Effectiveness of a mixed reality simulation program for dyspnoea care on new nurses’ clinical competency: A mixed-methods study

Abstract

Aims

This study developed and evaluated a mixed reality (MR) simulation program for dyspnoea care (D-MRSim) to enhance new nurses’ competency.

Background

Managing dyspnoea requires advanced skills, which many new nurses lack. Innovative training, such as MR simulation, may bridge this gap.

Design

This study used a mixed-methods design combining a randomized controlled trial and focus group interviews (FGIs).

Methods

A total of 59 new nurses at a university hospital in South Korea participated, with random assignment to the experimental (n = 30) or control group (n = 29). The experimental group received D-MRSim (three 2-hour sessions over 6 weeks), while controls received equivalent traditional training. D-MRSim, developed using the ADDIE (Analysis, Design, Development, Implementation, Evaluation) model, was evaluated through surveys, performance assessments and FGIs.

Results

The experimental group demonstrated significant improvements over time in confidence (χ² = 82.27, p < 0.001), problem-solving ability (χ² = 23.06, p < 0.001) and knowledge (χ² = 33.34, p < 0.001). The experimental group showed significantly greater knowledge (F = 10.17, p = 0.002) and higher clinical performance both immediately (Z = -6.12, p < 0.001) and 3 months after the program (Z = -5.55, p < 0.001). FGIs identified four themes: (1) Realistic learning environment, (2) Effective learning approach, (3) Clinical competency enhancement and (4) Limitations and need for improvement.

Conclusions

D-MRSim effectively enhanced new nurses' clinical competency by providing an immersive learning environment with repeated practice opportunities, serving as a valuable educational tool for improving clinical skills and adaptability.

1 Introduction

The growing complexity of healthcare environments, coupled with rapid technological advancements, has made clinical adaptation and competency development for new nurses a critical issue in nursing education ( ^{Fernandez et al., 2020}). The ineffective management of complex clinical problems, such as dyspnoea, can lead to severe patient outcomes, including increased morbidity and mortality ( ^{Sethi et al., 2023}). Although nursing curricula cover respiratory assessment and intervention, newly graduated nurses often struggle to apply this knowledge effectively in real-life clinical situations, particularly under time-sensitive and high-stress conditions ( ^{Liu et al., 2023}). This challenge intensified during the COVID-19 pandemic, as many new nurses reported feeling unprepared to manage respiratory problems, resulting in heightened stress and diminished confidence ( ^{Chen et al., 2021; Ji and Lee, 2021}). Dyspnoea, a potentially life-threatening symptom that requires rapid assessment and intervention, poses specific challenges for newly graduated nurses due to their limited clinical experience ( ^{Santus et al., 2023; Prosen and Ličen, 2023}).

To address these challenges, the field of nursing education is incorporating a variety of innovative teaching methodologies, with simulation-based education attracting considerable attention, which can offer clinical experiences that closely resemble real-life scenarios in a safe environment ( ^{Foronda et al., 2020}). However, traditional simulation methods are limited in their ability to replicate environmental elements and capture the complexity of actual clinical settings, potentially restricting the development of critical thinking and practical skills ( ^{Liu et al., 2023; Jagatheesaperumal et al., 2024}).

Advances in technology have led to the introduction of virtual reality (VR) and augmented reality (AR), providing enhanced experiential learning opportunities ( ^{Foronda et al., 2020; Tang et al., 2020}). Furthermore, mixed reality (MR) integrates the strengths of both VR and AR, creating an environment where physical and virtual elements can seamlessly coexist and interact ( ^{Frost et al., 2020}). MR offers a dynamic educational setting through visual, auditory and tactile stimuli ( ^{Hauze et al., 2019; Kim et al., 2023}), fostering deeper learning and enhancing clinical skills retention ( ^{Garcia et al., 2021}). Compared with VR, MR reduces side effects such as dizziness, allowing for prolonged usage, thereby making it an effective educational method that enhances learner engagement and learning outcomes ( ^{Kim et al., 2021}). MR also allows safe, repetitive practice across diverse clinical scenarios, providing real-time feedback and facilitating data collection to objectively assess and improve learner performance ( ^{Kim et al., 2021; Moon et al., 2024}). These attributes highlight MR as an innovative tool with significant potential to advance nursing education.

Current research on MR-based nursing education predominantly gravitates towards nursing students, with limited studies focusing on clinical nurses, particularly new nurses ( ^{Kim et al., 2021; Moon et al., 2024}). Despite the growing incorporation of MR in nursing education, there is a significant gap in research on MR simulation programs specifically designed to enhance the competency of novice nurses in managing patients with respiratory distress. This gap is concerning given that dyspnoea—a high-risk condition requiring immediate and accurate clinical intervention—is among the most challenging symptoms for new nurses to manage, necessitating systematic and comprehensive training ( ^{Liu et al., 2023}).

