Characterising oral microbial signatures for periodontal disease in the NHANES population.

BackgroundThe oral microbiome has been increasingly recognized for its role in systemic health through the oral–lung axis. However, population-level evidence linking oral microbial diversity and composition with chronic respiratory diseases (CRD) remains limited.MethodsWe analyzed data from 4,384 adults in the 2009–2012 National Health and Nutrition Examination Survey (NHANES), defining CRD by self-reported chronic obstructive pulmonary disease (COPD), asthma, emphysema, or chronic bronchitis. Oral rinse samples underwent 16S ribosomal RNA (16S rRNA) V1–V3 sequencing. Alpha diversity, including observed amplicon sequence variants (ASVs), Faith’s phylogenetic diversity (Faith’s PD), Shannon–Weiner index, and Simpson index, and beta diversity, including Bray–Curtis, weighted UniFrac, and unweighted UniFrac distances, were assessed. Associations with CRD were examined using weighted logistic regression and restricted cubic splines (RCS). Differential genus abundance was identified by Wilcoxon tests with false discovery rate correction. A random forest model integrated microbial and clinical features. An independent hospital cohort was additionally profiled by 16S rRNA sequencing, and genus-level differences were assessed with linear discriminant analysis effect size (LEfSe) to validate NHANES findings.ResultsHigher alpha diversity was inversely associated with CRD risk; each standard deviation increase in observed ASVs and Faith’s PD reduced CRD odds by 19 and 17%, respectively (p < 0.05). Beta diversity showed significant community-level separation by CRD status (p = 0.01). Several genera, including Rothia and Veillonella, were enriched in CRD, whereas Prevotella, Haemophilus, and Neisseria were more abundant in non-CRD individuals. The random forest model achieved an area under the curve (AUC) of 0.65. In the hospital cohort, compositional shifts were consistent with NHANES findings, and LEfSe confirmed the depletion of Alloprevotella and Peptostreptococcus in CRD patients.ConclusionOral microbial diversity and composition were significantly associated with CRD across both a representative U. S. population and a hospital cohort. Select genera and diversity indices may serve as non-invasive biomarkers for respiratory health, warranting further validation in longitudinal and mechanistic studies.

Background and Objective: Accumulating evidence has linked periodontal disease with hypertension, yet the relationship between the diversity of oral microbiota and blood pressure remains unclear. Our objective was to investigate the association between characteristics of oral microbiome diversity and hypertension in the general population. Methods: The data were extracted from the 2009-2012 National Health and Nutrition Examination Survey (NHANES), where we included participants aged 20 years or older who had indicators of hypertension. Participants with conditions such as pregnancy, periodontal disease, cardiovascular diseases (CVD), chronic obstructive pulmonary disease (COPD), diabetes, and cancer were excluded from the analysis. Oral microbiome α-diversity in adults was analyzed through oral rinse samples, encompassing measures such as observed amplicon sequence variants (ASVs), Faith’s Phylogenetic Diversity, the Shannon-Wiener Index, and the Simpson Index. Differences in oral microbiome α-diversity among various hypertension groups were described using weighted mean values. The association between oral microbiome α-diversity and hypertension was analyzed using multivariable logistic regression models. Results: Between 2009 and 2012, among 3,228 participants (weighted mean [SE] age, 36.54 [0.64] years; 1,408 males [weighted 43.62%]), there were 601 cases of hypertension (increasing to 771 when adopting the new diagnostic criteria for hypertension). Significant differences were observed in the levels of oral microbiota α-diversity among different hypertension groups. Multivariate logistic regression analyses revealed an independent association between levels of oral microbiome α-diversity and the risk of hypertension, after controlling for confounding factors. For each one-standard deviation (SD) increase in ASVs and Faith's Phylogenetic Diversity, the adjusted odds ratios (aORs) were 0.85 (95% confidence interval [CI]: 0.75-0.95) and 0.85 (95% CI: 0.75-0.95), respectively. This association persisted across varying diagnostic criteria for hypertension. Conclusions: Our findings indicate diverse oral microbiomes among hypertension groups, where greater α-diversity inversely correlates with hypertension risk, independent of diagnostic criteria. This implicates oral microbiota dysbiosis, potentially from environmental factors, as an additional hypertension risk factor, emphasizing the oral microbiome's potential in hypertension development.

