Lower Oral Microbial Alpha Diversity and Greater Community Heterogeneity Among Adults with Depression: A Cross-Sectional Analysis of NHANES 2009-2012.

Background: Growing evidence links microbial communities to mental health, but studies examining oral microbial diversity in depression have produced inconsistent findings and have often included small samples. Using a large population-based sample, we examined whether oral microbial diversity, overall community composition, and between-person heterogeneity differed by depression status and whether associations with alpha diversity persisted after adjustment for sociodemographic, behavioral, and oral-health factors. Methods: We analyzed oral microbiome data from the 2009–2012 National Health and Nutrition Examination Survey (NHANES) using the DADA2 dataset with the SV1032 sequence variant removed. Depression was defined using a PHQ-9–based DSM-IV algorithm requiring five or more symptoms, including depressed mood or anhedonia. Four measures of microbial diversity within each participant were compared by depression status using one-way analysis of variance and Cohen’s d. Differences in overall microbial composition between groups were assessed using PERMANOVA across three beta-diversity measures. To evaluate between-person heterogeneity, we compared how far each participant’s microbial profile was from the average profile of their group (the group centroid). Multivariable logistic regression models examined associations between each alpha-diversity measure and depression after adjustment for sociodemographic, behavioral, and oral-health factors. Results: The final sample included 7,681 adults (49.3% female; 4.9% with depression; mean age, 42 years). Participants with depression had lower oral microbial diversity across all four alpha-diversity measures (all p < 0.05), although the differences were small (Cohen’s d < 0.21). Overall microbial composition also differed by depression status, but these differences were small (PERMANOVA R² ≤ 0.0021; all p = 0.001). Participants with depression showed greater variability in microbial composition from one individual to another on two of the three beta-diversity measures (Bray–Curtis, p = 0.010; unweighted UniFrac, p = 0.001; weighted UniFrac, p = 0.337). After adjustment for sociodemographic, behavioral, and oral-health factors, greater alpha diversity remained associated with lower odds of depression across all four measures (all p < 0.05). Conclusions: Depression was associated with lower oral microbial diversity after adjustment for sociodemographic, behavioral, and oral-health factors. Participants with depression also showed greater between-person variation in microbial composition on two of the three beta-diversity measures, suggesting that depression may not be associated with a single, uniform microbial pattern. Although these differences were small, longitudinal studies are needed to determine their direction and underlying mechanisms. This study was supported by NIH/National Institute of Dental and Craniofacial Research.