The Office for National Statistics (ONS) mortality figures for week 16 (11-17 April) were published this morning at 9:30am. They make for grim reading.
This is the peak of the coronavirus epidemic in the UK. (As the ONS figures use date of registration rather than date of death the peak is shifted by a few days – though the report does have a graph for actual day of death this time, which is an improvement on previous editions). The figures will be coming down from next week (18-24 April), though it appears that the care home death curve is running behind the hospital curve by several days, which will lengthen the peak a little.
It’s the worst week for deaths since comparable records began in 1993: 22,351 deaths, which is 11,854 more than the 5 year average (‘extra deaths’). Of these 8,758 or 74 per cent mention Covid-19 on the death certificate, meaning 3,096 or 26 per cent (over a quarter of the extra deaths) are non-Covid related. (Covid-19 could have been missed, but equally Covid-19 deaths could be with rather than from the virus.)
Less than a quarter of Covid deaths in week 16 (23 per cent or 2,050 deaths) occurred in care homes. This is interesting because reports from elsewhere in Europe suggest around 50 per cent of all Covid deaths occur in care homes. The reason the UK proportion is so much lower than elsewhere has not yet been explained as far as I know. Is it because Covid deaths in UK care homes aren’t being identified (58 per cent of the extra care home deaths – that is, 2,795 deaths – were from non-Covid causes, down from 67 per cent the previous week) or because in other countries care home deaths are being wrongly attributed to Covid rather than other causes, especially the impact of lockdown, isolation and change? The big question is what is killing the other several thousand extra care home residents in the UK each week?
In terms of cause of death more broadly, 37 per cent of week 16 Covid-19 deaths (3,220 of 8,758) mention pneumonia, leaving nearly two thirds as non-pneumonia. I previously suggested this might indicate a death with rather than from Covid-19, but reports since have suggested it might often be due to cytokine storm (a lethal immune system response), which if true is not good at all as it makes the disease more deadly to more people. England appears to be one of the worst affected countries, and unusually the virus is killing more younger people here than elsewhere. There is currently no accepted explanation for this. The Oxford Centre for Evidence Based Medicine (CEBM) have identified the UK, US, Italy, France, Spain and Belgium as the six worst affected countries, together accounting for three quarters of worldwide deaths so far. As possible explanations they are exploring environmental factors such as temperature and humidity.
Deaths registered with Covid-19 to 17 April total 19,112. Since this includes the peak we can estimate the total number by the end of the epidemic will be a bit less than double this, say 30,000-35,000. This is deaths with rather than from Covid-19. This is as bad as (or slightly worse than) the worst flu years such as 2015.
As upsetting and unpleasant as this is, a deadly epidemic at this scale does not justify the unprecedented move of shutting down the economy and confining people to their homes, which has huge costs of its own, including in lives. In terms of whether a worse second wave will hit if the lockdown is lifted, it is worth repeating that the peak in deaths on 8 April corresponds to a peak in infections three weeks earlier, which is several days prior to the lockdown coming into effect, suggesting the lockdown is not what brought the epidemic under control. Rather it is much more likely due to a combination of the lighter measures introduced earlier, increased public awareness and a pre-existing resistance in the population to the virus.