A comprehensive literature review identified and analyzed existing studies on burnout and its prevention strategies, characterizing burnout as a chronic response to prolonged stress in work or academic settings. The review emphasized online learning as a key approach to mitigating burnout, highlighting the importance of providing emotional support; enhancing educator training; and using diverse, interactive tools to foster student engagement and motivation [30]. In addition, several strategies aimed at enhancing collaboration to combat stress and burnout in health care, including interprofessional education (IPE), medical movies, massive open online courses (MOOCs), and simulation-based interprofessional education (SIMBIE) incorporating co-debriefing and psychosocial support, will be discussed in detail in the following sections.

Among various methodologies, such as mentorship and supervision [31,32], improving collaborative practice has demonstrated success for boosting HCW effectiveness and efficiency. Collaborative practice is when “multiple health workers from different professional backgrounds provide comprehensive services by working with patients, their families, careers and communities to deliver the highest quality of care across settings” [33]. It enhances effectiveness and efficiency by focusing on 4 key competencies: values and ethics, roles and responsibilities, communication, and teams and teamwork [34]. These competencies foster effective communication, satisfaction with high-quality teamwork, interprofessionalism, and a positive attitude with mutual respect for diverse health care disciplines [35,36]. Therefore, collaborative practice is crucial for enhancing patient safety in emergency departments [37-40]. Studies suggest that collaborative practice skills should be developed early in professional education [41-44].

IPE, where “two or more professions learn about, from and with each other to enable effective collaboration and improve health outcomes” [33], is more effective for teaching collaborative practices than other methods, such as traditional lectures [45], meeting discussions [42], video-based education [46], or simulation-based education involving only one profession [46]. IPE fosters positive perceptions of interprofessionalism through experiential learning, helping HCWs develop the 4 key competencies essential for collaborative practices. For over a decade, IPE has been integrated into university studies to promote teamwork [47,48], enhance understanding of professional roles and responsibilities [49], strengthen communication skills [50,51], and expand interprofessional knowledge [52,53].

A meta-analysis review study by Sezgin and Bektas [54] found IPE significantly improves communication competency (n=7; 95% CI 0.26 to 0.82; P<.001) and teams-and-teamwork competency (n=4; 95% CI 0.25 to 0.56; P<.001) among HCWs. However, the analysis included only 8 randomized controlled trials (RCTs)—too few for subgroup analysis of key intervention characteristics. Similarly, the meta-analysis by Marion-Martins and Pinho [55] showed IPE enhances values-and-ethics competency by fostering cross-professional mutual respect (P=.007; n=1) and improves roles-and-responsibilities competency (n=2; P=.004), yet the limited number of studies in each analysis restricts the generalizability of these findings. Overall, meta-analyses have shown that IPE interventions significantly improve health care systems (12 studies: standardized mean difference [SMD]=1.37, 95% CI 0.92 to 1.82 [56]; 6 studies: mean difference 7.19, 95% CI 2.61 to 11.77 [55]). A recent scoping review also highlighted IPE’s positive impact on organizational culture, climate, and staff attachment [57]. Additionally, IPE initiatives improve health care professionals’ work environments and strengthen multidisciplinary team effectiveness [58].

Based on previous evidence-based studies, we can conclude that IPE is an effective methodology for fostering collaborative practices, indirectly alleviating workplace stress, reducing burnout, and decreasing turnover intentions while enhancing HCWs’ well-being, career satisfaction, and perceived service quality. Despite these benefits, Frenk et al, for The Lancet Commissions [59], criticized health education curricula for being fragmented and focused on a single profession, failing to prepare students for team-based clinical practice [59]. Traditional siloed education persists, hindering collaboration and patient safety [60].

The integration of films and television series into medical education, a practice known as cinemeducation [61], has been demonstrated to effectively enhance learning outcomes [62,63]. Cinemeducation has been widely implemented across various disciplines, including medical diagnostics [64], nursing [65], pharmacology [66,67], psychiatry [68-70], and psychology [71,72]. Importantly, cinemeducation extends beyond the mere screening of films in classrooms; it is grounded in a structured pedagogical framework that involves a sequence of carefully designed steps before, during, and after the educational activity [73,74]. Surprisingly, to the best of our knowledge, no study has examined the direct relationship between cinemeducation and stress or burnout among HCWs. Most studies investigated the effectiveness of cinemeducation for alleviating stigma, one of the greatest barriers to mental health treatment [75-79] that indirectly contribute to levels of stress [80,81] and burnout [82,83].

The study by Zeppegno et al [84] indicated that cinemeducation has the potential to reduce stigma, foster positive attitudes toward psychiatry, and enhance students’ ability to manage anxiety when confronted with others’ distress, with effects lasting up to 6 months. The pilot study by Rehl et al [85] demonstrated the effectiveness of cinematic virtual reality (cine-VR) training—a combination of cinemeducation and VR—for reducing stigmatizing attitudes among osteopathic medical students toward patients with opioid use disorder. The study also reported increased empathy, aligning with Vygotsky’s theory of learning, which emphasizes learning through collaboration with others using reflection and authentic activities in real-life situations [86]. Similarly, the study by Kontos et al [87] used a filmed version of a research-based theatrical production to reduce caregivers’ stigmatizing toward dementia. The study by Hawke et al [88] demonstrated that film-based interventions could reduce stigma among health care service providers toward individuals with bipolar disorder, with effects persisting for 1 month. Additionally, the pilot study by Linton et al [89] found that the film “Wounded Healer” significantly reduced stigma toward mental illness among health care students. Based on the reviewed studies, there is no doubt that cinemeducation is an effective approach for reducing stigma among HCWs and students.

