Validating a tool for estimating the risk of severe COVID-19 outcomes in Scottish adults

What is it about?

QCovid is a computer tool that was created to estimate a person’s risk of being hospitalised or dying due to catching coronavirus. It was made originally using data for adults in England aged 19-100 years. Updates to the original tool, named QCovid 2 and 3, added in coronavirus vaccination records. You can find out more about QCovid at https://www.qcovid.org/. Doctors record health data differently in England and Scotland. Hence, it is important to make sure that the tool works just as well in Scotland before we start to use it. In this study, we used the EAVE II national COVID-19 data platform, which contains health records for everyone in Scotland (5.4 million people). Certain information has been removed (like name and address) so that you can’t tell who specifically a record belongs to. Instead, it replaces each person’s name (and other identifiable information) with a list of random numbers and letters. That way, we can see that multiple records belong to the same person, but not who they are. Between December 2020 and June 2021, we found data for almost half a million adults with COVID-19. Overall, the tool was very good at predicting hospital admission and death in our data, However, it predicted both would be slightly more common than they were in in reality. In particular, the tool was slightly less accurate with older people.

Featured Image