This initiative aims to improve flourishing and quality of life of secondary school students, reduce mental distress (e.g., depression and suicidal ideation), enhance their understanding of mental health (e.g., mental health literacy) and help-seeking intention, and foster a supportive school environment (e.g., school climate-caring relationship, and sense of community). Also, this initiative aims to improve students' process of change in psychological (e.g., mattering, emotion regulation, empowerment) and social (e.g., trust belief) aspects and mental health awareness (e.g., mental health stigma). The feasibility, acceptability, and sustainability of the programme from multiple perspectives (e.g., students, student leaders, and stakeholders) will also be evaluated. In addition, the cost-effectiveness of delivering this programme (e.g., the incremental cost-effectiveness ratio (ICER)) among secondary schools in Hong Kong will be assessed.

The programme will be implemented among students in 130 local secondary schools over three academic years. The first is a pilot phase (Year 1), which 40 schools will implement the intervention and student participants will be evaluated at pre- (T0) and post-intervention (T1) using questionnaires. In this stage, participatory research will be conducted before and after the intervention among students, student leaders, and stakeholders in 20 pilot schools to co-design the intervention, ensuring the programme meet the actual wellness needs of youth. In following two academic years, an additional 90 schools will participate in a cluster randomized controlled trial (RCT) with a 1:1 ratio between intervention and waitlist control groups. Each year, 45 schools will implement the intervention. Summative evaluation will be conducted among RCT schools at T0 and T1, and 3-month follow-up (T2). Quantitative data be collected to assess the effectiveness of intervention, and qualitative data will provide understanding of students' and stakeholders' perspectives of the intervention implementation. Cost outcomes will include intervention costs and cost savings, calculated from the payer (i.e., JC/government) perspective using administrative records or validated tools. The primary outcome of cost-effectiveness will be the quality-adjusted life-years (QALYs) of students. Cost of implementing the intervention program and QALYs will be used to evaluate the cost-effectiveness of the intervention, for example, estimate the incremental cost-effectiveness ratio (ICER).

1. Background

1. Improving mental wellness of adolescents in Hong Kong has important and far-reaching significance Mental health issues have significant impact on children and adolescents worldwide, particularly with a young onset. Among these issues, anxiety and depression are the most reported and are among the top ten causes of disability-adjusted life-years for individuals aged 10-24 years (Collaborators, 2020; Fusar-Poli, 2019; Kessler et al., 2005). The prevalence of anxiety and depression has been steadily increasing over the past few decades. In Hong Kong, a study involving 9,518 secondary school students revealed a moderate to severe levels of depression, as measured by the Center for Epidemiological Studies-Depression (CES-D) (Wu et al., 2016). Additionally, another study conducted among 3,136 secondary school students in Hong Kong found that 54.3% and 65,8% of both males and females scored above the cut-off for mild depression, as assessed by the CES-D (She et al., 2021). Specifically, 54.3% of males and 65.8% of females surpassed the cut-off. It is crucial to address these challenges and provide appropriate resources and interventions to promote mental well-being among children and adolescents.

   The significance of mental health problems among adolescents cannot be overstated. Adolescence is a critical period of development, marked by numerous physical, emotional, and social changes. Mental health problems during this stage can have long-lasting effects on the individual's overall well-being and future trajectory. These issues not only impact the affected adolescents themselves but also have far-reaching consequences for their families, communities, and society. Extensive evidence has shown that anxiety and depression are associated with significant adverse consequences among adolescents, such as substance abuse, poor physical health, underachievement in schools, harmful social outcomes, and subsequent depression in later life, (Johnson et al., 2018; Ranasinghe et al., 2016). Moreover, they may induce long-term effects throughout life, and even affect the mental health of offspring (Avenevoli & Merikangas, 2006; Clayborne et al., 2019; Collishaw et al., 2016; Vismara et al., 2022). Recognizing and addressing mental health problems among adolescents is vital to ensure their healthy and successful transition into adulthood and to foster a resilient and thriving society.

