OBJECTIVE: To determine whether the effect of hip exercise on patellofemoral pain is mediated through changes in hip muscle strength or psychological factors. DESIGN: Secondary mediation analysis of a randomized clinical trial, in which 218 participants with patellofemoral pain were randomly assigned to receive foot orthoses or hip exercises. METHODS: Pain (Knee injury and Osteoarthritis Outcome Score pain subscale) and number of pain-free squats at 12 weeks were the outcomes for this mediation analysis, as they are pathognomonic of patellofemoral pain. Hip strength dynamometry (abduction, adduction, and external rotation) and psychological characteristics (pain catastrophizing, kinesiophobia, and anxiety) measured at 6 weeks were considered as potential mediators. We used mediation analysis to decompose the total effect of treatment on the outcome into (1) the "indirect effect" (ie, the portion acting through the mediator) and (2) the "direct effect". RESULTS: The effect of hip exercise on pain and squats was not mediated by any of the strength or psychological mediators analyzed. All indirect effects were small and showed wide 95% confidence intervals (CIs) that contained zero (eg, for pain-free squats: abduction strength -0.13; 95% CI -0.49 to 0.23; Tampa Scale of Kinesiophobia -0.17; 95% CI -0.64 to 0.30). CONCLUSION: Hip strength improved after hip exercise, yet strength did not mediate improvements in pain and pain-free squats, and alternative psychological mediators were not implicated.

Full text (sometimes free) may be available at these link(s):      help

• DOI
• PubMed
• PDF locator
• publisher