Graphical abstract

Introduction

Background/rationale

Manikin simulation (MK) training is currently a widely used tool for teaching basic life support (BLS) and is recommended because it places learners in a semi-authentic context, enabling them to perform chest compressions and coordinate BLS steps [1,2]. However, training both healthcare professionals and bystanders remains challenging, as survival rates after out-of-hospital cardiac arrest are still low worldwide, despite evidence that early and well-conducted BLS can double survival rates [3,4].

Immersive tools provide innovative learning experiences and high learner immersion, enhancing student engagement and motivation [5]. In the context of BLS training, virtual reality (VR) offers users a simulated, immersive experience within a 3-dimensional environment but does not permit chest compressions to be performed [6]. Conversely, mixed reality (MR) blends virtual and real environments, enhancing interaction between learners and both physical and virtual worlds, allowing chest compressions to be practiced on a physical manikin within a virtual environment [6,7]. However, the efficacy of new immersive tools for teaching BLS remains controversial, particularly regarding their ability to provide sufficient realism and effective skill transfer compared to traditional MK, highlighting the need for studies comparing both teaching methods [8,9]. The cost of the device must also be considered.

Objectives

The study tested the following hypothesis: MR would be non-inferior to manikin-based simulation for medical students’ BLS training. The aim of this study was thus to compare the effectiveness of an MR device with MK training for teaching BLS to medical students.

Methods

Ethics statement

The study was approved by the Sorbonne University Ethics Committee (CER-2023-ART-SIM), declared to the French Data Protection Authority (no., 2232141), and registered on ClinicalTrials.gov (NCT06832072). Written informed consent was obtained from all students.

Study design

This was a non-inferiority randomized controlled trial comparing manikin-based simulation and MR. Its reporting followed the Consolidated Standards of Reporting Trials (CONSORT) guidelines for non-inferiority randomized trials [10].

Setting

The study took place from December 2023 to May 2024 at Sorbonne University’s Faculty of Health (Fig. 1). After randomization, participants received BLS training using either MR or MK, with one teacher per group who was blinded to the study’s purpose. In the MK group, students attended a 2-hour session comprising a 15-minute BLS lecture and 9 simulation scenarios using a low-fidelity manikin (Supplement 1). Each scenario involved 2–3 students, included debriefing, and utilized a tablet (Laerdal SimPad; Laerdal Medical) for real-time feedback on chest compression quality. The teacher-to-student ratio ranged from 1:16 to 1:20.

In the MR group, 4 students attended a 32-minute course, which included 22 minutes of individual MR training (15 minutes of theoretical BLS instruction) using an HTC Vive Focus 3 headset, followed by 10 minutes of debriefing. The individual MR training included a step-by-step BLS reminder (15 minutes total), including chest compression practice. Students participated in 3 standardized mixed-reality scenarios (varied contexts) with a full field-of-view headset, immersing them in a virtual environment where the physical manikin torso appeared as a lifelike patient. They interacted with virtual characters, performed chest compressions on the physical manikin, and received real-time visual feedback via a device developed by D’un Seul Geste (Fig. 2). The teacher-to-student ratio was 1:4.

Training used simulation manikins: ResusciAnne-QCPR and LittleAnne-QCPR (Laerdal Medical), respectively, for the MK and MR groups. Both groups used the same torso; only the MK group used a full-body manikin.

Participants

Third-year medical students enrolled in the mandatory “Cardiac arrest level 1” course were eligible. They were informed about the potential use of MR and could decline participation or refuse the intervention. Students who opted out received standard BLS simulation training and were excluded from the study. Students absent or physically unable to perform chest compressions were also excluded.

Outcomes

The primary outcome was participants’ overall BLS performance at 1 month post-course because immediate post-training evaluations tend to overestimate skills, whereas a delayed assessment provides a more reliable measure. Secondary outcomes included chest compression depth and rate, percentage of optimal chest compressions, no-flow time, and time to call for a defibrillator. MR device tolerance and learner satisfaction were assessed immediately after training, while confidence in performing cardiopulmonary resuscitation (CPR) was evaluated at 1 month.

