Validity of ActiGraph child-specific equations during various physical activities

Area of Science:

• Pediatric physical activity monitoring
• Biometric data analysis
• Exercise physiology

Background:

• Accurate assessment of physical activity in children is crucial for understanding health behaviors.
• ActiGraph devices are widely used for objective physical activity monitoring.
• Existing prediction equations for ActiGraph data in children require validation.

Purpose of the Study:

• To evaluate the validity of seven child-specific ActiGraph prediction equations/cut points.
• To compare the accuracy of these equations in estimating energy expenditure and time spent in different physical activity intensities (sedentary, light, moderate, vigorous).

Main Methods:

• Forty boys and 32 girls (aged 12 years) participated.
• Activity data were collected using ActiGraph GT3X+ and energy expenditure using Cosmed K4b(2).
• Participants underwent structured activities and a 2-hour free-living monitoring period.

Main Results:

• The Crouter vector magnitude two-regression model (Cvm2RM) and Puyau methods provided the best estimates of energy expenditure (MET(RMR)) during free-living conditions.
• The Crouter vertical axis two-regression model (Cva2RM) showed the closest estimation of time spent in sedentary, light, moderate, and vigorous physical activity.
• All tested prediction equations significantly underestimated vigorous physical activity (VPA) and exhibited substantial individual prediction errors.

Conclusions:

• Cvm2RM and Puyau equations are recommended for estimating energy expenditure in children.
• Cva2RM is suggested for classifying activity intensity levels.
• Researchers should be aware of the limitations and individual prediction errors associated with these methods.