Ebola Virus Explained — How It Works and Why It's So Lethal

Ebola virus explained in one sentence: a group of related filoviruses that kill roughly half of the people they infect, spread only through direct contact with body fluids, and have never gone pandemic. The World Health Organization (WHO) puts average case fatality near 50 percent, with past outbreaks ranging from 25 to 90 percent. In 2026 the Democratic Republic of the Congo (DRC) is fighting the largest Bundibugyo virus outbreak on record, caused by a species with no licensed vaccine or treatment. What follows covers how the viruses work, how they travel, and what medicine can and cannot do.

Educational content, not medical advice. Talk to your doctor before changing any medication, treatment, or routine. Information here reflects current research but every person's situation differs.

What "Ebola" actually refers to

"Ebola" is shorthand for a small family of viruses, not a single pathogen. They are filoviruses, named for their long, thread-like shape under the microscope, and their genetic material is RNA. The genus is now called Orthoebolavirus, and WHO counts six species in it. Three of them are known to cause large outbreaks in people: Ebola virus (formally Orthoebolavirus zairense, often called Zaire ebolavirus), Sudan virus, and Bundibugyo virus.

The distinction between species matters more than the shared name suggests. A vaccine or antibody built against one species does not automatically protect against another, a point the U.S. Food and Drug Administration (FDA) makes explicitly in its Ebola guidance. The table below sorts out the cast.

Species	Disease it causes	Human disease status	Source
Ebola virus (Zaire)	Ebola virus disease	Large outbreaks; licensed vaccines and antibodies exist	WHO, FDA
Sudan virus	Sudan virus disease	Large outbreaks; no approved vaccine or treatment	WHO
Bundibugyo virus	Bundibugyo virus disease	Large outbreaks; no approved vaccine or treatment	WHO
Taï Forest virus	Taï Forest virus disease	Listed by CDC as causing illness in people; rare	CDC HAN, Sept. 2025
Reston virus	None listed	Not among the four species CDC lists as causing human illness	CDC HAN, Sept. 2025
Bombali virus	None listed	Not among the four species CDC lists as causing human illness	CDC HAN, Sept. 2025

The related Marburg virus belongs to the same broad family and produces a similar hemorrhagic illness, but it is a separate virus and falls outside the six species above. For the underlying biology of how any virus gets from one cell to a million copies, see our guide to how viruses work from infection to replication.

The 2026 Bundibugyo outbreak: where things stand

As of mid-October 2026, the biggest Ebola-family event in the world is not caused by the Zaire species that dominates textbooks. It is Bundibugyo virus disease in the DRC. WHO received an alert on May 5, 2026 about an unknown illness with high mortality in Mongbwalu Health Zone, Ituri Province, including four health workers who died within four days. Laboratory confirmation came on May 15, the DRC declared its 17th Ebola outbreak since 1976, and on May 17 the WHO Director-General determined the event to be a Public Health Emergency of International Concern. The UK Health Security Agency's monitoring report, with data through October 14, 2026, gives the scale:

• DRC: 9,066 confirmed cases and 4,381 confirmed deaths as of October 11, across 65 health zones in seven provinces.
• Weekly change: 463 new cases and 233 new deaths since the previous week's report.
• Ituri: 6,695 cases and 3,099 deaths, with new cases decreasing there; North Kivu and Haut-Uele still have ongoing transmission.
• Uganda: 20 confirmed cases, including 2 deaths. WHO and Africa CDC declared Uganda's outbreak over on August 27, 2026, after a 42-day countdown.
• Kenya: 2 confirmed cases, including 1 death, the first imported from the DRC and reported on October 6, and the second a family member who developed symptoms on October 13 while in quarantine.

That report describes this as the largest recorded Bundibugyo outbreak and the second-largest Ebola disease outbreak to date, behind only the 2014 to 2016 epidemic in West Africa. Dividing confirmed deaths by confirmed cases gives roughly 48 percent, but that is a simple ratio of counts that are still moving. Many recent cases have not yet had an outcome, so the true fatality figure will not settle until the outbreak ends. Past Bundibugyo outbreaks, in Uganda in 2007 and the DRC in 2012, had fatality of roughly 30 to 50 percent, according to WHO.

The key gap: unlike Ebola virus disease caused by the Zaire species, WHO states there is no licensed vaccine or specific therapeutic for Bundibugyo virus. Early supportive care is the main tool, and WHO has advised clinical trials of candidate products.

