The Ebola Numbers That Vanished: How Testing Exposed the True Scale of Congo's Outbreak

On May 15, 2026, when the Democratic Republic of the Congo officially declared an Ebola outbreak caused by the rare Bundibugyo strain, the numbers looked catastrophic: 336 suspected cases and 87 suspected deaths [1]. Two weeks later, suspected cases had ballooned to 906 with 223 suspected deaths [2]. Then, between late May and early June, something striking happened. The suspected case count collapsed — from 906 to 116 — even as confirmed cases climbed sharply from 125 to 321 [3][4].

The explanation was not a miracle. It was testing.

The Arithmetic of Diagnosis

As laboratory capacity expanded in the outbreak's epicenter in Ituri Province, hundreds of people previously classified as suspected Ebola patients were tested and cleared. WHO announced on June 2 that hundreds of suspected cases "have been cleared out" through expanded diagnostic screening [5]. By that date, 774 samples had been collected and 648 analyzed — an 84% processing rate — with 125 testing positive, yielding a test positivity rate of 19.2% [2].

Source: WHO Disease Outbreak News

Data as of Jun 2, 2026CSV

That means roughly four out of five people flagged as suspected Ebola cases did not have Ebola. Many likely had malaria, typhoid, influenza-like illness, or other febrile conditions common in eastern Congo. The low clinical index of suspicion among healthcare providers, compounded by co-circulating arboviruses, made distinguishing Ebola from endemic diseases nearly impossible without laboratory confirmation [2].

Source: WHO/Africa CDC

Data as of Jun 2, 2026CSV

The confirmed case trajectory tells the opposite story from the suspected case count. While suspected cases fell by 87% between May 27 and June 1, confirmed cases grew by 157% over the same window — from 125 to 321 in DRC alone, with an additional 15 cases in Uganda [3][4]. This divergence reveals that the initial suspected case totals were not measuring Ebola. They were measuring fear, diagnostic limitations, and the breadth of febrile illness in a region where baseline healthcare infrastructure is minimal.

A Ratio Without Precedent

The gap between suspected and confirmed cases in this outbreak is historically unusual. At its peak, the 2026 Bundibugyo outbreak had a suspected-to-confirmed ratio of approximately 6.7:1 — meaning nearly seven people were flagged for every one who actually had Ebola [2][6].

Source: WHO/CDC

Data as of Jun 2, 2026CSV

Compare that to the early months of the 2014–2016 West Africa epidemic, where the ratio was roughly 1.45:1 by June 2014 (364 confirmed out of 528 total reported cases) [6]. The 2018–2020 DRC outbreak in North Kivu and Ituri, which dealt with Zaire ebolavirus in an active conflict zone, had a ratio closer to 1.8:1 [7].

The difference is partly explained by the pathogen itself. The Bundibugyo strain is rare — only two previous outbreaks, in 2007 and 2012 — and most clinicians in eastern Congo have never seen it [2]. Standard rapid diagnostic tests are optimized for the more common Zaire ebolavirus. Initial Ebola Xpert testing on samples from the Rwampara health zone returned negative; samples had to be sent to Kinshasa's Institut National de la Recherche Biomédicale (INRB) for full PCR and genomic sequencing to confirm Orthoebolavirus bundibugyoense [8]. The Bunia laboratory, the closest facility capable of returning PCR results, has a 24-hour turnaround time but limited throughput [8].

The Testing Infrastructure

The DRC's diagnostic capacity for this outbreak rests on a fragile foundation. The largest biosafety-level laboratory in eastern Congo, established through the World Bank's Regional Disease Surveillance Systems Enhancement (REDISSE) project, has become the central testing hub [9]. The World Bank has financed critical equipment to keep it operational, and on May 27, the Pandemic Fund Governing Board approved up to $220.6 million in grant financing for the response [9].

But the underlying constraints are structural. Laboratory equipment — flow cytometers, sequencers, PCR machines — is costly and difficult to maintain. Technicians often must be brought in from outside the country, and equipment breakdowns cause cascading delays [8]. As new suspected cases were reported daily throughout May, hundreds of samples accumulated untested [10].