MOOCs are characterized as (1) “massive,” enabling access to thousands of learners; (2) “open,” with no enrollment fees; (3) “online,” delivered via the web; and (4) “courses,” offering structured content aligned with specific learning objectives [90]. They were first introduced in 2008 by Stephen Downes and George Siemens [91]. MOOCs have since revolutionized distance learning, driven by growing demand for flexible, accessible education [92,93]. The COVID-19 pandemic further accelerated their adoption, making MOOCs a central focus in education, with over 16,300 courses offered by 950 universities and more than 180 million enrollments globally [94] on platforms such as Coursera, edX, FutureLearn, Thai MOOCs, and Udacity [95,96].

Pedagogically, MOOCs fall into 3 categories: cMOOCs (connectivist), xMOOCs (extension of something else), and iMOOCs (integrated) [97,98]. cMOOCs, based on Downes’ connectivist principles, emphasize networked learning; peer interaction; and evolving, learner-driven content [99-101]. In contrast, xMOOCs prioritize instructor-led, content-centered learning with limited interaction [102]. iMOOCs, developed by Universidade Aberta, blend the collaborative, flexible features of cMOOCs with the structured, assessed nature of xMOOCs, encouraging self-directed learning, peer engagement, and artifact-based assessment (eg, presentations, videos, mind maps) [98]. Due to their scalability and accessibility, MOOCs are highly effective for educational interventions across diverse global audiences [103]. In health care education, MOOCs serve various roles—from promoting health literacy among the public (eg, dementia education [104]) to supporting just-in-time training during health crises [105] and enhancing HCWs’ well-being and resilience against stress and burnout [106].

MOOCs are an effective platform for delivering evidence-based interventions—such as stress and burnout education and mindfulness training—to HCWs, improving resilience while reducing stress and burnout. Recently, Ricker et al [107] conducted a single-group cohort study evaluating the Physician Well-being Course, a 4.5-hour online program covering well-being fundamentals (sleep, nutrition, exercise, resilience, and mindfulness), followed by a 2-week self-selected resiliency activity (10 minutes daily). The course was offered to postgraduate year-1 residents who could join voluntarily. Among 87 enrollees, 53 (61%) completed the course, with meditation being the most frequently selected resiliency activity (36/53, 68%) and sleep being the most frequently reported wellness behavior (22/36, 61%). Postintervention assessments showed statistically significant improvements in emotional exhaustion, depersonalization, and resilience (P<.05, paired t test). The authors suggested a key strategy to obtain a relatively high completion rate was attributed to offering the course during the preresidency timing, when residents had additional time and energy. Limitations of the study were the lack of follow-up assessing long-term effects and a control group.

Peterson et al [106] conducted a single-group cohort study to evaluate a pilot program supporting nursing students with coping with stress and burnout during the COVID-19 pandemic. The intervention consisted of an 8-hour self-paced online course and a 1-hour Zoom support group addressing 7 objectives including stress recognition, crisis response, self-care planning, and psychological safety. Psychoeducational resources such as mindfulness and breathing exercises were embedded, and participants were encouraged to choose a “battle buddy” for peer connection. Of 360 enrollees, 224 completed both pre- and postcourse surveys. Significant improvements were observed in calmness, connectedness, coping capacity, and hopefulness (P<.001, paired t test), reflecting gains in psychological flexibility [108], a protective factor against mental health deterioration [109-111]. The program also enhanced a sense of agency [65] through shared experiences and personalized resilience strategies, contributing to reduced acute stress [112]. Notably, burnout risk decreased significantly (OR 0.58, 95% CI 0.4-0.9; P<.006), which the authors linked to reduced isolation and increased social connectedness—consistent with theoretical models that identify social support and community belonging as key protective factors [113]. However, the study lacked a control group and follow-up data to assess long-term effects.

The asynchronous nature of MOOCs enables participants to engage with content at their own convenience and pace, thereby accommodating demanding clinical schedules. Additionally, MOOCs can support virtual peer groups, which promote shared coping strategies, social connectedness, and a sense of community. These elements are critical mechanisms for fostering psychological resilience and enhancing HCWs’ sense of agency, ultimately contributing to a reduction in acute stress—even under challenging circumstances [112].

As elaborated in the IPE section, IPE is an effective method for fostering the 4 competencies essential for collaborative practices, resulting in the mitigation of stress and burnout among health care professionals. Among other educational strategies [114-118], a recent scoping review and related studies indicate that SIMBIE, defined as “when participants and facilitators from two or more professions are engaged in a simulated health care experience to achieve shared or linked objectives and outcomes” [119], is an effective method for teaching the 4 interprofessional competencies in the emergency department [120-123]. In addition to improving technical skills [118,124,125], evidence-based studies have also demonstrated that SIMBIE enhances nontechnical skills such as communication and teamwork [54,118,124-130], with effects sustained for up to 6 months [124]. This is particularly significant given the increasing recognition that nontechnical skills are critical determinants of patient safety and quality of care [131-133].

