select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
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:
“More good news, but still some important reasons for caution, in the latest results from the ONS infection survey. The estimated percentage of people that would test positive for the virus that causes Covid-19 has fallen in each of the four UK countries, for the most recent week (13-19 February) compared to the previous week, and that continues a pattern of decrease that started at about the turn of the year. However, the pattern of decreases isn’t the same across the whole UK.
“Over this most recent week, the decrease in positivity was very considerably faster in Wales and in Northern Ireland than it was in England and Scotland. In Wales, the estimated positivity went down by 41% of the previous rate in a week, and in Northern Ireland by a huge 46%, so that it almost halved. For England, the figure was 22%, and 18% for Scotland – so a fall of roughly one fifth in a week in those countries. These figures are subject to greater statistical uncertainty in Wales and Northern Ireland, and to a smaller extent Scotland, than in England, because fewer people are included in the surveys in the other three countries because of their smaller populations.
“As a result of these different speeds of decrease, the picture across the UK looks quite a bit different from the previous week. In that week, 6-12 February, the ONS estimates of the rate of testing positive were pretty similar in England, Wales, and Northern Ireland – there were small differences, but no clear statistical evidence that the rates differed. The rate was clearly lower in Scotland in that week. But for the week of 13-19 February, things look different. In England, ONS estimate that 1 in 145 people would test positive. In Wales, it’s 1 in 205, in Northern Ireland 1 in 195, and in Scotland 1 in 225. So now the big difference is between England and the other three nations. Because the number of people swabbed for the survey in Northern Ireland is relatively small, it’s actually not entirely statistically definite that the rate in Northern Ireland is lower than in England, though it very probably is, and the evidence that the rates in Wales and in Scotland are lower than in England is strong.
“This change in pattern is because of the very rapid declines in positivity rates in Wales and Northern Ireland in the most recent week – the rate in Scotland didn’t fall so fast, but it was already much lower than the rate in England. In fact the estimated positivity rate in Scotland for the most recent week is lower than it has been for any week since the ONS first published results for Scotland in October.
“Even the falls in Scotland and England, around a fifth lower in a week, are quite large. But they only take the numbers of infected people down to where they were around early October last year, when there was already quite a lot of concern about how high they were compared to the position in the summer. However, in October, infections were generally increasing and there was no mass vaccination – now, infections are decreasing pretty fast, and millions have been vaccinated already. So yes, good news, but there’s still some way to go.
Additional information
“In terms of possible effects of vaccination on different age groups, one might expect that rates of testing positive could be falling faster in the oldest age groups, since more of them have already been vaccinated. There is perhaps some indication of this in the data, but the statistical signal is not very strong (yet), partly because the number of people tested in individual age groups is relatively low, so there’s a lot of statistical uncertainty. The detailed pattern differs between the different countries.
“In England, where the number of people tested is greatest and the statistical accuracy is therefore highest, there’s evidence that the percentage positive decreased in all age groups, with the possible exception of people of secondary school age up to 24 years, where ONS say the trend is ‘uncertain’. (Their central estimate shows a decrease, but a small one compared to the other age groups.) The fastest decrease is in children from age 2 up to the end of primary school, where the number who would test positive fell by two fifths in a week – but the fall for the 70+ age group was large too (almost a third), and the infection rate in that oldest group remains lowest of all the age groups than ONS use for this analysis.
“ONS estimate that the infection rate fell, over the most recent week, in all English regions except possibly for Yorkshire and the Humber, where the trend is not certain. Rates appear to have fallen fastest in the South East, and more particularly in the East of England and the South West, and that does leave the rates of testing positive in those regions fairly clearly below the rates in other parts of the country.”
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.”