Comparison of children's free-living physical activity derived from wrist and hip raw accelerations during the segmented week

Robert Noonan, L.M. Boddy, Y Kim, Z.R. Knowles, Stuart Fairclough

Research output: Contribution to journalArticle

16 Citations (Scopus)
60 Downloads (Pure)

Abstract

This study assessed children’s physical activity (PA) levels derived from wrist-worn GENEActiv and hip-worn ActiGraph GT3X+ accelerometers and examined the comparability of PA levels between the two devices throughout the segmented week. One hundred twenty nine 9-10 year old children (79 girls) wore a GENEActiv (GAwrist) and ActiGraph GT3X+ (AGhip) accelerometer on the left wrist and right hip respectively for seven days. Mean minutes of light PA (LPA) and moderate-to-vigorous PA (MVPA) per weekday (whole-day, before-school, school and after-school) and weekend day (whole-day, morning and afternoon-evening) segments were calculated, and expressed as percentage of segment time. Repeated measures ANOVA examined differences in LPA and MVPA between GAwrist and AGhip for each time segment. Bland–Altman plots assessed between-device agreement for LPA and MVPA for whole weekday and whole weekend day segments. Correlations between GAwrist and AGhip were weak for LPA (r=0.18-0.28), but strong for MVPA (r=0.80-0.86). LPA and MVPA levels during all weekday and weekend day segments were significantly higher for GAwrist than AGhip (p<0.001). The largest inter-device percent difference of 26% was observed in LPA during the school day segment. Our data suggest that correction factors are needed to improve raw PA level comparability between GAwrist and AGhip.
Original languageEnglish
Pages (from-to)2067-2072
JournalJournal of Sports Sciences
Volume35
Issue number21
Early online date14 Nov 2016
DOIs
Publication statusE-pub ahead of print - 14 Nov 2016

Fingerprint Dive into the research topics of 'Comparison of children's free-living physical activity derived from wrist and hip raw accelerations during the segmented week'. Together they form a unique fingerprint.

  • Cite this