| The overall goal of the project was to refine our understanding of the effect of exposure to water chlorination by-products on the risk of birth defects.; A meta-analysis based on the five previous studies suggested small effects of chlorination by-products on the risk of any birth defect (summary odds ratio = 1.25, 95 confidence interval = 1.11–1.40), neural tube (1.49, 1.08–2.05) and urinary tract defects (2.27, 1.34–3.85), whereas results of respiratory system, cardiac and oral cleft defects were heterogeneous and inconclusive.; To assess the effect of water chlorination by-products on specific birth defects, we conducted a nationwide cross-sectional study of 184,676 births in years 1993–98 using information from the Birth Registry and Waterworks Registry in Norway. We used four exposure categories on the basis of municipal level information on chlorination (yes/no) and the level of color (mg Pt/L) representing the amount of natural organic matter: high (chlorination, color 20-), medium (chlorination, color 10–19.9) and low (chlorination, color <10) with no chlorination and low color (<10) as the reference category. We applied logistic regression analysis to estimate the risk of birth defects and adjust for maternal age, parity, centrality and population density. The risks of any birth defect (adjusted odds ratio = 1.13, 95% confidence interval = 1.01–1.25), cardiac (1.37, 1.00–1.89), respiratory system (1.89, 1.00–3.58), and urinary tract defects (1.55, 0.92–2.61) were associated with exposure (medium and high combined). Of the specific birth defects, only the risk of ventricular septal defects was significantly elevated with an exposure-response pattern, yielding an adjusted OR of 1.63 (1.02–2.58) for the medium and 1.81 (1.05–3.09) for the high exposure category.; Seasonal variation in the occurrence of birth defects provides indirect evidence of the causal role of environmental factors, such as prenatal exposure to chlorination byproducts, which has been reported to exhibit seasonality. To assess the seasonal variation in the occurrence of birth defects in Norway, we applied χ2 goodness-of-fit test for identification of heterogeneity in the monthly occurrence of birth defects. A statistically significant seasonal variation (α = 0.05) was indicated for any birth defect, respiratory defects, urinary tract defects, hydrocephalus and for Down's syndrome. The highest risk for respiratory defects was among infants born in March, for urinary tract defects in March and from September to November, for hydrocephalus in July and for Down's syndrome in February.; These results provide some evidence that prenatal exposure to chlornation byproducts may increase the risk of birth defects. The present study indicates a seasonal variation in the occurrence of respiratory defects, urinary tract defects, hydrocephalus and Down's syndrome in Norway. Further studies are needed to explain the reasons for seasonal variation, which are likely to represent environmental causes of these birth defects. |
