Chlorination Disinfection By-products and Risk of Stillbirths in England and Wales

Abstract

PP-30-179 Background/Aims: We investigated the association between trihalomethane (THM) concentrations in drinking water and stillbirths across a large population residing in 12 regions of England and Wales. THM concentrations in drinking water are routinely measured by the water companies; therefore, they are often used in epidemiological studies as a marker for disinfection by-products. The study comprises 2.79 million births during the period 1993–2001, of which 14,265 were all-cause stillbirths. Additionally, we investigated 2 cause-specific subgroups of stillbirths, as defined by Wigglesworth codes. Methods: A hierarchical mixture model was used to model routinely collected THM data from each water company. For the third trimester of pregnancy, the modeled concentrations for total THMs and individual THM species were categorized into predefined low, medium, and high exposure categories and linked, through postcode of the maternal address at birth, to the birth outcome data. Logistic regression was performed adjusting for potential confounding by maternal age (split into 5 categories), and socioeconomic status (as measured by quintiles of the Carstairs Index). Fifteen company/period specific analyses were carried out for each disease endpoint and THM exposure combination, and then company/period specific results were combined through meta-analysis. Results: Mean total THM concentrations ranged from 16.5 μg/L in the low exposure category to 71.9 μg/L in the high exposure category. Prevalence rates of all-cause and cause-specific stillbirths differed according to maternal age category and according to socioeconomic status (trend of increasing prevalence rates from least to most deprived group). Meta-analysis results were suggestive of a small excess risk of stillbirth associated with chloroform, but not with total THMs or the other individual THM species examined. Conclusion: There was no evidence to support an association of stillbirth or cause-specific stillbirth with total THMs; however, there was evidence suggestive of a small excess risk of stillbirth with levels of chloroform.