Immersive virtual reality versus traditional instruction for long-term retention of cardiopulmonary resuscitation skills in medical student education: A randomized controlled trial

ERCT Criteria Breakdown

• Level 1 Criteria

  • C

    Class-level RCT

    • Randomisation was performed at the individual student level within a single college, not at the class or school level, and no tutoring exception applies.
    • "Participants were randomly assigned in a 1:1 ratio to either the VR group (n = 64) or the Control group (n = 64). The random sequence was generated using a computer-based random number generator."
    • Relevant Quotes: 1) "A total of 128 undergraduate medical students with no prior professional first-aid certification were recruited. Participants were randomly assigned in a 1:1 ratio to either the VR group (n = 64) or the Control group (n = 64). The random sequence was generated using a computer-based random number generator." (p. 2) 2) "To strictly maintain allocation concealment, sequentially numbered, opaque, sealed envelopes were utilized. These envelopes were opened by an independent researcher exclusively after a participant had completed all baseline assessments..." (p. 2) 3) "Due to the 8:1 learner-to-instructor ratio and the use of shared mannequins, the 30-min practice window inherently involved rotation, observation, and waiting." (p. 3) Detailed Analysis: Criterion C requires randomisation at the class level (or stronger school level), unless the intervention is one-to-one/personal tutoring, in which case student- level randomisation is acceptable. Here the unit of randomisation was clearly the individual student: participants were "randomly assigned in a 1:1 ratio" via a computer-based random number generator and individual sealed envelopes, all within a single institution (Qujing Medical College). This is student-level, not class-level or school-level, randomisation, creating a contamination risk the criterion is designed to prevent. The intervention is a group-delivered training programme (control taught at an 8:1 learner-to-instructor ratio), not one-to-one tutoring, so the tutoring exception does not apply. Criterion C is not met because randomisation was conducted at the individual student level within a single institution rather than at the class or school level.
  • E

    Exam-based Assessment

    • Outcomes were measured with a custom multiple-choice knowledge test and a study-adapted checklist with bespoke scoring, not an externally administered standardised exam.
    • "A 20-item multiple-choice questionnaire was administered before and immediately after training to measure knowledge acquisition."
    • Relevant Quotes: 1) "Procedural proficiency was evaluated using a highly structured, 20-item checklist directly adapted from the official AHA 2020 BLS Skills Testing performance criteria. Content validity was established a priori by an expert panel of three senior emergency medicine physicians." (p. 3) 2) "Each item was scored dichotomously (0 points for incorrect/missed execution; 5 points for correct execution), resulting in a cumulative continuous score ranging from 0 to 100." (p. 3) 3) "A 20-item multiple-choice questionnaire was administered before and immediately after training to measure knowledge acquisition." (p. 4) Detailed Analysis: Criterion E requires a standardised, widely recognised exam-based assessment that was not specially designed for the study. The theoretical outcome was a 20-item multiple-choice questionnaire assembled for the study, i.e., a custom instrument. The procedural outcome was a 20-item checklist that, while "directly adapted from the official AHA 2020 BLS Skills Testing performance criteria," was modified with a bespoke dichotomous 0/5 scoring scheme and had its content validity re-established by the authors' own expert panel. Rather than administering a recognised standardised exam in its official form, the authors constructed study-specific instruments. Although anchored to AHA guidance, these function as researcher-assembled measures for this study. Criterion E is not met because the outcomes relied on a custom multiple-choice test and a study-adapted, re-scored checklist rather than an externally administered standardised examination.
  • T

    Term Duration

    • The final outcome was measured only one month after a single training session, far short of a full academic term (~3-4 months).
    • "Assessments were conducted at two distinct time points: immediately post-training (immediate) and 1 month later (1 month retention)."
    • Relevant Quotes: 1) "Participants engaged in a single, continuous 45-min session using Meta Quest 2 head-mounted displays..." (p. 3) 2) "Participants received 75 min of total instruction: a 45-min traditional lecture and live demonstration... followed by a 30-min practice session..." (p. 3) 3) "Assessments were conducted at two distinct time points: immediately post-training (immediate) and 1 month later (1 month retention)." (p. 4) 4) "although the 1-month follow-up period successfully captured the initial trajectory of skill decay, it remains insufficient to determine long-term retention." (p. 8) Detailed Analysis: Criterion T requires that outcomes be measured at least one full academic term (typically ~3-4 months) after the intervention begins; short interventions are acceptable only if term-long follow-up tracking is present. The intervention here was a single training session, and the latest outcome (the primary outcome) was measured at a 1-month retention test. One month is shorter than a single academic term, and the authors themselves note that the 1-month follow-up is "insufficient to determine long-term retention." Criterion T is not met because the interval from intervention to final measurement was only one month, less than one academic term.
  • D