Data sources/measurement

The primary outcome was assessed 1 month after the BLS course using the BLS performance score (Fig. 3). The assessor was independent, blinded, and uninvolved in the BLS training. This 10-item scale is based on the validated Cardiff score, modified per the 2021 European Resuscitation Council recommendations [11,12]. It covers BLS steps following the chain of survival, with each item scored as achieved (1) or not achieved (0), totaling 10 points. Each student’s performance was assessed via a standardized simulated cardiac arrest scenario following a standardized protocol and recorded in an electronic Case Report Form using Google Forms (Google LLC). Baseline BLS performance was not assessed, assuming randomization would balance groups.

Chest compression data were collected using the QCPR application (iPad ver. 6.2.3; Apple) connected to ResusciAnne (Laerdal Medical), following current guidelines (depth: 50–60 mm; rate: 100–120/min) [11]. Time metrics were recorded with a stopwatch. After each session, participants completed an online questionnaire assessing MR device tolerability (MR group) and overall course satisfaction (both groups). One month later, before the BLS performance assessment, they rated their confidence in performing BLS via a 5-point Likert scale questionnaire using Google Forms, from 1: “strongly disagree” to 5: “strongly agree” (Dataset 1, Supplement 2).

Bias

The study design carried a risk of loss to follow-up, estimated at 30% in the sample size calculation. Participants assessed at 1 month might have differed in motivation or other factors. Although baseline BLS skills were unknown, random allocation likely balanced group differences. Randomization applied to course dates rather than individuals, but students were randomly assigned to these dates. Satisfaction and confidence were self-reported and possibly influenced by MR novelty, but these were secondary outcomes.

Study size

Sample size was calculated to ensure non-inferiority on the overall CPR performance score, with a maximum acceptable difference of 1 point, requiring a minimum of 69 participants per group (138 total). The 1-point non-inferiority margin was based on prior studies and clinical judgment, as a difference of less than 1 point on a 10-point scale is unlikely to represent a meaningful disadvantage in real-life BLS performance [13]. Using R software (The R Foundation for Statistical Computing) and variance data from Nas et al. [13] (alpha risk 5%, beta risk 20%), reported interquartile ranges were converted to estimate a combined standard deviation (SD) of 2.35, requiring approximately 69 participants per group. Accounting for a 30% dropout rate, 198 subjects were required (99 per group) [13].

Randomization

A randomized sequence of MK or MR course dates was generated in Excel (Microsoft Corp.) by an investigator. A blinded administrator assigned students to each date. Each student received an anonymous identifier to ensure confidentiality. They were informed about the devices used in the study and could decline participation. Those who opted out received standard BLS training and were excluded.

Statistical analysis

Continuous variables are presented as mean±SD. Schuirmann’s one-sided t-test was used for the primary non-inferiority outcome, and Student t-test was used for superiority analysis of secondary outcomes. Categorical variables are presented as counts and percentages, compared using the chi-square or Fisher’s exact tests, depending on sample size. P-values were adjusted for multiple secondary outcome analyses using the Benjamini-Hochberg method [14].

Results

Between December 2023 and May 2024, 225 participants were enrolled: 109 in the MR group and 116 in the MK group (Fig. 4). The dropout rate was 31.1%, with 70 students (25 MR, 45 MK) missing the 1-month assessment. Data from 155 participants were analyzed: 84 (54.2%) in the MR group and 71 (45.8%) in the MK group (Table 1).

Main results

The mean global BLS performance scores were 6.4 out of 10 in the MR group and 6.5 out of 10 in the MK group. The mean difference between the 2 scores was 0.1 (95% confidence interval [CI], –0.45 to +∞; P<0.001) (Table 2). The mean depth of chest compressions was 50.3 mm in the MK group and 46.6 mm in the MR group, representing a difference of 3.7 mm (95% CI, –6.5 to –0.95; adjusted P=0.03). The percentage of chest compressions with the correct depth was 61.2% in the MK group compared to 43.8% in the MR group, representing a difference of 17.4% (95% CI, –29.3 to –5.5; adjusted P=0.02). No significant differences were observed in the mean rate of chest compressions, the percentage of chest compressions with optimal rate, the time to call for a defibrillator, or no-flow time between the 2 groups (Table 2).

