Cross-generational comparability of hip- and wrist-worn ActiGraph GT3X+, wGT3X-BT, and GT9X accelerometers during free-living in adults

ActiGraph accelerometers are frequently used to characterize physical activity, but free-living cross-generational comparability of newer models has not been verified. Participants (N = 70) wore GT9X and wGT3X-BT accelerometers at the hip and a sub-sample (n = 54) wore GT9X and either wGT3X-BT or GT...

Full description

Saved in:
Bibliographic Details
Published in:Journal of sports sciences 2020-12, Vol.38 (24), p.2794-2802
Main Authors: Clevenger, Kimberly A., Pfeiffer, Karin A., Montoye, Alexander H. K.
Format: Article
Language:English
Subjects:
Accelerometers
Actigraphy - instrumentation
Actigraphy - statistics & numerical data
Adult
ENMO
Equipment Design
Exercise
Fitness Trackers
Functional Laterality
Hip
Humans
MAD
measurement
monitor
Physical activity
Sedentary Behavior
Time Factors
Wearable Electronic Devices
Wrist
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:ActiGraph accelerometers are frequently used to characterize physical activity, but free-living cross-generational comparability of newer models has not been verified. Participants (N = 70) wore GT9X and wGT3X-BT accelerometers at the hip and a sub-sample (n = 54) wore GT9X and either wGT3X-BT or GT3X+ monitors at each wrist for 4 days. Vector magnitude (VM) counts, VM acceleration, Mean Amplitude Deviation (MAD), and Euclidean Norm Minus One (ENMO) were calculated (60-s epoch), and cut-points were used to determine percent of time spent in each intensity (sedentary/light, moderate, vigorous). Epoch-level correlation coefficients (r) were ≥0.73, and weighted kappa for intensity classifications ranged from 0.71 (ENMO, hip) to 0.98 (VM counts, non-dominant wrist). Monitors were equivalent for all outcomes, except ENMO (all locations/monitors), percent of time spent in sedentary/light (hip) and moderate (hip and non-dominant wrist) activity as classified by ENMO-based cut-points, and vigorous activity as classified by VM count cut-points (non-dominant wrist; p > 0.05). While epoch-level data were not identical, most outcomes were strongly related between models (e.g., MAD, VM) and equivalent once reduced to percent of time spent in each intensity. However, monitor output was not equivalent for the acceleration-based metric ENMO, suggesting that caution should be exercised when comparing this outcome among ActiGraph models.
ISSN:0264-0414
1466-447X
DOI:10.1080/02640414.2020.1801320