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

Accurate triage is essential for efficient emergency care; however, little is known about the comparative effectiveness of interactive lecture (IL), virtual reality (VR), and feedback-based virtual reality (FVR) for Emergency Severity Index (ESI) triage training among nursing students. This study investigated the effects of interactive lecture (IL), VR, and FVR on the precision of triage assessments and on nursing students’ satisfaction with the Emergency Severity Index (ESI).. A quasi-experimental pretest–posttest design with three arms was conducted among sixth-semester nursing students ( n = 64) at Shahid Beheshti University of Medical Sciences in Tehran, Iran. Participants were assigned non-randomly, according to their existing course sections and the scheduling of the training sessions, to IL ( n = 22), VR ( n = 22), or FVR ( n = 20). Triage accuracy was evaluated using a 10-item scenario-based assessment administered before the intervention, immediately after its completion, and one week later. Satisfaction was measured utilizing a validated 18-item questionnaire. Data analysis was performed using repeated-measures ANCOVA and one-way ANOVA, with a significance level set at p < 0.05. Among 64 nursing students, baseline demographic characteristics and GPA were comparable across groups; baseline triage accuracy was highest in the FVR group, and the between-group difference approached significance ( p = 0.055), so all comparisons were adjusted for baseline accuracy. Immediately after the intervention, mean triage accuracy was highest in the FVR group, followed by the VR and IL groups; however, the between-group difference was not statistically significant ( p = 0.328). The overall main effect of time was not statistically significant ( p = 0.391). At the one-week follow-up, accuracy declined in the VR and FVR groups but continued to improve in the IL group, resulting in a significant time × group interaction ( p = 0.028), suggesting different learning trajectories across instructional methods. Satisfaction was high and comparable across all groups ( p = 0.63), indicating similar acceptability. All three instructional methods were favorably regarded by students, and post-intervention triage accuracy did not differ significantly between them at either assessment point. The overall main effect of time was not statistically significant; however, the significant time × group interaction indicates that accuracy followed different trajectories across methods, with continued improvement in the IL group at one week. These findings do not establish the superiority of any single method. VR may serve as a complement to instructor-led teaching rather than a replacement for structured discussion and guided reasoning, and larger studies with longer follow-up are required to confirm these patterns. Not applicable.