ObjectivesThe oral microbiota plays a significant role in oral health. The present study aims to characterize variations in the oral microbiota relative to the collection site, the dynamics of biofilm accumulation, and inherent inter-individual differences.MethodsWhole stimulated saliva and tooth biofilm samples from the 16 defined tooth regions were collected after 1, 2, or 3 days without oral hygiene (accumulation time) in six healthy adults with no signs of active caries or periodontal disease. The routines and conditions before and between sample collections were carefully standardized. Genomic DNA was extracted, and the V3-V4 regions of the 16S rRNA gene were amplified by PCR and sequenced on an Illumina MiSeq platform. Sequences were quality controlled, amplicon sequence variants (ASVs) were clustered, and taxonomic allocation was performed against the expanded Human Oral Microbiome Database (eHOMD). Microbial community profiles were analyzed by multivariate modeling and a linear discriminant analysis (LDA) effect size (LEfSe) method.ResultsThe overall species profile in saliva and tooth biofilm differed between participants, as well as sample type, with a significantly higher diversity in tooth biofilm samples than saliva. On average, 45% of the detected species were shared between the two sample types. The microbiota profile changed from the most anterior to the most posterior tooth regions regardless of whether sampling was done after 1, 2, or 3 days without oral hygiene. Increasing accumulation time led to higher numbers of detected species in both the saliva and region-specific tooth biofilm niches.ConclusionThe present study confirms that the differences between individuals dominate over sample type and the time abstaining from oral hygiene for oral microbiota shaping. Therefore, a standardized accumulation time may be less important for some research questions aiming at separating individuals. Furthermore, the amount of DNA is sufficient if at least two teeth are sampled for microbiota characterization, which allows a site-specific characterization of, for example, caries or periodontitis.

It has been increasingly recognized that adults living alone have a higher likelihood of developing Major Depressive Disorder (MDD) than those living with others. However, there is still no prediction model for MDD specifically designed for adults who live alone. This study aims to investigate the effectiveness of utilizing personal health data in combination with a stacked ensemble machine learning (SEML) technique to detect MDD among adults living alone, seeking to gain insights into the interaction between personal health data and MDD. Our data originated from the US National Health and Nutrition Examination Survey (NHANES) spanning 2007 to 2018. We finally selected a set of 30 easily accessible variables encompassing demographic profiles, lifestyle factors, and baseline health conditions. We constructed a SEML model for MDD detection, incorporating three conventional machine learning algorithms as base models and a Neural Network (NN) as the meta-model. Furthermore, SHapley Additive exPlanations (SHAP) analysis was used to explain the impact of each predictor on MDD. The study included 2,642 adult participants who lived alone, of whom 10.6% (279 out of 2,642) had a PHQ-9 score of 10 or above, indicating the presence of MDD. The performance of our SEML model was robust, with an area under the curve (AUC) of 0.85. Further analysis using SHAP revealed positive correlations between the occurrence of MDD and factors such as sleep disorders, number of prescription medications, need for specific walking aids, leak urine during nonphysical activities, chronic bronchitis, and Healthy Eating Index (HEI) scores for sodium. Conversely, age, the Family Monthly Poverty Level Index (FMMPI), and HEI scores for added sugar showed negative correlations with MDD occurrence. Additionally, a U-shaped relationship was noted between the occurrence of MDD and both sleep duration and Body Mass Index (BMI), as well as HEI scores for dairy. The study has successfully developed a predictive model for MDD, specifically tailored for adults living alone using a stacked ensemble technique, enhancing the identification of MDD and its risk factors among adults living alone.

