Calibration and Validation of a Wrist Worn Accelerometer for 8 to 12 Year Old Children

Purpose: Currently, there are no methods to distill information from the most widely used wrist-mounted accelerometer. A field-based calibration study was conducted in order to determine cut point thresholds for this wrist-mounted accelerometer in children aged 8-12 years.

Methods: Forty-five children (25 developmental; 20 cross validation) aged 8-12 years, performed up to seven activities while wearing the wrist mounted ActiGraph on their non-dominant wrist. The four activities performed by all 45 children were: Rest, Enrichment/Drawing, Walking, and the Progressive Aerobic Cardiovascular Endurance Run. The other activities performed by a subset of children included: swimming, splashpad and playing on fixed playground equipment. All activities were performed in a summer day camp setting and represented free-living, unstructured activities used to mimic children’s sporadic activity habits. Each activity lasted 10 minutes with minutes 5 thorough 8.5 used for data analysis. Both direct observation and percentage of heart rate reserve were used to determine activity intensity.

Results: Receiver Operator Characteristic analyses were used to determine optimal cut point thresholds for sedentary (0-480 counts), light (481-2400 counts), moderate (2401-3120) and vigorous (3120+) physical activity for 15 second epochs. Area under the curve values were 0.84, 0.94, and 0.93 for light, moderate and vigorous thresholds, respectively. Cross-validation data showed high classifi cation agreement for all categories with areas under the curve being 0.799, 0.919, and 0.929, respectively.

Conclusion: This field-based calibration study suggests that physical activity intensities can be distilled from the most widely used wrist-mounted accelerometer.