Original research
Calibration of the GENEA accelerometer for assessment of physical activity intensity in children

https://doi.org/10.1016/j.jsams.2012.05.013Get rights and content

Abstract

Objectives

The purpose of the study was to establish activity intensity cut-points for the GENEA accelerometer via calibration with oxygen consumption (V˙O2).

Design

The study was a lab-based validation and calibration study.

Methods

Forty-four children, aged 8–14 years, completed eight activities (ranging from lying supine to a medium paced run) whilst wearing GENEA accelerometers at three locations (each wrist and at the right hip), an ActiGraph GT1M at the hip and a portable gas analyser. ActiGraph output and V˙O2 were used for assessment of concurrent and criterion validity, respectively. Pearson's r correlations were used to assess validity of the GENEA monitors at each location and location-specific activity intensity cut-points were established via Receiver Operator Characteristic curve analysis.

Results

The GENEA showed good criterion validity at both wrist locations (right: r = .900; left: r = .910, both p < 0.01), although the hip-mounted monitor demonstrated significantly higher criterion validity (r = .965, p < 0.05). Similar results were shown for concurrent validity (right: r = .830; left: r = .845; hip: r = .985, all p < 0.01). GENEAs, irrespective of wear location, accurately discriminated between all activity intensities (sedentary, light, moderate and vigorous) with the hip mounted monitor recording the largest area under the curve for each intensity (area under the curve = 0.94–0.99).

Conclusions

The GENEA can be used to accurately assess children's physical activity intensity when worn at either the wrist or the hip.

Introduction

The use of accelerometers to assess children's physical activity is by no means novel, and multiple devices have been validated for use within a child population, such as the Actical (Mini Mitter, Bend, OR, USA), the ActiGraph (ActiGraph, Pensacola, FL, USA) and the RT3 (Stayhealthy.com, Monrovia, CA, USA). Previous research has suggested that seven days of activity monitoring is optimal for a reliable estimate of habitual physical activity.1 However non-compliance resulting in a significant reduction of data is well documented2, 3, 4 and is thought to arise as most accelerometers are attached to a belt worn around the waist making it necessary to remove the monitor when changing clothes, sleeping and participating in some activities, e.g. contact sports, formal occasions.5 Non-compliance may also arise from children removing the monitor by choice (e.g. for physical comfort or social acceptance) or by necessity for water based activities (e.g. showering or swimming).3 Some monitors are also unable to record high resolution data for long periods of time, due to limited battery and memory capacity.6

In an attempt to combat these problems, the GENEA waveform triaxial accelerometer was developed (Unilever Discover, Colworth, UK; manufactured and distributed by ActivInsights Ltd., Kimbolton, Cambridge, UK). This lightweight (16 g), small (L36 mm × W30 mm × H12 mm) accelerometer collects data in three axes (vertical, anteroposterior and mediolateral) at a rate of up to 80 Hz (with the most recent model, the GeneActiv, recording at up to 100 Hz); it is designed to be worn on the wrist and is waterproof. It is hypothesised that the latter two qualities largely negate the need to remove the monitor and thus may lead to greater compliance during assessment of habitual activity in children.

The GENEA has been found to have high intra and inter-instrument reliability (coefficient of variation = 1.8% and 2.4%, respectively), good criterion-referenced validity when compared to a multi-axis shaking table (MAST; Instron Structural Testing Systems, Buckinghamshire, UK) (r = .97) and high concurrent validity with the ActiGraph GT1M (r = .74).6 Esliger et al.6 also found that, irrespective of whether the monitor was worn at the hip or the left wrist, the GENEA could be used to distinguish between sedentary, light, moderate and vigorous intensity activities in adults. This study goes some way to providing support for the GENEA worn at the wrist, however, if this monitor is to be used in child based studies a separate validation and calibration needs to be undertaken within this specific population.