Two features of this outbreak explain why it grew. Ituri has ongoing armed conflict, and WHO's first assessment noted that insecurity limited surveillance, rapid-response deployment, and sample transport. The first alert also showed that infection prevention in health facilities had failed: WHO described the four early health-worker deaths as evidence of "critical breaches" in infection control. WHO assesses the risk as very high in the DRC, high in neighboring countries, and low globally, and it advises against border closures or travel restrictions.

Where Ebola hides between outbreaks

Ebola viruses are zoonotic, meaning they live in animals and occasionally jump to people. WHO says fruit bats of the Pteropodidae family are thought to be the natural hosts. Spillover into people happens through contact with the blood, secretions, organs, or other fluids of infected animals, including fruit bats, chimpanzees, gorillas, monkeys, forest antelope, and porcupines.

A useful way to picture this is a pilot light. The reservoir burns quietly in its animal hosts for years, doing little visible harm to them, and a single spark of human contact is enough to ignite a larger fire. The pilot light itself is almost impossible to extinguish, so outbreak control focuses on stopping the fire from spreading once lit. Our explainer on how viruses jump from animals to humans covers why these spillover events are so hard to predict.

The disease was first recognized in 1976, when two simultaneous outbreaks occurred: Sudan virus in Nzara, in what is now South Sudan, and Ebola virus in Yambuku, in what is now the DRC. Since then, the DRC has recorded 17 outbreaks, more than any other country in the sources cited here, which makes it the place where response teams have the most experience.

How the virus gets into cells and what it does to the body

After exposure, the incubation period runs from 2 to 21 days, according to WHO and CDC. During that time the person feels well and is not contagious. Inside the body the virus multiplies first in immune cells and other tissues, and the immune system's early alarm is delayed or muted, which gives the virus a head start. That head start is part of why the illness can progress quickly once symptoms appear.

Think of the immune system as a building's fire alarm. In Ebola infection, the first detectors are partly disabled, so the alarm sounds late. When it finally does, it rings at maximum volume: an overshooting inflammatory response that damages the body's own tissues. Blood vessels become leaky, fluid shifts out of circulation, the clotting system goes awry, and organs struggle.

CDC describes the clinical course in two broad phases. Early, or "dry," symptoms are fever, aches, pains, and fatigue. Later, "wet" symptoms appear, including diarrhea, vomiting, and unexplained bleeding. The older label "hemorrhagic fever" can therefore mislead, since bleeding is one late sign among several and not the defining feature of every case.

In fatal cases the final stage is typically shock and multi-organ failure, not blood loss alone. Fluid loss from vomiting and diarrhea compounds the problem, which is why optimized supportive care improves survival whatever specific drug is used, a point WHO makes in its fact sheet.

How Ebola spreads, and why it does not travel like flu

Ebola spreads through contact between infected body fluids and another person's broken skin or mucous membranes (the lining of the eyes, nose, and mouth). CDC also lists infected animals and contaminated objects such as needles as routes. WHO adds that burial ceremonies involving direct contact with the body can drive transmission, and that health workers are often infected when infection control precautions are not strictly followed.

Two facts shape everything about outbreak control. First, a person is not contagious until symptoms begin and remains infectious while the virus is in their blood. Second, CDC states plainly that Ebola disease is not spread through airborne transmission. Compare that with influenza, which people can pass on before they feel ill; our guide to how seasonal flu works shows how different the transmission logic is.

A campfire is a better analogy than a wildfire. A wildfire jumps on the wind, but a campfire spreads only to what touches it or sits right beside it. Ebola's "fuel" is the people nursing the sick, preparing the dead for burial, or working in under-protected clinics, which is why outbreaks cluster in households, health facilities, and funerals.

Route	What the sources say
Caring for a sick person	Direct contact with blood or body fluids, especially without protective equipment (WHO)
Burial practices	Direct contact with the body of someone who died of the disease (WHO)
Health facilities	Infections follow lapses in infection control (WHO)
Animals	Contact with infected fruit bats, primates, antelope, or porcupines (WHO)
Survivors' semen	Transmission documented up to 15 months after recovery (WHO)
Air	Not a route of transmission (CDC)

How many people does one case infect? The WHO Ebola Response Team's analysis of the first nine months of the West African epidemic estimated a basic reproduction number (R0, the average number of new infections caused by one case in a susceptible population) of 1.71 for Guinea, 1.83 for Liberia, and 2.02 for Sierra Leone, based on early growth. Those are low numbers by respiratory-virus standards, but above the threshold of 1 at which an outbreak keeps growing without intervention.

