Major past Ebola epidemics

A summary of the two most significant Ebola epidemics preceding the current Bundibugyo virus outbreak of 2026: the epidemic in West Africa of 2014–2016, still the largest documented to date, and the epidemic in the provinces of North Kivu and Ituri in DR Congo of 2018–2020, which remained the second-largest until 2026. The emphasis is on the course, bringing the epidemic under control and the final outcome — that is, on the lessons that help us understand the current situation.

Key messages

• These two epidemics were caused by Ebola virus (EBOV; species Orthoebolavirus zairense), not by Bundibugyo virus, which is behind the current 2026 outbreak. For EBOV, registered vaccines and monoclonal antibodies are now available; for Bundibugyo they are not.
• West Africa 2014–2016 was the largest Ebola epidemic in history. It lasted more than two years, affected three poor post-war countries with extremely weak health infrastructure, reached the capital cities (Conakry, Monrovia, Freetown) and led to the first-ever UN mission responding to a health crisis (UNMEER, September 2014). The key to bringing it under control was a combination of massive international resources, the building of treatment centres, contact tracing and follow-up, and changes to burial practices, not any single measure on its own.
• North Kivu and Ituri 2018–2020 was the second-largest epidemic until 2026; the current outbreak has surpassed it in both cases and deaths. It took place in an active war zone, with attacks on health centres and dozens of response-team members killed. Even so, it was brought under control through a combination of ring vaccination (Ervebo), monoclonal antibodies (the PALM trial), Ebola treatment centres and community engagement. The case-fatality rate was higher (~66%) than in West Africa, because a large proportion of the sick died in the community before reaching the treatment centres.
• Lessons that also apply to Bundibugyo 2026: the main tools of control are the classic public-health tools — rapid detection, isolation, contact tracing and follow-up, safe burials and infection prevention in healthcare. Vaccines and therapeutics complement these tools but do not replace them, which is especially important for an outbreak for which a specific vaccine is still lacking.

West Africa 2014–2016

Course

The epidemic began at the end of December 2013 in the forested village of Méliandou in southern Guinea. The index case is retrospectively identified as a two-year-old boy, Émile Ouamouno, who died on 28 December 2013. The infection spread from the family into the prefectures of Guéckédou and Macenta, but for several months it remained unidentified — the area lacked diagnostic capacity and local people had no experience of Ebola (none of the previous 24 Ebola epidemics since 1976 had appeared in West Africa).

WHO was officially informed only on 23 March 2014, after laboratory confirmation at the Institut Pasteur in Dakar. By then the virus was already in wider circulation in Guinea, and cases began to appear in neighbouring Liberia (first confirmed case on 30 March) and Sierra Leone (first confirmed on 25 May). The tri-border region, where the borders of Guinea, Liberia and Sierra Leone converge near the town of Guéckédou, became the epicentre.

In April 2014 Guinea had 168 cases and 108 deaths across six prefectures; by the end of May the virus had reached Conakry, the capital with almost two million inhabitants. By then the situation was epidemiologically unprecedented: for the first time in history, Ebola had reached a large urban environment. By July 2014 it had also reached the capitals of Liberia (Monrovia) and Sierra Leone (Freetown). On 8 August 2014 WHO declared a Public Health Emergency of International Concern (PHEIC) — more than four months after the first official report, which was retrospectively criticised as a serious delay.

The peak of the epidemic came in October 2014. That month Liberia was reporting more than 500 new cases a week and Sierra Leone more than 700. At one point CDC was working with a worst-case scenario of 1.4 million infections by January 2015. The international response finally got under way on a large scale: on 16 September President Obama announced the deployment of 3,000 US soldiers to Liberia, the United Kingdom took the lead in Sierra Leone and France in Guinea. On 18 September the UN established UNMEER (United Nations Mission for Ebola Emergency Response), historically the first UN mission responding to a health crisis.

