ROCHESTER, Minnesota — When we're experiencing symptoms, one of the first things we might do is Google it.
A new study from Mayo Clinic found there's a strong correlation between keyword searches and COVID-19 outbreaks in the United States.
"Once we know that a certain area is a hot spot, in some ways — unless that area is very well-equipped — it might be too late," said Dr. Mohamad Bydon, a Mayo Clinic neurosurgeon and principal investigator at Mayo's Neuro-Informatics Laboratory.
The Neuro-Informatics team studied Google Trends data from January to March.
Researchers found strong correlations between keyword searches on Google Trends and COVID-19 outbreaks in parts of the country, according to the study. These connections were observed up to 16 days before the first reported cases in some states.
As the pandemic progressed, search terms evolved. For example, "loss of smell" eventually became a popular search.
"If we can better predict future hot spots, future areas of emergence, we can better direct personal protective equipment, testing resources, other health-related ... resources to those areas," Dr. Bydon said.
Dr. Bydon said they did not find it less predictive when looking at older populations.
"We did look at some of the states that have older populations... and areas that have older populations, specifically. We did not find that there was less of a predictive value to the online search tools in those areas. So even persons of older ages are also using these same online search tools and have wide access to them," Dr. Bydon said.
But using big data to predict disease outbreaks is not without criticism.
Google Flu Trends, which started in 2008 and stopped in 2015, was an attempt to identify flu outbreaks by tracking people's Google searches. According to an article published in Science in 2014, GFT ended up overestimating flu cases.
Dr. Bydon said he believes Google Trends is one tool of many that experts can use to help predict future COVID-19 outbreaks. According to the study, a combination of internet surveillance methods and traditional methods is likely the key to effective surveillance.
"In the past, attempts to use it have come up against the fact that the world was not quite as online and virtual as it is today," Dr. Bydon said.
Mayo Clinic recently introduced an interactive COVID-19 tracking tool that Dr. Bydon said utilizes online and social media data.
"This will be one of a multitude of things that get utilized in order to better track and trend both coronavirus and potentially future disease states," he said.