Wearable measured physical activity profiles and variances in energy expenditure

Abstract

Physical activity is decreasing, and sedentary time is increasing in modern society causing health-, social-, and economic harm at both individual and community levels. The prevalence of diseases directly related to low physical activity is expected to increase significantly, and better means are needed for their prevention.

The amount of energy expenditure is associated with numerous health markers and can be accurately assessed using accelerometers. Different types of activities compete for limited time within a day, so it is essential to accurately identify the unique types of activities and their effects on energy expenditure. Individual effects on energy expenditure vary, so to achieve more personalized health benefits, valid empiric data on the variations in energy expenditure between individuals is needed. Mixed effect models are a robust way to examine data collected from natural settings due to their flexibility and capability to adjust for clustered and dependent observations. For this reason, a linear mixed model is used in this study to model the variations in energy expenditure and the physical activity-induced energy expenditure.

Unique types of physical activity were significantly associated with total daily energy expenditure and to variances in physical activity induced energy expenditure. Empirical data model showed that higher activity levels were associated with higher total daily energy expenditure and transformed-variable model showed that weight-bearing light physical activity was the most significant contributor on physical activity induced energy expenditure in working-age Finnish adults. The models fit well into the data, but a significant amount of variances in total daily energy expenditure was explained by baseline differences which were unable to be modeled with the data. Future studies should focus on the interactions between the unique types of physical activity and analyze physical activity induced energy expenditure using compositional methods for a clearer picture of physical activity behavior and its effects on energy expenditure.