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

expert reaction to latest ONS data on deaths registered; and on antibodies from the ONS Infection Survey

The Office for National Statistics (ONS) have released provisional counts of the number of deaths registered in England and Wales in the week ending 19 February 2021 and the latest data from their COVID-19 Infection Survey on antibodies.


Dr Jason Oke and Prof Richard Stevens, Nuffield Department of Primary Care Health Sciences, University of Oxford, said:

“Provisional figures of the number of deaths registered in England and Wales in the week ending 19 February 2021 (Week 7) released by the Office for National Statistics (ONS) show Covid related deaths falling quicker in the over eighties than in people aged 50-79 or under 50.  Death registrations where Covid-19 is mentioned have fallen by 56% in the over 80s since their peak at the end of January, whereas deaths in people aged 50-79 registrations have declined at a slower rate (45%) and in the youngest age group by 42%.  In the first wave, the opposite pattern was seen after the peak – deaths in the over 80s were the slowest to fall across these three age groups.  This suggests these ONS data are now starting to show a modest effect so far of vaccination on deaths in the most vulnerable age group.”


Dr Zania Stamataki, Viral Immunologist, University of Birmingham, said:

“Following the new real-live data from PHE that vaccination reduced mortality in the 80+ group, the ONS reports a faster reduction of deaths in this group which was also the highest age group for detected antibodies in the national survey.  Together with the evidence for reduction of hospitalisations after both the Pfizer/BioNTech and the Oxford/AZ vaccines, the message is clear: vaccines work for those 80+ and 70+ and they are saving lives.

“The second highest age group for antibodies after the 80+ was the 16-24 group, which have not been offered the vaccine yet.  These data reveal the importance of this group in virus transmission, and we should look carefully at supporting the young to prevent spreading infection while the vaccines are rolled out to all ages.  We need to minimise the risk of transmission of new variants to keep our vaccines working for longer.”


Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“There’s more good news in the latest Covid-related bulletins from ONS, but reasons for caution too.  The figures on death registrations take the information up to the week ending 19 February.  They continue the trend, that started four weeks earlier, of weekly decreases in deaths from all causes and in deaths involving Covid-19.  All-cause deaths in England and Wales were down 10% in a week, and deaths involving Covid-19 by 28%, so more than a quarter, in a single week.  As always, it’s helpful to look at the number of excess deaths, which ONS calculate as the difference between the registered number of deaths from all causes and the average number registered in the corresponding week over the five years 2015-2019.  In the week ending 19 February, there were 2,182 excess deaths, that is, 19% more deaths that would be expected at this time of year.  Nearly a fifth more deaths than would usually occur isn’t good, but that excess is falling quite fast, and is more than a third down on the previous week.  In terms of excess deaths, the position is rather better in Wales than in England.  In Wales, the number of deaths from all causes in the most recent week was higher than the five-year average, but by a considerably smaller amount (9%), and the number of excess deaths was within the range found in the five comparator years.

“Deaths from all causes were above the five-year average in every one of the English regions, but by differing amounts in different regions.  There were almost a third more deaths than average for the time of year in the West Midlands, and over a quarter more in the East, the East Midlands and London, but only 6% more in the South West.  The fact that considerably more deaths are still happening than would be expected at this time of year shows that we’ve still got a long way to go. Even though deaths involving Covid-19 fell fast compared to the previous week, there were still over 4,000 of them in that single week ending 19 February.  Nearly 30% of all deaths last week mentioned Covid-19 on the death certificate, and for the great majority of those Covid-related deaths (86% of them), Covid-19 was certified as the underlying cause.