Beyond the three epicentres, limited outbreaks occurred in Nigeria (20 cases, 8 deaths; transmission stopped on 19 October 2014), Mali (8 cases, 6 deaths) and Senegal (1 case, 0 deaths). Secondary transmission among health workers also occurred in the USA (Dallas) and Spain (Madrid); isolated imported cases were reported by the UK and Italy.

After the October 2014 peak the decline began. By January 2015 Liberia was reporting only tens of new cases a week, with Sierra Leone gradually following over the winter and spring of 2015. On 8 May 2015 WHO declared Liberia free of active Ebola transmission, but a return of cases soon followed (including in a 17-year-old boy in June 2015) — a consequence of viral persistence in survivors. On 7 November 2015 WHO declared Sierra Leone free of active Ebola transmission and on 29 December 2015 also Guinea.

The definitive end of the PHEIC was declared on 29 March 2016. The final end of all outbreaks after recurrent flare-ups was declared on 9 June 2016, 42 days after the last negative test of the last patient. By then the epidemic had lasted almost 28 months in total.

Geographical distribution at the end of the epidemic: Guinea (3,814 cases / 2,544 deaths), Liberia (10,678 cases / 4,810 deaths), Sierra Leone (14,124 cases / 3,956 deaths). Before Ebola these countries were at various stages of post-conflict recovery (Liberia and Sierra Leone had emerged from civil wars in 2003 and 2002, while Guinea had an unstable post-dictatorship transition) and had extremely weak health systems — in 2014 Liberia had roughly 50 doctors for the whole country.

Control measures

The key elements of the response, which gradually took hold and which today are regarded as a model for future Ebola epidemics:

Ebola Treatment Centres (ETCs). The main operators were MSF (Médecins Sans Frontières), who had been warning since April 2014 that the epidemic was out of control, and national health workers. The centres provided isolation of the sick, supportive care (rehydration, correction of electrolytes, antibiotics for secondary infections), and at the same time acted as diagnostic hubs. In October 2014 only limited capacity was in operation; by December 2014 UNMEER had helped bring enough ETCs into operation to meet the target criterion of “70% of new patients in care”.

Safe and dignified burials. Traditional burial practices in West Africa (washing the body, a farewell kiss) were a major mechanism of transmission — in the terminal phase the body of the deceased carries the highest viral load. UNICEF and the Red Cross introduced safe-burial programmes, in which specially trained teams in protective suits took charge of the body, disinfected the surroundings and carried out a dignified burial in the presence of the family. UNMEER set itself a target of “70% safe burials” for December 2014. Changing burial practices was one of the most significant factors of control.

Contact tracing and follow-up. Each confirmed or probable case generated a list of contacts, who were monitored for 21 days (the maximum incubation period) and isolated at the first symptoms. This classic public-health tool was the basis of epidemic control. Failures or delays in tracing (e.g. because of the huge mobility of the population in Conakry) correlated with sharp rises in cases.

Community engagement and behaviour change. Anthropologists and local communicators gradually persuaded the population that Ebola was not a curse or a government hoax, but a specific biological threat. This was extremely difficult — in Womey, Guinea, in September 2014 a mob killed eight health workers and journalists and threw their bodies into a septic tank. Gradually winning trust took months; where it succeeded, it strongly helped to reduce transmission. The fall in infections in Liberia at the end of October 2014 went hand in hand with the acceptance of safe burials in communities.

Clinical trials of the rVSV-ZEBOV vaccine (Ervebo). The vaccine had been developed earlier (Canada’s PHAC, later licensed to Merck), but by 2014 it had no human data. In Guinea in 2015 a phase 3 trial with ring vaccination, “Ebola Ça Suffit!”, was carried out, in which, after a new case was identified, its contacts and contacts of contacts were vaccinated. Among 2,119 people vaccinated at least 10 days before exposure, no EVD occurred; vaccine efficacy was estimated at 100% (95% CI 69–100%). This trial changed history: it laid the basis for the registration of Ervebo in November 2019 and for the ring-vaccination strategy in subsequent outbreaks.

