The study, published in BMC Nursing, enrolled 64 sixth-semester nursing students and assigned them non-randomly to one of three training arms: a conventional interactive lecture, a virtual reality simulation, or a feedback-based version of the virtual reality training. The researchers chose this population deliberately, since triage is typically learned on the job and mistakes made by novices can cascade into delayed care for the sickest patients. The Emergency Severity Index requires trainees to synthesize vital signs, presenting complaints, and resource predictions into a rapid category assignment, a cognitive task that blends algorithmic reasoning with clinical intuition.

Each group received instruction tailored to its format. The interactive lecture group worked through the ESI algorithm with an instructor, engaging in structured discussion and guided reasoning about patient scenarios. The virtual reality groups navigated simulated emergency cases in a digital environment, with one arm receiving explicit feedback on their triage decisions and the other experiencing the simulation without that added layer. The design allowed the team to isolate two separate questions at once: whether immersive simulation beats classroom teaching, and whether embedding feedback inside the virtual environment adds measurable value on top of immersion alone.

To measure the outcome, the researchers administered a ten-item scenario-based assessment of triage accuracy at three time points: before the intervention, immediately after it, and again one week later. Satisfaction was captured with a validated eighteen-item questionnaire. The statistical machinery behind the analysis was correspondingly careful. Because the three groups were not randomized, the team used repeated-measures analysis of covariance, adjusting for baseline differences, alongside one-way analysis of variance for the satisfaction scores, with significance set at the conventional threshold of p less than 0.05.

The baseline picture already hinted at the complexity to come. Demographic characteristics and grade point averages were comparable across the three groups, but baseline triage accuracy was highest in the feedback-based virtual reality group, a difference that approached statistical significance at p equal to 0.055. Rather than ignore this imbalance, the investigators adjusted all subsequent comparisons for baseline accuracy, a methodological decision that guards against mistaking pre-existing skill differences for genuine effects of the teaching method. It is a reminder that in non-randomized educational research, the groups you compare are never quite the groups you wish they were.

Immediately after the intervention, the feedback-based virtual reality group posted the highest mean triage accuracy, followed by the standard virtual reality group and then the interactive lecture group. Yet the between-group difference at that point was not statistically significant, with p equal to 0.328. Nor was the overall main effect of time significant across the cohort, at p equal to 0.391. In plain terms, none of the three methods could claim to have produced reliably better triage decisions than the others in the immediate aftermath of training, a result that cuts against the widespread assumption that immersive technology automatically translates into superior learning outcomes.

The most intriguing finding emerged at the one-week follow-up, where the trajectories of the three groups visibly diverged. Accuracy declined in both virtual reality groups but continued to improve among students taught by interactive lecture, producing a statistically significant time-by-group interaction at p equal to 0.028. This pattern suggests that the instructional methods shaped not just how much students learned but how their learning evolved over time. One plausible interpretation, which the authors themselves raise, is that the structured discussion and guided reasoning inherent to interactive lecturing may consolidate decision-making frameworks more durably, while simulation-based learning may produce an initial surge of performance that fades without reinforcement.

Satisfaction, by contrast, was uniformly high and statistically indistinguishable across all three groups, with p equal to 0.63. Students enjoyed the virtual reality experience just as much as the lecture, and vice versa. This matters for educators weighing investment decisions, because it means acceptability is not the bottleneck. The bottleneck is efficacy, and on that measure the study is refreshingly honest: the authors explicitly state that their findings do not establish the superiority of any single method, and they caution that virtual reality may serve best as a complement to instructor-led teaching rather than a replacement for structured discussion and guided reasoning.

The study’s limitations deserve as much attention as its findings. The quasi-experimental design, with participants assigned according to existing course sections and scheduling rather than by randomization, leaves open the possibility of residual confounding despite the statistical adjustments. The sample of 64 students, split into groups of 22, 22, and 20, is modest, and small samples limit the power to detect real differences that may exist. The one-week follow-up window is short by the standards of skill retention research, and triage competence in a real emergency department involves pressures, interruptions, and team dynamics that no classroom assessment fully reproduces. The authors call for larger studies with longer follow-up to confirm the divergent learning trajectories they observed.

Even so, the research lands at a moment when medical and nursing schools worldwide are pouring resources into simulation technology, often on the premise that newer means better. This trial offers a more nuanced message: immersion is not a substitute for the cognitive scaffolding that a skilled instructor provides, and the durability of learning may depend less on the delivery medium than on the quality of reasoning practice it affords. For emergency departments struggling to standardize triage training, and for educators designing curricula under budget constraints, the takeaway is pragmatic rather than revolutionary. Virtual reality can earn its place in the toolkit, but the evidence here suggests it should sit alongside the lecture hall, not instead of it, while researchers continue the harder work of figuring out which combination produces clinicians who sort the sickest patients first, every time.

Subject of Research: Comparative effectiveness of virtual reality and lecture-based Emergency Severity Index triage training for nursing students

Article Title: Effects of virtual reality, feedback-based virtual reality, and interactive lecture–based ESI triage training on triage accuracy and nursing students’ satisfaction: a quasi-experimental study

Article References: Saeedpour, M., Pazokian, M., KhabazKhoob, M., Emadi, N., & Moosavi, S. (2026). Effects of virtual reality, feedback-based virtual reality, and interactive lecture–based ESI triage training on triage accuracy and nursing students’ satisfaction: a quasi-experimental study. BMC Nursing. https://doi.org/10.1186/s12912-026-05450-w

Image Credits: AI Generated

DOI: 10.1186/s12912-026-05450-w

Keywords: virtual reality, triage training, Emergency Severity Index, nursing education, nursing students, simulation, interactive lecture, triage accuracy, quasi-experimental study, clinical education, feedback-based learning, BMC Nursing