There is no uniform definition of the number of patients in a mass-casualty incident (MCI) as this is dependent on location, setting, and environment. In the literature, however, it is often characterized by many simultaneous patients,Reference Ferrandini Price, Arcos González and Pardo Ríos 1 –Reference Wolf, Bigalke and Graf 2 resulting in a discrepancy between required and available humanitarian and material resources.Reference Follmann, Ohligs and Hochhausen 3 –Reference Nilsson, Åslund and Lampi 5
In addition to transportation accidents, especially plane crashes and railway accidents, an MCI may be after environmental disasters; terrorist attacks; and chemical, biological, radiological, and nuclear emergencies (so-called CBRN hazards).Reference Fröhlich and von Meißner 6 According to a study by Coleman (2006),Reference Coleman 7 the number of MCIs has increased exponentially in the last century. Particularly, there has been an increase in environmental disasters.Reference Coleman 7 –Reference Wijkman and Timberlake 8
Regardless of the cause of a MCI, the focus in such incidents is no longer primarily on the best possible care for the individual, but on the best possible care for the maximal number of patients, true to the motto “to be most good for the most people.”Reference Jones, White and Tofil 4 , Reference Ingrassia, Ragazzoni and Carenzo 9 In such a case, the initial care of victims follows the guideline: triage, treatment, transport.Reference Bazyar, Farrokhi and Khankeh 10 This concept was already defined in 1934 by the military doctors Spire and Lombardy. 11 Here, first responders at the scene of the accident first classify the victims into 3 or 4 triage categories (TC), which determines the treatment and transport priority of the individual persons – before application of individual medical treatment concepts (Table 1).Reference Heinz, Hoffmann and Schweigkofler 12 MCI triage can be defined as initial sorting for life-saving interventions, followed by continuous sorting of patients over time as resources increase, the number and severity of patients is defined, and patients respond, unrespond, deteriorate, or decompensate while awaiting transport.
Table 1. Triage categories*
* according to Bhallah et al. (2015).Reference Barnett, Thompson and Semon 49
TC: Triage Categories.
In order to perform triage reliably and validly, standardized triage algorithms are used.Reference Christ, Grossmann and Winter 13 While some algorithms, for example, Sort, Assess, Lifesaving Interventions, Treat/Transport (SALT) as a “national standard for MCI triage,” can be used universally for any patient in a MCI, there are others that are more specialized.Reference Lerner, Schwartz and Coule 14 For example, Jump Simple Triage and Rapid Transport (JumpSTART) has been developed specifically for pediatric triage.Reference Jones, White and Tofil 4 , Reference Romig 15 In contrast, the triage algorithms modified Simple Triage and Rapid Transport (mSTART) and the Amberg-Schwandorf Algorithm for Primary Triage (ASAV), developed in Germany, are used particularly in MCIs.Reference Romig 15 –Reference Neidel, Salvador and Heller 17
Upon closer inspection, the percentage distribution among the various triage categories varies greatly according to the event scenario. For example, in landslides, 67% of those involved died, while 97% were only slightly injured after the occurrence of a tornado.Reference Juncken, Heller and Cwojdzinski 18 Average values for triage categories in MCI are planning sizes of TC I/II/III (red, yellow, and green) with 20/30/50%.Reference Heller, Brene and Kowalzik 19 The numbers of victims are occasionally classified into groups, presented in Table 2.
