising as the current generation of COVID vaccines were designed against the original Wuhan virus variant – so of course they will be less effective against the delta variant than a COVID-19 vaccine designed specifically against the delta variant.

“Whilst booster shots with the current generation of vaccines designed against the original Wuhan virus will boost levels of cross/reacting antibodies, overall, they will not be very effective against the delta variant as they are not specific for this virus. So the once the antibody levels drop again, the delta variant will break through.

“Furthermore, as long as the delta variant can still replicate and cause breakthrough infections in the population – vaccinated or otherwise – there is always the chance of new variants arising that are even more vaccine escape capable.

“Optimum vaccine protection will only be achieved when specific vaccines are designed for specific viruses – as we try to do against seasonal flu each year – but these take time to design, manufacture and distribute. But as SARS-COV-2 – like other respiratory viruses – mutates continuously, so the vaccines will always be playing catch-up.

“There is a lot of wishful thinking about the efficacy of the current generation of vaccines (designed against the older original Wuhan virus) against the new delta variant.

“Any protection from the current generation of vaccines is relying on some degree of cross-reactive immunity between the Wuhan and delta viruses – where some parts of the delta virus S protein still resemble parts of the Wuhan virus S protein – but some of the essential neutralising epitopes affected by the L452R and T478K S protein mutations in the delta variant are quite different from the original Wuhan virus – hence the breakthrough infections.

“Unfortunately, we really need to develop updated vaccines against the delta virus, specifically, to maintain longer-lasting, specific immunity to this variant.”



All our previous output on this subject can be seen at this weblink:



Declared interests

Dr Peter English: “Dr English is on the editorial board of Vaccines Today: an unpaid, voluntary, position. While he is also a member of the BMA’s Public Health Medicine Committee, this comment is made in a personal capacity.”

None others received.

in this section

filter RoundUps by year

search by tag