select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
The Department of Health and Social Care (DHSC) have announced that there are now 29,474 confirmed cases of COVID-19 in the UK, and there have been 2,352 deaths.
Prof Paul Hunter, Professor in Medicine, UEA, said:
“I think it is certainly too early to talk about green shoots. At the press briefing on the 31st of March data was shown to suggest that the outbreak was being brought under control. Much was made of the fact that since the high of 2885 new cases on the 27th March we had three days of somewhat reduced numbers, 2546, 2433 and 2619. It was not mentioned in the presentation that Michael Gove had just reported that days figures, 3009 new cases which were once again an increase on the previous highest daily report. Today there has been another very big increase in new cases up by 4324.
“So are there any green shoots remaining? Probably not, it is still too early to say with any confidence that the curve is now flattening out. It is never a good idea in any outbreak or epidemic to talk about a slowing in new cases just after a weekend when all sorts of other things conspire to reduce reporting. The number of reported cases will also be affected by our ability to consistently analyse enough samples. Changes in the number of samples done each day will also have an impact on the number of reported new cases.
“We may expect to see some signal by this time next week if the current social distancing measures are having an effect and sample testing is working well. Any impact of reported deaths are unlikely to be seen for at least another two to three weeks.”
Prof Sheila Bird, Formerly Programme Leader, MRC Biostatistics Unit, University of Cambridge, said:
Commenting on NHS England figures on deaths reported to 5pm on 31 March 2020 [based on 1 April 2020 release]:
“Day of death was specified for all 486 deaths reported to 5pm on 31st March, but only 96 (20%) of these deaths had occurred on 31st March 2020, whereas 30th (Monday: mode) and 29th March (Sunday) and 28th March (Saturday) accounted for 137 (28%); 103 (21%); and 44 (9%) of deaths respectively. The preceding Week 13 (21 – 27 March) accounted for a further 91 deaths (19%) with the remaining 15 late-reported deaths relating to Week 12 or earlier.
“The table below summarizes the nature of NHS Trusts’ somewhat-inevitable delay in the reporting of COVID-19 deaths. Reporting a delay of just 24 hours or even 3-days is clearly insufficient.
“When deaths are increasing, reporting delay means that we under-estimate the steepness. When death-counts begin to fall, we are slow to pick this up (and thereby sentenced to longer lockdown) because we have muddled in late reports of deaths which occurred on days when counts were still rising sharply.
“Statistical methods to adjust for reporting-delay need to be adopted once evidence has accrued about the stability of the reporting-delay distribution (by week, age-group and gender).
“Additional weekly-reporting by death-date (but a week in arrears) would enable weekly-totals to be almost-complete.”
Prof Eleanor Riley, Professor of Immunology and Infectious Disease, University of Edinburgh, said:
“There may indeed be some “green shoots” in there.
“Deaths were doubling at in less than 2 days, now it is edging towards 3 days.
“Cases were doubling in less than three days and now it is more than 3 days.
“But the trajectory of both is still up and at a steep rate.”
Prof Keith Neal, Emeritus Professor of the Epidemiology of Infectious Diseases, University of Nottingham, said:
“It is usual for the number of new cases to increase whilst on the epidemic curve and the rate of increase is slowing when looked at over the past week. We might expect to see an effect like this given the measures adopted one week ago (lockdown) and two weeks ago (work from home and isolate if you are ill or household contact is ill).
“One factor that needs to be clarified is if any of the new cases were healthcare workers which would increase the number of new infections but would not have been included previously causing an artificial increase in the epidemic curve. The government should clarify whether this is the case or not as they will be included eventually as testing is rolled out.
“It will take another 1-2 weeks before social distancing is having any effect on reducing the number of deaths as it reflects infections acquired 2-4 weeks ago.”
Prof Sheila Bird, Formerly Programme Leader, MRC Biostatistics Unit, University of Cambridge, said:
“Trusts’ reporting delay for COVID-19 deaths is more than a few days: as yesterday’s reporting-delay distribution showed. Hence, it is essential for England/UK to provide, ultimately, the correct count for COVID-19 deaths on 30 March 2020 by doing so about a week to 10 days in arrears.
“In the meantime, statistical science can help out: by making use of the reporting delay distribution – if stable – to estimate (with uncertainty) the expected number of COVID-19 deaths that will have occurred on 30th March 2020. Such an estimate could be provided about 3 to 4 days in arrears, and so on April 3rd (say) for COVID-19 deaths on 30th March 2020. Adjusting death-counts or case-counts for inevitable reporting-delays has an honourable history, not least during the AIDS epidemic in the 1980s and 1990s.”
* https://www.gov.uk/guidance/coronavirus-covid-19-information-for-the-public
All our previous output on this subject can be seen at this weblink:
www.sciencemediacentre.org/tag/covid-19/
Declared interests
None received.