Data collected by healthcare companies could help make flu forecasts more accurate, according to new research from the University of Texas at Austin and published in the journal PLOS Computational Biology.
Lauren Ancel Meyers, PhD, a professor of integrative biology, and postdoctoral researcher Zeynep Ertem, PhD, evaluated multiple datasets to determine which were the most predictive of flu incidence. Their goal was to create more accurate forecasts for when the next flu season would peak, how long it would last and how many people would get sick.
The researchers reviewed more than 600 flu-related datasets, and said the best predictions came from the athenahealth EHR. According to their analysis, information from athenahealth — such as as how many patients receive flu vaccinations, positive flu test results and flu-related prescriptions — combined with traditional surveillance data from the CDC improved flu forecasts by 15 percent.
Although the researchers were granted access to athenahealth's data, most researchers and public health agencies are unable to access similar information from healthcare companies on an ongoing basis. However, athenahealth does a form of flu tracking using data from its network. Athenahealth's flu dashboard can be found here.
"The message is that we should think more systematically about the data that fuel our disease forecasts," Dr. Meyers told ScienceDaily. "With powerful — and sometimes surprising — combinations of data, we can make earlier and more accurate predictions about emerging threats."