A scoping review methodology was chosen and conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, exclusively with its extension for scoping reviews (PRISMA-ScR) (8), as it allows for the comprehensive mapping of the broad, interdisciplinary body of research on disaster medicine education and training that integrates new technologies for educational purposes. Supplementary Table 1 presents the corresponding PRISMA-ScR Checklist.

Through this approach, we aim to address the following research questions:

To ensure a comprehensive review, we included all original studies that reported on technological innovations in DM education. Eligible study designs encompassed experimental, quasi-experimental, mixed-methods, and feasibility studies. The training programs targeted healthcare professionals including physicians, nurses, paramedics, and students or residents, and aimed to develop disaster-related knowledge and skills. Interventions had to incorporate technology-based educational methods, including but not limited to mixed reality, simulation platforms, mobile applications, e-learning tools, gamified systems, smart devices, sensor-based tools, or any communication and information technologies (ICTs). Only peer-reviewed articles published in English from the year 2000 onward were included, with no geographical restrictions.

We conducted a comprehensive literature search across four electronic databases: PubMed, Scopus, Web of Science, and IEEE Xplore. The search was conducted in July 2024. It covered publications from January 1, 2000, to the date of the final search.

The search strategy combined terms related to DM and emergency preparedness (e.g., disaster, mass casualty, emergency medicine) with terms related to education and training (e.g., education, training, simulation) and simulation technology (e.g., virtual reality, mobile app, e-learning, ICT, smart, gamification). Boolean operators (AND, OR) and truncation were applied to maximize sensitivity. Supplementary Table 2 presents the implemented search string.

Following the above eligibility criteria, titles and abstracts were first scanned independently by two reviewers (JGU, MAK), with the support of the software CADIMA (9), to select articles for in-depth analysis if both reviewers agreed upon. This web-based software platform streamlines the screening and data extraction process. After the initial screening process, reviewers assessed the full-text eligibility for inclusion. During the full-text screening phase, studies were selected only if there was an agreement among the reviewers, and a third researcher (BA) acted as arbitrator when there was no consensus. Subsequently, a comprehensive data extraction sheets was created to extract relevant information for thematic analysis. The primary author (JGU) extracted information about each included study, including the first author, publication year, country, study design, the number and type of participants, and details about the intervention. This information encompassed the type of simulation technology used, the comparator, and the training content. Additionally, results regarding the impact of the training and the methods employed to measure this impact, such as metrics and tools, were also collected.

Data from the included studies was collated and tabulated to provide a comprehensive overview of the use of simulation technology in DM education and training. A semi-quantitative analysis using descriptive statistics was conducted to summarize the key characteristics of the studies. After identifying the overall trends, a qualitative synthesis was performed to gain deeper insights into the main topic, as well as the most commonly used tools and metrics.

The search retrieved a total of 1,917 articles from the four databases. After removal of duplicated (n = 598), the titles and abstracts of 1,319 records were screened for eligibility. Of these, 161 articles were selected for full-text review by the authors, resulting in 32 studies that finally met the established inclusion and exclusion criteria to be included in this scoping review. This literature search process is presented in the PRISMA flow-chart (Figure 1).

Publications date from 2000 to 2024. 16 records were conducted in North America, particularly the United States, while the other half of studies are from East Asia and Europe. Over half of the studies adopted quasi-experimental designs (n = 17) (10–26), others utilized randomized controlled trial design (n = 12) (27–37), feasibility or pilot designs (n = 2) (38, 39) and qualitative methods (n = 1) (40). A comprehensive summary of the extracted data is provided in Table 1.

Eight studies targeted an audience composed in first place by Emergency Medical Services (EMS) professionals. Following, five studies (16, 21, 22, 33, 41) delivered to undergraduate medical students, four studies to professional nurses (13, 25, 26, 35), three (11, 14, 20) to undergraduate paramedicine students, and other three studies to undergraduate nursing students. One study (40) on mental health specialists, another on professional paramedics, and one last study (15) on civil protection operators. The number of participants per study ranges from 4 to 120. The total number of participants was 1.464 among all studies, professional nurses being the most frequents (n = 340), followed by EMS personnel (n = 300), and medical students (n = 263).

