Effectiveness of augmented reality-based education on hospital fire safety knowledge, attitudes, and behaviors: a single-blind randomised controlled trial

Background Hospital fire safety education is essential for patient safety and emergency preparedness. This study evaluated whether augmented reality (AR)-based education, which offers more contextual and interactive learning than traditional education, improved fire safety knowledge, attitudes, behaviors, and acceptance among hospitalized patients and accompanying caregivers. Methods This study was a single-blind, parallel-group randomised controlled trial with a 1:1 allocation ratio. A total of 163 hospitalized patients and accompanying caregivers were enrolled and randomly allocated to either the AR-based education group (n = 82) or the traditional education group (n = 81) using a computer-generated sequence concealed in sequentially numbered, opaque, sealed envelopes. Of these, 153 participants completed the study and were included in the effectiveness analysis, including 76 in the AR group and 77 in the traditional education group. The AR group received tablet-based fire safety education using a hospital-developed AR platform, whereas the control group received standard hospital fire safety education. Fire safety knowledge, attitudes, and behaviors were assessed before the intervention and before discharge. Acceptance of the educational method, including perceived ease of use and perceived usefulness, was assessed after the intervention. Generalized estimating equations were used to examine group, time, and group-by-time interaction effects. Results Both groups showed improvements in fire safety knowledge, attitudes, and behaviors after the intervention. The GEE analysis showed significant group-by-time interaction effects for knowledge (β = 1.338, 95% CI: 0.834–1.842, p  < 0.001) and behaviors (β = 3.964, 95% CI: 2.638–5.289, p  < 0.001), indicating greater improvements in the AR group than in the traditional education group. The group-by-time interaction for attitudes was not statistically significant (β = 1.140, 95% CI: -0.174 to 2.454, p  = 0.089). Participants in the AR group reported higher perceived ease of use and perceived usefulness than those in the traditional education group. Conclusions AR-based education was more effective than traditional education in improving hospital fire safety knowledge and behaviors among hospitalized patients and accompanying caregivers. Although attitudes improved after the intervention, the between-group effect was not statistically significant. The AR approach was also associated with higher acceptance, supporting its potential use as a patient education strategy for hospital fire safety preparedness. Trial Registration ClinicalTrials.gov, NCT07732959, registered on July 20, 2026 (retrospectively registered).