   More importantly, mental health issues increase risk of suicide which is a serious public health issue. Suicide-related behaviors are common among school-aged adolescents. According to the World Health Organization, suicide is the fourth leading cause of young people aged 15-29 (World Health Organization, 2023). In Hong Kong, the suicide rate among those aged 15-24 rose to 12.2 deaths per 100,000 people in 2022, compared with 6.2 per 100,000 in 2014 (South China Morning Post, 2023). It is believed that the suicide rate is under-reported in many countries due to the inferior death classification system, and the cultural and religious beliefs that may affect individual's views towards suicide (Beautrais et al., 2006). Losing a young life not only results in a significant societal loss but also inflicts immense psychological suffering upon their families (Goldsmith et al., 2002). Moreover, the unfortunate act of suicide can have a copycat effect, particularly when sensationalized by the media, which is especially prevalent in Asian countries (Chen et al., 2010). Interventions that promotes mental health among adolescents are of utmost importance, seeking to provide support and assistance to those at risk, ultimately saving lives and mitigating the devastating impact on individuals, families, and communities.

2. The need for youth-centered and school-based initiatives for youth's mental wellness Promoting mental health among adolescents is a significant challenge as many adolescents who have mental health problems are disinclined to seek help (Platell et al., 2020). Despite the availability of mental health program in the community, they are difficult to reach those at-risk youth to provide resources and support (Bradby et al., 2007; French et al., 2003). Numerous barriers to mental health support utilization are identified and include the lack of awareness of mental health issues (French et al., 2003), fear of stigma (Bradby et al., 2007), reliance of self-coping (Burgess et al., 2020), perceived ineffectiveness of mental health services (Platell et al., 2020), and fear of the possible consequences of the loss of privacy (Bradby et al., 2007). Additionally, there have also been critics that existing mental health programs are disconnected from their lived experience and fail to address their specific needs, which can lead to disengagement and lack of trust (Georgiadis et al., 2020). To increase acceptability of mental health services, it is important that adolescent mental health programs should be youth-centered. By prioritizing the perspectives, experiences, and needs of adolescents, mental health programs can be tailored and relevant to their unique challenges and circumstances, promoting better engagement and participation (Georgiadis et al., 2020). Also, actively involving adolescents in the planning, design, and implementation of mental health initiatives can promote empowerment them and give them a sense of ownership over their own well-being (Freire et al., 2022). By recognizing the agency of adolescents and involving them in decision-making processes, we can create a more comprehensive and holistic approach to adolescent mental health that addresses their specific needs, ultimately leading to better outcomes and improved overall well-being (Freire et al., 2022). Furthermore, school-based programme is recommended for adolescents as it can provide an easy ongoing access to them (Kern et al., 2017). As adolescents spend most of their time in the school, school-based programme is considered one of the most effective ways to promote mental health and help-seeking among adolescents (Kern et al., 2017).

2. Objectives

1. To evaluate the effectiveness of the multi-faceted programme in promoting flourishing, quality of life, and understanding of mental health (e.g., mental health literacy) and help-seeking intention, reducing mental distress (e.g., depressive symptoms and suicidal ideation and attempt), and fostering positive social environments (e.g., social support, school climate-caring relationship, and sense of community) among secondary school students in Hong Kong.
2. To evaluate the students' process of change in psychological (e.g., mattering, self-efficacy, resilience, emotion-regulation, empowerment, and stress reduction) and social (e.g., trust belief and engagement) aspects and mental health awareness (e.g., mental health stigma and help-seeking attitudes) throughout the programme.
3. To evaluate the feasibility, acceptability, and sustainability of the programme implementation among local secondary schools, including understanding the experiences and perspectives of students and stakeholders using a mixed-method approach.
4. To evaluate the cost-effectiveness (e.g., the incremental cost-effectiveness ratio (ICER)) of delivering the programme among secondary schools in Hong Kong.

3. Study design, randomization, and sample size planning The programme will employ a parallel cluster randomized controlled trial (RCT) alongside a quasi-experimental design, with secondary schools as the unit of allocation and individuals as the unit of analysis. A total of 130 secondary schools will be involved, comprising 40 pilot schools and 90 schools participating in the RCT. For the pilot schools, the quasi-experimental design will be conducted for pre- (T0) and post-intervention (T1) assessment with no control group. The RCT schools will be randomly assigned to the intervention or waitlist control conditions (ratio 1:1).

