select search filters
briefings
roundups & rapid reactions
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 18 June 2021.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“This week’s ONS bulletin on provisional numbers of deaths registered in England and Wales takes the data up to the week 12-18 June. That’s well clear of the abnormal patterns of late registration that make the figure harder to interpret on or just after bank holidays, so we don’t have to take that into account again until the end of August. I haven’t seen anything particularly surprising in the new data, or indeed seriously concerning.
“One aspect that is unwelcome is that the number of registered deaths in England and Wales where Covid-19 is mentioned on the death certificate has risen slightly in the most recent week. There were 102 of these deaths involving Covid-19 that week, compared to 84 the previous week. It’s not surprising that the number of deaths involving Covid-19 has increased, given that all measures of new cases and infections have been rising quite markedly in England and Wales since early to mid May. Unfortunately, despite the success of vaccinations, if infections and cases go up, there will be some increase in deaths. I was a little surprised not to see an increase last week. However, what wasn’t clear from the previous data was how big the increase in deaths would be. Before vaccination, the pattern would have been clear. Vaccinations will mean that the rise in deaths after a rise in infections won’t be as great as it was. But vaccination isn’t perfect for several reasons, including that not everyone is vaccinated, so it’s more difficult to predict exactly how deaths will change. While, of course, any death is distressing for friends and family, I’m encouraged by the fact that the increase in deaths is not large, not yet anyway. Death registrations involving Covid-19 are above 100 again in the most recent week (though only just), after a run of three weeks where they were below 100. But for the vast majority of weeks since the pandemic began, weekly deaths involving Covid have been very far above 100. (Apart from the recent run of three weeks below 100, there were just another two weeks around the end of August last year, and the very first week when Covid-related deaths were registered, last March.) So the most recent week’s number is still small. Of the latest week’s 102 deaths involving Covid-19, 74 were coded as having Covid-19 as the underlying cause of death. For the others, Covid-19 would have contributed to the death (so they aren’t people that were killed by something entirely unrelated but had happened to have a positive test).
“Of course, we’ll have to continue to keep a sharp eye on numbers of deaths, but so far at least, vaccination does appear to be keeping increases in deaths in check (together with improvements in treatment of those who become seriously ill). And we shouldn’t forget that death isn’t the only possible really bad effect of infection – some people who are infected will end up with long Covid or other serious long-term health consequences.
“Another way of looking at the numbers of deaths involving Covid-19 is to compare them with the total number of deaths from all causes. Back in late January, almost half of all registered deaths (46% of them in the worst week) involved Covid-19. In the latest week it’s just 1.1% (or 11 in every thousand total deaths). That percentage is again a little higher than the week before, when it was 0.8% (8 in every thousand), but the increase is very small. As always, most of the Covid-related deaths are in older people, with 72 of the 102 in the latest week being in people who were 70 or older. This means that the percentage of the Covid-19 deaths which are of people aged 70 and over has increased compared to what it was in May, but it’s still a lot less than early this year when over 80% of Covid-related deaths were of people aged 70 and over. And when the total number of Covid-related deaths is as low as it currently is, you’d expect detailed patterns by age to vary quite a bit from week to week, for no particular reason.
“So far I’ve been talking about the figures for England and Wales taken together, but in fact all of the Covid-related deaths in the most recent week were in England. No deaths with Covid-19 on the certificate were registered in Wales at all in the latest week, and that’s the first time there have been zero Covid-related deaths registered in a week in Wales since the pandemic began. That’s excellent news.
“In my comment last week, I expressed some minor concern that the numbers of registered deaths from all causes looked rather high compared to the average of the five years 2015-19. Though that might well have been a blip connected with the late May bank holiday, this week the total registered deaths from all causes in England and Wales is again above the five-year average, though only very slightly. (There were 9,459 registrations, which is just 42 more than the five-year average.) Apart from a couple of anomalous weeks around bank holidays, all-cause deaths have been below the five-year average every week since early March, until the most recent week and the week before that. It’s difficult to know exactly why all-cause deaths are close to the national average now, because we don’t yet have detailed data on the cause of most of them. But a possibility is as follows. It seems likely that all-cause deaths fell below the 5-year average in the spring for a combination of reasons. Probably some people, who might otherwise have died then, had unfortunately died from Covid-19 earlier in the pandemic. Probably lives were saved because the lockdown drastically reduced the transmission of other respiratory infectious diseases, such as influenza, so that people were not dying from those causes. Probably there were other reasons too. But now we’re out of the seasons where, in years past, many people would have died from flu and other respiratory infections, and Covid-19 deaths are at a low level, so it seems as if the direct effects of Covid-19 and the indirect effects because of the lockdown are falling out of the picture. Maybe it’s even a sign of things returning to normal that total deaths are returning to something close to the pre-pandemic averages.
“What’s still showing no sign of returning to normal, though, is that deaths in people’s own private homes are still running at around 100 a day more than the five-year average. Very few of those deaths at home now involve Covid-19. If you take the Covid-19 deaths out of the picture, excess deaths at home from non-Covid-causes have been running at something like this 100 a day level for well over a year now. We still haven’t had a detailed explanation of exactly why this is happening.”
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 its Advisory Committee. I am also a member of the Public Data Advisory Group, which provides expert advice to the Cabinet Office on aspects of public understanding of data during the pandemic. My quote above is in my capacity as an independent professional statistician.”