Clinical trials of monoclonal antibodies. ZMapp (Mapp Biopharmaceutical) was tested in the PREVAIL II trial in the late phase of the West African epidemic. The results were statistically inconclusive but suggested a favourable trend; ZMapp was therefore used as the control arm in the later PALM trial in the DRC in 2018–2019.

International funding and coordination. The US Congress allocated USD 5.4 billion through emergency supplemental appropriations (P.L. 113-235, December 2014), of which more than USD 2.5 billion went to the international response. Total international contributions reached about USD 5 billion from 70 countries at the start of 2015. Under the ASEOWA initiative, the African Union deployed 824 health specialists. The British army helped bring treatment centres into operation in Sierra Leone, and France did so in Guinea.

IHR (International Health Regulations) and PHEIC. WHO declared a PHEIC on 8 August 2014, but with a long delay — more than four months after the first official report and after calls from the field. The then WHO Director-General, Margaret Chan, was retrospectively criticised for putting off the decision. This experience was one of the main reasons for the subsequent WHO reforms.

What did not work or was harmful: blanket travel bans (introduced by several dozen countries) — both WHO and the UN Security Council rejected them as counterproductive, because they disrupted the flow of humanitarian aid and created space for panic in the media without any epidemiological benefit. Quarantines of entire neighbourhoods (e.g. West Point in Monrovia in August 2014) were counterproductive for community trust.

Outcomes

Final epidemiological data (WHO, 8 May 2016): 28,616 reported cases (suspected, probable and confirmed), 11,310 deaths, an apparent case-fatality rate of 40%. Among the 15,227 laboratory-confirmed cases the case-fatality rate was 74%. WHO itself states that these figures probably underestimate the true scale, especially in the earliest phase in Guinea.

Survivors and post-Ebola syndrome. The epidemic left approximately 17,000 survivors. Many suffer long-term from post-Ebola syndrome: arthritis, headaches, uveitis (inflammation of the eye, sometimes with viral persistence even months after clinical recovery), psychological problems and social stigmatisation. In men the virus was repeatedly detected in semen even 12–18 months after recovery; sexual transmission caused several documented reactivation outbreaks (including a flare-up in Liberia in 2015 and in Guinea in 2021, genetically derived from the original West African lineage).

Economic impact. The World Bank estimated the direct economic losses in Guinea, Liberia and Sierra Leone at USD 2.2 billion for 2015 alone. Secondarily — because of the collapse of health services — the epidemic led to thousands of additional deaths from malaria, HIV, tuberculosis, maternal complications and complications of childbirth. In some regions measles vaccination programmes fell back to the level of the 1980s.

The Ervebo vaccine (rVSV-ZEBOV). Clinical data from the “Ebola Ça Suffit!” trial in Guinea (2015) accelerated the registration process. WHO prequalification November 2019, FDA approval December 2019, conditional EMA marketing authorisation November 2019. The vaccine is a single-dose live attenuated vaccine based on a recombinant vesicular stomatitis virus expressing the EBOV glycoprotein (Kikwit 1995 variant). It is licensed for Ebola virus (EBOV); for disease caused by Bundibugyo virus it is not licensed, and the evidence on possible cross-protection is, according to WHO, limited and inconclusive.

Global health reforms. The failure to respond rapidly to the West African epidemic led to fundamental reforms:

• The WHO Health Emergencies Programme was set up on 1 July 2016 as a new structure within WHO with rapid-deployment capacity.
• The Coalition for Epidemic Preparedness Innovations (CEPI) was founded in January 2017 in Davos as a global mechanism for funding the development of vaccines against epidemic threats.
• The WHO R&D Blueprint introduces a list of priority pathogens with development priorities.
• The Independent Oversight and Advisory Committee (IOAC) was created as external oversight of the WHO emergencies programme.

Geographical and demographic consequences. In some districts Sierra Leone had thousands of orphans; Liberia spent years rebuilding its collapsed health system. The stigmatisation of survivors (especially women survivors and health workers) had long-term social consequences.

This epidemic definitively changed the understanding of Ebola epidemics: it showed that a local outbreak, given delayed detection and weak infrastructure, can grow into a global health crisis.

