Unpaid Responders and a Misfit Shot Let Congo’s Ebola Spread

Congo’s health ministry said Monday that Bundibugyo Ebola has killed 3,226 people among 6,686 confirmed cases, with figures running through Sept. 6. It is already the country’s largest Ebola outbreak and the second largest on record.

On Tuesday, WHO technical officer Luca Fontana said the response still needs 1,600 more treatment beds and about 5,000 more health workers. The people who would fill those jobs have spent the summer fighting the virus without pay, while the only licensed Ebola shot was built for a different species.

Congo’s Count Has Already Beaten 2018

The ministry’s Monday update put recoveries at 1,563 and said 819 patients were still in isolation or hospital care. The case fatality ratio stood at 48.3 percent. Cases have been recorded in 61 health zones across six provinces: Ituri, North Kivu, South Kivu, Haut-Uele, Tshopo, and Bas-Uele. Kinshasa has numbered this as the country’s 17th Ebola epidemic.

Congo declared the outbreak on May 15 after Bundibugyo virus was found in Mongbwalu, a gold-mining town in Ituri. WHO called it a public health emergency of international concern on May 17. Africa CDC followed a day later with a continental emergency. From the declaration to Sept. 6 is 114 days. U.S. CDC researchers, writing with data through Aug. 21, said the climb had already reached about 5,000 confirmed cases in 100 days.

That pace left behind the 2018 to 2020 North Kivu epidemic, which logged 3,470 cases and 2,287 deaths over about two years and had been Congo’s worst. Ituri is still the centre. As of Aug. 21 it accounted for 84 percent of confirmed cases. South Kivu, by the Monday update, had gone 97 days without a new confirmed case. The virus is not moving evenly. It is moving fast where people move.

WHO and Africa CDC announced an immediate initial release of 70,000 doses of Merck’s Ervebo shot on Aug. 20, split as 50,000 for front-line workers and 20,000 for a Phase 3 trial. The stockpile those doses came from was built for Zaire ebolavirus, the species behind West Africa’s 2014 to 2016 disaster and Congo’s 2018 fight. Bundibugyo is a different species.

HOW THIS OUTBREAK COMPARES

WHO has warned that if this climb holds, the death toll can pass West Africa’s. A CDC modeling note in June said low isolation rates could produce an epidemic on that scale. The recorded count is already past every prior Bundibugyo outbreak combined, and past Congo’s own 2018 to 2020 peak.

Most New Cases Were Never on a Contact List

The ministry can publish a daily total. It cannot see most of the chains that feed it. CDC staff abstracting Congo’s own situation reports found that for a stretch of the summer, only 15 to 20 percent of new cases had been listed as known contacts. The control target is above 90 percent. In plain terms, four fifths of new patients were surprises.

CONTROL GAPS IN LATE AUGUST

• Contact lists: Teams named 10.6 contacts per confirmed case against a target of at least 20, a sign that many exposed people were never written down.
• Deaths in the community: 59 percent of confirmed deaths occurred outside an Ebola treatment unit, against a target of zero.
• Testing: 72 percent of validated alerts got a laboratory test, against a target above 90 percent, and 24 percent of those tests came back positive.
• Burials: Only 49 percent of affected health zones had a safe burial team, against a target of 100 percent.

Alerts investigated within 24 hours last came in at 83 percent, short of the 90 percent bar, and even that figure was last published on Aug. 5. National bed occupancy sat at 64 percent, under the 80 percent ceiling, but some units were running at 140 percent. CDC’s field note, posted Sept. 1 with data through Aug. 21, said five core control measures had not met their targets and that the outbreak was still expanding.

Africa CDC’s Prof. Yap Boum told a briefing in August that experts judged only 30 to 40 percent of infections were being found, which is the basis for the agency’s warning that the true outbreak may be about three times the official file. Milder Bundibugyo illness, he said, often presents without bleeding, so it hides among other fevers. Jean Kaseya, Africa CDC’s director-general, said in early August that more than 70 percent of new infections were turning up in the community rather than among people already under watch.

Money has already moved, and Ervebo doses have left the stockpile. The death curve has not bent with them. The lists that would make ring vaccination and isolation work are still missing most of the people on them.

Who Is Carrying the Virus Through Ituri?

The index cluster sat in Mongbwalu, a gold town whose pits and trading paths pull men in and send them out again. WHO’s August risk paper named insecurity, thin clinics, funding gaps, and the lack of a licensed Bundibugyo vaccine as the drag on control, and it flagged population movement as a reason chains stay unnamed. Artisanal mining is not a side detail in Ituri. It is how a fever in one camp becomes a case in the next health zone.

