Abstract
OBJECTIVES: This study aims to test a conceptual mediation model wherein periodontitis is associated with mortality through direct pathways and indirectly via accelerated biological aging. MATERIAL AND METHODS: We analyzed six cycles of National Health and Nutrition Examination Survey data with mortality follow-up of 250 months. Weighted descriptive statistics were used to compare group characteristics, Kaplan-Meier analysis to evaluate periodontitis-mortality associations and generalized linear models to examine the links between periodontitis and biological aging. Cox proportional hazards models integrated with restricted cubic splines were utilized to explore the association between biological age and mortality, and mediation analyses quantified the mediation of biological aging. Additionally, age, gender, and smoking status subgroup analyses were conducted. RESULTS: Moderate/severe periodontitis was associated with a significantly elevated all‑cause mortality risk (18.31% vs. 10.88% in no/mild periodontitis) and greater biological age advancement (PhenoAge: 1.22 years; KDM: 0.68 years). Biological age acceleration exhibited a non-linear association with mortality, with hazard ratios rising sharply beyond a threshold (PhenoAge: 16.4 years; KDM: 31.8 years). Mediation analysis showed that biological age partially mediated the periodontitis-mortality association, with indirect effect hazard ratios of 1.085 (95% CI: 1.067-1.106) for PhenoAge advance and 1.027 (95% CI: 1.016-1.040) for KDM advance in all‑cause mortality, though the proportion mediated was modest and varied across subgroups. CONCLUSIONS: Our findings support the hypothesis that biological aging (assessed by PhenoAge and KDM advances) plays a significant, though partial, mediating role in the link between periodontitis and elevated mortality.