Comparison of physical activity metrics from two research-grade accelerometers worn on the non-dominant wrist and thigh in children

Duncan S. Buchan*, Benjamin D. Maylor

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)
13 Downloads (Pure)

Abstract

This study compared physical activity metrics from the activPAL (AP) worn on the thigh with the ActiGraph worn on the non-dominant wrist using open-source methods. Measures included average acceleration, intensity gradient (IG) and the minimum acceleration value of the most active X mins (MX). Fifty-two children (26 boys; age: 10.4 ± 0.6 years) provided ≥1 day (24h) of concurrent wear time from the activPAL and ActiGraph. Measures tended to be lower from the activPAL versus the ActiGraph. Poor agreement was evident for average acceleration but good for the IG. For the IG, the absolute and relative zones needed to reach equivalence was 4% and 0.4 SDs, respectively and for average acceleration were 10% and 1.2 SDs, respectively. Good agreement was evident for M60, M30, M20, M15 and M10 between devices. Regardless of the reference device used, equivalent estimates for the intensity gradient, M60, M30, M20, M15 and M10 were observed with relative and absolute equivalence zones being ≤4% and ≤0.5 SDs, respectively. The IG, M60, M30, M20, M15 and M10 appear good candidates for comparing activity data collected from the activPAL and ActiGraph. Future research can use the AP to report on sedentary behaviours as well as PA outcomes.
Original languageEnglish
Pages (from-to)80-88
Number of pages9
JournalJournal of Sport Science
Volume41
Issue number1
Early online date4 Apr 2023
DOIs
Publication statusPublished - 2023

Keywords

  • GGIR
  • agreement
  • equivalence
  • intensity gradient
  • average acceleration
  • MX

Fingerprint

Dive into the research topics of 'Comparison of physical activity metrics from two research-grade accelerometers worn on the non-dominant wrist and thigh in children'. Together they form a unique fingerprint.

Cite this