select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
The Office for National Statistics (ONS) have released provisional counts of the number of deaths registered in England and Wales in the week ending 20 November 2020.
Prof Sheila Bird, Formerly Programme Leader, MRC Biostatistics Unit, University of Cambridge, said:
“Provisional counts for COVID-mention deaths by date of occurrence are key data. They show a major November slow-down (the rate of increase has slowed), although not yet a major down-turn (when counts of COVID-mention deaths start to decrease) in England and Wales.
“My table, updated from last week, contrasts COVID-mention deaths in England & Wales (by week of occurrence) in early March to early May 2020 with early September to mid November 2020. The cumulative 2nd wave total of COVID-mention deaths that have occurred by 20th November is around 12,800 (to the nearest 100).
“Not all deaths which occurred in the week ending 20 November 2020 have yet been registered. Taking registration delay into account, we can expect that 2850 COVID-mention deaths may have occurred in England & Wales in the week ended 20 November 2020, see Table.
“This week gives further support for down-turn in COVID-mention deaths when registration delay is adjusted for. Happily, the most recent fortnightly multipliers in turquoise (2179/1263 = 1.73) and yellow (2590/1665 = 1.56) are decreasing. If the yellow multiplier continued to apply, we’d expect around 3400 COVID-mention to have occurred in the week ending 20 November. But, using registration-day as the basis for estimation, marvellously we are running substantially lower: 2850 (around the same number as occurred in the week ending 15 May, 11 weeks into Wave 1).”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“The latest ONS weekly data on death registrations in England and Wales doesn’t give us such positive news as we’ve recently had about numbers of coronavirus infections. For the latest week available, 14-20 November, deaths involving Covid-19 are up by about a tenth on the previous week. Excess deaths from all causes, which is the number by which that week’s total deaths are higher than the average form the previous five years, are also higher than in the previous week. I’ll be much happier when they stop rising – but the increases in the latest week are smaller than the increases in the previous week, which is something of a good sign. Also it takes some time – on average maybe three weeks – after a person is infected with the virus until they die, if, sadly, that happens. So the people whose deaths were registered in the most recent week available, 14-20 November, may have died before that week (because there can be delays in registering deaths), and in any case they would have generally been infected in late October or very early November. The levelling off or reduction in the numbers of people infected didn’t occur until about that time, or slightly later, so I’m not yet concerned that these figures on deaths are contradicting the positive news that we’ve seen recently in data on infections. That said, it’s worth noting that the number of deaths involving Covid-19 in the latest week, nearly 2,700 registrations, is at about the same level as in mid-May, when we were just beginning to have some slightly loosenings of the first lockdown. Things might well be beginning to turn round, but we’ve a long way to go still.
“The position isn’t exactly the same across the whole of England and Wales – excess deaths are higher in some regions than in others – but there were more deaths from all causes in the most recent week in every region of England and in Wales than the five-year average. In the hardest-hit regions – the North West and Yorkshire & The Humber, there were about 140 deaths in the latest week for every 100 deaths in that week averaged over the previous five years. The position wasn’t much better in the North East, East Midlands, West Midlands and Wales, where there were between about 125 and about 130 deaths in the latest week for every 100 in the five-year average.
“Those excess deaths figures are from deaths for all causes. Just on deaths that the ONS report as involving Covid-19 – these are not counted in the same way as deaths reported in the daily coronavirus statistics, but are deaths where Covid-19 is mentioned on the person’s death certificate, by the certifying doctor, as having contributed to their death. In some of these cases, Covid-19 would not have been regarded as the underlying cause of death, but in most of them, 88% of them this week, it is counted as the underlying cause. These are, very predominantly, not deaths where the person happened to have Covid-19 but their death was actually caused by something else.
“I have to mention yet again that the numbers of deaths at home are still running considerably above the average of the previous five years. Another 1,000 above the average this week. That’s about 40% of the five-year average, just as last week. In other words, for every 10 deaths at home that occurred in the corresponding week in the previous 5 years, there were 14 this year, in the week ending 20 November. Most of those deaths don’t have Covid-19 mentioned on the death certificate at all. ONS, and PHE, have done some previous analyses of excess deaths at home, and statistics of that sort can be helpful up to a point – but they can’t tell us what sort of a death these people had in their homes, or what the care was like at the end of their lives.”
Prof Sir David Spiegelhalter, Chair, Winton Centre for Risk and Evidence Communication, University of Cambridge, said:
“In the week ending November 20th, there were 12,535 deaths registered in England and Wales, compared to a five-year average of 10,380.
“This is an excess of 2,155, a 21% increase over the five-year average. And it is also substantially larger than the peak for this week over the last 10 years, which was 10,882 in 2019. This is far greater than could be explained by an ageing population.
“2,697 deaths had Covid on the death certificate, and it is important to note that these are not just deaths ‘with’ Covid. 88% of the Covid deaths, that is 2,361 people, were deaths ‘from’ Covid as the main underlying cause, only 12% were ‘with’ Covid as a secondary cause.
“In contrast, there were only 280 death with flu or pneumonia as the main cause of death, so Covid is currently causing more than 8 times the number of deaths than flu or pneumonia.
“It is encouraging that deaths that were not caused by Covid were slightly below the five-year average. We might expect some deaths that would normally occur now to have been brought forward by the first wave, but this still suggests that the collateral damage of the measures against the pandemic have not yet had an impact on overall mortality.
“Between September 5th and November 20th, 12,907 deaths involving Covid were registered in the UK, and there have been roughly 3,000 since then, making 16,000 altogether in the second wave. Sadly, the prediction that the second wave would involve tens of thousands of Covid deaths looks like it will be fulfilled, and we can expect this second-wave total to rise to over 20,000 by Christmas.
“Once again, there were over 1,000 additional deaths in private homes compared with normal, a 40% increase. This seems to be a long-term change in the way people are dying in this country, and deserves close attention.”
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 Trustee of the SMC and a member of the Advisory Committee, but my quote above is in my capacity as a professional statistician.”
Prof Sir David Spiegelhalter: “DJS is a paid non-executive director of the UK Statistics Authority, that oversees the work of the ONS.”
None others received.