Regarding the content, fifteen training courses aimed to the carrying out of execution of patient triage during MCIs, followed by four studies (13, 18, 26, 29) on decontamination in Chemical, Biological, Radiological, Nuclear, and high yield Explosives (CBRNE) disaster scenarios, and another four (24, 30, 37, 41) on general concepts of disaster management. Two studies (16, 21) on flood and earthquake evacuations, two studies on outbreak/pandemic preparedness (25, 38), and two other studies (35, 40) on mental health support in disasters. Lastly, the studies on hospital disaster preparedness (17), civil protection rescuing procedures (42), and in EMS executive management (20), were the least represented.

Technology-based trainings were primarily delivered using various forms of virtual reality (VR) simulation; from thirteen articles using desktop-based simulation (14, 18, 20, 25, 27, 28, 30, 33, 34, 36, 37, 41, 43), to five studies (11, 15, 16, 26, 38) utilizing fully immersive simulation methods. Successively, e-learning platforms were reported in four articles (10, 17, 24, 35), followed by high-fidelity mannequins (12, 22, 23), augmented reality (31, 39), mobile-based technology (21, 40), 360° immersive simulation (13), video-based trainings (29), and text-based mobile messaging (32). The relationship between technology used and topics taught can be appreciated in Figure 2.

Examples of desktop-based simulations range from simple disaster footage projected on screen walls and trainees' individual screen-based multi-patient scenario (30), to the VR system CAVE, which is a full-immersion virtual environment enclosed by walls, floor, and ceiling, creating a realistic replica of a disaster using sophisticated three-dimensional computer-based imaging (27), the serious game “60 Seconds to Survival” (44), a tabletop virtual system (14), the online virtual simulation “Second Life” (45), simulation model of a regional EMS system that replicates the course of action after a 911 call (20), and the XVR training software (33). In these, extended reality (XR) accessories such as joysticks were utilized, and victims were simulated by avatars, sometimes replicating standardized patients. Head-mounted displays and tracking sensors were used by fully immersive simulations, exposing participants to the sensorial challenges of close-to-real disaster scenarios, to train and evaluate their behavior during exercises.

Notably, in all but one of the included studies, participants actively engaged with the technology themselves. In contrast, McCoy et al. (39) assessed the feasibility of a disaster course delivered via tele-simulation. In their study, an instructor used smart glasses to stream a live, interactive MCI scenario while acting as a paramedic evaluating victims and verbalizing key clinical information to remote learners.

Twenty-one studies had a least two training groups, four of which compared a type of virtual reality simulation (fully immersive, 360° immersive, augmented reality, and desktop-based) with traditional lecture sessions (11, 29, 35, 41), and other 11 studies with conventional live training methods (13, 16, 24–27, 33, 34, 36, 43, 44). The remaining five studies implemented a variety of other digital technologies; for instance, e-learning platform were used in two studies comparting with text-based traditional education and live simulation (24, 35). One study used video-based footage to compare CBNRE training performance against traditional lecture (29). Another study compared a text-based messaging mobile application with traditional voice over two-way radio during disaster simulation (32). One study used a mobile-based simulation app comparing with paper-based hazard maps to teach flood evacuation steps (21). Lastly, one study didn't introduce a comparator, rather confronted the performance of both groups on CBNRE disaster scenarios (18).

The remaining 11 articles had only one group that undertook either one or more tech interventions (10, 12, 14, 15, 17, 20, 22, 23, 38–40). Among these, two were feasibility studies of fully immersive and augmented reality (38, 39), and one compared two triage systems using the same desktop-based VR simulation method (SALT vs. SMART) (14).

Finally, it is worth noticing that the included articles revealed a research trend over the last 20 year, accentuated on the study of mass casualty triage, which can be appreciated in Figure 3.

The majority of the studies taught triage, either alone (10–12, 14, 22, 23, 27, 31–34, 36, 39, 43, 44) or in combination with other related content (13, 17, 18, 26, 29, 30, 41). Assessed as triage accuracy and time to triage, alongside with knowledge acquired, treatment/intervention accuracy, and performance correctness, these studies investigated the use of technology-enhanced educational strategies to train healthcare professionals and students in triage protocols under disaster conditions. Table 2 provides an insightful overview of this thematic cluster.

Desktop-based virtual reality was the most commonly used single tech to teach triage for MCIs (14, 18, 27, 30, 33, 34, 36, 41, 43, 44). These platforms immersed learners in virtual MCI scenarios where they were required to perform patient assessments, prioritize interventions, and allocate resources. Notably, some VR environments were found to offer better data capture and time-stamped data on triage actions, enhancing the granularity of performance assessment (14), although challenges such as user navigation difficulties (30) and low scenario novelty (44) were also reported.