Additionally, studies demonstrated unique advantages of SIMBIE over other educational strategies for learning IPE. First, SIMBIE offers a safe [134,135], controlled [127,136], and realistic [121,137] environment where learners from various health care professions can practice collaboration and teamwork without posing any risk to real patients [44,127]. Second, SIMBIE enables repetitive and deliberate practice, allowing learners to refine their skills by learning from mistakes in a risk-free setting—something not always feasible in real clinical environments [138-143]. Third, SIMBIE facilitates active authentic experiential learning by immersing learners in realistic clinical situations that prepare them for real-world practice [137,144,145]. Unlike passive formats, substantial research in adult education arguably emphasizes that active participation significantly enhances learning effectiveness [118,146-152]. Authentic experiential learning of SIMBIE benefits its uses in stress inoculation (SIT) training. Couarraze et al [153] reported positive effects on stress, anxiety, and burnout of anesthesia and critical care workers after attending simulation training based on critical situation exposure: The effects lasted for at least 1 week. Fourth, SIMBIE can expose learners to rare and complex medical conditions that they may not encounter during typical clinical rotations [145,154], thereby equipping them to manage a broader range of clinical scenarios more effectively. Studies have shown that using simulation as a teaching strategy significantly reduces State-Trait Anxiety Inventory (STAI) scores, with this reduction persisting at a 1-week posttraining follow-up. These findings underscore the benefits of simulation-based education, particularly for residents in anesthesia and intensive care [153]. Similarly, Shamputa et al [155] highlighted the effectiveness of virtual IPE initiatives for fostering interprofessional collaboration (IPC), particularly during the COVID-19 pandemic.

Beyond economical [156], logistical [157], scalability [158], and resilience-related [159] benefits, virtual SIMBIE has shown potential for enhancing educational outcomes. A recent meta-analysis reported that virtual SIMBIE significantly improves students’ clinical reasoning and performance [39]. A systematic review and meta-analysis by Jiang et al [151] found no significant differences between virtual and real-world SIMBIE in terms of improving knowledge, procedural skills, clinical reasoning, or communication skills. Liaw et al [160] reported that desktop VR–induced stress levels of medical and nursing students, measured using blood pressure and heart rate, were comparable to the level induced by face-to-face simulation with during simulated rapidly deteriorating patient situations. Similarly, Shamputa et al [155] highlighted the effectiveness of virtual IPE initiatives at fostering IPC, particularly during the COVID-19 pandemic. Although virtual SIMBIE offers standardized and immersive practice scenarios, development of tacit knowledge for clinical practice is facilitated through debriefing [161-164]. Students consistently rated debriefing as essential to their learning experience, a finding supported by prior research [165-167]. Therefore, it is imperative that facilitators of virtual SIMBIE receive appropriate debriefing training to effectively guide these critical reflective processes [93].

Debriefing in the context of simulation-based health care education refers to “the facilitated discussion between two or more individuals in order to guide reflection and review performance, with the intent of gaining insight and understanding such that future performance is improved” [168]. Debriefing is regarded as an interactive, bidirectional, and reflective conversation that involves some degree of facilitation—whether by a facilitator, multiple facilitators (co-debriefing [169]), or the learners themselves (self-guided debriefing [170])—to support the reflective learning process [168,171,172]. When conducted effectively, debriefing thus serves as a critical component of learning in SIMBIE [163,173-175]. Its importance lies in its role as a process of reflection-on-action [176,177], which is a central element of Kolb’s [178] experiential learning cycle. Experience alone during simulation is insufficient to produce learning; rather, it is the intentional reflection on that experience that facilitates deeper understanding [164,173,174,179-182].

Attending debriefing sessions has been studied as a potential strategy for reducing stress and preventing burnout. Although a small study reported no significant effect on burnout scores, most participants appreciated the emotional and social support offered through these sessions and recommended debriefing as a beneficial approach for junior medical residents [183]. Attending debriefing sessions has been shown to significantly reduce the risk of burnout and alleviate work-related posttraumatic stress among intensive care staff, even after accounting for resilience and other contributing factors [184].

Several studies demonstrated that team debriefing after critical events provides various aspects of job and personal resources [185]: for example, (1) psychosocial support [186-193], (2) improvement in teamwork and interprofessional relationships [190,191,194-197], (3) learning and performance improvement [191,193,196-198], and (4) team-based cultural enhancement [189,190,199]. The Job Demands-Resources (JD-R) model [185] hypothesizes that the resources—both job-related (eg, knowledge, skills, abilities, social support) and personal (eg, self-efficacy)—function as a protective “buffer” that mitigates the adverse impact of job demands on individual strain, thereby reducing the risk of burnout [185,200,201].

This hypothesis is supported by both qualitative and quantitative evidence. Qualitative studies highlight the benefits of SIMBIE and debriefing interventions for enhancing emotional support and team reflection [140,186,187,190,191,193,195,202-206]. Quantitative studies further support these findings. For example, although Gunasingam et al [183] found no statistically significant reduction in burnout scores, participants valued the emotional and social support gained from debriefing sessions. Similarly, Colville et al [184] reported that debriefing was significantly associated with reduced burnout and work-related posttraumatic stress among intensive care staff, even after adjusting for resilience and other factors. These findings underscore debriefing as a time-efficient, evidence-based strategy for mitigating burnout in both simulation-based and real-world clinical environments.

