What if you could track people getting sick just by analyzing how they surf the Web?
Researchers from Google and the Centers for Disease Control and Prevention tried that back in 2009. They linked the number of Google searches for flu-like symptoms with the percentage of doctor's visits related to the flu. The findings suggested that search patterns alone could reveal how many people probably had or were about to get sick with flu.
From the work, Google Flu Trends was born. But critics quickly found that the estimates weren't as accurate as first thought. The algorithm underestimated the number of flu cases in 2009 and overestimated them in 2012. Google responded by tweaking its algorithm in 2009, 2013 and again in late October.
But the challenge to make online disease sleuthing more accurate continues. On Monday, researchers suggested Wikipedia searches might forecast the flu and a paper published Oct. 30 in the Royal Society of Open Science said combining Google Flu Trends with historic data on flu levels gives us the most accurate look yet.
Shots decided to take a look at a few of the public health issues we've followed in the past and what we've learned from their limitations.
Copyright 2022 NPR. To see more, visit https://www.npr.org. 9(MDAzMjM2NDYzMDEyMzc1Njk5NjAxNzY3OQ001))
