Water Chlorination By-Products and Congenital Anomalies in England and Wales

Abstract

ISEE-400 Objective: We have conducted a study to examine the relation between trihalomethanes (THMs) and congenital anomalies in 12 water regions in the United Kingdom. Material and Methods: The congenital anomaly data were provided by the National Congenital Anomalies Registry (NCAR), the Local Congenital Anomalies registries (LCARs), and the national Terminations registry. THM data based on sampling of tap water within water supply zones were provided by the water companies and linked to the birth data via residential postcode using a Geographical Information System (GIS). For each birth, a modeled estimate of THM exposure was obtained for the first trimester of pregnancy. We used 3 categories of total THM (TTHM) exposure (<30, 30 to <60, and 60+ mg/L), total brominated exposure (<10, 10 to <20, and 20+ mg/L), and bromoform exposure (<2, 2 to <4, and 4+ mg/L). Limited information on potential confounders was available on sex, maternal age, and measures of socioeconomic status (SES) at small-area scale. A total of 2,605,226 pregnancy outcomes were included in the study. Results: In general, there did not seem to be an association between THM exposure and risk of congenital anomalies. For some congenital anomaly subgroups, there tended to be a negative exposure response relationship with TTHM exposure, but only for exomphalos, and cleft palate only and broad urinary defects as part of a multiple syndrome were they statistically significant. The exposure-response relationships with total brominated THMs and bromoform for some subgroups of congenital anomalies tended to be mostly positive but not statistically significant, with the exception of obstructive urinary defects, which showed a negative exposure-response relationship with brominated THMs, which was statistically significant (P < 0.05), and isolated respiratory defects and some groups of cleft lip and palate, which showed a negative relationship with bromoform, which was not statistically significant. Conclusions: We found little evidence for a relationship between trihalomethanes and congenital anomalies.