On 16 May 2026 the WHO Director-General declared a Public Health Emergency of International Concern (PHEIC) for the outbreak of Ebola disease caused by Bundibugyo virus in the Democratic Republic of the Congo and Uganda [4]. As of 27 July 2026, DRC has reported 3,360 cumulative laboratory-confirmed cases and 1,487 confirmed deaths (case-fatality ratio ≈ 44 %) across 48 health zones spanning five provinces, with 733 patients currently hospitalised in isolation at Ebola Treatment Centers (ETC) and treatment facilities [1,2]. Uganda has reported 20 laboratory-confirmed cases including 2 deaths [3].
This report presents a stochastic, metapopulation transmission model built using the GLEAM framework [5], incorporates transmission occurring in the hospital, community, and through funerals, and it explicitly models isolation protocols through ETC. The model is calibrated in two stages using an Approximate Bayesian Computation approach on observed cross-border importations from Ituri to Uganda, and on the cumulative laboratory-confirmed count in Ituri at 18 July 2026 (2,090 cases) assuming a case ascertainment rate as low as 40 %.
Key findings
1
Posterior R₀ ≈ 2.02 (90 % CrI [1.72, 2.47]); empirical doubling time 12.9 d (90 % CrI [10.7, 15.6]). Estimates pool the three calibration runs with equal prior weight, shown here for the iso50 reference scenario (n = 812); the other isolation scenarios give essentially the same posterior.
2
Forward projection to 1 August 2026 in DRC. Cumulative symptomatic cases: iso30 5,551 [3,352 – 8,245], iso50 4,991 [2,981 – 7,318], iso75 4,414 [2,645 – 6,363]. Active cases (I + H + ETC): iso30 1,287 [651 – 2,342], iso50 870 [423 – 1,630], iso75 462 [206 – 946]. New symptomatic cases per day: iso30 158 [75 – 298], iso50 95 [44 – 195], iso75 42 [16 – 96].
3
Scenario outcomes by 30 September 2026 in DRC. Probability of ≥ 20,000 cumulative symptomatic cases: < 1 % under 75 % isolation, 20 % under 50 %, 75 % under 30 %. High isolation (75 %) keeps cumulative size below 20,000 in nearly all trajectories and drives the outbreak toward control by end of September. The intermediate 50 % target contains transmission but still allows one in five trajectories to exceed 20,000 cumulative cases. Low isolation (30 %) fails to control the outbreak: three quarters of trajectories exceed 20,000 cases.
The intervals reported below are predicted under the assumed model and intervention scenario. Calibration extends only to the 18 July 2026 confirmed-case evidence. Projections beyond that date are model-driven and depend on assumptions about isolation timing and intensity.
2 — Calibration and posteriors
Two-stage calibration and posteriors
The model is calibrated by Approximate Bayesian Computation in two stages. Stage 1 conditions on observed cross-border importations from Ituri to Uganda, requiring exactly one Uganda importation by 14 May 2026 and exactly one further importation in the window 14 to 18 May 2026. Stage 2 conditions the Stage-1 posterior on the cumulative laboratory-confirmed count in Ituri at 18 July 2026 (2,090 cases) allowing ascertainment as low as 40 %, requiring the simulated cumulative symptomatic onsets in Ituri by 18 July to lie in the window [2,090, 5,225].
The filtered posterior — pooled with equal prior weight across the three calibration runs and shown here for the iso50 reference scenario (n = 812) — yields a median R₀ of 2.02 (90 % CrI [1.72, 2.47]) and a median empirical doubling time of 12.9 d (90 % CrI [10.7, 15.6 d]), measured per draw by log-linear fit to cumulative symptomatic onsets in Ituri over the pre-isolation window before 15 May 2026. The two other isolation scenarios yield essentially the same posterior (median R₀ shifts by less than 0.05).
Figure 1 — Filtered posterior: R₀ and empirical doubling time
Posterior densities over the accepted trajectories under the iso50 reference scenario (n = 812), pooled with equal prior weight across the three calibration runs. R₀ is taken directly from the joint posterior; the doubling time Td is measured per draw by log-linear fit to cumulative symptomatic onsets in Ituri (C ∈ [10, 1000]) over the pre-isolation window before 15 May 2026. KDE on linear axes (Gaussian kernel, Scott bandwidth); dashed line marks the median; shaded band the 90 % credible interval; dots below the baseline show a 250-trajectory subsample.
