Drinking water chlorination and delivery outcome-a registry-based study in Sweden.

By using the Swedish Medical Birth Registry and official data on drinking water chlorination, three cohorts were identified and compared: women who lived in areas where drinking water was disinfected with chlorine dioxide, women who lived in areas that used sodium hypochlorite disinfection, and women who lived in areas where there was no chlorination of the drinking water. There was a statistically significant increase in short gestational duration and low birth weight and especially in short body length and very small head circumference in areas using sodium hypochlorite, but no significant effects on these variables were found in areas using chlorine dioxide. No effects on congenital malformations, childhood cancer, infant mortality, low Apgar score, neonatal jaundice, or neonatal hypothyroidism were associated with either of the two drinking water chlorination methods. Because the exposure information in this study was based on the chlorination method and not the amount of byproducts in the water, the general lack of significant effects could be due to a low concentration of such byproducts.