Learner satisfaction was rated 4.9/5 in the MR group versus 4.7/5 in the MK group (difference, 0.2 points; 95% CI, 0.07 to 0.33; adjusted P=0.02). In the MK group, 99.1% recommended the training, compared to 92.7% in the MR group, representing a difference of 6.5% (95% CI, 1.3% to 11.7%; adjusted P=0.04). One month post-training, the MR group reported a mean confidence score of 3.9/5 compared to 3.6/5 in the MK group (difference, 0.3 points; 95% CI, 0.11 to 0.58; adjusted P=0.02). No significant differences were found in the other questionnaire items (Table 3). Among the 109 MR participants, 93 (85.3%) reported no symptoms, while 3 (2.8%) reported dizziness and nausea, and another 3 (2.8%) reported headaches and eye pain. Additional outcomes and detailed BLS performance scores are provided in Tables 4 and 5.

Discussion

Key results

This study validates the non-inferiority hypothesis of MR compared to traditional MK for teaching BLS, as evaluated 1-month post-training through a simulated scenario. Chest compression depth was greater and more frequently aligned with guidelines in the MK group [11].

Interpretation

MR was non-inferior to MK for teaching BLS to medical students, as assessed 1 month after training, despite reduced instructional time. This supports MR’s potential as a time-efficient, individualized approach to learning BLS skills. However, overall performance scores remained modest in both groups, consistent with previous studies, and possibly reflecting students’ early training stage and limitations inherent to single-point assessments compared to continuous evaluations [15,16]. This highlights the importance of repeated training to consolidate BLS skills, a challenging goal due to limited time and organizational complexity [1].

Although MR incorporated a real manikin enabling chest compressions (unlike full VR), chest compression quality remained superior in the MK group. This may result from visual distortion or limited tactile feedback in MR, potentially hindering accurate assessment of compression depth. These findings must be considered in context with the study’s primary outcome—overall BLS performance—selected for its clinical relevance and comprehensive reflection of learners’ skills across all resuscitation steps.

The most notable differences between the 2 training methods were instructor-to-student ratio and total instructional time. MR training was significantly shorter (32 minutes vs. 120 minutes), potentially providing a pedagogical advantage through time-efficient learning. Additionally, the lower instructor-to-student ratio in MR (1:4 vs. 1:16–20) may have enhanced learner engagement by providing more personalized instruction and a more motivating learning environment. However, nuance is necessary, as the instructor debrief occurred only during the final minutes in MR sessions, while MK instructors remained with learners throughout.

Finally, satisfaction and confidence were slightly higher in the MR group, possibly reflecting increased engagement due to MR’s immersive nature or favorable ratio. Yet, these effects may be driven by technological novelty rather than true pedagogical benefits, and students may prefer traditional simulation due to its broader range of scenarios [15].

Comparison with previous studies

Although previous studies employed different assessment scales, their findings align with ours, reporting no significant differences in learners’ technical skills [16,17]. Results regarding chest compression quality varied, with no consistent advantage for either modality, likely because compression quality was not a primary outcome in the current study [16,18]. While essential, chest compression quality must be interpreted within the broader context of complete resuscitation sequences [4,11]. This remains an area for improvement but does not challenge overall observed performance.

Consistent with earlier research, confidence in performing CPR was higher in the MR group, and the device was well tolerated, with few adverse events [17]. Increased confidence may facilitate quicker initiation and more effective compressions, potentially improving survival outcomes after out-of-hospital cardiac arrest [19].

Limitations and strengths

First, the higher dropout rate in the MK group may introduce selection bias, as participants completing the 1-month assessment could differ in motivation or other unmeasured factors. Second, although satisfaction and confidence were higher in the MR group, differences were small and may lack educational significance.

