Relationship between physical activity measured using accelerometers and energy expenditure measured using doubly labelled water in Indian children.

Background/objectives: To test the association between physical activity measured using accelerometer counts (Actigraph) and energy expenditure (EE) measured using the doubly labelled water (DLW) method in free-living children in India. The aim of this study was to explore the usefulness of Actigraphs in estimating EE.
Subjects/methods: Total EE (TEE) was measured in 58 children aged 8-9 years over a period of 2 weeks using the DLW technique. Physical activity level (PAL) was estimated from TEE, and the basal metabolic rate was predicted from weight. Physical activity was measured simultaneously using the Actigraph accelerometers (MTI AM7164 and GT1M). TEE was also calculated from the Actigraph counts using a published equation.
Results: TEE (mean: 6.6 vs 5.7 MJ, P=0.04) and Actigraph counts (counts/minute: 557 vs 465, P=0.02; total counts: 445 534 vs 354 748, P=0.004) were higher in boys than in girls. There were no significant correlations between either total Actigraph counts (r=0.15, P=0.3) or counts/minute (r=0.18, P=0.2), and TEE estimated using DLW. Similarly, there were no significant correlations between Actigraph counts and PAL (r=0.10, P=0.5; r=0.17, P=0.2, respectively). The Bland-Altman analysis showed poor agreement between TEE estimated using the DLW method and TEE derived from the Actigraph equation.
Conclusions: Activity measured using Actigraph accelerometers was not related to TEE and PAL derived using the DLW technique in children in Mysore. Actigraphs may not be useful in predicting EE in this setting, but may be better used for judging activity patterns.