Stress experienced within teams engaged in collaborative practice can be broadly categorized into 2 types, based on the differential impact of stressors: individual stress and team stress [207-209]. One of the most widely accepted definitions of stress is provided by Lazarus and Folkman [210], who conceptualize individual stress as “a particular relationship between the person and the environment that is appraised by the person as taxing or exceeding his or her resources and endangering his or her well-being” [211]. Weaver et al [208] extended the definition to the team context, defining team stress as “a particular relationship between the team and its environment, including other team members, that is appraised by the team members as taxing or exceeding their resources and/or endangering their well-being.” Fundamentally, team stress represents a distinct collective psychological construct, whereas individual stress reflects a psychophysiological phenomenon experienced at the individual level [207].

Although individual stress and team stress occur at different levels, they can exert reciprocal and cross-level influences on performance at both levels [209]. Cross-level studies have revealed how specific team-level stressors—such as role ambiguity [212], team climate [213], and team conflict [214]—affect individual team members’ attitudes, behaviors, and emotional states in ways that are relevant to their performance. Conversely, studies have also demonstrated that individual-level stressors, including workload [215] and time pressure [216], can significantly impact team performance. For instance, Savelsbergh et al [217] found that excessive team-level quantitative workload negatively affects both individual and team performance by inhibiting team learning behaviors and further impairs individual performance indirectly through elevated individual workload [217]. Additionally, Kamphuis et al [218] reported that team-level intervention strategies can modulate the effectiveness of individual-level interventions aimed at enhancing individual performance. Collectively, these findings underscore that both individual performance and team performance are shaped by complex interactions between team-level stressors—reflecting characteristics of the team environment—and individual-level stressors—reflecting the attributes and experiences of individual team members.

Several studies have reported that prolonged exposure to occupational distress can ultimately lead to burnout [187,219-229]. This burnout induction process can be explained by the JD-R model [185], which is 1 of the 2 leading theoretical frameworks on burnout [230]. According to the JD-R model, burnout occurs when the availability of resources—both job-related (eg, knowledge, skills, abilities, social support) and personal (eg, self-efficacy)—is insufficient over time to buffer the negative effects of job demands on individual strain [185,200,201]. The model hypothesizes that increasing access to resources—physical (eg, assistance with wearing personal protective equipment [PPE]), psychological (eg, positive feedback from peers or supervisors), social (eg, peer support), or organizational (eg, stress-coping programs)—can help mitigate the risk of burnout.

Our cinemeducation was created on the basis of vicarious learning, which is the modification of an observer’s behavior that is similar to that of a model by watching the model’s behavior be reinforced or punished [231]. The effectiveness of cinemeducation stems from its ability to simplify the comprehension of complex situations by presenting real-life role-playing scenarios that illustrate both effective and ineffective practices. This approach enhances the understanding of human behaviors and their impact on colleagues and patients. Movies in cinemeducation serve as a unique and engaging learning tool [232,233], capable of stimulating debate [234] and providing insights into students’ perspectives on topics that might otherwise remain unexpressed [235-237] or be overshadowed by technical considerations [238]. Cinemeducation thus transforms implicit aspects of human behavior, teamwork, and professional interactions into explicit, tangible learning experiences, fostering deeper comprehension, critical reflection, and meaningful discussions.

Furthermore, movies effectively evoke emotions that foster active participation and deeper learning [239]. They leverage vicarious learning by allowing viewers to emotionally connect with characters [240]. Emotional experiences tend to be more vividly remembered [241-243], facilitating vicarious learning among students, a process supported by the Yerkes-Dodson law [244]. Our arguments on the effectiveness of cinemeducation have been further supported by empirical evidence suggesting that movies are effective at reducing stigma by fostering active learning through emotional engagement [245-247] and by imparting values centered on lived experiences [88]. Moreover, movies could represent a highly complex form of symbolic content that conveys nuanced and richly layered audiovisual information to students through storytelling. Beyond merely delivering content, cinema functions as both a tool and a reflective space, akin to psychotherapeutic sessions, allowing viewers to project aspects of their own psyche onto movies [248]. This process enables viewers to immerse themselves in the narrative, fostering self-reflection and deeper engagement. Additionally, cinema is believed to provide a contemporary experiential medium that momentarily detaches viewers from their daily lives, engaging their unconscious mind in a manner comparable to hypnosis and dreaming [249]. These immersive experiences contribute to the formation of subjective world views, a concept theoretically grounded in Jungian [250] and Hillmanian [251] perspectives on the relationship between images and archetypes. According to this framework, archetypal experiences—facilitated through engagement with cinematic imagery—support learning from both cognitive and emotional perspectives [252]. Such experiences encourage students to engage in discovery and self-reflection, aligning with the principles of inductive learning [253].

Cinemeducation for teaching professionalism as well as stress and burnout education was delivered to HCWs using xMOOCs. xMOOCs represent a practical and scalable solution to deliver interventions that mitigate stress and prevent burnout to health care professionals. The asynchronous nature of xMOOCs enables participants to engage with content at their own convenience and pace, thereby accommodating demanding clinical schedules.

