Environmental pathogen surveillance in cities without universal piped wastewater infrastructure

Although wastewater surveillance for assessing community health and well-being is now mainstream, most cities in low- and middle-income countries lack conventional wastewater services. In these settings, environmental surveillance beyond conventional wastewater offers the potential to inform public health responses, design interventions intended to reduce exposures, and to evaluate infection control programs. To explore these potential use cases, we measured pathogens, source-tracking markers, and fecal indicator bacteria in wastewater treatment plant (WWTP) influent and effluent, wastewater surface discharges, impacted river water, impacted soils, open drains, stormwater, and fecal sludges from onsite sanitation in Maputo, Mozambique. We detected a wide range of pathogens by multi-parallel RT-qPCR across all matrices, revealing a nuanced picture of pathogen flows in the city and suggesting the potential for exposures beyond those typically included in studies of sanitation and health. We developed a regression model with multiple pathogens as the dependent variable and observed lower pathogen concentrations in direct wastewater discharges (mean difference -1.4 log10 per liter, 95% CI: -1.7, -1.1), WWTP effluent (-0.97 log10, 95% CI: -1.5, -0.47), water from open drains (-2.0 log10, 95% CI: -2.5, -1.6), impacted river water (-3.0 log10, 95% CI: -3.7, -2.4), and stormwater (-4.7 log10, 95% CI: -7.0, -3.3) compared to WWTP influent. We further observed that a one standard deviation increase in 7-day cumulative precipitation was associated with an increase in the pathogen concentration in all matrices (0.11 log10, [0.04, 0.19]). Despite lower concentrations of pathogens in matrices compared to WWTP influent, frequent detection of pathogens indicates clear potential to use environmental pathogen surveillance to inform public health responses in cities lacking universal conventional wastewater, with a wide range of promising applications.