Table 2. Number of victims*
* according to Kluger et al. (2020).Reference Kluger, Coccolini, Catena and Ansaloni 50
As MCIs are rare, but significant, events, adequately trained personnel are essential. To respond quickly to such an event, it is important to identify patients in need of immediate treatment as quickly as possible. This is primarily done through screening, which is carried out in the form of triage by professional first responders, some of whom are non-medical (e.g., firefighters and civil protection officials). 20 It is currently unclear which form of training/education guarantees the most sustainable increase in knowledge and skills. In this context, not only the comparison of reliability, accuracy, and time expenditure of different triage algorithms must be addressed, but also the procedures used to provide background knowledge on disasters. In this regard, some studies show that, in addition to conventional teaching methods such as lectures, textbooks, or group discussions,Reference Delnavaz, Hassankhani and Roshangar 21 increasingly more media-based methods are being used. Besides already established techniques, such as Power-Point presentations and video-based learning (VBL),Reference Curtis, Trang and Chason 22 more advanced tools, such as online platformsReference Wiese, Love and Goodman 23 and video games, are also being usedReference Eccles, Hopper and Turk 24 –Reference Mohan, Rosengart and Fischhoff 26 to impart learning objectives. In addition, the influence of active application of what has been learned in the form of simulation exercises (SE)27 or case-based learning (CBL), in which participants apply their acquired knowledge to real-life scenarios and develop solutions with the help of an expert, 28 is increasingly being investigated. Furthermore, it is to be examined which tools, whether technical or non-technical, can optimize triage and, thus, improve patient outcomes.Reference Behmadi, Asadi and Okhovati 29 –Reference Brooks, Curnin and Owen 30 This means that a triage can not only be carried out manually by using paper cards, but also electronically, for example, using a tablet PC.Reference Bolduc, Maghraby and Fok 31 Likewise, manual triage can also be supported with the aid of an aide-memoire (i.e., a pocket card that represents the triage algorithm.)Reference Cuttance, Dansie and Rayner 32
This Scoping review aims to analyze the didactic and content structure of current studies on the topic of prehospital triage training in MCI.
Methods
Search Strategy and Identification of Relevant Studies
This pilot study was conducted as an initial examination of the topic using a limited database, limited search terms, and a limited study period. To find relevant studies, the electronic databases PubMed, Medline, Psyndex, and grey literature were screened with the search strategy basing on 3 main topics:
-
• mass casualty incident as a special event in disaster medicine
-
• the aspect of preparation for such an event
-
• triage as a special procedure for risk stratification.
The following search term was used:
((mass casualty incident) OR disaster OR (multiple casualty incident) OR (mass casualty event)) AND (training OR preparedness OR simulation OR exercise OR intervention) AND (triage).
The inclusion frame was based on the time period from January 2011-December 2024 to ensure that the results reflect current training methods. Only clinical trials with full-text access in English were included (35 by Medline/ PubMed, 3 by Psyndex). In evidence-based medicine and clinical research, clinical trials are systematic surveys that aim, among other things, to evaluate the performance, efficacy, and benefits of specific treatments and interventions (as in the present case, to prepare for a prehospital triage training in MCIs). Clinical trials are conducted to answer scientific questions using a standardized, high-quality procedure. Because the selected PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews) search string yielded more than 1700 sources, this pilot study included only clinical studies that could be selected using the online search engine. These clinical studies provide an overview and guidance for a subsequent systematic review.
Additionally, 5 studies were added via the grey literature search. The sources of the grey literature were internal documents (the authors’ institute maintains one of the world’s largest sources of maritime medical documentation), conference reports, and unpublished theses. Study titles and abstracts were screened according to PRISMA-ScR. This pilot study applied the PRISMA-ScR checklist and aimed to test and refine the scoping review methodology, including data extraction, coding, and analysis, before a full-scale systematic review is conducted (Figure 1). Finally, the full texts of the remaining articles were reviewed for eligibility (Figure 1). Two researchers independently assessed the identified studies and decided whether to include them. If there were different assessments, a third researcher made the final decision.
Figure 1. Flow chart for selection process of suitable references.
Studies were selected under the inclusion and exclusion criteria shown in Table 3.
Table 3. Inclusion and exclusion criteria
Data Extraction and Analysis
The included studies were collected and analysed using Microsoft excel (version 365, Microsoft corporation, 2024, Germany) and the following criteria were extracted: author, title, year, country, population ((investigation vs. control) group, sample size, participation rate, MCI level), study design (control group, follow-up, randomized, pre-test/post-test, cross-over examination [different order in 2 separate intervention parts]). In addition, the types of interventions were stratified concerning the structure (whether a simulation took place as part of the intervention or as a measuring method) and the didactic concept (media-based interventions (power-point [ppt]-presentation, watching videos, technical support during the exercise, video game, online platform) vs. conventional interventions (lecture, textbook reading, group discussion/questions). Finally, the applied outcome parameters (triage accuracy, knowledge, efficiency of triage time, over/undertriage rates, and donning and doffing of personal protective equipment [PPE]) as well as the studies’ main results were summarized. This scoping review did not require submission to an ethics committee or an Institutional Review Board.