The randomization process of the cluster RCT will be conducted based on the geographical banding or districts of the schools, ensuring that each banding encompasses schools with similar demographic and socio-economic profiles. Schools within each band or district will be further stratified prior to randomization, based on factors like school size, student demographics, and resource availability, to ensure comparability. After that, a comprehensive list of schools within each band or district will then be randomly assigned to either the intervention or waitlist control group, using a computer-generated sequence. During the intervention, participants and program facilitators will not be blind to the allocation status due to the nature of the intervention. The RCT schools will be assessed at pre- (T0), post-intervention (T1), and 3-month follow-up (T2). The waitlist condition group (N=45) will implement the intervention after the T2 evaluation and be assessed another round of evaluation after implementing the intervention.

To estimate the number of RCT schools, students' flourishing will be the primary outcome of this programme. Based on Schalkwyk et al.'s well-being intervention for improving adolescents' flourishing, after the intervention, the mean difference of flourishing (measured by MHC-SF, as a continuous variable) between two groups was around 3 (SD 15). Assuming an intra-cluster correlation coefficient (ICC) is 0.07 [0.05-0.10] for flourishing outcome, which has been estimated for the mental well-being indicators among adolescents. If it is estimated that 200 students each school participate in the programme, to ensure 80% power and a 5% significance level, the number of school clusters will be 29 for each arm. Therefore, at least 58 schools are needed in the RCT phase and 26,000 students are roughly reached during the pilot and RCT phase.

4. Target groups Local junior students in Form 1 to 3 from130 designated secondary schools in Hong Kong. The inclusion criteria include 1.) Chinese student in Hong Kong secondary school, 2.) Studying at Form 1 to Form 3 at the time of recruitment, 3.) Competence in comprehending written Chinese, and 4.) Competence in speaking Cantonese. Meanwhile, the exclusion criteria include 1.) Non-Chinese student, 2.) Not studying at Form 1 to 3 in Hong Kong secondary school, 3.) Incompetence in comprehending written Chinese, and 4.) Incompetence in speaking Cantonese.

5. Recruitment, data collection, and ethical approval To ensure comprehensive coverage and diversity in the study sample, 130 secondary schools across the 18 districts of Hong Kong will be invited to the programme. The programme will be conducted in two phases spanning across 3 academic years. The first academic year serves as a pilot phase involving 40 secondary schools based on school geographical banding, where the intervention will be initially implemented and evaluated. The pilot phase includes pre- (T0) and post-intervention (T1) evaluations using questionnaires, alongside the participatory research with students, student leaders, and stakeholders among 20 schools. The pilot phase helps to test the intervention and evaluation are practical and applicable in real-world setting, and the participatory research gather and incorporate multiple perspectives to develop and improve the intervention tailored to youth's actual needs.

Following the pilot phase, the programme will incorporate a cluster RCT design with 90 secondary schools, which will be randomly assigned to either an intervention group or waitlist control group, with 45 schools in each. The intervention group will implement the intervention in the second academic year; evaluation will be conducted at schools in both groups at T0, T1, and 3-month follow-up (T2). After the follow-up evaluation, in the third academic year, the intervention will be extended to the waitlist control schools, ensuring equitable access to the intervention while maintaining the integrity of the RCT design; another round survey will be conducted to evaluate the intervention among these schools at T1 as well. In addition, qualitative interviews for formative evaluation will be conducted among students, student leaders, and stakeholders at 20 RCT schools (10 in the intervention arm and 10 in the waitlist control arm) to explore deeper insights into their perspectives for the intervention implementation

6. Implementation and evaluation components

The pilot phase involves 40 secondary schools across different districts in Hong Kong in the intervention implementation and evaluation, and 20 pilot schools will join the participatory research. This phase serves significant purposes: first, it allows for the development and improvement of the intervention through participatory research. By involving students, student leaders, and stakeholders in co-designing the tailored intervention, the programme aims to meet the specific needs and context of each school. This collaborative approach can foster a sense of ownership among participants, which is critical for its successful adoption and sustainability. Second, the pre- and post-evaluation will provide support on the initial efficacy and feasibility of the intervention. It will enable us to identify any logistical challenges, resource requirements, and potential barriers to implementation that may not be apparent without a real-world trial. The assessment is implicative for adjusting the implementation before scaling up to RCT phase. Third, in addition to evaluate the overall effectiveness of the intervention, a series of subgroup analyses will also be conducted to understand the differential effects of the intervention across different contexts. The findings may help us tailor our programme to diverse student needs and enhance the intervention's overall effectiveness. Specially, here are several aspects of the subgroup analyses:

