Online searches have been used to study different health-related behaviours including identifying disease outbreaks. However, as several reasons can motivate individuals to seek online information, studying which online behaviours reflect real-life disease can help us improve current systems and implement timely public health measures. This is especially important during a pandemic such as COVID-19, making it a natural laboratory to study this subject.

Using online (search trends, Twitter posts) and offline (surveys, media reports, number of infected people) data, we are developing a model to identify different offline events that can explain the observed online behaviour(s). This will allow us to nowcast future waves of COVID-19 infection and offer a methodology that can be applied to other outbreaks.

This project is being developed in collaboration with Carlota Louro and received funding through an FCT PhD fellowship.