Evidence for a detection ceiling in the 2026 Bundibugyo virus disease epidemic in the Democratic Republic of the Congo: an analysis of publicly reported aggregate surveillance data
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
The 2026 Bundibugyo virus disease (BVD) epidemic in the Democratic Republic of the Congo (DRC) was declared on 15 May 2026 and determined a public health emergency of international concern on 17 May 2026. Public surveillance reporting consists of cumulative counts by report date; no line list with symptom-onset dates is available. Widely circulated characterisations — that this is the fastest-growing Ebola outbreak on record, that reported cases are doubling every 22 days, and that the case fatality ratio (CFR) is 37.5% — rest on these aggregates. We examined what the published data actually support.
Methods
We assembled twelve published anchor points (15 May–13 July 2026) from WHO, WHO AFRO, NICD and the DRC Institut National de Santé Publique; one was recovered by back-calculation and checked against the directly reported subsequent total. We computed mean daily incidence between anchors and the within-interval death-to-case ratio. We reconstructed symptom-onset dates by Richardson–Lucy deconvolution with right-truncation correction under assumed onset-to-report delays (mean 5, 7 and 9 days), estimated the instantaneous reproduction number (Cori method), and computed three CFR estimators: crude, resolved-case, and outcome-delay-adjusted. Provincial CFRs used exact binomial intervals.
Results
Confirmed cases plateaued at 40–52 per day for eighteen days (25 June–13 July). Over the same period the within-interval death-to-case ratio rose from 0.28 to 0.58. Reconstructed Rt was 1.28 (95% CrI 1.15–1.41) on 7 July, having fallen from approximately 2.9 in mid-May. Growth on the reconstructed onset curve corresponded to a doubling time of approximately 90 days, against 22 days computed from cumulative counts. CFR estimates were 37.5% (crude), 50.2% (outcome-delay-adjusted) and 67.3% (resolved-case). Crude provincial CFR was 34.9% (95% CI 32.7–37.1) in Ituri, 58.2% (50.7–65.5) in North Kivu and 81.0% (58.1–94.6) in newly affected provinces.
Conclusions
A flat case count accompanied by a rising death-to-case ratio is difficult to reconcile with a transmission plateau and is consistent with saturated case detection. Reported case counts appear to have substantially decoupled from transmission, and cannot presently distinguish control from detection failure. Doubling times computed from cumulative totals are artefacts. Test volume and positivity by health zone are the critical missing denominators.
