Original research
Validation and comparison of ActiGraph activity monitors

https://doi.org/10.1016/j.jsams.2011.04.003Get rights and content

Abstract

Objective: To compare activity counts from the ActiGraph GT3X to those from the ActiGraph GT1M during treadmill walking/running. A secondary aim was to develop tri-axial vector magnitude (VM3) cut-points to classify physical activity (PA) intensity. Methods: Fifty participants wore the GT3X and the GT1M on the non-dominant hip and exercised at 4 treadmill speeds (4.8, 6.4, 9.7, and 12 km h−1). Vertical (VT) and antero-posterior (AP) activity counts (counts min−1) as well as the vector magnitudes of the two axes (VM2) from both monitors were tested for significant differences using two-way ANOVA's. Bland–Altman plots were used to assess agreement between activity counts from the GT3X and GT1M. Linear regression analysis between VM3 counts min−1 and oxygen consumption data was conducted to develop VM3 cut-points for moderate, hard and very hard PA. Results: There were no significant inter-monitor differences in VT activity counts at any speed. AP and VM2 activity counts from the GT1M were significantly higher (p < 0.01) than those from the GT3X at 4.8, 9.7 and 12 km h−1. High inter-monitor agreement was found for VT activity counts but not for AP and VM2 activity counts. VM3 cut-points for moderate, hard, and very hard PA intensities were 2690–6166, 6167–9642, >9642 counts min−1. Conclusion: Due to the lack of congruence between the AP and VM2 activity counts from the GT1M and the GT3X, comparisons of data obtained with these two monitors should be avoided when using more than just the VT axis. VM3 cut-points may be used to classify PA in future studies.

Introduction

ActiGraph (Pensacola, FL) activity monitors are widely used in physical activity (PA) research. The uniaxial ActiGraph 7164 model was extensively used in the mid 1990s and early 2000s.1, 2, 3 The 7164 model was subsequently replaced with the technologically superior, Micro-Electro-Mechanical System (MEMS) uniaxial GT1M model. Studies have shown that activity counts measured in the vertical (VT) plane using the 7164 and GT1M monitors are not significantly different from each other.4, 5, 6 This allows inter-monitor (7164 and GT1M) comparisons of results among different studies. In October 2008, ActiGraph enabled dual axes measurement capability in the GT1M by unlocking the antero-posterior (AP) axis. Users can now obtain activity counts from the VT, AP, and a composite vector magnitude of these two axes (VM2) from the GT1M.

In 2009, ActiGraph discontinued the GT1M and released the triaxial GT3X activity monitor. The GT3X measures acceleration in three individual orthogonal planes (VT, AP, and medio-lateral (ML)) and provides activity counts as a composite vector magnitude of these three axes (VM3). Signal processing specifications of the GT1M and the GT3X are identical and ActiGraph states that there are no intra-axis differences in activity counts from these two activity monitors (Personal Communication with John Schneider, ActiGraph Vice-President for research and development, November 2009). To date, no study has empirically examined if there are differences in activity counts between these two monitors. This needs to be investigated since researchers potentially apply PA classification techniques developed on the older GT1M to data collected using newer models. Additionally, determining output similarities between the GT1M and the GT3X may also allow the comparability of results among different studies that did not use the same model of activity monitor. Therefore, the primary aim of this investigation was to determine if there are differences in activity counts obtained from the multiple axes of the GT1M and the GT3X. A secondary aim was to develop VM3 activity count cut-points to classify PA intensity. It is possible that advanced PA classification techniques using pattern recognition will be available for the GT3X in the future. However, these methods remain in the early stages of development. Therefore, despite the limitations of using simple accelerometer PA cut-points, this analytic technique remains the method of choice to classify PA intensity from accelerometer output.