Fully immersive VR and 360° VR simulations provided a more sensorial engaging experience, used to replicate high-pressure disaster environments. Studies using these methods (11, 13, 38) reported strong user engagement and perceived training value. However, technical barriers such as complex controls, hardware discomfort, and high costs were consistently mentioned. Augmented reality and smart-glasses-based interventions offered in two studies (31, 39) real-time overlays of clinical information or tele-simulation perspectives. These innovations were found to improve decision accuracy and broaden remote training possibilities, yet were limited by battery life, technical compatibility with eyewear, and the need for stable connectivity infrastructure.

E-learning modules (10, 17) and video-based trainings (29) provided more accessible formats for large-scale deployment. These studies showed consistent post-intervention improvements in knowledge and practical application, with Báez et al. reporting skill retention at one-month follow-up (10). Curtis et al. found video-based learners performed better in personal protective equipment (PPE) use and decontamination tasks than those taught via lecture (29). However, these methods lacked the experiential dimension of immersive platforms.

High-fidelity mannequins and live-actions simulations, featured in three studies (12, 22, 23), continued to play a valuable role in the hands-on skills development. While learners reported increased confidence and improved teamwork skills, high-fidelity mannequins were occasionally limited in replicating physiological responses.

Lastly, only one study (32) explored a different approach to disaster communication through text-based messaging mobile application against voice over radio, reaching improved information accuracy during hospital response to an MCI simulation.

In all these virtual environments, a variety of MCI scenarios were simulated, from urban area explosions (12, 13, 22, 23, 27, 31, 34), CBRNE events (18, 26, 29, 30), natural hazards such as earthquakes, floods and tornados (32, 36, 43, 44), mass shootings (39, 43, 44), car crashes (14, 33), and other non-specified MCIs scenarios (10, 11, 30, 41).

Measurements tools and metrics varied across studies. Most used scores, checklists, or pre/post-knowledge tests, while some conducted surveys with Likert scales. One study (18) implemented debriefing and focus group discussion to record participants experiences. Only three articles conducted follow-up assessment within their methods, completing post-test within 2 weeks, 1 month, and 2 months (10, 27, 30).

Comparative studies revealed mixed findings, highlighting either no improvement or no change in improvement in comparison to traditional methods (11, 27, 31, 33, 36, 43). For instance, while Knight et al. and Cicero et al. supported VR's superiority over traditional card-sort or lecture-based training (34, 44), others such as Shubeck et al. found participants preferred live-action training due to its perceived realism and greater emotional engagement (36). Moreover, Follman et al. highlighted a trade-off between quality and efficiency, noting that improvements in triage accuracy with augmented reality technology came at the cost of longer assessment times (31).

This scoping review synthesized 32 original studies published between 2000 and 2024 that examined the use of technological tools in DM education and training. In doing so, it addressed the primary research question by mapping current evidence on how simulation technologies have been utilized to enhance knowledge acquisition, technical skills development, decision-making, and learners' engagement in disaster settings. The review also provided insights into the secondary research question by analyzing the outcome measures and evaluation strategies used to assess training impact, revealing substantial variability and lack of standardization across studies.

Mass casualty triage was the most prominent topic in the included studies and the over where digital training approaches were most actively developed. Over two-thirds of the included studies addressed triage either as the primary learning objective or as a key element of broader disaster preparedness curricula.

The reviewed studies employed a range of digital modalities to simulate mass casualty incidents and evaluate learners' ability to assess, prioritize, and manage multiple victims. These simulations commonly focused on structured protocols such as START or SALT, and measured outcomes like triage accuracy, speed, and decision-making under pressure.

Although most studies reported positive short-term outcomes such as improved knowledge or triage accuracy, relatively few demonstrated statistically significant advantages of technology-enhanced methods over traditional pedagogical approaches such as lectures, tabletop exercises, or live-action simulations. Several studies, particularly those comparing VR with traditional simulations, found no significant differences in performance outcomes (27, 33, 43). Moreover, some participants expressed a preference for live-action scenarios, citing higher perceived realism and emotional engagement (36).

