UC Berkeley

Prenatal organophosphate (OP) pesticide exposure has been associated with adverse neurodevelopment, including decreased cognition, increased hyperactivity and inattention, and higher risk for autism-related traits in multiple studies. However, OP pesticide use has decreased in recent decades and data gaps exist regarding the neurodevelopmental impacts of current-use pesticides, such as neonicotinoids, glyphosate, and pyrethroids. Moreover, most people, particularly those living in agricultural areas, are exposed to multiple pesticides and research is needed to examine the effects of these mixtures that may interact synergistically to adversely impact health and neurodevelopment. Mounting evidence also suggests that the effects of environmental toxicants are due in part to causal interactions with social stressors and biologic factors, e.g., genetics, however studies investigating the neurodevelopmental impacts of environmental exposures have typically treated social factors as confounders. Failure to account for potential effect modification by these factors may underestimate the impact of environmental neurotoxicants, particularly among the most vulnerable populations where exposures to environmental and non-chemical stressors are likely to co-occur. Additionally, previous exposure assessment and epidemiology studies investigating pesticide exposures and subsequent health effects have largely relied on the analysis of dialkylphosphate (DAPs) metabolites, non-specific biomarkers of OP pesticide exposure, from random spot urine samples. DAPs have higher inter- and intra-individual variability and data gaps exist regarding the extent to which concentrations from spot samples may approximate internal dose from the “gold standard” 24-hour urine samples, which has implications for pesticide risk assessment and the establishment of regulatory guidelines. This dissertation aims to examine the validity of using DAPs assessed from random spot urine samples to estimate cumulative OP pesticide dose, the associations between applications of mixtures of agricultural pesticides near the home during pregnancy and early childhood with adolescent neurobehavioral functioning, and the joint effects of use of mixtures of agricultural pesticides near the home during pregnancy and adverse childhood experiences (ACEs) on adolescent neurobehavioral outcomes. Chapter 1 provides a general introduction to human exposure to agricultural pesticides and highlights the background, significance, and specific aims for each study/chapter. Chapter 2 examines the validity of using first morning void (FMV) and random non-FMV urine samples to estimate cumulative 24-hr OP pesticide dose among participants in the Child Validation Study (CVS). For this study, investigators collected urine samples over seven consecutive days, including two 24-hr samples, from 25 children living in the agricultural Salinas Valley, California. These analyses employed measurements of urinary DAP metabolites, data on nearby agricultural pesticide applications, and daily dietary intake data to estimate internal dose from exposure to a mixture of OP pesticides according to the United States (US) Environmental Protection Agency (EPA) Cumulative Risk Assessment guidelines. Dose estimates from volume- and creatinine-adjusted same-day FMV and non-FMV spot urine samples were compared to the “gold standard” estimates from 24-hr samples. Non-FMV samples had relatively weak ability to predict 24-hr dose (R2=0.09-0.38 for total DAPs) and tended to underestimate the percentage of samples exceeding regulatory guidelines. Models with FMV samples or the average of an FMV and non-FMV sample were similarly predictive of 24-hr estimates (R2 for DAPs=0.40-0.68 and 0.40-0.80, respectively, depending on volume adjustment method). Findings for these analyses suggest that reliance on non-FMV samples for risk assessments may underestimate daily OP dose and the percentage of children with dose estimates exceeding regulatory guidelines. Chapter 3 examines associations of applications of mixtures of agricultural pesticides within 1 kilometer (km) of the home during pregnancy and early childhood (ages 0-5 years) and adolescent internalizing and externalizing behaviors in the Center for the Health Assessment of Mothers and Children of Salinas (CHAMACOS) cohort. These analyses employed linear mixed effects Bayesian Hierarchical Models (BHM) to examine associations with maternal- and youth-reported behavioral and emotional problems from the Behavior Assessment System for Children, 2nd edition (BASC-2) at ages 16 and 18 years (n=593). Associations between pesticide applications and neurobehavioral outcomes were largely null. There were some trends of modestly increased internalizing behaviors and attention problems in association with OP insecticide use near the home during the prenatal period. In the postnatal period, a two-fold increase in glyphosate applications was associated with more youth-reported depression (β=1.2; 95% Credible Interval (CrI): 0.2, 2.2) and maternal-reported internalizing behaviors (β=1.23; 95% CrI: 0.2, 2.3) and anxiety (β=1.2; 95% CrI: 0.2, 2.3). There were some protective associations with imidacloprid, a neonicotinoid, during the prenatal period, particularly in sex-specific analyses. This study extends previous work by considering the neurobehavioral effects of potential exposure to mixtures of pesticides. Chapter 4 examines interactions of applications of pesticide mixtures near the home during pregnancy and childhood adversity with adolescent neurobehavioral outcomes among CHAMACOS participants. These analyses employed linear mixed effects BHM to examine the joint effect of applications of 11 agricultural within 1 km of maternal homes during pregnancy and youth-reported Adverse Childhood Experiences (ACEs) with maternal and youth-reported internalizing behaviors, hyperactivity, and attention problems at ages 16 and 18 years. Overall, there was little evidence of modification of exposure-outcome associations by ACEs. Malathion use near the home during pregnancy with associated with increased internalizing behaviors among those with high ACEs from both maternal report (β = 1.9; 95% CrI: 0.2, 3.7 for high ACEs [3+] vs. β = -0.1; 95% (CrI): -1.2, 0.9 for low ACEs [0-2]) and youth report (β = 2.1; 95% CrI: 0.4, 3.8 for high ACEs vs. β = 0.2; 95% CrI: -0.8, 1.2 for low ACEs). Results were stronger among males for both maternal and youth report of internalizing behaviors. Applications of malathion and dimethoate were also associated with higher youth-reported hyperactivity and/or inattention among those with high ACEs. There was no evidence of effect modification by ACEs for any other pesticides. This analysis builds upon previous studies by considering co-exposure to mixtures of agricultural pesticides and social adversity. It is the first to examine interactions of chemical and non-chemical stressors on neurobehavioral development assessed during adolescence or early adulthood. Chapter 5 highlights the major findings for each chapter/study, conclusion, and future directions.