©
Shutterstock
Academics at the University of Oxford and the London School of Hygiene & Tropical Medicine (LSHTM), working on behalf of NHS England, and in partnership with TPP and NHSX, have analysed the pseudonymised health data of over 17.4 million UK adults to discover the key factors associated with death from COVID-19.
This is the largest study on COVID-19 conducted by any country to date, and therefore gives the strongest evidence on risk factors associated with COVID-19 death.
Compared to white people, people of Asian and Black ethnic origin were found to be at a higher risk of death. Contrary to prior speculation, this increased risk could only be partially explained by other features, such as pre-existing medical conditions or deprivation: this means further work must be done to fully understand why these patient groups are at such increased risk of death. Additionally, people from deprived social backgrounds were also found to be at a higher risk of death: again, this finding could not be explained by other risk factors.
Results confirmed that men are at increased risk from COVID-19 death, as well as people of older ages and those with uncontrolled diabetes. People with more severe asthma were also found to be at increased risk of death from COVID-19.
The study was done via the OpenSAFELY analytics platform, a new secure mechanism for electronic health records analysis in the NHS. The platform ‘plugs in’ to the current NHS records storage system to minimise the security risks associated with transferring and storing data elsewhere. Using OpenSAFELY, trusted analysts can run large-scale computation across pseudonymised patient records in the highly secure data centres where they are already stored for routine care. This delivers analyses quickly and safely while preserving patient privacy.
Professor Liam Smeeth, Professor of Clinical Epidemiology at LSHTM, NHS doctor and co-lead on the study, says: ‘We need highly accurate data on which patients are most at risk in order to manage the pandemic and improve patient care. The answers provided by this OpenSAFELY analysis are of crucial importance to countries around the world. For example, it is very concerning to see that the higher risks faced by people from BME backgrounds are not attributable to identifiable underlying health conditions’.
Dr Ben Goldacre, Director of the DataLab in the Nuffield Department of Primary Care Health Sciences at the University of Oxford, NHS doctor and co-lead on the study, says: ‘During a global health emergency we need answers quickly and accurately. That means we need very large, very current datasets. The UK has phenomenal coverage and quality of data. We owe it to patients to keep their data secure; and we owe it to the global community to make good use of this data. That’s why we have developed a new highly secure model, taking the analytics to where the data already resides.’
Further analyses using OpenSAFELY are already underway, including investigation into the effects of specific drugs routinely prescribed in primary care. The platform can also be used to evaluate COVID-19 spread with innovative approaches to modelling; predict local health service needs; assess the indirect health impacts of the pandemic; track the impact of national interventions; and inform exit from lockdown.