This study examines the association between denture use and all-cause mortality risk among Chinese edentulous elderly, using data from the Chinese Longitudinal Healthy Longevity Survey (CLHLS, 2008-2014 waves). A cohort of 8444 adults aged [Formula: see text] with complete tooth loss was analyzed, with denture use recorded as a binary variable (yes/no) and mortality verified via follow-up linkages. Missing data were handled via multiple imputation (MICE), and propensity score matching (PSM) balanced covariates between denture users and non-users. Kaplan-Meier survival analysis demonstrated significantly improved survival probabilities among denture users compared to non-users (log-rank p = 0.0033). Multivariate Cox proportional hazards models, after full adjustment for potential confounders, consistently revealed that denture non-users faced a 15.8% higher mortality risk (HR = 1.158, 95% CI 1.082-1.239, P < 0.001). Mediation analyses elucidated potential pathways, identifying BMI as a significant partial mediator (indirect effect = - 0.010, P < 0.001) in the denture-mortality relationship. In contrast, neither dietary diversity (DDS: p = 0.105) nor daily staple food intake (DSFA: p = 0.190) demonstrated significant mediating effects. The robustness of these findings was confirmed through comprehensive sensitivity analyses, including complete-case analysis and evaluation of pre-propensity score matched cohorts, with all approaches yielding consistent results. This study provides robust evidence that denture use is significantly associated with reduced all-cause mortality among edentulous elderly adults, with BMI serving as a key mediating factor in this protective relationship.
No clinical trial protocols linked to this paper
Clinical trials are automatically linked when NCT numbers are found in the paper's title or abstract.PICO Elements
No PICO elements extracted yet. Click "Extract PICO" to analyze this paper.
Paper Details
MeSH Terms
Associated Data
No associated datasets or code repositories found for this paper.
Related Papers
Related paper suggestions will be available in future updates.