March 31, 2020

expert reaction to latest DH figures on COVID-19 cases and deaths in the UK

The Department of Health and Social Care (DHSC) have released the latest figures for COVID cases, 25,150 (up by 3,009 on yesterday), and deaths, 1789 (up by 381). 

Prof Sheila M. Bird, formerly Programme leader at MRC Biostatistics Unit, University of Cambridge, said:

“Day of death was specified for each of 334/367 deaths (91%) reported today by NHS Trusts in England. This reporting standard makes transparent the nature of NHS Trusts’ inevitable delay in the reporting of COVID-19 deaths.

“Of these 334 date-specified deaths, reported by 5pm on 30th March, only 39 (12%) had actually occurred on 30th March 2020 versus 40% on 29th March and 19% (62) on 28th March.

“In fact, Week 13 (21-27 March 2020) accounted for 98/334 (29%) of the COVID-19 deaths reported on 31st March by NHS England as having occurred by 5pm on 30th March 2020.

“The extent of reporting delay underscores the need for an additional weekly summary which reports by death-date (but a week in arrears).  This would enable NHS Trusts’ reporting of Week 13 COVID-19 deaths to be almost-complete in time for a weekly-summary presented on Monday, 6th April; and NHS Trusts could, more easily, be asked report also on the gender and birth-year of those who had died in Week 13. On Monday 13th April, we could expect a weekly-summary of COVID-19 deaths that had occurred in Week 14 (28 March-3 April).”

Prof Sir David Spiegelhalter, Chair, Winton Centre for Risk and Evidence Communication, University of Cambridge, said:

“The extreme day-to-day variation in reported Covid-19 deaths is far more than we would expect from chance variability, and must be due to reporting practices.  Some deaths occurred many days ago, and there seems to be fewer reported over the weekend.  So great caution is required in interpreting daily figures”

“Over the last week, reported deaths rose on average by an estimated 21% each day (uncertainty interval 6% to 38%), similar to Italy at the same stage of the epidemic around 2 weeks ago, which if continued would lead to the daily count doubling within 4 days.  Italy’s death rate appears to have now levelled off at around 900 a day, and we can hope that our death rate will also flatten off due to the measures started a few weeks ago.  It is a good sign that the number of new cases per day is now stable.  But this will be a very tough time.” 

Dr Simon Clarke, Associate Professor in Cellular Microbiology, University of Reading, said:

“The news of a further 381 deaths in hospitals, on the back of a lower than expected increase yesterday, is shocking but sadly unsurprising and we can expect to see the UK’s COVID19 death-toll to continue to head higher over coming weeks. There will be days when the figures are comparatively low, but there will also be days when we see distressingly large increases.  It’s therefore important to look at trends over a number of consecutive days, rather than draw conclusions from any single day’s figures.

“Yesterday, the Chief Scientist presented a number of charts showing the long-term path of the outbreak. While he cautioned against reading too much into the day-to-day figures, he did point out that the UK trajectory was similar to that seen in France, and some people took the low-ish numbers as providing some hope that the UK might be showing signs of successfully flattening the peak of cases.

“Today’s terrible uptick in deaths is a reminder that only time will tell if the measures the UK has taken is having enough of an effect to avoid the tragic scenes playing out in Italy and Spain. It also reminds us the importance of sticking to the advice on staying home and avoiding contact with others. We cannot afford to be complacent in the face of this coronavirus.”

Prof Keith Neal, Emeritus Professor of the Epidemiology of Infectious Diseases, University of Nottingham, said:

“The increase in new cases is in line with what might be expected, three days ago there were 2885 new cases.  The big uncertainty today is how many of the new diagnoses are health care workers given testing started yesterday. It is important to know how many of these are health care workers so as to interpret the trend of the previous numbers which were only hospitalised patients.”

Prof Jim Naismith FRS FMedSci, Rosalind Franklin Institute and Oxford University, said:

“We can’t forget that behind these numbers are awful personal losses. Some of NHS Trusts reporting deaths in today’s NHS press release are apparently counting deaths over several days, others Trusts for one day, some Trusts stop at the 29th other Trusts report deaths on the 30th. It is unclear how these numbers relate to the overall daily number from the UK government.  

“However, it does appear deaths from previous days are only now being reported, this will have artificially decreased the previous daily totals and have increased today’s totals. Scientists have consistently warned that we cannot judge our progress in curbing the epidemic by a single day’s reported number of deaths.

“Although priority must be clearly given to treating the sick, it would good for someone in government to spend some effort quite quickly ensuring consistent reporting standards for the number of deaths before they are made public. Simply releasing numbers with such apparent differences in reporting periods is not helpful. Fundamentally, we must remember that only by continuing to follow government guidance on social distancing will we decrease the growth in the number of new deaths and in time the number of new deaths. Sadly we have to accept that because deaths are very much a lagging indicator, we will not see the effect of the government’s most recent measures in the daily number of deaths for some time.”

Dr James Gill, Honorary Clinical Lecturer, Warwick Medical School, said:

“The UK and the world’s media have not previously been exposed to the level of granular data as is being made available during this Coronavirus outbreak. A note of caution should be sounded on putting excess weight on day to day variations, even if they are appearing to show an improving picture in terms of infection rate. We are beginning to see the suggestion that social isolation and distancing is having an effect on the new case data as has been shown by the ONS release earlier today. 

“It cannot be strongly emphasised enough that what matters from a data perspective is the overall trend, rather than the day to day figures. As heart breaking as it is to see the death toll of COVID19 infected patients rising, the numbers of new infections may be trending downwards. What we may be seeing is that as patients are testing positive for coronavirus, unfortunately some of those patients will deteriorate and possibly die. Those deaths will be displaced in time from that patient’s initial positive test result. As there is a small suggestion that over the last two days there has been a reduction in new cases, hopefully due to this time lag, we will similarly see a reduction in deaths. It must however, once again, be reiterated that a pattern of change over two days should not have undue significance placed on it, as we previously saw similar drop in new infections between the 21st and the 23rd of March. There is however no risk in keeping our fingers crossed.”

All our previous output on this subject can be seen at this weblink: www.sciencemediacentre.org/tag/covid-19/

Declared interests

None received.