Association between the Framingham steatosis index and osteoarthritis: evidence of mediation by phenotypic age

Abstract The Framingham Steatosis Index (FSI) is a validated surrogate marker of hepatic steatosis and metabolic dysfunction. Emerging evidence suggests that metabolic abnormalities and accelerated biological aging may contribute to osteoarthritis (OA). However, whether FSI is associated with OA and whether this relationship is mediated by Phenotypic Age (PhenoAge) remain unclear. This study conducted a cross-sectional analysis of 9279 U.S. adults from the National Health and Nutrition Examination Survey, including 948 participants with OA. Weighted logistic regression models were applied to evaluate the association between FSI and OA. Restricted cubic splines and two-piecewise linear regression were used to assess nonlinear relationships and threshold effects. Subgroup analyses were performed to examine effect modification. Exploratory mediation analysis was conducted to quantify the mediating role of PhenoAge in the FSI–OA relationship. Higher FSI was significantly associated with OA (Model 3: OR = 1.18, 95% CI 1.12–1.25, p < 0.001), with a clear dose–response pattern across quartiles ( p for trend < 0.001). A nonlinear association was observed with an inflection at FSI = – 1.798; OA prevalence increased steeply below this threshold. Subgroup analyses showed consistent positive associations across strata. Exploratory mediation analysis suggested that PhenoAge may partially account for this association between FSI and OA, accounting for 23.03% of the total effect. FSI was positively associated with OA prevalence in U.S. adults, and Phenotypic Age may partially account for this association. Future longitudinal studies are needed to clarify causality and determine whether improving metabolic health or slowing biological aging can reduce OA risk.