Parameter
Median
90 % CrI
R₀
2.02
[1.72 – 2.47]
Td (days)
12.9
[10.7 – 15.6]
3 — Forward projection at 1 August 2026
Posterior predictive distributions, DRC · 1 August 2026 · three isolation scenarios
Forward projections evaluate three target isolation rates, 30 %, 50 % and 75 %, each phased in at 2 percentage points per day from 20 May 2026 across DRC (17 basins plus the Goma catchment) until the scenario target is reached. Uganda is held at a constant 80 % isolation from the start of the intervention window. The isolation Q determines the probability with which each new symptomatic case is routed to the isolation pathway (Iiso → HETC → ETC). Isolation-tracked cases are identified faster than routine cases (onset-to-HCF 2 d versus 5 d on the routine track) and progress to the ETC compartment, where transmission stops and any deaths receive safe and dignified burial. A small residual nosocomial channel (10 % of the baseline HCF rate) is retained during the 3-day HCF → ETC transit, reflecting imperfect IPC during transfer; community deaths and routine-HCF deaths still contribute to funeral transmission.
The Stage-2 posterior predictive distributions in DRC at the projection date 1 August 2026, two weeks after the evidence-filter date, are summarised below across the three target isolation scenarios (30 %, 50 %, 75 %). At this near-term horizon the three cumulative distributions still overlap heavily (medians 4,414 – 5,551), while active cases and daily incidence show clearer scenario separation as the higher-isolation trajectory transitions to declining incidence and the lower-isolation trajectory continues to grow.
Figure 2 — Posterior predictive distributions · DRC · 1 August 2026 · three isolation scenarios
Filtered posterior predictive distributions in DRC on 1 August 2026 under three target isolation scenarios (rows: 30 %, 50 %, 75 %). Columns: cumulative symptomatic cases (left), active cases I + H + ETC (middle), and new symptomatic cases per day (right). Shared log x-axis per column. Gaussian KDE on log-transformed values; dashed line marks the median; shaded band the 90 % credible interval. n = 660 (iso30), 812 (iso50), 905 (iso75).
Scenario
Cumulative
Active
Daily new
30 %
5,551
1,287
158
50 %
4,991
870
95
75 %
4,414
462
42
Medians shown; 90 % CrIs in Fig 2. n = 660 / 812 / 905.
4 — Long-term scenario projections
Long-term scenario projections through 30 September 2026
To bridge the near-term predictive posteriors (Section 3) with the long-term outlook that follows, we first show the predicted trajectories of active and daily-new symptomatic cases through mid-August (Fig 3) and the estimated effective reproduction number Rt through end-September (Fig 4). These two views clarify the dynamics that drive the 30 September outbreak-size and controllability metrics (Figs 5 and 6).
Predicted trajectories · 1 July – 15 August 2026
Between 1 July and 15 August 2026 the three scenarios diverge as the isolation cap is approached. Under 75 % isolation active symptomatic cases turn over and both active cases and daily new onsets decline steadily through the window; under 50 % isolation active cases continue to grow slowly, consistent with the effective reproduction number holding just above 1 (see Fig 4); under 30 % isolation the outbreak continues to grow at a faster rate. The 50 % and 90 % predictive-interval bands together with 15 sampled trajectories per panel indicate the within-scenario variability.
Figure 3 — Predicted trajectories in DRC · 1 July – 15 August 2026
Active symptomatic and daily new symptomatic cases in DRC across the three isolation scenarios (rows: active symptomatic, daily new symptomatic; columns: iso30, iso50, iso75). Bands: 50 % (darker) and 90 % (lighter) predictive intervals; solid line: median; thin lines: 15 individual trajectories per panel.
Scenario
Active @ 15 Aug
Daily @ 15 Aug
30 %
1,864
227
50 %
1,008
110
75 %
361
33
Medians shown; 90 % PIs in Fig 3.
Effective reproduction number · 1 June – 30 September 2026
The effective reproduction number Rt (Cori estimator on the daily onsets, 7-day sliding window) is shown from 1 June through 30 September 2026 for each scenario. All three start near R₀ ≈ 2 in early June, when the isolation ramp has just begun and its coverage is still small. As the cap is approached the scenarios diverge cleanly: iso75 crosses R = 1 in mid-July and stabilises at 0.77 (90 % PI 0.55 – 0.96) by September; iso50 stays just above 1 at 1.18 (0.97 – 1.39); iso30 remains around 1.42 (1.21 – 1.67). Only the iso75 asymptote is unambiguously below 1, which explains the ordering of the outbreak-size and controllability metrics that follow.
Figure 4 — Effective reproduction number Rt in DRC · 1 June – 30 September 2026
Effective reproduction number Rt (Cori 7-day sliding window) from 1 June to 30 September 2026, one panel per isolation scenario. Median (solid), 50 % PI (darker band) and 90 % PI (lighter band). Horizontal dashed reference at R = 1. Dotted horizontal line marks the asymptotic median (median Rt over 1 – 30 September) shown as an annotation on each panel.
Scenario
Asymptotic Rt
90 % PI
30 %
1.42
[1.21 – 1.67]
50 %
1.18
[0.97 – 1.39]
75 %
0.77
[0.55 – 0.96]
Median of Rt over 1 – 30 September 2026.