Section snippets

Methods

Fifty healthy participants (28 men and 22 women; mean age ± SD = 26.9 ± 7.7 years) from Amherst, MA and surrounding areas volunteered to participate in this study. Participants were given detailed information about study procedures and provided researchers with written informed consent. This study was approved by the University of Massachusetts Institutional Review Board.

Activity monitors were initialized to collect data in 1 s epochs. The GT3X was initialized to collect data in the VT, AP, and ML

Results

The final sample for the comparison analysis was comprised of 32 participants [mean ± SD age, height, weight, and body mass index: 28.0 ± 9.0 years, 173.2 ± 8.5 cm, 71.6 ± 12.3 kg, and 23.8 ± 3.6 kg m−2]. No significant differences among VT activity counts from the GT1M and GT3X were observed during any of the 4 speeds (Fig. 1a). The Bland–Altman plot showed a high agreement between VT activity counts from the GT3X and GT1M monitors (Fig. 2a). The mean bias for VT activity counts (GT3X  GT1M) was −50 ± 382 counts

Discussion

This study compared activity counts between the ActiGraph GT1M and GT3X activity monitors during treadmill walking and running. Significant inter-monitor differences in AP and VM2 activity counts were found (Fig. 1b and c). These differences resulted in poor agreement between AP activity counts from the two activity monitors and moderate agreement between VM2 activity counts from the GT1M and the GT3X (Fig. 2b and c).

Inter-monitor discrepancy in AP activity counts may be attributable to

Conclusion

The present study showed that VT activity counts from the GT1M and the GT3X are comparable, while those from the AP axis and VM2 are dissimilar. Our results suggest that if data have been collected in more than just the VT axis, a direct comparison of findings among studies must be avoided when one study used the GT1M and the other the GT3X. This study provides new VM3 cut-points for the ActiGraph GT3X, allowing researchers to use a superior activity monitor to classify PA intensity.

Practical implications

  • Physical activity prediction models developed from previous uniaxial ActiGraph accelerometers (e.g. Freedson cut-points) can be used with GT3X vertical axis activity counts.

  • Due to dissimilarities in activity counts between the GT1M and the GT3X in the antero-posterior axis, direct comparisons of findings among studies using different monitors to measure physical activity must be avoided.

  • Until advanced techniques are developed using triaxial data from the GT3X, the VM3 activity count cut-points

Acknowledgements

We thank all of the participants of the study and Cheryl Howe, Kate Lyden, Natalia Petruski, and Sarah Kozey-Keadle who assisted with various aspects of the project.

The present study did not receive any external funding.

Patty Freedson is a paid member of the Scientific Advisory Committee for Actigraph, Inc. (Pensacola, FL).

References (16)

  • P.S. Freedson et al.

    Calibration of the Computer Science Applications, Inc. accelerometer

    Med Sci Sports Exerc

    (1998)
  • D. Hendelman et al.

    Validity of accelerometry for the assessment of moderate intensity physical activity in the field

    Med Sci Sports Exerc

    (2000)
  • A.M. Swartz et al.

    Estimation of EE using CSA accelerometers at hip and wrist sites

    Med Sci Sports Exerc

    (2000)
  • K. Corder et al.

    Comparison of two ActiGraph models for assessing free-living physical activity in Indian adolescents

    J Sports Sci

    (2007)
  • D. John et al.

    Comparison of four ActiGraph accelerometers during walking and running

    Med Sci Sports Exerc

    (2010)
  • S.L. Kozey et al.

    Comparison of the ActiGraph 7164 and the ActiGraph GT1M during self-paced locomotion

    Med Sci Sports Exerc

    (2010)
  • H. Rosdahl et al.

    Evaluation of the Oxycon mobile metabolic system against the Douglas bag method

    Eur J Appl Physiol

    (2010)
  • J. Nilsson et al.

    Changes in leg movements and muscle activity with speed of locomotion and mode of progression in humans

    Acta Physiol Scand

    (1985)
There are more references available in the full text version of this article.

Cited by (0)

View full text