Wearables may be the key to detecting the flu even before a patient begins to show symptoms, according to a new study in JAMA.
The small study zeroed in on two cohorts of participants that volunteered to be infected with the H1N1 virus and the rhinovirus. Researchers developed digital biomarker models that pulled wearable data for early detection of the viruses and severity prediction, which was aimed at the time frame after pathogen exposure and before symptoms developed.
The models developed included a swath of biometric data including heart rate, skin temperature, electrodermal activity and movement, however, it is important to note that the prediction models were different for the two cohorts.
Researchers found that not only was the wearable model able to detect the pre-symptomatic flu, the prediction model was able to distinguish between mild and moderate infection.
“This cohort study suggests that the use of a noninvasive, wrist-worn wearable device to predict an individual’s response to viral exposure prior to symptoms is feasible,” authors of the study wrote. “Harnessing this technology would support early interventions to limit presymptomatic spread of viral respiratory infections, which is timely in the era of COVID-19.”
Researchers found that the detection model for the H1N1 virus could distinguish between infections and noninfections with an accuracy of up to 92%, with 90% precision, 90% sensitivity and 93% specificity, within 24 hours after the inoculation.
The model for the rhinovirus could distinguish with an accuracy of 88%, with a 100% precision, 78% sensitivity and 100% specificity at the time of symptom onset, which was 36 hours after the inoculation.
At the 24-hour mark, the prediction model was able to distinguish between mild and moderate infections at an accuracy of 90% for the H1N1 virus, and 89% for the rhinovirus.
HOW THEY DID IT
The data was collected from 31 participants with H1N1 and 18 participants with the rhinovirus.
Data for the H1N1 group was collected from September of 2017 to May of 2018. The participants were all between the ages of 18 and 55, with a mean age of 36.2. Data from the rhinovirus group was collected from Sept. 14 to Sept. 21 of 2015. Participants in the rhinovirus group were all between the ages of 20 and 34, with a mean age of 22.
The study excluded individuals who were pregnant, were breastfeeding or who smoked. Participants with a history of chronic respiratory, allergy or other significant illness were also excluded.
In the H1N1 study, participants wore the E4 wearable from Empatica one day before and 11 days after the inoculation, according to the study. Participants in the rhinovirus group also wore the E4 wristband, only for four days before and five days after the inoculation.
This isn’t the first study to look at wearable data virus detection. A 2020 study published in The Lancet Digital Health found that resting heart rate and sleep duration data collected from Fitbit devices could help inform timely and accurate models of population-level flu trends.
In March 2020, the Scripps Research Translational Institute announced the launch of DETECT (Digital Engagement and Tracking for Early Control and Treatment), which was focused on combining heartrate, activity and sleep data from a range of wearable devices to the onset of disease.
Wearables may also be key to detecting lingering symptoms of COVID-19. Research published in JAMA Network Open out of the DETECT study found that wearable data could help the medical community understand the lasting impacts of the virus on health.