select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
The Office for National Statistics (ONS) have released the latest headline results from their COVID-19 Infection Survey.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“There’s a lot of positive news in this week’s release of headline results from the ONS Coronavirus Infection Survey (CIS). This takes the estimates of the numbers of people across the UK who would test positive for the virus up to the week ending 15 January. That’s a little more than a week on from last week’s data release, which covered the week ending 6 January for most of the UK but 7 January for Scotland. During that time, the total of the ONS estimates for the numbers of people in the four UK countries who would test positive on any day, on average, fell to about 3.4 million, from about 4.3 million in last week’s release. That’s a fall of roughly a fifth in a week, which is pretty substantial. As always with estimates based on a survey, there’s some statistical uncertainty around these figures, but there’s no real doubt that the number of infected people has gone down quite a lot.
“The position does differ from one UK country to another. ONS estimate that the trend was definitely downward in three of the countries, but not in Northern Ireland, where they say the trend was ‘uncertain’ in the most recent week, but upwards over the two weeks ending 15 January. That looks like a reasonable assessment – the estimated number of people there who would test positive in the most recent week is slightly up (by 5%, or about 5,000 people) compared to last week’s release, but the margin of statistical error is relatively wide for Northern Ireland because the number of people swabbed for the survey is relatively small, so it’s far from certain that there was a real increase. The decreases in the other three countries do look real, though. The numbers went down by about a fifth in England and in Scotland, and by even more (about a third) in Wales, though the margin of error is relatively large in Wales too. As a result, the latest estimates are that about 1 in every 25 people would test positive in Wales, and about 1 in 20 in each of the other UK countries.
“Does this mean the Omicron wave is over? No it doesn’t. Levels of infection at 1 in 20 or 1 in 25 are still high, with estimated infection levels in each country the highest they have ever been since the CIS began, apart from the previous two weeks (and in Northern Ireland, the highest they have ever been since the CIS began there). But it does mean that the recent very rapid increases have stopped, and in many areas the tide has turned. That’s not true yet in Northern Ireland, and as I’ll explain, it’s not entirely true in all the English regions or all the age groups. And the numbers have been falling for only about a week so far – it’s possible that might not continue, or at least not continue so rapidly. But my feeling is that it will continue, and that the English regions and Northern Ireland will start showing the same downward trend before long, so these numbers have cheered me up. We still do have to be careful, but we can be optimistic that things are showing clear signs of improvement.
“Headline results are given for individual regions, and separate age groups, in England only (as usual). However, as always, there are wider margins of error on the regional and age group estimates than for the whole country, simply because fewer people are swabbed for the survey in a single region or age group than for the whole country.
“For the regions, ONS consider that the infection level fell in the week ending 15 January, compared to last week’s results, in seven of the nine regions, but they say that the trend in two of them (the North East and the South West) is ‘uncertain’. The margin of error for the North East is relatively wider than for all the other English regions, because fewer people are swabbed there, which is in turn because its population is smaller than for all the other regions, so the trend there always tends to be harder to call. In fact the point estimate of the infection rate there increased, compared to the week before, by about 10%, so from around 1 in 15 of the population to around 1 in 10. In all the other regions, the estimated infection rate fell compared to the previous week, though by considerably more in some regions than others. The decrease in London and in the North West was around a third, while in the West Midlands, the East of England, and the South West it was around 10% or less (and in the South West, ONS feel that the trend is too uncertain to be sure it really is a decrease). The estimated rate of testing positive in the South West is lowest of all the regions, at about 1 in 25, but that’s still a pretty high infection rate. In the other regions in the South of England (South East, East and London) and in the East Midlands, the estimated rates are about 1 in 20, and in the remaining three regions (West Midlands, North West, Yorkshire and the Humber) about 1 in 15.
“On age groups, ONS report that rates fell in every group apart from the youngest (age 2 to school year 6), where the rate of testing positive increased. That increase in the youngest group isn’t surprising, given that they would have been back at school throughout the most recent week. The rate for that youngest group went up by about a tenth in a week, from about 1 in 15 to nearer 1 in 10. It’s much more encouraging that the infection rate did not go up in children of secondary school age – in fact their estimated rate fell, from about 1 in 15 to about 1 in 20. The CIS data can’t tell us the reasons for that difference in trends between younger and older children, but it could have something to do with the fact that vaccinations are available for many children of secondary school age, but not for most children or primary school age and younger. Rates in all the age groups from school year 11 (16 or 17) right up to age 69 clearly fell.
“However, the infection rate in those aged 70+ barely changed over the latest week, just a very slight fall. ONS do regard it as a decrease, though, but we’ll have to wait for the more detailed CIS results on Friday to see exactly why they think that. It’s still true that infection rates are lowest in the 70+ group, with somewhere between 1 in 30 and 1 in 35 infected, but that remains a relatively high infection rate for that group. The infection rate for the next group down, age 50-69, is lower than all the other groups apart from 70+, at about 1 in 25, with all the other groups between school year 7 and age 49 having estimated rates of about 1 in 15 or 1 in 20.
“Overall, mostly I see encouraging signs of decrease in the age group estimates, but the slower decrease in the oldest group is a little concerning. I’m rather more concerned, though, about the increase in infection rates in young children, partly for the sake of their own health, partly because absence from school and illness in those children will affect their education, and partly because they might infect their families. That infection rate in the youngest group will be the first thing I look at in next week’s release.”
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.”