Using Real-Time Google Search Interest as a Predictive Tool for COVID-19 Cases in the United States

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is an ongoing global pandemic with over 23 million associated cases and 800,000 associated deaths. There is a surplus of proposed predictive models (n>145) for COVID-19 that have emerged in academic literature; however, many of these predictive models have proven unreliable or biased.1 Several studies have looked at Google Trends data as a possible predictive tool in the last months.2-12 In this retrospective study, we looked at the predictive value of the Google Trends Tool as it applies to COVID-19 Cases and Reported Onset of Symptoms in the US. We looked back at Google Trends data for search interest of common COVID-19 search terms: "coronavirus" and "covid-19" from January 2020 through mid-June 2020 and compared that data to Centers for Disease Control (CDC) data on COVID-19 Cases Reported and Reported Date of Initial Onset of Symptoms.13 Google Trends is a free online tool that allows a user to quantify the search interest for a keyword or phrase over time.14 Significant strong positive correlation was found between CDC Reported Date of Initial Symptoms for Cases data and Search Interest for both terms "covid-19" and "coronavirus." Google Trends is a free and easy to access tool that may have utility as a predictive instrument with regards to the current COVID-19 pandemic. The Google Trends Tool may offer new insight and predictive value for medical decision making during current and future outbreaks in near-real time at a very granular level allowing states, cities and military bases to prepare.