North Kivu and Ituri (DR Congo) 2018–2020

Course

The epidemic was declared on 1 August 2018 by the DRC Ministry of Health as the country’s 10th Ebola epidemic — just eight days after the official end of the 9th epidemic in Équateur province (May–July 2018, 54 cases, 33 deaths, Bikoro and Iboko). Sequencing at INRB Kinshasa confirmed that the two epidemics were genetically independent; they were two separate zoonotic introductions of Ebola virus (EBOV).

The index cases of the 10th epidemic were identified in the Mabalako health zone in North Kivu province. A cluster of 26 cases of acute haemorrhagic fever with 20 deaths was reported in mid- to late July 2018; samples from six patients sent to INRB were positive for Ebola virus (EBOV). From Mabalako the virus spread to other health zones in North Kivu, gradually into Ituri (first confirmed case on 13 August 2018), and later also into South Kivu. The main affected towns were: Beni, Butembo, Katwa, Mabalako, Mangina.

The epidemic ran in three waves:

1. First wave (August 2018 – February 2019). Spread from Mabalako via Beni to Butembo, gradually into further health zones. The vaccination programme began almost immediately (rVSV-ZEBOV under an emergency-use protocol), which limited the geographical extent. By 1 October 2018 more than 12,000 people had already been vaccinated.

2. Second wave (March–August 2019). Escalation in Butembo and Katwa, where the epidemic developed most. This wave was markedly affected by attacks on health workers and treatment centres (see control measures). In June 2019 the first cross-border transmission to Uganda occurred: a five-year-old Congolese boy, who had reached the Mpondwe-Lhubirira border crossing with his family, tested positive in Bwera and died. On 17 July 2019 WHO declared a PHEIC, after two earlier meetings at which the emergency committee had not recommended declaring a PHEIC. In this wave the epidemic also reached Goma, the capital of North Kivu with a million inhabitants and a border crossing into Uganda and Rwanda.

3. Third wave (September 2019 – June 2020). A gradual decline. By October 2019 new cases were reported mainly from Mandima and Mambasa in Ituri. The end of transmission was finally achieved through a combination of more intensive vaccination, better diagnostics and gradually earned community trust. On 25 June 2020 WHO declared the end of the epidemic, 42 days without a new case since the last discharged patient.

Geographical distribution at the end: North Kivu province was the most affected, Ituri second, South Kivu marginally, plus 4 cross-border cases in Uganda (Kasese District, June 2019; all from one family, brought from the DRC; three of them returned to the DRC, where they died, and are therefore counted in the DRC statistics).

The epidemic took place in an area affected by more than 20 years of armed conflict between the DRC army (FARDC), the Mai-Mai militias, the Allied Democratic Forces group (ADF, originally a Ugandan Islamist group operating in the DRC) and other actors. In 2018 North Kivu had more than five million internally displaced people, high mobility (economic migration for work in the mines, commercial movement across borders), low trust in state institutions and a multigenerational experience of violence. These factors defined both the course and the control measures.

Control measures

The key elements of the response were, compared with 2014–2016, markedly more technologically advanced (a vaccine and therapeutics were available), but operationally much harder (a war zone):

Ring vaccination with the rVSV-ZEBOV vaccine (Ervebo). Deployed from August 2018 under an expanded-access regimen. The principle: after a new case was identified, its contacts and contacts of contacts (the “ring”), plus front-line health workers, were vaccinated. In total, more than 265,000 people were vaccinated, of whom 102,515 were followed up for safety. Real-world effectiveness, according to a later retrospective study (Lancet Infectious Diseases 2024, test-negative design), was 84% (95% CrI 70–92). NEJM 2023 separately published an analysis from the DRC that showed an EVD incidence 0.43× lower among the vaccinated (95% CI 0.29–0.65). Among front-line health workers the incidence 30 days after vaccination was only 0.4 cases per 1,000 people. The vaccine requires a cold chain at −60 to −80 °C, which was a significant operational problem.

