A Mobile Game (Safe City) Designed to Promote Children’s Safety Knowledge and Behaviors: Protocol for a Randomized Controlled Trial

Background Children have high levels of curiosity and eagerness to explore. This makes them more vulnerable to danger and hazards, and they thus have a higher risk of injury. Safety education such as teaching safety rules and tips is vital to prevent children from injuries. Although game-based approaches have the potential to capture children’s attention and sustain their interest in learning, whether these new instructional approaches are more effective than traditional approaches in delivering safety messages to children remains uncertain. Objective The aim of this study is to test the effectiveness of a game-based intervention in promoting safety knowledge and behaviors among Hong Kong school children in Grades 4-6. It will also examine the potential effect of the game-based intervention on these children’s functioning and psychosocial difficulties. Methods This study comprises the development of a city-based role-playing game Safe City, where players are immersed as safety inspectors to prevent dangerous situations and promote safety behavior in a virtual city environment. The usability and acceptability tests will be conducted with children in Grades 4-6 who will trial the gameplay on a mobile phone. Adjustments will be made based on their feedback. A 4-week randomized controlled trial with children studying in Grades 4-6 in Hong Kong elementary schools will be conducted to assess the effectiveness of the Safe City game–based intervention. In this trial, 504 children will play Safe City, and 504 children will receive traditional instructional materials (electronic and printed safety information). The evaluation will be conducted using both child self-report and parent proxy-report data. Specifically, child safety knowledge and behaviors will be assessed by a questionnaire involving items on knowledge and behaviors, respectively, for home safety, road safety, and sport-related safety; child functioning will be assessed by PedsQL Generic Core Scales; and psychosocial difficulties will be assessed by the Strength and Difficulties Questionnaire. These questionnaires will be administered at 3 time points: before, 1 month, and 3 months after the intervention. Game usage statistics will also be reviewed. Results This project was funded in September 2019. The design and development of the Safe City game are currently under way. Recruitment and data collection will begin from September 2020 and will continue up to March 1, 2021. Full analysis will be conducted after the end of the data collection period. Conclusions If the Safe City game is found to be an effective tool to deliver safety education, it could be used to promote safety in children in the community and upgraded to incorporate more health-related topics to support education and empowerment for the larger public. Trial Registration ClinicalTrials.gov NCT04096196; https://clinicaltrials.gov/ct2/show/NCT04096196 International Registered Report Identifier (IRRID) PRR1-10.2196/17756

because of their advantages on risk, acceptance and effectiveness over previous street side/field training methods, game-based VR training requires expensive production and can be difficult for young children as they may have not yet learned the safe ways to navigate environments alone (4). To overcome these VR game limitations, another popular genre of video and computer games is roleplaying games (RPG) which gives the game developer more flexibility in designing developmentally appropriate adventure story with attractive characters and solvable challenges to engage the young participants. Like VR game, RPG allows players to strategize and interacts with objects and resources, but instead of training players in realistic simulated hazard situations, RPG players create their own avatar/character customized with unique attributes, skills and traits to play and advance in the cyber world. Furthermore, RPG can be combined with social game elements such as chat and items exchange to foster collaboration and communication among players (5). Some of these games are referred as massively multiplayer online role-playing game (MMORPG). RPGs or MMORPGs can facilitate learning by increasing student motivation and critical thinking and problem-solving skills (6, 7). Preliminary evidence showed that MMORPGs can lead to greater improvement in English communicative skills than traditional teaching method (8). In a recently published pilot RCT on HIV prevention, young adolescents participating in the smartphone game-based intervention with RPG techniques showed a larger gain in sexual health-related knowledge and self-efficacy compared with the control group (9). However, there is a lack of robust scientific evidence on RPG for injury prevention, highlighting the need to conduct this proposed project which will be able to demonstrate whether RPG can be used as a health promotion tool to prevent injury and potentially other health risk behaviors in Hong Kong.

