📊 epidemiology

Cost-effectiveness of geographically targeted versus ring vaccination campaigns in response to an outbreak of Ebola Virus Disease caused by Orthoebolavirus zairense

While ring vaccination generally outperforms geographically targeted vaccination in reducing cases, deaths, and costs when effective contact tracing is feasible, the latter serves as a more effective alternative in remote or conflict-affected settings where surveillance and tracing are severely constrained.

Original authors: Nedjati-Gilani, G., Cox, V. M., Pearson, C. A. B., Skarp, J. E., Johnson, R., Ebengo, D. M., Mbala-Kingebeni, P., Niyukuri, D., Morgenstern, C., Hinsley, W., Riley, S., Hunter, A., McCabe, R., Bagayok

Published 2026-07-23

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Nedjati-Gilani, G., Cox, V. M., Pearson, C. A. B., Skarp, J. E., Johnson, R., Ebengo, D. M., Mbala-Kingebeni, P., Niyukuri, D., Morgenstern, C., Hinsley, W., Riley, S., Hunter, A., McCabe, R., Bagayoko, A., Costa, A., Lambach, P., Kallay, R., Aberle-Grasse, E., Pallas, S., Muhoza, P., Ferguson, N. M., Whittles, L. K., Hauck, K. D., Cori, A.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine a game of "hot potato" where the potato is a dangerous virus, and the goal is to stop it from being passed around a crowded room. In the world of infectious diseases, scientists have long known that the best way to stop a virus from spreading is to find the person holding the potato, then quickly vaccinate everyone they've touched, and everyone those people touched. This is called "ring vaccination." It's like drawing a protective circle around the hot potato so it can't jump to anyone else. But what happens if the room is chaotic, the people are hiding, or the virus is moving too fast to track who touched whom? In those messy situations, health workers sometimes try a different tactic: instead of chasing specific people, they vaccinate everyone in the specific neighborhood where the potato was found. This is called "targeted geographic vaccination." The big question for scientists is: in a real-world emergency, which strategy saves more lives and uses resources better? It's a crucial puzzle because outbreaks don't always happen in perfect, calm conditions, and health teams need to know which tool to grab from their kit when things get tough.

This paper is a high-tech simulation lab where researchers built a digital version of a virus outbreak to test these two strategies against each other. They didn't just guess; they created a complex, computer-generated world based on real data from the Democratic Republic of the Congo, complete with virtual families, neighborhoods, and a virus that behaves just like the real thing. They ran thousands of different scenarios, changing how fast the virus spreads, how quickly they found the first case, and how well the health team could do their job.

The results are a bit like a sports match where the winner depends entirely on the weather. When the health team is working well—finding cases quickly and tracing contacts effectively—the "ring vaccination" strategy is the clear champion. In these good scenarios, the ring strategy prevented about 20% more cases than the geographic approach, used fewer vaccine doses, and saved more money overall. It's like having a precise sniper who hits the exact targets needed to stop the game.

However, the story changes when the situation is a disaster. If the health team is struggling—maybe because of conflict, fear, or a lack of resources—and they can't find or trace contacts well, the "targeted geographic" strategy actually performs slightly better. In these worst-case simulations, vaccinating the whole neighborhood prevented more infections and stopped the biggest, most dangerous outbreaks from growing out of control, all while using fewer vaccine doses than the struggling ring strategy.

But here is the most important twist the paper reveals: the choice between these two vaccination methods matters much less than how well the health team does the basics. The simulations showed that improving how fast they find the virus and how well they isolate sick people had a far bigger impact on the outcome than which vaccination strategy they picked. Whether they used rings or neighborhoods, if the response was slow or messy, the virus still spread. The authors suggest that while ring vaccination is the best standard plan when things are under control, targeted geographic vaccination is a smart backup plan for when things go wrong. Ultimately, the paper suggests that no single vaccine strategy is a magic wand; the real key to winning the game is a strong, fast, and well-organized response team.

Problem Statement
The World Health Organization (WHO) recommends ring vaccination (RV)—immunizing contacts and contacts-of-contacts of confirmed Ebola Virus Disease (EVD) cases—as the primary strategy to contain outbreaks of Orthoebolavirus zairense. However, in remote or conflict-affected settings, effective contact tracing required for RV is often impeded by factors such as population mobility, social stigma, and community mistrust. Consequently, responders have increasingly adopted Targeted Geographic Vaccination (TGV), which immunizes entire communities near a case regardless of known contact status. While previous modeling studies have evaluated single strategies or used stylized scenarios, there has been a lack of direct comparison between RV and TGV across a range of realistic outbreak settings, particularly regarding cost-effectiveness and operational constraints. This study addresses the gap in understanding which vaccination strategy is optimal under diverse epidemiological and operational conditions.