For the experimental groups, we implemented SIT [194] to induce the steeling effect through the integration of cinemeducation, coping-with-stress strategies via MOOC, SIMBIE, and co-debriefing. SIT is a structured cognitive behavioral intervention aimed at fostering resilience by equipping individuals with the tools needed to effectively manage stress. The approach consists of 3 interrelated and overlapping phases: conceptualization, skills acquisition and consolidation, and application and follow-through.

In the conceptualization phase, participants were introduced to the cognitive, emotional, and behavioral manifestations of stress using cinemeducation through medical movies and coping-with-stress strategies. A flipped classroom approach via cinemeducation and MOOC were used to optimize preparedness for the coming SIMBIE sessions [306]. Presession engagement with medical content via xMOOC platforms supported the development of interprofessional knowledge and encouraged positive attitudes IPE, Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS), and stress coping strategies [307]. This model enhanced learner engagement and emotional investment, thereby maximizing the effectiveness of in-session SIMBIE activities [308]. The integration of stress management into clinical education has been widely recognized as essential to the development of professional competencies [309].

In the skills acquisition and consolidation phase, learners developed a repertoire of coping mechanisms—including relaxation techniques; cognitive restructuring; Identify, Situation, Background, Assessment, and Recommendation (ISBAR) [49]; closed-loop communication; and collaborative communication strategies—within SIMBIE sessions that approximate real-world clinical stressors, such as time constraints and multisensory demands (visual and auditory stimuli). Skills consolidation was reinforced through co-debriefing, which provided structured opportunities for critical reflection and feedback.

In the final application and follow-through phase, learners applied these skills during a second round of SIMBIE and were progressively intensified to balance cognitive load and emotional challenge while maintaining learners within their zone of proximal development. SIT not only supports the development of self-regulation and problem-solving capabilities but also enhances self-efficacy by helping participants reframe stressors as challenges rather than threats. Moreover, it fosters a psychologically safe learning environment in which learners can engage with high-stakes situations without fear of failure—ultimately promoting more adaptive coping strategies, reduced anxiety, and improved performance under pressure.

Participants viewed a 75-minute medical movie developed in-house that depicted an interprofessional emergency team managing high-stress scenarios. The film emphasized communication, shared mental models, team reasoning, team support, and collaborative problem-solving for stress mitigation. Viewing was conducted individually and asynchronously, with no postfilm group discussion. Although cinemeducation is evidence-based for reducing stigma [84,85,88], its impact on stress or burnout has not been previously studied.

The MOOCs comprised 7 lessons covering essential topics, including Interprofessional Education Collaborative core competencies for IPC practice, TeamSTEPPS principles, team-based clinical reasoning for diagnosis, stress management and coping strategies, IPE for patient safety, and ethical principles for collaboration. A 15-minute segment focused specifically on stress management and coping strategies, addressing significant stressors faced by emergency department personnel, such as disease outbreaks, heavy workloads, and interpersonal challenges. These stressors were categorized as external (eg, workplace pressures and unexpected events) or internal (eg, personality traits like perfectionism and difficulties with work-life balance). The module emphasized problem-solving approaches over emotional reactions to challenges, aiming to mitigate the long-term effects of stress, including burnout symptoms like emotional exhaustion, depersonalization, and reduced personal accomplishment. All participants were required to complete both the medical movie and MOOCs as well as pass the pre- and posttest exams to receive a certificate and meet the eligibility criteria for the flipped classroom before attending the SIMBIE session.

To promote psychological safety, we implemented structured prebriefing strategies based on best practices [135,168,257]. Facilitators and learners introduced themselves, shared prior experiences, and established ground rules emphasizing confidentiality, active participation, and a focus on performance improvement rather than individual critique [162,171,172,284]. Effective prebriefing must communicate to learners that they are entering a unique, controlled environment for reflective practice where making errors is acceptable and expected as part of the learning process [256].

A “fiction contract” was cocreated to encourage engagement and suspend disbelief despite simulation limitations [310-312]. Learners were oriented to the simulation space, equipment, and environment to reduce anxiety and enhance participation [182,313-315].

For co-debriefing, facilitators conducted prebriefing discussions to align objectives, align areas of expertise, clarify roles, agree on debriefing methods, clarify who will lead different elements, and address potential challenges [168,169]. This precoordination ensured a smooth, cohesive, and collaborative debriefing process and helped establish a shared mental model between co-debriefers [169]. Facilitators also attended to implicit cues—such as tone, expressions, and body language—to foster trust and a positive psychological climate [316-318].

The 3D SIMBIE simulations, available in both computer-based and VR formats, were designed to progressively increase in complexity and stress, replicating high-acuity emergency scenarios reflective of real-world challenges during the COVID-19 pandemic. Each simulation integrated both technical and TeamSTEPPS-based nontechnical learning objectives in the context of a multidisciplinary emergency department encounter. Participants worked as a 6-member health care team to diagnose and manage a complex case involving a 70-year-old male patient with COVID-19, chronic obstructive pulmonary disease, hypertension, diabetes mellitus, and a documented allergy to ceftriaxone. The presence of a distressed spouse added an emotionally charged layer to the scenario, requiring participants to apply both clinical and interpersonal skills.

