Abstract
OBJECTIVE: This study analyzed data from the US population to examine how oral microbiome diversity and diet quality individually and synergistically affect frailty. METHODS: This study included 6,283 participants aged 20 years or older from the 2009-2010 and 2011-2012 NHANES cycles. A frailty index (FI) consisting of 36 items was developed, with items related to nutritional status excluded. The diversity of the oral microbiome was assessed using α-diversity, including observed ASVs, the Shannon-Weiner index, Faith's phylogenetic diversity (PD), and the Simpson index. Dietary quality was assessed using Dietary Inflammatory Index (DII), Dietary Approaches to Stop Hypertension (DASH), Mediterranean Diet Score (MED), and Alternate Healthy Eating Index (AHEI). Multivariable logistic models were employed to examine the separate and combined associations of oral microbiome diversity and four dietary quality scores with FI, with interaction effects were explored. Several subgroup analyses and sensitivity analyses were conducted to assess the robustness of our findings. Furthermore, the mediation analysis was used to explore oral microbiome diversity as a mediator in the relationship between dietary scores and FI. RESULTS: Both oral microbiome diversity and dietary quality scores showed significant individual associations with FI. Jointly, those in the highest tertile of oral microbiome diversity and the lowest tertile of DII had lower FI [β(Observed ASVs) (95% CI) = -2.544(-3.678,-1.411); β(Faith's PD) (95% CI) = -2.688(-3.783,-1.593); β(Shannon-Weiner index) (95% CI) = -2.359(-3.333,-1.386); β(Simpson index) (95% CI) = -1.93(-2.879,-0.981)], compared to participants in the lowest tertile of oral microbiome diversity and the highest tertile of DII. A significant interaction between oral microbiome diversity (Observed ASVs and Faith's PD) and DII in relation to FI reduction was found (P for interaction(Observed ASVs*DII) = 0.032, P for interaction(Faith's PD*DII) = 0.014). Other dietary scores showed similar joint associations of oral microbiome diversity with FI, but no significant interactions were observed. Further mediation analysis indicated that the proportion of DII's effect on FI mediated through Observed ASVs, Faith's PD, and the Shannon-Weiner index was 8.7%, 7.5%, and 3.4%, respectively. CONCLUSION: This study demonstrates that a high-quality diet and greater α-diversity of oral microbiota are significantly associated with a reduced risk of frailty. Notably, the interaction between DII and the diversity of the oral microbiota exerts a particularly substantial influence on frailty risk.