Wastewater-informed agent-based modelling of hepatitis E transmission dynamics

Hepatitis E virus (HEV) is considered a predominantly foodborne pathogen in developed settings. During COVID-19 lockdown periods, however, HEV concentrations in wastewater at a treatment plant in Munich, Germany decreased, suggesting that pandemic-related behaviour changes inadvertently influenced transmission. In contrast, reported cases and wastewater data from a smaller catchment showed no comparable decline. To assess whether the observed reduction is compatible with a near-exclusively foodborne infection and to reconcile the contrasting signals across surveillance modalities, we developed a stochastic, individual-level model of HEV transmission, shedding, and ascertainment in Munich. Using Approximate Bayesian Computation, we calibrated this model to wastewater and case data from 2020-2023, first separately and then jointly. Posterior parameter estimates indicated a substantial decline in transmission during lockdowns to about 35-40% of the non-lockdown level, with the 95% credible interval entirely below 1 (no change). Joint inference suggested that possible modest lockdown-associated increases in diagnosis probabilities and higher measurement variability in the smaller catchment masked this effect in clinical and small-scale wastewater data, respectively. These findings demonstrate how wastewater-based surveillance, used alongside reported cases, can enable more robust parameter inference for models of under-reported pathogens like HEV, thereby supporting informed public health risk assessments.