Submitted to: Obesity Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/20/2001
Publication Date: 3/25/2002
Interpretive Summary: Interest has increased in the determination of children's precise activity levels because of the fact that overweight and obesity are becoming such widespread phenomena among U.S. children, which bodes poorly for their future as active, fit, and healthy adults. Developing technology has produced small activity monitors that can be attached to parts of the body to measure activity, even tiny movements while at rest. However, these activity monitors have been calibrated in terms of resting metabolic rate equivalents (METs), and MET thresholds have been defined for various levels of activity in adults only. That creates a problem because children have different resting metabolic rates than adults. We wanted to validate and test the performance and sensitivity of two types of activity monitors, the Actigraph and the Actiwatch, in measuring various levels of activity of a group of children. We also wanted to calibrate the monitors in terms of threshold levels for sedentary, light, moderate, and hard physical activit in children. We tested the monitors in a group of 26 children in a room respiration calorimeter in the lab as well as in field settings, as they pursued activities at various energy levels. We achieved all our objectives, and found that both monitors provide useful tools with which to measure physical activity levels in children.
Technical Abstract: Purpose: This study was designed 1) to validate accelerometer-based activity monitors against energy expenditure (EE); 2) to test performance of the monitors placed above the iliac crest (hip) or on the lower leg; 3) to test monitors under field conditions; and 4) to calibrate threshold levels for sedentary, light, moderate and hard physical activities in children. Methods: Computer Science and Applications Actigraph (CSAAG) and Mini-Mitter Actiwatch (MMAW) monitors, placed above the hip or on the lower leg of 26 children aged 6-16 y, were validated and calibrated against continuous 6-h measurements of EE by respiration calorimetry, activity by microwave detector and heart rate by telemetry in a room calorimeter in 26 children, ages 6-16 y. Results: Mean correlations between EE and counts were slightly higher for the MMAW-hip (r=0.78 +/- 0.06) and MMAW-leg (r=0.80 +/- 0.05) than the CSAAG-hip (r=0.66 +/- 0.08) and CSA-leg (r=0.73 +/- 0.07). The CSA and MMAW performed similarly on the hip (inter- instrument r=0.88) and on the lower leg (inter-instrument r=0.89). Threshold levels corresponding to AEE of 0.015, 0.05 and 0.10 kcal.kg- 1.min-1 and heart rates of 90, 130 and 160 bpm were derived from the regression of AEE on activity counts. Conclusion: CSAAG and MMAW monitors provide useful tools for the assessment of physical activity levels in children.