Publication Date

8-2024

Date of Final Oral Examination (Defense)

3-29-2024

Type of Culminating Activity

Thesis

Degree Title

Master of Public Health

Department

Public and Population Health

Supervisory Committee Chair

Cynthia Curl, Ph.D.

Supervisory Committee Member

Uwe Reischl, Ph.D.

Supervisory Committee Member

Carly Hyland, Ph.D.

Abstract

The herbicide glyphosate is the most heavily used agricultural chemical in the world, and exposure has been linked to carcinogenic and reproductive effects, although this remains controversial. Glyphosate has a short biologic half-life, challenging the assessment of associations between chronic exposure and potential health effects. Long-term glyphosate exposure has often been evaluated using a single urine sample, potentially resulting in exposure misclassification and resulting in systematic underestimation of the association between glyphosate exposure and health outcomes. To determine the minimum number of urine samples required to accurately characterize chronic glyphosate exposure during pregnancy, we collected and analyzed 453 urine samples from a cohort of 40 pregnant women (average samples per participant = 11). Utilizing this original dataset, I assigned the participants included in this study to a quartile of exposure using their specific gravity-adjusted geometric mean glyphosate concentrations, which I considered their “true” quartile assignment. Using bootstrap resampling, I randomly selected limited numbers of urine samples from each participant to build new bootstrapped datasets. I then compared the agreement rate between the quartile assignment in the bootstrapped datasets and the original quartile assignment. These results will inform future glyphosate by increasing our understanding of the number of urine samples required to accurately assess chronic exposure.

DOI

https://doi.org/10.18122/td.2307.boisestate

Available for download on Saturday, August 01, 2026

Included in

Public Health Commons

Share

COinS