The hip adduction moment as a proxy for joint loading: a comparison between walking and running

Abstract

Hip adduction moment (HAM) is commonly used as a surrogate for hip joint loading; however, its validity across different gait tasks and modifications remains unclear. Understanding how gait modification influences this relationship is essential for developing effective load-management strategies for individuals with or at risk of hip osteoarthritis. This study examined the relationship between HAM and hip joint reaction force (HJRF) during walking and running under natural and foot progression modification conditions in healthy adults. Musculoskeletal modelling in OpenSim was used to estimate HJRF, and inverse dynamics were applied to calculate HAM. Spearman’s correlation coefficients were used to assess the relationship between HAM and HJRF across gait conditions. Paired t-tests and linear mixed models, accounting for speed where applicable, assessed differences between natural and modified gait.

During walking, early stance HAM was strongly correlated with HJRF in both natural (ρ = 0.923, p < 0.001) and toe-in conditions (ρ = 0.851, p < 0.001). Toe-in walking significantly reduced HAM (p = 0.017) and HJRF (p = 0.037) compared to natural walking. During running, peak HAM had lower but significant correlation with HJRF in the natural condition (ρ = 0.703, p < 0.001), but no significant association was observed during toe-out running (p = 0.065). Toe-out running significantly decreased HAM (p = 0.006) while HJRF remained unchanged (p = 0.941). Gait modification influences the coupling between HAM and HJRF, particularly during dynamic tasks such as running. While HAM is a valid indicator of hip loading during walking, it becomes less representative under altered dynamic conditions. These findings highlight the limitations of using single-plane moments as a surrogate for joint loading.