Results
Study Design and Aim of Study
Seventeen (n = 17) sources were included in the data extraction phase. Six of the studies were published in North America, 5 in Europe, 5 in Asia, and 1 in Australia. The populations of the studies fell into 4 basic categories: 1. Students (of a medical profession) (n = 6), 2. Nurses, paramedics, and firefighters (n = 5), 3. Emergency Medical Services (EMS) providers (not further specified) (n = 5), and 4. Physicians (n = 1).
Sample sizes varied from 6-292 participants, with a mean of 82 (standard deviation [SD] 62) participants. Participation rates varied from 67%-100% (x̄ 94.2%; SD 8.6%) at baseline, and 67% at follow-up (3), if conducted. Based on the study intervention, 16 references (94.1%) used a control group and divided the entire study population accordingly. The duration of intervention ranged from 15 minutes to a 3-day training course. Only 3 studies had a follow-up measure.Reference Nilsson, Åslund and Lampi 5 , Reference Curtis, Trang and Chason 22 , Reference Aghababaeian, Sedaghat and Tahery 33 The interval between intervention and last data collection ranged from 2 weeks-6months.
In this study, 16 studies (94.1%) were randomized and 9 (52.9%) used a pre-post comparison of measurements. In addition, 4 studies had a cross-over design,Reference Ingrassia, Ragazzoni and Carenzo 9 , Reference Wiese, Love and Goodman 23 , Reference Aluisio, Daniel and Grock 34 –Reference Follmann, Ruhl and Gösch 35 in which all participants took part in 2 interventions, but went through them in a different order. Of the 17 studies, 11 (64.7%) used exclusively quantitative methods, 1 (5.9%) used an exclusively qualitative method. and 5 studies (29.5%) followed a mixed-methods approach. The results of the study population and design of the analyzed studies can be found in Table 4.
Table 4. Study population and design in the 17 identified references
AM, Aide-memoire; ASAV, Amberg-Schwandorf algorithm for primary triage; CBL, case-based learning; CG, control group; EMS, emergency medical service; IG, intervention group; JumpSTART, jump simple triage and rapid transport; MCI, mass-casualty incident; mSTART, modified simple triage and rapid transport; SALT, sort, assess, lifesaving interventions, treat/transport; SE, simulation exercise; VBL, video-based learning.
* All participants took part in 2 interventions but went through them in a different order.
Basically, the studies can be divided into the following main research areas: The studies by Wolf et al. (2014)Reference Wolf, Bigalke and Graf 2 and Jones et al. (2014)Reference Jones, White and Tofil 4 (11.8%) dealt with the comparison of different triage algorithms and a total of 5 studies (29.4%), with ways to optimize the triage process (with electronic aidsReference Follmann, Ohligs and Hochhausen 3 , Reference Bolduc, Maghraby and Fok 31 , Reference Follmann, Ruhl and Gösch 35 –Reference Apiratwarakul, Cheung and Tiamkao 36 and with the triage support provided by an aide-memoire).Reference Cuttance, Dansie and Rayner 32 The remaining 10 studies (58.8%) compared different teaching methods.Reference Nilsson, Åslund and Lampi 5 , Reference Ingrassia, Ragazzoni and Carenzo 9 , Reference Curtis, Trang and Chason 22 , Reference Wiese, Love and Goodman 23 , Reference Aghababaeian, Sedaghat and Tahery 33 , Reference Aluisio, Daniel and Grock 34 , Reference Cicero, Whitfill and Walsh 37 –Reference Zheng, Yuan and Huang 40
In addition, the number of victims within the different simulation exercises varied from 3Reference Aluisio, Daniel and Grock 34-50,Reference Adler, Erfurt and Metz 46 corresponding to MCI numbers of victims I-III (Table 4). On average, 17 victims (SD 9.8), corresponding to MCI numbers of victims II, were triaged in the studies. Depending on the study and intervention group actors, manikins, virtual patients, and/or paper-based patients represented the victims. In some of the simulation exercises, further measures were taken to enhance the realism by using moulage, but also high- and low-fidelity manikins, footages, and, in one simulation exercise, an elaborate coulisse. Different scenarios were applied in the simulation exercises. A corresponding distribution according to the superordinate categories (cf. introduction) can be seen in Figure 2.
