ob体育

Update, as of March 10th

Background

Ugandan authorities declared an outbreak of Sudan virus disease, a type of Ebola disease caused by the Sudan virus species Orthoebolavirus sudanense (SUDV), on January 29, 2025. Since its discovery in 1976, this virus has caused several outbreaks in Uganda and near the border between South Sudan and the Democratic Republic of the Congo. As of Feb. 27, 2025, there have been nine confirmed cases and one death in and around Kampala, the capital city of Uganda. The outbreak includes the presumed index case, family members of the index case, and healthcare workers who treated the index case.

To inform preparedness and decision-making, we outline three plausible scenarios for the outbreak in Uganda over the next three months, including implications for the United States. The Center for Forecasting and Outbreak Analytics (CFA), in collaboration with the ob体育 East Africa Viral Hemorrhagic Fever Outbreak Response team, developed these scenarios by integrating insights from previous outbreaks of Ebola disease and modeling work. We use a simple model to examine how the level of active case detection efforts and potential for transmission can impact the progression of the outbreak in Uganda.

In a recent rapid risk assessment, ob体育 assessed the current risk posed by the outbreak of Sudan virus disease in Uganda to the U.S. general population as low, with moderate confidence.

Potential Outbreak Scenarios

We outline three plausible scenarios for how the outbreak in Uganda could progress over the next three months:

• Optimistic: A small outbreak of Sudan virus disease (<25 total cases) with no ongoing transmission in three months. In this scenario, Uganda could experience limited spread among healthcare and household contacts, similar to the 2012 outbreak of Sudan virus disease in Uganda, which totaled 6 reported cases.
• Pessimistic: A medium-sized outbreak of Sudan virus disease (25–300 total cases). There may be ongoing transmission in three months, with indications that the outbreak is declining. In this scenario, Uganda could experience expanded transmission among healthcare workers and their household contacts, similar to the 2022 outbreak of Sudan virus disease in Uganda, which totaled 164 reported cases. In this scenario, there may be spread to neighboring countries.
  • Additional resources may be needed for rapid detection of Sudan virus disease, isolation of cases, contact tracing, treatments, and vaccinations Aas the number of cases in the outbreak increase.
  • The risk of importation to the United States may increase if the outbreak in Uganda grows or if there is spread to other neighboring countries, due to increases in returning healthcare workers or travelers from the region. Historically, the majority of Ebola cases in the United States during outbreaks in other countries have been among returning healthcare workers and travelers.
• Worst-case: A large outbreak of Sudan virus disease (more than 300 cases) with evidence of exponential growth in three months. In this scenario, Uganda would likely experience sustained community transmission, as was observed in the 2014 outbreak of Ebola virus disease in Guinea, Liberia, and Sierra Leone (28,610 total reported cases). Spread to neighboring countries is most likely in this scenario.
  • In this scenario, substantial additional resources may be needed for rapid detection and control of ongoing spread, including additional medical staff and facilities.
  • Of the three potential scenarios, importation to the United States is most likely to occur in this scenario. If a large outbreak occurs in Uganda, the public health impact for the U.S. general population may be significantly higher. Additionally, there may be increased pressure for U.S. assistance and increased fear in the United States as the outbreak in Uganda grows.

We continue to examine the latest outbreak data and evidence, combined with modeling approaches, to determine which scenario is most likely to occur in the next three months. Below, we identify a set of indicators that we use to assess which scenario is most likely occurring. Indicators are not all-inclusive. Not all indicators need to be met for the assessment of a given scenario, and we rely on expert judgment to determine which scenario is most likely occurring.

Assessment of Current Scenario

As of Feb. 27, 2025, most indicators suggest that we are currently in the optimistic scenario:

• There are no new reported cases since February 7, 2025, and 100% of confirmed cases were among the contacts of the index case.
• Of the traced contacts, the majority have completed quarantine and been discharged.
• Uganda has experience containing Ebola outbreaks and is demonstrating a high capacity to respond; the country has activated an emergency operations center, and point of entry and exit screening.
• There are no reported cases in other regions of Uganda or neighboring countries.
• A randomized ring vaccination effort began on February 3, as part of an ongoing vaccine trial for a SUDV vaccine. We do not yet know if this vaccine will be effective at stemming transmission but, if effective, this vaccine could contribute to slowing growth of the outbreak.