Outbreak history in numbers

The table below comes from CDC's list of the largest Ebola outbreaks. Counts mix confirmed and probable cases, and agencies sometimes count them differently, so treat the figures as the official record rather than exact tallies.

Outbreak	Cases	Deaths (fatality)
2014 to 2016, West Africa	28,610	11,308 (39%)
2018 to 2020, DRC (10th outbreak)	3,470	2,287 (66%)
2000, Uganda (Sudan virus)	425	224 (53%)
1976, DRC (first recognition)	318	280 (88%)
1995, DRC (Kikwit)	315	254 (81%)
1976, Sudan	284	151 (53%)
2007, DRC (Kasai Occidental)	264	187 (71%)
2022, Uganda (Sudan virus)	164	55 (34%)
2003, Republic of the Congo	143	128 (89%)
2007, Uganda (Bundibugyo)	131	42 (32%)

Source: CDC, Ebola outbreak history. The 2018 to 2020 totals include four cases confirmed in Uganda in 2019.

Two more recent entries sit outside that table. The Bulape outbreak in Kasai Province, DRC, was declared on September 4, 2025; WHO later recorded 64 cases and 45 deaths (a fatality of 70.3 percent), and the UKHSA reports it ended in December 2025. The 2026 Bundibugyo outbreak described above now exceeds every outbreak in the table except West Africa.

Why does fatality swing from 32 percent to 89 percent? The sources above do not assign a single cause, but several plausible factors work together. The species differs, as the Bundibugyo and Sudan rows show lower figures than the Zaire rows. Access to supportive care differs. And the way cases are counted matters: if mild cases go undetected, the denominator shrinks and fatality looks higher, and if recent patients have not yet died or recovered, it can look lower.

The 2014 to 2016 West African epidemic and what it changed

CDC's accounting of the West African epidemic is 28,610 cases and 11,308 deaths, a fatality of 39 percent, with Guinea, Liberia, and Sierra Leone the core affected countries. Imported or linked cases also occurred in Italy, Mali, Nigeria, Senegal, Spain, the United Kingdom, and the United States. It remains the largest Ebola outbreak in history.

The 39 percent figure sits oddly beside the early number most people remember. A WHO Ebola Response Team analysis published in the New England Journal of Medicine in 2014, using data through September 14, 2014, reported 4,507 probable and confirmed cases and 2,296 deaths, and estimated case fatality at 70.8 percent among people whose outcomes were known. The two numbers answer different questions. One divides by everyone ever counted, including people who recovered, died, or were still sick; the other divides only by patients whose outcome was recorded, which in a fast-growing outbreak skews toward deaths because they are recorded sooner than recoveries.

The epidemic showed how a virus previously seen in smaller outbreaks could spread through several countries when it reached towns and cities with stretched health systems. Its most direct scientific legacy is the vaccine trial described next, which ran in Guinea, and the FDA cited a Guinea study from the epidemic in approving the first Ebola vaccine.

Ring vaccination and the vaccines that exist

The key trial was called Ebola ça Suffit ("Ebola, that's enough"), a ring vaccination study in Guinea and Sierra Leone published in The Lancet in 2017. Ring vaccination works like a firebreak: once a case is confirmed, health teams list that person's contacts and the contacts of those contacts, and vaccinate that ring so the virus has nobody susceptible left to reach.

Clusters of contacts were randomly assigned to immediate vaccination or vaccination delayed by 21 days. Among the 4,539 contacts in the 51 immediate-vaccination clusters there were no cases with onset at least 10 days after randomization, versus 16 cases among the 4,557 contacts in the 47 delayed clusters. The vaccine's estimated efficacy was 100 percent, with a 95 percent confidence interval of 68.9 to 100 percent. The interval is wide because there were few cases, and the lower bound is the honest summary of what the trial could rule out.

Product	Type	Covers	Regulatory status
Ervebo	Vaccine	Zaire ebolavirus	FDA approved Dec. 19, 2019 for adults 18 and older; WHO's advisory group recommended it for outbreak response in July 2024
Zabdeno and Mvabea	Two-vaccine regimen (Janssen)	Ebola virus disease	Approved, per WHO; dates not stated on the WHO fact sheet
Candidates for Sudan virus	Various	Sudan virus	None approved; WHO describes several in development
Candidates for Bundibugyo virus	Various	Bundibugyo virus	None licensed (WHO, May 2026)

The FDA's guidance on Ebola countermeasures is blunt about the limits: products approved for Zaire ebolavirus "cannot be directly applied" to Sudan ebolavirus outbreaks or outbreaks caused by other orthoebolaviruses, because different species may not respond the same way. That is the practical reason the 2026 Bundibugyo outbreak has relied on classic outbreak control rather than vaccination. To understand why vaccines train the body the way they do, see how vaccines actually work.

