Background: Accelerometers can provide accurate estimates of moderate-to-vigorous physical activity (MVPA). However, one of the limitations of these instruments is the inability to capture light activity within an acceptable range of error. The purpose of the present study was to determine the validity of different activity monitors for estimating energy expenditure (EE) of light intensity, semi-structured activities.
Methods: Forty healthy participants wore a SenseWear Pro3 Armband (SWA, v.6.1), the SenseWear Mini, the Actiheart, ActiGraph, and ActivPAL monitors, while being monitored with a portable indirect calorimetry (IC). Participants engaged in a variety of low intensity activities but no formalized scripts or protocols were used during these periods.
Results: The Mini and SWA overestimated total EE on average by 1.0% and 4.0%, respectively, while the AH, the GT3X, and the AP underestimated total EE on average by 7.8%, 25.5%, and 22.2%, respectively. The pattern-recognition monitors yielded non-significant differences in EE estimates during the semi-structured period (p = 0.66, p = 0.27, and p = 0.21 for the Mini, SWA, and AH, respectively).
Conclusions: The SenseWear Mini provided more accurate estimates of EE during light to moderate intensity semi-structured activities compared to other activity monitors. This monitor should be considered when there is interest in tracking low intensity activities in groups of individuals.