Does the measure of the foot strike angle using 2D video provide a good inter-rater reliability? A retrospective study

Details

Supervisors

Nguyen, Anh Phong

Faculty

Faculté des sciences de la motricité

Degree label

Master [120] en sciences de la motricité, orientation générale, à finalité spécialisée: kinésithérapie musculo-squelettique

Abstract

Objective: Investigate inter-rater reliability in the measurement of foot strike angle using a HD camera and Kinovea during a trail. Method: In this retrospective study, the FSA of the 55 runners were taken at the end of the 6 laps of the 26km Clecy scientific trail by 3 observers, and this in a completely independent way. Measurements were carried out using motion analysis software and repeated in an excel table. The measurements were then submitted to various statistical tests such as the analysis of the ICC (2.1), the calculation of Cohen's weighted kappa, or reported in a Bland and Altman diagram. Main results: Weighted Cohen’s kappa coefficient and their 95% IC were calculated to compare operator 1 and operator 2; OP2 and OP3 and finally for OP1 and OP3. The obtained values are: 0,70575 (0,63130 - 0,78020) ; 0,73703 (0,66335 - 0,81071) ; 0,81109 (0,75198 - 0,87019). The ICC value (2,1) comparing the values obtained by the operators is 0.9559 with a 95% CI from 0.9568 to 0.9636. Finally, Bland and Altman plots have been realized following the same methodology as Cohen's weighted. The means difference and the limits of agreement were equal to 2.5 (-6,3 - 11,3), -1,1 (-8,5 - 6,3) and 1,2 (-7,3 - 9,7). Authors' conclusion: The results showed a good degree of inter-rater agreement. Its simplicity and reliability should be a precious resource in the medical community. It would allow for example the setting up of larger studies trying to demonstrate a correlation between race pattern and running related injuries and consequently expand prevention.