Executive Summary
Since the publication of Spread Risk Assessment #1 on 18 May 2026, the outbreak has expanded geographically and additional exported cases have been confirmed. As of 22 May 2026, the Democratic Republic of the Congo has reported 101 laboratory-confirmed cases (up from 33), 904 suspected cases, and 119 suspected deaths across health zones in Ituri (Bunia, Rwampara, Nyankunde, Mongbwalu) and North Kivu provinces. One additional case with travel history from the affected northern provinces has been detected in Bukavu, South Kivu; this is not currently considered evidence of established local transmission. Kampala, Uganda has reported five laboratory-confirmed cases — including one death — within the 15–16 May cluster: three are importations from DRC (two travelers detected on 15–16 May plus one retrospective case traced to 14 May travel), and the remaining two are local contacts of the traveling cases rather than independent importations. In total, four independent importation events have been observed: three to Kampala and one to Bukavu (South Kivu) [1,2,3].
Preliminary IPC assessments in affected facilities revealed major gaps in screening, triage, and patient isolation, with four healthcare worker deaths reported to date. Cross-border transmission risk remains elevated due to active population movement, mining activities, insecurity, and the porous DRC–Uganda border.
This report updates the relative importation-risk distribution under the confirmed multi-province footprint spanning Ituri, North Kivu, and Kampala, and re-estimates the underlying outbreak size using detected importations within an Approximate Bayesian Computation (ABC) framework coupled to the underlying mobility model, across three importation-detection-efficiency scenarios (100%, 66%, and 50%, corresponding to 4, 6, and 8 true exported cases given the four observed). Throughout the report, relative importation risk (RR) denotes the conditional probability that a potential exported case arriving abroad arrives at a given destination, normalized so that values across all candidate destinations sum to one.
Methods Overview
Mobility Data
Global air travel patterns are derived from the International Air Transport Association (IATA) and the Official Airline Guide (OAG), providing passenger-level origin–destination flows between international airports. These data are complemented by short-scale commuting patterns capturing daily movements between adjacent subpopulations, representing the local spread of disease through routine human activity within and across health zones. The modeling approach follows the GLEAM (Global Epidemic and Mobility) computational framework [4,5].
Relative Importation Risk
For each potential destination Y, the model estimates the probability P(Y) that an infected individual originating in the affected area travels to Y, conditional on at least one exported case occurring. The output is a relative risk distribution across all candidate destinations, normalized so that values represent the share of importation probability assigned to each location.
Outbreak Size Estimation
The number of infections at the origin is inferred using an Approximate Bayesian Computation (ABC) approach. Given the observed international exportations as data D, the posterior distribution P(θ|D) over outbreak size θ is approximated by repeatedly simulating outbreaks of varying sizes, generating synthetic exportation events under the mobility model, and retaining parameter values whose simulated outputs are consistent with the observations. Two parameters condition the inference: the incubation period of BVD (2–21 days; average ~10 days based on the 2007 Uganda outbreak) and the importation-detection efficiency abroad. Three importation-detection-efficiency scenarios — 100%, 66%, and 50% — are reported to bound sensitivity to undercounting, corresponding to true export counts of 4, 6, and 8 given the four independent events observed (three to Kampala plus one to Bukavu, South Kivu) [1,2,3].
Analysis under the Current Expansion
Under the current expanded footprint, with Ituri, North Kivu, and Kampala treated as co-active sources of exportation, relative importation risk is concentrated in the cross-border corridors connecting the affected provinces to neighbouring countries. The highest-risk destinations are Kasese (Uganda, RR 37%), Kigali (Rwanda, RR 17%), Pakuba (Uganda, RR 16%), Bukavu (South Kivu, RR 15%), Arua (Uganda, RR 7%), and Cyangugu (Rwanda, RR 6.5%). The single laboratory-confirmed case in Bukavu with travel history from the affected northern provinces is consistent with this model-predicted corridor.
Inclusion of Kampala as an active source — driven by the cluster of laboratory-confirmed importations from DRC — redistributes a portion of total exportation probability toward destinations connected to the Ugandan capital by air, including Kigali and the broader East African aviation network. Outside the immediate region, the model assigns negligible absolute probability (< 1%) to intercontinental destinations under the current scenario.
P(Y): relative probability of a potential carrier arriving at Y, conditional on ≥1 exported case. Confirmed-case provinces (Ituri, North Kivu, Kampala) shown in dark blue; risk gradient in red–orange tiers. Hover for values.
