A bibliographic search is carried out in Web of Science (WoS) within the Topic field, including the terms “mass gathering” and “COVID” and that are within the period 2020–2024. The search yielded 199 results, with three works discarded due to duplicity. Two reviewers (P.L & J.M.R.R.) independently screened all manuscript titles and abstracts identified in the literature search as potentially relevant. Disagreements were resolved by consulting a third independent reviewer (G.G.A). During the different stages of the selection process, references were managed using Excel. A total of 29 papers are identified that make direct reference in their abstracts to decision support techniques.

The three authors of this work independently moved on to the complete reading phase of these 29 works, 19 being discarded in a consensual manner for the following reasons: no reference to MGs or hypothetical MGs (n = 7), correspondence, commentary, editorial, opinion or letter (n = 6), review (n = 3), no article peer-reviewed (n = 2); not decision-making technique (n = 1) with finally, 10 studies were included in this review.

Data were extracted by one reviewer (P.L.) and verified by the second reviewer (J.M.R.R). Data were extracted using specific marking colors to match the outcomes of interest. MGs can be classified according to the type of event as religious, sports, cultural, political or musical (8). We will develop the selected works more broadly according to the type of event.

A narrative synthesis of the selected works is carried out, classifying them according to the decision support tools used. No meta-analysis was performed. We focus our data analysis on identifying, by fully reading the works, the different decision-making techniques, as well as synthesizing the main conclusions of each of the works. We rely on Microsoft^® Excel ^® to build a table and radial visualization graphs where each of the decision tools are observed being present in the selected works.

The study, conducted at the University of October 6 in Egypt (9), assesses the risk of mass gatherings during the COVID-19 pandemic in Egypt, following the WHO Strategic Response Plan. The implementation of intra-action reviews at mass events at the university suggested that the high preparedness reduces the transmission of COVID-19. Coordination among the plan's 10 pillars was crucial to preventing transmission during student meetings.

During the United States Republican Convention (10), the successful use of the Johns Hopkins University risk assessment tool at the Republican National Convention is described, which allowed for the safe holding of the event. These data-driven strategies helped protect communities and the local health system, providing lessons applicable in the reopening of schools and public services.

In this type of event, we identified three jobs. In the study dedicated to the effect of non-pharmaceutical interventions in Malaysia (11), a heterogeneous SEIR (Susceptible, Exposed, Infected, Removed) model, which is a mathematical model for the study of infectious diseases was used to assess the impact of non-pharmaceutical interventions following the second wave of COVID-19 in Malaysia. The study showed that the motion control command was effective in reducing transmission. Statistical analyses provided relevant information about the local dynamics of the disease and aided in the decision-making of the Malaysian Ministry of Health.

In the Hajj Risk Management work in 2020 (12), Saudi Arabia successfully implemented measures to mitigate COVID-19 risks during the Hajj in 2020, limiting participation to 1,000 pilgrims. No cases of COVID-19 were identified among participants or staff, highlighting the effectiveness of mitigation strategies implemented by the Saudi government to prevent outbreaks at mass events.

Finally, and related to the Hajj, agent-based simulations are used to model risky contacts between pilgrims during the Hajj (13). The results indicated that as the number of pilgrims increased, it was more difficult to maintain physical distancing, suggesting that contact management is key to assessing transmission risks in future events.

Related to sporting events we find three other works. The first refers to the impact of heat and COVID-19 on the Tokyo 2021 Olympic Games (Japan) (14), where the interaction between the increase in COVID-19 cases and heat illness in Tokyo during the 2021 Olympic Games was analyzed. The authors highlighted that the double burden of COVID-19 and heat-related illnesses could overwhelm health care systems if adequate countermeasures are not put in place.

Secondly, within this type of event we have the assessment of the impact of COVID-19 on the 2022 FIFA World Cup in Qatar (15). This study looked at the impact of COVID-19 during the 2022 FIFA World Cup, revealing a significant increase in cases during the event, but with low mortality rates. The study emphasizes the importance of vaccination and effective collaboration between organizers and health authorities to manage risks at mass events.

Finally, and related to the Tokyo Olympic Games, transmission scenarios are studied through a model of multiple branching processes (16), evaluating the potential transmission of COVID-19 during the Tokyo 2020 Olympic Games. It was estimated that preventive measures could significantly reduce cases, underlining the importance of keeping transmission levels below epidemic levels to avoid contagion between groups.

Our selection of publication highlights two that cannot be classified in the previous points since they address more than one type of MGs.

Malaysia's response to the COVID-19 pandemic (17) used DEMATEL (Decision Making Trial and Evaluation Laboratory) and Fuzzy Rule-Based techniques to assess responses to the pandemic in Malaysia. Movement control orders, international travel restrictions, and the cancellation of mass gatherings were identified as key factors in preventing COVID-19 transmission in the country.

On the other hand, observations of behavior during mass events in Wales (18), an observational study during mass events in Wales showed that personal protective behaviors, such as social distancing and mask wearing, were influenced by the design of the environment and social norms. The results suggest that system-level changes may improve adherence to healthy behaviors in future challenges.

The most identified decision support tools in our selection of documents have been risk assessment (10, 12–14) which is widely used to assess the potential risks associated with mass gatherings and implement appropriate measures and simulation (9, 11–13) is applied to predict the spread of the virus and assess the impact of different protective measures. To a lesser extent, Bayesian analysis (10), decision support systems (11, 13), statistical modeling (11), data analysis (17, 18), multicriteria decision making (17), fuzzy logic (17) and scenario analysis (16) have been recognized.