Accordingly, the aim of this study was to develop an MR simulation program for dyspnoea care (D-MRSim) to enhance new nurses’ competency and evaluate its effectiveness. The specific objectives were as follows: 1) To develop a tailored D-MRSim designed to improve the competency of new nurses in patient care; 2) To assess the impact of D-MRSim on new nurses’ confidence, problem-solving ability, knowledge and clinical performance in dyspnoea care; 3) To explore the experiences of new nurses who participated in D-MRSim.

1.1 Conceptual framework

This study was based on Kolb’s Experiential Learning Theory ( ^{Kolb, 1984}), which outlines a four-stage learning cycle: concrete experience, reflective observation, abstract conceptualization and active experimentation. D-MRSim was designed to align with this theory, incorporating pre-briefing, simulation practice and debriefing to enhance learning outcomes.

In the concrete experience stage, new nurses engaged in immersive MR simulation scenarios involving dyspnoea patients, gaining practical experience. The reflective observation stage allowed participants to reflect on their actions during debriefing, reinforcing knowledge and building confidence ( ^{Cant and Cooper, 2017; Fey and Jenkins, 2015}). During the abstract conceptualization stage, nurses integrated their observations and reflections, deepening their understanding and problem-solving ability ( ^{Geeta, 2023}). Finally, in the active experimentation stage, nurses applied the acquired skills in actual clinical settings, bridging the gap between theory and practice, which improved their confidence and clinical performance ( ^{Thomas and Revell, 2016}).

By leveraging Kolb’s learning cycle, D-MRSim effectively enhanced critical clinical competencies, such as knowledge, confidence, problem-solving skills and clinical performance. Simulation-based education effectively supported the clinical adaptation of new nurses, facilitating the integration of theoretical knowledge with hands-on practice ( ^{Shin et al., 2015}; ^{Franklin and Lee, 2014}), thus preparing them to care for dyspnoea patients in complex healthcare environments.

2 Methods
2.1 Study design

This study employed a mixed-methods research design to evaluate the effectiveness of D-MRSim in enhancing new nurses’ competency. A randomized controlled pretest-posttest design was used for the quantitative component and focus group interviews (FGIs) were conducted for the qualitative aspect ( ^{Fig. 1}). The study was registered with the National Institute of Health, Korea Disease Control and Prevention Agency (Registration Number: KCT0009087).

2.2 Participants & Allocation

The study participants were new nurses who joined a tertiary general hospital in South Korea in 2023, with employment duration of less than one year. The inclusion criteria were: (1) nurses within their first year of employment with no prior nursing experience, (2) willingness to participate in the D-MRSim program, (3) comprehension of the study purpose and (4) agreement to attend all training and evaluation sessions. Exclusion criteria were: (1) previous employment as a registered nurse; (2) prior experience with MR or similar simulation programs; (3) history of motion sickness or visual impairment affecting MR use; or (4) unwillingness to participate.

The sample size was calculated using G*Power 3.1.9.7. According to ^{Padilha et al. (2019)}, a medium effect size (ES) (f = 0.25) was applied. Based on an alpha level (α) of 0.05, a power (1-β) of 0.95, two groups, three repeated measurements and an assumed medium effect size of 0.5 for correlations between time points, the minimum required sample size was determined to be 44 participants (22 per group). Accounting for potential dropouts, a total of 60 participants (30 per group) were recruited. A post-hoc power analysis was conducted using G*Power version 3.1.9.7, which showed a significance level of α = 0.05, an effect size of d = 2.11 and a power of 1-β = 0.96.

This study adhered to the CONSORT (Consolidated Standards of Reporting Trials) checklist for randomized controlled trials ( ^{Supplementary 1}). Randomization was conducted using a random number table generated via the Social Psychology Network’s research randomizer. While the clinical performance evaluator was blinded to group assignments during assessments, complete double-blinding was not feasible due to the nature of the intervention. The experimental group used the D-MRSim equipment in a designated training room, while the control group received traditional training. One participant from the control group withdrew during the second follow-up survey due to resignation, resulting in final group sizes of 30 (experimental) and 29 (control) ( ^{Fig. 2}).

2.3 D-MRSim development

The D-MRSim was developed using the ADDIE (Analysis, Design, Development, Implementation, Evaluation) model using Microsoft HoloLens 2 and Microsoft Dynamics 365 Guides for the MR environment. Each phase of development yielded specific insights and improvements to the program:

During the analysis phase, a literature review focusing on “mixed reality,” “nursing education,” and “dyspnoea” was conducted alongside a needs assessment survey of new nurses. Key learning requirements identified included managing sudden dyspnoea, arterial blood gas analysis (ABGA) and physician communication. The analysis revealed that new nurses particularly struggled with rapid clinical decision-making in respiratory distress situations.

In the design phase, we created detailed learning objectives: (1) performing physical assessments for dyspnoea patients; (2) recognizing and describing dyspnoea changes; (3) using situation, background, assessment, recommendation (SBAR) technique for physician communication; and (4) demonstrating oxygen administration procedures. Microsoft Dynamics 365 Guides was selected for its stability and user-friendly interface after evaluating various MR platforms.

