Abstract
Frailty is associated with numerous adverse health outcomes, and insulin resistance (IR) is a known risk factor for its development. However, it remains unknown whether non-insulin-based IR indicators are associated with frailty prevalence. This study investigated the association between the estimated glucose disposal rate (eGDR, a noninvasive IR marker) and frailty, and examined the mediating role of systemic inflammation. Using National Health and Nutrition Examination Survey data (1999-2018), we performed weighted logistic regression and employed restricted cubic splines to analyze the eGDR-frailty relationship in adults ≥20 years. Mediation analysis quantified the contributions of 4 inflammatory biomarkers (systemic immune-inflammation index, systemic inflammation response index (SIRI), neutrophil-to-lymphocyte ratio, neutrophil-to-albumin ratio). Stratified analyses with interaction tests were visualized via forest plots. The study included 17,384 individuals (mean age 60.40 ± 15.26 years; 53.6% female). After full adjustment, each unit increase in eGDR was associated with a 16% reduction in frailty risk (odds ratio = 0.84; 95% confidence interval 0.83-0.86; P < .001). Restricted cubic splines confirmed a non-linear inverse association (P-nonlinear < .05). Mediation analysis revealed modest but significant mediation by inflammatory markers: neutrophil-to-albumin ratio showed the strongest effect (6.68%; P < .001), followed by SIRI (2.32%) and followed by SIRI (2.32%) and both neutrophil-to-lymphocyte ratio and systemic immune-inflammation index (approximately 1.15%). Subgroup analyses demonstrated a consistent protective association between higher eGDR and reduced frailty prevalence across all subgroups (all P-interaction > .05). Higher eGDR exhibits an L-shaped inverse association with frailty in US adults, partially mediated (1.14-6.68%) by systemic inflammation (particularly the neutrophil-to-albumin ratio). These findings support targeting metabolic-inflammatory crosstalk for frailty prevention.