The increasing availability and use of low-cost consumer-grade wearables has led to a huge surge in research from both academics and industry exploring how these devices can assist in healthcare and research.
Now a team of researchers from Singapore have demonstrated that wearable devices can not only identify groups of people with similar patterns of daily activity, but are also able to predict a variety of health markets – including those related to the risk of heart disease like obesity, high blood pressure and high blood sugar.
Writing in PLoS Biology, team led by Weng Khong Lim from SingHealth Duke-NUS Institute of Precision Medicine, report on ‘comprehensively studied’ 233 normal volunteers by integrating data from wearable sensors with lifestyle questionnaires, cardiac imaging, sphingolipid profiling, and clinical measurements of various heart and metabolic disease markers.
“Apart from showing that wearable sensors can be used to identify groups of volunteers with distinct behavioural and demographic characteristics, we showed that resting heart rate (RHR) from wearables performed better than step counts in predicting heart and metabolic disease risk markers,” wrote the team.
Notably, the authors also found wearable activity data can identify active individuals at increased risk of having enlarged hearts – a condition also known as "athlete's heart" that is commonly thought only to affect competitive athletes.
Wearable insights
According to the researchers, the people in the study could be stratified into distinct clusters based on daily activity patterns.
Furthermore, these clusters were marked by distinct demographic and behavioural patterns, they said.
“While resting heart rates (RHRs) performed better than step counts in being associated with cardiovascular and metabolic disease markers, step counts identified relationships between physical activity and cardiac remodelling, suggesting that wearable data may play a role in reducing over-diagnosis of cardiac hypertrophy or dilatation in active individuals,” the team wrote.
"An enlarged left ventricle could be caused by heart disease or harmless adaptation to sustained exercise, and these two conditions share overlapping features,” noted senior author Professor Stuart Cook. “Activity data from wearables may help us identify individuals more likely to have this condition due to exercise, and are therefore at risk of misdiagnosis in the clinic.”
Furthermore, that activity data is predictive of circulating levels of a class of lipids known as ceramides, which have been associated with obesity, diabetes and heart disease.
"Compared to their more sedentary counterparts, active volunteers had lower levels of circulating ceramides,” commented another senior author Associate Professor Khung Keong Yeo. “In the past, researchers studying the interaction between lifestyle and lipid metabolism would have relied on questionnaires or expensive experimental studies."
Source: PLoS Biology
Volume 16, Issue 2, Open Access, doi: 10.1371/journal.pbio.2004285
“Beyond fitness tracking: The use of consumer-grade wearable data from normal volunteers in cardiovascular and lipidomics research”
Authors: Weng Khong Lim, et al