Show Hide image

Up to one in 17 care home residents may have died since the Covid-19 outbreak started. Why?

After 20,836 people died in the five weeks to 17 April, care homes have become the new epicentre of the disease. 

As many as one in 17 care home residents may have died in just five weeks since the start of the Covid-19 outbreak. New data released on 28 April showed that the number of care home deaths has gone beyond “alarming” and is becoming a national scandal.

That was before figures released today (29 April) by Public Health England showed that 29.1 per cent of all care homes in England had experienced an outbreak of Covid-19 this month.

We need to be clear on what we do and don’t know. While there are a number of different official and semi-official counts of how many people have lost their lives because of Covid-19, none is perfect  or perfectly up to date. Counting Covid-19 deaths is a difficult task. It’s even more difficult when they occur in care homes, where many people are already frail and vulnerable, and have a number of serious existing medical conditions.

What we do know is how many people are dying in care homes overall and how many people we would normally expect to die based on past figures. A total of 7,316 people died in care homes in the week to 17 April. In a normal year, that figure would be 2,154.

In the five weeks to 17 April, 20,836 people died in care homes  9,679 more than we would expect for those weeks. Only around a third of those deaths were officially recorded as related to Covid-19.

The total population of care homes in England and Wales was estimated at 352,000 people in the 2011 Census. While that number is very likely to have risen since due to the UK’s growing and ageing population, if still current it would mean as many as one in 17 care home residents died in the five weeks after the Covid-19 outbreak started. In a normal year, the ratio would be nearer to one in 32. 

Why are only 3,093 care home deaths officially connected to Covid-19? The number is based on deaths registered in England and Wales where coronavirus was mentioned on the death certificate, even if it was not the primary cause of death. It also represents the number of deaths registered in a week, not the number of actual deaths. Registrations can happen a few days after a death has actually occurred, so the total number of deaths is likely to be higher. Put simply, there is a lag in the data.

That might explain part of the gap, but it is highly unlikely to explain it completely. There are two other possible reasons for it. Firstly, not all deaths caused by Covid-19 might be registered as such on the death certificate. Testing for the virus has been largely restricted to seriously ill patients in hospital for much of the pandemic in the UK; only now is testing being expanded to symptomatic pensioners, asymptomatic patients, and staff in care homes. Many doctors might be reluctant to commit mention of the disease to a death certificate without a test, or at least reasonable clinical evidence. Elderly patients with existing comorbidities might not have had obvious symptoms of the disease for long, if at all, before dying. That doesn’t mean that they didn’t have Covid-19 and that it didn’t contribute to their death. 

In such cases, coronavirus could be reported as pneumonia or other respiratory problems on death certificates. Data from Public Health England shows a steep increase in the number of acute respiratory outbreaks, with most of them happening in care homes.

Secondly, some people might have died because of Covid-19, but not necessarily with Covid-19. With emergency services stretched, care-home staffing levels sometimes reduced through sickness or self-isolation, and no precedent for how to handle such an outbreak, the pandemic may have incidentally contributed to more people dying from other conditions than would be typical. 

How many deaths reflect the first factor and how many the second? The question is politically as well as epidemiologically important. Here it would be useful to have better data to give a picture of what is happening inside care homes.

We can start with testimony from primary sources. Mary*, a care-home kitchen worker from Derby, thinks not enough is being done. “Measures were put in place too late,” she says, adding that workers don’t know who has and who doesn’t have Covid-19.

Dee*, a 40-year-old carer in Bath, says: “Like many people, I am frustrated that testing was not done sooner, especially if the government knew about Covid-19 in January or February.”

We do know that the number of deaths in care homes has risen each week since the start of the Covid-19 outbreak, and for many such homes are now the area of greatest concern. There were 3.4 times more care-home deaths than expected registered in the week to 17 April. By contrast, the mortality rate outside care homes was just 1.4 times higher than expected during the same week.

While care home residents represent only 0.6 per cent of the population of England and Wales, they account for more than a third (36 per cent) of all unexpected deaths since the outbreak began.

Vic Rayner, director of the National Care Forum, says care homes are “no where near a peak” of coronavirus deaths and that “the front line of this virus has moved into care homes”. There is some evidence he is right: excess deaths outside care homes rose to 56 per cent above “normal” levels at the start of April, and have since increased to 80 per cent. Excess deaths inside care homes were at 240 per cent above “normal” levels in the week to 17 April  and the lines are quickly diverging even further.

If we want to know exactly why so many people are dying in our care homes – and why care homes have so quickly become the focus of the outbreak in the UK – the expansion of the testing regime is a good place to start. There is no doubt, however, that it has come late in the day. 

As Anita Astle of the National Care Association says: “We tested ten people who were asymptomatic within our care home and we received one positive result. “Had we not done the test, we wouldn't have known that person had the virus until they showed symptoms. By that time, they probably would have infected other people.”

Nicu Calcea is a data journalist at New Statesman Media Group