    Documented Control Group

    • The control group's size, demographics, baseline performance, and business-as-usual conditions are clearly documented.
    • "Participants received 75 min of total instruction: a 45-min traditional lecture and live demonstration by a certified instructor, followed by a 30-min practice session using standard CPR mannequins."
    • Relevant Quotes: 1) "there were no statistically significant differences between the Control group and the VR group in terms of age (20.27 +/- 1.07 vs. 20.47 +/- 1.08 years; p = 0.307) or gender distribution (p = 0.858)... baseline theoretical knowledge and operational capabilities... were comparable between the two groups (77.83 +/- 6.93 vs. 78.45 +/- 4.69; p = 0.557)." (p. 4) 2) "Table 1. Baseline characteristics of participants... Control group (n = 64)... Age 20.27 +/- 1.07... Baseline assessment score 77.83 +/- 6.93." (Table 1, p. 5) 3) "Participants received 75 min of total instruction: a 45-min traditional lecture and live demonstration by a certified instructor, followed by a 30-min practice session using standard CPR mannequins." (p. 3) Detailed Analysis: Criterion D requires clear documentation of the control group's characteristics, size, baseline performance, and conditions. The paper documents the control group as n = 64, reports its age and gender distribution and baseline assessment scores in Table 1, confirms baseline equivalence with the intervention group, and describes precisely what the control group received (75 min of traditional lecture, demonstration, and mannequin practice with instructor feedback). This allows a proper comparison at baseline. Criterion D is met because the control group's size, demographics, baseline scores, and instructional conditions are fully documented.

• Level 2 Criteria

  • S

    School-level RCT

    • Randomisation was at the individual student level in a single college, not at the school level.
    • "Participants were randomly assigned in a 1:1 ratio to either the VR group (n = 64) or the Control group (n = 64)."
    • Relevant Quotes: 1) "This randomized controlled trial was conducted at the Qujing Medical College. A total of 128 undergraduate medical students... were recruited. Participants were randomly assigned in a 1:1 ratio to either the VR group (n = 64) or the Control group (n = 64)." (p. 2) 2) "The random sequence was generated using a computer- based random number generator." (p. 2) Detailed Analysis: Criterion S requires randomisation at the school level, i.e., entire educational institutions or implementation units randomly assigned to conditions. In this study the randomisation unit was the individual student, and the entire trial took place within a single institution (Qujing Medical College). No schools, sites, or units were randomised. Criterion S is not met because randomisation occurred at the individual student level within one institution rather than at the school level.
  • I

    Independent Conduct

    • The intervention tools were third-party products (the AATE VR CPR Simulator software and AHA guideline content), the outcome was scored by blinded independent assessors, and the authors declared no conflicts of interest.
    • "all evaluations were conducted by two independent, AHA-certified BLS instructors who were strictly blinded to the participants' group allocation." (p. 3)
    • Relevant Quotes: 1) "The intervention was delivered via the 'CPR Simulator' software (AATE VR), which was explicitly programmed according to the 2020 American Heart Association (AHA) BLS guidelines." (p. 3) 2) "the VR group employed standalone consumer-grade VR head-mounted displays (e.g., Meta Quest 2) with a custom CPR application." (p. 2) 3) "To minimize subjective assessment bias, all evaluations were conducted by two independent, AHA-certified BLS instructors who were strictly blinded to the participants' group allocation... a third senior, blinded adjudicator evaluated the performance to establish the final score." (pp. 3-4) 4) "The authors declare no conflicts of interest." (p. 9) Detailed Analysis: Criterion I requires the study to be conducted independently from the designers of the intervention, to reduce bias in implementation, measurement, and analysis. The intervention here was not authored by the research team: the VR training used a third-party commercial software product (the AATE VR "CPR Simulator") and content derived from the American Heart Association's BLS guidelines. The developers of these intervention tools did not participate in the trial's data collection, analysis, or conclusions. Mirroring the standard's example where an intervention provider does not participate in the evaluation, the developers of the VR software and CPR protocol are independent of the evaluation. In addition, the primary procedural outcome was scored by two independent, AHA-certified assessors strictly blinded to group allocation, with a blinded adjudicator for discrepancies, and the authors declare no conflicts of interest, with funding from provincial education grants rather than the technology vendor. Criterion I is met because the intervention tools were designed by third parties (AATE VR and the AHA), the outcome was assessed by blinded independent raters, and no conflicts of interest were declared.
  • Y