Probability of a large outbreak by 30 September 2026
The probability that the cumulative number of symptomatic cases in DRC exceeds 10,000 by 30 September 2026 is 98 % under 30 % isolation, 69 % under 50 % and 9 % under 75 %. For the higher threshold of ≥ 20,000 cases, the corresponding probabilities are 75 %, 20 % and < 1 %.
Figure 5 — Probability of a large outbreak in DRC by 30 September 2026
Probability that cumulative symptomatic cases in DRC exceed 10,000 (left) and 20,000 (right) by 30 September 2026, by target isolation rate. Bars show the fraction of filtered posterior trajectories above each threshold; n = 660 (iso30), 812 (iso50), 905 (iso75).
Target isolation
P(≥ 10,000)
P(≥ 20,000)
30 %
98 %
75 %
50 %
69 %
20 %
75 %
9 %
< 1 %
Controllability — daily incidence on 30 September 2026
As a measure of controllability we compute the probability that daily new symptomatic onsets in DRC exceed 10 cases on 30 September 2026. Under 30 % isolation the probability is 100 %. Under 50 % isolation the probability is 100 %. Under 75 % isolation it drops to 63 %, indicating that a substantial share of trajectories cross below the 10-daily-onsets threshold by that horizon under strong isolation while sustained low-level transmission persists in the remainder.
Figure 6 — Probability of > 10 daily symptomatic onsets in DRC on 30 September 2026
Probability that daily new symptomatic onsets in DRC exceed 10 on 30 September 2026, by target isolation rate. Bars show the fraction of filtered posterior trajectories above 10 daily new cases on that date; n = 660 (iso30), 812 (iso50), 905 (iso75).
Target isolation
P(daily > 10)
30 %
100 %
50 %
100 %
75 %
63 %
5 — Methods
Methods
Model framework
The framework is GLEAM (Global Epidemic and Mobility Model) [5], an Africa-wide metapopulation network coupled by daily commuting matrices and origin-destination passenger flows from IATA Passenger Intelligence Services and OAG Aviation Analytics. The within-basin compartmental dynamics are simulated as a discrete-time multinomial chain-binomial process. At each time step the number of individuals exiting a compartment is drawn from a binomial with exit probability 1 − exp(−Σᵢ λᵢ · Δt), and exits are partitioned across the available destination compartments by a multinomial whose probabilities are proportional to the per-route hazards. We do not assume any mobility reductions between subpopulations: no behavioral reduction, no border closure is explicitly modeled.
Compartmental structure
The compartmental structure follows a Legrand-like SEIHFR scheme [7,8]. However, in order to model the impact of isolation and safe burial protocols, we increase the complexity of the model to incorporate three potential outcomes an infectious individual could experience (see Fig. 7 for details):
Community Branch: the individual does not get admitted to any healthcare facility. They are able to infect people in the community in their infectious stage and, if death occurs, through unsafe burial practices.
HCF-only Branch: the infected individual is infectious for an average time of 5 days until they are admitted to a HCF where they can generate new infections through nosocomial transmission and, if death occurs, through unsafe burial practices.
Isolation Branch: individuals are infectious for an average time of 2 days before they are moved to a HCF for an average time of 3 days and have a transmissibility that is 10 % of individuals in a HCF in the HCF-only branch. They are then moved to an ETC, where they no longer contribute to ongoing transmission.
The funeral compartment F is fed only by community deaths and regular-HCF deaths (deaths in ETC receive safe and dignified burial and bypass F). Transmission rates by compartment: βI in all three community-infectious compartments (Icom, IcomH, Iiso), βH in regular-track H, βH,iso = 0.1 · βH in the HCF → ETC transit compartment HETC, zero in ETC, and βF in F.
Natural-history parameters
Compartmental dwell times and severity values are based on literature. The Legrand-type SEIHFR formulation [7,8] supplies the framework, Wamala et al. [9] are the source for the original 2007–2008 Uganda BVD outbreak data on which some of the BVD-specific natural-history estimates rest. Isolation-pathway case-finding timings (Tiso and The) follow the 2018 North Kivu RVEAP response [6].
