Environmental influences on menstrual disturbances and uterine leiomyomata
Embargo Date
2027-05-09
OA Version
Citation
Abstract
Menstrual disturbances and gynecologic conditions affect the health and quality of life of millions of individuals, yet they are understudied relative to their public health burden. Environmental exposures can influence reproductive health outcomes, and may represent potential targets for public health interventions. In this dissertation, we estimated the effects of three environmental factors — the neighborhood socioeconomic environment, use of chemical hair straighteners, and metal biomarkers — on the risk of abnormal uterine bleeding (AUB), dysmenorrhea, and uterine leiomyomata (UL).First, we conducted a cross-sectional study of the associations of neighborhood socioeconomic disadvantage with the prevalence of AUB and dysmenorrhea. We analyzed data from Pregnancy Study Online (PRESTO), an internet-based cohort of North American pregnancy planners. We operationalized neighborhood disadvantage using the Area Deprivation Index (ADI) and defined AUB and dysmenorrhea based on self-reported menstrual cycle characteristics. We observed that living in a more disadvantaged neighborhood (relative to the least disadvantaged within-state quartile) was associated with a higher prevalence of menstrual disturbances. After adjustment for individual-level household income, the association between neighborhood disadvantage and prevalence of AUB was attenuated, whereas neighborhood disadvantage remained positively associated with the prevalence of dysmenorrhea.
Second, we conducted a cross-sectional study of the associations of chemical hair straightener (“relaxer”) use with the prevalence of AUB and dysmenorrhea in PRESTO. We observed positive associations of current (vs. never) relaxer use with the prevalence of AUB and dysmenorrhea. Furthermore, measures reflecting higher cumulative lifetime exposure to hair relaxers, including younger age at first use, ever experiencing burns during application, and greater frequency of use, were positively associated with the prevalence of dysmenorrhea.
Third, we conducted a prospective study of the associations of 17 metal biomarkers measured in whole blood with UL incidence during 20 months of follow-up. We analyzed data from the Study of Environment, Lifestyle and Fibroids (SELF), a cohort of reproductive-aged Black individuals in the Detroit, Michigan area. We used Bayesian Kernel Machine Regression (BKMR), a flexible statistical mixture method that allows for non-linear associations and interactions between mixture components while controlling for co-pollutant confounding. We observed a weak, positive association between the metal mixture and incident UL during 20 months of follow-up, which was driven by cadmium. We observed stronger positive associations of cadmium and mercury with incident UL among participants with vitamin D deficiency (<20 ng/mL).
Our results illuminate novel connections between the social and physical environment and reproductive health. If supported by further research, our findings may inform public health interventions to prevent menstrual disturbances and UL.
Description
2025