Article Text

Download PDFPDF
117 Validation of Accelerometers for Measurement of Physical Activity Energy Expenditure in South Asians and Europeans
  1. Saima Afaq1,
  2. Sian-Tsung Tan1,
  3. Uzma Afzal1,
  4. Marie Loh1,
  5. Anthony Dimarco2,
  6. Jaspal Kooner1,
  7. John Chambers1
  1. 1Imperial College London
  2. 2Oxford University Hospitals NHS Trust


Background South Asians are at ~2 fold higher risk of cardiovascular disease (CVD) than Europeans. Physical inactivity is an independent risk factor for CVD, and may be more common in South Asians than Europeans. However, published studies of physical activity in South Asians have all relied on self-report by questionnaire, an approach of limited accuracy, confounded by linguistic, cultural and reporting bias. Accelerometers are validated tools for measurement of physical activity energy expenditure (PAEE) in Europeans, but are influenced by body habitus and device positioning. The accuracy of accelerometers in South Asians is not known.

Aims To identify the accelerometer and accelerometer position that most accurately quantifies physical activity amongst South Asians, and to compare the accuracy of the chosen accelerometer between South Asians and Europeans.

Methods We investigated 57 South Asians and 29 Europeans undergoing an exercise treadmill test (ETT) for clinical indications. All participants completed the treadmill test while simultaneously wearing 3 commercially available accelerometers in manufacturer recommended positions: i. Actigraph GT3X (3 positions - ankle, waist and wrist); ii. Geneactiv (3 positions - ankle, waist and wrist) and iii. Actiheart (chest only). Each participant thus had 7 device-position combinations. Personal and medical history was collected by questionnaire, followed by anthropometric measurements (height, weight, waist and hip circumference). We used linear regression and Bland-Altman plots to investigate the relationships between PAEE measured by the accelerometers (Measured) and actual PAEE on the treadmill (Actual).

Results The relationship between Measured and Actual PAEE was highest for the Actigraph GT3X in the ankle position (regression beta: 0.55+/-0.03, Table). Wrist-worn devices consistently displayed poor PAEE prediction. Measured and Actual PAEE were similar amongst South Asians and Europeans for all accelerometers and positions (P > 0.05), including the ankle-worn Actigraph GT3X. However, even the ankle-worn Actigraph GT3X underestimated PAEE, especially at high workload (Figure).

Conclusion The Actigraph GT3X is the most accurate device for measuring PAEE amongst both South Asians and Europeans. PAEE is best measured by the ankle-worn accelerometers followed by waist-worn accelerometers. Our results will inform the design of future studies to investigate the relationship between physical activity and CVD in the two populations.

Relationship between Measured and Actual energy expenditure [beta (SE), from regression] in South Asians and Europeans.

Abstract 66 Table 2

Procedural characteristics

  • Cardiovascular disease
  • physical activity
  • South Asians

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.