Human behaviour in epidemics
Public health organisations traditionally monitor disease outbreaks—like influenza—through observing and testing patients in hospitals and GPs.
Unfortunately, people may be contagious before they feel sick or reach the hospital.
Digital online health surveillance systems like FluTracking attempt to help monitor health by providing a near real-time platform for self-reporting symptoms.
In this project, we examine user participation in FluTracking and use statistical methods to model and understand how users behave in this voluntary health surveillance system. We then find that these behaviours can substantially affect estimates of disease prevalence.