Size of the Epidemic
As of 22 May 2026, four independent importation events have been observed: three to Kampala (two travelers detected on 15–16 May and one retrospective case traced to 14 May travel) and one to Bukavu, South Kivu (travel history from the affected provinces). The two additional Kampala cases that bring the Ugandan total to five are local contacts of the traveling cases and do not constitute independent importations. We report three importation-detection-efficiency scenarios — 100% (the observed baseline of 4 events), 66% (true count of 6 given 4 observed), and 50% (true count of 8 given 4 observed) — to bound the inference against plausible additional undetected exportations.
Under the most conservative scenario (100% importation detection), the posterior median outbreak size is 1,047 infections (90% CrI: 443–2,043). Under the mid-range scenario (66% importation detection), the median is 1,495 infections (90% CrI: 739–2,644). Under the most expansive scenario (50% importation detection), the median rises to 1,944 infections (90% CrI: 1,056–3,224).
Across all three scenarios, the lower 5th percentile of the posterior remains well above 400 infections, consistent with the surveillance gaps documented in the WHO situation report and the reported healthcare-worker deaths [1].
| Sensitivity: importation-detection efficiency abroad | |||
|---|---|---|---|
| Scenario | Median | 90% CrI | Context |
| 100% importation detection | 1,047 | [443–2,043] | observed baseline (3 Kampala + 1 Bukavu) |
| 66% importation detection | 1,495 | [739–2,644] | moderate undercounting allowance |
| 50% importation detection | 1,944 | [1,056–3,224] | substantial undercounting allowance |
Limitations & Assumptions
Model Scope
The mobility model does not incorporate sociodemographic attributes of travelers (age, economic status, occupation, or pre-existing medical conditions), any of which may modulate both exposure risk at origin and care-seeking behavior at destination. Travel probability within a catchment area is treated as independent of these attributes and of specific location within the catchment.
Case importations are modeled as statistically independent events. Real-world events involving multiple related cases (for example, a family cluster traveling together) are counted as a single importation event. To the extent that such clusters are common, the model will under-count the number of distinct exposure events while correctly counting the number of distinct geographic seedings.
Data Limitations
Mobility data may not fully capture informal cross-border movement, which is substantial in the Ituri–Uganda corridor. The outbreak-size inference is sensitive to the assumed importation-detection-efficiency scenario; actual importation-detection efficiency in Uganda is not directly observed and may vary by port of entry, healthcare-seeking pathway, and clinical presentation.
Epidemiological Assumptions
The incubation period of Bundibugyo virus (2–21 days; average ~10 days) constrains the window within which an exported case could reasonably travel while infectious or pre-symptomatic. The model does not predict downstream transmission chains following importation events: risk estimates reflect the probability of a single importation and do not account for the probability of sustained local transmission in the receiving location.
References
- [1] World Health Organization Regional Office for Africa. Ebola Bundibugyo Virus Disease Outbreak — Democratic Republic of the Congo | Uganda. Weekly External Situation Report 01. Data as of 18 May 2026. Available at: afro.who.int
- [2] World Health Organization. Epidemic of Ebola disease caused by Bundibugyo virus in the Democratic Republic of the Congo and Uganda determined a public health emergency of international concern. WHO Statement, 16 May 2026.
- [3] Centers for Disease Control and Prevention. Ebola Disease: Current Situation. CDC Situation Summary, 18 May 2026.
- [4] Balcan D, Gonçalves B, Hu H, Ramasco JJ, Colizza V, Vespignani A. Modeling the spatial spread of infectious diseases: the GLobal Epidemic and Mobility computational model. Journal of Computational Science. 2010;1(3):132–145.
- [5] Davis JT, Chinazzi M, Perra N, Mu K, Pastore y Piontti A, Ajelli M, Dean NE, Gioannini C, Litvinova M, Merler S, Rossi L, Sun K, Xiong X, Longini IM, Halloran ME, Viboud C, Vespignani A. Cryptic transmission of SARS-CoV-2 and the first COVID-19 wave. Nature. 2021;600:127–132.
- [6] Wamala JF, Lukwago L, Malimbo M, et al. Ebola hemorrhagic fever associated with novel virus strain, Uganda, 2007–2008. Emerging Infectious Diseases. 2010;16(7):1087–1092.
- [7] International Air Transport Association (IATA). Passenger Intelligence Services. IATA, 2026.
- [8] Official Airline Guide (OAG). Aviation Analytics. OAG, 2026.