Third, while overall performance scores were modest, MR’s non-inferior results despite reduced exposure time suggest instructional efficiency. However, evaluation was limited to a single follow-up scenario, potentially not capturing long-term skill retention. Fourth, despite variations in instructional time and instructor involvement, both training modalities shared identical theoretical and practical content, enabling valid educational effectiveness comparisons. Differences in guidance levels could have influenced learners’ perceptions and outcomes. Rather than limiting, this may indicate instructional efficiency, as MR achieved non-inferior performance despite shorter exposure, and absolute performance was similarly modest in both groups. Additionally, a subset of students from both groups received CPR training in another course before follow-up, potentially confounding results despite balanced distribution.

Generalizability

All participants were third-year medical students, a homogeneous group in training level and BLS skills, limiting results' generalizability to non-healthcare professionals, a relevant target audience. Beyond demonstrating non-inferiority in overall performance, MR offers advantages including learner autonomy, shorter duration, self-assessment, and skill refreshment opportunities. However, scaling this model in typical educational settings poses challenges, primarily due to the required low instructor-to-student ratio, despite shorter duration. Further research should evaluate MR’s time efficiency for instructors and explore broader educational integration. By merging real-world and digital elements, MR creates immersive experiences potentially enhancing learner engagement and motivation, especially among younger learners [7,20].

Suggestions

Further research should optimize MR as a BLS teaching method, particularly in extending its use to the general population.

Conclusion

MR is non-inferior to MK for teaching BLS, as evaluated 1 month after training through a BLS simulation scenario. However, chest compression depth was better in the control group. Despite this, MR offers significant advantages, including personalized learning and reduced training time, while maintaining high participant satisfaction and skill levels. Further research should explore the broader applicability of MR training across different learner populations.

Supplementary materials

Fig. 1.

Study protocol and randomization. M0, after the course assessment; M1, 1-month assessment; MK, manikin group; MR, mixed-reality group; CPR, cardiopulmonary resuscitation; eCRF, electronic Case Report Form.

Fig. 2.

Illustration of the mixed-reality device: torso (1), victim in the immersive reality (2), global virtual scene (3), and chest compression quality feedback (4).

Fig. 3.

Ten-item assessment scale to assess the overall basic life support (BLS) performance at 1-month evaluation. CPR, cardiopulmonary resuscitation; CC, chest compression; AED, automated external defibrillator.

Fig. 4.

Flow chart of participant participation and analysis. MR, mixed-reality group; MK, manikin group.

