A Data-driven Framework for Modeling Environmental Exposure Mixtures, Biological Aging Acceleration, and Chronic Disease Risk in U.S. Adults

Environmental exposure biomarkers (EEBs) reflect the internal burden of pollutants, yet the joint effects of multiple exposures on biological aging and chronic disease risk remain insufficiently characterized. We analyzed 8,582 adults from the 2013-2016 National Health and Nutrition Examination Survey (NHANES). Mixed exposure was characterized using 74 EEBs. Phenotypic age acceleration and biological age acceleration were used as aging outcomes. Weighted quantile sum (WQS) regression, Bayesian kernel machine regression (BKMR), and LASSO regression were applied to identify key exposure components associated with aging acceleration. Logistic and Cox regression models were then used to evaluate the associations between aging indicators and chronic disease risks. Higher mixed EEB exposure was significantly associated with accelerated aging, reflected by increases in both phenotypic age and biological age. WQS models identified arsenobetaine, copper, and tin as major contributors to phenotypic age acceleration, whereas selenium, zinc, and MHNCH contributed most strongly to biological age acceleration. Moreover, each one-year increase in phenotypic age was associated with a 30.0\% higher risk of dyslipidemia and a 14.3\% higher risk of metabolic-associated fatty liver disease, while each one-year increase in biological age was associated with a 59.7\% higher risk of chronic obstructive pulmonary disease and a 48.5\% higher risk of anemia. This study proposes a unified data-driven analytical framework that integrates exposure mixtures, biological aging, and disease risk modeling. The findings highlight the importance of evaluating mixed exposures rather than single pollutants and may support risk stratification and prevention strategies in environmental health.