How Did Ebola Spread Unseen for 3 Months in DRC?

February's Secret: How the DRC Ebola Outbreak Got a Three-Month Head Start

Somewhere in Ituri Province, around February 2026, the DRC Ebola outbreak began moving through human bodies. Nobody counted it. Nobody sequenced it. Nobody called anyone. That silence lasted roughly three months, and in the arithmetic of outbreaks, three months is a catastrophe waiting to finish writing itself.

Viral sequencing, the painstaking business of reading a pathogen's genetic text and tracing its family tree, now places the true origin of this epidemic in February 2026. The official declaration came on May 14, when the DRC Ministry of Health announced an Ebola outbreak in the Mongbwalu health zone of Ituri Province. That two-sentence gap, February to May, is not a clerical error. It is a data point about what happens when early-warning infrastructure quietly collapses.

Here is the strange part. Surveillance systems are not passive nets that catch whatever falls into them. They require money, people, and maintained trust in local communities. Funding cuts to the International Rescue Committee, one of the organizations providing that on-the-ground early-warning capacity, contributed directly to the delayed detection. The virus did not hide cleverly. The listening posts were simply understaffed.

Three months of undetected spread is not a biological fact alone. It is a policy outcome. By the time May 14 arrived and the official count began, the epidemic had already accumulated invisible momentum — a head start that responders have been trying to close ever since.

Meet Bundibugyo: The Ebola Strain That Arrived Without a Vaccine

The Ebolavirus genus contains six distinct species. Most people have heard of one: the Zaire strain, the villain of the 2014 West Africa epidemic that killed more than 11,000 people. Bundibugyo ebolavirus is a different animal entirely — rarer, less studied, and for now, unmatched by any licensed vaccine or treatment on the planet.

Here is the strange part. When Zaire ebolavirus returned to eastern DRC in 2018, responders had a weapon: Ervebo, a vaccine that proved effective in the field and helped end that outbreak. That tool is now useless. Bundibugyo is genetically distinct enough that Ervebo offers no meaningful protection against it. The 2026 outbreak has effectively reset the clock to a pre-vaccine era of response, and the WHO's declaration of a Public Health Emergency of International Concern on May 17 — just three days after the official outbreak announcement — signaled how seriously that fact was being taken.

Now hold that thought, and consider the speed. The 2018 Zaire outbreak took 235 days to reach its first 1,000 confirmed cases. This outbreak reached the same milestone in 40 days. That is not a marginal difference — it is a different category of event. Epidemiologists do not yet have a clean explanation for the acceleration. Surveillance collapse, undercounting, community transmission patterns, the particular biology of Bundibugyo — all are candidates, none are proven. What is clear is that the response inherited a virus that travels faster than its predecessor, armed only with the tools that worked against a different enemy. The numbers make that mismatch visible. The reasons behind them remain honestly open.

The Numbers That Made This the Deadliest Ebola Crisis in DRC History

By August 16, 2026, the confirmed death toll stood at 2,325. That number needs a moment to land: 2,325 people dead, from a single outbreak, in a single country, in roughly three months of counted time. It is the highest mortality count ever recorded in the Democratic Republic of the Congo's 17 previous encounters with Ebola.

Here is the strange part. A single outbreak number tells you the size; the rate tells you the terror. On one 24-hour stretch in mid-August, epidemiologists logged 101 new confirmed cases. That is not a peak from the early chaotic weeks, when surveillance was thin and the counts were catching up. That is August, months into a declared emergency, with international responders on the ground.

The total case count reached 4,945. Ituri Province, where the virus first surfaced in February, remains the epicenter, accounting for 4,194 of those cases and 1,838 of the deaths. But Ituri's dominance does not mean the virus stayed local. It has now reached 55 health zones spread across six provinces, ranging from North Kivu and South Kivu in the east to Bas-Uelé pushing westward. That is not one fire. That is an archipelago of fires.

Now hold that thought. Each new province means a fresh population with no prior exposure to this strain, new health workers learning protocols under pressure, new community trust to earn from scratch. The geography of spread is, in practical terms, a map of how much harder the response becomes with every additional health zone. These numbers are not a historical record to file away. They are the live dimensions of the problem as it stands today.

