Validation of two accelerometers to determine mechanical loading of physical activities in children

The purpose of this study was to assess the validity of accelerometers using force plates (i.e., ground reaction force (GRF)) during the performance of different tasks of daily physical activity in children. Thirteen children (10.1 (range 5.4-15.7) years, 3 girls) wore two accelerometers (ActiGraph GT3X+ (ACT), GENEA (GEN)) at the hip that provide raw acceleration signals at 100 Hz. Participants completed different tasks (walking, jogging, running, landings from boxes of different height, rope skipping, dancing) on a force plate. GRF was collected for one step per trial (10 trials) for ambulatory movements and for all landings (10 trials), rope skips and dance procedures. Accelerometer outputs as peak loading (g) per activity were averaged. ANOVA, correlation analyses and Bland-Altman plots were computed to determine validity of accelerometers using GRF. There was a main effect of task with increasing acceleration values in tasks with increasing locomotion speed and landing height (P > 0.001). Data from ACT and GEN correlated with GRF (r = 0.90 and 0.89, respectively) and between each other (r = 0.98), but both accelerometers consistently overestimated GRF. The new generation of accelerometer models that allow raw signal detection are reasonably accurate to measure impact loading of bone in children, although they systematically overestimate GRF.