Accumulating evidence links a range of adverse health effects with exposure to per- and polyfluoroalkyl substances (PFAS), compounds increasingly detected in global drinking water sources. Their strong carbon-fluorine bonds confer resistance to chemicals and extreme temperatures, making them difficult to degrade in the natural environment. Therefore, they disperse through water, persist in soil, and accumulate in animals, leading to their ubiquitous presence. A recent national survey revealed that all participants who provided a serum sample had at least one PFAS found in their blood. PFAS exposure has been linked to multiple health effects, including immune suppression, cancer, elevated cholesterol, and developmental effects. Identifying sources of contamination and exposure pathways is essential for promoting effective public health strategies to mitigate risks and prevent health problems. Individuals can be exposed to harmful compounds through various pathways, such as consuming marine foods, drinking water, and using everyday consumer products. With numerous potential sources of personal exposure, it can be challenging to identify the predominant source for targeted exposure reduction interventions. Therefore, our overall objective was to quantify the contribution of drinking water and dietary factors to PFAS serum concentrations in two populations: a contaminated community and a general population of pregnant women.Chapter 2 applies source apportionment methods, principal component analysis-multiple linear regression (PCA-MLR) and positive matrix factorization (PMF) in a Midwestern community with known PFAS drinking water contamination predominant in perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS) to verify the predominant mixture in blood. Both models identified the predominant mixture in the contaminated historical drinking water as the most potent contributor to serum PFAS. The relative source contribution from drinking water was above 44% for the entire cohort and increased to 54% when restricted to those participants with the highest drinking water concentrations. Chapter 3 extends the investigation to a broader range of exposure factors, including general diet and consumption of local and home-produced foods. Using a bioinformatics approach that applied elastic net with ordinary least square regression and multiple variate regression, drinking tap water was identified as a significant predictor. On average, each additional year of drinking tap water was associated with a 7% increase in serum ΣPFAS (p < 0.01). Consumption of home produced eggs was associated with a 62% increase in serum ΣPFAS for each additional serving per day (p=0.01), whereas egg consumption, in general, was associated with a 54% decrease (p=0.03).Additionally, consumption of self-caught games was associated with a 68% increase in serum PFOA and PFOS (p<0.02). These factors are all consistent with sources of PFAS in drinking water and diet that we have documented in the community. Proceeding to Chapter 4, we investigated dietary predictors, lifestyle, and health condition’s association with serum PFAS among pregnant women. A random forest model was adopted to identify essential features and quantify relative contributions. After controlling for demographic and drinking water covariates, serum ΣPFAS was positively associated with the consumption of other cereal grains and was associated with an increase in serum PFOA. (p=0.01). A positive association was found in pre-prepared meals with sum PFAS, PFOS, PFOA, and PFHxS, though it was nonsignificant. This project enhanced our understanding of the relative contribution of water and diet to the prenatal population. Chapter 5 provides an overview of the research conducted in this thesis and suggests future studies. It highlights the importance of drinking water and diet as sources of PFAS exposure that persist with the phase-out of some PFAS and continued use of many others.
Read