Figure 2. Distribution of emergency scenarios of triage exercises.
Didactic and Content Structure of Triage Exercises
Most MCI response studies involve a study group that is trained in an MCI triage system or process and then confronted with a simulated MCI. The studies often include a study group and a control group (without previous training in an MCI triage system), both using a known or other or their usual MCI triage system or process performing triage or sorting of matched simulated patients.
As shown in Table 5, 16 studies (94.1%) collected results directly during a simulation exercise in which participants were asked to perform realistic triage during a fictional mass-casualty incident – 13 studies (76.5%) used the simulation as a measuring method (e.g., to assess the triage time efficacy) and/or 7 studies (41.2%) as part of the intervention (e.g., triage with technical support).
Table 5. Types of interventions in the 17 identified studies
AM, Aide-memoire; CBL, case-based learning; CG, control group; IG, intervention group; mSTART, modified simple triage and rapid transport; ppt, PowerPoint; SE, simulation exercise.
Most of the interventions were based on a combination of different teaching methods. 15 studies (88.2%) used media-based interventions as a didactic method; 5 of them in both the intervention group and control group. It should be noted that the 5 control groups that used media-based intervention did so exclusively in the form of power-point presentations.Reference Jones, White and Tofil 4 , Reference Nilsson, Åslund and Lampi 5 , Reference Curtis, Trang and Chason 22 , Reference Cuttance, Dansie and Rayner 32 , Reference Zheng, Yuan and Huang 40 The media-based interventions implemented in the intervention groups were further divided into studies with video watching (n = 5),Reference Curtis, Trang and Chason 22 , Reference Aghababaeian, Sedaghat and Tahery 33 , Reference Motola, Burns and Brotons 38 –Reference Zheng, Yuan and Huang 40 with technical support during a simulation exercise (n = 4)Reference Follmann, Ohligs and Hochhausen 3 , Reference Bolduc, Maghraby and Fok 31 , Reference Follmann, Ruhl and Gösch 35 –Reference Apiratwarakul, Cheung and Tiamkao 36 (in these cases, in the form of virtual reality glasses and telemedical counselling or electronic triage), with the use of a video gameReference Ingrassia, Ragazzoni and Carenzo 9 , Reference Cicero, Whitfill and Walsh 37 (n = 2), as well as with the use of an online platform (n = 1)Reference Wiese, Love and Goodman 23 (Table 5).
A total of 9 studies (52.9%) used conventional teaching methods in at least 1 of their study arms.Reference Wolf, Bigalke and Graf 2 , Reference Nilsson, Åslund and Lampi 5 , Reference Ingrassia, Ragazzoni and Carenzo 9 , Reference Curtis, Trang and Chason 22 , Reference Wiese, Love and Goodman 23 , Reference Cuttance, Dansie and Rayner 32 , Reference Aluisio, Daniel and Grock 34 , Reference Xia, Li and Chen 39 –Reference Zheng, Yuan and Huang 40 In the control groups, these were mostly lectures, while in the intervention groups, lectures and reading material, as well as group discussions and question rounds, were more often used.
