select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
The government have released the latest NHS Test and trace statistics.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“While is certainly isn’t encouraging that the number of positive cases from NHS Test and Trace was up 27% in the week ending 12 August compared to the week ending 5 August, that certainly doesn’t necessarily mean that the number of infected people in England has gone up by that amount. It could have done so – but the trouble with these data is that the number of positive cases depends on who was tested. And it may well not depend just on the number of people tested, but on what kind of people they were. If, somehow, people who really have a Covid-19 infection have become more likely to come forward for testing in the latest week, that could put up the number of positive tests even if the number of people infected in the country as a whole hasn’t increased and even if (as did happen) the total number of people tested in Test and Trace went down slightly. I’m not saying that this is definitely what has happened – but it’s a possibility.
“It might help to look also at a different source of data on new infections – the Office for National Statistics (ONS) Infection Survey. That presents data each week from a survey of a representative sample of people in England, who are not chosen for testing because of any symptoms or other characteristics. So the Infection Survey data are not affected by people being tested specifically because they have symptoms, unlike the Test and Trace count of positive tests. Last week’s ONS infection survey report (14 August) concluded that the number of new infections per day was lowest at the end of June, rose slowly after that, but that there is evidence that the rise is levelling off. The number of positive cases from NHS Test and Trace tracks quite closely the rates of new infections estimated from a model in the ONS survey – lowest in late June, increasing after that. But there is statistical uncertainty in the ONS numbers, because they are based on samples, and there is also uncertainty in the Test and Trace numbers as a representation of new infections in the community, because they depend on who comes forward for testing and why. So the data are from both sources are consistent with there being a levelling off in the increase in new cases since the end of June. Really the trends are not very clear, though. We’ll maybe know more tomorrow, when the latest week’s results from the ONS survey are published.”
Dr Yuliya Kyrychko, Reader in Mathematics, University of Sussex, said:
“The latest figures show a significant growth in the number of new positive cases. Since the percentage of positive cases has increased by 27% compared to last week, while the number of tests has slightly reduced, this suggests that the increase in positive cases cannot be easily attributed to changes in testing, but rather suggests a genuine growth in the level of infection. Most worryingly, the increase in Pillar 1 cases, which are the ones done in hospitals, has increased by a staggering 34% compared to last week, indicating that the number of people requiring hospital visits/care workers has grown even more substantially.”
Dr Konstantin Blyuss, Reader in Mathematics, University of Sussex, said:
“Latest data suggest that as the number of people who have to be tested increases, this results in delaying test results, as shown by a substantial drop in test turnaround times since last week, with fewer than 15% of home tests being returned within 48 hours. This suggests that it is really important to ensure there is adequate capacity for quick testing to ensure small numbers of local cases do not turn into large outbreaks. With many European countries currently experiencing substantial growth in COVID-19 infections, and people travelling more widely, it is essential that the track-and-trace system is able to cope with the potential additional cases arising from people returning to the UK from abroad.”
Prof Paul Hunter, Professor in Medicine, The Norwich School of Medicine, University of East Anglia, said:
“The fact that positive diagnoses have risen at a time when the number of tests are remaining fairly static does suggest that the incidence of COVID-19 in the community is now beginning to rise again. The additional sharp rise in Pillar 1 tests being positive in particular supports this assumption. Because pillar 1 tests are done in NHS facilities they are more likely driven by clinical need than are pillar 2 tests and are a somewhat better indicator of the number of people becoming ill enough to seek medical attention.
“Clearly if this trend continues the demands on the test track and trace service and on the NHS will increase over coming weeks.”
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 member of the SMC Advisory Committee, but my quote above is in my capacity as a professional statistician.”
None others received.