Advertisement

Almost one third of Covid deaths in July and August 'primarily caused by other conditions'

The Oxford study found people may have been included in the Covid-19 figures if doctors believed the virus might have exacerbated their condition - Justin Setterfield/Getty Images Europe
The Oxford study found people may have been included in the Covid-19 figures if doctors believed the virus might have exacerbated their condition - Justin Setterfield/Getty Images Europe
Coronavirus Article Bar with counter
Coronavirus Article Bar with counter

Coronavirus was not the main cause of death for nearly one third of recorded Covid-19 victims in July and August, research by Oxford University has found.

Analysis shows that around 30 percent of people included in the coronavirus death toll by the Office for National Statistics (ONS) over the summer months had died primarily from other conditions.

It means someone who suffered a heart attack, or even died in a road traffic accident, may have been included in the figures if they had also tested positive for coronavirus at some point, or if doctors believed the virus may have exacerbated their condition.

Throughout the entire pandemic, around one in 13 people currently classed as Covid-19 victims did not have the disease as an underlying cause of death.

It means 3,877 deaths (7.8 per cent) in which coronavirus was not the primary cause have been included in the figures. In July and August, that number jumped to 28.8 per cent of all registered deaths, meaning Covid-19 was not the main cause of death in 465 of 1,617 recorded victims (listen to the podcast below, which discusses whether Britain's death toll could be set to increase again).

Early in the pandemic, the World Health Organisation (WHO) warned that even if coronavirus appeared on the death certificate as a "significant condition", the death should not be included in the figures. WHO guidelines state that such deaths "are not due to Covid-19 and should not be classified as such".

Dr Jason Oke, part of the team from the Centre of Evidence-Based Medicine at Oxford that uncovered the data, said many people had been dying "with" coronavirus but not "from" it.

"At the beginning of the epidemic we only saw this in a few cases, but this is increasing because a lot more people have now had Covid," he said. "The true death rate is an important thing to know because it gives us an idea of impact. Clearly Covid was having a  massive impact in March and April, but we need to know if the disease is now taking the same toll as before.

"The impact now seems to be lessening, and if that is true – which it certainly looks like at the moment, because there doesn’t seem to be the same fatality rate (see graphic below, showing fatality rates across Europe) – then that will guide decisions in managing risk, so it's important to get this number right."

The team uncovered the figures after comparing the number of registered deaths where coronavirus was not the main cause to the ONS data.

During the pandemic, 263,826 deaths were recorded altogether in England and Wales and 218,143 were from causes other than coronavirus. That suggests that some 45,683 people died "from" coronavirus rather than the 49,560 figure published by the ONS.

Experts at Oxford are concerned that the problem with over-counting will get worse as more people in the population contract coronavirus. They are worried it means thousands more people died at the peak because of the pandemic response rather than because of the virus and have been wrongly included in the Covid-19 statistics.

"In the lockdown, there may have been even more deaths that were not caused by Covid, but were caused by the actions of lockdown – and that is important to know," added Dr Oke.

The team is the same one that discovered that Public Health England (PHE) had been wrongly counting people as having died from the virus if they had ever tested positive.

That led to Matt Hancock, the Health Secretary, ordering an urgent review of the data, and in August the UK death toll was revised downwards by 5,377 to 41,329 and now only include deaths that occur within 28 days of a positive test.

However, Paul Hunter, professor in medicine at the University of East Anglia, said it was very difficult for doctors to be able to tell how much of an impact Covid-19 had on an underlying condition.

"It's a difficult judgment call to make," he said. "Say, for example, you've got a patient who has got leukaemia, they get Covid, and a couple of weeks later they die.

"There is evidence they had got some degree of pneumonia, so what do you put as the primary cause of death? How much is because of Covid is not easy to say. The primary cause is leukemia, but they might not have died if they had not got Covid."