The idea for an app to track the spread of COVID-19 came from professor Tim Spector, an epidemiologist at King's College London. He used his startup company Zoe Global Limited to build the Covid Symptom Tracker app in collaboration with King's College London, Guy's and St Thomas' Hospitals. Initially the project was UK-based, where there is open membership. In the US at a later date various cohorts from existing studies were added, including from Nurses' Health Study. This research was done in collaboration with Massachusetts General Hospital. The project website states that "The app is and always will be free, and any data you provide will not be used for commercial purposes." The tracker was released as a trial for 5,000 twins, using patients involved in other ZOE research projects. It was later expanded for use for non-twins. The app entered the UKApp Store and Play Store on 24 March, and the US App and Play stores on 29 March. Within 24 hours of being available in the UK, the app had been downloaded over 1 million times. As of May 2020, the app has been downloaded by over 3 million people, including 2 million Britons. By 17 July the number exceeded 4 million.
Research
The COVID Symptom Study requires users to reveal their location. Users answer personal information including age, gender and location, as well as report if they have any underlying chronic conditions. They also answer questions related to common COVID-19 symptoms, and input any illness or symptoms that they have, as well as stating whether they have been tested for COVID-19. The app then uses statistical analysis to predict whether the symptoms are consistent with COVID-19, and is designed to be able to distinguish between COVID-19 symptoms and symptoms of other seasonal flus. The app does not have any contact tracing functionality. Based on the data inputted into the app, the tracker suggests that 2 million people in the UK aged between 20 and 69 may have had COVID-19, and that as of May 2020, 280,000 people in that age range currently have symptoms consistent with COVID-19. The study also estimates the risk level to health workers, compared with the general public. Research based on the app was described in papers in Science on 5 May 2020 and in Nature on 11 May 2020.
