Can Internet Searches Predict an Outbreak?
Aim: The aim of this literature review is to provide an introductory overview of the use of Google Trends for public health surveillance, its strengths and limitations, and areas for further research.
Methods: A literature search was conducted through PubMed to identify publications which used Google Trends data (GTD) or an existing GTD model as a source to investigate an infectious disease.
Results: Google Trends has been used to model the incidence of a range of diseases, including influenza, dengue fever, HIV, pertussis, and malaria, with varying degrees of accuracy. Models frequently correlate with reported case data at a negative time lag, providing warnings of outbreaks earlier than traditional systems. Case data can be incorporated into models for greater accuracy. News/media bias, a small population size, and various sociodemographic factors are recurring themes noted to reduce accuracy.
Discussion: With the potential to monitor disease incidence in real time and improve existing modelling solutions, Google Trends represents an exciting new frontier for epidemiology and public health. However, these tools should be positioned as an adjunct to traditional public health surveillance rather than a replacement. More real-world testing in diverse cultural settings is necessary to better understand its strengths and limitations across the digital divide.