In order to better prepare for the flu season, Google Flu Trends has been reconfigured to include data from the Centers for Disease Control and Prevention (CDC).
Google Flu Trends was designed to predict flu outbreaks in the United States and 29 other countries weeks before the CDC’s program, FluView. Since launching the Flu Trends website in 2008, Google has continued to improve the ways in which the program collects data. Flu Trends uses search information from millions of Google searches to predict when and where the flu will hit next.
In “The Parable of Google Flu: Traps in Big Data Analysis”, an article published in Science magazine, it was revealed that Flu Trends’ data collection methods for the 2012-2013 flu season were flawed. The article claimed that Flu Trend was overestimating doctor visits for the flu when compared to the reports from the CDC. After the article was published, several media reports questioned the reliability and relevancy of Flu Trends.
Google has plans to issue a technical paper in response, detailing how the CDC data integrates with Google’s data. Their system isn’t designed to be a replacement for traditional surveillance methods, but rather as a way to alert health professionals to outbreaks earlier, improving their odds against the flu.
The quantity of the data doesn’t mean that one can ignore issues with the information, its collection, and its interpretation. The biggest challenge that Google will face is collecting valid and reliable data that can be used for scientific analysis.
Google is committed to strengthening their role in the healthcare industry. In June, the company introduced GoogleFit, a health and fitness data tracker for Android mobile devices. Google Glass has also been used in the medical field, with physicians using the technology during surgeries and as a teaching aid in medical schools.
Author: Lauren Daniels