Clinical trials of monoclonal antibodies — the PALM trial. “Pamoja Tulinde Maisha” (Swahili for “Together Save Lives”) was a randomised controlled trial of three experimental therapies: REGN-EB3 (Inmazeb, Regeneron, a three-antibody cocktail), mAb114 (Ebanga, Ridgeback, a single antibody) and remdesivir (Gilead) against the control ZMapp (Mapp Biopharmaceutical, a three-antibody cocktail). The trial began on 20 November 2018 at four ETCs (Beni, Katwa, Butembo, Mangina). By 9 August 2019, 681 patients had been enrolled (out of a planned 725).

Preliminary DSMB results on 9 August 2019 led to the early termination of the trial, because REGN-EB3 and mAb114 clearly dominated. The final 28-day mortality published in NEJM 2019:

• ZMapp (control): 49.7% (84/169)
• Remdesivir: 53%
• mAb114: 35.1% (61/174), p=0.007 vs. ZMapp
• REGN-EB3: 33.5% (52/155), p=0.002 vs. ZMapp

The key finding was that in patients with low viral load (NP-Ct >22, i.e. those arriving early) the 28-day mortality was only 11% with REGN-EB3 and 6% with mAb114, that is, survival over 90%. This is a dramatic change from the historical case-fatality rate of 60–90% for Ebola virus (EBOV) infection. Conversely, in patients with high viral load (late arrival, multi-organ failure) the mortality was still 64–70% even with the best therapeutics. The conclusion: early detection and early admission to a treatment centre remain key.

On the basis of the PALM trial, the FDA registered Inmazeb in October 2020 and Ebanga in December 2020, both as treatment for Ebola virus (EBOV) infection. For Bundibugyo these products are not registered, and their clinical benefit cannot be automatically assumed.

Ebola Treatment Centres (ETCs). The main operators were MSF, ALIMA (The Alliance for International Medical Action), the DRC Red Cross and national health workers. Centres in Beni, Butembo, Katwa, Mangina, Mambasa and other localities combined isolation, supportive care, the administration of monoclonal antibodies (after the PALM trial began), contact tracing and community communication.

Safe and dignified burials. As in West Africa, this was a key tool. In the DRC the burial teams were organised by the Red Cross, often with the assistance of FARDC or MONUSCO for security.

Community engagement. Anthropologists (including the team of Juliet Bedford from Anthrologica) worked with local leaders, churches and traditional healers. In North Kivu community engagement was especially difficult — a combination of distrust of “Kinshasa” as a distant government, myths about Ebola as a “political weapon” during the election period (December 2018), and traditional burial rituals.

Security and attacks on health workers. A distinctive feature of this epidemic were attacks on health workers and health facilities. In 2019 alone WHO documented more than 300 attacks on health workers and health facilities in North Kivu/Ituri, with 6 deaths of health workers and 70 injuries:

• 24 February 2019: attack on the ETC in Katwa, destruction of the centre and withdrawal of MSF.
• 9 March 2019: attack on the ETC in Butembo, injury of three health workers and the death of a Congolese police officer.
• 19 April 2019: attack on the University Hospital in Butembo, the death of Dr Richard Mouzoko Kibounga, a WHO epidemiologist from Cameroon, who was chairing a meeting with health workers.
• 8 May 2019: attack by 50 militiamen in Butembo near WHO accommodation; a firefight with DRC forces.
• 27 November 2019: attack on the response teams’ base in Biakato Mines and on the coordination office in Mangina, the death of four response-team members and one police officer.

After the April 2019 attack, health workers in Butembo threatened to strike; MSF withdrew operations from some facilities. FARDC and MONUSCO strengthened their presence at ETCs, but the militarisation of the health response was controversial — it risked blurring the line between health workers and security forces and, paradoxically, increasing distrust. A modelling study published in Eurosurveillance 2020 estimated that the attacks in Butembo increased transmissibility and the number of cases. After an adjusted strategy was introduced, including expanded ring vaccination around clusters and the engagement of local health workers, this outbreak was gradually brought under control.

