Modelling faecal pathogen flows and health risks in urban Bangladesh: Implications for sanitation decision making

Faecal-oral infections are a major component of the disease burden in low-income contexts, with inadequate sanitation seen as a contributing factor. However, demonstrating health effects of sanitation interventions - particularly in urban areas - has proved challenging and there is limited empirical evidence to support sanitation decisions that maximise health gains. This study aimed to develop, apply and validate a systems modelling approach to inform sanitation infrastructure and service decision-making in urban environments by examining enteric pathogen inputs, transport and reduction by various sanitation systems, and estimating corresponding exposure and public health impacts. The health effects of eight sanitation options were assessed in a low-income area in Dhaka, Bangladesh, with a focus on five target pathogens (Shigella, Vibrio cholerae, Salmonella Typhi, norovirus GII and Giardia). Relative to the sanitation base case in the study site (24% septic tanks, 5% holding tanks and 71% toilets discharging directly to open drains), comprehensive coverage of septic tanks was estimated to reduce the disease burden in disability-adjusted life years (DALYs) by 48-72%, while complete coverage of communal scale anaerobic baffled reactors was estimated to reduce DALYs by 67-81%. Despite these improvements, a concerning health risk persists with these systems as a result of effluent discharge to open drains, particularly when the systems are poorly managed. Other sanitation options, including use of constructed wetlands and small bore sewerage, demonstrated further reductions in local health risk, though several still exported pathogens into neighbouring areas, simply transferring risk to downstream communities. The study revealed sensitivity to and a requirement for further evidence on log reduction values for different sanitation systems under varying performance conditions, pathogen flows under flooding conditions as well as pathogen shedding and human exposure in typical low-income urban settings. Notwithstanding variability and uncertainties in input parameters, systems modelling can be a feasible and customisable approach to consider the relative health impact of different sanitation options across various contexts, and stands as a valuable tool to guide urban sanitation decision-making.