Heavy metal exposure and dental caries in children and adolescents: a NHANES-based analysis integrating machine learning approaches

Dental caries remains a major public health concern among children and adolescents. While established risk factors such as diet and oral hygiene are well recognized, the association between environmental heavy metal exposure and pediatric dental caries is not fully understood. This study examined the relationship between blood heavy metal concentrations and dental caries in a nationally representative pediatric population. Data from 15,194 participants aged 5–19 years were obtained from the National Health and Nutrition Examination Survey (NHANES). Dental caries was assessed using the DMFT/S index. Blood lead, cadmium, and mercury concentrations were measured by inductively coupled plasma mass spectrometry. Multivariable logistic regression models were used to evaluate associations after adjustment for sociodemographic, nutritional, and hematological covariates. Multiple machine learning models were constructed to assess predictive performance, and SHAP values were applied to interpret feature importance. Higher blood concentrations of lead, cadmium, and mercury were independently associated with increased odds of dental caries (lead: OR = 1.10, 95% CI: 1.02–1.18; cadmium: OR = 1.18, 95% CI: 1.10–1.27; mercury: OR = 1.14, 95% CI: 1.06–1.22). Subgroup analyses demonstrated heterogeneity by age and socioeconomic status, with stronger associations for cadmium observed among adolescents and higher-income groups. Among the machine learning models, XGBoost showed the best predictive performance (AUC = 0.75), with cadmium ranking as the most influential feature. Blood heavy metal concentrations were significantly associated with dental caries in children and adolescents, suggesting that environmental exposure may be an important correlate of pediatric caries risk. 1. Elevated blood lead, cadmium, and mercury levels were independently associated with increased dental caries likelihood in children and adolescents, even after adjusting for sociodemographic and nutritional confounders. 2. Cadmium exhibited stronger associations with caries in adolescents and high-income subgroups, highlighting age- and socioeconomic-specific heterogeneity. 3. XGBoost outperformed other machine learning models (AUC = 0.75), with cadmium identified as the most influential predictor through SHAP value analysis. 4. Associations between heavy metals and dental caries were observed across income strata, including higher-income populations. 5. These findings highlight environmental heavy metal exposure as a potential, yet underrecognized, correlate of pediatric dental caries risk.