Abstract
This study examined the compliance of children wearing wrist- and hip-mounted ActiGraph GT3X+ accelerometers and compared estimates of sedentary behaviour (SB) and physical activity (PA).
One hundred and eighty-eight 9-12-year-old children wore a wrist- and hip-mounted accelerometer for 7 days. Data were available for 160 (hip) and 161 (wrist) participants. Compliance was analysed using 9 different wear-time criteria. Output comparisons were compared against one wear-time criteria (any 4 days at 10-hours per day). Time spent in SB and PA was calculated using R-package GGIR.
Wear-time for the wrist (15.6 to 17.4 h.d−1) was significantly greater than the hip (15.2 to 16.8 h.d−1) across all wear-time criteria (all P < 0.05). Moderate-strong associations were found between time spent in SB (r = 0.39), LPA (r = 0.33), MPA (r = 0.99), VPA (r = 0.82) and MVPA (r = 0.81) between the two device placements (All P < 0.001). The wrist device detected more minutes in LPA, MPA, VPA and MVPA whereas the hip detected more minutes in SB (all P = 0.001). Estimates of time in SB and all activity outcomes from the wrist and hip lacked equivalence.
The GT3X+ when worn at the wrist promotes greater compliance than at the hip. Minutes in SB and PA calculated from raw accelerations at the hip and wrist provide contrasting estimates and cannot be directly compared. Further calibration and correction factors may facilitate the comparison of findings in studies that estimate time spent in SB and PA intensities captured from the wrist and hip.
One hundred and eighty-eight 9-12-year-old children wore a wrist- and hip-mounted accelerometer for 7 days. Data were available for 160 (hip) and 161 (wrist) participants. Compliance was analysed using 9 different wear-time criteria. Output comparisons were compared against one wear-time criteria (any 4 days at 10-hours per day). Time spent in SB and PA was calculated using R-package GGIR.
Wear-time for the wrist (15.6 to 17.4 h.d−1) was significantly greater than the hip (15.2 to 16.8 h.d−1) across all wear-time criteria (all P < 0.05). Moderate-strong associations were found between time spent in SB (r = 0.39), LPA (r = 0.33), MPA (r = 0.99), VPA (r = 0.82) and MVPA (r = 0.81) between the two device placements (All P < 0.001). The wrist device detected more minutes in LPA, MPA, VPA and MVPA whereas the hip detected more minutes in SB (all P = 0.001). Estimates of time in SB and all activity outcomes from the wrist and hip lacked equivalence.
The GT3X+ when worn at the wrist promotes greater compliance than at the hip. Minutes in SB and PA calculated from raw accelerations at the hip and wrist provide contrasting estimates and cannot be directly compared. Further calibration and correction factors may facilitate the comparison of findings in studies that estimate time spent in SB and PA intensities captured from the wrist and hip.
Original language | English |
---|---|
Pages (from-to) | S218-S219 |
Number of pages | 2 |
Journal | Journal of Physical Activity and Health |
Volume | 15 |
Issue number | 10 (Supplement 1) |
DOIs | |
Publication status | Published - 20 Oct 2018 |