The development phase involved creating a teaching plan using Microsoft Dynamics 365 Guides. A detailed description of the MR interface design, including representative screenshots and question formats, is provided in ^{Supplementary 2}. Content validity was evaluated by six experts (two professors specializing in adult nursing and four nursing managers with over 20 years of clinical experience, all holding a master's degree or higher), achieving a content validity index (CVI) of 1.0. Expert feedback led to three major improvements: enhanced patient communication realism, addition of positive reinforcement elements and better alignment with nursing duties.

Implementation began with a preliminary survey of nine new nurses (September 18, 2023) to assess program operability and usability (score: 85.13/100). This testing revealed several key challenges: user unfamiliarity with HoloLens devices (reported by 6 participants), occasional tracking issues (3 participants) and navigation difficulties (4 participants). In response, we enhanced the pre-briefing with detailed device instructions and developed standardized troubleshooting procedures ( ^{Supplementary 3}).

The finalized program structure included three 2-hour sessions over 6 weeks, with scheduled practice time between sessions (maximum 2 hours/week). Each session consisted of pre-briefing (15 minutes), simulation practice (45 minutes), debriefing (30 minutes) and self-directed practice (30 minutes). The simulation scenarios progressed through three stages: initial presentation, condition worsening and improvement.

Throughout implementation, we systematically documented technical issues and participant experiences. No serious adverse events occurred, though minor technical difficulties affected 5 % of sessions, primarily involving device connectivity (3 %) and tracking accuracy (2 %). All issues were resolved through standard troubleshooting procedures. The program's development was supported by a research grant from the Korean Society of Nursing Science, with no commercial conflicts of interest.

This systematic development process yielded several key lessons: (1) detailed device orientation is crucial for successful implementation; (2) standardized troubleshooting procedures enhance program reliability; and (3) progressive scenario complexity helps build user confidence. These insights may guide future MR-based nursing education developments.

2.4 Intervention procedures
2.4.1 Experimental group

The experimental group received D-MRSim consisting of three 2-hour sessions over 6 weeks. Each session included pre-briefing (15 minutes), simulation practice (45 minutes), debriefing (30 minutes) and self-directed practice (30 minutes). Participants used Microsoft HoloLens 2 to engage with dyspnoea care scenarios, performing patient assessment, clinical decision-making, oxygen therapy, high-flow nasal cannula (HFNC) application and ABGA procedures. Scenarios progressed through initial symptom presentation, condition deterioration and improvement following interventions. Between sessions, participants could access the D-MRSim equipment for additional practice (maximum 2 hours/week).

2.4.2 Control group

The control group received traditional dyspnoea care matched in duration and content to the experimental group. Sessions consisted of didactic lecture (30 minutes), procedure demonstrations (30 minutes), hands-on practice with standard equipment (45 minutes) and discussion/feedback (15 minutes). Lectures and demonstrations were conducted by experienced clinical educators using conventional simulation mannequins, covering identical theoretical content (respiratory physiology, dyspnoea assessment, oxygen therapy, HFNC, ABGA interpretation). The control group used the same case scenarios and could practice in the standard simulation laboratory between sessions (maximum 2 hours/week).

2.5 Instruments

This study used a structured questionnaire to assess the general characteristics of the study participants, as well as their confidence in dyspnoea care, problem-solving ability, knowledge of dyspnoea care and education satisfaction. Clinical performance in dyspnoea care was evaluated by a single trained instructor.

2.5.1 Confidence in dyspnoea care

The confidence in dyspnoea care scale was developed by four clinical nurse educators, each possessing a master’s degree or higher, at the research hospital. The tool comprises 14 items, divided into three sub-domains: oxygen therapy (5 items), HFNC (4 items) and ABGA (5 items). Each item is rated on a Likert scale ranging from 0 (completely disagree) to 10 (strongly agree), with higher scores indicating greater confidence in dyspnoea care. The validity of this instrument was assessed by six experts, including two professors specializing in adult nursing and four nursing managers with over 20 years of clinical experience at the research hospital, resulting in a CVI of 0.96. In this study, the instrument demonstrated a Cronbach’s α of 0.96, indicating reliability.

2.5.2 Problem-solving ability

Problem-solving ability was measured using the tool developed by ^{Lee et al. (2008)}. This tool consists of 30 items divided into five sub-domains: problem clarification (6 items), solution search (6 items), decision making (6 items), solution implementation (6 items) and evaluation and reflection (6 items). Each item is rated on a Likert scale ranging from 1 (very rarely) to 5 (very often), with higher scores indicating a greater problem-solving ability. According to ^{Lee et al. (2008)}, the overall reliability of the instrument showed a Cronbach’s α of 0.93. In this study, the Cronbach’s α was 0.99.