THE GROUPS THE DAILY TALLY SKIPS

• Pit crews and traders: Gold work around Mongbwalu keeps people moving across zone lines that contact tracers treat as fixed.
• Displaced households: Armed conflict in the east has emptied villages and filled roads, so a burial in one place seeds illness in another.
• Checkpoint staff: Six screening posts in Haut-Uele stopped filing reports in mid-August after pay dried up, which punches holes in the map the ministry still publishes.
• School-age children: Classes resumed on Sept. 1 across a country already counting cases in six provinces, even as the main teachers’ union called for a boycott over pay.

The Syndicat des enseignants du Congo asked for a monthly wage of $500, or the franc equivalent, and told members not to restart lessons. The government promised an open, calm return. Temperature checks at school gates do not replace contact lists that already miss most new patients. Classrooms pack together the age group that moves between home, market, and church, which is exactly the mix that made West Africa so hard to stop.

Imported cases have already shown how far a single traveler can go. Uganda recorded 20 confirmed infections and two deaths, then declared its outbreak over on July 28. France treated one returning health worker. Germany recorded an imported case. WHO still rates the risk very high inside Congo, high for land neighbours, and low for the rest of the world, with no sustained spread outside the country. The next jump, if it comes, will likely follow the same roads the miners already use.

Front-Line Staff Have Gone Months Unpaid

Dr. Biensi Kano sits on the epidemiological surveillance committee in Bunia, Ituri’s capital. “Since the Ebola virus disease outbreak was declared, we’ve been demanding payment for our work,” he said, after staff in the epicentre began walking out in July. Nurses, drivers, grave diggers, and decontamination crews have described the same gap: risk without wages, from the May 15 declaration onward.

I hope their conscience tells them that I deserve to be paid.

Bondele, front-line Ebola responder, Ituri

Bondele, a veteran of the 2018 response who came back for this one, said he had worked unpaid since mid-May. “With risks like that, we deserve salaries.” In Rwampara, health worker Bahati Claude asked how two months could pass with no pay at all. In Bunia, a nurse who gave his name as Jean-Pierre asked how long they were expected to risk their lives for free. A nurse named Kayimpa, a mother of four, said she had not received a single government paycheck and that her name was misspelled on the payroll, which stalled her file.

Public Health Minister Roger Kamba has not denied the mess. He has said 15 to 20 percent of names submitted for some recent Ebola payments were fictitious, and that money existed but lists were being padded. On one centre’s roster of 240 people, officials could find only 40. Ghost names on a payroll are not a bookkeeping quirk in an Ebola zone. They delay the real staff, and they give strikers a reason to shut the gate.

HOW THE PAY DISPUTE HIT THE RESPONSE

1. May 15, 2026: Congo declares Bundibugyo virus disease. Front-line workers later say wages and outbreak bonuses stop from this date.
2. July 13, 2026: Staff at Rwampara General Hospital strike over unpaid salaries and bonuses, shut the site, and burn a tire in the road.
3. July 25, 2026: Doctors, nurses, and guards halt work at the Elikya treatment centre in Bunia over two months of unpaid performance bonuses.
4. Mid-August 2026: Six checkpoints in Haut-Uele stop reporting during a pay strike; weeks later several still are dark, and three Ituri posts join them.
5. Sept. 6, 2026: The ministry’s file reaches 6,686 confirmed cases and 3,226 deaths, while pay disputes still interrupt isolation, hygiene, and screening.

Hygiene crews struck at a Bunia treatment centre in early August and disrupted admissions. Decontamination workers later walked out and left infection-control jobs unfinished in Ituri. By Aug. 19, only five of 18 responders staffing North Kivu checkpoints had been paid. Community-engagement teams in North Kivu and Tshopo had gone three months without wages. Congo’s difficulty paying clinic staff did not start in May. It is now deciding whether contact tracing, the one tool that still works without a matching vaccine, can function at all.

The Licensed Shot Does Not Match This Virus

There are currently no licensed vaccines for Bundibugyo virus disease, and none for Sudan virus or Taï Forest virus either. Ervebo is licensed only for Ebola virus disease caused by Zaire ebolavirus. WHO reviewed the animal and lab file and said there is not enough evidence to know whether Ervebo works against Bundibugyo. Some studies hint at a degree of protection, WHO said, but the results are limited and uncertain. WHO does not currently recommend using Ervebo to answer a Bundibugyo outbreak.

Kinshasa asked for the stockpile anyway. The International Coordinating Group released 70,000 doses, with Gavi putting $7 million to ship 70,000 doses and another $6 million from its First Response Fund. Kamba launched the campaign in Kisangani, capital of Tshopo, in late August, and said the first shots would go to health workers and to people who had been near patients. A later WHO emergency note, dated Aug. 31, said off-label use should sit inside research protocols and that a ring randomized trial should generate the efficacy data that still do not exist. Full protection after Ervebo, even against the virus it was built for, takes 10 to 14 days.

