The Tailored Evidence-based Enhancements in Mental Health-Gamified and Individualized Follow-Up Treatment for Suicide (TEEM-GIFTS) : Implementing a Gamified mHealth Intervention to Reduce Post-Discharge Suicide Risk in Patients With Mental Disorders Using Multiphase Optimization Strategy (MOST)

The Tailored Evidence-based Enhancements in Mental health (TEEM) projects, adhering to implementation science principles, encompass various studies aimed at improving mental health care in China. Among these is the Gamified and Individualized Follow-Up Treatment for Suicide (GIFTS) study. Based on the multi-phase optimization strategy (MOST), this is a mixed-methods study with three phases. The selection phase involves forming the gamified mHealth suicide risk management strategy and system through community participatory research. The optimization phase involves testing and adjusting the intervention elements for the gamified mHealth strategy through a factorial design randomized controlled trial (RCT) to evaluate the effectiveness of different combinations of intervention elements in reducing suicide risk. The confirmation phase will assess the optimized strategy based on the Implementation Outcomes Framework (IOF), focusing primarily on patient outcomes obtained through a standard RCT.

1. Selection phase Based on literature review, the research team has identified several gamification elements, including avatar, points, badges, missions, level & leaderboards, team mates & team missions, venting space, mood sharing, achievement and titles, and rewards, etc. During developing the gamification mHealth app (TEEM), the investigators categorize the elements into four types of intervention elements as Routine Tracker, Mind Care, Social Connectivity, and Achievement & Rewards; meanwhile, under the community-based participatory research (CBPR) framework, the investigators aim to explore the appropriate content of gamification mHealth interventions, the acceptance and feasibility of such interventions among discharged psychiatric patients, as well as structural and perceptual barriers during implementation.

In specific, the research team will recruit discharged patients with mental disorders and their lay health care supporters (LHSs) who are usually their family members, psychiatrists and nurses, psycho-crisis intervention team members, community mental health workers and mental health social workers as the community team, and will categorize the community team into three sub-groups, the patients-LHSs group, the clinic mental health service provider group (psychiatrists and nurses, and psycho-crisis intervention team members), and the community mental health service provider group (community mental health workers and mental health social workers). The investigators will conduct three focus group interviews in each sub-group and ten to fifteen cases of individual in-depth interview with the community to avoid bias in focus groups and to protect privacy related to personal experience in suicide and suicide intervention. There will be scheduled meetings with the community to discuss and revise the intervention strategy before implementation. Purposive sampling will be applied to recruit participants for the community team. For each type of sub-group, there will be five to eight members 2. Optimization phase The research team will conduct a a factorial design randomized controlled trial (RCT) to evaluate the effectiveness of different combinations of intervention elements in reducing suicide risk. There are four intervention elements as mentioned, thus there will be 16 groups in the factorial design RCT. The research team will recruit discharged patients with mental disorders from Shenzhen Mental Health Center/Shenzhen Kangning Hospital (SKH).

The research team sets the standard for optimizing interven elements as follows: The experimental group implementing the intervention element and the experimental group not implementing the intervention element should have an effect difference in reducing the suicide risk by 35%, which will be used as the basis for calculating the sample size of this study. This study aims to detect a 35% difference (δ) in the implementation of the intervention elements with a test level of α=0.05 and a test power (1-β) of 80%. A total of 260 patients need to be recruited for this study. Assuming a loss to follow-up rate of 20% and considering the number of experimental groups, the final sample size is 320 patients.

In accordance with the factorial design, the optimization phase will consist of 16 groups, and this is due to the four types of intervention elements and their 16 possible intervention conditions. Thus, after recruitment and the baseline survey, participants will be assigned into Group 1 to Group 16 by simple randomization in R program, with 20 participants in each group. Further, participants in each group will only have access to a single intervention condition of elements assigned to their group. Noted, the intervention app TEEM will be developed during the selection phase, and there will be an Android/iOS version of TEEM and a mini-program in the WeChat platform. Participants, LHSs, nurses who perform recruitment and baseline survey, and investigators who perform follow-ups will be blinded to the assignment.