Two papers just published in the medical journal The BMJ provide new insights into the risk of dying from COVID-19.
In the first, a leading statistician reports that deaths from the virus show an association with age that fairly closely matches the “normal” age-related risk of death from all other causes.
In the second, researchers describe a possible prediction model developed after identifying four distinct risk groups based on data from patients admitted to hospital in Britain.
For his study, David Spiegelhalter from the Winton Centre for Risk and Evidence Communication in Cambridge analysed death certificate data for England and Wales between 7 March and 26 June.
For the general population aged over 55, he calculated that the risk of catching and then dying from COVID-19 during this time was equivalent to experiencing around five weeks extra risk above the “normal” annual risk of death.
However, this risk decreased steadily with age, corresponding to just two additional days above the “normal” annual risk for school children.
The death rate over the 16-week period was around 12-13% higher for each year older, corresponding to doubling for every five to six additional years of age, and this relation is consistent from childhood to old age.
Spiegelhalter points out that his analysis refers to averages over populations, and although age seems to be the overwhelmingly dominant influence on mortality, clearly other factors, such as pre-existing medical conditions, affect individual risk.
He also stresses that these are observed historical rates in the population and cannot be quoted as the future risks of getting COVID-19 and dying.
Nevertheless, he suggests that “[n]ormal risk appears a reasonable comparator for interpreting both population and infection fatality risks”.
The “4C Mortality Score”, described in the second paper, was created by the ISARIC Coronavirus Clinical Characterisation Consortium, involving researchers from Imperial College London and the Universities of Edinburgh, Glasgow and Liverpool.
They collected data from 35,463 adults (median age 74) with COVID-19 admitted to 260 hospitals across England, Scotland and Wales between 6 February and 20 May 2020. Measures included comorbidities, respiratory rate, blood oxygen concentration, level of consciousness, urea, and C-reactive protein, a chemical linked to inflammation.
Data were entered into the model to give a score ranging from 0-21 points. They found that patients with a score of 15 or more had a 62% mortality compared with 1% mortality for a score of 3 or less, which could provide a guide to treatment strategies.
Four risk groups were created. One in every hundred patients in the low-risk group was found to be at risk of dying. It was 10 in a hundred patients in the intermediate-risk group, 31 in a hundred in the high-risk group and 62 in a hundred in the very high-risk group.
To validate the model, they tested it on a further 22,361 patients admitted to the same hospitals between 21 May and 29 June 2020 and found similar score performance, even after taking account of other potentially important factors.
Finally, they compared the model with existing risk scores and found that it demonstrated high discrimination for mortality with excellent calibration.
The authors do note some limitations to the observational study, and suggest that the approach should be further validated to determine its applicability in other populations.