Effects of Interventions During Exercises
In the 17 studies analysed, the parameter triage accuracy was surveyed most frequently (16; 94.1%), followed by knowledge and triage time efficacy (each 8; 47.1%). The 2 studies comparing an intervention vs. non-intervention concluded that an intervention is always associated with a significant increase in knowledge and practical skills compared to no intervention.Reference Motola, Burns and Brotons 38 –Reference Xia, Li and Chen 39 In the first study by Motola et al., both groups were informed about a disaster scenario, with the intervention group additionally receiving a specific demonstration via videos on personal protective equipment and the handling of weapons of mass destruction. However, the control group did not receive such pre-scenario-based video training. In the study by Xia et al., only the intervention group - in contrast to the control group - were specially trained in disaster preparedness by a nursing education program.
Also, in the remaining studies, all intervention groups as well as the control groups that participated achieved better results in terms of knowledge and practical skills in view of the pre-post comparison after the implementation of this intervention. Two studies described that there was no difference between different triage algorithms (JumpSTART vs. SALT; mSTART vs. ASAV).Reference Wolf, Bigalke and Graf 2 , Reference Jones, White and Tofil 4 However, manual triage was found to be associated with improved patient outcome with both aide-memoire supportReference Cuttance, Dansie and Rayner 32 and VR goggles which display the triage algorithm directly in the field of view, as opposed to a triage exercise with telemedicine guidance.Reference Follmann, Ohligs and Hochhausen 3 , Reference Apiratwarakul, Cheung and Tiamkao 36 In addition, in 4 studies, a significantly faster process was measured with electronically performed triage compared to manual triage. However, Bolduc et al. (2018)Reference Bolduc, Maghraby and Fok 31 observed that this time gain does not correlate with improved patient outcome.
Simultaneously, in the study by Wiese et al. (2021),Reference Wiese, Love and Goodman 23 there was no significant difference in the increase in triage accuracy and practical skills between virtual simulation and in-person, so-called immersive, simulation. Similar results were found in the studies by Ingrassia et al. (2015) and Cicero et al. (2019).Reference Ingrassia, Ragazzoni and Carenzo 9 , Reference Cicero, Whitfill and Walsh 37 Nevertheless, in the study by Wiese et al. (2021),Reference Wiese, Love and Goodman 23 the virtual simulation exercise was associated with a greater knowledge gain than in the control group.
In the study by Nilsson et al. (2015),Reference Nilsson, Åslund and Lampi 5 case-based learning in addition to a presentation-based lecture was not shown to result in better triage accuracy. However, case-based training was shown to be superior to live simulation in terms of knowledge gain.Reference Aghababaeian, Sedaghat and Tahery 33
In total, 42.9% of media-based interventions show a comparable and 57.1% a greater training effect than conventional teaching methods. According to 4 studies, technical and non-technical aids increase the triage accuracy. When comparing VBL and traditional teaching methods, there was no significant difference in the gain of self-efficacy and knowledge, but VBL was superior to traditional teaching methods in the acquisition of practical skills.Reference Curtis, Trang and Chason 22 In contrast, the study by Aghababaeian et al. (2013)Reference Aghababaeian, Sedaghat and Tahery 33 found that VBL was significantly inferior to live simulation in gaining knowledge and skills. A distribution of the outcome parameters collected in the studies can be found in Table 6.
Table 6. Outcome parameters and main results of the simulation exercises
AM, Aide-memoire; ASAV, Amberg-Schwandorf algorithm for primary triage; CBL, case-based learning; IG, intervention-group; JumpSTART, jump simple triage and rapid transport; PPE, personal protective equipment; SE, simulation exercise; VBL, video-based learning.
Generally, there is a broad spectrum of studies on didactic methods in MCI triage. Therefore, this pilot study included only clinical studies to identify highly standardized and high-quality intervention studies to provide an overview of didactic concepts in triage exercises. For a subsequent systematic review, further (non-clinical) studies should also be considered.
A limitation of this pilot study is the lack of other databases, comprehensive search terms that could capture more mass casualty incident training, more MCI triage algorithms, and a longer study period to ensure that methods used in the past are captured, even if they involved less advanced technologies (e.g., tabletop or live exercises). The populations of the identified studies were different (students, nurses, paramedics and firefighters, emergency services, and doctors), so that the challenges of didactic concepts in triage exercises must be presented from different perspectives. The high average participation rate (94.2%) suggests that the results in this pilot study may be representative of a larger pool of sources.
