COVID-19 tool predicts chances of hospitalisation in people who are most vulnerable

The five percent of people in Britain predicted by a new tool to be at highest risk from Covid-19 accounted for three-quarters of deaths during the first wave of the pandemic, researchers reported Wednesday. As countries worldwide grapple with a second wave of disease, the risk-assessment method — which also predicts the chances of hospitalisation — could help identify the small percentage of the population most in need of being shielded from the virus, they reported in BMJ, a medical journal. “The tool provides nuanced information on people’s risk of serious illness due to Covid-19 and is designed for use by clinicians with patients to reach a shared understanding of risk,” the authors said in a statement. To develop the new application, called QCOVID, researchers from across Britain compiled data from six million patients, including age, height-weight ratio, ethnicity, and pre-existing conditions — such as high blood pressure and diabetes — known to increase the risk of serious outcomes after infection. They then tested the approach on 2.2 million patients — most of whom did not have Covid-19 — to see how well it predicted hospitalisation and deaths during two periods, late January to the end of April, and May 1 to June 30. More than three-quarters of those who died from the virus were in the top five percent of those predicted to be at maximum risk. While the tool effectively profiled those facing the worst odds, it did not identify which factors caused fatal outcomes, the researchers cautioned

A ‘striking tool’

It also only identified risk relative to other members of society but not the absolute risk of severe illness or death, which can change depending on infection rates and precautionary measures such as social distancing, mask-wearing and hand washing. Experts not involved in the research said it held real potential. “This provides support for the concept of targeted shielding,” said Mark Woolhouse, a professor of infectious disease epidemiology at the University of Edinburgh, describing the study as “a very striking result.” “If ways could be found to protect this five percent — such as regular testing of their closest contacts —- it could make a substantial difference to the public health burden of Covid-19.” But the model is only as good as the follow-up policies put in place, cautioned David Strain, a senior clinical lecturer at the University of Exeter. “Without a clear implementation plan this tool does not yet have a purpose,” he told the Science Media Centre in London. Strain also pointed out that the conditions in which the virus spreads continue to evolve. “These data are derived from hospitalisations and deaths during the first wave,” he noted. “Behaviour — and thus risk of transmission — has changed significantly during this time, some for the better (such as the widespread acceptance of masks) and some for the worse.” The model, for example, does not input information about self-isolation, whether someone is working in a high-risk job, or is in a neighbourhood with a high rate in infection. Aware of this problem, the authors said they designed the tool so that it can be updated as new data arrives. Since January, there have been about 750,000 cases of Covid-19 in Britain, and nearly 44,000 deaths. Worldwide, the virus has infected more than 40 million people, and killed more than 1.1 million.