Symbol
Meaning
Value
Source
TE
Latent period (Erlang-2 mean)
8 d (prior U(5, 10))
[8, 9]
TI
Icom dwell (community, no HCF)
9.6 d
[7, 8]
Tsh
Onset to HCF admission, routine track
5 d
[8]
Tiso
Onset to HCF admission, iso track
2 d
[6]
The
HCF to ETC transit, iso track
3 d
[6]
TH
HCF stay, routine track
4.6 d
[7, 8]
TETC
ETC stay
4.6 d
assumed equal to TH
TF
Funeral period
2 d
[7, 8]
pH
HCF admission probability (Legrand baseline)
0.80
[8, 9]
δnh
CFR, non-hospitalised
0.50
supportive-care gradient
δh
CFR, regular HCF
0.45
supportive-care gradient
δetc
CFR, ETC
0.30
best clinical management
R₀ split
Community : HCF : Funeral at baseline
0.44 : 0.22 : 0.34
[8]
κiso
HCF iso transmission factor (βH,iso / βH)
0.1
assumed (residual nosocomial)
Figure 7 — Compartmental structure of the model
Twelve compartments, three potential outcomes at E₂ exit. Latent stages E₁ → E₂ feed three parallel infectious-community sub-compartments: Icom (community-only fate, dwell TI), IcomH (regular HCF fate, dwell Tsh = 5 d) and Iiso (isolation fate, dwell Tiso = 2 d). Routine cases progress to H with full nosocomial transmission rate βH; isolated cases progress through HETC (residual rate 0.1 · βH) to ETC (no transmission, safe burial). The funeral compartment F (transmission rate βF) is fed only by community and regular-HCF deaths.
Seeding
The epidemic originates in Ituri, DRC, with 5 infectious cases on the inferred start date. Every other basin (including all of Uganda) begins fully susceptible and is reached only through mobility coupling.
Inference — two-stage calibration
The model is calibrated by Approximate Bayesian Computation in two stages. Stage 1 conditions on observed cross-border importations from Ituri to Uganda, requiring exactly one Uganda importation by 14 May 2026 and exactly one further importation in the window 14 to 18 May 2026. Stage 2 conditions the Stage-1 posterior on the cumulative laboratory-confirmed count in Ituri at 18 July 2026 (2,090 cases) allowing ascertainment as low as 40 %, requiring the simulated cumulative symptomatic onsets in Ituri by 18 July to lie in the window [2,090, 5,225].
Parameter
Prior
Notes
R₀
U(1.2, 5.0)
Basic reproduction number
Incubation period TE
U(5, 10) d
Erlang-2 (two-stage latent)
Outbreak start date
U(15 Feb, 15 Mar 2026)
Date of first case in Ituri
Ascertainment sensitivity
The Stage-2 evidence filter assumes a specific ascertainment rate that maps the laboratory-confirmed count to a credible window on simulated cumulative onsets. The baseline analysis allows ascertainment as low as 40 % (window [2,090, 5,225]).
Posterior medians of R₀ and Td are stable when the ascertainment floor is relaxed or tightened; a formal sensitivity table at alternative floors on the 18 July 2026 anchor is deferred to a subsequent update.
Isolation mechanism
Forward projections evaluate three target isolation rates, 30 %, 50 % and 75 %, each phased in at 2 percentage points per day from 20 May 2026 across DRC (17 basins plus the Goma catchment) until the scenario target is reached. Uganda is held at a constant 80 % isolation from the start of the intervention window. The isolation Q determines the probability with which each new symptomatic case is routed to the isolation pathway (Iiso → HETC → ETC). Isolation-tracked cases are identified faster than routine cases (onset-to-HCF 2 d versus 5 d on the routine track) and progress to the ETC compartment, where transmission stops and any deaths receive safe and dignified burial. A small residual nosocomial channel (10 % of the baseline HCF rate) is retained during the 3-day HCF → ETC transit, reflecting imperfect IPC during transfer; community deaths and routine-HCF deaths still contribute to funeral transmission.
Empirical doubling-time estimation
The doubling time Td reported in §2 is measured directly from each accepted trajectory rather than derived analytically from R₀. For each draw we extract daily new symptomatic onsets in Ituri from the iso50 scenario, build the cumulative series C(t), and fit log-linear by ordinary least squares: log C(t) = a + r · t, then Td = ln 2 / r. The fit window is C(t) ∈ [10, 1000], truncated at 14 May 2026 (the last day before the isolation ramp begins).
Limitations and Assumptions
Ascertainment assumption.
The Stage-2 evidence filter allows ascertainment as low as 40 % for the laboratory-confirmed Ituri count at the 18 July 2026 snapshot. Posterior medians of R₀ and Td are stable when the ascertainment floor is relaxed or tightened; the upper credible-interval bounds move modestly with the floor.
Calibration targets.
Stage 1 uses only evidence of importations from Ituri → Uganda importations (exact-count, no tolerance band); within-Ituri incidence is not directly fit in Stage 1. Stage 2 adds a single cumulative count at one date. Alternative acceptance rules or additional within-basin time-points would shift the posteriors.
Care-pathway parameters.
Isolation-track timings (Tiso = 2 d, HCF → ETC ≈ 3 d) follow the 2018 North Kivu RVEAP response, which may not exactly match the 2026 operational reality. The residual nosocomial channel in the iso branch (0.1 · βH) is a modelling assumption, sensitivity-testable.
Scope.
