Estimating the effectiveness of syndromic screening at airports for Bundibugyo ebolavirus disease

This study has not yet been peer reviewed.

Background An outbreak of Bundibugyo ebolavirus disease (BVD) affecting Ituri Province, Democratic Republic of the Congo (DRC), with confirmed spread to Uganda, was declared a Public Health Emergency of International Concern by WHO in May 2026. Airport syndromic (fever) screening is often considered as one of a suite of measures to limit international spread of viral haemorrhagic fevers. We estimated how effective combined exit and entry syndromic screening would be at detecting BVD-infected air travellers.

Methods We used a stochastic individual-based simulation model, adapted from a screening model originally developed for 2019-nCoV, to estimate the proportion of infected travellers who would be detected at exit screening, become severely ill during the flight, be detected at entry screening, or remain undetected throughout. Natural-history parameters (incubation period and onset-to-severe-disease interval) were derived from the 2007 Uganda outbreak and a Bayesian re-analysis of the 2012 Isiro (DRC) outbreak line list, since patient-level data from the current 2026 outbreak are not yet available. We modelled a representative 12-hour DRC/Uganda-to-international connecting itinerary with 86% screening sensitivity at both exit and entry, and conducted sensitivity analyses varying flight duration, the asymptomatic/afebrile fraction, the epidemic’s doubling time, and screening sensitivity.

Results Under the baseline scenario, we estimate that 73% (95% CrI: 68–77%) of BVD-infected travellers would evade combined exit and entry screening entirely, primarily because most board their flight before symptom onset. Among those who successfully boarded, 92% (95% CrI: 90–94%) would arrive undetected. Even under a theoretical best-case scenario of 100% screening sensitivity at both stages, 72% of infected travellers would still go undetected on arrival. The undetected fraction increased further during active epidemic growth, rising to as much as 85% under the fastest estimated doubling time, and was most sensitive to the incubation period, ranging from 36–82% undetected across incubation periods of 1–14 days.

Conclusion Syndromic airport screening alone is unlikely to meaningfully reduce the risk of international BVD spread via air travel, because most infected travellers depart while still pre-symptomatic. Resources are likely better directed towards outbreak control at source, clinician preparedness and referral pathways in receiving countries, and structured post-departure self-monitoring guidance for travellers from affected regions. Full methods, results, sensitivity analyses and an interactive Shiny app for exploring screening parameters are available in the accompanying report, the preprint on medRxiv, and the GitHub repository.