Wastewater-based estimation of temporal variation in shedding amount of influenza A virus and clinically identified cases using the PRESENS model

Abstract

Wastewater-based estimation of infectious disease prevalence in real-time assists public health authorities in developing effective responses to current outbreaks. However, wastewater-based estimation for IAV remains poorly demonstrated, partially because of a lack of knowledge about temporal variation in shedding amount of an IAV-infected person. In this study, we applied two mathematical models to previously collected wastewater and clinical data from four U.S. states during the 2022/2023 influenza season, dominated by the H3N2 subtype. First, we modeled the relationship between the detection probability of IAV in wastewater and FluA case counts, using a logistic function. The model revealed that a 50 % probability of IAV detection in wastewater corresponds to 0.53 (95 % CrI: 0.35–0.78) cases per 100,000 people, as observed in clinical surveillance over two weeks. Next, we applied the previously developed PRESENS model to IAV wastewater concentration data from California, revealing rapid and prolonged virus shedding patterns. The estimated shedding model was incorporated into an extended version of the PRESENS model to assess the variability in the relationship between IAV concentrations and case numbers across other states, including Massachusetts, New Jersey, and Utah. As a result, our analysis demonstrated the effectiveness of normalizing IAV concentrations against PMMoV (Pepper mild mottle virus) to accurately understand spatial distribution patterns of IAV prevalence. We successfully estimated FluA case counts from wastewater concentrations within a factor of two for 80 % of the states, covering 34 % of the population monitored by wastewater surveillance. Importantly, wastewater-based estimates provided real-time or leading insights (0–2 days) compared to clinical case detection in three states, enabling early understanding of the incidence trends despite delays in data publication. These findings highlight the potential of wastewater surveillance to predict outbreaks, providing valuable lead time over traditional methods by accounting for the lag between detection and public reporting of case data.