Nature: OpenSAFELY: What are the factors associated with COVID-19 deaths in 17 million patients?

The spread of COVID-19is rapidly affecting mortality rates worldwideThere is an unprecedented urgency to understand who is most likely to have serious consequences, and to do so, we need new ways to analyze large data sets in a timely mannerNHS staff created OpenSAFELY: a safe health analysis platform covering 40 per cent of patients in England, keeping patient data in an existing data centre of a major primary care electronic health record provider17,278,392 adults recorded primary health care records with 10,926 COVID-19-related deaths under false namesCOVID-19-related deaths are associated with male (risk ratio (HR) 1.59, 95% confidence interval (CI) 1.53-1.65), older and poor (both with strong gradients),diabetes, severe asthma, and various other medical conditionsthe risk of blacks and South Asians was higher than that of Caucasians (HR 1.48, 1.30-1.69 and 1.44, 1.32-1.58, respectively)therefore, the study quantified a range of clinical risk factors for COVID-19-related deaths in the largest cohort study to dateOpenSAFELY is rapidly increasing the number of patient records, and researchers say it will update and expand the results regularly