Exposome-wide association study of thyroid function using U.S. National Health and Nutrition Examination Survey data

Previous epidemiologic studies examining thyroid function and chemical exposures have typically focused on a single or a limited number of chemical classes, often neglecting the effects of chemical mixtures. This study addressed this gap by exploring the associations between exposure to hundreds of chemicals and thyroid function using an exposome-wide association study (ExWAS) approach and National Health and Nutrition Examination Survey (NHANES) data. We analyzed data from three NHANES cycles (2007-2008, 2009-2010, and 2011-2012), which include measures of thyroid function (free and total triiodothyronine [T3], free and total thyroxine [T4], thyroid-stimulating hormone [TSH]) and chemical biomarker concentrations from 9,082 participants. For adolescents (aged 12-19 years) and adults (aged ≥20 years), we employed multiple regression by accounting for survey weights to identify biomarkers associated with thyroid function test levels and used Bayesian group weighted quantile sum (BGWQS) regression to assess the effects of chemical mixtures on these measurements. After adjusting for multiple comparisons, we found in single exposure scenarios that 44 and 67 biomarkers were associated with at least one thyroid function measure in adolescents and adults, respectively (adjusted p-value <0.05). In scenarios involving mixed chemical exposures, groups such as pesticides, sodium/iodide symporter (NIS) inhibitors, and metals were associated with alterations in thyroid hormones or TSH across both age groups. Volatile organic compounds were specifically linked to lower T4 levels in adolescents, whereas phenols and parabens were associated with lower TSH levels exclusively in adults. Although limited by the cross-sectional data, this study identified chemical biomarkers linked to thyroid function.