Validation of Different Accelerometers to Determine Mechanical Loading in Children

Purpose: Impact loading of bone is thought to be an important prevention for osteoporosis throughout life. The purpose of this study was to assess validity of accelerometers that provide raw acceleration signals with ground reaction force (GRF) for daily physical activities in children. Understanding error of these devices is a precondition for their field-based use over time in observational and intervention studies interested in mechanical loading of bone.

Methods: Thirteen children (10.1 yr (SD 3.0), 3 girls) wore two accelerometers (ActiGraph GT3X+ (ACT), GENEA (GEN)) at the hip that are able to provide raw acceleration signals at 100Hz. Participants completed different tasks (walking, jogging, running, drops from boxes of 10, 20, 30 cm height, rope skipping, and dancing) on a force plate. GRF (in bodyweight) was collected for one step per trial (10 trials) for ambulatory movements and for all dropbox (10 trials), skipping and dance procedures. Accelerometer outputs as peak loading (g) per activity were averaged. Descriptive statistics, correlation analyses and Bland-Altmann plots were used to determine validity of accelerometers to assess GRF.

Results: There was a main effect on the task level with increasing acceleration values by increasing locomotion speed and boxdrop height (p<0.001). Accelerations of ACT and GEN correlated with GRF for all activities (r=0.90 and 0.89, respectively) and between each other (r=0.98), but both accelerometers consistently overestimated GRF.

Conclusions: The validity of the accelerometers available to estimate GRF by force plate is sufficient to assess impact loading of bones in children.