JMIR Publications

ABSTRACT

Background:

In nursing care for high-acuity, low-occurrence (HALO) medical devices, a single operational error can have fatal consequences, yet nurses rarely have opportunities to practice; the left ventricular assist device (LVAD) is a representative case requiring a safe environment for repeated practice. Virtual reality (VR) offers such practice opportunities, but its usability as a learning tool must be validated before its educational effectiveness is examined. Usability evaluations to date have largely relied on a general usability perspective, which may miss defects embedded in the clinical context, and validation by raters with domain expertise is rare.

Objective:

This study aimed to identify clinical and safety-related usability defects—those a general usability evaluation is unlikely to capture—by having domain experts with direct LVAD nursing experience evaluate the usability of a VR-based LVAD nursing education program.

Methods:

We applied an explanatory sequential mixed methods design. In the quantitative phase, 30 nurses with LVAD nursing experience completed the VR program and then responded to standardized usability measures (eg, the System Usability Scale [SUS]) and an evaluation based on the metaverse usability heuristics of Omar et al. In the qualitative phase, we conducted in-depth interviews on the stage-by-stage training scenarios with 9 participants selected on the basis of their response patterns. We systematically compared and integrated the quantitative and qualitative results from the two phases.

Results:

Thirty nurses participated in the quantitative evaluation and 9 in the qualitative interviews. The mean SUS score was 65.2 (SD 14.7); 43.3% of participants rated the program Good or better, 40.0% OK, and 16.7% Poor. The overall heuristic score was 3.96 (SD 0.59) out of 5; domains related to physical discomfort (eg, cybersickness) and motion-tracking lag scored lowest with the largest between-rater variation. Qualitative analysis yielded three themes: foundational competence building and psychological preparation through hands-on performance; immersion arising from the reproduced environment alongside sensory and interface limitations; and the position of VR within blended learning. The mixed methods integration identified five integration points with no discordance; notably, the domain experts identified a tactile and auditory dimension of safety confirmation that the standardized measures did not cover.

Conclusions:

A mixed methods evaluation involving raters with clinical domain expertise suggests that such expertise complements a general usability perspective in the usability validation of VR education for HALO medical devices. VR was well suited as a foundational module within blended learning, and this approach may be applied beyond LVAD to the design and evaluation of education for other HALO medical devices. Clinical Trial: This study was reviewed and approved by the Institutional Review Board (IRB) of Samsung Medical Center (Approval No.: 2025-08-104-001).