Feasibility and validity of the ActiGraph GT3X accelerometer in measuring physical activity of Malawian toddlers

Abstract

Aim: To test the feasibility and validity of the ActiGraph GT3X accelerometer in measuring physical activity of rural Malawian toddlers.

Methods: Fifty-six children aged 16.0-18.5 months wore the accelerometer on their right hip for 7 days. We analysed days with a minimum of 600 min of wear time, excluding night time and periods when the unit registered zero for 20 consecutive minutes. The first and last days were excluded as they were incomplete. Accelerometer counts were compared with coded free play video recordings to define median accelerometer counts for sedentary, light, moderate and vigorous activity. Count cut points were defined for moderate to vigorous physical activity, with predictive validity assessed using a second set of video recordings.

Results: Median wear time was 797 min/day, with 79% of participants completing at least four eligible days. Accelerometer counts were significantly higher for observed moderate to vigorous physical activity, than lighter activity, with cut points of 208 counts/15 sec for vector magnitude and 35 counts/15 sec for vertical axis, showing sensitivity of 94.2% and 84.1% and specificity of 90.9% and 84.6%, respectively.

Conclusion: The accelerometer proved a feasible and valid method of assessing physical activity among Malawian toddlers.

Keywords: ActiGraph accelerometer; Infants; Motor activity; Physical activity; Validation studies.