2.5.3 Knowledge of dyspnoea care

The knowledge of dyspnoea care scale was developed by four clinical nurse educators, each possessing a master’s degree or higher, at the research hospital. The instrument consists of 10 items covering the following areas: respiratory acidosis (2 items), oxygen therapy (2 items), non-invasive ventilation (2 items), HFNC (2 items) and ABGA (2 items). Each item is scored 0 for an incorrect answer and 1 for a correct answer, with total scores ranging from 0 to 10. Higher scores indicate greater knowledge in dyspnoea care. The instrument's validity was assessed by six experts, including two professors specializing in adult nursing and four nursing managers with over 20 years of clinical experience at the research hospital, achieving a CVI of 1.0. The instrument demonstrated a 95 % confidence interval ranging from 5.80 to 6.50, indicating the precision of knowledge estimation. A narrower confidence interval reflects greater accuracy and reliability in the instrument’s measurement ( ^{Cook and Stubbendieck, 1986}).

2.5.4 Clinical performance in dyspnoea care

The clinical performance in dyspnoea care scale was developed by four clinical nurse educators, each possessing a master’s degree or higher, at the research hospital. The instrument consists of 15 items divided into three domains: assessment of patients with dyspnoea (3 items), intervention for patients with dyspnoea (7 items) and post-intervention management of patients with dyspnoea (5 items). Each item was designed to be evaluated by the instructor, with scoring as follows: 0 points for not performed or performed incorrectly, 1 point for immature or insufficient performance and 2 points for correct performance ( ^{Supplementary 4}). Higher scores indicate a higher level of clinical performance in dyspnoea care. The validity of this instrument was assessed by six experts, including two professors specializing in adult nursing and four nursing managers with over 20 years of clinical experience at the research hospital, resulting in a CVI of 0.97. In this study, the instrument demonstrated reliability with a Cronbach’s α of 0.85.

2.5.5 Education satisfaction

Education satisfaction was measured using an instrument developed by ^{Jung (2005)}. This instrument consists of 10 items rated on a Likert scale ranging from 1 (not at all) to 5 (very much), with higher scores indicating greater education satisfaction. In the study by ^{Jung (2005)}, the instrument demonstrated a Cronbach’s α of 0.75. In this study, the Cronbach’s α was 0.98.

2.6 Data collection

Data collection was conducted at three time points: before intervention, immediately after intervention and 3 months post-intervention, from October 10, 2023, to July 26, 2024. The experimental group underwent D-MRSim, while the control group participated in traditional lectures and practical training using the same dyspnoea care scenario. For clinical performance evaluation, a trained instructor who was blinded to group assignments conducted standardized assessments using structured checklists. All evaluations were conducted in the same simulation environment to maintain consistency. To prevent bias, participants were instructed not to discuss their group assignment with the evaluator.

Quantitative data were collected through self-report questionnaires measuring confidence in dyspnoea care, problem-solving ability, knowledge of dyspnoea care and education satisfaction. Clinical performance was evaluated through direct observation by the blinded instructor during standardized patient scenarios. For qualitative data, 17 participants from the experimental group volunteered for FGIs. Four FGI sessions, each with 4–5 participants, were conducted for 90–100 minutes in a quiet conference room at the research hospital. The female corresponding author, who had extensive experience in qualitative research, served as the mediator, while the first male author took field notes. All interviews were audio-recorded with participant consent and transcribed verbatim by a research assistant. The participants were informed about anonymity and their right to withdraw from the interview. A semi-structured interview guide ( ^{Supplementary 5}) was used to ensure consistent exploration of participants’ experiences across all focus groups.

2.7 Data analysis
2.7.1 Analysis of quantitative data

The collected data were analyzed using the SPSS/WIN 26.0 software (IBM Corp., Armonk, NY, USA). The general characteristics of the experimental and control groups were determined, and preliminary homogeneity was verified by frequency, percentage, Chi-square test and independent t-test. Shapiro-Wilk test was performed to assess the normality of data distribution. To analyze changes in outcomes across the three time points (pre-intervention, immediately after intervention and 3 months post-intervention), repeated measures analysis of variance was used for normally distributed data, while the Friedman test was employed for non-normally distributed data. Comparisons of post-intervention scores or pre-post mean differences between groups were performed by either independent t-test or Mann-Whitney U test, depending on the data distribution. Statistical significance was set at p < 0.05 for all tests. The reliability of the instrument was confirmed using Cronbach's alpha. ES were determined using Cohen's d to quantify the magnitude of differences among the groups, with values of 0.2, 0.5 and 0.8 representing small, medium and large effects, respectively ( ^{Lakens, 2013}).

2.7.2 Analysis of qualitative data

Qualitative data were analyzed using NVivo version 12 (QSR International, Burlington, MA, USA). Content analysis of the FGI data was conducted following the method outlined by ^{Graneheim and Lundman (2004)}. Initially, an independent researcher assigned a unique number to each participant and transcribed the FGI data while maintaining anonymity. The corresponding author and the primary researcher then examined the accuracy of transcription by cross-referencing with field notes and recordings. The corresponding author conducted a detailed analysis by thoroughly reading the data to understand the context, identify meaningful units and generate codes. These codes were systematically categorized based on similarities and differences to capture explicit and observable aspects of the text. To ensure the rigor and reliability of the findings, the analysis was reviewed and refined through discussions with a third-party nursing professor who is an expert in qualitative research. FGI data were analyzed using content analysis methods, adhering to the COREQ (Consolidated Criteria for Reporting Qualitative Research) guidelines ( ^{Supplementary 6}).

