This was a prospective, randomized controlled parallel group trial in a tertiary university hospital in Thailand. It was approved by the Khon Kaen University institutional committee for ethics in human research and registered with the Thai Clinical Trials Registry [HE611533]. Written informed consent was obtained from each participant prior to enrollment.

Participants were fifth-year medical students at the Khon Kaen University Faculty of Medicine who attended the ALS training course offered by the faculty’s CPR unit in April 2019. Eligibility criteria included not having participated in any type of ALS course in the past and never having played Resus Days. After enrollment, participants were randomly assigned to one of two groups (control group or game group) at a 1:1 allocation ratio based on a computer-generated random number list [26].

The sample size for analysis of covariance of two levels and two covariates was determined based on our pilot data using power analysis. We hypothesized that the difference in knowledge between the two groups was approximately 5%, and that there would be moderate consistency between pretest and posttest scores. The power analysis was determined using an alpha of .05 and power of 0.80. This resulted in an estimated desired effect sample size of at least 22 subjects in each group.

Prior to any intervention, we tested students’ baseline ALS knowledge using a multiple-choice pretest. Students in the game group then played Resus Days using their own smartphone for 1 hour in normal mode. After this, students in both groups attended an ALS training course taught by CPR instructors. Students in the game group were allowed to play Resus Days as much as they wished during the 2-day ALS training course.

Resus Days was developed by Rath Panyowat. It is a mobile simulation game that allows health care professionals to practice CPR through repetitive playing. Players must resuscitate a simulated patient in various scenarios by choosing treatment methods based on the ALS algorithm until the patient’s cardiac rhythm becomes normal. There are two game modes: normal and physician. In normal mode, a player receives hints regarding diagnosis and treatment and gains points if the prescribed treatment is correct. In physician mode, there are no hints and points are subtracted if the prescribed treatment is incorrect. There are a total of 7 scenarios including cardiac arrest, bradycardia, tachycardia, and several simulated megacode (mixed algorithm) scenarios. To win in each scenario, the player must strictly follow the ALS algorithm within the time allotted (Figure 1).

The ALS training course was held over 2 days. The first day of the program consisted of approximately 6 hours of lectures, and the second day consisted of approximately 8 hours of hands-on workshops, skill station activities, simulation-based training with high-fidelity ALS simulator mannequins with the SimPad system (Laerdal Medical). The student-to-instructor ratio was 1:6. At the end of the course, student ALS knowledge was reassessed using multiple-choice questions in accordance with ALS guidelines, and their practical skills were evaluated individually by examiners blinded to group allocation. We used a questionnaire to assess student attitudes and perceptions regarding both the course and the game.

Students received a score from 0 (if the patient died) to 3 (if patient survived and the most appropriate treatments were prescribed) for each scenario. The highest possible overall score was 21 points (see Multimedia Appendix 4).

Categorical variables were expressed as frequencies and percentages. Continuous data were expressed as means and standard deviations. Differences between groups in terms of baseline characteristics were compared using an independent sample t test. An analysis of covariance model was used to compare posttest scores between the two groups adjusted for baseline score measurements. The Pearson chi-square test was used to compare the skill-test pass rate and other binary variables. The correlation between game scores before ALS training and knowledge pretest scores and between game scores post-ALS training and knowledge posttest scores were analyzed using Pearson correlation coefficients. All data analyses were performed using Stata version 10 (StataCorp LLC).

ALS certification is mandatory for all medical students entering their clinical years. This study demonstrates the merits of using a serious video game to augment ALS knowledge, especially regarding the ALS algorithm. In addition, the game group had a higher pass rate on the ALS skill test and indicated greater satisfaction with their training.

Previous studies have also examine the effectiveness of games in helping students practice their resuscitation skills. Creutzfeldt et al [18], for example, conducted a study that evaluated the efficacy of game technology as part of a pre-BLS training program for medical students, which led to improved BLS knowledge and performance. These improvements were associated with adherence to the guidelines, which resulted in students performing chest compressions with the appropriate frequency. However, that study had a limited sample size, and its findings required confirmation. Our findings are consistent with those of Creutzfeldt et al [18], but we had a larger sample size and applied game-based learning to training in ALS, which has a more complicated algorithm. The game group exhibited clear improvement with regard to ALS algorithm knowledge. They also demonstrated improvements in terms of skill, but not to a statistically significant extent.