MSF deployed approximately 50 internationally mobile staff and 480 locally hired workers in Ituri Province [10]. WHO, the Africa CDC, and the International Federation of Red Cross and Red Crescent Societies have all mobilized response teams [11]. International donors pledged $500 million, with the United States announcing $80 million in additional aid for a total commitment exceeding $112 million [12][13].

The USAID-Shaped Hole

The funding landscape has shifted dramatically since the last major DRC Ebola response. USAID, which spent over $516 million on the 2018–2020 Kivu/Ituri outbreak, was effectively dismantled by funding cuts in 2025. The agency officially closed on July 1, 2025, with 80% of global health awards terminated and $12.7 billion pulled from global health funding [14].

The consequences are measurable. No Disaster Assistance Response Team (DART) — the rapid-deployment unit that has been central to US outbreak responses since 2018 — was deployed for the 2026 outbreak [14]. Experienced epidemiologists and logistics specialists who previously staffed these responses have been "largely been fired" and "eliminated from those positions," according to NPR reporting [15]. Dr. Jean Kaseya of the Africa CDC stated on the second day of the declared outbreak: "We don't have manufacturing for PPE" [14].

"The U.S. has historically played a much larger role — a leadership role — in these actions than any other country," one public health expert told NPR. "No one has stepped in to fill the gap with the departure of the U.S. from WHO and the elimination of foreign assistance programs" [15].

Who Carries the Burden

The outbreak is concentrated in Ituri Province, which accounts for 88% of confirmed DRC cases — 299 of 321 confirmed cases as of June 2 spread across 15 health zones [3]. The health zones of Bunia (37 cases), Rwampara (33 cases), and Mongbwalu (20 cases) bear the heaviest load [2]. Mongbwalu, a gold-mining town of 130,000 people, sits at the outbreak's epicenter [12]. Bunia, the provincial capital with over one million residents, has also reported hundreds of suspected (though largely unconfirmed) cases [12].

Healthcare workers have been disproportionately affected. Twenty percent of new positive cases are among healthcare workers, with 16 confirmed healthcare worker infections in DRC as of late May [2][5]. The case fatality rate among confirmed cases stands at approximately 14% — significantly lower than the 25–40% observed in previous Bundibugyo outbreaks in 2007 and 2012, and well below the 70.3% fatality rate of the 2025 Kasai Province Zaire ebolavirus outbreak [2][7][16].

Contact tracing, the backbone of Ebola containment, remains inadequate. Only approximately 25% of identified Ebola patient contacts are being reached by tracking workers [5]. The Africa CDC reported that contact tracing coverage improved from 7% to 45% as the response scaled, but these figures vary by source and health zone [4]. In Ituri and North Kivu, 2,635 contacts had been identified; in Uganda, 436 [2].

No Vaccine for This Virus

The most consequential difference between this outbreak and recent Ebola emergencies is pharmaceutical. The Ervebo vaccine, which proved effective through ring vaccination strategies during the 2018–2020 DRC outbreak and helped contain subsequent flare-ups in 2025, targets the Zaire ebolavirus [17]. The Bundibugyo strain is a different species entirely. There is no approved vaccine or therapeutic for Bundibugyo ebolavirus [4][17].

This eliminates the single most effective containment tool deployed in recent Ebola responses. Ring vaccination — immunizing contacts of confirmed cases and contacts of those contacts — was credited with containing outbreaks in South Kivu (September 2025, contained in 67 days) and North Kivu (December 2025) [17][18]. Without it, the 2026 response depends entirely on traditional public health measures: isolation, contact tracing, safe burial practices, and community engagement.

Community Distrust and the Cost of False Designation

In Mongbwalu, Dr. Richard Lokudi, a hospital director, described the challenge: "People think it's a mystic illness. Or they don't trust the hospital" [12]. Community members have alleged that "hospital personnel are injecting people with the illness" [12].

The distrust has turned violent. Health facilities have been attacked multiple times. Angry crowds burned isolation tents. Soldiers fired warning shots to disperse mobs. People attempted to retrieve bodies from treatment centers due to delays in burial protocols [12]. At least three attacks were launched against health centers in the outbreak zone [12].

The reclassification of suspected cases compounds this dynamic. The WHO case definition for suspected Ebola is deliberately broad: any person with fever and at least three additional symptoms, or fever with contact history, or unexplained bleeding [6]. In a region where malaria, typhoid, and other febrile illnesses are endemic, this definition captures large numbers of people who do not have Ebola. Those individuals may be subjected to quarantine, isolation, movement restrictions, and social stigma before testing confirms they are negative.