2.7.3 Rigour

To ensure the rigor of the study, the qualitative research evaluation criteria proposed by ^{Lincoln and Guba (1985)} were applied. These criteria included truth value, applicability, consistency and neutrality. To establish truth value, FGI recordings were transcribed and cross-checked with the original data and the analysis results were verified by three participants. For applicability, the study context was described in detail and purposive sampling was conducted to include new nurses from various departments. Data collection continued until data saturation was reached. Consistency was ensured by meticulously recording the research process and having two experienced researchers collaborate on the evaluation. To maintain neutrality, collected data were anonymized and reflective analytical memos were written after each interview. Additionally, semi-structured questions were employed during FGIs and a neutral stance was maintained throughout the process.

2.8 Ethical considerations

This study was conducted after receiving approval from the Institutional Review Board (IRB) of the research hospital (blinded for review) and written informed consent was obtained from all participants. The voluntary nature of participation, anonymity and confidentiality were explained to all participants. They were informed they could withdraw from the study at any time without any negative consequences. Confidentiality of data regarding research participation and personal information was maintained throughout the study.

3 Results
3.1 Participant recruitment & baseline characteristics

The total number of participants was 59 (30 in the experimental group and 29 in the control group). A comparison of the general characteristics between the two groups revealed no statistically significant differences in age, gender, or work department distribution, confirming the homogeneity of the groups ( ^{Table 1}).

3.2 Differences in dependent variables between and within groups

Confidence in dyspnoea care (χ² = 82.27, p < 0.001), problem-solving ability (χ² = 23.06, p < 0.001) and knowledge of dyspnoea care (χ² = 33.34, p < 0.001) were significantly improved over time for all participants. In comparison to the control group, the experimental group showed a significant increase in knowledge (F = 10.17, p = 0.002), with a medium effect size observed immediately after the program (ES = 0.51) and 3 months post-program (ES = 0.71). Clinical performance in dyspnoea care was also significantly improved in the experimental group immediately after the program (Z = -6.12, p < 0.001, ES = 2.50) and 3 months later (Z = -5.55, p < 0.001, ES = 2.11), indicating a very large effect size. Differences in confidence in dyspnoea care, problem-solving ability and education satisfaction between groups were not statistically significant ( ^{Table 2}, ^{Fig. 3}).

3.3 Data analysis of FGIs

FGI participants comprised 17 new nurses (3 males and 14 females) aged between 23 and 26 years. The participants’ working departments were as follows: 5 (29.4 %) in the internal medicine ward, 4 (23.5 %) in the surgical ward, 4 (23.5 %) in the internal medicine intensive care unit (ICU), 3 (17.7 %) in the surgical ICU and 1 (5.9 %) in the emergency room (ER).

Four major themes emerged from FGI content analysis. Subtopics and representative quotes for each theme are presented in ^{Table 3}.

Theme 1. Realistic learning environment of D -MRSim

• 1) High similarity to clinical practice Participants noted that D-MRSim closely resembled actual clinical settings, making them feel as if they were in an actual hospital environment.
• 2) Immersive learning experience

The simulation provided a high level of engagement, with participants expressing that the immersive nature of MR enhanced their concentration and learning.

Theme 2. D-MRSim as an effective learning approach

• 1) Systematic learning structure Step-by-step guidance and immediate feedback were highlighted as crucial features that helped participants understand and master the learning content.
• 2) Opportunities for repeated learning and practice

The ability to revisit challenging content and engage in hands-on practice was considered beneficial for reinforcing learning.

Theme 3. Clinical competency enhancement through D-MRSim

• 1) Improved practical clinical skills Participants felt that D-MRSim improved their ability to handle complex medical devices and procedures, expecting these skills to transfer to actual clinical settings.
• 2) Improved clinical judgment and coping skills

The simulation helped enhance decision-making abilities and preparedness for emergencies.

• 3) Increased confidence in clinical application

Practicing in a realistic environment boosted the participants' confidence in applying their skills to actual patient care.

Theme 4. Limitations and need for improvement in D-MRSim

• 1) Technical limitations Participants experienced difficulties with device operation and issues with screen clarity, indicating a need for technical improvements.
• 2) Need for improvement and expansion

There was a demand for a wider variety of clinical scenarios and the inclusion of more complex procedures in D-MRSim.

4 Discussion

This study evaluated the effectiveness of D-MRSim designed for new nurses and explored their experiences. The findings indicated that D-MRSim significantly enhanced new nurses’ knowledge and clinical performance in dyspnoea care. Furthermore, qualitative analysis showed that D-MRSim contributed to an immersive learning environment, thus serving as an effective learning method and strengthening clinical competency.

