A Reexamination of the Relationship between Foot Strike Angle and Early Stance Loading Variables during Running

This study aimed to compare nonlinear and piecewise linear models for FSA with VLR/VTA during running, controlling for competitive running experience.

A total of 475 U.S. Army trainees completed surveys on their participation in sports and a running gait assessment on an instrumented treadmill while VLR, VTA, and FSA were measured (via force plates, tibial-mounted accelerometers, and high-speed cameras, respectively). A 3-min warm-up was given, and then participants ran for 1 min at 2.68-2.91 m·s -1 , followed by 14 strides of data collection. Multiple regression was used to assess the prediction of VLR and VTA by FSA, including nonlinear/piecewise terms and controlling for competitive running experience.

A nonlinear, quadratic relationship was found for FSA × VLR ( R2 = 0.25, P < 0.001) and VTA ( R2 = 0.14, P < 0.001), and with a significant effect of running experience ( P < 0.02). Models showed similar shapes but better fit for those with (VLR: R2 = 0.38; VTA: R2 = 0.16) versus without (VLR: R2 = 0.21; VTA: R2 = 0.11) competitive running experience. Finally, VLR peaked at similar FSAs as in previous work (≈10°-12° current vs 5°-10° previous). Piecewise linear regression produced similar results to quadratic models ( R2 difference < 0.02).

Our results show that nonlinear or piecewise linear models can be used to represent the relationship between VALR/VTA and FSA. Although VLR/VTA were lowest with the lowest FSAs (i.e., most forefoot), participants with the highest FSAs (i.e., most rearfoot) showed reduced values compared with those typical of a more moderate/flatter rearfoot strike pattern.