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. 


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

“The latest figures from the ONS survey provide some of the best quality data about levels of infections in the community. The survey has been running with large sample sizes and consistent methods for many months, so should not be susceptible to changes in the types or numbers of tests being carried out, or other variables.

“For me, the most significant trend is the small increase in infections among secondary school-aged children. This appears to suggest that children mixing in schools is having an impact on infections, despite all the additional control measures and new testing regimes introduced in schools.

“While children are at much lower risk of serious disease from covid, we have seen in past waves and in other countries that children can keep outbreaks going, providing a reservoir of virus that could put more vulnerable populations at risk. It should be remembered that no coronavirus vaccine yet produced provides 100 per cent protection against infection.

“The data also show the possible first sign of an increase in infections among those aged 35-49 years. This is the group that would include many parents of secondary school pupils. It is also the age group which is still largely unvaccinated, and while this middle-aged group is at less risk than the extremely elderly, many people this age have died or had long-term health impacts from catching Covid-19.”


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

“Today’s results from the ONS Infection Survey suggest that the decline in COVID-19 infections seems to have slowed or even plateaued. Indeed there does seem to have been a small rise in positive rates in secondary school aged children (years 7 to 11). The ONS Infection Survey Includes all positive samples whether from symptomatic individuals or not.

“For most of the past year there was a fairly close correlation between the estimated prevalence of COVID from the ONS survey and from the incidence of COVID as reported each day from the daily reports of new cases. However, we are now seeing this correlation becoming less close as a result of the vaccine programme rolling out. Largely because as would have been predicted from the phase 3 trials of the Oxford AstraZeneca vaccine the vaccines are better at preventing severe disease than at preventing asymptomatic infection.

“Looking at the age breakdown of daily cases from the DHSC dashboard1 it appears that case numbers are still declining in all age groups over 20 years, especially in the over 60s. Also daily admissions to hospital are still declining in all age groups over 20.

“So should we be worried about this apparent plateauing of cases and a possible uptick in school aged children? Clearly this is something that needs to be monitored closely over coming weeks, but it may not be something that will derail the roadmap out of lockdown. Firstly the uptick in children is not yet sufficiently big to indicate that this is definitely a real increase and not just a random variation. Secondly at present this does not yet appear to be causing any slowing in the decline of daily admissions to hospital, though it may still be a little early to expect to see any clear impact.    

“In my view, this latest data does not put the next phase of easing out of lockdown in doubt. The main change on the 29th March is that restrictions on meeting outsides will be eased and we know that transmission of COVID outdoors is much less efficient than indoors. So this Monday’s relaxation is unlikely in my view to affect transmission rates very much at all. Then with the school Easter break coming up we are unlikely to see any further upward pressure on transmission before the 12th April. After that of course further relaxations will depend on data available towards the end of April. But with recent declines in prevalence slowing whilst case numbers are still relatively high we do need to be careful about not going too fast out of lockdown.”



Prof Lawrence Young, Professor of Molecular Oncology, Warwick Medical School, said:

“It was inevitable that infections would start to level off and even increase as we ease from lockdown. We are just beginning to see the impact of students returning to school, particularly in Wales and Scotland where schools went back earlier. Every change as we slowly release from lockdown is likely to have an effect on the levels of infection. This is why the roadmap is taking a cautious approach – leaving gaps between each relaxation so that the impact on infections can be carefully monitored. We need to remain vigilant particularly keeping an eye on local outbreaks and increases – as evidence by increased infections in the North West, Yorkshire and The Humber. While vaccination and the current lockdown are clearly bringing down overall infection levels, there’s no room for complacency as we lift further restrictions. We need to continue the pace of vaccine rollout, effectively spot and stamp out local outbreaks (test, trace and isolate) and prevent compromising the situation by stopping the importation of virus variants (border control).”


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

“This week’s bulletin from ONS on their infection survey doesn’t provide such positive news as some other data we’ve had in the past couple of weeks. I wouldn’t say there is huge cause for concern, but the new data do remind us that things are far from being back to normal (whatever that means exactly).

