Assessment of the impact of media coverage on COVID-19-related Google Trends data: infodemiology study

Sousa Pinto, Bernardo; Antó, Aram; Czarlewski, Wienia; Antó i Boqué, Josep Maria; Almeida Fonseca, João; Bousquet, Jean

doi:http://dx.doi.org/10.2196/19611

Assessment of the impact of media coverage on COVID-19-related Google Trends data: infodemiology study

License

Citation

Sousa-Pinto B, Anto A, Czarlewski W, Anto JM, Fonseca JA, Bousquet J. Assessment of the impact of media coverage on COVID-19-related Google Trends data: infodemiology study. J Med Internet Res. 2020; 22(8):e19611. DOI: 10.2196/19611

Abstract

Background: The influence of media coverage on web-based searches may hinder the role of Google Trends (GT) in monitoring coronavirus disease (COVID-19). Objective: The aim of this study was to assess whether COVID-19-related GT data, particularly those related to ageusia and anosmia, were primarily related to media coverage or to epidemic trends. Methods: We retrieved GT query data for searches on coronavirus, cough, anosmia, and ageusia and plotted them over a period of 5 years. In addition, we analyzed the trends of those queries for 17 countries throughout the year 2020 with a particular focus on the rises and peaks of the searches. For anosmia and ageusia, we assessed whether the respective GT data correlated with COVID-19 cases and deaths both throughout 2020 and specifically before March 16, 2020 (ie, the date when the media started reporting that these symptoms can be associated with COVID-19). Results: Over the last five years, peaks for coronavirus searches in GT were only observed during the winter of 2020. Rises and peaks in coronavirus searches appeared at similar times in the 17 different assessed countries irrespective of their epidemic situations. In 15 of these countries, rises in anosmia and ageusia searches occurred in the same week or 1 week after they were identified in the media as symptoms of COVID-19. When data prior to March 16, 2020 were analyzed, anosmia and ageusia GT data were found to have variable correlations with COVID-19 cases and deaths in the different countries. Conclusions: Our results indicate that COVID-19-related GT data are more closely related to media coverage than to epidemic trends.