Logo image
Back
Per- and polyfluoroalkyl substances (PFAS) in U.S. mothers and children: sociodemographic determinants, immune outcomes, and epidemiologic mixture analysis perspectives
Dissertation

Per- and polyfluoroalkyl substances (PFAS) in U.S. mothers and children: sociodemographic determinants, immune outcomes, and epidemiologic mixture analysis perspectives

Jessie A. Gleason
Doctor of Philosophy (Ph.D.), Drexel University
Aug 2026
DOI:
https://doi.org/10.17918/00011589
pdf
Gleason_Jessie_20263.86 MB
PDF Embargoed Access, Embargo ends: 30 Sep 2027

Abstract

Per- and polyfluoroalkyl substances (PFAS) are a large group of synthetic chemicals that are highly persistent in the environment. Several PFAS are detected in blood serum of nearly all people in the United States, and exposure begins in utero. Evidence suggests PFAS may adversely affect child development and immune function. Using data from the NIH Environmental Influences on Child Health Outcomes (ECHO) public-use dataset, this dissertation examined sociodemographic patterns of prenatal PFAS exposure among U.S. pregnant women and evaluated associations between prenatal PFAS exposure and childhood infections and allergic conditions. Compared with White mothers, Black mothers had lower levels of PFOA, PFOS, and PFHxS, and lower detection of PFOSA, EtFOSAA, PFDoDA, and PFPeA and higher detection of PFBS and PFHxA. Compared with White mothers, Asian mothers had higher levels of PFNA, PFDA, and PFUnDA and greater weekly fish consumption did not attenuate associations. Hispanic mothers had lower levels of most of the PFAS analytes studied compared with non-Hispanic mothers. Higher levels of maternal education were associated with increased levels of most PFAS analytes. Child health outcomes were derived from parent-reported surveys collected at age three years and included respiratory infections, respiratory hospitalizations, and allergic conditions. Multivariable logistic regression was used to estimate associations per doubling of PFAS concentrations, adjusting for maternal demographic and behavioral factors and cohort. Overall, maternal PFOS, PFHxS, and PFNA were associated with increased odds of pneumonia and PFOS was further associated with allergic rhinitis and surgery for treatment of any one of three common allergy conditions (tonsil removal, ear tube, and adenoids) select childhood infections and allergic outcomes. Larger cohort studies with medically confirmed outcomes are needed to confirm these findings. Because PFAS occur as complex mixtures, we also conducted a focused review of epidemiologic studies evaluating PFAS mixtures, in relation to birth weight as a case example and commonly studied outcome. Nineteen studies that reported both single-pollutant and mixture analyses were reviewed to compare findings across modeling approaches, mixture composition, and interpretation of model outputs. Most studies reported lower birth weight with increasing PFAS mixture exposure. PFOA and PFOS were identified as important contributors to the overall mixture effect and were also significant in single-pollutant models observed in several studies. Effects sizes of individual PFAS were of similar magnitude to those observed from mixture models, and there was no clear evidence of synergism or antagonism in the overall effect of PFAS mixtures on birth weight, suggesting PFAS have non-interactive effects. Future studies should carefully consider mixture composition and apply interpretable mixture methods, such as quantile g-computation and Bayesian kernel machine regression, to evaluate biologically, toxicologically, and regulatorily relevant PFAS groupings.

Metrics

1 Record Views

Details

Logo image