Related Experiment Videos
Comparison of Accelerometer-Based Cut-Points for Children's Physical Activity: Counts vs. Steps.
Cheryl A Howe1, Kimberly A Clevenger2, Ryann E Leslie3
1School of Applied Health Sciences and Wellness, Ohio University, 1 University Terrace, Grover Center E154, Athens, OH 45701, USA. howec@ohio.edu.
Children (Basel, Switzerland)
|August 8, 2018
Summary
Accelerometer-based cadence may misclassify children's moderate-vigorous physical activity (MVPA). This study found cadence had lower agreement than counts/min for estimating MVPA during free play.
Area of Science:
- Pedometry and physical activity monitoring
- Biomechanical analysis of human movement
- Childhood physical activity and health
Background:
- Accelerometers are used to measure children's physical activity (PA), including moderate-vigorous PA (MVPA).
- Current technology allows for step counting and activity intensity estimation using accelerometer data.
- Accurate interpretation of accelerometer-based metrics like cadence (steps/min) is crucial for reliable MVPA assessment in children.
Purpose of the Study:
- To evaluate the relationship and agreement between accelerometer-derived cut-points (cadence and counts/min) and measured MVPA in children.
- To compare the accuracy of different accelerometer metrics in estimating children's free-play PA intensity.
Main Methods:
- Forty children (8-12 years) participated in the study.
- Participants engaged in 6-10 games while wearing both a portable metabolic analyzer and a GT3X+ accelerometer.
- Measured and estimated MVPA were compared using correlation, kappa statistics, sensitivity, and specificity.
Main Results:
- Games generated an average intensity of 6.3 METs, 64.5 steps/min, and 3318 (vertical) or 5350 (vector magnitude) counts/min.
- Cadence showed a stronger relationship (r=0.50) with measured MVPA intensity than vertical or vector magnitude counts/min (r=0.38).
- Agreement between measured and estimated MVPA varied significantly by cut-point, with low agreement observed for cadence-based cut-points.
Conclusions:
- While accelerometer-derived metrics correlate with measured PA intensity, cadence-based methods may misclassify children's free-play MVPA.
- Accelerometer counts/min demonstrated variable but sometimes substantial agreement with measured MVPA, depending on the cut-points used.
- Further research is needed to refine accelerometer algorithms for accurate assessment of children's PA, particularly during free-play activities.