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

expert reaction to latest data from the ONS Infection Survey

The Office for National Statistics (ONS) have released the latest data from their COVID-19 Infection Survey.


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

“The latest results from the ONS infection survey are encouraging, very good news in nearly every detail.  But there’s a very important difference between a piece of good news and everything being fine now.  It’s a bit like when a relative has been very ill in hospital and you hear that they are getting better.  Of course that’s great, but it doesn’t mean that their treatment can stop, or that they can leave hospital now, or that there’s no chance they might start getting sicker again.

“What’s excellent about these ONS results is that they show evidence that the number of infections is falling in all four countries in the UK, in all but one region of England, and in all the age groups that ONS divide the population into.  In England, for example, ONS estimate that between 1 in 75 and 1 in 85 of the community population would test positive for the virus, if they were tested on a day between 31 January and 6 February.  The week before, the range was 1 in 60 to 1 in 50.  That’s roughly 150,000 fewer infected people in England.  The rate of testing positive is reasonably similar in Wales and in Northern Ireland to the position in England, though there’s more statistical uncertainty.  The rate is lower in Scotland – between 1 in 130 and 1 in 175.  But there’s good evidence that all these rates are falling.

“What’s not so good, yet, is that the infection rate in all these countries is still pretty high.  High and falling is much better than high and increasing, but really we need to see low and falling rates.  In England, for instance, the rate in the most recent week is similar to what it was in early November, when the Government had just found it necessary to impose a lockdown.  Our imaginary ill relative is getting better, but they are still very ill and we shouldn’t imagine that things can just go back to normal yet.

“One small dark cloud on the horizon in these results is that, in one English regions, the South West, ONS can’t say that infections are falling.  Their assessment is that infections are levelling off there in the most recent week, after some evidence of a fall in infection rates in previous weeks.  But the rate in the South West is still estimated to be lower than in several other regions (though not all of them). I can’t say whether this small difference in trends there has anything to do with the virus variant first discovered recently in Bristol, and the ONS data can’t tell us about that.  In any case it’s not a huge difference compared to trends elsewhere – but it does again show that we’ve got to be careful.”

Additional information

“The ONS Infection Survey carries out repeated swab tests on a sample of people that is reasonably representative of the community population of each of the four UK countries.  Because these people are tested only in order to check the progress of the pandemic, the estimates won’t be affected by biases that can arise in the daily counts of new confirmed cases because of changes in the types of test being carried out or in the types of people who ask to be tested (or are asked to have a test).  That said, the trends in the rates measured in the ONS survey do broadly reflect the trends in new confirmed case numbers – both are decreasing reasonably fast.  I wouldn’t expect the exact numbers to match up anyway.  The ONS survey (and the REACT-1 survey in England) are not measuring the same thing as the counts of new confirmed cases.  That’s because the surveys measure prevalence – the percentage of people who would test positive, which includes people who have just very recently been infected but also people who were infected some time ago but would still test positive (and might still be able to infect others).  (On average, according to one study, people continue to test positive for about 17 days after they would first test positive.)  The new confirmed cases measure incidence – the number of new cases, not including people who were a new case the day before or the week before.  Currently, new confirmed cases are falling faster that the total positive infections measured by ONS, but that doesn’t mean that either set of numbers is wrong, only that they are measuring different things.

“I haven’t commented on the information in the ONS report on different virus variants.  That’s because the picture is quite complicated.  The results from swab tests can’t show exactly which variant someone has.  They can show whether someone has an infection that could be the variant that was first observed in Kent in November, though it’s possible that a few of those cases do not in fact have that variant.  That ‘Kent’ variant is now dominant across the whole UK.  But the swab tests can’t distinguish between the old virus variants that were around before the ‘Kent’ variant emerged and some of the other new variants, such as the South African variant.  Full sequencing of the samples would be required for that, and that takes longer and can’t always be done.  Also there’s a lot of statistical uncertainty about the detailed data on possible variants from these ONS results.  So I’ll leave any comments on this to others with more expertise on the matter than I have.”


Prof James Naismith, Director of the Rosalind Franklin Institute, and University of Oxford, said:

“The UK has suffered through two deadly waves of covid19.  In terms of lives lost per million, the UK has been one of the worst performers.  The roll out of the safe and effective vaccines is a true bright spot.  We owe it to families who have lost loved ones, those who livelihoods have been destroyed,  those whose education has been derailed and those who have suffered isolation or health problem to squarely face what went wrong and why, what went right and why and how we can do better in future.

“Today’s prevalence numbers (for the week ending 6th Feb) show prevalence dropping across all age groups, all UK nations and in almost all UK regions.  The prevalence in England has dropped to around 1 in 80 people, Scotland has 1 in 150.  It would appear that across England prevalence may have stopped falling in the South West at roughly 1 in 100.  It is to be hoped that we can continue to drive down prevalence to much lower values.  We are still well above levels seen in the summer.

“I find the data on the so called Kent variant sobering.  It has become dominant in all the UK nations.  Fortunately for us, the vaccines are highly effective against it.  Had this variant arrived in the UK prior to imposition of social distancing and well before the vaccines, the consequences would have been unbearable with deaths rates and disruption way above anything we have seen anywhere.  Despite the lockdown, this variant is only slowly being suppressed.  We have been lucky.  Had we not locked down and cancelled Christmas for much of the South of England, the death toll, already high, would have been much higher still.  It is therefore important that we properly asses risks before reducing social restrictions.  In estimating the harm removing restrictions could cause we need to consider infection prevalence, vaccination progress, to what extent the vaccine alters transmission and new variants.  In holding restrictions we need to fully consider the harm these do.  In science, changing one’s mind in response to data is a vitally important skill.  Those siren voices who at every turn have opposed lockdown or urged immediate lifting have unenviable record of advocating policies that would have killed tens of thousands.  Returning to as normal a life as possible, is something every single one of us wants.

“I am filled with foreboding for less well off countries.  The new variant poses a deadly threat to their people and their economies.  Both humanitarian concerns and selfish ones, (virus circulating widely will lead to new strains), demand we help these countries.  I applaud and strongly support the UK government commitment to the COVAX initiative.”



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