Ross Clark

28 July 2021

12:11 PM

How many people have been taken into hospital and are dying with Covid, and how many have been admitted to hospital and died because of the virus has been one of the fundamental questions of the pandemic. We have been bombarded with daily statistics which have never attempted to put this into context.

An overall Covid death toll of 129,000 – by the government’s count – means little without knowing who has been dying. Statistics relating to the age profiles of the dead and their medical histories have cropped up from time to time – showing that those who have died are, on average, over 80 and that many suffered from serious and often multiple pre-existing conditions. But these statistics have slipped out of the day-to-day conversation.

Today, however, comes leaked data which puts an interesting light on data for hospitalisations. Statistics from NHS trusts shows that only 44 percent of patients who have been counted as ‘Covid’ patients had tested positive for the infection before they were admitted to hospital. The other 56 per cent only tested positive after admission to hospital either in a PCR test to confirm suspected Covid, routine screening – or they were only diagnosed later in their hospital stay.

There is still a lot of context missing from the figures. They don’t tell us, for example, how many people had Covid as a contributing factor in their admission. And they don’t distinguish between a patient who fell off their bicycle and was found to be infected with Covid, or one who went into hospital with a lung condition which had been caused or made worse by previously undiagnosed Covid. Nevertheless, these figures do highlight the problem of continuously feeding the public a daily digest of statistics which have not been telling us the whole story.

With new Covid infections having fallen over the past week, a lot more people suddenly seem to want to question the daily figures. The same arguments that were used by those trying to allay fears about rising infections earlier in the year – that recorded infections are partly a product of the number of tests being performed, for example, and that the number of tests has fallen now schools have broken up – are now being employed by people who don’t seem to want to believe that cases are really falling.

One lesson for the next pandemic is not to roll out daily statistics of cases, hospitalisations and deaths without contextualising those figures. Knowing how many people are seriously ill with the disease, how many are dying from it, means nothing without knowing who those people are. Covid has killed in large numbers – but the act of simply feeding us a figure for people who have died while infected with a virus could equally be used to make out that an innocuous strain of the common cold was a deadly disease.