Speaker
Description
Wastewater monitoring of pathogens offers the potential to track a wide range of infectious diseases for public health, but current methods for analyzing transmission dynamics are primarily tailored to SARS-CoV-2, which is detected at high concentrations in municipal sewage. To support robust surveillance of pathogens with lower concentrations in wastewater and limited clinical validation data, we developed a Bayesian semi-mechanistic model that incorporates uncertainties in pathogen-specific infection and shedding dynamics, and introduces a coherent model for digital PCR (dPCR) data.
We applied this model to estimate the effective reproduction number Rt and epidemic growth rate during two seasonal waves of SARS-CoV-2, influenza A virus (IAV), and respiratory syncytial virus (RSV) using measurements from differently sized wastewater treatment plants in Switzerland. Despite 10- to 50-fold lower IAV and RSV concentrations compared to SARS-CoV-2, the model produced real-time Rt estimates consistent with retrospective data and captured the uncertainty from viral shedding kinetics for each pathogen.
The model also generated short-term forecasts of concentration measurements with little bias during rising and falling infections, and overdispersion only at very low concentrations. Rt estimates were robust to non-daily sampling and outliers in the measurement data. Our approach enables the reliable quantification of transmission dynamics from wastewater data in real-time, without the need for smoothing, imputation, or outlier detection prior to model fitting. By combining interpretable parameterization, rigorous uncertainty quantification, and real-time assessment of model predictions, this method can support the wastewater-based surveillance of undermonitored infectious diseases using only a few wastewater samples per week.
The method has been implemented in the open-source R package EpiSewer with the aim to facilitate wastewater-based monitoring of a broad range of pathogens.
| Expedited Notification | No thanks, I do not require Expedited Notification |
|---|
