Google Searches Predict Flu Outbreaks?

Flu early warning system from Google

/ Author:  / Reviewed by: Joseph V. Madia, MD

(RxWiki News) If you feel like you're coming down with something like flu, what source of information do you turn to first? For many people, the answer is Google.

Researchers at Johns Hopkins have found that data from Google Trends on searches for “flu” match up with the numbers of people coming into hospital emergency rooms with flu-like symptoms.

They say that Google Trends could potentially be a powerful tool to warn hospitals to prepare for an increase in flu patients.

"Google searches can predict health patterns."

The study was led by Dr. Richard Rothman, an emergency medicine physician and researcher at the Johns Hopkins University School of Medicine in Baltimore.

Dr. Rothman's research team tracked Google Flu Trends in Baltimore starting in January 2009 and ending in October 2010. Google Trends is an internet search tool that shows how popular a search term is, relative to other searches on Google.

They found that spikes in searches for flu in Baltimore correlated with spikes in hospital ER visits for patients complaining of flu-like symptoms.

The study authors say these strong results mean that Google Flu Trends is potentially a real-time model for predicting flu outbreaks.

Such a model could provide an early warning system for hospitals and health care workers, who could use the advance notice to increase the numbers of staff and beds on hand to treat flu patients.

Currently, hospitals rely on case reports created by the US Centers for Disease Control and Prevention during flu season. But these reports are often weeks old by the time they reach healthcare providers.

That's why there's so much potential for Google Flu Trends, the study authors say. They provide information on a daily basis on a city-level as a flu spreads in real time.

The study was published in the January 9, 2012 issue of Clinical Infectious Diseases.

Reviewed by: 
Review Date: 
January 11, 2012
Last Updated:
October 8, 2012