The Ebola Numbers Don't Add Up: Inside the Growing Gap Between Official Counts and the Likely Scale of the DRC Outbreak

On May 29, 2026, the Democratic Republic of Congo's Ministry of Health reported 1,262 suspected and confirmed Ebola cases and at least 241 deaths [1]. The International Rescue Committee says those numbers represent a fraction of the true toll. With four out of five contacts going untraced, diagnostic cartridge shortages delaying confirmations, and the virus having circulated for months before detection, the actual caseload may be several times higher than what surveillance systems are capturing [2].

The question is not whether cases are being missed. The question is how many — and what that means for a response already stretched to breaking.

A Virus With a Head Start

The first known case — a healthcare worker in Bunia, the capital of Ituri Province — developed symptoms on April 24, 2026, and died at a local medical center [3]. But the outbreak was not officially declared until May 15, and the virus was not confirmed as Bundibugyo ebolavirus until that same day [4]. The IRC estimates the virus had been spreading undetected since before March, giving it roughly three months of silent transmission before any formal response began [2].

By the time the WHO declared the outbreak a public health emergency of international concern (PHEIC) on May 17 — just two days after confirmation — there were already 8 confirmed cases, 246 suspected cases, and 80 suspected deaths across at least three health zones in Ituri Province [3]. That speed alarmed epidemiologists. Ruth McCabe, a modeler at Imperial College London, estimated based on case fatality rates from prior Bundibugyo outbreaks (historically 30–50%) and current death counts that actual cases could already have exceeded 1,000 by mid-May [5].

The case trajectory since then has confirmed those fears. By May 24, confirmed cases had risen to 101 with 904 suspected, and by May 29 the confirmed count reached 282 with 980 suspected [1][6].

Source: WHO / DRC Ministry of Health

Data as of May 29, 2026CSV

The Anatomy of Undercounting

The IRC's central claim rests on a specific metric: only 20% of known contacts are currently being traced [2]. In Ebola response, contact tracing is the primary tool for identifying new cases before they become new transmission chains. When 80% of contacts are lost to follow-up, the surveillance system is, by definition, missing the majority of new infections.

Rachel Howard of the IRC stated: "When four out of five contacts are not being traced, it becomes incredibly difficult to contain the outbreak or even understand its true scale" [2].

Several structural factors compound the tracing gap:

Diagnostic bottlenecks. Shortages of diagnostic cartridges and testing backlogs are slowing case confirmation [2]. Most rapid diagnostic tests deployed in the field were designed to detect the more common Zaire ebolavirus, not the Bundibugyo species [7]. This means even when suspected cases reach healthcare facilities, confirmation is delayed or impossible.

Patient flight. At least seven confirmed Ebola patients have reportedly left treatment centers, creating unmonitored chains of transmission in the community [2].

Geographic inaccessibility. The outbreak is centered in Ituri Province and has spread across 11 health zones, with cases also reported in North Kivu (including Butembo and Goma) and South Kivu [8]. Poor roads and active combat restrict movement of surveillance teams and medical supplies.

Historical Precedent: How Bad Can Undercounting Get?

The 2014–2016 West Africa Ebola epidemic — the largest in history — demonstrated that early undercounting can mask exponential growth. A capture-recapture analysis in Montserrado County, Liberia, estimated the true number of cases was at least three-fold higher than reported during the study period [9]. The CDC's own modeling at the time incorporated correction factors to account for underreporting, acknowledging that official figures substantially understated the epidemic's scale [10].

The structural conditions driving undercounting in 2026 are, in several respects, worse than in 2014. Eastern DRC is an active conflict zone — a challenge West Africa did not face at comparable scale. There is no approved vaccine or therapeutic for Bundibugyo ebolavirus, whereas Zaire ebolavirus vaccines (notably rVSV-ZEBOV) became available during the later stages of the West Africa response and were central to the 2018–2020 DRC outbreak response [4][7]. And the international public health infrastructure has been weakened by US funding cuts and institutional losses since 2025 [11].