References

• 1. Greif R, Lockey A, Breckwoldt J, Carmona F, Conaghan P, Kuzovlev A, Pflanzl-Knizacek L, Sari F, Shammet S, Scapigliati A, Turner N, Yeung J, Monsieurs KG. European Resuscitation Council Guidelines 2021: education for resuscitation. Resuscitation 2021;161:388-407. https://doi.org/10.1016/j.resuscitation.2021.02.016 ArticlePubMed
• 2. Greif R, Bhanji F, Bigham BL, Bray J, Breckwoldt J, Cheng A, Duff JP, Gilfoyle E, Hsieh MJ, Iwami T, Lauridsen KG, Lockey AS, Ma MH, Monsieurs KG, Okamoto D, Pellegrino JL, Yeung J, Finn JC, Baldi E, Beck S, Beckers SK, Blewer AL, Boulton A, Cheng-Heng L, Yang CW, Coppola A, Dainty KN, Damjanovic D, Djarv T, Donoghue A, Georgiou M, Gunson I, Krob JL, Kuzovlev A, Ko YC, Leary M, Lin Y, Mancini ME, Matsuyama T, Navarro K, Nehme Z, Orkin AM, Pellis T, Pflanzl-Knizacek L, Pisapia L, Saviani M, Sawyer T, Scapigliati A, Schnaubelt S, Scholefield B, Semeraro F, Shammet S, Smyth MA, Ward A, Zace D. Education, implementation, and teams: 2020 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Resuscitation 2020;156:A188-A239. https://doi.org/10.1016/j.resuscitation.2020.09.014 ArticlePubMed
• 3. Yan S, Gan Y, Jiang N, Wang R, Chen Y, Luo Z, Zong Q, Chen S, Lv C. The global survival rate among adult out-of-hospital cardiac arrest patients who received cardiopulmonary resuscitation: a systematic review and meta-analysis. Crit Care 2020;24:61. https://doi.org/10.1186/s13054-020-2773-2 ArticlePubMedPMC
• 4. Tanaka H, Ong ME, Siddiqui FJ, Ma MH, Kaneko H, Lee KW, Kajino K, Lin CH, Gan HN, Khruekarnchana P, Alsakaf O, Rahman NH, Doctor NE, Assam P, Shin SD; PAROS Clinical Research Network. Modifiable factors associated with survival after out-of-hospital cardiac arrest in the Pan-Asian Resuscitation Outcomes Study. Ann Emerg Med 2018;71:608-617. https://doi.org/10.1016/j.annemergmed.2017.07.484 ArticlePubMed
• 5. Yahya LB, Naciri A, Radid M, Chemsi G. Immersive simulation in nursing and midwifery education: a systematic review. J Educ Eval Health Prof 2024;21:19. https://doi.org/10.3352/jeehp.2024.21.19 ArticlePubMed
• 6. Alcazar Artero PM, Pardo Rios M, Greif R, Ocampo Cervantes AB, Gijon-Nogueron G, Barcala-Furelos R, Aranda-Garcia S, Ramos Petersen L. Efficiency of virtual reality for cardiopulmonary resuscitation training of adult laypersons: a systematic review. Medicine (Baltimore) 2023;102:e32736. https://doi.org/10.1097/MD.0000000000032736 ArticlePubMedPMC
• 7. Barteit S, Lanfermann L, Barnighausen T, Neuhann F, Beiersmann C. Augmented, mixed, and virtual reality-based head-mounted devices for medical education: systematic review. JMIR Serious Games 2021;9:e29080. https://doi.org/10.2196/29080 ArticlePubMedPMC
• 8. Dubreucq E, De La Vega SB, Bouaoud J, Philippon AL, Thiebaud PC. Impact of virtual, augmented or mixed reality in basic life support training: a scoping review. Clin Simul Nurs 2025;99:101672. https://doi.org/10.1016/j.ecns.2024.101672 Article
• 9. Cheng A, Fijacko N, Lockey A, Greif R, Abelairas-Gomez C, Gosak L, Lin Y; Education Implementation Team Task Force of the International Liaison Committee on Resuscitation (ILCOR). Use of augmented and virtual reality in resuscitation training: a systematic review. Resusc Plus 2024;18:100643. https://doi.org/10.1016/j.resplu.2024.100643 ArticlePubMedPMC
• 10. Piaggio G, Elbourne DR, Pocock SJ, Evans SJ, Altman DG; CONSORT Group. Reporting of noninferiority and equivalence randomized trials: extension of the CONSORT 2010 statement. JAMA 2012;308:2594-2604. https://doi.org/10.1001/jama.2012.87802 ArticlePubMed
• 11. Olasveengen TM, Semeraro F, Ristagno G, Castren M, Handley A, Kuzovlev A, Monsieurs KG, Raffay V, Smyth M, Soar J, Svavarsdottir H, Perkins GD. European Resuscitation Council Guidelines 2021: basic life support. Resuscitation 2021;161:98-114. https://doi.org/10.1016/j.resuscitation.2021.02.009 ArticlePubMed
• 12. Whitfield RH, Newcombe RG, Woollard M. Reliability of the Cardiff Test of basic life support and automated external defibrillation version 3.1. Resuscitation 2003;59:291-314. https://doi.org/10.1016/s0300-9572(03)00246-6 ArticlePubMed
• 13. Nas J, Thannhauser J, Vart P, van Geuns RJ, Muijsers HE, Mol JQ, Aarts GW, Konijnenberg LS, Gommans DH, Ahoud-Schoenmakers SG, Vos JL, van Royen N, Bonnes JL, Brouwer MA. Effect of face-to-face vs virtual reality training on cardiopulmonary resuscitation quality: a randomized clinical trial. JAMA Cardiol 2020;5:328-335. https://doi.org/10.1001/jamacardio.2019.4992 ArticlePubMedPMC
• 14. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Methodol 1995;57:289-300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x Article
• 15. Sakr A, Abdullah T. Virtual, augmented reality and learning analytics impact on learners, and educators: a systematic review. Educ Inf Technol 2024;29:19913-19962. https://doi.org/10.1007/s10639-024-12602-5 Article
• 16. Sungur H, van Berlo ZM, Luwa LM. Enhancing cardiopulmonary resuscitation training with mixed reality: improving cardiopulmonary resuscitation performance and enjoyment. Cyberpsychol Behav Soc Netw 2024;27:379-386. https://doi.org/10.1089/cyber.2023.0411 ArticlePubMed
• 17. Giacomini F, Querci L, Dekel BG. Mixed reality versus mass or self-directed training for adolescents’ basic life support instruction: a prospective, randomized pilot study. Open Anesth J 2023;17:e258964582307180. https://doi.org/10.2174/18743218-v17-230822-2023-20 Article
• 18. Rushton MA, Drumm IA, Campion SP, O’Hare JJ. The use of immersive and virtual reality technologies to enable nursing students to experience scenario-based, basic life support training-exploring the impact on confidence and skills. Comput Inform Nurs 2020;38:281-293. https://doi.org/10.1097/CIN.0000000000000608 ArticlePubMed
• 19. Lund-Kordahl I, Mathiassen M, Melau J, Olasveengen TM, Sunde K, Fredriksen K. Relationship between level of CPR training, self-reported skills, and actual manikin test performance: an observational study. Int J Emerg Med 2019;12:2. https://doi.org/10.1186/s12245-018-0220-9 ArticlePubMedPMC
• 20. Madden J, Pandita S, Schuldt JP, Kim B, S Won A, Holmes NG. Ready student one: exploring the predictors of student learning in virtual reality. PLoS One 2020;15:e0229788. https://doi.org/10.1371/journal.pone.0229788 ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by  

