Google searches ‘can predict Covid outbreaks two weeks before they happen’

A rise in people Googling terms such as “Covid symptoms”, “loss of smell” and “face mask” can predict coronavirus outbreaks more than two weeks before they happen, researchers have found.

Scientists at the Mayo Clinic in the US looked at what people were searching for in the weeks before the first coronavirus cases occurred in 50 states.

They found that searches for ten keywords and phrases relating to the pandemic rocketed before an outbreak occurred, and believe monitoring trends could give public health officials a head start in fighting the virus.

Traditional surveillance, including widespread testing and public health reporting, can lag behind the incidence of infectious disease. However, the web analysis picked up disease 16 days before the first cases were reported in some states.

“If you wait for the hotspots to emerge in the news media coverage, it will be too late to respond effectively,” said study leader Dr Mohamad Bydon,  a Mayo Clinic neurosurgeon.

“Our study demonstrates that there is information present in Google Trends that precedes outbreaks, and with predictive analysis, this data can be used for better allocating resources with regards to testing, personal protective equipment, medications and more.

“In terms of national preparedness, this is a great way of helping to understand where future hotspots will emerge.”

The team found that, as well as symptoms and masks, people also began searching for “Lysol”, a strong disinfectant, and Googling the location of coronavirus testing centres. Searchers were also keen to find out about tests and antibodies. 

Previous studies have found that social media sites such as Twitter are useful for picking up outbreaks of swine flu (H1N1), Mers (Middle East Respiratory Syndrome), measles and Sars (Severe Acute Respiratory Syndrome). The data can also help public health teams spot where disease is declining, the team believes.

The research was published in the journal Mayo Clinic Proceedings. 

Source Article