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, and the government have released the latest estimates for the COVID-19 growth rate and R value.
Dr Simon Clarke, Associate Professor in Cellular Microbiology, University of Reading, said:
“On the day when it seems likely that the number of Covid-19 fatalities in the UK will overtake the country’s tally of civilian deaths during the second world war, the latest Covid-19 report from ONS is sobering. The data continues to show a mixed picture across England, with clear and continued increases in the numbers of people infected in some areas and a slowing in the rate of decline in some areas that are under Tier 3 and even some increases in areas in the top tier. Nationally the decrease in numbers which we’d previously seen has clearly gone into reverse. New infections continue to increase in London and the East of England, but the situation in the East Midlands, North East and South East is now deteriorating. If this continues, as seems likely, the numbers of hospitalisations and deaths can also be expected to increase in the coming weeks; policy makers and members of the public should reflect on what this means for the festive period.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“Given the recent messages from other sources of data and analysis, it’s not at all surprising that the latest ranges for R and for the growth rate for the pandemic, from SAGE and the Government Office for Science are higher than last week’s estimates, and last week’s were in turn a bit higher than those for the week before. This week’s range for R, for the whole UK, is 1.1 to 1.2, so entirely above 1, meaning that the pandemic is growing again. Last week’s range was 0.9 to 1.0, indicating that R was probably less than 1 and the pandemic might not be growing. But these estimates are based to some extent on data on hospital admissions and deaths, and if those events do occur, they don’t happen immediately after someone is infected. This means that the R number ranges can’t reflect the most up-to-date position. It’s now very clear from other data sources that the pandemic was in fact growing last week, on average across the whole of the UK, and that growth continues. The range for the UK growth rate is now +1% to +4% per day, meaning that each day, on average, there will be between 1% and 4% more new infections that there were the day before. A growth rate of +1%, if it continues, would mean a doubling of cases every 10 weeks, but a +4% daily rate corresponds to a doubling time of about two and a half weeks. So there’s quite a lot of uncertainty – but the key point is that both ends of the range indicate a growing number of infections, and if the true growth rate is near the top end of the range, infections will be growing quite fast. The estimation process generally looks at current (and to some extent past) data, so it can’t yet have fully taken into account the very recent changes to tiers in England, let alone any interventions that haven’t happened yet, such as consequences of the formation of “Christmas bubbles” or restrictions that are already announced for after Christmas in Wales and Northern Ireland.
“The R number and growth rate ranges for the English regions and for the other UK countries do differ from one another, though again the picture does match what has already been seen from other data sources. The position looks slightly better in Scotland than (on average) in the other UK nations, though there is still a considerable chance that R is above 1 there. For England as a whole, the R range goes slightly higher than for the whole UK, at 1.1 to 1.3, and the lower end of the growth rate range is also slightly higher (the England range is +2% to +4% per day). R and the growth rate are estimated as highest in the East of England, the South East, and London. The ranges are lower in the North East and Yorkshire, and in the North West, but even there, the range for R goes from slightly below 1 to above 1, so the epidemic may (or may not) still be growing there.
“These figures do correspond, in broad terms, to what’s known about the current status of the pandemic from other sources. But it’s important to take all available sources of data and analysis into account. Other sources include the infection surveys from Imperial College (REACT-1)* and the ONS**, which test swabs from reasonably representative samples of the community population of England (and, in the case of the ONS infection survey, from the other UK nations too). There are also the daily counts of new confirmed cases, presented on the dashboard at coronavirus.data.gov.uk, and widely reported in the media, and similar data can be found in the weekly Test and Trace reports. A potential issue with the data on new confirmed cases is that the number of positive tests does depend on how many people, and which people, present themselves for testing, and that can vary over time, though that doesn’t seem to be a major complication for the interpretation of the results at present. But in any case that issue doesn’t really apply to the survey data, because people are tested only to measure infection levels across the country. And the R number and growth rate estimates are based on wider sources of data than swab test results. All of these sources feed into the overall picture, and in broad terms, they are now all saying the same thing. In England as a whole, the counts of confirmed cases fell from early November until the very end of November or the first day or two of December, and then began rising again. That matches the estimates of the number of people with a current infection from the ONS infection survey – the estimate published today for the most recent week (6 to 12 December) is higher than the estimate for the previous week (29 November to 5 December). The rise since early December does not show up in the REACT-1 overall estimates, because they are based on swabs taken only up to 3 December, before the increase in infections really got going – though the REACT-1 researchers did find early signs of increases in London and some adjoining areas. Confirmed cases have also been rising in the other UK nations, again starting around the end of November in Scotland and in Northern Ireland, though earlier and faster in Wales (beginning in mid-November). So everything is pointing in the same direction, and unfortunately that direction is not encouraging at all. ONS have published another very useful regular weekly bulletin*** today, for England, putting together data from these sources and others.
“One further point about the ONS Infection Survey. The data on new confirmed cases relate to new infections, and the R number and growth rate estimates also relate to new infections. The estimates from the ONS survey (and from REACT-1) are for all infections – that is, they estimate what is known as the prevalence of positive tests, i.e. the proportion of people who would test positive. That includes people who are newly infected, but also people who were infected some time ago but are still testing positive. The ONS survey has previously published estimates of the rate at which new infections are occurring each day – the so-called incidence rate – and that is more similar in meaning to the counts of new confirmed cases, and to the growth rate of new infections in the R and growth rate ranges. The ONS survey can do this because it tests people repeatedly and so can tell, approximately, when someone’s infection actually began. However, there have been issues with the method of producing the incidence rate estimates from the ONS data, because, as the survey progresses, people move from being tested weekly to being tested monthly. Because of those issues, ONS last week paused the production of the incidence rate estimates, and this week they have announced that they aim to re-introduce the incidence rate estimates on a better footing from some time (still to be announced) in January.”
**https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/18december2020
Prof Linda Bauld, Professor of Public Health, University of Edinburgh , said:
“As we look ahead to Christmas and New Year, current findings from the ONS Infection Survey and the publication of the latest R values and growth rates provide real cause for concern.
“Infections levels are rising in large parts of England with the exception of the North West and Yorkshire and Humber. In Northern Ireland, recent decreases have stalled, in Wales there has been a substantial growth in infections and in Scotland the percentage testing positive has risen. Following last week where the R value was hovering around 1 for the UK as a whole, it is now clear that it is above one and new infections are growing each day.
“This is why the easing of restrictions for a five day period over Christmas increasingly looks like the wrong decision at the wrong time. Everyone recognizes the importance of this holiday period for many around the country, and the need to see loved ones who have been cut off from friends and family for long periods. But by allowing travel around the UK and changing guidance to allow household mixing indoors we are setting ourselves up for a miserable January with tough restrictions. If the NHS is put under even more pressure and staff are absent due to Covid-19 or caring responsibilities, then a worst case scenario is that this Christmas easing could even delay one of the main routes out of this pandemic – vaccine rollout. “
ONS Infection Survey Results: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveypilot/latest
R number and Growth rates:
https://www.gov.uk/guidance/the-r-number-in-the-uk
All our previous output on this subject can be seen at this weblink:
www.sciencemediacentre.org/tag/covid-19
Declared interests
None received.