Jul 22, 2019 | Updated: 09:15 AM EDT

Friends, Not Fitbit, More Predictive of Health, Says Studies

Jun 28, 2019 10:29 AM EDT

Friends, Not Fitbit, More Predictive of Health, Says Studies
(Photo : Photo by Barbara Johnston/University of Notre Dame.)

It is possible for people to assume about themselves with the use of wearable fitness trackers. Looking at their heart rate is easy to determine where they quite felt the stress of that presentation at work this morning, or think themselves healthier based on the number of steps they have taken by the end of the day.

However, according to a new study in the Public Library of Science journal, PLOS ONE, to get a better reading on the overall health and wellness of people, they would be better off looking at the strength and structure of their circle of friends.

Even those past research have revealed how beliefs, opinions, and attitudes spread throughout people's social networks, researchers at the University of Notre Dame were interested in what the structure says about the state of health, happiness, and stress.

Nitesh V. Chawla, Frank M. Freimann, Professor of Computer Science and Engineering at Notre Dame, Director of the Interdisciplinary Center for Network Science and Applications and the lead author of the study, said that they were interested in the topology of the social network, what does people's position within their social network predict about their health and well-being? What the researchers discovered was the social network structure provides a significant improvement in predictability of wellness states of an individual over using the data derived from wearables, like the number of steps or heart rate.

In their survey, the participants wore Fitbits to capture health behavior data including steps, sleep, heart rate, and activity level and completed studies and self-assessments about their feelings of stress, happiness, and positivity.

Then, Chawla and his team analyzed and modeled the data, using machine learning alongside the social network characteristics of an individual, including degree, centrality, clustering coefficient, and several triangles. These traits are indicative of properties like connectivity, social balance, reciprocity, and closeness within the social network. The research revealed a strong correlation between social network structures, heart rate, number of steps, and level of activity.

The structure of the social network provided significant improvement in predicting one's health and wellbeing compared to looking at health behavior data from the Fitbit alone. A simple instance is when the social network structure is combined with the data derived from wearables, the machine learning model achieved a 65 percent improvement in predicting happiness, 54 percent improvement in predicting one's self-assessed health prediction, 55 percent improvement in predicting positive attitude, and 38 percent improvement in predicting success.

Chawla noted that when they heard that health and wellness programs driven by wearables at places of employment are not working, they should be asking, is it because they are taking a single dimensional view where they give the employees the wearables and forget about it without taking the step to understand the role social network play in heal.

©2017 ScienceTimes.com All rights reserved. Do not reproduce without permission. The window to the world of science times.
Real Time Analytics