Methodology
The authors developed a spatially-explicit, individual-based stochastic transmission model (ebolasim) calibrated to outbreaks in the Democratic Republic of the Congo (DRC). The model generates a demographically realistic synthetic population in North-East DRC, capturing household, extended family, and community contact structures. It simulates EVD transmission, disease progression (including postmortem transmission via funeral practices), and outbreak response interventions, including safe and dignified burials, isolation in Ebola Treatment Centres (ETCs), contact tracing, and vaccination.

The study simulated 45 distinct scenarios varying:

Transmission intensity: Low ( $R_0=1.5$ ), medium ( $R_0=1.8$ ), and high ( $R_0=2.1$ ).
Detection timing: Early (20 undetected infections), medium (40), and late (100).
Response effectiveness: Worst-case, moderate-case, and best-case parameter sets for case ascertainment, contact tracing, and vaccination.
Sensitivity scenarios: Pairing best-case TGV parameters with worst-case or moderate-case non-vaccination responses to account for perceived higher acceptability of TGV in stigmatized environments.

For each scenario, 500 stochastic outbreak simulations were performed. Outcomes measured included total cases, deaths, epidemic duration, vaccine doses used (estimated at the 95th percentile for planning), and the Incremental Net Monetary Benefit (INMB) of TGV relative to RV.

Key Contributions

Direct Comparative Analysis: Unlike prior studies that often evaluated strategies in isolation or retrospectively, this work directly compares reactive RV and TGV across a comprehensive matrix of transmission and response scenarios.
Operational Realism: The model integrates complex operational constraints, including delays in detection, variable contact tracing success, and the specific mechanics of ring formation versus geographic deployment.
Cost-Effectiveness Integration: The study calculates INMB using outbreak response cost estimates from the 2018-2020 Nord Kivu outbreak, providing an economic dimension to the epidemiological comparison.
Policy-Relevant Thresholds: By estimating vaccine dose requirements at the 95th percentile, the study offers practical guidance for stockpiling and resource planning under tail-risk scenarios.

Results

Context-Dependent Performance: No single strategy dominates across all scenarios. The relative performance of RV and TGV is heavily dependent on the effectiveness of the broader outbreak response (surveillance, contact tracing, isolation).
Moderate-to-Best Case Responses: In scenarios with effective case detection and contact tracing, RV generally outperforms TGV. Under moderate response conditions with medium detection delays, RV resulted in approximately 18% fewer cases (210 vs. 256) and 15% fewer deaths compared to TGV. RV also consistently required fewer vaccine doses and demonstrated a higher probability of positive INMB (being more cost-effective) in 59–67% of simulations.
Worst-Case Responses: Under severe operational constraints where contact tracing is insufficient, TGV slightly outperforms RV. In worst-case scenarios, TGV prevented more infections and curbed the largest outbreaks more effectively than RV, requiring fewer doses (e.g., a 21% reduction in mean outbreak size in specific sensitivity analyses). However, the cost-effectiveness (INMB) between the two strategies in these scenarios was largely indistinguishable.
Impact of Detection and Response: The study found that improvements in outbreak detection timing and the efficacy of non-vaccination response measures (ascertainment, isolation) had a substantially greater impact on reducing epidemic size than the choice between RV and TGV. For instance, moving from a moderate to a best-case response scenario reduced outbreak size by 35%, whereas switching from TGV to RV reduced it by only 18% in the same context.
Vaccine Dose Requirements: In moderate and best-case scenarios, RV required fewer doses than TGV. Conversely, in worst-case scenarios, TGV required fewer doses. Under high transmission and worst-case response, both strategies struggled to control transmission, with estimated dose requirements reaching up to 1.5 million for RV and 690,000 for TGV, though these figures are interpreted with caution due to uncontrolled epidemic growth in simulations.

Significance and Claims
The authors conclude that while vaccination is a critical component of the outbreak response toolbox, it is not a panacea. The primary determinant of successful EVD control remains robust surveillance, rapid case detection, and effective isolation.

Strategic Recommendation: When supported by effective case detection and management, RV remains the preferred approach due to its superior epidemiological impact and resource efficiency.
Contingency Planning: TGV should be considered a viable contingency option, particularly in settings where operational constraints (such as conflict or mistrust) severely limit the feasibility of contact tracing and ring formation.
Decision-Making: The study emphasizes that decision-makers cannot confidently identify a priori which strategy is more cost-effective under challenging conditions due to considerable stochastic variability. However, the findings support a pragmatic approach where the vaccination strategy is chosen based on the specific outbreak context without sacrificing overall impact, provided that core response measures are strengthened.

The paper explicitly notes that its findings apply to Orthoebolavirus zairense and does not address the 2026 outbreak caused by Orthoebolavirus bundibugyoense, for which no vaccines are currently approved.

More like this