Activity monitors could predict flu outbreaks up to three weeks earlier, study suggests

Sarah Knapton
Population changes in heart rate and sleep patterns were found to spike during flu outbreaks - Getty Images
Population changes in heart rate and sleep patterns were found to spike during flu outbreaks - Getty Images

Activity monitors can predict flu outbreaks up to three weeks earlier than current surveillance methods raising hopes that health services could be better prepared for upcoming waves of illness.

Researchers from Scripps Research Translational Institute in California, analysed sleep and heartbeat data from 47,249 Fitbit users between March 2016 to March 2018.

The data showed spikes in poor sleeping and faster heart rates in areas in the weeks before flu outbreaks were reported.

Usually the emergence of a flu outbreak takes between one and three weeks to spot, as the cases filter through from doctors to public health officials. 

But researchers say notifications could now happen in real-time, meaning that outbreak response measures could be quickly enacted, such as ensuring patients stay at home, wash hands, and deploying antivirals and vaccines.

Study author Dr Jennifer Radin, of Scripps said: “Responding more quickly to influenza outbreaks can prevent further spread and infection, and we were curious to see if sensor data could improve real-time surveillance at the state level. 

“We demonstrate the potential for metrics from wearable devices to enhance flu surveillance and consequently improve public health responses.” 

It is estimated that an average of 600 people a year die from the flu in Britain, although in some years, such as 2008-2009, it rose to 13,000.

Past studies using crowdsourced data, such as looking for reports of flu on Twitter, were not found to be accurate as it was difficult to separate out the activity of individuals with influenza from heightened awareness or related to media during flu season.

In the new study, researchers looked for periods when average resting heart rate was above the usual norm, while weekly average sleep had fallen significantly. 

The users were then arranged by which state they lived in, and the proportion of users above the threshold each week was calculated. 

That data was compared to weekly estimates for influenza-like illness rates reported by the U.S. Centers for Disease Control (CDC).

It is the first time heart rate trackers and sleep data have been used to predict flu, or any infectious disease, in real-time. 

And the researchers believe that with greater volumes of data it may be possible to refine the outbreak area to county or city-level.

Dr Cecile Viboud, of the US government’s National Institutes of Health, which funded the research. Said:  USA, says: “The ever expanding “big data” revolution offers unique opportunities to mine new data streams, identify epidemiologically-relevant patterns, and enrich infectious disease forecasts.” 

The research was published in The Lancet Digital Health