Risks and benefits of drinking water treatment : focusing on child health and prenatal development

Sammanfattning: Drinking water is our most important food item and its access is indispensable for a wellfunctioning society. Recent Swedish water-borne outbreaks have demonstrated a system vulnerability and a clear need for improved knowledge on how variations in drinking water quality affects our health. This thesis explores this topic and consists of two large-scale epidemiological studies. The specific objectives were to i) obtain data on drinking water consumption patterns among adults, ii) assess whether changes in drinking water treatment and/or raw water source–aiming to increase pathogen reduction–affected the risk of gastrointestinal illness (GII) and iii–iv) assess if gestational exposure to by-products from drinking water chlorination war associated with the risk of adverse reproductive outcomes. In a longitudinal cohort, we collected repeated information on tap water consumption and GII episodes via a monthly SMS among 5,200 participants during several periods in 2012-2016. The study was conducted in two parts of Sweden, in populations of neighbouring municipalities. In Paper I, we found that 99.8% of adults were consumers of cold tap water, while the consumption of bottled water was low. This lends support to the use of large register-based studies to assess the associations between drinking water exposures and health. In Paper II, we assessed if changes in municipal drinking water production and/or of raw water source affected the risk of GII. These changes encompassed either switching ground water treatment plant, changing from a surface to a ground water treatment plant or switching the surface water treatment plant and raw water source, all resulting in increased pathogen reduction in the drinking water. We observed no differences in the risk of GII among adults, however, among children, a 24% relative risk reduction in GII was observed after switching surface water treatment plant and raw water source. The indications that children are the most sensitive population to drinking water related GII are in line with previous findings. In a nationwide register-based study, we assessed whether gestational exposure to chlorination by-products, trihalomethanes (TTHM), was associated with small for gestational age (SGA), preterm delivery or congenital malformations. We included more than 620,000 children born during 2005–2015 of mothers residing in Swedish localities (≥10 000 inhabitants) and where information on trimester specific TTHM exposure was available. The exposure was categorized into no chlorination, <5, 5–15, and >15 μg TTHM/L and stratified by treatment (hypochlorite and chloramine). In Paper III, we found indications of a dosedependent multivariable-adjusted association of TTHM with risk of SGA in areas using hypochlorite, odds ratio (OR) 1.20 (95% confidence interval [CI]: 1.08-1.33) when comparing the highest exposed population to the unexposed. In Paper IV, TTHM was dosedependently associated with malformations, but only in areas using chloramine. Comparing the population with highest exposure to the unexposed, ORs of 1.82 (95% CI: 1.07–3.12), 2.06 (95% CI: 1.53–2.78), 1.77 (95% CI: 1.38–2.26) and 1.34 (95% CI: 1.10–1.64) were seen for malformations of the nervous system, urinary system, genitals and limbs, respectively. The findings indicate that chlorination by-products may be associated with several adverse reproductive outcomes. Congenital malformations linked to chlorination by-product from chloramine use has not previously been highlighted and needs further attention.