select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
The government have published the latest estimates for the COVID-19 R number and growth rates.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“This week’s revision to the UK Government’s ranges for R and the growth rate of new infections is a little discouraging, in that the overall position seems to be a bit worse than last week’s revision. But I don’t think there’s any great cause for concern. You might well expect infections to stop going down so quickly as they have, or even rise by a certain amount, as lockdown restrictions start to be removed, schools reopen and so on. But there’s no sign in these new figures that infections are increasing, not yet at any rate, only that they might be falling more slowly that was the case in recent weeks. Also, the roll-out of vaccinations means that any effect of changes in the numbers of new cases on serious illness, hospitalisations and deaths is likely to be much smaller than it would have been in the past. So let’s not be complacent, but these new figures don’t make me anxious.
“The revised range for R for the whole of the UK is 0.7 to 0.9. The lower end of that range has gone up – last week the range was 0.6 to 0.9. The fact that the lower end has increased doesn’t mean that R has definitely increased – for example, if R were 0.8 last week and this week, that’s inside both ranges. R might have gone up a little, but the whole range of possible values is below 1.0, the figure which would indicate that infections were not changing.
“Also, the range for the daily growth rate of new infections for the UK is still entirely below zero. This week it runs from -5% to -2%. Putting it another way, SAGE are saying that, for every 100 infections today, there will be between 95 and 98 tomorrow, so the number of infections continues to fall rather than rise. A growth rate of -5% per day, which means a rate of decrease of 5% per day, if it continued at that level, would mean that the number of new infections would halve roughly every two weeks. A growth rate of -2% per day corresponds to new infections halving in about five weeks. Last week the UK growth rate range was -6% to -3%, so this week’s range has increased at both ends. That indicates that infections might be falling slightly more slowly than before, though the actual rate may not have changed. (A growth rate of -4%, that is, a rate of decrease of 4% a day, is in the interval for both weeks, for instance.)
“The position looks slightly worse for England as a whole, and for the English regions. The R range for England this week is 0.8 to 1.0, higher at both ends than last week’s range of 0.7 to 0.9, and now it goes right up to 1.0. That’s compatible with R for England having gone up a little, and it is not longer so certain that R for England is below 1.0. But even R at 1.0 doesn’t mean that infections are increasing, just that they may have levelled off – and 1.0 is the upper end of the range, so SAGE are still saying that R is likely to be below 1.0 and therefore that cases are continuing to fall. The same broad conclusion applies to the range for England for the growth rate. This week’s range runs from -4% (new cases decreasing by 4% a day) to 0 (no change up or down in daily cases). That’s higher than last week, when the range was from -6% to -2%, but the worst case in it is still that infections aren’t changing, not that they are increasing. The range most likely indicates that cases are continuing to fall, but not so fast as they have been falling in recent weeks.
“The same goes for the individual English regions. In all but two of them the R range goes right up to 1.0, the exceptions being London and the South West where the upper limit is 0.9, but it does not go above 1.0 for any of them. For four of the regions, the growth rate range now goes right up to 0 (no change in daily cases), but for the other three (London, North West, South West) its upper end is 1%, a 1% daily decrease in new cases, and none of the ranges goes above zero. Given that the most likely values for R and the growth rate are near the middle of their ranges, not at one end, again it looks as if new cases are falling in all regions, but not quite so fast as in recent weeks.
“As always, this data release does not cover the other UK countries outside England, apart from including them in the overall UK ranges, but it provides links to the latest R estimates from the devolved administrations. They all give ranges for R that are below 1, though in Scotland the range does go right up to 1, and it ends only just below 1 in Northern Ireland. Again the indication is that cases are falling, but may be falling slowly.”
Further information:
“These trends are arguably a bit more positive than shown in today’s bulletin from ONS on their infection survey, which reports that the fall in infections has probably levelled off in England, Wales, and Northern Ireland, and that infections may be increasing (quite slowly) in Scotland. But the difference between the trends in the two sources of data is not very big, I’d say. To some extent the R and growth rate estimates lag behind the latest trends, because to some extent they are based on data on events like hospitalisation and deaths that occur some time after a person is newly infected. So in a way they can be seen as a kind of average over the past few weeks, rather than exactly today. The ONS latest data apply to last week (week ending 20 March). So these ranges and the ONS infection survey results don’t completely match up on dates, and the R and growth rate ranges are based on wider sources of data than just infections. So it’s not surprising or worrying if they don’t match exactly.”
All our previous output on this subject can be seen at this weblink:
www.sciencemediacentre.org/tag/covid-19
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.”