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 23 July 2021.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“For deaths involving Covid-19, the increase in registrations for the latest week (17-23 July) compared to the previous week isn’t surprising. There has been a lot of discussion about the reasons for the downturn in the daily counts of new confirmed cases (on the dashboard at coronavirus.data.gov.uk), but that did not start until about 20 July in England and Wales (in terms of the date when the person’s positive test was taken). There’s inevitably a delay between changes in new cases and changes in patterns of deaths, because if, sadly, a person dies from Covid-19, that won’t happen (on average) for two or three weeks after their symptoms begin. So the people whose deaths were registered in the most recent week in this ONS bulletin would have been infected well before confirmed cases began to fall. In Scotland, confirmed cases began to fall earlier than in the rest of the UK, at about the end of June. But falls in deaths from Covid-19, according to the dashboard, didn’t begin for about three weeks after that. If the same kind of pattern happens in England and Wales, we shouldn’t expect to see falls in the numbers of registered deaths in this ONS bulletin, where Covid-19 is mentioned on the death certificate, for at least another two weeks, and quite probably three. In fact, the increase in such deaths for the latest week compared to the week before is quite large – the number grew by half between those two weeks. There were increases between those two weeks in England and Wales considered separately, and indeed also in Scotland and Northern Ireland, and in every region of England except one. (There was no change in the North-East.) Another feature indicating that the increase is substantial is that the numbers of deaths classified by when the person died, rather than when their death was registered, increased in the latest week compared to the previous week in England, in Wales, and in every region of England. The number of deaths classified by date of occurrence is always incomplete for the latest week in these figures, because it can take time after someone dies for their death to be registered and coded, so the fact that there have been increases even allowing for the latest week’s counts being incomplete shows that the increases are real, rather than being caused by some sort of registration delays, and that they are pretty large. (Deaths classified by date of occurrence fell between those two weeks in Scotland and in Northern Ireland, though we’ll have to wait till later bulletins to see whether those decreases are real or are caused by incomplete counts for the latest week.)
“So the increase in deaths involving Covid-19 isn’t in itself too concerning, but some of the details of what has changed are more worrying, and I’ll say more about that below. In the wider context, if I look at other data too, I’m not (yet) hugely concerned, but the details do add to the uncertainty about how things are going to develop over the rest of the summer and into the autumn.
“Registered deaths involving Covid-19 increased most in the latest week, compared to the week before, in older groups of people. The total number of registrations in England and Wales went up by 50% in the latest week compared to the week before, but in those in their 70s it went up by 58%, and by 65% in people aged 80 and over. The bulletin doesn’t give any indication why this is so in terms of the vaccination status or pre-existing health conditions of the people who died. But in the most recent week, more than two thirds of registered deaths – 68% – were of people aged 70 and over. Back in late May and early June, when numbers of registered deaths involving Covid-19 were low, that proportion was considerably lower, between 50% and 60%. In the latest week, death registrations involving Covid-19 in England and Wales made up 3.4% of all deaths, and that percentage hasn’t been that high since mid-April. Of the deaths registered in England and Wales in the latest week where Covid-19 was mentioned on the death certificate, it was coded as the underlying cause of death in 85% of them. That’s a higher percentage than in most recent weeks – apart from one single week, it hasn’t been that high since February when total Covid-related deaths were running at around 3,000 a week rather than the current figure of 327.
“All these details are rather concerning, and would be more concerning if there weren’t signs from the dashboard figures that new infections are now decreasing, so that numbers of deaths might well also start decreasing soon. But those decreases in cases haven’t yet been confirmed by other data sources, such as the ONS Covid-19 Infection Survey. There are reasons why a decrease in new cases might not have shown up in the other sources yet, even if it is really happening, but it does all add to the uncertainty about how things are going to develop in the next few weeks. Personally I’m reasonably optimistic that things aren’t going to get too bad, but I really can’t be at all sure of that. As people keep saying, and as we absolutely mustn’t forget, Covid hasn’t gone away, and we’re not out of the woods yet.
“The figures on registrations of deaths from any cause for the most recent week haven’t changed much compared to the week before – up by 47, which is less than a 1% increase. But they do continue the pattern of the previous two weeks, in being above the average for the corresponding weeks in 2015-19. That comes after a long run since the start of March where they were below the five-year average, apart from a couple of weeks where registration figures were disrupted by bank holidays. In the latest week, death registrations in England and Wales from all causes were 7% higher than the five-year average. There hasn’t been an excess deaths figure that large since the week ending 26 February. Back then, the excess was because of Covid deaths (nearly 3,000 registered that week). But now, it largely isn’t directly because of Covid-19 deaths. There were 327 Covid-related deaths registered in the latest week, but the number of excess deaths was considerably higher at 651. Since we don’t yet know what actually caused most of those excess deaths, apart from Covid, it’s not yet possible to give an explanation.”
Prof Hannah Cloke, Professor of Hydrology, University of Reading, said:
“These latest data on excess mortality from the ONS for the week ending 23 July shows around 800 more deaths than we would expect at this time of year. Less than half of deaths registered suggest Covid-19 as a cause.
“While the true cause of these extra deaths will need to be investigated more closely, this was the week of the heatwave across a large part of the country, with maximum temperatures reaching the 30s and minimum temperatures not dipping much below the high teens.
“We know from past experience that in such hot conditions, vulnerable people can die. This data is the first indication that this is what happened during the July heatwave. The heat may have just killed the equivalent of a jumbo-jet full of people over the space of a week.
“This suggests that the Met Office was right to issue its first ever warning for heat risk. But it also suggests that telling people about the risk of heat is not enough. We need to urgently review the UK’s plans to cope with current and future heatwaves, or we will be watching more people die from heatwaves year after year. It would be unacceptable for society to accept this number of deaths from a flood or an accident. It is entirely unacceptable to accept this number of deaths from heat, especially when we know we are facing a growing problem in the future.”
Prof Paul Hunter, Professor in Medicine, The Norwich School of Medicine, University of East Anglia, said:
“The past three weeks of ONS death statistics show reported deaths exceeding the average for the same week for the years 2015 to 2019. In the period 3rd to 23rd July there were 1,667 excess deaths recorded of which only 728 (44%) were recorded as being due to COVID. So although COVID will have contributed to these excess deaths it cannot explain the majority.
“I suspect that some of these excess deaths will be related to the heat wave that we saw in the middle of the month.
“In a systematic review of the risk of death during heat wave https://jamanetwork-com.uea.idm.oclc.org/journals/jamainternalmedicine/fullarticle/413470 the biggest risk factors for death during a heat wave were being confined to bed, not leaving home daily, being unable to care for oneself and pre-existing psychiatric illness. Amongst a range of other factors that were associated with a reduced risk of death was increased social contact.
“The message here is that during a heatwave, even during COVID, it is important that we still visit our more vulnerable friends, relatives and neighbours because that is one of the ways that we can reduce deaths.”
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.”
None received.