Inferring Transmission Dynamics of Respiratory Syncytial Virus from Houston Wastewater

2025/11/21 by Palacio, Jose R., Ensor, Katherine B., Keller, Sallie A. +4
Medicine · Mathematics · #Respiratory viral infections research #SARS-CoV-2 detection and testing #COVID-19 epidemiological studies

paper · doi:10.48550/arxiv.2511.17816

Abstract

Wastewater-based epidemiology (WBE) is an effective tool for tracking community circulation of respiratory viruses. We address estimating the effective reproduction number (R_t) and the relative number of infections from wastewater viral load. Using weekly Houston data on respiratory syncytial virus (RSV), we implement a parsimonious Bayesian renewal model that links latent infections to measured viral load through biologically motivated generation and shedding kernels. The framework yields estimates of R_t and relative infections, enabling a coherent interpretation of transmission timing and phase. We compare two input strategies-(i) raw viral-load measurements with a log-scale standard deviation, and (ii) state-space-filtered load estimates with time-varying variances-and find no practically meaningful differences in inferred trajectories or peak timing. Given this equivalence, we report the filtered input as a pragmatic default because it embeds week-specific variances while leaving epidemiological conclusions unchanged.

Citations

• Understanding and managing uncertainty and variability for wastewater monitoring beyond the pandemic: Lessons learned from the United Kingdom national COVID-19 surveillance programmes
• Endemic-epidemic models with discrete-time serial interval distributions for infectious disease prediction

Related

• Semiparametric inference of effective reproduction number dynamics from wastewater pathogen surveillance data
• The signal is not flushed away: Inferring the effective reproduction number from wastewater data in small populations
• Wastewater surveillance using differentiable Gaussian processes
• Uncovering dynamics between SARS-CoV-2 wastewater concentrations and community infections via Bayesian spatial functional concurrent regression
• Inference for stochastic kinetic models from multiple data sources for joint estimation of infection dynamics from aggregate reports and virological data