While we assess that we are currently in the optimistic scenario—which indicates that no remaining transmission is likely in three months—we note that it remains possible that a much larger outbreak could occur. In 2014, the Ebola virus in West Africa started as a small outbreak that later grew very large.

Modeling Insights

Modeling can be used to assess how public health response activities may affect the progression of the Sudan virus disease outbreak in Uganda. CFA developed a mathematical branching process model to estimate the cumulative number of cases over time and to examine the role of transmission potential, active case detection, and the percent of contacts traced and quarantined. The transmission potential is defined as the reproductive number (R₀), or the number of new infections caused by each infection. Active case detection efforts include both contact tracing and subsequent quarantine of confirmed contacts of cases and are a tool that can reduce the need for non-targeted interventions, such as restrictions on gatherings. This model is stochastic, with built-in random variation, so we run many individual simulations to produce a range of possible outcomes and estimates of associated uncertainty.

Model results show that increasing the probability of successful active case detection—or percent of all potentially infected contacts currently tracked and quarantined—reduces the probability of ongoing transmission after 3 months, for both high and low levels of transmission potential (Figure 1).

View Larger

Show More

Additional modeling assesses four combinations of low and high levels of transmission potential and active case detection (Figure 2). Low and high active case detection were defined by the time for active detection and the probability of successful active case detectionB. Modeling results indicate that for low transmission potential, the number of total cases stays low, and the outbreak most likely follows the optimistic scenario defined above at both high and low levels of active case detection. Speeding up and expanding active case detection can increase the likelihood of achieving an optimistic scenario, with no new transmission within three months (Table 1, Figure 2). If transmission potential is high, improved active case detection decreases the likelihood of a worst-case scenario (Figure 2). However, for high transmission potential, even high levels of active case detection may be insufficient in completely stopping transmission within the next three months (Figure 1). These results indicate that when the potential for transmission is high, it is possible that other public health measures—beyond contact tracing—may be necessary to stem the outbreak.

View Larger

Show More

We note that our current modeling approach makes several assumptions, including the following:

• The model assumes one index case. If there is more than one transmission chain occurring in this outbreak, modeling results may under-estimate the total expected cases.
• Over the course of the outbreak, the levels of active and passive detection do not change. In a real-world scenario, rates of active and passive detection may shift as more resources are dedicated to outbreak response.
• The model does not account for all public health interventions that may occur over the course of an outbreak.
• The model assumes a Poisson distribution for the number of people each infected person infects. This assumption is conservative in that more transmission chains continue than using other distributions, resulting in an under-estimate of the probability of containment; in simulations that fall into the worst-case scenario, this assumption could underestimate the total number of cases compared to other distributions.

Key Uncertainties

This scenario assessment is intended to provide a high-level overview of what could occur in the next three months, following the outbreak of Sudan virus in Uganda. We note several areas of uncertainty:

• Current size and geographic scope of the outbreak, and completeness and timeliness of existing surveillance data.
• Timeliness and effectiveness of active case detection efforts in ongoing outbreak.
• Impact of the location of the outbreak in the urban setting of Kampala, Uganda. Most previous outbreaks of Sudan virus in Uganda have occurred in more rural areas.
• Effectiveness of the SUDV vaccine currently in trial.

Appendix - Additional Modeling Methods

The simulation begins with a defined number of infected people who each then infect a randomly generated number of secondary cases. This number is drawn from a Poisson distribution with mean equal to the reproductive number (R₀). In real-world terms, this means that all infected persons are assumed to have the same potential to infect others (i.e., there's less potential for "superspreading" than likely exists in reality). In the model, infected persons can be detected one of two ways: 1) "passive" detection, which means that the infection is detected without regard to whether the source of the infection (i.e., the infector) was also detected; or 2) "active" detection, which is the detection of secondary cases through contact tracing. An infection can be actively detected only if its infector was also detected (either actively or passively). This means that there can be no active detections until at least one infection in a transmission chain is passively detected. The model is parameterized in terms of probabilities of "effective" active and passive detection, which means the probability that an infection is not only detected but also successfully isolated and therefore prevented from causing any onward transmission. For the purposes of this analysis, we kept passive detection parameters constant.