All figures and statistics in this report are aggregated to DRC (17 DRC basins plus the Goma catchment). Cross-border spillover to Uganda, Rwanda, South Sudan, etc., is not reported here (see Report #6 for international dissemination).
Mobility and seasonality.
Mobility is held static across the simulation, with no behavioral reduction, no border closure modeled, and no seasonality in transmissibility. Informal cross-border movement may be under-captured.
Isolation policy.
The isolation ramp (2 percentage points per day from 20 May 2026 across DRC plus the Goma catchment; Uganda held at 80 %) and the case-finding timings are applied homogeneously across all symptomatic cases. Real-world coverage varies by location and over time.
6 — References
References
1World Health Organization Regional Office for Africa. Ebola Bundibugyo Virus Disease Outbreak — Democratic Republic of the Congo | Uganda. Weekly External Situation Report. Available at: https://insp.cd/category/sitrep/
2Institut National de Santé Publique, République Démocratique du Congo. Situation reports, Maladie à Virus Bundibugyo. Available at: https://insp.cd/category/sitrep/
5Balcan 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. J Comput Sci. 2010;1(3):132–145.
6Ajelli M, Merler S, Fumanelli L, Pastore y Piontti A, et al. Vaccination strategies for Ebola in the Democratic Republic of Congo: the WHO-Ebola modeling collaboration. International Journal of Infectious Diseases. 2025;153:107779.
7Legrand J, Grais RF, Boelle PY, Valleron AJ, Flahault A. Understanding the dynamics of Ebola epidemics. Epidemiology and Infection. 2007;135(4):610–621.
8Gomes MFC, Pastore y Piontti A, Rossi L, Chao D, Longini I, Halloran ME, Vespignani A. Assessing the international spreading risk associated with the 2014 West African Ebola outbreak. PLOS Currents Outbreaks. 2014.
9Wamala 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.
Epistorm · Northeastern University / CFA–CDC Insight Net
On 16 May 2026 the WHO Director-General declared a Public Health Emergency of International Concern for the outbreak of Ebola disease caused by Bundibugyo virus in DRC and Uganda. As of 27 July 2026 DRC has reported 3,360 cumulative laboratory-confirmed cases and 1,487 confirmed deaths (case-fatality ratio ≈ 44 %) across 48 health zones spanning five provinces, with 733 patients hospitalised in isolation. We present a stochastic, metapopulation transmission model built using the GLEAM framework, which incorporates transmission in the hospital, the community and through funerals, and which explicitly models isolation protocols through Ebola treatment centres (ETC). The model is calibrated by Approximate Bayesian Computation in two stages, on observed cross-border importations from Ituri to Uganda, and on the cumulative laboratory-confirmed count in Ituri at 18 July 2026 (2,090 cases), allowing ascertainment as low as 40 %. The accepted posterior (three calibration runs pooled with equal prior weight, shown for the iso50 reference scenario, n = 812) yields R₀ = 2.02 (90 % CrI [1.72, 2.47]) and an empirical doubling time of 12.9 d (90 % CrI [10.7, 15.6]). Forward projections in DRC at 1 August 2026 give median cumulative cases of 5,551 under 30 % isolation, 4,991 under 50 % isolation, and 4,414 under 75 % isolation. Under three isolation scenarios (30 %, 50 %, 75 %) phased in from 20 May 2026 at 2 percentage points per day, the probability of ≥ 20,000 cumulative cases in DRC by 30 September 2026 is 75 %, 20 % and < 1 % respectively. High isolation drives the outbreak toward control by end of September, while lower isolation leaves it uncontrolled.
1 Background
On 16 May 2026 the WHO Director-General declared a Public Health Emergency of International Concern (PHEIC) under the International Health Regulations (2005) for the outbreak of Ebola disease caused by Bundibugyo virus (BVD) in the Democratic Republic of the Congo and Uganda. As of the 27 July 2026 snapshot used in this report, DRC has reported 3,360 cumulative laboratory-confirmed cases and 1,487 confirmed deaths (case-fatality ratio ≈ 44 %) across 48 health zones spanning five provinces, with 733 patients hospitalised in isolation at Ebola treatment centres (ETCs) and treatment facilities. Uganda has reported 20 laboratory-confirmed cases including 2 deaths.
The outbreak originated in Ituri (Ituri Province, DRC) and has produced two attributed cross-border importations to Uganda by mid-May. Local isolation interventions are being rolled out across the affected basins. This report estimates the posterior burden of the outbreak in DRC and projects forward trajectories under three counterfactual isolation scenarios, conditioning on the observed importation pattern and on the cumulative laboratory-confirmed case count at 18 July 2026.
All figures and statistics in this report are aggregated to DRC (17 DRC basins plus the Goma catchment). Cross-border spillover to Uganda, Rwanda, South Sudan, etc., is not reported here (see Report #6 for international dissemination).
