The 'Covid deaths' which are not caused by Covid

6 May 2021

2:46 AM

6 May 2021

2:46 AM

Registered Covid deaths fell to just one on Monday, leading many to comment that the epidemic in Britain is effectively over. One day’s statistics don’t mean an awful lot, especially over a bank holiday, but what about the wider picture? Over the UK as a whole, there have been 90 deaths over the past seven days, a fall of 41.2 per cent over the previous seven day period – although that, too, may be affected by the bank holiday.

A more in-depth analysis, offering more context – although a little out of date – is provided by the latest weekly analysis of deaths from all causes, published today by the Office for National Statistics and covering the week ending 23 April. It shows that in that week there were 9,941 deaths in England and Wales, 497 fewer than the previous week and 5.3 per cent lower than the five-year average. It was the seventh week in which deaths were lower than the five-week average, suggesting that that second wave of Covid-19 is now thoroughly well over – we are seeing fewer deaths than normal at this time of year, perhaps because some people who would otherwise have died now had already been carried off by Covid-19. Of those 9,941 deaths registered in the week to 23 April, 260 (2.6 per cent of the total) mentioned Covid-19 on the death certificate.

Confusingly, the government uses two different definitions of a Covid death. Public Health England (PHE) – which produces the figures quoted in the opening paragraph – defines a Covid death as one which occurs, from any cause, within 28 days of someone testing positive for Covid-19. The ONS, by contrast, defines a Covid death as one where Covid-19 or ‘novel coronavirus’ is mentioned anywhere on a death certificate. There are obvious weaknesses to both. In the case of the PHE figures, they include deaths which have nothing to do with Covid-19 – you could be run over by a bus and still be counted. As for the ONS figures, they could feasibly miss people who have never been tested for Covid (a large number of people who died in care homes early on in the pandemic never were tested). On the other hand, they include people who may really have died from other causes.

Of the 260 people whose death certificates mentioned Covid-19 in the week up to 23 April, only 176 (67.7 per cent) mentioned the disease as an underlying cause. In other words, in a third of ‘Covid deaths’, the real cause was something else; Covid was at most a contributory factor. The further the vaccination programme progresses, the fewer deaths we can expect to have Covid as underlying cause. This is because the vaccines have been found to be more effective at preventing serious illness and death than they are at preventing infection. As a result, we are bound to see more and more cases of people who died of some other disease while they were mildly infected with Covid-19.

But the figures which arguably give the best overall view of the lethality of the pandemic are those published by the Institute and Faculty of Actuaries as part of its Continuous Mortality Investigation. These look at mortality as a whole, and don’t just compare it with a raw, five-year average as the ONS does – they adjust for population growth and for an ageing population by looking at the standardised mortality rate. The latest figures, published today, show that cumulative mortality for 2021 up to 23 April is running just 2 per cent ahead of the average for the years 2011 to 2020.

We have become conditioned to hearing frightening daily death figures, which are often crudely converted into ‘Jumbo jet-loads’. Yet the wider picture is of overall mortality running only slightly ahead of a normal year.

Got something to add? Join the discussion and comment below.

Show comments