The 2026 outbreak has effectively reset the clock to a pre-vaccine era of response.

Why Almost Half of Patients Are Dying — and Why That Number Is Getting Worse

In June, the case fatality ratio for this outbreak sat at roughly 20 percent. Grim, but within a range epidemiologists have learned to work inside. By mid-August, that number had climbed to 46.8 percent. The direction is the wrong one.

Thomas Parisch, a public health specialist with MSF, named the mechanism plainly: "Normally, as an outbreak progresses, the case fatality ratio should fall. Instead, we're still seeing many cases detected very late." Late detection means the virus has already done most of its damage by the time anyone with a kit and a treatment protocol arrives. The body has less time to fight; the numbers show it.

Here is the strange part. More than 70 percent of deaths are occurring not in clinics or treatment centers, but in homes, in villages, in the community. Jean-Paul Malo Lotsima, a civil society official in Djugu, put it simply: "People prefer to be treated at home, thinking it is a case of poisoning or some other illness." That preference is not irrational distrust born from nowhere. It has a history, fed by years of misinformation about what this disease is and where it came from.

The consequence is invisible mortality. A death at home, without a diagnosis, goes uncounted. The official toll of 2,325 is almost certainly an undercount of the actual scale.

Distrust has also turned physical. Misinformation about safe burial practices — the rituals that matter most in transmission control — has led to direct attacks on healthcare facilities. Workers trying to reach the sick cannot always get there safely. The response loses ground not just to the virus, but to the fear surrounding it.

Airlifts Over Frontlines: The Logistics of Responding When Roads Are War Zones

Picture the 2014 West Africa outbreak response, which was brutal but at least navigable by truck. Now remove the roads. Active armed frontlines in eastern DRC have turned ground supply routes into frontlines themselves, leaving the United Nations Humanitarian Air Service to build a logistics bridge by air, shuttling critical supplies between Entebbe in Uganda and Bunia in Ituri Province. When your delivery network is a fleet of aircraft operating over a conflict zone, the word "bottleneck" takes on a different weight.

The comparison to previous responses is hard to miss. In 2018, the Zaire-strain outbreak unfolded in a difficult region, but health workers could reach most communities by road, by river, by foot. Here, the geography is governed by armed groups, and that changes everything: the speed, the cost, the reach, the morale of responders.

The human toll extends far past the case count. During this crisis, the number of people requiring humanitarian assistance in the region grew from 14.9 million to 18.6 million. That gap — 3.7 million additional people — reflects displacement, collapsed supply chains, and a health system stretched past its frame.

Uganda, at least, demonstrated what containment can look like with functional infrastructure and open borders. It recorded 20 cases and 2 deaths before declaring its portion of the outbreak over on July 28, 2026. Now hold that thought: cross-border surveillance is still running, because a suppressed fire near a dry forest is not the same as a fire that never was.

Two Candidate Vaccines, One Imported Case, and the Questions the Ebola Response Cannot Answer Yet

In July 2026, clinical trials began in Ituri for two candidate vaccines specifically designed against the Bundibugyo strain. That is genuinely good news. It is also, in the same breath, a reminder of how far there is still to go: neither product is licensed, neither is deployable at scale, and the path from "trial started" to "vials in a health worker's hands" is measured in months at best, more likely years.

Here is the strange part. While those trials were enrolling their first participants, a single imported case appeared in France in August 2026. One person. One flight, presumably. One demonstration that a six-province outbreak is not a regional problem sealed behind a border.

Now hold that thought, because three questions sit at the center of this crisis with no answers attached. The original animal-to-human spillover — the moment in approximately February 2026 when the DRC Ebola outbreak first crossed into a person — has no identified source. The true scale of the epidemic, given that 70 percent of deaths occur at home and outside any counting system, is genuinely unknown. And no one can say with honesty when a deployable Bundibugyo vaccine will exist.

We still don't know. That, honestly, is the best part about science, and the worst part about an Ebola outbreak in the DRC that is not waiting for the science to catch up.