Relationships of triglyceride–glucose index and body mass index with 5-year all-cause mortality in patients with diabetes and comorbid hypertension: Evidence from two prospective cohort studies

Asthma diagnosis poses challenges due to underreporting of symptoms, misdiagnoses, and limitations in existing diagnostic tests. Machine learning (ML) offers a promising avenue for addressing these challenges by leveraging demographic and clinical data. In this study, we aim to compare different ML diagnostic models and obtain the most valuable features for asthma diagnosis using data from the National Health and Nutrition Examination Survey (NHANES) dataset. A total of 8,888 participants with available asthma diagnosis data from the 2017–2018 NHANES survey were included. After careful selection of variables related to asthma, various ML algorithms including Support Vector Machine (SVM), Random Forest (RF), AdaBoost (ADA), XGBoost (XGB), K-Nearest Neighbors (KNN), Naive Bayes (NB), and Multi-Layer Perceptron (MLP) were evaluated. SVM and ADA emerged as top performers with the highest area under the curve (AUC) scores of 0.72 and 0.71, respectively. RF exhibited high accuracy but low precision. Feature interpretation using SHapley Additive exPlanations (SHAP) values identified significant predictors such as close relative asthma history, dietary fat intake, and chronic bronchitis. Feature reduction experiments showed promising results without significant loss in predictive performance. Our findings demonstrate the potential diagnosis ability of ML algorithms, particularly SVM and ADA, in asthma diagnosis by incorporating diverse clinical and demographic factors. In addition, close relative asthma history, dietary fat intake, and chronic bronchitis could be suggested as the valuable asthma diagnosis features. These outcomes can bring promising results in early diagnosis of asthma.

Background and Aims: The oral microbiome, comprising over 700 bacterial species along with a variety of viruses and other microbial taxa, is colonized as early as the prenatal period and is continuously shaped by various factors throughout life. While its primary role in maintaining oral health is well-recognized, recent studies suggest that the oral microbiome may also play a significant role in affecting systemic health condition, whereas the evidence from human studies is still lacking. We aimed to investigate whether oral microbiome alpha diversity was significantly related to premature mortality among U.S. adults. Materials and Methods: This study utilized data from 7,956 participants in the 2009–2010 and 2011–2012 cycles of the National Health and Nutrition Examination Survey (NHANES). Oral microbiome alpha diversity metrics, including observed Amplicon Sequence Variants (ASVs), Faith’s Phylogenetic Diversity, the Shannon-Weiner Index, and the Simpson Index, were calculated using microbial sequences obtained from oral rinse samples. Premature mortality was defined as death occurring before the age of 80 and was ascertained through linkage with the National Death Index till December 31, 2019. Results: During an average follow-up period of 8.84 years (SD: 1.54 years), 436 premature deaths (5.48%) were recorded among the 7,956 participants. After adjusting for demographic factors, smoking, alcohol drinking, and extra dental cleansing, higher alpha diversity metrics were significantly associated with a reduced hazard of premature mortality. Specifically, for each standard deviation increase, the HR were 0.81 (95% CI: 0.71–0.93) for observed ASVs, 0.82 (95% CI: 0.71–0.94) for Faith’s Phylogenetic Diversity, 0.83 (95% CI: 0.73–0.96) for the Shannon-Weiner Index, and 0.82 (95% CI: 0.72–0.95) for the Simpson Index. Adjustments for dietary factors attenuated these associations to be null. Conclusion: Our findings suggest that alpha diversity of the oral microbiome may be not a strong risk factor of premature mortality independently of dietary factors among U.S. adults.

Asthma exacerbations significantly contribute to morbidity and healthcare burden, yet accurate risk prediction remains challenging. This study aimed to develop and validate a machine learning (ML)-based model to predict the exacerbation risk in adult asthma patients. This study analyzed data from the National Health and Nutrition Examination Survey (NHANES) collected between 2007 and 2012, comprising a cohort of 1,480 adult participants diagnosed with asthma. A total of 37 candidate features were assessed, and feature selection via least absolute shrinkage and selection operator (LASSO) identified seven key predictors. Nine ML models were developed and evaluated using the area under the curve (AUC) to determine the optimal model. The best-performing model underwent further validation using calibration curves, precision-recall (PR) curves, and decision curve analysis (DCA). Shapley Additive Explanations (SHAP) was applied to interpret the model's predictions. A light gradient boosting machine (LightGBM) model showed the best predictive performance, with an AUC of 0.902 in the training set and comparable discrimination in the validation set. Model performance in the validation cohort showed consistent results across calibration analysis, PR curves, and DCA. Model interpretability was examined using SHAP and a web-based calculator was developed to support individualized risk estimation. This ML-based model demonstrated strong predictive accuracy and could serve as a valuable tool for risk assessment of adult asthma in clinical practice.