The purpose of the present study therefore was to develop physical activity intensity cut-points for use with the GENEA when assessing the intensity of children's physical activity. To achieve this, the output was calibrated with a criterion measure of physical activity (indirect calorimetry) allowing accelerometer output cut-points to be established for sedentary, light, moderate and vigorous physical activity for use with children. Additionally, concurrent validity of the GENEA, relative to the ActiGraph, was investigated.

Section snippets

Methods

An opportunistic sample of 44 apparently healthy children (n = 26 females, n = 18 males) aged between 8 and 14 y (X¯=10.9±1.9) from Devon, UK were recruited for the present study. Children had a mean height of 150.3 cm (±13.0), mass of 41.8 kg (±10.9), body mass index (BMI) of 18.3 kg m−2 (±2.8) and waist circumference of 64.0 cm (±8.3). Participants peak V˙O2 ranged from 27.9 ml kg−1 min−1 to 58.4 ml kg−1 min−1 (X¯=41.6±7.7). The Ethics Committee at the University of Exeter granted approval for the study.

Results

Table 1 shows the intensity of each activity performed by participants, given in METs, along with the corresponding accelerometer output. Increases in MET values coincided with an increase in accelerometer output, for both the GENEA and the ActiGraph. The exception to this concerns the computer game activity (Nintendo Wii) where the wrist worn GENEA monitors recorded a high number of counts while the MET values remained relatively low. A series of group X activity mixed model ANOVAs revealed no

Discussion

The aim of the present study was to establish validity for the GENEA waveform triaxial accelerometer, and calibrate the output against energy expenditure to determine activity intensity cut-points for sedentary, light, moderate and vigorous physical activity in a child population.

The expected pattern of increasing MET values corresponding with increasing accelerometer counts was observed, with only one exception in this protocol: the interactive computer games (boxing on Nintendo Wii). This was

Conclusion

The GENEA has demonstrated high criterion and concurrent validity, irrespective of whether it is worn on either wrist or the hip, and has been established as peer to the ActiGraph. Cut-points for use with the GENEA when worn at the wrist or hip are presented to enable the classification of sedentary, light, moderate and vigorous intensity activity. It therefore has the potential to be used in future child related studies and may facilitate increased compliance rates.

Practical implications

  • The GENEA accelerometer is a valid measure of children's activity intensities whether worn at the wrist or the hip.

  • GENEA output and ActiGraph counts showed similar criterion validity as measures of children's physical activity.

  • Cut-points are now available to use with the GENEA to determine the intensity of children's physical activity.

Acknowledgements

The authors would like to thank Charlotte Benjamin and Johanna Hänggi for the help with the data collection, and the participants for their involvement in this research.

References (16)

  • S. Trost et al.

    Using objective physical activity measures with youth: how many days of monitoring are needed?

    Med Sci Sports Exerc

    (2000)
  • R. Colley et al.

    Quality control and data reduction procedures for accelerometry-derived measures of physical activity

    Health Rep

    (2010)
  • P.R.E. Crocker et al.

    Feasibility of using the tritrac motion sensor over a 7-day trial with older children

    Pediatr Exerc Sci

    (2001)
  • R. Troiano et al.

    Physical activity in the United States measured by accelerometer

    Med Sci Sports Exerc

    (2008)
  • K.R. Evenson et al.

    Assessment of differing definitions of accelerometer non-wear time

    Res Q Exerc Sport

    (2009)
  • D.W. Esliger et al.

    Validation of the GENEA accelerometer

    Med Sci Sports Exerc

    (2011)
  • V.T. Van Hees et al.

    Estimation of daily energy expenditure in pregnant and non-pregnant women using a wrist worn tri-axial accelerometer

    PLoS ONE

    (2011)
  • J.E. McLaughlin et al.

    Validation of the cosmed K4B2 portable metabolic system

    Int J Sports Med

    (2001)
There are more references available in the full text version of this article.

Cited by (223)

View all citing articles on Scopus
View full text