Researchers at the University of Chicago created models to predict how flu spreads across the U.S. each year, using geographic movements, demographics and social interactions, according to a study published in eLife.
To create the models, the researchers used several large datasets, which listed healthcare visits, geographic movements and demographics for more than 150 million people over the course of nine years.
The study authors examined healthcare records from Truven MarketScan, a database that holds de-identified patient data from over 40 million U.S. families. They then analyzed the 2003-11 flu seasons, marking the insurance claims that indicated patients received treatment for flu-like symptoms.
Here are five things to know about the study.
1. The study found seasonal flu outbreaks originate in warm, humid areas in the south and southeastern parts of the country and migrate away from the coasts, moving northward.
2. Before the flu starts to migrate from U.S. coasts, the counties where flu outbreaks originated tended to be near the Gulf of Mexico or the Atlantic Ocean, the study found.
3. The researchers also examined 1.7 billion geo-located messages from Twitter over a 3.5-year period to capture how people traveled between counties on a week-by-week basis. For example, if a person regularly sends tweets from home, then tweets from work or while visiting family in the next county, the study authors determined this would create a movement pattern between those two counties.
4. The study authors found people in the southern part of the country have a high degree of social connectivity. In these areas, there are a greater number of tight-knit communities, which can increase the population's chances of spreading the flu.
5. These models allowed the researchers to recreate three years of historical flu data. As the first reports of the flu start coming each fall, these tools could be used as part of health officials' prevention efforts, said Andrey Rzhetsky, PhD, senior study author.
"For example, if flu-like symptoms are being reported in one county, you could tell people in neighboring counties to stay away from crowds, or you could focus vaccination efforts in certain places in advance," he said. "It could be used essentially as a weather forecast for the flu."