From January 2021 through July 2026, that same ICG stockpile had already sent more than 56,000 Ervebo doses into Zaire-strain responses in Congo. The tool is familiar. The virus is not. Thomas Geisbert, a virologist who helped build rVSV Ebola vaccines, has said an early blend left Bundibugyo out because it had caused so few outbreaks. “We guessed wrong.” Candidate Bundibugyo-specific shots are in development, and Gavi has pointed to a $40 million First Response Fund line to speed them. They will not arrive in time for the patients already in Ituri’s tents.

WHO’s Africa director, Prof. Mohamed Janabi, said in August that new animal data suggested Ervebo might offer some cross-protection against death, and that a technical group wanted the licensed shot prioritized for a randomized trial. He also said the file must still lead.

Promising development on #Ebola: new data from animal studies suggest that Ervebo, the only licensed vaccine against Ebola Zaire, may offer some cross-protection against #Bundibugyo virus, particularly against death from the disease.@WHO’s Technical Advisory Group recommends… pic.twitter.com/sLTTbJKiG4

— Prof. Mohamed Janabi (@ProfJanabi) August 7, 2026

That trial is the honest use of a Zaire shot in a Bundibugyo epidemic. Using the same product as if it were a proven ring around Mongbwalu is a different claim, and it is the claim WHO has refused to make. This newsroom has already set out how licensed Ebola shots miss Bundibugyo. The September counts have only made that mismatch more expensive.

Beds Arrived Faster Than the People to Staff Them

Fontana told reporters in Geneva on Tuesday that nearly 1,400 treatment beds are now up across 59 centres, after more than 900 were built in the first two months. He called it one of the fastest Ebola treatment build-outs he had seen, including during West Africa. Patients, he said, are arriving from villages far beyond Bunia, so the next beds have to sit closer to new clusters, not only in the original epicentre.

Yet, despite this extraordinary effort, we estimate that another 1,600 beds will be required.

Luca Fontana, technical officer, WHO Health Emergencies Programme, Geneva briefing

WHO wants about 3,000 beds. The staffing math is blunt: three health workers per patient bed, or about 9,000 people for that planned capacity. Fontana said nearly 4,000 already work inside the centres, so the gap is about 5,000 recruits. “The health ministry is identifying medical staff from other provinces to bring them in. The issue is to identify them fast enough to meet growing needs.” Médecins Sans Frontières has more than 1,700 staff across Ituri, North Kivu, South Kivu, Haut-Uele, and Tshopo. That is a large partner footprint. It is not 5,000 extra pairs of hands.

A 50-bed centre costs about $500,000 to run, Fontana said, and a single-use protective suit runs about $25 per task. WHO spokesperson Christian Lindmeier said a six-month WHO-Africa CDC plan is costed at $115 million, with $112.4 million received or offered, a $2.6 million gap he expects to widen as the map grows. The pool of groups that can actually run an Ebola centre, Fontana added, has shrunk after global donor cuts. Beds without payrolls become the same problem as lists without contacts: a response that exists on paper while 59 percent of the dying never reach a ward.

Why Earlier Bundibugyo Outbreaks Stayed Small

Bundibugyo virus had caused only two known human outbreaks before this year. Uganda in 2007 and 2008 counted 149 cases and 37 deaths. Congo in 2012 counted 36 cases and 13 deaths. Both were stopped with the old tools: find the sick, list the contacts, isolate, bury the dead without infecting the family. Those files were small enough that a missing vaccine did not decide the ending. This file is not.

Uganda’s 2026 imported cluster, 20 confirmed cases and two deaths, ended on July 28 after detection in six days and a response measured in hours. Congo’s side of the same border is a war economy with gold pits, unpaid burial teams, and treatment units that some families still fear. CDC’s August snapshot found that most operational indicators were below the levels that end Ebola epidemics, and that data for several others were simply missing. The 2018 to 2020 Zaire outbreak in these same provinces lasted about two years and still killed 2,287 people, even with Ervebo in the rings. This species has no such ring that WHO will stand behind.

Fontana said Kinshasa is now hunting clinicians in quieter provinces and trying to move them east before the next row of tents opens. The ministry’s Monday ledger is already past every Bundibugyo outbreak on record, and past Congo’s own worst Ebola count, 114 days from the declaration. The unpaid surveillance officer in Bunia, the miner leaving Mongbwalu, and the child who went back to class on Sept. 1 are the people who decide whether 3,226 is a midpoint or a floor.

Disclaimer: This article is news reporting and analysis of a public-health emergency, and it is for information only. It is not medical advice, a diagnosis, or a recommendation to take or refuse any vaccine, treatment, or travel decision. Anyone who may have been exposed to Ebola, or who has fever or bleeding after time in an affected area, should seek care from a qualified physician or local health authority and follow official isolation guidance. Case counts, death totals, vaccine-trial plans, and pay-status claims reflect the ministry, WHO, CDC, and Africa CDC figures cited here as of the dates given and will change as the outbreak and the response change.