November 19, 2021

expert reaction to latest data from the ONS COVID-19 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 weekly report from ONS on their Coronavirus Infection Survey (CIS) takes the estimates of positivity rates on tests for the virus up to the week 7-13 November. Much of my comment on last week’s release would also apply this week, in fact. The new report is pretty encouraging in terms of the trends in infection, particularly coming after last week’s report that also showed decreasing infection in much of the UK. However, it does still indicate that rates of infection are quite high. Things aren’t yet anywhere near back to the pre-pandemic normal.

“As last week, ONS estimate that the rate of testing positive for the virus that can cause Covid-19, in the community populations of England and of Wales, continued to decrease in the most recent week. In Scotland, ONS estimate that the rate of testing positive remained level in the latest week. In Northern Ireland, ONS concluded that the trend in the most recent week was uncertain. However it is always more difficult to be clear about short-term trends in the smaller UK countries, particularly Northern Ireland which is the smallest, because the survey is designed in such a way that fewer people are swabbed to estimate the positivity rate in the smaller countries, so that the statistical margins of error are wider.

“It’s good to see a pattern predominantly of decreases, but let’s not get carried away. It doesn’t apply in two of the four UK countries, and the general pattern of decreases in the CIS has only been going for two weeks really. In England, ONS estimate that about one person in every 65 in the community population would have tested positive in the most recent week, with a margin of error from 1 in 60 to 1 in 70. Last week’s estimate was 1 in 60, so it isn’t a huge downward shift in positivity. And 1 in 65 is a high positivity rate, make no mistake – before the latest peak in England in the last few weeks, the last time the rate was that high was around the end of January this year, when we were still locked down. Of course, now that vaccination has had such an effect on hospitalisations and deaths, a rate this high isn’t as concerning as it was earlier in the pandemic, but it’s still too high for comfort in my view.

“In Wales, the positivity estimate for the most recent week is about 1 in 55, so on the face of it, it looks as if the infection rate is rather higher than in England – but the statistical margin of error in Wales runs from 1 in 45 to 1 in 70, wider than for England because fewer people provide samples in Wales, so it isn’t entirely certain that the rates in the two countries do differ. In Northern Ireland, the estimated positivity rate for the latest week is about 1 in 65, the same as the England estimate, though the margin of error for Northern Ireland is very wide (1 in 50 to 1 in 85). Until very recently, estimated positivity in Northern Ireland was lower than in England and Wales, but the recent decreases in England and Wales mean that that’s no longer the case. Though ONS say the trend in Scotland is ‘level’, the position there does still look more favourable than in England and Wales, with an estimated positivity rate for the most recent week of 1 in 95 and a margin of error that doesn’t overlap with those for England or Wales.

“As always, ONS also provide estimates for the regions of England, and for subgroups of the English population by age – and as always, because fewer people are swabbed in a single region or age group than for the whole country, the statistical margins of error are wider for these regional and age group estimates than for the whole of England. ONS look at seven different age groups in England. In the latest week they estimate that positivity is decreasing in four of the seven (school years 7 to 11, school year 12 to age 24 years, ages 35-39, and ages 70 and over). In the other age group of children, age 2 years to school year 6, they estimate that positivity was decreasing for the latest two weeks but that the trend in the most recent week was uncertain. For the other two groups (ages 25-34 and ages 50-69), ONS regard the trends as uncertain. It’s good news that ONS did not detect a clear increase in any of these age groups, and it’s particularly good news that they have seen a very marked decrease in positivity in children of secondary school age (school years 7 to 11). The estimate for the latest week is the positivity is still quite high in that group, roughly 1 in 30 – but the rate has fallen by well over half since about 20 October, when almost 1 in 10 of this age group were estimated to test positive and there was a lot of concern about them driving infections on a major scale. The decrease in the positivity rate in this group started falling, according to the ONS estimates, before the most common half-term dates in England and has now continued  for a couple of weeks after the holiday, so it seems not just to be some kind of half-term effect. But the positivity rate remains highest in this age group, compared to all the others except younger children (age 2 to school year 6), where the estimated positivity rate is now roughly the same at about 1 in 30. The positivity rates remain a lot lower in adult age groups, even young adults, and for those aged 70 and over, for example, the estimate is 1 in 190 testing positive.

“ONS are estimating that the positivity rate decreased in the latest week in four of the nine English regions (North West, Yorkshire and The Humber, West Midlands and South West), and they say the trend was uncertain in the other five. On the whole, regional estimates of positivity differ rather less than they have done fairly recently, and it’s good to see that the peak in positivity in the South West, peaking in mid to late October, has now fallen back so that the estimated positivity level there is now similar to the English average. Positivity levels are now highest in the Midlands regions – about 1 in 50 in the East Midlands and about 1 in 55 in the West Midlands – and again lowest in London at about 1 in 80 – but these are not large regional differences, taking into account the fairly wide margins of statistical error.

“The question is, how long will this general pattern of decreases in positivity go on? That’s hard to say, I think. I hope it does continue, but I certainly can’t be sure about that. We’re getting further into the cold season of the year when respiratory infections tend to do their worst. Also, the numbers of new confirmed infections on the dashboard at coronavirus.data.gov.uk have shown some signs of increasing in the last couple of weeks, though the increase has not been great yet, and small movements in the dashboard cases and in the CIS do not always coincide. I think we have to wait and see.”

Coronavirus (COVID-19) Infection Survey, UK: 19 November 2021 https://www.ons.gov.uk/releases/coronaviruscovid19infectionsurveyuk19november2021

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.”