“Our hopes for getting numbers of deaths down to somewhere close to pre-pandemic levels rest to a considerable extent on vaccination.  It takes time for any effects of vaccination to show up in reductions in deaths – all the evidence from trials and real-world data so far do indicate that the vaccines in use in the UK will reduce deaths, but, if someone unfortunately dies from the effects of the virus, this would typically be three weeks or more after they were first infected.  Also, it takes two or three weeks after a person is vaccinated before the vaccine can build up immunity to the virus in their body.  So any effect of vaccination on numbers of deaths wouldn’t start to show up until maybe six weeks after vaccination.  That delay is why, for instance, data on the effects of the Oxford/AstraZeneca vaccine on preventing deaths couldn’t be included in the PHE research on vaccine effectiveness that was published yesterday1.  That vaccine came into use later than the Pfizer/BioNTech vaccine, and though the PHE study could and did look at the effects of both vaccines on hospitalisation (where the delay between vaccination and any effect is shorter), and at the effect of the Pfizer/BioNTech vaccine on deaths, it couldn’t look yet look at the effect of the Oxford/AstraZeneca vaccine on deaths, because not enough time had passed since that vaccine was used on a big scale.  Because of that delay, I wouldn’t expect to see very clear signs of vaccination reducing death registrations in the oldest group (aged 80+) that were vaccinated first.  But there are some very encouraging small signs. Deaths involving Covid-19 fell faster in people aged 80+ than in the under 80s in the most recent week compared to the week before – a 31% fall in people aged 80+ compared to a 25% fall in people aged up to 79.  The fall in Covid-19 deaths between the two previous weeks (week ending 5 February to week ending 12 February) was also faster in the 80+ group than the under 80s.  The difference between age groups isn’t huge, and there’s no way to be certain from counts of deaths whether it’s definitely due to vaccination (and not, say, to people in the oldest age groups being particularly careful about social distancing and lockdown measures), but I certainly regard this as a positive sign that vaccines are probably starting to have a visible effect where it counts.

“The other ONS release today is the latest data from their infection survey on antibody prevalence.  This provides information on how many people in the community populations (aged 16+) of the four UK countries would test positive for antibodies to the virus that causes Covid-19, for the period from 15 January to 11 February.  People can develop these antibodies either as a result of an infection with the virus, or because of vaccination – the test used cannot distinguish whether it was infection or vaccination (or both) that caused the antibodies to develop.  The previous release of data, which covered the position during January, already showed a fairly clear effect of vaccination in England, where the percentage of people aged 80+ who would test positive for the antibodies had increased to a high level (41%).  However, this was not really apparent in the other three UK countries.  In the latest data, though, there seems to be clear evidence of a vaccination effect in all four countries.  In England, for the latest period, 56% of the 80+ age group would test positive for antibodies – more than half of them.  In the next groups down the age scale in England, there seems to be an effect of vaccination too.  An estimated 1 in 4 of those aged 75-79 would test positive for antibodies, and about 1 in 6 for the 70-74 age group.  The antibody positivity rate for the 75-79 group has doubled, compared to the previous report.  The rise for the 70-74 group is a bit smaller but still very substantial.  I wouldn’t expect to see as strong an effect of vaccination on antibody positivity in these age groups as in the 80+ group, because people in their 70s started receiving vaccine later, but the effect is certainly there.  It isn’t (yet) as clear in the other countries – the antibody positivity rates for the 80+ group are 21% in Scotland, and 19% in Wales.  In Northern Ireland the report gives data only for a larger age group, 70+, because not so many people are sampled for this survey in Northern Ireland (because the population is lower).  The percentage for the 70+ group there is 18%.  But vaccination got going slightly later in the other three countries than in England, and the balance of vaccinations in the oldest groups between people in care homes and people living in their own private homes was different, initially, between countries.  (The last point matters, because the survey doesn’t include people who live in care homes.)  Because vaccinations numbers have increased and any difference between vaccination rates in care homes and private homes has reduced, I’d expect the other countries to catch up with England on antibody levels on this survey in the older age groups quite quickly.  The rates are already quite a lot higher in each of the countries than they were in the previous bulletin.”





Deaths registered weekly in England and Wales, provisional: week ending 19 February 2021:

Coronavirus (COVID-19) Infection Survey antibody data for the UK: 2 March 2021:



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



Declared interests

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  I am also a member of the Public Data Advisory Group, which provides expert advice to the Cabinet Office on aspects of public understanding of data during the pandemic.  My quote above is in my capacity as an independent professional statistician.”

None others received.

in this section

filter RoundUps by year

search by tag