Before you believe “the outbreak is slowing”
A worked example from my desk, dated so you can see the method.
On 12 September I checked whether the DRC Bundibugyo Ebola outbreak was slowing. Three sources, read in full:
- WHO Disease Outbreak News 2026-DON617 (data to 7 Sep): 6,757 confirmed cases, 3,267 deaths, crude CFR 48.3%.
- ECDC's outbreak page (data to 12 Sep): 7,200 confirmed, 3,475 deaths, 923 patients in isolation.
- UKHSA week 36 and the DRC ministry's own 13 Sep update, to cross-check the direction.
Answer: not yet visible in the national counts. Ituri, the epicentre, added 1,114 cases in 21 days; North Kivu added 453.
The caveat that governs any “slowing” story: WHO says part of the increase may reflect strengthened surveillance, laboratory capacity, and data reconciliation. Daily notification counts are not a clean transmission signal in either direction. A story that reads them as one is guessing.
Where a real turn would show first: 21-day per-zone windows and onset-date curves, before the daily line moves.
That is the whole method. Read the primary. State the limit. Name what would change the answer.
Free sample brief, same build: https://telegra.ph/DRC-Bundibugyo-outbreak--is-it-slowing-09-18 The desk: https://telegra.ph/Verified-research-briefs--sources-attached-09-18