select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
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, University of Reading, said:
“As the economy bounces back from lockdown and we approach the autumn and the return of schools and Universities, the news that most recent data show that 1 in 70 people in England were infected with the Covid-19 virus is sobering. This is more evidence that the UK is ‘running hot’ with our management of Covid-19 and we are likely to see numbers increase in the weeks ahead due to a number of reasons. Cooler autumn weather, leading to increased indoor social mixing, is likely to drive further increases in the coming weeks. The end of summer holidays and return of people to work and education, without clear guidance on physical measures to avoid transmission such as mask wearing or social distancing, is also likely to push up infection rates.
“Compared to the same week1 in 2020, where ONS estimated 28,200 or 1 in 1,900 people in England were infected with Covid-19, the data sets for the week ending 20 August this year estimate that 756,900 or 1 in 70 are infected. That number was 1 in 80 just a week earlier.
“This means that community infections are 26 times more common than a year ago, when the population was unvaccinated, and the country was 3 months into its re-opening.
“This time, the vaccine programme is keeping down the number of hospitalisations and deaths but increasing numbers of community infections still translate into growing numbers of very sick covid patients, and an unnecessarily high burden on the NHS.
“Treatments unrelated to Covid-19 like chemotherapy or planned operations become more difficult when unsuspecting patients and staff who may well be asymptomatically infected because of the high community numbers, take the virus into hospital. Making sure that vulnerable patients are protected against such community infections presents a very real burden on NHS time and resources.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“Well, this is a bit depressing. The latest weekly bulletin from the ONS Covid-19 Infection Survey (CIS) takes the prevalence estimates up to 21 August, and ONS conclude that the number of people who would test positive for the virus that can cause Covid-19 is increasing in all four of the UK countries. The total of their central estimates for the number of people testing positive across the UK in the week ending 21 August is 10% up on the previous week. This isn’t surprising. The daily numbers of new confirmed cases on the dashboard at coronavirus.data.gov.uk have been rising in all four UK countries since the start of August. It’s important to see whether that is confirmed in the CIS estimates, because the CIS results come from a representative sample of people who are tested only to measure the progress of the pandemic, so they are not affected by changes in who is being routinely tested, which can bias the dashboard counts. But, in broad terms anyway, these results do match what’s been on the dashboard. Also, modelling results have suggested increasing infection levels, though they differ on how big an increase and exactly when.
“Basically, the pattern in the new CIS results is that infection levels are really high in England and in Northern Ireland. They are quite a lot lower in Wales and in Scotland, but confirmed case numbers on the dashboard have been rising quickly in both of those countries recently, so things look problematic there too. In England, ONS estimate that 1 in 70 people in the community population would have tested positive for the virus in the week to 21 August. The week before, the estimate was 1 in 80 – so infection levels have increased, albeit not all that quickly. However, the infection level for the most recent week in England is still higher in the most recent week than it has been since the CIS started there in April last year, apart from a six week period at the height of the winter peak from mid-December to the end of January. In Northern Ireland, the CIS estimate for the most recent week is that 1 in 40 people would test positive, that is, over 43,000 people in a community population of about 1.8 million. There’s some statistical uncertainty about that figure – the margin of error runs from 1 in 35 to 1 in 55 – but it’s the highest infection estimate in Northern Ireland since the CIS began there in September last year.
“For Wales, the estimate is that 1 in 120 people would have tested positive in the most recent week, and in Scotland, 1 in 140. However, the estimated number of infected people in Scotland in the most recent week is much higher than the week before, when an estimated 1 in 200 would have tested positive. Schools in Scotland generally reopened after the summer break on various dates in the latest week covered by the CIS. It’s just about possible that the reopening might have contributed a bit to the increase in infections, but the effect on the latest figures is unlikely to be large, given that it generally takes a few days after an infection contact for the infection to become detectable.
“It’s true that vaccines have much reduced the risk that someone will end up in hospital or die, if they become infected with the virus. But they haven’t reduced the risk to zero. The last time that infections were at the level they now are in England, according to the CIS, was the end of January. Back then, in England, there were around 2,300 daily hospital admissions of Covid-19 patients, and 1,100 deaths a day involving Covid-19 according to ONS figures based on death registration. The corresponding daily figures for England for the latest available weeks are about 770 hospital admissions and about 80 deaths. (The death registration figure is for the week ending 13 August, so the week before the latest CIS figures. The hospital admissions figure is for the same week as the CIS figures.) And across the whole UK, the number of daily deaths within 28 days of a positive test, on the dashboard, is now about 100. Obviously the position is better than it was at the end of January – but it’s still not good, and the latest dashboard figures and models indicate that things are going to get worse in the short term. What I’d like to hear is an explanation of what policy actions are being taken by the UK government to take this into account, with an explanation of the choices that have been made, even if the choice is not to change anything. I’m not in the business of proposing policies myself – that’s not my job. But I’ve heard very little about policy on Covid-19 for England recently, apart from the welcome encouragement for people to get vaccinated, and some changes in the rules for foreign travel. What’s the plan, please?”
Further information
“The figures that I’ve mentioned so far are all measures of the total numbers who would test positive, regardless of whether they have just become infected or have already been infected for some time. As usual, ONS also present incidence estimates, which count the average number of new infections occurring daily. Also as usual, these are two weeks behind the prevalence estimates for technical reasons, so they go up only until the week ending 6 August. There’s a considerable level of statistical uncertainty in these incidence estimates, particularly in the smaller UK countries where the CIS sample sizes are smaller than in England. For instance, for the week ending 6 August, in Northern Ireland, ONS’s central estimate is that there were about 3,200 new infections a day, but the margin of statistical error runs from 2,200 to 4,300 new infections a day. Because of these wide margins of error, it’s hard to be sure of the trends. In all four countries, the estimated daily number of new infections in the week ending 6 August wasn’t clearly changed from the figure for the previous week. That does roughly correspond to the picture on the dashboard estimate of new confirmed cases. The weeks ending 31 July and 6 August broadly correspond to the time when, in all four countries, the trend in new confirmed cases on the dashboard was changing from a decrease to an increase, so that the numbers didn’t change much between those weeks. I’d expect there to be signs of increase in the CIS incidence estimates next week.”
https://www.ons.gov.uk/releases/coronaviruscovid19infectionsurveyuk27august2021
All our previous output on this subject can be seen at this weblink:
www.sciencemediacentre.org/tag/covid-19
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.