This study develops a dual-level machine learning framework for risk stratification and phenotyping of hearing loss (HL) in patients with chronic kidney disease (CKD) using data from the National Health and Nutrition Examination Survey (NHANES). From a cohort of 3,402 CKD patients, feature selection via univariate and multivariate logistic regression identified key predictors, which were used to construct predictive models with nine machine learning algorithms. The eXtreme Gradient Boosting (XGBoost) model demonstrated superior performance, with mean area under the curve (AUC) values of 0.984 (training), 0.984 (validation), and 0.939 (testing). SHapley Additive exPlanations (SHAP) interpretation identified age as the predominant risk determinant. Subsequent Gaussian mixture modeling (GMM) clustered patients into two distinct subtypes: a low-risk subgroup (n = 1,075) with a 1.58% HL prevalence and a high-risk subgroup (n = 2,316) characterized by older age, elevated blood urea nitrogen and bicarbonate levels, and a 48.2% HL prevalence. A classifier trained on these subtypes achieved discrimination (AUC = 0.99974). A clinically a web-based tool was also developed based on the six most influential features. The findings establish a dual-level predictive framework integrating explainable ML and unsupervised clustering for HL risk assessment in CKD. This approach provides a robust strategy for the precision screening of high-risk subpopulations and supports the integration of hearing assessments into routine CKD.

The ratio of red blood cell distribution width (RDW) to albumin concentration (RAR) has emerged as a reliable prognostic marker for mortality in patients with various diseases. However, whether RAR is associated with mortality in the general population remains unknown. To explore whether RAR is associated with all-cause and cause-specific mortality and to elucidate their dose-response association. This population-based prospective cohort study used data from participants in the 1998-2018 US National Health and Nutrition Examination Survey (NHANES) and from the UK Biobank with baseline information provided from 2006 to 2010. Included participants had complete data on serum albumin concentration, RDW, and cause of death. The NHANES data were linked to the National Death Index records through December 31, 2019. For the UK Biobank, dates and causes of death were obtained from the National Health Service Information Centre (England and Wales) and the National Health Service Central Register Scotland (Scotland) to November 30, 2022. Potential associations between RAR and the risk of all-cause and cause-specific mortality were evaluated using Cox proportional hazards regression models. Restricted cubic spline regressions were applied to estimate possible nonlinear associations. In NHANES, 50 622 participants 18 years of age or older years were included (mean [SD] age, 48.6 [18.7] years; 26 136 [51.6%] female), and their mean (SD) RAR was 3.15 (0.51). In the UK Biobank, 418 950 participants 37 years of age or older (mean [SD], 56.6 [8.1] years; 225 038 [53.7%] female) were included, and their mean RAR (SD) was 2.99 (0.31). The NHANES documented 7590 deaths over a median (IQR) follow-up of 9.4 (5.1-14.2) years, and the UK Biobank documented 36 793 deaths over a median (IQR) follow-up of 13.8 (13.0-14.5) years. According to the multivariate analysis, elevated RAR was significantly associated with greater risk of all-cause mortality (NHANES: hazard ratio [HR], 1.83 [95% CI, 1.76-1.90]; UK Biobank: HR, 2.08 [95% CI, 2.03-2.13]), as well as mortality due to malignant neoplasm (NHANES: HR, 1.89 [95% CI, 1.73-2.07]; UK Biobank: HR, 1.93 [95% CI, 1.86-2.00]), heart disease (NHANES: HR, 1.88 [95% CI, 1.74-2.03]; UK Biobank: HR, 2.42 [95% CI, 2.29-2.57]), cerebrovascular disease (NHANES: HR, 1.35 [95% CI, 1.07-1.69]; UK Biobank: HR, 2.15 [95% CI, 1.91-2.42]), respiratory disease (NHANES: HR, 1.99 [95% CI, 1.68-2.35]; UK Biobank: HR, 2.96 [95% CI, 2.78-3.15]), diabetes (NHANES: HR, 1.55 [95% CI, 1.27-1.90]; UK Biobank: HR, 2.83 [95% CI, 2.35-3.40]), and other causes of mortality (NHANES: HR, 1.97 [95% CI, 1.86-2.08]; UK Biobank: HR, 2.40 [95% CI, 2.30-2.50]) in both cohorts. Additionally, a nonlinear association was observed between RAR levels and all-cause mortality in both cohorts. In this cohort study, a higher baseline RAR was associated with an increased risk of all-cause and cause-specific mortality in the general population. These findings suggest that RAR may be a simple, reliable, and inexpensive indicator for identifying individuals at high risk of mortality in clinical practice.

