An investigation of a novel three-dimensional activity monitor to predict free-living energy expenditure.

Abstract

The aim of this study was to assess the capability of the 3dNXTM accelerometer to predict energy expenditure in two separate, free-living cohorts. Twenty-three adolescents and 14 young adults took a single dose of doubly labelled water and wore a 3dNXTM activity monitor during waking hours for a 10-day period while carrying out their normal routines. Multiple linear regression with backward elimination was used to establish the strength of the associations between various indices of energy expenditure, physical activity counts, and anthropometric variables. 3dNXTM output accounted for 27% and 35% of the variance in the total energy expenditure of the adolescent and young adult cohort, respectively. The explained variance increased to 78%, with a standard error of estimate of 7%, when 3dNXTM output was combined with body composition variables. The 3dNXTM accelerometer can be used to predict free-living daily energy expenditure with a standard error of estimate of 1.65 MJ in adolescents and 1.52 MJ in young adults. The inclusion of anthropometric variables reduces the error to approximately 1 MJ. Although it remains to cross-validate these models in other populations, early indications suggest that the 3dNXTM provides a useful method of predicting energy expenditure in free-living individuals.