Navigating the Dilemma of Mining Medical Data for Insights

Big medical datasets pose a serious problem. Thousands of patients’ health records are an enormous risk to personal privacy. But they also contain an enormous opportunity – they could show us how to provide better treatments or more effective health policies.

A system called OpenSAFELY has been designed to solve this problem, with the help of a computer code “robot”.

Professor Ben Goldacre, director of the Bennett Institute for Applied Data Science at the University of Oxford, explains how it works. Presenter: Tim Harford Producer: Tom Colls Production co-ordinator: Janet Staples Sound mix: Hal Haines Editor: Charlotte McDonald