“For many weeks since the peak of the latest pandemic wave around the turn of the year, these ONS bulletins showed very clear evidence of decreasing rates of infection across all four UK countries. Last week ONS reported that the estimated rate of testing positive for a current infection was still falling in England, and possibly in Wales too, but it had levelled off in Northern Ireland, and there were signs of a small increase in Scotland. In today’s bulletin they take the data up to the week ending 20 March, and there’s more evidence that the previous decreases have levelled off, or even reversed. In England, ONS estimate that 1 in 340 people in the community population would have tested positive for the virus that can cause Covid-19, if tested in the week ending 20 March. That’s the same estimate as the previous week. The estimate for Wales is 1 in 450, a very slightly lower rate than the week before, and in Northern Ireland it is 1 in 320, again very similar to the week before. In Scotland the estimate is 1 in 240, which is a higher rate than the previous week when it was 1 in 275. So the position appears to be that infection rates are about the same in England and in Wales, with Scotland a bit higher and Wales a bit lower.

“All these figures are subject to some statistical uncertainty, which is quite considerable in Scotland and (even more so) Wales and Northern Ireland because fewer people are swabbed for the survey there than in England. So we can’t be completely certain that the infection rate differs between the countries at all, and the trends are far from being entirely certain. ONS’s view is that the previous decrease in people testing positive has probably levelled off in England, Wales and Northern Ireland, but in Scotland it may be increasing. I think those are reasonable conclusions, but they can’t be definite.

“They aren’t too alarming, I’d say. Lockdowns have been eased in various ways in all four countries, and it was always a pretty safe bet that reopening schools would change trends. But the positivity rates certainly aren’t powering upwards fast, and even the increase in Scotland is really pretty slow. Given that the vaccination roll-out is showing very clear evidence of reducing the risk of severe illness and death, for people unfortunate enough to get infected, a levelling off or a moderate rise in the infection rate is unlikely to have the serious consequences for the NHS than it would have before vaccinations were available. And, as the weather warms up, we can spend more time outside (once the rules permit) and that will put a downward pressure on infections.

“But we shouldn’t be complacent. The rate of testing positive is sticking around the level that it was in late September or the start of October, and in England (where the data are the most reliable) the estimated rate is currently six times as high as it was back in August last year. Not everyone can be vaccinated, and no vaccine is 100% effective.

“This week, ONS have resumed the publication of estimates of the daily number of new infections (the incidence rate, in the jargon). What I’ve described so far is about estimates of all infections (the so-called prevalence rate), including people newly infected but also people who people who may have been infected for some time. Comparing these new ONS estimates of new infections with the counts of new confirmed cases on the dashboard at gives a pretty clear indication that a large proportion (quite possibly over half) of new infections just don’t show up in the dashboard counts of new confirmed cases. That’s quite probably, for many of them, because they don’t have any symptoms. But at least some of the infected people with no symptoms can pass the infection on to others. That’s another reason for continuing caution.”


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

“Today’s ONS data are less reassuring than last week; they serve as a caution but not yet evidence we need to change course.

“The prevalence of the virus (measured one week ago) shows an increase in Scotland and, it now appears, in part of the North of England.

“Overall, the numbers in England (and for the UK) have levelled off and are showing no further decline, at roughly 3 people in 1000.

“What has to be understood is that the new variants are harder to control and it seems likely that the limited reopening of schools has started to show up in data.

“Scotland has gone into this cycle first – its numbers levelled off around 2 weeks ago.  Scotland should therefore be particularly closely watched with respect to decisions about further relaxation across the UK.  If cases continue to rise in Scotland, it is safe to assume that this will happen everywhere else a week or two later.

“When is too early to relax social restrictions? The governments in the UK have made clear they will be guided by data not dates. I am afraid this is correct, frustrating as it is.  We seem unlikely to lower prevalence further and may be lucky to keep at current levels, although more outdoor living will help as the weather improves. 

“Realistically, we only have the vaccines as a way out.  The successful vaccination campaign has already saved lives.

“Even in a full-blown third wave, the number of hospitalisation and deaths for every thousand people who become infected would be much lower because of vaccination.  However, no one who has not been vaccinated is entirely safe from serious illness and a large wave of infection in young people will lead to hospitalisations.

“The vaccines are safe, and very effective at saving lives and preventing hospitalisations. We have still to work out how successful they are at limiting the spread of the virus.  Even if they are 100% effective at limiting spread, the biggest effects upon the spread of the virus will come as we approach the herd immunity threshold.  That is to say, going from 65 % to 75 % vaccination has much more effect that going from 0 to 10 %.  I stress that vaccinating 50 % of the population is a huge achievement that will reduce spread but it is not yet enough to prevent a third wave in the young were we to completely unlock.



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



Declared interests

Prof 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