GRB(4): How to avoid the problem of contamination within schools?
A: Thank you for pointing out the problem of potential contamination between intervention and control group. Since the main intervention is delivered through a web/app-based platform, we will provide unique login credentials for students randomized into the intervention group. Such unique login credentials will be required to access the game and participants would be advised not to share and exchange such information with their peers. This information has been added in the revised proposal (clean version, P.6). Under this scheme, we are confident that contamination would be minimized.
To evaluate the degree of contamination, we will also ask in the questionnaire whether the students in the control group accessed the game through other peers' login credentials. Sensitivity per protocol analysis will be conducted using this information. Meanwhile, we would continue to seek funding for upgrading the game to serve other health education purposes and for use in multiple platforms.

GRB(6): An ethics approval from a recognized ethics committee is required A:
We are in process of applying for the ethics approval. The scanned copy of the ethics approval will be submitted to the GRB by email as specified in the decision letter.

GRB(7)
: Clarify if the following budget item: "Incentives for participants" ($9,200) should be re-named as "Award for participants".

A:
Thank you very much for your suggestion. The following item "Incentives for participants" have been re-named as "Award for participants".

GRB(8):
Provide justification and breakdown for the following budget item: "Game design and development" ($780,000).

A:
Thank you very much for your comments. Justification and breakdown of the budget item "Game design and development" has been added, as below: Game programming: $270,000 A game programmer will be hired for 12 months to develop codebase for the proposed game.

Gaming component designing: $280,000
A game designer will be hired for 12 months to design the game content including creating goals, rules, challenges, game point allocation, level system that could produce a desirable gaming environment for the users to interact and engage in the game.

Graphics or Art design: $200,000
A game artist will be hired for 10 months to create 2D art for the visual elements of a video game, such as characters, vehicles, props, scenery, background, objects, colors, textures, and clothing.

Game sound design: $30,000
A sound designer will be hired at the final stage of the game design to generate and adjust audio elements for the game.

R1(1): Promoting children's safety knowledge and behaviors via a game-based approach is novel. The proposed work is highly relevant to the thematic priority "Injury prevention". A:
Thank you very much for your comments.

R1(2):
The aims and objectives are clearly defined. They are well informed by scientific evidence.
As the applicants mentioned, there are quite limited number of digital safety games on the market. But it is useful to provide more evidence support for using the digital games for enhancing health-related knowledge and behaviors (other than safety).

A:
Thank you very much for your comments. We have elaborated the scientific evidence on digital games in section (b) (ii) and (iii).

R1(3a): The implementation plan is detailed and appropriate for the project.
Sample size has been adequately justified.

The proposed timeline is feasible. A:
Thanks for your comments.

R1(3b):
The only concern is the potential problem of contamination. The same school has students in both intervention and control group. How to ensure that contamination won't happen? A: Thank you for pointing out the problem of potential contamination between intervention and control group. Since the main intervention is delivered through a web/app-based platform, we will provide unique login credentials for students randomized into the intervention group. Such unique login credentials will be required to access the game and participants would be advised not to shared and exchange such information with their peers. This information has been added in the revised proposal (clean version, P.6). Under this scheme, we are confident that contaminations would be minimized. To evaluate the degree of contamination, we will also ask in the questionnaire whether the students in the control group accessed the game through other peers' login credentials. Sensitivity per protocol analysis will be conducted using this information.

R1(4): The indicators are clearly defined and outcome evaluation tools are appropriate for the purposes A:
Thanks for your comments R1(5): The team has a strong track record in health promotion for children and youth.

The team consists of cross-sector collaborations among academic institutions and NGO.
A: Thanks for your comments.

R1(6):
It was mentioned that the preliminary design for the game has already been completed at the time of the application. Need to clarify whether the budget requested is to cover the remaining work. A: Thanks for your comments. The preliminary design of the game, including the main goals, mode of gaming, and target audience of the game, was developed by the project team during the application process. However, a professional game development team with expertise in different areas of game development including game programming, game design, graphics designs, and audio design is needed and essential for producing an attractive, playable and immersive game.