2 Results
2.1 Posterior of R₀ and the doubling time
The accepted posterior — three calibration runs pooled with equal prior weight, shown for the iso50 reference scenario (n = 812) — yields a median basic reproduction number R₀ of 2.02 (90 % CrI [1.72, 2.47]) and a median empirical doubling time of 12.9 d (90 % CrI [10.7, 15.6 d]), measured by log-linear fit to cumulative Ituri onsets over the pre-isolation window before 15 May 2026. The other two isolation scenarios (iso30, iso75) yield essentially the same posterior.
Figure 1. Filtered posterior over the accepted trajectories under the iso50 reference scenario (n = 812), pooled with equal prior weight across the three calibration runs. R₀ posterior (left) and empirical doubling-time posterior (right). KDE on linear axes (Gaussian kernel, Scott bandwidth); dashed line marks the median; shaded band the 90 % credible interval; dots below the baseline show a 250-trajectory subsample.
2.2 Predictive posteriors at 1 August 2026
The filtered posterior predictive distributions in DRC at the projection date 1 August 2026 give the following medians. Cumulative symptomatic cases: iso30 5,551 [3,352, 8,245]; iso50 4,991 [2,981, 7,318]; iso75 4,414 [2,645, 6,363]. Active cases (I + H + ETC): iso30 1,287 [651, 2,342]; iso50 870 [423, 1,630]; iso75 462 [206, 946]. New symptomatic cases per day: iso30 158 [75, 298]; iso50 95 [44, 195]; iso75 42 [16, 96]. At this near-term horizon the three cumulative distributions still overlap heavily (medians 4,414 – 5,551), while active cases and daily incidence show clearer scenario separation as the higher-isolation trajectory transitions to declining incidence and the lower-isolation trajectory continues to grow.
Figure 2. Filtered posterior predictive distributions in DRC on 1 August 2026 under three target isolation scenarios (rows: 30 %, 50 %, 75 %; columns: cumulative, active, daily new). Shared log x-axis per column. n = 660 / 812 / 905.
2.3 Long-term scenario projections through 30 September 2026
To bridge the near-term predictive posteriors of Section 2.2 with the long-term outlook that follows, we first show the predicted trajectories of active and daily-new symptomatic cases through mid-August (Figure 3) and the estimated effective reproduction number Rt through end-September (Figure 4). These clarify the dynamics that drive the 30 September outbreak-size and controllability metrics (Figures 5 and 6).
Between 1 July and 15 August 2026 the three scenarios diverge as the isolation cap is approached. Under 75 % isolation active symptomatic cases turn over and both active cases and daily new onsets decline steadily through the window; under 50 % isolation active cases continue to grow slowly, consistent with Rt holding just above 1; under 30 % isolation the outbreak continues to grow at a faster rate.
Figure 3. Predicted trajectories in DRC · 1 July – 15 August 2026. Active symptomatic and daily new symptomatic cases across the three isolation scenarios. Bands: 50 % and 90 % predictive intervals; solid line: median; thin lines: 15 individual trajectories per panel.
The effective reproduction number Rt (Cori estimator on daily onsets, 7-day sliding window) starts near R₀ ≈ 2 in early June across all three scenarios, when the isolation ramp has just begun. As the cap is approached the scenarios diverge: iso75 crosses R = 1 in mid-July and stabilises at 0.77 (90 % PI 0.55 – 0.96) by September; iso50 stays just above 1 at 1.18 (0.97 – 1.39); iso30 remains around 1.42 (1.21 – 1.67). Only iso75 crosses cleanly below R = 1.
Figure 4. Effective reproduction number Rt in DRC · 1 June – 30 September 2026, one panel per isolation scenario. Median (solid), 50 % PI (darker) and 90 % PI (lighter). Horizontal dashed reference at R = 1; dotted horizontal reference at the September asymptotic median.
Forward projections evaluate three target isolation rates (30 %, 50 % and 75 %), each phased in at 2 percentage points per day from 20 May 2026 across DRC + Goma; Uganda is held at a constant 80 % isolation. The probability that cumulative symptomatic cases in DRC exceed 10,000 by 30 September 2026 is 98 % under 30 % isolation, 69 % under 50 % and 9 % under 75 %. For the higher threshold of ≥ 20,000 cases, the corresponding probabilities are 75 %, 20 % and < 1 %.
Figure 5. Probability that cumulative symptomatic cases in DRC exceed 10,000 (left) and 20,000 (right) by 30 September 2026, by target isolation rate. n = 660 / 812 / 905.
As a complementary measure of controllability, we compute the probability that daily new symptomatic onsets in DRC exceed 10 cases on 30 September 2026. Under 30 % isolation the probability is 100 %. Under 50 % isolation the probability is 100 %. Under 75 % isolation it drops to 63 %, indicating that a substantial share of trajectories cross below the threshold under strong isolation while sustained low-level transmission persists in the remainder.
