COVID-19 vaccine hesitancy proved to be a major barrier to higher uptake, but it is unclear whether interventions targeting hesitancy could result in substantial prevention benefits. Epidemic models that represent vaccine decision-making psychology can provide insight into the potential impact of vaccine promotion interventions in the context of the COVID-19 pandemic and future epidemics of vaccine-preventable diseases. We coupled a network- and agent-based model of SARS-CoV-2 transmission with a social-psychological vaccination decision-making model in which vaccine side effects and breakthrough infections could "nudge" individuals towards vaccine resistance while spikes in COVID-19 hospitalizations could nudge them towards vaccine willingness. This model was parameterized and calibrated to represent the COVID-19 epidemic in Georgia, USA from January 2021 to August 2022. We modelled counterfactual scenarios in which increases to resistant-to-willing nudges were combined with decreases to willing-to-resistant nudges. We compared cumulative vaccine doses administered, SARS-CoV-2 incidence, and COVID-related deaths across scenarios. Increasing the probability of hospitalization-prompted resistant-to-willing nudges increased vaccine uptake by as much as 5.4 % and decreased SARS-CoV-2 incidence by as much as 4.0 %. In contrast, decreasing the probability of breakthrough infection-related willing-to-resistant nudges had a negligible impact on further vaccination and disease outcomes. Vaccine promotion interventions that address community-level factors influencing decision-making may have a greater ability to avert SARS-CoV-2 infections than those targeted to individual vaccination and infection history. Additionally, reactive vaccine promotion interventions may have only limited prevention benefits in the short term, suggesting that attention should be paid to formulating interventions that accurately anticipate the case curve.

The COVID-19 pandemic has increased interest in wastewater-based epidemiology (WBE) as a reliable and cost-effective framework for monitoring the spread of microbes. However, WBE frameworks have rarely been applied to the study of fecal–oral transmissible diseases, except for poliomyelitis. Here, we investigated the presence of hepatitis A virus (HAV) and hepatitis E virus (HEV) in wastewater in Munich. We collected wastewater samples between July 2020 and November 2023. A total of 186 samples were processed using centrifugation and analyzed for HAV- and HEV-RNA using RT-qPCR. As a reference, we used notification data from clinically or laboratory-diagnosed hepatitis A and E cases. Lockdown stringency levels were derived from official documentation. Our results show that 87.6% of wastewater samples were positive for HEV at concentrations of 9.0 × 10¹ to 2.5 × 10⁵ copies/L, while HAV was only detectable in 7.5% of the samples at viral loads of 4.6 × 10¹ to 2.4 × 10³ copies/L. We also detected differences in HEV concentrations but not in case numbers when comparing lockdown and no-lockdown periods. This study covers all but the first lockdowns in Bavaria. We present a unique real-world dataset evaluating the impact of lockdown interventions on hepatitis A and E case numbers, as well as on the concentrations of HAV and HEV in wastewater. Person-to-person spread and eating out appear to have contributed to the transmission of HEV. In addition, the consistently high HEV concentrations in sewage support the findings of serological studies, indicating a substantial burden of undetected subclinical infections.

Disease surveillance and epidemiological modeling are critical to guide public health interventions, but model performance depends on data availability and quality. While clinical reports often suffer from under-ascertainment and delays, wastewater-based surveillance (WBS) can rapidly capture community infection dynamics by detecting viral RNA from both symptomatic and asymptomatic cases. However, WBS data can be difficult to interpret. Here, we present a coupled model of infectious disease and wastewater dynamics designed for scalability to large cities. We calibrate the model to the first COVID-19 wave in the city of Munich and quantify how sampling protocols, precipitation infiltration, viral decay, normalization strategies, and intervention timing can shape the relationship between wastewater measurements and disease prevalence, thereby improving the interpretability and practical value of wastewater data for epidemiological decision-making. We find that when appropriate sampling, normalization, and analysis strategies are utilized, wastewater data can reveal changes in prevalence earlier than clinical reports and provide advance warning of upcoming increases in disease burden. Our results guide WBS design and integration into predictive early-warning systems. Our modeling framework is generalizable to other COVID-19-like pathogens to help enable robust, cost-efficient disease monitoring.

