COVID-19 Epidemiology and Google Searches

Abstract

We read “Increased internet search interest for GI symptoms may predict COVID-19 cases in US hotspots” by Ahmad et al with interest. The authors compared search volume for gastrointestinal (GI) symptoms with coronavirus disease 2019 (COVID-19) incidence in 15 states to observe that searches for the terms ageusia, loss of appetite, and diarrhea correlated with disease burden at 4 weeks. In our own analysis of Google Trends, we made similar observations with a few distinctions. We assessed COVID-19 plus diarrhea searches and United States COVID-19 epidemiology by using the Pearson correlation coefficient. We used Centers for Disease Control and Prevention (CDC) data for reported incidence and mortality (deaths per capita), cross-referencing with U.S. census data.Our findings suggest that diarrhea searches do not correlate well with disease burden; however, although analysis of online searches for GI symptoms and COVID-19 is not likely to be a good substitute for more traditional epidemiologic methods, search activity could still be useful as part of a more complex model. As you have concluded, Google Trends is a valuable tool, and it is our responsibility to carefully understand and refine its role in this global pandemic.