Age-Based and Time Trends of Chemical Exposure Biomarkers in US Population 1999-2014
Individuals are exposed to complex chemical mixtures driven by behavioral, socioeconomic, and demographic factors. To systematically address the interactions between chemical exposure and demographic and lifestyle factors, we have developed an untargeted approach to study the 1999-2014 National Health and Nutrition Examination Survey (NHANES) data of 229 chemical biomarkers in 82,091 study participants. The present study aims: 1) to determine chemical exposure differences between young and older individuals; 2) to identify chemicals with increasing exposure in the US population by determining biomarker time trends; and 3) to evaluate key determinants that explain these patterns, particularly considering half-lives in human. We use a series of multivariate linear regression models that include age, sex, study years, ethnicity, smoking behaviors, poverty income ratio, and creatinine levels as covariates and chemical biomarker concentration as the outcome. Metabolites from chemicals in personal care products such as benzophenone-3, parabens, and triclosan are higher in young individuals, while polychlorinated biphenyls, cadmium, and trans-nonalchor are elevated in the older population. Concentrations of 1-pyrene, mono-isobutyl phthalate, 2-phenanthrene, 2-napthol, and perfluorononanoic acid are significantly increasing over time. Correlation analysis between the half-lives in human and age coefficient (yearly increase in the log-transformed biomarker concentration) demonstrates a positive correlation with R2 = 0.5 when considering over 50 chemicals with half-lives of more than 1000 hours. Systematically studying the NHANES biomarker dataset by applying an untargeted approach allows for the identification of expected and unexpected exposure trends by age and over time. These findings can be utilized to prioritize chemicals for toxicological evaluation or targeted environmental health interventions.