COVID-19 Pandemic Can Now Be Tracked Through Google Searches

Internet search terms related to COVID-19 have been "highly correlated" with cases and deaths linked to the novel coronavirus in the U.S., researchers say.

A new study—titled Trends and Prediction in Daily New Cases and Deaths of COVID-19 in the U.S.—builds on academic research conducted in China which suggested internet searches could be used as a model to track the scope of the pandemic.

In China, web search interest was determined to be correlated with daily incidence of COVID-19, but the theory had not yet been applied to the U.S.—until now.

Results appeared to be promising, the team said. "Search terms related to COVID-19 are highly correlated with the trends in daily new cases and new deaths of COVID-19 in the USA. Therefore, an internet search-interest based model may be used to predict development and peak-time of COVID-19 outbreak," the academic paper says.

The authors on the new study, available via Xia & He Publishing, extracted information about new infections and deaths in the U.S. from population data sets, including the dashboard maintained by researchers from Johns Hopkins University.

To judge internet search interest, some of the most common COVID-19-related terms were identified via Google Trends. They included "COVID-19," "COVID," "coronavirus," "pneumonia," "high temperature," "cough," "COVID heart," and "COVID diabetes."

The study, based on data from between March 1 to April 7, 2020, tracked if the search interest followed a similar pattern to the eventual infection and death statistics.

"Recent research in China has found compelling evidence that rises in daily Chinese COVID-19 cases and death rates can be predicted based on what people are searching for over the preceding couple of weeks," explained Graham Mills, a managing director of who holds a medical PhD from the University of Cambridge.

"The authors of this study wanted to understand if the same relationship between online searching and COVID-19 incidence existed in the U.S. Through this analysis, they found that it did. An increase in the Google searches of 'covid,' covid heart,' and 'covid pneumonia' for example were all followed within 12-13 days by a matched increase in the number of daily Covid-19 cases, and within 19-20 days by a matched increase in the daily COVID-19 death rate."

Ultimately, "COVID," "COVID pneumonia" and "COVID heart" were found to be the top three search terms in regards to a correlation with daily cases and new deaths.

But despite the correlations, the researchers found prediction accuracies of the search model were low—indicating it may not be of use as a long-term predictive tool.

"We found very high correlation in retrospective modelling but low accuracy in prediction, suggesting that the search-interest based model may be more helpful in predicting daily-incidence peak or early outbreak than post-peak or post-intervention trends," said the research paper, cited with seven separate authors.

"By finding as many leading indicators of how case and death rates will trend within a region, governments can best prepare and allocate resources for the scale of the problem they are likely to face over the coming months," Mills noted.

"The implication from this is that there exists a wealth of rapidly evolving information online, as simple as Google search trends, that can yield powerful and actionable insights to governments on how to best handle the COVID-19 pandemic."

He added: "The challenge, though, is that over the last three months the rate at which data has been shared on pandemic dynamics has accelerated exponentially."

In the U.S., COVID-19 cases continue to rise. At the time of writing, there has been at least 968,203 infections in the U.S., and the virus has been linked to over 54,931 deaths.

Google logo
The Google logo is seen on a phone in this photo illustration in Washington, DC, on July 10, 2019. ALASTAIR PIKE/AFP/Getty