Figure 6. Probability that daily new symptomatic onsets in DRC exceed 10 on 30 September 2026, by target isolation rate. n = 660 / 812 / 905.
3 Methods
3.1 Model framework
The framework is GLEAM (Global Epidemic and Mobility Model) [5], an Africa-wide metapopulation network coupled by daily commuting matrices and origin-destination passenger flows from IATA Passenger Intelligence Services and OAG Aviation Analytics. The within-basin compartmental dynamics are simulated as a discrete-time multinomial chain-binomial process. At each time step the number of individuals exiting a compartment is drawn from a binomial with exit probability 1 − exp(−Σᵢ λᵢ · Δt), and exits are partitioned across the available destination compartments by a multinomial whose probabilities are proportional to the per-route hazards. We do not assume any mobility reductions between subpopulations: no behavioral reduction, no border closure is explicitly modeled.
3.2 Compartmental structure
The compartmental structure follows a Legrand-like SEIHFR scheme [7,8]. To model the impact of isolation and safe burial protocols, we incorporate three potential outcomes an infectious individual could experience (Figure 7):
Community Branch: the individual does not get admitted to any healthcare facility. They infect people in the community in their infectious stage and, if death occurs, through unsafe burial practices.
HCF-only Branch: the individual is infectious for an average of 5 days until admission to a HCF, where they can generate new infections through nosocomial transmission and, if death occurs, through unsafe burial practices.
Isolation Branch: the individual is infectious for an average of 2 days before being moved to a HCF for 3 days with transmissibility 10 % of the HCF-only branch, and is then moved to an ETC where they no longer contribute to ongoing transmission.
The funeral compartment F is fed only by community deaths and regular-HCF deaths (deaths in ETC receive safe and dignified burial and bypass F). Transmission rates by compartment: βI in all three community-infectious compartments (Icom, IcomH, Iiso), βH in regular-track H, βH,iso = 0.1 · βH in the HCF → ETC transit compartment HETC, zero in ETC, and βF in F.
Figure 7. Compartmental structure of the model. Twelve compartments, three potential outcomes at E₂ exit. Latent stages E₁ → E₂ feed three parallel infectious-community sub-compartments. Routine cases progress to H (βH); isolated cases progress through HETC (0.1 · βH) to ETC (no transmission, safe burial). F (βF) is fed only by community and regular-HCF deaths.
3.3 Natural-history parameters
Compartmental dwell times and severity values are based on literature. The Legrand-type SEIHFR formulation [7,8] supplies the framework, Wamala et al. [9] are the source for the original 2007–2008 Uganda BVD outbreak data on which some of the BVD-specific natural-history estimates rest. Isolation-pathway case-finding timings (Tiso and The) follow the 2018 North Kivu RVEAP response [6].
Symbol
Meaning
Value
Source
TE
Latent period (Erlang-2 mean)
8 d (prior U(5, 10))
[8, 9]
TI
Icom dwell (community, no HCF)
9.6 d
[7, 8]
Tsh
Onset to HCF admission, routine track
5 d
[8]
Tiso
Onset to HCF admission, iso track
2 d
[6]
The
HCF to ETC transit, iso track
3 d
[6]
TH
HCF stay, routine track
4.6 d
[7, 8]
TETC
ETC stay
4.6 d
assumed = TH
TF
Funeral period
2 d
[7, 8]
pH
HCF admission probability (Legrand baseline)
0.80
[8, 9]
δnh
CFR, non-hospitalised
0.50
supportive-care gradient
δh
CFR, regular HCF
0.45
supportive-care gradient
δetc
CFR, ETC
0.30
best clinical management
R0 split
Community : HCF : Funeral at baseline
0.44 : 0.22 : 0.34
[8]
κiso
HCF iso transmission factor (βH,iso/βH)
0.1
assumed (residual nosocomial)
3.4 Seeding
The epidemic originates in Ituri, DRC, with 5 infectious cases on the inferred start date. Every other basin (including all of Uganda) begins fully susceptible and is reached only through mobility coupling.
3.5 Two-stage calibration
The model is calibrated by Approximate Bayesian Computation in two stages. Stage 1 conditions on observed cross-border importations from Ituri to Uganda, requiring exactly one Uganda importation by 14 May 2026 and exactly one further importation in the window 14 to 18 May 2026. Stage 2 conditions the Stage-1 posterior on the cumulative laboratory-confirmed count in Ituri at 18 July 2026 (2,090 cases) allowing ascertainment as low as 40 %, requiring the simulated cumulative symptomatic onsets in Ituri by 18 July to lie in the window [2,090, 5,225].