Despite the growing interest in simulation technology for DM education, our review found that the evaluation of training effectiveness remains inconsistent and largely unstandardized. Outcome measures across the included studies varied widely, with most relying on short-term knowledge quizzes, self-reported confidence, or simplified checklists. This pattern reflects what Cook et al. (7) described as a recurring challenge in digital learning environments, where the complexity of technologies often outpaces the development of appropriate evaluation frameworks, making it difficult to assess effectiveness beyond superficial metrics (42). In our review, none of the included studies employed structured tools, and only a few used validated instruments or follow-up assessments (30, 41). Several factors may explain this gap, including the lack of disaster-specific evaluation frameworks (46), and practical constraints that favor the use of simple, low-resource assessment methods over validated, behavior-based instruments (47).

At the same time, the increased reliance on simulation-based training is not unique to disaster medicine. Virtual reality and other immersive technologies are being increasingly adopted across health professions education, showing promising results in areas such as cardiopulmonary resuscitation and emergency care training. As highlighted by Trevi et al., simulation is emerging as both an effective and cost-effective modality in broader clinical education contexts (48). This further underscores the urgency of developing robust, transferable evaluation strategies that can be adapted across disciplines and scenarios, including but not limited to disaster response training.

These findings are consistent with those of Voicescu et al. (49), who reported a widespread mismatch between the educational objectives of disaster management programs and the strategies used to evaluate their outcomes. While many programs aimed to develop applied competencies these were often measured using basic tools that capture only surface-level cognitive gains. Our review reinforces this observation in the context of technology-enhanced training: although many interventions sought to build operational triage capabilities or situational awareness through immersive or interactive modalities, their impact was typically assessed using low-resolution, knowledge-based instruments.

Previous research indicated that simulation fidelity—the extent to which and educational environment replicates real-world conditions—plays an important role in shaping learning outcomes (50). Across several studies in our review, participants reported that immersive VR and high-fidelity simulation environments improved their engagement, emotional involvement, and ability to make rapid triage decisions under pressure (13, 16, 38). These tools commonly provided real-time feedback, sensory immersion, and dynamic scenarios that stimulated the cognitive and emotional challenges of mass causality incidents, supporting faster decision-making and triage. In contrast, desktop-bases simulation and e-learning modules, while useful for foundational knowledge, were often perceived as less realistic and less helpful in preparing learners for the stress and ambiguity of mass casualty incidents (30, 43).

This difference in learner perception aligns with the broader simulation literature, which emphasize that emotional, physical, and conceptual fidelity are essential to effective experiential learning, particularly in high-stakes, team-based scenarios like disaster response. Zechner et al. (51) echoes this in their mixed reality prototype study, demonstrating that the incorporation of realistic environmental cues—such as visual distraction and situational variability—along with adaptive scenario challenges, improved participants' sense of preparedness by more closely replacing the dynamic and unpredictable nature of real–world MCIs. Chang et al. (52) similarly found that tactile feedback from a capillary refill simulator resulted in more accurate diagnostic judgments compared to video–only instruction.

Furthermore, Weinstein et al. (53) concluded that effective MCI simulation must balance high physical conceptual, end emotional fidelity. This assertion is also reflected in our review, suggesting that hybrid and multi-modal formats hold promise, even if they were only explored in a few studies (25, 31).

These converging findings suggest that the effectiveness of technology-enhanced disaster education appears to depend less on the type of technology used and more on how well it is integrated into a coherent, immersive, and learner centered training ecosystem. Rather than novelty or format alone, realism, interactivity, and scenario flexibility appear to be the key drivers of meaningful learning. As digital tools become increasingly accessible and sophisticated, the next challenge may lie in ensuring their use is aligned with clear educational goals and embedded in structured, outcome-based training programs.

This review identified triage as both a central of current educational efforts and a key area for future research. Its prominence in literature and operational relevance makes it an ideal testbed for intervention studies.

Notably, no study in this review addressed the use of Artificial Intelligence (AI), Machine learning, or adaptive learning systems is DM education, despite being included in the search strategy. Further research could investigate how AI-enable platforms might support dynamic scenario generation, personalized feedback, or real-time assessment in high-pressure training environments.

To move the field forward, educators and training developers are encouraged not only to adopt emerging technologies, but to integrate them onto pedagogically sound curricula that emphasize realism., feedback, and behavioral assessment. Building on this review, our forthcoming experimental study will examine the use of a mobile application to teach triage principles to medical students using tabletop simulation design.