Association between the cardiometabolic index and mortality risk in US adults: Data from the National Health and Nutrition Examination Survey (NHANES) 1999-2018.

Health care practitioners who care for adolescents transitioning to adulthood often face incongruent recommendations from pediatric and adult guidelines for treatment of lipid levels. To compare the proportion of young people aged 17 to 21 years who meet criteria for pharmacologic treatment of elevated low-density lipoprotein cholesterol (LDL-C) levels under pediatric vs adult guidelines. We performed a cross-sectional analysis of the National Health and Nutrition Examination Survey (NHANES) population. Surveys were administered from January 1, 1999, through December 31, 2012, and the analysis was performed from June through December 2014. Participants included 6338 individuals aged 17 to 21 years in the United States. To estimate the number and proportion of individuals aged 17 to 21 years in the NHANES population who were eligible for statin therapy, we applied treatment algorithms from the 2011 Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents of the National Heart, Lung, and Blood Institute and the 2013 Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults from the American College of Cardiology and American Heart Association. After imputing missing data and applying NHANES sampling weights, we extrapolated the results to 20.4 million noninstitutionalized young people aged 17 to 21 years living in the United States. Of the 6338 young people aged 17 to 21 years in the NHANES population, 2.5% (95% CI, 1.8%-3.2%) would qualify for statin treatment under the pediatric guidelines compared with 0.4% (95% CI, 0.1%-0.8%) under the adult guidelines. Participants who met pediatric criteria had lower mean (SD) LDL-C levels (167.3 [3.8] vs 210.0 [7.1] mg/dL) but higher proportions of other cardiovascular risk factors, including hypertension (10.8% vs 8.4%), smoking (55.0% vs 23.9%), and obesity (67.7% vs 18.2%) compared with those who met the adult guidelines. Extrapolating to the US population of individuals aged 17 to 21 years represented by the NHANES sample, 483 500 (95% CI, 482 100-484 800) young people would be eligible for treatment of LDL-C levels if the pediatric guidelines were applied compared with only 78 200 (95% CI, 77 600-78 700) if the adult guidelines were applied. Application of pediatric vs adult guidelines for lipid levels, which consider additional cardiovascular risk factors beyond age and LDL-C concentration, might result in statin treatment for more than 400 000 additional adolescents and young adults.

Historic evidence suggests that use of high-dose combined oral contraceptives (OCs) (containing >50 microg of estrogen and>or=1mg progestin) places women at increased risk for periodontal diseases. Since the mid-1970s, OC formulations have dramatically changed. This study investigated the association between OC use and periodontal diseases among 4,930 National Health and Nutrition Examination Survey (NHANES) I and 5,001 NHANES III premenopausal U.S. women, aged 17 to 50 years, before and after the reduction of hormone levels in OCs. Data for this cross-sectional study came from the first (NHANES I, 1971 to 1974) and third (NHANES III, 1988 to 1994) NHANES studies. The prevalence of OC use in the U.S. premenopausal female population in NHANES I was 22% and in NHANES III, 20%. Using multivariable logistic regression, a protective association between current OC use and gingivitis was suggestive but not significant in both NHANES I (odds ratio [OR]=0.65; 95% con- fidence interval [CI]: 0.42 to 1.01) and NHANES III (OR=0.80; 95% CI: 0.61 to 1.02) surveys. Current OC use was also associated with a decreased risk of periodontal disease in NHANES I (OR=0.36; 95% CI: 0.13 to 0.96) and a non-significant association in NHANES III (OR=0.73; 95% CI: 0.50 to 1.07). This analysis failed to validate the theory that earlier high- or current low-dose OC use is associated with increased levels of gingivitis or periodontitis and suggests an important reexamination of the perceived association between OC use and periodontal diseases.

Periodontal Disease as a Marker of Malnutrition Risk in Community-Dwelling Older Adults.

Periodontal diseases in the Australian adult population