Parameter
Prior
Notes
R0
U(1.2, 5.0)
Basic reproduction number
Incubation period TE
U(5, 10) d
Erlang-2 (two-stage latent)
Outbreak start date
U(15 Feb, 15 Mar 2026)
Date of first case in Ituri
3.6 Ascertainment sensitivity
The baseline analysis allows ascertainment as low as 40 % (window [2,090, 5,225]). Under this floor the accepted posterior has n = 660 (iso30), 812 (iso50), 905 (iso75). Tighter or looser floors leave posterior medians of R₀ and Td essentially unchanged; the upper credible-interval bounds shift modestly with the floor.
3.7 Isolation mechanism
Forward projections evaluate three target isolation rates, 30 %, 50 % and 75 %, each phased in at 2 percentage points per day from 20 May 2026 across DRC (17 basins plus the Goma catchment) until the scenario target is reached. Uganda is held at a constant 80 % isolation from the start of the intervention window. The isolation Q determines the probability with which each new symptomatic case is routed to the isolation pathway (Iiso → HETC → ETC). Isolation-tracked cases are identified faster than routine cases (onset-to-HCF 2 d versus 5 d on the routine track) and progress to the ETC compartment, where transmission stops and any deaths receive safe and dignified burial. A small residual nosocomial channel (10 % of the baseline HCF rate) is retained during the 3-day HCF → ETC transit, reflecting imperfect IPC during transfer; community deaths and routine-HCF deaths still contribute to funeral transmission.
3.8 Empirical doubling-time estimation
The doubling time Td is measured directly from each accepted trajectory rather than derived analytically from R₀. For each draw we extract daily new symptomatic onsets in Ituri from the iso50 scenario, build the cumulative series C(t) and fit log-linear by ordinary least squares: log C(t) = a + r · t, then Td = ln 2 / r. The fit window is C(t) ∈ [10, 1000], truncated at 14 May 2026 (the last day before the isolation ramp begins). The fit window is C(t) ∈ [10, 1000], truncated at 14 May 2026 (the last day before the isolation ramp begins).
3.9 Limitations and assumptions
Ascertainment assumption. The Stage-2 evidence filter allows ascertainment as low as 40 % for the laboratory-confirmed Ituri count at the 18 July 2026 snapshot. Posterior medians of R₀ and Td are stable when the ascertainment floor is relaxed or tightened; the upper credible-interval bounds move modestly with the floor. Calibration targets. Stage 1 uses only evidence of importations from Ituri → Uganda importations (exact-count, no tolerance band); within-Ituri incidence is not directly fit in Stage 1. Stage 2 adds a single cumulative count at one date. Alternative acceptance rules or additional within-basin time-points would shift the posteriors. Care-pathway parameters. Isolation-pathway timings follow the 2018 North Kivu RVEAP response, which may not exactly match 2026 operational reality. The residual nosocomial channel in the iso branch (0.1 · βH) is a modelling assumption. Scope. All figures and statistics are aggregated to DRC (17 DRC basins plus the Goma catchment). Cross-border spillover is not reported here (see Report #6). Mobility and seasonality. Mobility is held static across the simulation, with no behavioral reduction, no border closure modeled, and no seasonality. Isolation policy. The isolation ramp and case-finding timings are applied homogeneously across all symptomatic cases; real-world coverage varies.
References
[1] World Health Organization Regional Office for Africa. Ebola Bundibugyo Virus Disease Outbreak, Democratic Republic of the Congo | Uganda. Weekly External Situation Report. https://insp.cd/category/sitrep/
[2] Institut National de Santé Publique, République Démocratique du Congo. Situation reports, Maladie à Virus Bundibugyo. https://insp.cd/category/sitrep/
[3] Ministry of Health, Uganda. Ebola disease, situation updates. https://evd-daily.health.go.ug/
[4] World Health Organization. Disease Outbreak News, Ebola disease caused by Bundibugyo virus, Democratic Republic of the Congo and Uganda (DON 606). https://www.who.int/emergencies/disease-outbreak-news/item/2026-DON606
[5] 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. J Comput Sci. 2010;1(3):132–145.
[6] Ajelli M, Merler S, Fumanelli L, Pastore y Piontti A, et al. Vaccination strategies for Ebola in the Democratic Republic of Congo: the WHO-Ebola modeling collaboration. International Journal of Infectious Diseases. 2025;153:107779.
[7] Legrand J, Grais RF, Boelle PY, Valleron AJ, Flahault A. Understanding the dynamics of Ebola epidemics. Epidemiol Infect. 2007;135(4):610–621.
[8] Gomes MFC, Pastore y Piontti A, Rossi L, Chao D, Longini I, Halloran ME, Vespignani A. Assessing the international spreading risk associated with the 2014 West African Ebola outbreak. PLOS Currents Outbreaks. 2014.
[9] Wamala JF, Lukwago L, Malimbo M, et al. Ebola hemorrhagic fever associated with novel virus strain, Uganda, 2007–2008. Emerg Infect Dis. 2010;16(7):1087–1092.