Treatments: what the PALM trial showed

Until recently, treatment meant supportive care and hope. The turning point was the PALM trial, run in the DRC during the 2018 to 2020 outbreak and published in the New England Journal of Medicine in 2019. It randomized patients to four treatments: the antibody cocktail ZMapp, the antiviral remdesivir, the single antibody mAb114, and the three-antibody cocktail REGN-EB3. Of 681 patients enrolled, 673 received a study treatment, and an independent safety board stopped the trial in August 2019 after two of the treatments crossed the prespecified efficacy threshold.

Treatment	28-day mortality
ZMapp (comparison arm)	about 50%
Remdesivir	about 53%
mAb114 (ansuvimab)	35.1% (61 of 174)
REGN-EB3 (Inmazeb)	33.5% (52 of 155)

Source: review summarizing the PALM trial (Frontiers in Immunology). Percentages for ZMapp and remdesivir are rounded in that source.

The FDA approved Inmazeb (REGN-EB3) on October 14, 2020 and Ebanga (ansuvimab-zykl, the mAb114 antibody) on December 21, 2020, both for infection with Zaire ebolavirus. WHO strongly recommends either antibody for Ebola virus disease. Put the numbers in perspective, though: about one in three treated patients still died within 28 days. These antibodies are a major advance and not a cure, and for Sudan virus and Bundibugyo virus no specific treatment is approved at all.

How outbreaks are actually stopped

Ebola is controllable because it needs close contact, does not spread through the air, and has an incubation period that opens a window for intervention. The ingredients of a successful response are well established, and when any one is missing, outbreaks can slip away.

1. Rapid diagnosis. Laboratory testing, typically polymerase chain reaction (PCR, a method that detects viral genetic material), confirms cases. In the 2026 outbreak, an initial Ebola test on 20 samples from Rwampara came back negative for Ebola virus, and the cause was identified as Bundibugyo virus only after the national reference laboratory in Kinshasa analyzed further samples.
2. Isolation. Patients are separated so they cannot infect others. For suspected cases in the United States, CDC advises keeping a patient isolated until a negative test on a specimen collected at least 72 hours after symptom onset.
3. Contact tracing. Everyone exposed is listed and monitored for 21 days, the outer edge of the incubation period.
4. Ring vaccination. Where a vaccine exists, contacts and contacts of contacts are vaccinated.
5. Safe and dignified burials. Trained teams handle bodies while respecting local customs.
6. Community trust. Control depends on cooperation. Communities that distrust outside responders hide cases and avoid clinics.
7. Health-worker protection. Protective equipment, training, and screening at clinic entrances.

The 2026 outbreak shows what happens when these links weaken. WHO's first report counted 65 listed contacts in the DRC, 15 of them high-risk, and said follow-up was weak because of insecurity. Four health workers died within four days at one hospital. The Bulape outbreak in 2025 also included four health workers among its first 37 confirmed cases, according to CDC.

For a sense of how Ebola compares with other dangerous pathogens, our overview of the deadliest viruses on the planet sets contact-spread viruses beside those with very different transmission routes.

What survivors live with

Surviving Ebola is not always the end of the illness. WHO notes that the virus can persist in immune-privileged sites, meaning places the immune system reaches poorly, including the testicles, the eyes, and the brain. Relapse without a new infection is rare but documented, and the virus may persist in the placenta, amniotic fluid, fetus, and breast milk.

Semen is the best-studied example. WHO reports transmission through infected semen up to 15 months after clinical recovery. Where testing programs exist, male survivors are offered monthly semen tests until two consecutive negatives; without a program, WHO advises safer sex practices for 12 months. WHO does not recommend isolating survivors whose blood has tested negative, because they do not pose a risk through everyday contact.

WHO lists clinical and psychological effects among the consequences of surviving. Because survivors who test negative in blood are not a risk through everyday contact, explaining that distinction clearly to communities is part of the medical response, not an extra.

Five things people get wrong about Ebola

"It is airborne." CDC states that Ebola disease is not spread through airborne transmission. The risk concentrates among caregivers, health workers, and funeral attendees.

"Everyone who gets it dies." WHO's range of 25 to 90 percent across past outbreaks shows that survival is common, and it is more common with good supportive care and, for Zaire ebolavirus, modern antibodies.