    Year Duration

    • Outcomes were measured only one month after a single session, far below 75% of an academic year, and criterion T is not met.
    • "Assessments were conducted at two distinct time points: immediately post-training (immediate) and 1 month later (1 month retention)."
    • Relevant Quotes: 1) "Assessments were conducted at two distinct time points: immediately post-training (immediate) and 1 month later (1 month retention)." (p. 4) 2) "although the 1-month follow-up period successfully captured the initial trajectory of skill decay, it remains insufficient to determine long-term retention... The absence of an extended follow-up limits our ability to evaluate the ultimate durability of VR-based training." (p. 8) Detailed Analysis: Criterion Y requires that outcomes be tracked for at least 75% of a full academic year (~9-10 months). The longest interval here was a 1-month retention test. Additionally, per the standard, if criterion T (Term Duration) is not met then Y cannot be met, and T is not met in this study. The authors explicitly acknowledge the follow-up was too short to judge durability. Criterion Y is not met because tracking lasted only one month, well short of a full academic year, and the term- duration prerequisite is unmet.
  • B

    Balanced Control Group

    • The delivery modality itself is the treatment variable; practice repetitions and content were matched (control had more clock time), so no confounding extra resources favoured the intervention.
    • "the pedagogical architectures of both groups were operationally equivalent in providing standardized theoretical input, real-time corrective feedback, and identical task repetitions, isolating the delivery modality as the primary independent variable."
    • Relevant Quotes: 1) "To rigorously evaluate the pedagogical effect of the VR medium, the intervention and control conditions were strictly matched in terms of practical repetitions, feedback architecture, and true active engagement, despite inherent differences in their absolute clock times." (p. 3) 2) "the pedagogical architectures of both groups were operationally equivalent in providing standardized theoretical input, real-time corrective feedback, and identical task repetitions, isolating the delivery modality as the primary independent variable." (p. 3) 3) "Participants received 75 min of total instruction..." (control) vs. "Participants engaged in a single, continuous 45-min session..." (VR); each required "a minimum of three full [simulated] resuscitation cycles." (p. 3) 4) "Both groups received identical content based on American Heart Association guidelines." (Abstract) Detailed Analysis: Criterion B asks whether any extra time/budget given to the intervention group was matched by the control group, unless the additional resource is itself the treatment variable. Comparing resources: both arms delivered the same AHA-based content and required an identical practice density (a minimum of three full resuscitation cycles). In terms of clock time, the control group actually received more (75 min vs. 45 min for VR), so the intervention group was not advantaged on instructional time or practice repetitions. The principal difference is the delivery modality (immersive VR with algorithmic multimodal feedback vs. traditional demonstration), which the authors explicitly frame as "the primary independent variable." The VR device and its objective feedback are integral to, and inseparable from, the intervention being tested; the control condition represents the appropriate business-as-usual comparator. There is thus no separable, unmatched extra resource that confounds the contrast. Criterion B is met because practice repetitions and content were matched (with the control receiving more clock time), and the VR delivery modality itself, not an add-on resource, is the explicit treatment variable.