On the other hand, one factor cuts the other direction: the Bundibugyo strain has a historically lower case fatality rate (30–50%) compared to Zaire ebolavirus (which can exceed 60–90% in some outbreaks) [5][7]. If the true caseload is substantially higher than reported while deaths remain relatively contained, it could indicate either significant undercounting of mild or asymptomatic infections, or a lower-virulence strain — or both. This distinction matters for response strategy but remains unresolved given current diagnostic limitations.

A War Zone as an Epidemic Zone

The outbreak is unfolding in one of the most volatile regions on the African continent. Ituri Province and the neighboring Kivus are contested by multiple armed groups, including the Allied Democratic Forces (ADF), CODECO militias, and the Rwanda-backed M23 movement [8][12].

WHO Director-General Tedros Adhanom Ghebreyesus described the situation as a "catastrophic collision of disease and conflict," stating: "We cannot build community trust or isolate the sick while bombs are falling" [8].

Nearly 1 million people live in dense displacement camps in Ituri Province alone [12]. Over the preceding months, more than 100,000 civilians had been displaced by violence in the immediate outbreak zone [12]. These camps — overcrowded, lacking sanitation, and largely beyond the reach of formal health surveillance — represent ideal conditions for viral transmission and the worst conditions for case detection.

The broader humanitarian context amplifies the risk. According to the UN-backed Integrated Food Security Phase Classification (IPC), nearly 10 million people across Ituri, North Kivu, South Kivu, and Tanganyika provinces face acute hunger in the first half of 2026, with 26.5 million nationally experiencing high acute food insecurity [8]. As Tedros noted: "Hunger and disease are old companions. People weakened by hunger are far more vulnerable to infections" [8].

Healthcare Workers: Both Victims and Indicators

Healthcare worker infections serve as a sentinel indicator of how widely a virus is circulating and how effectively infection prevention measures are functioning. In this outbreak, the signal is concerning.

At least six healthcare workers have died, including two doctors in recent days [2]. At the Bunia Centre Medical Evangelique Nyakunde, all reported Ebola cases were frontline health professionals, with four dying within a four-day span [13]. In Uganda, two of the country's seven confirmed cases were healthcare workers [8].

MSF (Doctors Without Borders) has identified healthcare-associated transmission as a major driver of the outbreak and is training health workers in infection prevention while constructing a 65-bed Ebola treatment center in Mongbwalu, considered the current epicenter [13]. But healthcare worker deaths do more than reflect transmission — they accelerate it. When staff flee or facilities close in response to infections, communities lose their only point of contact with the formal health system, and cases that would otherwise be detected vanish from surveillance entirely.

Cross-Border Spread and Regional Risk

The outbreak crossed into Uganda by mid-May, with cases confirmed in Kampala — a city of over 3 million people and a major regional hub with international air connections [4][6]. As of late May, Uganda had reported 7 confirmed cases and 1 death, all epidemiologically linked to the DRC [6]. Uganda subsequently closed its high-traffic border with the DRC [14].

The cross-border dynamics are shaped by long-established trade routes and displacement corridors between eastern DRC and Uganda, Rwanda, Burundi, and South Sudan. Ituri Province borders both Uganda and South Sudan, and population movement across these borders is continuous and largely unmonitored. South Sudan's surveillance infrastructure is among the weakest in the region, raising particular concern about undetected spillover [12].

The DRC itself also faces internal spread. Cases have been detected in North Kivu — including the cities of Butembo and Goma — and South Kivu, representing expansion well beyond the initial Ituri focus [8]. Goma, a city of over 2 million near the Rwandan border, was a major site of transmission during the 2018–2020 outbreak and would represent a significant escalation if sustained local transmission is established.

The Funding Gap

The financial response to the 2026 outbreak has been markedly slower and smaller than in previous Ebola emergencies.

