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 for deaths in England an Wales, including deaths from COVID-19.
Prof David Leon, Professor of Epidemiology, London School of Hygiene & Tropical Medicine, said:
“Today the Office for National Statistics published updated counts of weekly deaths from any cause and Covid-19 that were registered in England and Wales up until 23 May that occurred up to 15 May (week 20). The main report focuses on deaths by week according to date of registration of the death. From a scientific perspective it is more meaningful to look according to the week in which the death occurred which can be done using the data made available today.
“In the most recent week (ending 15 May) there 10684 deaths occurred of which 2637 had Covid mentioned on the death certificate (25%). These numbers are provisional and will be slight undercounts as some of the deaths that occurred in this most recent week will still not have been registered by 23 May.
“The total number of deaths occurring in care homes was 3264 of which 1127 mentioned Covid on the death certificate (35%). The equivalent numbers for deaths in people’s own homes were 2698 total death and 84 Covid deaths (3%). For hospitals there were a total of 4073 deaths of which 1368 were Covid (34%).
“These numbers confirm a steady downward trend in both deaths from all causes and from Covid that has been evident since the middle of April. This downward trend is seen in all settings (places of death). In care homes and hospitals total deaths in this most recent week ending 15 May have more than halved compared to their respective peaks (peak in care homes in week ending 17 April with 7376 deaths; peak in hospitals in week ending 10 April with 9159 deaths).
“For this most recent week ending 15 May, for England and Wales as a whole there were only 1146 more excess deaths relative to than would have been expected based on the 5-year average for 2015-19. This compares with a peak of 10,849 excess deaths for the week ending 10 April. This is a very large proportional fall of almost 90% and suggests that mortality rates in England and Wales are likely to reach normal pre-pandemic levels over the next few weeks on the assumption that there is no substantial upsurge in numbers of Covid cases.
“However, one important feature of mortality in England and Wales remains highly atypical even in this most recent week. This concerns where the deaths are occurring compared to earlier years. In the week ending 15 May there was a slight deficit of deaths (-412) occurring in hospitals compared to the weekly average for 2015-19, but an excess of deaths in care homes (1271) and in people’s own homes (435). These data strongly suggest that people who would in previous years been admitted to hospital in a serious condition and died there, are instead dying in their place of residence.
“Do these results suggest that people are dying because they do not get to hospital? Here the data are more difficult to interpret. However, if one takes away the number of Covid deaths from the total, there is in fact now emerging evidence that mortality rates from all causes other than Covid may already be at or even below the average from previous years. In care homes there was only an excess of 144 such non-Covid deaths occurring in the week ending 15 May. This number is likely to be a slight underestimate – but is far smaller than the 2903 excess non-Covid deaths in care homes at the peak of the epidemic.
“It remains striking that in the past 5 weeks (weeks 16-20) there has been a consistent weekly deficit of non-Covid deaths in hospitals ranging from -1800 to -2000. How far hospitals will return to “normal” in terms of numbers of deaths from conditions other than Covid is unclear.
“Overall these new data suggest that so called acute “collateral” deaths that are the result of failure to treat non-Covid conditions (e.g. from heart attacks) are no longer likely to be an important contributor to total mortality. However, these data cannot tell us about the future possible burden of excess deaths from conditions which may occur over the next few years due to disruption and delay to diagnosis and treatment of potentially life-threatening chronic conditions as in the case of come types of cancer.”
Prof Carl Heneghan, Director of the Centre for Evidence-Based Medicine, and Professor of Evidence-Based Medicine, University of Oxford, said:
“There were 43% (4,385) more deaths than we’d expect for this week of the year (5-year average 10,188).
“Some of the excess registration in week 20 are due to late registrations from the Bank Holiday weekend on Friday the 8th of May.
“In the first 20 weeks of this year, there have been 274,491 registered deaths – 21% (49,138) more than we’d expect when comparing to the 5-year average of 225,353 deaths.
“Analysis by 5-year age bands shows there have been no excess deaths in those under 35 years compared with the five-year average.
“In contrast, over 75s account for just over three-quarters of the excess deaths this year. The very elderly, over 90-year-olds, account for nearly one in 4 of the excess deaths.”
Prof Sir David Spiegelhalter, Chair of the Winton Centre for Risk and Evidence Communication, University of Cambridge, said:
“Over the seven weeks up to 15th May, as the NHS focussed on Covid, around 8,800 fewer non-Covid deaths than normal occurred in hospitals.
“These were not ‘exported’ to care homes, since fairly few care-home residents normally die in hospitals.
“Instead it appears that these contributed to the huge rise in deaths at home during this period, from the normal 16,300 to 28,600. Only 1,800 of these 12,300 extra deaths were labelled as Covid, leaving 10,500 excess non-Covid deaths.
“If up to 8,800 of these deaths would normally have occurred in hospitals, this would leave at least 1,700 unexplained non-Covid deaths at home.
“This also means that the 12,800 extra non-Covid deaths in care homes are not explained by deaths moving from hospitals.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“Looking at the broad picture from this week’s data, the general pattern from the past few weeks continues. There continue to be very encouraging signs – the total number of deaths in England and Wales in the latest week (ended 15 May) where COVID-19 was mentioned on the death certificate is again down on the previous week, as are the separate numbers in hospitals, care homes, hospices and at home.
“Some of the other details are not so encouraging, but this almost certainly has a lot to do with the Friday of the previous week, 8 May, being a bank holiday when very few death registrations occurred. Assuming that deaths, that would have been registered on that day if it hadn’t been a bank holiday, were instead registered in the following week, that could explain why the total number of deaths from all causes was up for the week ending 15 May compared to the week ending 8 May. The same was true for excess deaths (the difference between the number of deaths from all causes that week, and the average number in that week for the past 5 years). We certainly mustn’t take these increases to be caused by an official or unofficial loosening of lockdown measures. The figures released next week, which will include the data for the week ending 22 May, won’t be affected by that bank holiday and might give clearer indications, though one week’s data can never tell the whole story anyway.
“But with deaths from all causes still running at around 4,000 a week above the average weekly figure of around 10,000 for this time of year, we certainly aren’t out of the woods yet.”
All our previous output on this subject can be seen at this weblink: www.sciencemediacentre.org/tag/covid-19
Declared interests
Prof Kevin McConway: “Prof McConway is a member of the SMC Advisory Committee, but his quote above is in his capacity as a professional statistician.”
None others received.