• Level 3 Criteria

  • R

    Reproduced

    • This specific 2026 trial has not been independently replicated; cited VR CPR trials are separate studies, not replications of it.
    • "This resilience aligns with findings from recent randomized controlled trials indicating that VR training is noninferior, and often superior, to traditional methods in skill retention..."
    • Relevant Quotes: 1) "This resilience aligns with findings from recent randomized controlled trials indicating that VR training is noninferior, and often superior, to traditional methods in skill retention-particularly regarding compression depth and frequency..." (p. 7) 2) "Issleib M, Geiger A... Virtual reality as a training method for basic life support of the medical staff on general wards: a randomized controlled non-inferiority trial. Virtual Real. 2025..." (Reference 10) 3) "Nas J, Thannhauser J, Vart P, et al. Effect of face- to-face vs virtual reality training on cardiopulmonary resuscitation quality: a randomized clinical trial. JAMA Cardiol. 2020..." (Reference 7) Detailed Analysis: Criterion R requires that this specific study (or its central claim in the same context and design) be independently replicated by a different team in a peer- reviewed outlet. This is a newly published 2026 trial. The other VR CPR RCTs it cites (e.g., Nas et al. 2020, Issleib et al. 2025) are independent studies with different designs and populations, not replications of this particular Qujing single-session VR-versus- traditional 1-month-retention trial. An internet search (July 2026) returned only such unrelated VR CPR trials and systematic reviews, with no study reproducing this specific trial. No independent reproduction of this specific study was identified. Criterion R is not met because no independent replication of this specific study exists; the cited trials are distinct studies rather than reproductions.
  • A

    All-subject Exams

    • Only CPR knowledge and skill were assessed (a single specialised domain), and criterion E is not met, so the all-subject requirement fails.
    • "Outcomes included theoretical knowledge (multiple-choice questionnaire) and procedural skills (checklist)..."
    • Relevant Quotes: 1) "Outcomes included theoretical knowledge (multiple- choice questionnaire) and procedural skills (checklist), assessed immediately post-training and at a 1-month follow-up." (Abstract) 2) "The prespecified primary outcome for this study was the procedural skill score evaluated at the 1-month retention test." (p. 3) Detailed Analysis: Criterion A requires assessment of all main subjects using standardised exam-based assessments, and it explicitly depends on criterion E being met. Here only a single, highly specialised competency (CPR/basic life support) was measured, via a custom knowledge test and a study-adapted checklist. No other subjects were assessed, and criterion E (standardised exam-based assessment) is not met, which alone precludes A. Criterion A is not met because only one specialised domain was measured and the prerequisite criterion E is not satisfied.
  • G

    Graduation Tracking

    • Follow-up ended at one month with no tracking to graduation, and criterion Y is not met.
    • "although the 1-month follow-up period successfully captured the initial trajectory of skill decay, it remains insufficient to determine long-term retention."
    • Relevant Quotes: 1) "Assessments were conducted at two distinct time points: immediately post-training (immediate) and 1 month later (1 month retention)." (p. 4) 2) "although the 1-month follow-up period successfully captured the initial trajectory of skill decay, it remains insufficient to determine long-term retention... Future multicenter trials with extended follow-up periods (e.g., 6 months)... are warranted." (p. 8) Detailed Analysis: Criterion G requires tracking participants through to graduation from their educational stage, and it depends on criterion Y being met. Follow-up here stopped at one month after training; there is no tracking of students toward graduation. An internet search (July 2026) for subsequent publications by these authors tracking this cohort to graduation returned none. Criterion Y is also not met, which independently precludes G. Criterion G is not met because measurement ended at one month with no graduation tracking, and the year-duration prerequisite is unmet.
  • P

    Pre-Registered

    • No trial registry entry or pre-registered protocol is reported; only ethics approval is mentioned.
    • Relevant Quotes: 1) "The study protocol was reviewed and approved by the Ethics Committee of the Center for Laboratory and Practical Training, Qujing University of Medicine & Health Sciences (Approval No. L20240131)." (p. 4) 2) "The sample size was determined a priori... The prespecified primary outcome for this study was the procedural skill score evaluated at the 1-month retention test." (pp. 3-4) Detailed Analysis: Criterion P requires a publicly pre-registered protocol (e.g., ClinicalTrials.gov or ISRCTN) posted before data collection, including hypotheses, methods, and planned analyses, with a verifiable registration date. The paper reports ethics-committee approval and states the primary outcome and sample size were prespecified, but it provides no trial-registry name, registration ID, or registration date. A search of the Chinese Clinical Trial Register (ChiCTR) and ClinicalTrials.gov (July 2026) found no registration for this trial. Internal a priori specification and ethics approval do not constitute public pre-registration. Criterion P is not met because no public trial registration or pre-registered protocol with a verifiable pre-data-collection date is reported.