Jiwon Oh

ORCID Identifier(s)


Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Earth and Environmental Science


Earth and Environmental Sciences

First Advisor

Hyeong-Moo Shin


Pregnant women are exposed to per- and polyfluoroalkyl substances (PFAS) primarily via ingestion of contaminated food, water, and house dust. PFAS in pregnant women can be transferred to their child through the placenta during pregnancy. Many long-chain PFAS have been frequently detected in blood of pregnant women, amniotic fluid, and umbilical cord blood for the last few decades. Prenatal exposure to PFAS has shown potential to adversely affect offspring’s neurodevelopment in laboratory animals. However, epidemiological evidence on associations between prenatal PFAS exposure and child neurodevelopment remains inconsistent. My dissertation research consists of four sub-studies whose objectives are (1) to examine associations between prenatal maternal serum PFAS concentrations and the risk of autism spectrum disorder (ASD), (2) to examine associations between prenatal exposure to PFAS and cognitive development in infancy and toddlerhood, (3) to examine associations between cord blood PFAS concentrations and cognitive development in infancy and toddlerhood, and (4) to investigate longitudinal changes in maternal PFAS concentrations from pregnancy to two years postpartum. The first, second, and fourth studies used data from the MARBLES (Markers of Autism Risk in Babies – Learning Early Signs), a longitudinal cohort of children with a first degree relative who was diagnosed with ASD. The MARBLES children were administered to Mullen Scales of Early Learning (MSEL) for cognitive functions at 6, 12, 24, and 36 months of age and clinically diagnosed with ASD at 36 months of age. Nine PFAS were quantified in maternal serum collected during the 1st, 2nd, and 3rd trimester of pregnancy and at 3, 6, and 24 months after delivery. In the first study, I performed Poisson regression analyses to examine associations of ASD risk with individual PFAS as well as combined PFAS using a principal component analysis (PCA). I observed that prenatal maternal serum concentrations of perfluorooctanoate (PFOA) and perfluorononanoate (PFNA) were associated with increased risk of child ASD. In the second study, I used multiple linear regression models to examine cross-sectional associations of PFAS with the MSEL Composite and four subscale scores at each time point and generalized estimating equations to examine associations between PFAS and longitudinal changes in the MSEL scores over the four assessment time points. I also classified trajectories of the MSEL Composite scores into low- and high-score groups and fit Poisson regression models to estimate relative risk (RR) of a high-score group versus a low-score group. I observed that prenatal maternal serum PFOA concentrations were inversely associated with child MSEL Composite scores at 24 and 36 months of age. When assessing longitudinal changes in the scores over the four time points, PFOA was associated with a negative slope for Composite scores and all four subscales. When examining RR between low- and high-score groups, PFOA was associated with increased risk of having lower and/or decreasing Composite scores. In the third study, I used a Japanese population to confirm the findings from the second study. I used the same statistical models to examine the cross-sectional and longitudinal associations between cord blood PFOA and PFOS concentrations and MSEL Composite and subscale scores assessed at 4, 6, 10, 14, 18, 24, 32, and 40 months of age. MSEL Composite scores were inversely associated with cord blood PFOA at 18 months of age, but not at other ages. When accounting for changes in scores from 4 to 40 months of age, PFOA and PFOS were inversely associated with Fine Motor scores and positively associated with Receptive Language scores. In the fourth study, I fit separate linear mixed models during pregnancy, early postpartum (from delivery to 6 months postpartum) and late postpartum (from 6 to 24 months postpartum) to estimate percent changes for each sub-period. During pregnancy, linear and branched perfluorooctane sulfonate (n- and Sm-PFOS), linear PFOA (n-PFOA), and PFNA concentrations changed -4% to -3% per month. During early postpartum, perfluorohexane sulfonate and n-PFOA concentrations changed -6% and -4%, respectively, per month, and Sm-PFOS and PFNA concentrations changed -1% per month. During late postpartum, n-PFOS, Sm-PFOS, and PFNA concentrations changed -1% per month.


Per- and polyfluoroalkyl substances, Prenatal exposure, Autism spectrum disorder, Cognitive functions, Concentration changes, Pregnancy, Early postpartum, Late postpartum


Earth Sciences | Physical Sciences and Mathematics


Degree granted by The University of Texas at Arlington