As part of the EU-funded INTARESE project, which contribued to the development of this Toolbox, a case study was done to assess the health impacts of domestic water supply and use. 

After the initial screening phase, the majority of the case study was focused on disinfection by-products (DBPs). These serve as an example of a complex exposure.  Moreover, in contrast to many other drinking water contaminants, populations served with disinfected water supplies are exposed to some DBP components irrespective of their water ingestion habits:  volatile components such as THMs are absorbed through the skin (percutaneous absorption) and through the lungs (inhalation).  This near-ubiquitous exposure in areas served with disinfected water makes the assessment of health impacts of DBP exposure particularly relevant and of considerable policy interest.

Over 600 disinfection by-products have been identified.  Epidemiological evidence has been suggestive of an association between exposure to DBPs and bladder cancer, adverse birth outcomes related to prematurity and growth, and birth defects.  In the vast majority of epidemiological studies these health outcomes have been linked to concentrations of total trihalomethanes (a marker of DBP exposure that is used in monitoring) at place of residence (or place of residence of the mother in the case of birth outcomes).  Total trihalomethanes may not be the best marker of exposure to the whole chemical mixture, particularly since they may exhibit threshold carcinogenic properties and are only weakly correlated to many other, more mutagenic chemicals in the mixture.  Their advantage, however, is that they they are routinely measured as part of drinking water monitoring programmes in most developed countries.  This helps to increase the statistical power of the analyses, by allowing investigation of larger study populations.  On the other hand, this benefit may be outweighed by the uncertainties that arise from using an exposure metric that fails completely to capture the true exposure of the population.

Toxicological studies, in contrast, have focused on individual chemical constituents of the DBP mixture.  This clearly helps to improve the specificity of the analysis, but in doing so creates uncertainties of interpretation when dealing with chemical mixtures in which the putative agent is unknown or under some dispute.  Other assessments of cancer risk associated with exposure to DBPs have employed cancer potency factors attributed to THMs.  This approach, however, can lead to a highly overexaggerated estimate of risk, which may in fact be zero.