International coordination. WHO coordinated the response with national health authorities and partners including CDC, MSF, Africa CDC, UNICEF and the World Bank. UN Security Council Resolution 2439 (2018) called for the protection of workers responding to Ebola, but unlike in 2014–2016 no separate UNMEER-type UN mission was established.

What worked and what did not. The early deployment of the vaccine worked (the impact on North Kivu was demonstrable), the clinical trials of therapeutics (PALM changed the treatment paradigm), and the cooperation of INRB with international partners (rapid sequencing). What did not work: the initial underestimation of the security situation, the unsustainably long time before the PHEIC was declared (almost a year), and communication failures in some communities, where health workers were perceived as “government agents” during the election period.

Outcomes

Final epidemiological data (DRC MoH, WHO): 3,481 cases (3,323 confirmed, 158 probable), 2,299 deaths, 1,162 survivors. The apparent case-fatality rate of 66% was markedly higher than in West Africa. The main reason was the high proportion of deaths outside treatment centres: some of the sick never reached an ETC at all and others arrived late because of distrust and fear.

Health workers among the sick. Health workers made up about 12% of all cases. Vaccinating front-line health workers was one of the first priorities and had a measurable effect — see above, an incidence of 0.4/1,000 30 days after vaccination.

The Ervebo vaccine — registration. Data from the DRC were key for WHO prequalification (November 2019), conditional EMA marketing authorisation (November 2019) and FDA approval (December 2019). Ervebo is today part of the recommendations of the WHO Strategic Advisory Group of Experts on Immunization for the response to outbreaks caused by Ebola virus (EBOV).

A second vaccine — Zabdeno/Mvabea (Johnson & Johnson). A two-dose regimen (Zabdeno as the first dose + Mvabea as the booster dose), developed in parallel. The EMA approved it in July 2020. It is suitable for preventive vaccination in periods without an active outbreak (it requires 8 weeks between doses), not for ring vaccination during an acute outbreak.

Monoclonal antibodies — registration. The FDA approved Inmazeb (REGN-EB3, Regeneron) in October 2020 and Ebanga (mAb114, Ridgeback) in December 2020. Both drugs are registered as treatment for Ebola virus (EBOV) infection. These drugs dramatically changed the prognosis of patients admitted early.

Recurrence and persistence. In June 2020, in the last days of the North Kivu/Ituri epidemic, the 11th Ebola epidemic broke out in Équateur province, genetically distinct. In February 2021 in Guinea (the N’Zérékoré area) a new outbreak then broke out, genetically derived from the original West African lineage of 2014–2016, which showed that the virus can persist in survivors for more than five years and trigger a new outbreak. Likewise, in August 2022 in North Kivu (Beni Health Zone) a small outbreak broke out (1 confirmed case), genetically related to the 10th epidemic, which supported the hypothesis of a recurrent source from a survivor.

Lessons for the future.

• Ring vaccination is feasible even in a conflict zone, provided a suitable vaccine exists; for Bundibugyo no such approved vaccine yet exists.
• For Ebola virus (EBOV) infection, the PALM trial showed that early detection and the early administration of effective monoclonal antibodies markedly reduce mortality (from the historical 60–90% to less than 10% in patients with low viral load).
• The PALM trial showed that a high-quality randomised clinical trial is feasible even during an acute epidemic in a conflict zone — a precedent for future trials, including trials for Bundibugyo.
• Attacks on health workers are a specific threat. The response must rest mainly on community engagement, not on militarisation.
• Viral persistence in survivors requires long-term medical follow-up and programmes for male survivors.

What this means for Bundibugyo 2026

• The previous successes with the Ervebo vaccine and monoclonal antibodies concern mainly Ebola virus (EBOV); they cannot be automatically transferred to Bundibugyo virus.
• The most useful lessons are organisational: rapid detection, isolation, contact tracing, safe burials, protection of health workers and work with communities.
• Clinical trials during an epidemic are possible, but they require community trust, national leadership, logistics and a clear distinction between a research candidate and an approved treatment.