As a limitation of this study, due to the heterogeneity in the aims and methods of the included studies, it is difficult to evaluate the effectiveness of different triage algorithms. Furthermore, regarding media-based interventions and technical and non-technical aids, there is a lack of knowledge about long-term effects, particularly as many studies used a pre-/post-test based design. It is also hardly possible to give general recommendations on didactic triage models – particularly as learning preferences vary from person to person. It can also be assumed that the different professional groups (from students to emergency physicians) and age groups in the selected studies prefer different forms of learning concepts. All the mentioned aspects should be the subject of subsequent longitudinal studies comparing different triage concepts in MCI.
Additionally, in this study only studies were included published in English, so relevant studies published in other languages may have been overlooked.
Discussion
The lack of national guidelines often leads to divergent triage processes. Therefore, standardized model core criteria for mass triage of casualties were developed. These consist of 4 core criteria (general considerations, global triage, life-saving measures, and individual triage category assessment).Reference Lerner, Cone and Weinstein 41 Much of the existing literature focuses on general training sessions (e.g., exercises, lectures, mixed methods training, etc.) conducted to improve the participants’ prehospital disaster response.Reference Baetzner, Wespi and Hill 42 There is a large need for training of different occupational groups regarding behaviour and task areas/assignment during MCI. The present results show that there are many approaches to optimizing training output with the help of media-based interventions (Table 5). This results not only in financial benefits, but also in better accessibility for users due to the unrestricted/flexible access in time and place.Reference Karakuş, Duran and Yavuz 43 In part, this was reflected in higher effect sizes.Reference Treloar, Hawayek and Montgomery 44
It is noticeable that none of the studies simulated an MCI level greater III. This can be explained by requirements for further logistics, organization, and time, as well as the higher costs that would be incurred if a simulation exercise was extended to a victim number greater than 50. It should be noted, however, that significantly larger numbers of victims must be expected, especially in the context of a CBRN emergency (e.g., a nuclear attack or the outbreak of a pandemic). Because the resulting triage incidents cannot be managed alone, training in the context of a group operation, in which communication and teamwork are also instilled, should be considered in the future.
It is also necessary to conduct studies with a follow-up period as long as possible to gain a sense of the necessity for periodic repetition of training. The longest follow-up period of the studies reviewed in this paper was 6 months.Reference Nilsson, Åslund and Lampi 5 As evidenced in the study by Curtis et al. (2018),Reference Curtis, Trang and Chason 22 quality of performance declines over time. This is particularly because MCIs rarely occur, so that independent and regular application of the skills learned can only be expected to a limited extent. Thus, the recommendation of the 8th Sighting Consensus Conference 2019 of the Federal Office of Civil Protection and Disaster Assistance Germany 20 is to train first responders annually to maintain competence and to focus on the practical application of the skills. However, the extent to which an annual cycle is sufficient remains to be seen until high-quality studies with a correspondingly long follow-up period have been conducted. It must also be considered that different teaching methods are associated with different knowledge acquisition and retention rates. More active teaching methods are repeatedly described as superior to passive ones, such as didactic lectures.Reference Karakuş, Duran and Yavuz 43 Accordingly, the study by Aghababaeian et al. (2013)Reference Aghababaeian, Sedaghat and Tahery 33 also showed in the follow-up that role plays have a more lasting effect than watching educational videos.