"There is an Ebola vaccine, so Ebola is solved." The licensed products cover the Zaire species. For Sudan virus and Bundibugyo virus, including the 2026 outbreak, there are no approved vaccines.

"You can catch it from someone who looks healthy." Per WHO and CDC, people do not transmit before symptoms begin. This is why monitoring contacts for 21 days works.

"The fatality rate is a property of the virus." It is a property of the virus, the patient, the health system, and the arithmetic used to count. The West Africa figures of 70.8 percent and 39 percent are both defensible and describe different things.

The takeaway

Ebola is frightening because the disease is severe, but it is also unusually legible. It travels by touch, it announces itself with symptoms before it can spread, and every major outbreak has been brought down by the same unglamorous work of testing, isolating, tracing, and earning trust. What has changed since 2014 is that for one species, Zaire, medicine now has a vaccine and two antibodies that tilt the odds. The 2026 Bundibugyo outbreak is a reminder that the species matter: where the tools do not exist, outcomes depend on how quickly and fully the basics are carried out, in places where conflict and fragile health systems make that hardest.

Frequently asked questions

Q: Can Ebola become airborne? A: There is no evidence that any orthoebolavirus that infects people spreads through the air. CDC states that Ebola disease is not spread through airborne transmission. Documented spread requires contact with body fluids of a sick or deceased person, or with contaminated objects.

Q: Can someone with Ebola infect others before symptoms start? A: No. WHO and CDC both state that a person is not contagious until symptoms appear. The incubation period of 2 to 21 days is why exposed contacts are monitored for 21 days instead of being treated as infectious from the start.

Q: How is Ebola diagnosed? A: Laboratory testing of a blood sample, usually by PCR, confirms infection, and the assay must match the species. In the 2026 outbreak, initial tests for Ebola virus came back negative because the cause was Bundibugyo virus, so confirmation required a different test.

Q: Why does fatality vary so much between outbreaks? A: Fatality has ranged from 25 to 90 percent in past outbreaks, per WHO. Species, access to supportive care and antibody treatment, patient age, and how cases are counted all affect the figure, and confirmed-case ratios from ongoing outbreaks shift as outcomes are recorded.

Q: Is there a vaccine or treatment for Bundibugyo virus? A: Not as of WHO's May 2026 notice. WHO states that, unlike Ebola virus disease from the Zaire species, there is no licensed vaccine or specific therapeutic for Bundibugyo virus, and it advises clinical trials of candidates. Optimized supportive care is the mainstay.

Q: How long can survivors pass on the virus? A: WHO reports transmission through infected semen up to 15 months after recovery. It advises monthly semen testing where programs exist, or safer sex practices for 12 months otherwise, and it does not recommend isolating survivors whose blood has tested negative.

Q: Why hasn't Ebola caused a global pandemic? A: It depends on close contact and does not spread through the air, and people are not infectious until symptomatic, so cases can be recognized and isolated. CDC's history lists imported or linked cases in several countries outside West Africa during 2014 to 2016, including Spain, the United Kingdom, and the United States, and none grew into an outbreak on the scale of Guinea, Liberia, or Sierra Leone.

Sources

• WHO. Ebola disease fact sheet, updated April 24, 2025.
• WHO. Disease Outbreak News: Ebola disease caused by Bundibugyo virus, DRC and Uganda, May 16, 2026 (revised May 26, 2026).
• UK Health Security Agency. Outbreaks under monitoring, week 41, week ending October 11, 2026.
• CDC. Health Alert Network notice CDCHAN-00524, September 18, 2025.
• CDC. Ebola outbreak history.
• FDA. Ebola medical countermeasures.
• Henao-Restrepo AM, et al. Efficacy and effectiveness of an rVSV-vectored vaccine in preventing Ebola virus disease: final results from the Guinea ring vaccination, open-label, cluster-randomised trial. The Lancet, 2017. Abstract summary at Epicentre.
• Mulangu S, et al. A randomized, controlled trial of Ebola virus disease therapeutics. New England Journal of Medicine, 2019. Figures as summarized in Frontiers in Immunology review (PMC8435780).
• WHO Ebola Response Team. Ebola virus disease in West Africa: the first 9 months of the epidemic and forward projections. New England Journal of Medicine, 2014. PAHO record.

For related coverage, see our H5N1 bird flu guide and the piece on how AI is used to predict viral evolution.

---

Read next:

• Mpox (Monkeypox) Explained — How the Virus Works in 2026
• Dengue Virus Explained — Why It's Spreading and How It Works
• HIV Explained — How the Virus Works and Why It's Treatable Now