The quantitative results showed confidence in dyspnoea care and problem-solving ability improved over time in both groups without significant between-group differences. Similarly, ^{Alamrani et al. (2018)} reported that both simulation education and traditional education methods enhanced critical thinking and confidence. While quantitative measurements showed no group differences, FGI data revealed participants experienced “increased confidence in clinical application” following D-MRSim, suggesting benefits not fully captured by standardized instruments. This improvement likely stems from gradual skill development through repeated clinical experiences rather than teaching methods alone. Team-based learning approaches are recommended for enhancing problem-solving skills ( ^{Yeung et al., 2023}), which may explain why D-MRSim’s focus on individual capabilities had limited impact on problem-solving ability.

D-MRSim significantly enhanced knowledge of dyspnoea care. Notably, even 3 months after the program's completion, the knowledge level of the experimental group remained higher than that of the control group, with the effect size increasing from medium to large. Therefore, the realistic learning experience and opportunities for repeated practice provided by D-MRSim may be effective in retaining knowledge over the long term. This suggests that the multimodal learning experience provided by MR creates stronger and more durable knowledge encoding through simultaneous engagement of visual, auditory and tactile pathways, addressing a key challenge in nursing education: ensuring critical knowledge remains accessible long after initial training. FGI results confirmed this finding, as participants reported that the visual and auditory elements felt more realistic and closely resembled actual clinical settings. Additionally, D-MRSim allowed learners to repeatedly learn from difficult choices, providing immediate feedback that supports the metacognitive process, enabling self-reflection within a structured learning environment and contributing to long-term knowledge retention ( ^{Kosior et al., 2019; Moon et al., 2024}).

D-MRSim significantly enhanced clinical performance. In comparison to the control group, the experimental group demonstrated higher clinical performance, both immediately after the program and 3 months later, with a very large effect size. This represents compelling evidence of MR’s educational value, as clinical performance is ultimately the most relevant outcome for patient care quality. MR’s unique contribution appears to be its ability to bridge the theory-practice gap by creating contextual learning environments where skills are acquired in conditions resembling real clinical practice. The improvement in clinical performance may be attributed to the “high similarity to clinical practice” and “opportunities for repeated learning and practice” highlighted by participants in FGIs. Similarly, a study by ^{Cho and Kim (2024)} found that immersive learning through simulation could effectively enhance clinical performance by offering realistic clinical experiences, with debriefing sessions playing a significant role in reinforcing this effect.

The results of qualitative analysis revealed several advantages of D-MRSim. Participants positively highlighted the realistic learning environment, systematic learning structure and opportunities for repeated practice provided by D-MRSim. These features have consistently been identified as significant benefits when applying MR technology to nursing education ( ^{Kim et al., 2021; Moon et al., 2024}). In particular, the participants emphasized the “high similarity to clinical practice” and the “immersive learning experience,” noting that realistic voices, resembling interactions with actual patients and medical staff, reduced the need for unnecessary imagination. Additionally, the detailed step-by-step guidance provided by D-MRSim, such as the preparation and application of A-line and HFNC, allowed the participants to experience the entire process of patient care. This experience was imprinted on new nurses as a valuable opportunity for developing practical skills ( ^{Dicheva et al., 2023}).

MR applications in nursing education are expanding, though still emerging compared with traditional methods. Current applications primarily focus on fundamental skills like intravenous cannulation ( ^{Kim et al., 2023}), physical assessment ( ^{Frost et al., 2020}) and cardiac emergency management ( ^{Moon et al., 2024}). While previous studies demonstrated MR’s effectiveness for specific procedures such as catheterization and wound care ( ^{Viglialoro et al., 2021}), our D-MRSim uniquely advances the field by teaching complex clinical reasoning for respiratory distress management. Unlike applications focusing on isolated skills, our program integrates comprehensive clinical judgment from assessment to evaluation, specifically designed for new nurses managing respiratory complications. These findings suggest that D-MRSim may serve as an innovative educational method that effectively addresses the limitations of traditional nursing education.

The limitations of D-MRSim and the need for further improvement were also identified. Addressing the technical limitations highlighted by the participants, particularly difficulties in device operation and screen clarity, is crucial for narrowing the gap between virtual reality and actual reality. Enhancing these aspects can improve the learners’ initial learning curve and sense of immersion ( ^{Viglialoro et al., 2021}). Furthermore, there was a demand for simulations that cover a wider range of clinical scenarios and more complex medical procedures, which indicates the direction for developing future MR-based nursing education programs. The findings are consistent with those from previous studies that have emphasized the necessity for continuous technical enhancements and content expansion when implementing MR technology in nursing education ( ^{Frost et al., 2020; Jagatheesaperumal et al., 2024}).

There are some limitations in this study. The true educational value of MR simulation lies in its immersive environment that enhances knowledge retention and clinical performance through psychological and environmental fidelity, allowing for more effective transfer of learning to actual clinical situations, particularly for complex skills like dyspnoea management. First, the study was conducted at a single institution, which may limit the generalizability of the results. Second, although a 3-month follow-up was conducted to assess the long-term effects, a longer follow-up period would be beneficial to determine the optimal timing of additional training to enhance long-term retention and effectiveness. Third, while we emphasized the benefits of repeated practice opportunities in MR simulation, we did not systematically collect data on the actual frequency and duration of individual practice sessions beyond the structured program components, which limits our ability to quantitatively assess the relationship between practice frequency and learning outcomes. Fourth, as this study focused specifically on caring for patients with dyspnoea, further research is required to evaluate the effectiveness of MR simulation in other clinical contexts.

