Real-time flu prediction may be possible using wearable devices - expert comment

A new study published in The Lancet Digital Health has found that data from wearable devices on heart rate and sleep can predict flu in real-time better than current surveillance methods.
Digitally colourized TEM image of H1N1 influenza particles. Credit: NIAID

The research used anonymous data collected from wearable devices, such as Fitbits and other smartwatches and fitness trackers, to retrospectively identify weeks with elevated resting heart rate and changes to routine sleep. Resting heart rate tends to increase during infectious episodes, allowing tracking and prediction of influenza-like illness.

Dr Rosalind Eggo is Assistant Professor in Public Health Epidemiology at the London School of Hygiene & Tropical Medicine, working on infectious disease modelling to predict how diseases like influenza, Ebola and zika spread through populations and how they can be controlled.

Commenting on this study, Dr Eggo said:

“Wearable technologies have promise to provide new data sources for tracking infectious diseases such as influenza. This study is an example of an interesting use of this data for flu surveillance.

“However, it is critical to fully understand the strengths and weaknesses of these types of data before relying on them for monitoring infectious diseases. Further analysis is required to gauge how reliable these data are over time, how specific these measurements are for flu, and how representative Fitbit users are of the whole population.”

In December, Dr Eggo was invited to take part in The Royal Institution Christmas Lectures, where she explained how infectious diseases spread through a population, and how this spread can be slowed or halted completely through vaccination. Watch her talk from 22:05 here.

Past studies using crowdsourced data, such as Google Flu Trends and Twitter have found variable success on their own as these methods tend to overestimate rates during epidemics.

According to The Lancet Digital Health, further prospective studies will need to be done to help differentiate between infectious versus non-infectious forecasting.



Jennifer M Radin, Nathan E Wineinger, Eric J Topol, Steven R Steinhubl. The Lancet Digital Health. Harnessing wearable device data to improve state-level real-time surveillance of influenza-like illness in the USA: a population-based study. DOI: 10.1016/S2589-7500

LSHTM Scholarship Fund

Every year, LSHTM donors, alumni and friends donate to help widen access to education through the provision of scholarships.

Scholarships are transformational, have the capacity to change the lives of the students who receive them, and alleviate the financial burden of study for talented and motivated recipients.