Technical training included time-sensitive interventions such as intubation, ventilator management, laryngeal mask airway insertion, cricothyroidotomy, hyperkalemic crisis, use of PPE, and coordination of portable chest imaging. Clinical reasoning was guided toward formulating an accurate diagnosis and implementing a safe treatment plan under pressure, closely mirroring the demands of real-world emergency care clinical environments and emphasizing decision-making under pressure, as illustrated by the multiple stressor events shown in Figures 2 and 3. Nontechnical training focused explicitly on the 5 TeamSTEPPS domains—team structure, communication, leadership, situation monitoring with the STEP framework, and mutual support—using strategies such as ISBAR, closed-loop communication, and real-time psychosocial support. Communication occurred via open microphones to simulate authentic interprofessional interaction, and learners engaged with the patient’s spouse through a branching dialogue system that enabled them to select empathetic responses in real time.

An interactive game-based interface further reinforced TeamSTEPPS competencies as our main learning objectives. Each participant was equipped with a HP bar, a symbolic “health point” indicator representing emotional resilience under pressure. The HP bar depletes over time, particularly during high-stakes decision points or delays in clinical action such as managing a “can’t intubate, can’t ventilate” situation, video laryngoscope battery failure, or sudden hypotension. Mutual support was emphasized as teammates could replenish each other’s HP by pressing a “plus” button—an interactive metaphor for peer support and collaborative coping. The HP bar thus served both as a visual representation of stress load and as a positive reinforcement mechanism for TeamSTEPPS-aligned behaviors. These features were designed to enhance psychological fidelity and simulate cognitive load realism by integrating both clinical complexity and emotional dynamics.

This study was conducted during the COVID-19 pandemic, during which all learners were limited to online lectures without clinical exposure. Given these constraints, along with the logistical challenges of assembling multiprofessional student teams and facilitators, we designed the study to ensure both research feasibility and standardization. To address safety concerns for both participants and researchers, a control group (Group A) was exposed to a 3D computer-based SIMBIE simulation without debriefing. All participants provided informed consent after being fully briefed on the study’s potential risks and benefits. Inclusion criteria required that participants have a PHQ-9 score within the normal range, to minimize psychological risk. Trained research assistants were present throughout the simulation to observe and provide support as needed. Participants were also informed of their right to withdraw from the study at any time without penalty. These measures were implemented to uphold ethical standards and ensure fairness, psychological safety, and relevance to the context of the pandemic.

The 3D SIMBIE and 3D VR SIMBIE platforms were developed by our ER-VIPE team, which comprises professionals from various disciplines including emergency physicians, nurses, pharmacists, medical technologists, radiologic technologists, communication arts specialists, instructional designers, architects, psychologists, and experts in the humanities and education. This interdisciplinary team collaborated with a specialized group of engineers in immersive learning technologies for health care education. Both platforms use advanced simulation software to create realistic, interactive environments that enable participants to engage in high-stakes medical scenarios.

The key distinction lay in the level of immersion: The 3D version used standard computer input devices (eg, screen, keyboard, and mouse), while VR SIMBIE used headsets and controllers to deepen emotional engagement and presence. This allowed participants to physically move and interact within the environment in a fully immersive and realistic manner. The VR format enhances the sense of presence, making participants feel as though they are “inside” the clinical scenario, which may result in greater cognitive and emotional stress compared with the 3D desktop version. Figures 2 and 3 illustrate the user interface, visual comparisons, and experiential differences between both platforms. In addition, they show the health point (HP) bar, which can reach a maximum of 50 points, with the ability to restore a teammate’s energy up to 5 times, adding 10 points each time. Each simulation concluded with a structured co-debriefing session, enabling participants to reflect on clinical decision-making and TeamSTEPPS performance, consolidating targeted nontechnical teamwork competencies.

Following the initial 30-minute SIMBIE session, participants engaged in a 60-minute oral co-debriefing session, guided by TeamSTEPPS as the primary learning framework. This session provided a safe, nonjudgmental environment for team-based problem-solving and crisis management. Facilitated by instructors, it aimed to enhance experiential learning, reinforce team dynamics, and foster a no-blame culture. After the co-debriefing, participants in Group B replayed the 3D computer-based SIMBIE for a final 30-minute session, while participants in Group C replayed the 3D VR SIMBIE for their concluding 30-minute session.

In this study, co-debriefing was conducted using two complementary conceptual frameworks designed to optimize critically reflective learning: (1) the PEARLS (Promoting Excellence and Reflective Learning in Simulation) framework [169] and (2) the “divide and conquer” approach [169]. The PEARLS framework offers a straightforward and structured method for facilitating consistent, reliable exchanges of information during co-debriefing. It involves 4 sequential phases: reactions, during which learners can voice emotionally salient concerns or choose discussion points; description, during which facilitators and learners establish a shared, factual account of what happened; analysis, which explores performance gaps and encourages critical discussion; and summary, which consolidates the discussion into key learning points.