In their systematic review, Baetzner et al. (2022) observed different types of disaster training programs, including traditional (e.g., lectures, real-life scenario training) and technology-based training (e.g., computer-assisted learning, instructional videos).Reference Baetzner, Wespi and Hill 42 Most programs consisted of more than 1 method. Although all methods have been shown to be effective, the literature indicates that technology-based methods often produce similar or better training results than traditional trainings.Reference Baetzner, Wespi and Hill 42 The requirements for information technology (IT)-supported management in triage training during a mass accident event are diverse. Particular attention must be paid to the attitudes and technology affinity of first responders when dealing with this new technology. This must be considered when developing e-triage systems.Reference Adler, Erfurt and Metz 46
Overall, the studies did not provide information about the baseline knowledge of the participants prior to the intervention. However, it should be noted that an intervention always leads to an increase in knowledge and improvement of practical skills. This is particularly evident in the studies of Motola et al. (2015) and Xia et al. (2020),Reference Motola, Burns and Brotons 38 –Reference Xia, Li and Chen 39 which compare intervention versus no intervention. Different triage algorithms have only been investigated and compared to a limited extent, so that the current state of research as presented in this paper does not fully reflect their importance. This can be explained by the fact that the publication period for the study selection was set to the last 12 years. Accordingly, further studies on this topic should be conducted and reflected in future reviews, together with studies that have already taken place.
Both studies comparing different triage algorithms showed that there was no significant difference in triage accuracy or over-/undertriage rates.Reference Wolf, Bigalke and Graf 2 , Reference Jones, White and Tofil 4 These findings are consistent with those of a large-scale review published in 2019.Reference Bazyar, Farrokhi and Khankeh 10 Many triage algorithms are complex and, as a result, are not used or are used incorrectly by responders.Reference McKee, Kerins and Hamilton 47 Accordingly, it is important for future studies to increase research on assistive techniques for manual triage. For example, studies have shown that electronic assistance systems and aide-memoires help to improve the application accuracy of algorithms.Reference Baetzner, Wespi and Hill 42 , Reference Aigner, Kaiser and Miksch 48 The studies by Follmann et al. (2019),Reference Follmann, Ohligs and Hochhausen 3 Bolduc et al. (2018),Reference Bolduc, Maghraby and Fok 31 and Cuttance et al. (2017)Reference Cuttance, Dansie and Rayner 32 examined in this literature review also came to comparable conclusions.
It should be criticized that only 3 of the studies examined (17.6%) measured self-efficacy,Reference Curtis, Trang and Chason 22 , Reference Wiese, Love and Goodman 23 , Reference Zheng, Yuan and Huang 40 because according to Barnett et al. (2014),Reference Barnett, Thompson and Semon 49 perceived self-efficacy is one of the most important predictors of willingness to respond. For example, more than 50% of participating emergency medicine residents in the study of Curtis et al. (2018)Reference Curtis, Trang and Chason 22 reported that they did not receive sufficient training during their residency training to feel confident in handling disaster medicine. Furthermore, it must be noted that while knowledge forms the basis of functioning, it is not sufficient to ensure appropriate practical skills.Reference Barnett, Thompson and Semon 49 Accordingly, the study by Aluisio et al. (2016)Reference Aluisio, Daniel and Grock 34 can be criticized for the fact that only knowledge, but not practical skills, was surveyed within the framework of the outcome parameters. Consequently, in future studies, care should be taken to survey the development of practical skills during the course of the intervention to ensure an unrestricted picture of training success.
Conclusion
In summary, most of the studies follow a randomized controlled design and the results were predominantly collected using quantitative methodology. Different interventions to optimize training success were explored. A great interest in supplementing and/or replacing conventional training methods with media-based procedures can be seen in this review. At the same time, the studies considered allow the conclusion that many of the tested media-based methods have an equivalent, sometimes better training effect, although data on the long-term success of the interventions are not available due to lack of or too short follow-up periods.
Consequently, the present study findings provide an overview for a subsequent systematic review on the topic of prehospital triage training in MCI. In view of the diverse didactic concepts for triage training and the individuality of first responders, a combination or at least a variety of training offers seems advantageous. Future research should focus on how to create and evaluate the effects of different existing training concepts.
Author contribution
Benedict Oldenburg and Clara Wichtmann contributed equally as co-senior authors.
Funding statement
This research did not receive any specific grant from funding agencies in the public, commercial, or non-for-profit sectors.
Competing interests
All authors declare that no conflicts of interest exist.
Open access