This study aimed to retrospectively assess the cost-effectiveness of various COVID-19 vaccination strategies in Ethiopia. It involved healthcare workers (HCWs) and community participants; and was conducted through interviews and serological tests. Local SARS-CoV-2 variants and seroprevalence rates, as well as national COVID-19 reports and vaccination status were also analyzed. A cost-effectiveness analysis was performed to determine the most economical vaccination strategies in settings with limited vaccine access and high SARS-CoV-2 seroprevalence. Before the arrival of the vaccines, 65% of HCWs had antibodies against SARS-CoV-2, indicating prior exposure to the virus. Individuals with prior infection exhibited a greater antibody response to COVID-19 vaccines and experienced fewer new infections compared to those without prior infection, regardless of vaccination status (5% vs. 24%, p < 0.001 for vaccinated; 3% vs. 48%, p < 0.001 for unvaccinated). The cost-effectiveness analysis indicated that a single-dose vaccination strategy is optimal in settings with high underlying seroprevalence and limited vaccine availability. This study underscores the need for pragmatic vaccination strategies tailored to local contexts, particularly in high-seroprevalence regions, to maximize vaccine impact and minimize the spread of COVID-19. Implementing a targeted approach based on local seroprevalence information could have helped Ethiopia achieve higher vaccination rates and prevent subsequent outbreaks.

Epidemiological modeling is critical to guide public health interventions, but model performance depends on data availability and quality. While clinical reports suffer from under-ascertainment and delays, wastewater-based surveillance (WBS) can rapidly capture community infection dynamics by detecting viral RNA from both symptomatic and asymptomatic cases. However, WBS data can be difficult to interpret. Here, we present a coupled model of infectious disease and wastewater dynamics designed for scalability to large cities. We calibrate the model to the first COVID-19 wave in Munich and quantify how sampling protocols, precipitation, viral decay, normalization strategies, and intervention timing shape the relationship between wastewater measurements and disease prevalence. We find that under appropriate normalization and analysis strategies, wastewater data can provide advance warning of increases in disease burden. Our results guide WBS design and integration into predictive early-warning systems, and our framework is generalizable to other COVID-19-like pathogens, thereby enabling robust disease monitoring.

Wastewater-based epidemiology (WBE) was introduced early in the COVID-19 pandemic to provide continuous, anonymous, population-level information on SARS-CoV-2 circulation at a time when individual testing was constrained by shortages of reagents, staff, logistics, and funding. It has been difficult to assess whether wastewater data retain their epidemiological value once clinical testing declines, because most published records cover only the first one or two pandemic years. We monitored SARS-CoV-2 in Munich wastewater continuously from April 2020 to April 2025, spanning all lockdown and post-lockdown phases, the vaccination campaigns, and the first post-pandemic years. Samples were collected once or twice weekly as qualified morning grab samples from six residential catchments, and as 24-h composite samples from the wastewater treatment plant (WWTP) Gut Grosslappen. A total of 1100 samples were analyzed by RT-qPCR or droplet digital PCR (ddPCR), and compared with official notification data for the corresponding catchment areas. Weekly viral loads correlated significantly with reported cases at every site (Spearman ρ 0.47 to 0.79), with no consistent lead or lag at weekly resolution. Correlations were higher for the residential catchments (ρ 0.62 to 0.79) than for the WWTP (ρ 0.47 and 0.58), although this difference did not persist when the comparison was restricted to an interval in which both site types were sampled concurrently. The ratio of wastewater viral load to reported cases remained near baseline while testing was mandatory, and rose markedly after mandatory testing ended in March 2023. This shows that notification data increasingly underestimated infection numbers while wastewater continued to track circulation. Wastewater surveillance therefore becomes more informative, not less, as clinical surveillance contracts.