Developing an Exposure Burden Score for Chemical Mixtures Using Item Response Theory, with Applications to PFAS...

Area of Science:

• Environmental health sciences
• Toxicology
• Biostatistics

Background:

• Quantifying total exposure to chemical mixtures independent of health outcomes is challenging.
• A summary metric for chemical mixture exposure is needed for biomonitoring, risk assessment, and health calculators.

Purpose of the Study:

• To develop and apply a novel exposure burden score method for chemical mixtures.
• To estimate exposure burden for per- and polyfluoroalkyl substances (PFAS) mixtures.
• To assess associations between PFAS burden scores and cardio-metabolic outcomes in the U.S. population.

Main Methods:

• Applied item response theory (IRT) to National Health and Examination Survey (NHANES) data (2017-2018) from 1,915 participants.
• Quantified a latent PFAS burden score using serum concentrations of eight PFAS biomarkers.
• Used linear regression to estimate associations between PFAS burden scores and cardio-metabolic outcomes, comparing results to summed PFAS concentrations.

Main Results:

• PFAS burden scores and summed PFAS concentrations showed moderate-to-high correlation.
• PFAS burden scores demonstrated generally stronger associations with cardio-metabolic outcomes (e.g., total cholesterol) compared to summed concentrations.
• Isomers of PFOS were most informative for the PFAS burden scores.

Conclusions:

• Item response theory (IRT) provides a robust method for quantifying chemical mixture exposure burden, reducing measurement error.
• The IRT-based PFAS burden score enhances the ability to detect associations with health outcomes.
• An accompanying PFAS burden calculator is available to facilitate broader research applications.