• Enhancing virtual reality applications for adult basic life support: insights from a comparative analysis
  Nino Fijačko, Benjamin S. Abella, Vinay M. Nadkarni, Špela Metličar, Anne-Astrid Agten, Robert Greif
  Virtual Reality.2026;[Epub]     CrossRef
• Enhancing STEM and STEAM Education at the Grade Level Through Mixed Reality Applications: A Meta-analytical Study
  Segun Michael Ojetunde, Umesh Ramnarain
  Canadian Journal of Science, Mathematics and Technology Education.2026;[Epub]     CrossRef
• Collocated mixed reality for basic life support training in medical students: a randomised pilot feasibility trial
  David Maiershon, Jonathan Giron, Vered Daitch, Yael Feinstein, Hadas Leibovich-Kot, Shlomit Tovbis, Tohar Levartovsky Ovadiya, Hadar Lavi, Ben Hazan, Maxim Lantsman, Alon Basker, Miri Mizrahi Reuveni
  Resuscitation Plus.2026; : 101383.     CrossRef
• IMPACTOS DO USO DE SIMULAÇÃO IMERSIVA NA CAPACITAÇÃO EM ATENDIMENTO À PARADA CARDIORRESPIRATÓRIA
  Iago Brenner Farias Leal, Izabelly Ferreira de Andrade, Yan Carlos de Sousa Diniz, Lara Maria Ferreira Lopes Valéri Pinto, Maria Helena Vieira Pereira Marques, Francisca Evelyn Abreu de Lira, Thaís Helena Gomes de Sousa, Maria Isabelly Araújo Ferreira, A
  Revista Multidisciplinar do Nordeste Mineiro.2025; 16(1): 1.     CrossRef