INTRODUCTION: There is lack of understanding of the relationship between knee adduction moment (KAM) reductions and improvements in pain or function in patients with knee osteoarthritis (KOA). Moreover, there is no systematic review describing the longitudinal relationship between KAM changes and subsequent changes in pain and/or physical function. We aimed to: 1) investigate the relationship between changes in KAM induced by non-surgical biomechanical interventions and consecutive changes in pain and/or physical function in patients with medial KOA and; 2) compare this relationship for different interventions. EVIDENCE ACQUISITION: We considered eligible all RCTs using biomechanical interventions aimed to reduce KAM in KOA patients, that measured pain/function. We used Cohen's d effect size to quantify outcome measurements. EVIDENCE SYNTHESIS: Fourteen papers reporting 11 studies were identified. Braces were tested in 6 studies, insoles in 5 studies, shoes in 3 studies and gait retraining in 2 studies. Methodological differences were large among studies. Large effect sizes (>0.8) changes in pain/function were observed with interventions having at least a small KAM effect size (>0.2), suggesting an association between KAM and pain/function changes. A linear trend was observed between inter-intervention KAM and VAS pain effect sizes, based on 4 studies. No firm conclusions could be drawn for the different intervention types. CONCLUSIONS: There was a trend toward larger KAM reductions leading to larger improvements in pain/function in non-surgical biome-chanical interventions. Additional high-quality RCT with consistent methodology are needed to fully characterize the association between KAM and pain/function changes.