Enhanced cholera surveillance was conducted at the two official cholera treatment facilities in Uvira, Democratic Republic of the Congo, covering January 2017 through December 2023. Surveillance combined routine case detection with systematic rapid diagnostic case testing to improve case ascertainment over the study period. These facility-based data formed the backbone of analyses used to infer trends in observed cholera burden around the time of the mass campaign.
The investigation incorporated repeated, representative population surveys that measured healthcare-seeking behavior, self-reported vaccination status, population mobility, and antibody profiles. Serologic sampling was used to estimate seroincidence and thereby infer infection dynamics that are not fully visible from facility-attended case counts alone. Rapid diagnostic testing of suspected cases at facilities provided additional, time-resolved data on case confirmation.
A Bayesian inference framework integrated the multiple data streams into an ensemble of mechanistic cholera transmission models. Models accounted for time-varying transmissibility, realistic immunity dynamics following infection and vaccination, and decline in population-level vaccine coverage driven by population turnover. The ensemble approach combined model structures and parameter uncertainty to produce probabilistic estimates of outcomes under observed and counterfactual vaccination scenarios.
The mass vaccination campaign in 2020 employed Euvichol Plus and reached 66% coverage with at least one dose in the target population of Uvira. The study explicitly modeled the achieved coverage as well as hypothetical scenarios of higher coverage to quantify the incremental benefits of broader vaccination. Loss of effective coverage over time due to births, deaths, and migration (population turnover) was incorporated in the modeling to reflect realistic declines in vaccine-derived protection at the population level.
Using ensemble counterfactual modeling, the authors estimated that the 2020 mass vaccination averted 56% (95% Credible Interval: 34–81) of infections and deaths over the subsequent three years. Translated into facility-attended outcomes, this impact corresponds to a median estimate of 2,350 averted facility-attended cases (95% CrI: 890–8,960). Estimated averted deaths—counting both facility and community mortality—had a median of 44 (95% CrI: 16–150). These estimates reflect combined direct and indirect effects captured by the transmission models and the integrated surveillance and serologic data.
The authors simulated alternative vaccination scenarios to assess potential additional gains. Vaccinating the entire eligible population of Uvira was estimated to avert 64% (median, 95% CrI: 43–87) of cases and deaths. The analysis highlighted the influence of population turnover: when turnover was halved in model scenarios, estimated impact rose to a median of 71% (95% CrI: 52–90). This indicates that maintaining durable population protection depends substantially on demographic dynamics that erode coverage over time.
Although the mass oral cholera vaccine (OCV) campaign produced a substantial reduction in infections and deaths in this endemic setting, several factors limited its total achievable benefit. Imperfect coverage (66% with at least one dose), ongoing high transmission intensity in Uvira, and rapid loss of population-level protection because of turnover reduced the realized impact relative to what might be possible with higher coverage and slower turnover. The authors emphasize that OCVs are an important tool but not a standalone solution in high-burden, endemic contexts.
The study concludes that successful cholera control in Uvira will require multisectoral interventions beyond vaccination, notably sustainable improvements in water, sanitation, and hygiene infrastructure to reduce transmission. The authors recommend setting and communicating realistic expectations for mass vaccination programs in highly endemic areas to maintain public confidence in current-generation OCVs. Modeling results suggest that campaigns combined with strategies to reduce population turnover effects and to increase and sustain coverage would yield greater public health gains.
Strengths of the work include enhanced, multi-year surveillance, integration of representative population surveys and serology, and the use of a Bayesian ensemble of mechanistic models to account for uncertainty in transmission and immunity. Limitations reported by the authors are tied to the underlying data and modeling assumptions; specific modeling choices and their implications were addressed within the ensemble approach to reflect structural and parameter uncertainty. Data and code needed to reproduce primary analyses are available at the project repository: https://github.com/GenevaIDD/uvira-ocv-impact.
The study was funded by Gavi (M&E 9166 09 20 A16) and the Wellcome Trust (221688/Z/20/Z). Ethical approvals were obtained from the Johns Hopkins Bloomberg School of Public Health (IRB00015785), the London School of Hygiene and Tropical Medicine (25365), and the University of Kinshasa School of Public Health (ESP/CE/65/2021). The authors declared no competing interests.
Overall, integrating enhanced surveillance, serology, and mechanistic transmission models provided evidence that the 2020 mass OCV campaign in Uvira substantially reduced cholera burden but that broader, sustained control will require complementary water and sanitation interventions and strategies to preserve population-level immunity.