Source: WHO / KFF / CNN

Data as of May 31, 2026CSV

WHO approved $3.9 million from its Contingency Fund for Emergencies [3]. The US State Department mobilized $23 million in emergency funding within two days of learning of the outbreak [15]. But the WHO's Contingency Fund had received only $5.4 million in total donor contributions for all of 2026, and was near exhaustion [15].

The broader context is one of systematic disinvestment. US humanitarian assistance to the DRC collapsed 96% from fiscal year 2024 to a partially reported $35 million in fiscal year 2026, while health sector funding, despite a partial recovery, remained 50% below its FY2024 baseline [11]. The dismantling of USAID eliminated institutional knowledge and field presence. Dennis Carroll, former director of USAID's Emerging Pandemic Threats program, told NPR: "The people that had those years of experience, they've largely been fired" [16].

The US withdrawal from the WHO further complicated coordination. Other major donors — including the United Kingdom, Germany, and Canada — also cut foreign aid budgets in 2025 [11]. No single country or institution has filled the resulting gaps.

Evaluating the IRC's Claims

The IRC is simultaneously a humanitarian responder and an advocacy organization. Its institutional incentives — fundraising, political leverage, media attention — align with emphasizing crisis severity. This does not mean its claims are wrong, but it means they warrant independent verification.

On the core undercount claim, the evidence from other sources is broadly corroborative. The WHO's own data shows a ratio of roughly 3.5 suspected cases for every confirmed case as of late May, reflecting acknowledged diagnostic limitations [1]. Imperial College London's independent modeling reached similar conclusions about likely case numbers exceeding official counts [5]. MSF, which operates independently of the IRC and has its own field presence, has described the outbreak as "rapidly evolving" and has committed to a large-scale response consistent with an outbreak exceeding official figures [13].

The 20% contact tracing rate cited by the IRC has not been publicly disputed by the WHO or DRC health authorities, and is consistent with the operational constraints documented by multiple organizations working in eastern DRC [2][8].

Where greater skepticism is warranted is on the IRC's framing of the outbreak as potentially "the deadliest on record" [17]. The 2014–2016 West Africa epidemic killed more than 11,000 people across three countries. The 2026 outbreak, while serious, has not yet reached that scale, and the Bundibugyo strain's lower historical case fatality rate may limit absolute mortality even if case counts grow substantially. The IRC's use of this framing appears designed to mobilize urgency rather than reflect a current quantitative assessment.

Previous outbreak projections by humanitarian organizations have had a mixed record. During the 2018–2020 DRC outbreak, early worst-case scenarios projected tens of thousands of cases; the final count was approximately 3,481 cases and 2,299 deaths [4]. Projections are inherently uncertain, and organizations tend to model worst-case scenarios to justify resource mobilization — a strategy that is defensible from a public health standpoint but can overstate probable outcomes.

What the Numbers Mean

If the IRC and independent modelers are correct that the true caseload is significantly higher than the confirmed count of 282, the implications cascade.

A higher true caseload with roughly similar death counts would push the apparent case fatality rate downward. The confirmed CFR currently stands above 30%, but if thousands of milder or undetected cases exist, the effective CFR could be substantially lower. This would be consistent with Bundibugyo virus's historical profile — less lethal than Zaire ebolavirus but still a serious pathogen [5][7].

A lower CFR might seem like good news, but it carries its own risks. Less severe cases are harder to detect and more likely to circulate in communities, making containment more difficult. Patients who feel well enough to travel spread the virus along trade routes and displacement corridors, which is likely how the virus reached Kampala [14].

The most consequential implication is for the response timeline. Every major Ebola outbreak has demonstrated that early, aggressive intervention is far more effective and less costly than delayed response. The 2014 West Africa epidemic was largely a story of missed early warnings and delayed mobilization. The current combination of a three-month head start for the virus, a funding environment that is 50–75% below historical baselines, the absence of a strain-specific vaccine, and active armed conflict in the outbreak zone creates conditions where the gap between official counts and reality is not just an academic question — it determines whether the response is calibrated to the actual threat.