Determination of cut-off points for the Move4 accelerometer and assessment of energy expenditure in children and adolescents aged 6–16 years using manual wheelchairs: a validation and calibration study

BACKGROUND: The present study aimed to determine activity intensity and validate the Move4 accelerometer and derive cut-off points to accurately classify wheelchair-based physical activity (PA) in children and adolescents. METHODS: A calibration and validation study with a test-retest design was performed in 24 children and adolescents (mean age 9.9 ± 3.2 years; 50% girls) using manual wheelchairs. Participants used a manual wheelchair as their primary mode of mobility and were able propelling the wheelchair independently. Participants completed standardized activities while wearing Move4 sensors at the wrist, upper arm, and chest. In a subsample of 7 children and adolescents, energy expenditure was recorded using portable metabolic cart. Mean Amplitude Deviation (MAD) and Movement Acceleration Intensity (MAI) were processed to derive cut-off points for sedentary, light, moderate, and vigorous PA using CART models, and validity was assessed via Metabolic Equivalent (MET) and heart-rate responses. RESULTS: Both MAD and MAI values increased systematically across activity intensities, supporting good criterion validity. MET values showed progression from sedentary to vigorous tasks. Intensity-specific cut-off points were successfully established for all sensor placements, with consistently higher thresholds for MAI. Sensitivity and specificity analyses indicated the most balanced performance for the upper arm (MAD) and chest (MAI). Test-retest comparisons showed greater variability for MAD-based cut-offs, whereas MAI thresholds demonstrated higher stability, with deviations of approximately 4-5%. CONCLUSIONS: While both MAD and MAI validly reflected PA intensity, MAI showed greater temporal stability and test-retest robustness. These findings support the use of accelerometery for classifying PA in children and adolescents using wheelchairs and underscore the importance of selecting appropriate metrics and sensor positions.