The reaction phase helps to establish a shared and psychologically safe foundation for learning and reflection by allowing learners to express their immediate thoughts and emotional responses [315]. For instance, facilitators began with questions such as, “What emotions are you experiencing right now?” to validate participants’ initial reactions and set a supportive tone for reflection. The inclusion of this phase not only supports psychosocial well-being but also may enhance learner engagement by increasing the relevance and resonance of subsequent discussions. In the description phase, debriefers introduce observations coupled with evaluative judgments and may shorten the phase if learners appear to have a shared understanding of the case. This is followed by the analysis phase, in which facilitators guide group reflection using structured prompts. Examples of such prompts included: “Can you describe what was on your mind during the most intense moments?” “In what ways did your team interact and support each other under pressure?” “What specific actions did you take to address the issue, and how successful were they?” and “Are there any other methods you think might have worked in that situation?” Facilitators also applied the plus/delta technique, asking questions such as “What aspects of the scenario went smoothly?” and “What elements could be improved next time?” In addition, the advocacy-inquiry model was used to promote deeper insight into decision-making. For example, a facilitator might say “I’d like to talk about mutual support in TeamSTEPPS. I observed three unsuccessful intubation attempts. I was concerned about the effect of prolonged hypoxia. What was going through your mind as your team worked to establish the airway?”—encouraging participants to share their thought processes and mental models. Finally, the summary phase concluded the session with a review of critical insights. Facilitators asked learners to identify key lessons they planned to apply in future clinical practice, such as, “What are the main takeaways from this experience that you would carry into real-life situations?”

Additionally, we implemented the “divide and conquer” approach during co-debriefing, assigning a lead role—typically the medical debriefer—to guide the session [59,163,319]. Under this model, co-debriefers conveniently divide thematic responsibilities based on the TeamSTEPPS framework, thereby minimizing redundant cognitive processing. This division of labor helps reduce the intrinsic and extraneous cognitive load experienced by co-debriefers—a key consideration in accordance with cognitive load theory, which posits that reducing unnecessary mental burden enhances co-debriefers’ performance and consequently learning outcomes of learners [311,320].

To prevent power imbalances among co-debriefers, we followed strategies suggested by Oriot et al [293] on agreeing in advance about the techniques, approaches, and educational strategies we will use—including our use of nonverbal communication. This is important because, as Holmes and Mellanby [164] noted, “debriefers debrief slightly differently, everybody has a slightly different focus, different speeds and different process for how you do it.” Establishing nonverbal communication allows co-debriefers to “authorize” each other to speak using pre-agreed signals. This approach helps prevent interruptions or conflicts, ensuring that one debriefer does not undermine the other’s process. Additionally, we also conducted a “post-debriefing huddle,” which includes peer coaching and provides an opportunity for co-debriefers to review challenges, resolve misunderstandings, and collaboratively develop strategies for future sessions [168,290].

The subscale of trait anxiety from the short version of the Spielberger STAI [323] was ordered and paid for with the necessary permissions obtained. It is used to assess an individual propensity to anxiety. Trait anxiety data were collected and included as a control variable in the analysis. There are 5 items, and the items use a 4-point Likert scale (1=not at all to 4=very much so). An example is, “In general, I feel that difficulties are piling up so that I cannot overcome them.” The Cronbach α of the original scale was found to be acceptable (α=0.91). This measure was translated into Thai using a back-translation method by J Chavanovanich, PhD (email, August 21, 2024; see Multimedia Appendix 4).

The short-version DSSQ [324] was used to assess the level of subjective stress state. Professor Helton granted us permission for its use. The DSSQ is a 24-item multidimensional measurement comprising task engagement, distress, and worry. The items use a 5-point Likert scale (1=not at all to 5=extremely). An example item is, “I was motivated to do the task.” DSSQ is a well-established tool for assessing acute stress, recognized for its reliability and validity. It consistently achieves high internal consistency, with Cronbach α for the original scale found to be acceptable, exceeding 0.80 across all dimensions. This measure was translated into Thai using a back-translation method by J Chavanovanich, PhD (email, August 21, 2024). We chose psychological tools, such as the DSSQ, instead of physiological measures like heart rate and heart rate variability because they allow for more detailed differentiation between engagement (items: 2, 5, 11, 12, 13, 17, 21, and 22), distress (items: 1, 3, 4, 6, 7, 8, 9, and 10), and worry (items: 14, 15, 16, 18, 19, 20, 23, and 24), offering a comprehensive understanding of stress states (see Multimedia Appendix 4).

A 6-item personal burnout subscale from the CBI [322] was used to assess the level of prolonged physical and psychological exhaustion. The items use a 5-point Likert scale (1=never/almost never to 5=always). An example item is, “How often are you physically exhausted?” The measure was translated into Thai using a back-translation method by J Chavanovanich, PhD (email, August 21, 2024). Permission to translate and use the CBI for this research was granted by the National Research Centre for the Working Environment by T Clausen, PhD (email, June 28, 2022). The CBI has demonstrated strong psychometric properties across various studies. It exhibits high internal consistency, with Cronbach α coefficients typically exceeding 0.90 for both the overall scale and its subscales (personal, work, and patient burnout). Construct validity was supported by confirmatory factor analysis, and convergent validity was demonstrated through strong correlations with a previously validated measurement [322,325-327]. We focused exclusively on personal burnout, as the participants were students who were neither working nor patients. Concentrating on personal burnout was expected to provide a more direct response to the research question and increase the likelihood of successful study completion, although this may reduce validity (see Multimedia Appendix 4).

