Exploring virtual worlds for scenario-based repeated team training of cardiopulmonary resuscitation in medical students.

Background: Contemporary learning technologies, such as massively multiplayer virtual worlds (MMVW), create new means for teaching and training. However, knowledge about the effectiveness of such training is incomplete, and there are no data regarding how students experience it. Cardiopulmonary resuscitation (CPR) is a field within medicine in high demand for new and effective training modalities.

Objective: In addition to finding a feasible way to implement CPR training, our aim was to investigate how a serious game setting in a virtual world using avatars would influence medical students' subjective experiences as well as their retention of knowledge.

Methods: An MMVW was refined and used in a study to train 12 medical students in CPR in 3-person teams in a repeated fashion 6 months apart. An exit questionnaire solicited reflections over their experiences. As the subjects trained in 4 CPR scenarios, measurements of self-efficacy, concentration, and mental strain were made in addition to measuring knowledge. Engagement modes and coping strategies were also studied. Parametric and nonparametric statistical analyses were carried out according to distribution of the data.

Results: The majority of the subjects reported that they had enjoyed the training, had found it to be suitable, and had learned something new, although several asked for more difficult and complex scenarios as well as a richer virtual environment. The mean values for knowledge dropped during the 6 months from 8.0/10 to 6.25/10 (P = .002). Self-efficacy increased from before to after each of the two training sessions, from 5.9/7 to 6.5/7 (P = .01) after the first and from 6.0/7 to 6.7/7 (P = .03) after the second. The mean perceived concentration value increased from 54.2/100 to 66.6/100 (P = .006), and in general the mental strain was found to be low to moderate (mean = 2.6/10).

Conclusions: Using scenario-based virtual world team training with avatars to train medical students in multi-person CPR was feasible and showed promising results. Although we found no evidence of stimulated recall of CPR procedures in our test-retest study, the subjects were enthusiastic and reported increased concentration during the training. We also found that subjects' self-efficacy had increased after the training. Despite the need for further studies, these findings imply several possible uses of MMVW technology for future emergency medical training.