Exposure to disinfection by-products and risk of birth defects: A systematic review and dose-response meta-analysis

The disinfection of drinking water has been a major public health achievement, significantly reducing waterborne diseases. However, there are also some drawbacks in this practice, particularly the fact that chlorinated disinfectants can generate disinfection by-products (DBPs) which in turn have been associated with adverse health effects, including birth defects. The extent to which such adverse reproductive effect may occur following chlorinated water consumption, and the possible threshold of exposure involved, are however not entirely clear. This systematic review and dose-response meta-analysis aimed to evaluate the association between DBP exposure and the risk of congenital anomalies, with a focus on exposure thresholds. We conducted systematic search in Scopus, PubMed, and Web of Science up to March 18, 2025, to retrieve the observational studies assessing DBP exposure and reporting congenital anomalies and birth-related outcomes. We performed a dose-response meta-analysis to assess exposure thresholds. A total of 31 studies met the inclusion criteria. The meta-analysis comparing the highest versus the lowest exposure levels found a positive association between trihalomethanes (THMs) and prevalence at birth of cardiovascular, musculoskeletal, neural tube, urinary tract defects, as well as most growth-related birth outcomes. The dose-response analysis revealed a nonlinear association, with increased risks for anencephaly, musculoskeletal, and orofacial defects at THM levels exceeding 60 μg/L, and for term low birth weight at levels above 30 μg/L. A plateau effect was observed for urinary tract defects at 30 μg/L, while low birth weight and small-for-gestational-age infants showed a nearly linear association with DBP exposure. These findings suggest that high THM exposure, particularly above certain thresholds, increase the risk of growth and urinary tract defects, findings that are also supported by biological plausibility. Given existing regulatory limits (80-100 μg/L in US and EU), our results indicate the need for stricter standards to protect vulnerable populations such as pregnant women.