R1(7):
The developed digital game could be used after the completion of this project. Upon completion of this project, flyers about the Safe City game will be made and distributed in seminars and workshops as well as sent to all local schools, news media, education groups and safety organizations to publicize the game and increase its usage in the community. Information about the game will also be posted in social networking websites (e.g. Facebook). Safe City evaluation results will be shared in various meetings, conferences and events to raise public awareness and usage of the game. This information has been added in the revised proposal (clean version, P.11).
Meanwhile, we would continue to seek funding for upgrading the game to serve other health education purposes and for use in multiple platforms.

R1(8a): Strengths: A well written proposal and a strong team
A: Thank you very much for your comments.

A:
As described above, we will minimize contamination through unique login credentials for each student in the intervention group and will evaluate the effect of continuation in sensitivity per protocol analysis.

A:
The sustainability plan has been strengthened by adding "Upon completion of this project, flyers about the Safe City game will be made and distributed in seminars and workshops as well as sent to all local schools, news media, education groups and safety organizations to publicize the game and increase its usage in the community. Information about the game will also be posted in social networking websites (e.g. Facebook). Safe City evaluation results will be shared in various meetings, conferences and events to raise public awareness and usage of the game" in the revised proposal (clean version, P. 11).

R2(1):
The project is innovative in using a digital game based experiential learning approach to promote child safety knowledge and behaviors, addressing injury prevention thematic priority needs among children in HK, rather than the traditional safety programs in formal education settings lacking practice and interactive elements to enhance student learning motivation and behavioral changes.

A:
Thank you very much for your comments.

R2(2):
Aim and objectives were well articulated. The proposed strategy was well supported by related scientific evidence.

A:
Thank you very much for your comments.

R2(3):
The target group is over 1,000 P4-6 children. The implementation plan and milestones are well designed including the content of the city game in achieving its stated objectives.

A:
Thank you very much for your comments.

R2(4):
RCT evaluation plan well designed with control group and well selected indicators, evaluating the 5 specific objectives of the project, comparing the effectiveness of the project approach with traditional strategy in changing children's safety knowledge and behaviours, enhancing their health-related quality, and reducing their psychosocial difficulties A: Thank you very much for your comments.

Reviewer 3 R3(1a):
The investigators argued that traditional health education approaches are usually classroom-based and use one-way method to deliver the information from the instructor to the audience, however, this is too subjective and lack of literature support. School-based and parent oriented approaches are still the most effective way for children in learning. I don't pretty agree with the team and will still believe that the best way to learn safety is to have it be integrated to the school system and in the family. There does not exist any conflicts with parent-child collaborative decision making as cited in (iii) A: Thank

R3(4):
In such a similar training game design study, outcomes can be divided into 2 types: task-specific and domain-specific outcomes. The team identifies task-specific game outcomes as primary outcomes, and domain knowledge such as safety knowledge and clients' self-perceived reflections such as HRQOL and psychosocial difficulties as secondary outcomes.
The secondary outcomes are appropriate but the investigators fail to make sure of the task-specific outcomes.
The task-specific outcomes should be the data obtained from the game which provides big data from more than one thousand cases and the difference in responses and the interactions in different safety domains such as unsafe knowledge, unsafe behaviours, unsafe environments, unsafe management, etc. Other parameters such as correct and incorrect responses, preservation responses, cues level, reaction time, total time of game, total points of game, etc. have not been mentioned in the evaluation. The outcome can also be either reducing the likelihood and consequence of accidents, or both.

R3(5):
The research team consists of experts in various fields including injury, child health, etc. but not in game design.

A:
We understand the reviewer's concern in our game design expertise, but we would like to highlight our experience in game design for health promotion.