Descriptive statistics were summarized using means (SDs) and medians (IQRs). Frequencies and percentages were used for categorical data. Comparisons between groups in demographic characteristics at baseline were conducted using the Wilcoxon Mann-Whitney U test, Kruskal-Wallis H test with the Dunn post hoc test, chi-square test, or Fisher exact test, as appropriate. Generalized estimating equations (GEEs) were used to analyze changes over time and assess differences in improvement scores for burnout and DSSQ between groups, with adjustments for anxiety. DSSQ engagement, distress, and worry scores were measured using specific items from the DSSQ. Engagement scores among Groups A, B, and C were obtained from items 2, 5, 11, 12, 13, 17, 21, and 22. Distress scores were derived from items 1, 3, 4, 6, 7, 8, 9, and 10, while worry scores were obtained from items 14, 15, 16, 18, 19, 20, 23, and 24. To assess changes over time and compare improvements in burnout and DSSQ scores between groups, GEEs were used, with adjustments for anxiety as a covariate. Both an intention-to-treat (ITT) analysis and a per-protocol (PP) analysis were performed. Missing data were imputed using the last observation carried forward method. Statistical significance was set at a 2-tailed P value of <.05 for all analyses. Stata version 15 (Stata Corp) [328] was used for data analyses.

To assess the magnitude of intervention effects on burnout reduction, Cohen d was calculated for each of the 5 intervention conditions using mean differences and pooled SDs. Five effect sizes were computed according to the structure of the study. A negative effect size indicated a reduction in burnout compared with baseline, while a positive effect size indicated an increase in burnout.

A power analysis for a repeated measures multivariate analysis of variance (MANOVA) conducted in G*Power (version 3.1.9.4) with a within-between interaction indicated a required sample size of 82 participants to detect an effect size of 0.35, with an α level of .05 and a power of .80. To account for potential dropout, 10% was added, resulting in a total sample size of 87 participants.

The DSSQ [329] was selected as the primary tool in this study due to its ability to assess momentary stress states across 3 dimensions: task engagement, distress, and worry. Designed for real-time application, the DSSQ is particularly well-suited for simulation-based educational research [330]. It enabled the detection of subtle, immediate changes in learners’ stress responses before and after each ER-VIPE session. In contrast, the Perceived Stress Scale-10 [331] is a globally recognized instrument widely used in health care studies. It provides a reliable measure of general perceived stress over the past month and is appropriate for large-scale screening.

Following exposure to a medical movie and MOOCs, Group B had significantly higher DSSQ engagement scores, adjusted for confounding factors, than Group A (mean difference 3.93; P<.001). Group C had nonsignificantly higher scores than Group A (mean difference 2.07; P=.08), and nonsignificantly lower scores than Group B (mean difference –1.86; P=.11). After participating in the SIMBIE process all 3 groups demonstrated positive trends in DSSQ engagement scores, adjusted for confounding factors. Group B scored higher than Group A, though the difference was not significant (mean difference –0.31; P=.77). Similarly, Group C had nonsignificantly higher scores than Group A (mean difference –1.17; P=.27) and nonsignificantly lower engagement scores than Group B (mean difference –0.86; P=.42). See Figure 4, Table S3 in Multimedia Appendix 7, Table S4 in Multimedia Appendix 8, and Figure S1 in Multimedia Appendix 9.

Focusing on DSSQ-distress following exposure to a medical movie and MOOCs, the results indicated negative trends across all 3 groups. Group B exhibited nonsignificantly higher distress scores than Group A, with a mean difference of 0.70 (P=.43). Similarly, Group C had nonsignificantly higher distress scores than Group A, with a mean difference of 0.82 (P=.36) and nonsignificantly lower distress scores than Group B, with a mean difference of 0.12 (P=.90). After participating in the SIMBIE process and adjusting for confounding factors, only Group B exhibited negative trends in DSSQ distress scores. However, the differences for all 3 groups were not significant. Group B scored higher than Group A, but the mean difference was not statistically significant (mean difference –0.69; P=.21). Similarly, Group C had nonsignificantly higher scores than Group A (mean difference –0.38; P=.49) and nonsignificantly lower scores than Group B (mean difference 0.31; P=.57). See Figure 4, Table S3 in Multimedia Appendix 7, Table S4 in Multimedia Appendix 8, and Figure S1 in Multimedia Appendix 9.

Regarding DSSQ-worry following exposure to a medical movie and MOOCs, all 3 groups exhibited positive trends. Group B had nonsignificantly lower worry scores than Group A, with a mean difference of –0.13 (P=.90). Group C also had nonsignificantly lower worry scores than Group A (mean difference –0.30; P=.77) and nonsignificantly higher scores than Group B (mean difference –0.17, P=.87). After participating in the SIMBIE process and adjusting for confounding factors, all 3 groups had negative trends in DSSQ-worry, though no significant differences were detected. Group B had nonsignificantly lower worry scores than Group A (mean difference 0.69; P=.45). Similarly, Group C had nonsignificantly lower worry scores than Group A (mean difference 0.97; P=.30) and nonsignificantly higher scores than Group B (mean difference 0.28; P=.77). See Figure 4, Table S3 in Multimedia Appendix 7, Table S4 in Multimedia Appendix 8, and Figure S1 in Multimedia Appendix 9.