Research from the 2007–2008 Bundibugyo outbreak in Western Uganda documented that social distancing and community stigma directed at suspected cases — many later cleared — was linked to lasting psychological distress [19]. Rwanda and Uganda closed their borders with Congo; Canada imposed a 90-day entry ban on Congolese residents [12]. For individuals wrongly suspected, the personal and economic consequences can be severe, and formal compensation mechanisms are largely nonexistent.

Nearly one million people live in displacement camps in Ituri Province, where the humanitarian crisis — seven million displaced across eastern Congo — predates and exacerbates the outbreak [12][13]. Goma airport has been closed since January 2025, further complicating logistics [13].

Surveillance Improvement or Undercount?

The steelman case for surveillance improvement is straightforward: prior to testing scale-up, clinicians had no way to distinguish Ebola from dozens of other febrile illnesses. The high suspected-to-confirmed ratio reflects the broad case definition doing exactly what it was designed to do — casting a wide net. As testing resolves uncertainty, the numbers converge toward reality. The 19.2% test positivity rate, while low compared to other Ebola outbreaks, is consistent with an outbreak occurring against a background of high endemic disease burden [2].

The counterargument is equally grounded. Only 25% of identified contacts are being reached [5]. The outbreak may have been spreading undetected since January 2026, with a four-week gap between the presumed index case's symptom onset on April 25 and laboratory confirmation on May 15 [2]. Insecurity in eastern Congo — an active conflict zone — renders some communities inaccessible to response teams. Community deaths reported outside health facilities are difficult to investigate. Dr. Esther Sterk of MSF stated: "Every day there are many community deaths and suspected patients arriving" [12].

WHO Director-General Tedros Adhanom Ghebreyesus, who traveled to Kinshasa on May 28, acknowledged the tension: "We cannot build community trust or isolate the sick while bombs are falling" [13]. Dr. Abdou Sebushishe estimated "beyond six months before this outbreak could be put under control" [5].

The honest answer is that both things are true simultaneously. Testing is clarifying the picture — and the picture remains incomplete.

When Does It End?

WHO considers an Ebola outbreak ended after 42 consecutive days — two maximum incubation periods — have passed since the last possible exposure to a confirmed case [20]. This criterion has been criticized as insufficient. During the 2014–2016 West Africa epidemic, Liberia's outbreak was declared over three separate times before it was actually contained, as the virus persisted in survivors' body fluids in immunologically privileged sites [20][21].

Researchers have proposed alternatives: a preliminary end-of-outbreak declaration after 63 days from the last detected case's symptom onset, followed by 90 days of enhanced surveillance [21]. After the 42-day period, WHO recommends maintaining a combination of active and passive surveillance for at least six months [20].

For the Bundibugyo outbreak, the end remains distant. With confirmed cases still climbing, no vaccine available, contact tracing reaching only a fraction of identified contacts, and community resistance to public health measures, the 42-day clock has not started. The question of what surveillance infrastructure will remain after the outbreak is premature — but the pattern from previous DRC outbreaks suggests that the laboratories, contact tracing networks, and emergency operations centers built for response tend to deteriorate once international attention and funding move on.

The World Bank's Emergency Operations Center in Kinshasa, rehabilitated four years ago with international financing, includes attached warehouses with emergency supply stockpiles [9]. Whether that infrastructure endures depends less on epidemiology than on politics and budgets.

What the Numbers Actually Say

The drop from 906 to 116 suspected cases is a data quality event, not an epidemiological one. It means the surveillance system is getting better at distinguishing Ebola from everything else. The simultaneous rise in confirmed cases — from 12 to 321 in less than three weeks — means the outbreak itself is still growing.

Both numbers matter. The falling suspected count is evidence that testing works. The rising confirmed count is evidence that the outbreak is not under control. Reading one without the other produces a misleading picture — either false reassurance or unjustified alarm.

The 2026 Bundibugyo outbreak is testing the international public health system at a moment of diminished capacity: no applicable vaccine, a gutted US aid apparatus, active armed conflict in the outbreak zone, and deep community distrust of health authorities. The numbers are clearer than they were three weeks ago. The situation is not.