5 Conclusion

This study demonstrated that D-MRSim significantly improved nursing knowledge and clinical performance in dyspnoea care by providing an immersive learning environment and opportunities for repeated practice, contributing to a higher level of clinical competency among new nurses. Furthermore, the realistic learning environment provided by D-MRSim offered experiences that closely resemble actual clinical situations, suggesting its effectiveness in increasing confidence in clinical application. However, technical limitations, such as difficulties in device operation, were identified, indicating the need for future improvements. Overall, D-MRSim showed potential as an innovative tool in nursing education, with the capability to effectively enhance the clinical performance of new nurses. Future research should focus on evaluating the development and long-term effectiveness of D-MRSim in various clinical situations, which will further strengthen the effectiveness of nursing education using MR simulation and expand its applicability in diverse educational settings.

Ethical Approval

The study was approved by the Chonnam National University Hospital Institutional Review Board (Approval number: CNUH-2023–298).

Ethical Compliance

This research was conducted in compliance with all applicable ethical standards. All methods were carried out in accordance with relevant guidelines and regulations. Informed consent was obtained from all subjects involved in the study.

Funding Sources

This study was supported by the Research Grant of Korean Society of Nursing Science in 2023.

Acknowledgments

I would like to express our appreciation to the nurses and clinical nurse educators for participating in our study and for their valuable responses.

CRediT authorship contribution statement

JEONG Hye Won: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Funding acquisition, Formal analysis, Conceptualization. Hwi Gon Jeon: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Conceptualization.

Authorship Contribution

All authors have made substantial contributions to the conception and design of the study, data acquisition, analysis, and interpretation. All authors have been involved in drafting the manuscript or revising it critically for important intellectual content and have approved the final version to be published.

Originality

The authors confirm that this work is original and has not been published elsewhere, nor is it currently under consideration for publication elsewhere.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.nepr.2025.104397.

Appendix A Supplementary material

Supplementary material

Supplementary material

Supplementary material

Supplementary material

Supplementary material

Supplementary material

Table 1

Comparison of baseline characteristics between the two groups (N = 59).

Variables		Total	Exp. (n = 30)	Cont. (n = 29)	χ 2 /t	p
M ± SD or N (%)	N (%)	N (%)
Age in years	≤ 24	24.42 ± 3.29	20 (82.8)	24 (66.7)	2.01	.156
	>24		10 (17.2)	5 (33.3)
Gender	Male	10 (16.9)	6 (20.0)	4 (13.8)	0.40	.731 †
	Female	49 (83.1)	24 (80.0)	25 (86.2)
Department	Medical/Surgical ward	31 (52.5)	15 (50.0)	16 (55.2)	0.265	.876
	ICU	12 (20.4)	6 (20.0)	6 (20.7)
	Others (OR, ER, Delivery room)	16 (27.1)	9 (30.0)	7 (24.1)

Table 2

Comparison of dependent variables between groups and within groups (N = 59).

Variables	Group	Exp. ( n = 30)	Cont.( n = 29)	Sources	F/χ 2 ( p)	t/Z ( p)	ES (CI)
		M ± SD
Confidence in dyspnea care	Pre a	4.79 ± 1.73	4.55 ± 1.73	Time	82.27 § (<.001) Post hoc: a<b,c	0.40 (.692)	-
	Post 1 b	8.60 ± 1.58	8.16 ± 1.51	Group	1.84 (.180)	−1.38 † (.169)	0.29 (−0.23–0.80)
	Post 2 c	8.44 ± 1.38	7.62 ± 1.94	Time x Group	0.56 ￡ (.525)	−1.55 † (.122)	0.38 (−0.14–0.89)
	Post 1-pre	3.81 ± 2.16	3.60 ± 2.77			0.32 (.751)	-
	Post 2-pre	3.64 ± 2.01	3.07 ± 2.58			0.96 (.342)	-
Problem-solving ability	Pre a	3.83 ± 0.73	3.90 ± 0.81	Time	23.06 § (<.001) Post hoc: a<b,c	−0.57 † (.569)	-
	Post 1 b	4.33 ± 0.64	4.34 ± 0.58	Group	0.02 (.878)	−0.02 † (.982)	−0.02 (−0.53–0.49)
	Post 2 c	4.31 ± 0.70	4.17 ± 0.77	Time x Group	0.60 (.540)	−0.86 † (.388)	0.19 (−0.32–0.70)
	Post 1-pre	0.50 ± 0.66	0.44 ± 0.66			0.37 (.715)	-
	Post 2-pre	0.48 ± 0.70	0.27 ± 0.97			0.94 (.349)	-
Knowledge of dyspnea care	Pre a	6.30 ± 1.34	6.00 ± 1.34	Time	33.34 § (<.001) Post hoc: a<b<c	−1.01 † (.314)	-
	Post 1 b	7.63 ± 1.07	7.17 ± 0.71	Group	10.17 (.002)	−1.78 † (.074)	0.51 (0.01–1.02)
	Post 2 c	8.33 ± 1.32	7.38 ± 1.37	Time x Group	1.14 ￡ (.321)	−2.54 † (.011)	0.71 (0.18–1.23)
	Post 1-pre	1.33 ± 1.85	1.17 ± 1.67			−0.28 † (.782)	-
	Post 2-pre	2.03 ± 1.88	1.38 ± 2.03			1.29 (.207)	-
Clinical performance in dyspnea care	Post 1	26.13 ± 2.46	18.34 ± 3.67			−6.12 † (<.001)	2.50 (1.82–3.18)
	Post 2	27.27 ± 3.24	19.52 ± 4.05			−5.55 † (<.001)	2.11 (1.48–2.76)
Education satisfaction	Post 1	4.80 ± 0.39	4.77 ± 0.47			−0.31 † (.754)	0.07 (−0.44–0.58)