Several studies have reported beneficial effects of game-based training on medical student learning outcomes [18,21-23]. A large number of the medical students in our study had positive attitudes about Resus Days, indicating that the game helped them become more familiar with various cardiac arrest scenarios prior to the ALS training course. These findings are in line with those of a study by Cheng et al [22], which found that game-based learning helped learners develop a holistic understanding of scientific concepts. Furthermore, some students stated that playing the game helped them memorize the CPR algorithm and medications. This may be due to improvements in cognitive function as a result of repetitive playing. Step-by-step game playing, in which scenarios become increasingly complicated, has been shown to stimulate critical thinking and increase engagement [18,21].

Another interesting finding was the low correlation between student game scores and their knowledge scores. This can be explained by the fact that some students may not have been familiar with the game mechanics at first, resulting in them getting low scores despite being knowledgeable about ALS and having good pretest or posttest scores. Because of this, we do not recommend using a player’s game score as a substitute for traditional testing. However, regardless of players’ scores, their knowledge was significantly improved by playing the game. This is partially because the game provided the correct answers at the end of each scenario even if the player failed.

Several methods have been proposed to prepare trainees before attending ALS training courses. We found that engaging in game-based preparation prior to precourse ALS lectures (consisting of precourse algorithm rehearsal and lectures accompanied by Microsoft PowerPoint slides or lecture notes) was more beneficial than conventional precourse lectures alone. Although it was only an add-on intervention, the observed benefits demonstrated the ability of gamified training to improve a complicated medical course.

Most of the studies that have demonstrated the benefits of game-based learning have been conducted in high-income countries [20]. This study, however, was conducted in a low-middle income country and found that, even in this setting, game-based learning was practical and effective. In the near future, this may be used to help train large numbers of medical students or personnel in remote areas who have little or no access to traditional classroom-based lectures. However, if game-based learning is to replace conventional ALS lectures, a randomized controlled trial using a well-designed game may be necessary.

One strength of our study is that it is the first to demonstrate the effectiveness of ALS pretraining using a smartphone game and addresses the feasibility of gamified learning in training students to treat highly complicated medical conditions. Second, we used a game that is available in an online store and therefore available to anyone with access to a compatible device. Although previous studies have developed their own games as tools for research [15,17,19], the initial design of such a study is costly and time consuming. Moreover, using a custom game specifically developed for a particular study would limit its generalizability. Third, as our study was a randomized controlled trial, differences in baseline characteristics between students in the game and control groups were limited. Finally, we performed a comparison of gamified learning with a standard ALS training course that was provided by certified CPR providers and that met international standards.

However, our study also had some potential limitations. First, this was a single-center study, which limits the generalizability of the results. Second, we evaluated students based on their knowledge scores, which are imperfect indicators of how someone will perform in real-world scenarios. Third, the sample size was only large enough to evaluate student knowledge scores and not their results on the ALS skill test. Fourth, most of the participants in this study were medical students who were familiar with smartphone games. It is unclear how effective this kind of gamified learning would be in an older population such as senior doctors. Fifth, the students in the game group were assigned to play Resus Days for 1 hour before attending the lecture and were allowed to play as much as they wished during the training course; however, we did not have specific protocol and did not collect any data regarding the frequency of gameplay during the 2-day training. Finally, the content in this game was developed by an individual physician based on standard guidelines, not by a content expert or trusted organization.

Our results demonstrate that incorporating a serious smartphone game (Resus Days) in medical students’ ALS training leads to better algorithm knowledge scores than conventional training alone. In addition, students in the game group indicated high satisfaction with their training. Although there was no significant difference between the two groups in terms of their pass rates on the ALS skills test, those who trained using the serious game tended to do better.