Cross-connections in drinking water distribution networks: Quantitative microbial risk assessment in combination with fault tree analysis and hydraulic modelling

Abstract

Deficiencies in drinking water distribution networks, such as cross-connections, may lead to contamination of the drinking water and pose a serious health risk to consumers. Cross-connections and backflows are considered among the most severe public health risks in distribution networks. The aim of this paper was to provide a framework for estimating the risk of infection from cross-connection and backflow events. Campylobacter, norovirus, and Cryptosporidium were chosen as reference pathogens for this study. The theoretical framework was constructed based on the fault tree analysis methodology. National aggregated cross-connection incident data was used to calculate the probability of a contamination event occurring in Swedish networks. Three risk cases were evaluated: endemic, elevated, and extreme. Quantitative microbial risk assessment (QMRA) was used to assess daily risk of infection for average national estimates. The framework was also evaluated using local data from the Gothenburg network. The daily risk of infection from cross-connection and backflow events in Swedish networks was generally above an acceptable target level of 10−6 for all reference pathogens and modelled cases; the exception was for the Gothenburg system where the risk was lower than 10−7. An outbreak case study was used to validate the framework results. For the outbreak case study, contaminant transport in the network was simulated using hydraulic modelling (EPANET), and risk estimates were calculated using QMRA. The outbreak simulation predicted between 97 and 148 symptomatic infections, while the epidemiological survey conducted during the outbreak reported 179 cases of illness. The fault tree analysis framework was successfully validated using an outbreak case study, though it was shown on the example of Gothenburg that local data is still needed for well-performing systems. The framework can help inform microbial risk assessments for drinking water suppliers, especially ones with limited resources and expertise in this area.