Repeatability of tibial acceleration measurements made on children during walking and running

To determine the between-visit reliability of an accelerometer as a measure of lower-extremity impact acceleration at a variety of gait speeds in children.

Absolute reliability assessment.

Ten children with no known gait pathology attended two testing sessions, three weeks apart. A tri-axial accelerometer was fixed to the child's distal tibia to measure peak positive acceleration responses while walking and running on the treadmill at three different speeds (comfortable walking, threshold walking, and jogging). Reliability of the average and standard deviation Peak Positive Acceleration (avgPPA and sdPPA, respectively) was calculated by intra-class correlation coefficients (ICC) and Minimum Detectable Change (MDC).

Excellent reliability was indicated with ICC values for avgPPA of 0.90, 0.95, and 0.81 for comfortable walking, threshold walking, and jogging, respectively. Moderate reliability was found for the sdPPA measures. MDC values were calculated to be 18%, 26%, and 23% for comfortable walking, threshold walking, and jogging, respectively, indicating the amount by which an avgPPA value would need to change to ensure that the change is greater than a measurement error.

An accelerometer attached to the distal tibia is practical for use in a clinical environment to collect lower extremity acceleration data in children. Clinicians can utilise this technique for assessing a change following an intervention, such as biofeedback gait retraining.