Table 3

Examples of quotations from FGIs (N = 17).

Themes	Categories	Codes	Examples of quotations
Realistic learning environment of MR simulation	High similarity to clinical practice	Reproduction of actual clinical settings	“ It felt as though I was genuinely working in a clinical setting.” (Group 3-Participant 3)
		Implementation of actual clinical situations	“ I was astonished by how closely the patient's breathing sounds, and the monitor alarm resembled that in real life. It genuinely felt as though I was in a hospital room.” (Group 2-Participant 2)
	Immersive learning experience	Elevated concentration	“ It was not monotonous, and I felt more engaged because the simulation wasn't dull; instead, it unfolded vividly before my eyes.” (Group 3-Participant 1)
		Realistic interaction	“ I preferred it because it felt more realistic, as I was actively engaged in the simulation rather than merely discussing it.” (Group 1-Participant 4)
MR simulation as an effective learning approach	Systematic learning structure	Sequential, step-by-step guidance	“ It was easier to understand because the arrows indicated the step-by-step preparation process.” (Group 2-Participant 3)
		Immediate feedback mechanism	“ When reporting to the doctor, it was helpful to watch a video with model answers after completing the report, as it allowed me to receive immediate feedback.” (Group 1-Participant 4)
	Opportunities for repeated learning and practice	Possibility of repeated learning	“ Since I could revisit the parts that are difficult to understand multiple times, it felt as though someone was right next to me demonstrating the process.” (Group 1-Participant 1)
		Opportunity for hands-on practice	“ It was beneficial to be able to perform it myself after observing it, rather than merely watching how it’s done or listening to an explanation before doing it independently.” (Group 4-Participant 4)
Clinical competency enhancement through MR simulation	Improved practical clinical skills	Enhanced proficiency in operating advanced medical equipment	“ I effectively used Airvo during my practice and provided the patient with a clear explanation of the potential side effects.” (Group 2-Participant 3)
			“ I believe I effectively performed ABGA on the A-line and assembled Airvo according to the video during practice.” (Group 4-Participant 4)
		Enhanced proficiency in executing clinical procedures	“ I believe that by studying and remembering the basic oxygen delivery methods I learned today, I will be able to provide more effective nursing care to dyspnea patients.” (Group 3-Participant 1)
	Improved clinical judgment and coping skills	Enhanced clinical decision-making skills	“ It was beneficial to be able to experience everything simultaneously—the patient's condition, the equipment in front of me, the test results, and more.” (Group 3-Participant 2)
		Enhanced emergency response capability	“ If I apply this simulation to emergency situations in clinical practice, I will be better equipped to remember the appropriate actions in the event of an actual occurrence.” (Group 4-Participant 2)
		Enhanced skills in patient education and communication	“ I feel that I have not been providing sufficient explanations to the patient thus far; from now on, I realize the need to practice explaining the purpose of the treatment to the patient in a clear and easy-to-understand manner.” (Group 4-Participant 2)
	Increased confidence in clinical application	Practical experience	“ Since I personally performed the procedure from start to finish as if it were an actual situation, I believe I will be able to respond quickly and accurately in clinical practice when a patient experiences difficulty breathing.” (Group 2-Participant 4)
		Enhanced confidence in clinical practice	“ I believe it will not be difficult the next time I actually manage a patient.” (Group 2-Participant 1)
Limitations and need for improvement in MR simulation	Technical limitations	Operational difficulty	“ Operating the HoloLens was challenging. I had to concentrate to press the next button; however, maintaining a steady focus was difficult.” (Group 3-Participant 4)
		Issues with screen clarity	“ I think it was initially difficult to get accustomed to because operating the HoloLens was uncomfortable, and the vision appeared blurred.” (Group 3-Participant 2)
	Need for improvement and expansion	Request for diverse scenarios	“ I would also like to experience a situation where Airvo is applied, and nebulizer therapy is administered during extubation.” (Group 4-Participant 1)
			“ I would like to try connecting and setting up a ventilator directly for a patient, as well as experience an intubation situation.” (Group 4-Participant 4)