Introduction: This article focuses on the effectiveness of virtual reality (VR) simulation training for intravenous (IV) cannulation. We analyzed the learning process in skills acquisition from groups with limited experience in comparison with the performance of experts.
Methods: Novices (n = 20) and intermediates (n = 23) trained on nine IV cannulation scenarios with a high-fidelity VR simulator (Virtual IV). Another three scenarios were performed before and after training (pre- and postassessment). Pre-/postassessment was also undertaken by a group of 10 experts for comparative evaluation. Performance metrics included the completion time and an error score that was obtained after quantitative analysis of the errors committed during training. The learning curves were evaluated by recording the number of attempts required to successfully perform each training scenario. A performance was considered successful when the error score was below a predetermined threshold.
Results: The learning curves of intermediates and novices demonstrated a clear plateau at the sixth and eighth scenario, respectively. These plateaus were reached after 15 (intermediates) and 23 (novices) attempts. There was a highly significant reduction in the time and errors between pre- and postassessment (P < 0.01). Before training, the performance (errors, time) of the three groups was significantly different to one another (novices: lowest, experts: highest, P < 0.05). After training, the performance of intermediates and novices was equivalent to that of the experts (P > 0.1). Intermediates committed fewer critical and more noncritical errors than novices, and vice versa (P < 0.05).
Conclusions: Simulation training enhanced the skills of inexperienced subjects significantly. The VR simulator demonstrated construct validity for three different levels of experience. The number of attempts over a series of equal difficulty scenarios provides a valuable alternative to the traditional measures of the learning curve.