1. School banding/districts, reflecting varying socio-economic and educational environments. This will help understand how the intervention performs across different contexts;
2. Other school contextual factors, such as school size and available resources. This will help understand how the diverse characteristics of schools affect the intervention's outcomes;
3. Participants' baseline mental health status. This will inform us about the intervention effectiveness and suitability for students with different mental health needs;
4. Completion rate of the intervention, including students' engagement in the WARP-based self-management sessions and wellness clubs' activities. This will help optimize the participation strategies in following implementation; The pilot phase is important, which ensure the intervention plan is both effective and adaptable to the needs of diverse school communities in Hong Kong. The evaluation of the whole programme includes participatory research, summative evaluation, formative evaluation, and cost evaluation.

The outcome measures include students' mental well-being outcomes (e.g., flourishing, depression, anxiety, and suicidal ideation and attempt), quality of life, understanding of mental health (e.g., mental health literacy) and help-seeking intention, social environment (e.g., social support, school climate-caring relationship, and sense of community). The measures about process of change will involve psychological (e.g., mattering, self-efficacy, resilience, emotion regulation, empowerment, and perceived stress) and social (trust belief and engagement) process and mental health awareness (e.g., mental health stigma and help-seeking attitudes).

Data collection will be using questionnaires, with the details as below:

i) Outcome measures Mental outcomes

1. Flourishing Students' psychological wellbeing is measured by the Mental Health Continuum-Short Form (MHC-SF) (Keyes, 2006; Palmer et al., 2023). It is a 14 items validated measure for Chinese adolescent (Guo et al., 2015). Each item is scored from 0 (Never) to 5 (every day). Among the 14 items, 3 are emotional well-being questions and 11 are positive functioning questions. Adolescents are considered to be "flourishing" if they score 4 or above to at least 1 of the emotional questions and to at least 6 of the positive functioning questions; considered to be "languishing" if they score 1 or below to 1 or more of the 3 emotional questions and to 6 or more of the 11 positive functioning questions; considered to have "moderate mental health" if they are neither flourishing nor languishing.

2. Depression and Anxiety Students' depression and anxiety will be assessed with the 4-item Patient Health Questionnaire-4 (PHQ-4), an ultra-brief self-report questionnaire that consists of a 2-item anxiety scale (GAD-2) and a 2-item depression scale (PHQ-2) (Kroenke et al., 2009; Löwe et al., 2010). The PHQ-2 consists of the two core criteria for depressive disorders, with the following two items determining depression: "Feeling down, depressed or hopeless" and "Little interest or pleasure in doing things". The two items of the GAD-2 assess the core criteria for generalized anxiety disorders: "Feeling nervous, anxious, or on edge" and "Not being able to stop or control worrying". This tool has been validated and used in adolescent populations (Wörfel et al., 2016). This 4-item survey consists of two parts naming depression (two items) and anxiety (two items) in which each item is graded on a 4-point Likert scale scoring from 0 (Not at all) to 3 (Nearly every day). The total score of each part is 6. Scale scores of ≥3 was suggested as cut-off points between the normal range and probable cases of depression or anxiety respectively (Kroenke et al., 2003, Kroenke et al., 2007, Löwe et al., 2005). PHQ-4 scores go from normal (0 -2), mild (3-5), moderate (6-8) to severe (9 -12).

3. Suicidal ideation and attempt Suicidal ideation and attempt will be assessed by 2 questions from the PHQ-A, "Has there been a time in the past month when you have had serious thoughts about ending your life?" and "Have you ever, in your whole life, tried to kill yourself or made a suicide attempt?" (Johnson et al., 2002)

   Social environment

4. Social support A 3-item subscale of the Comprehensive Inventory of Thriving (CIT) will be employed to measure the level of social support perceived from the environment for adolescents (Su et al., 2014). Each item of the subscale is graded on a 5-point Likert scale scoring from 1 (Strongly disagreed) to 5 (Strongly agreed). The three questions are 'There are people I can depend on to help me', 'There are people who give me support and encouragement' and 'There are people who appreciate me as a person'. Higher score indicates higher level of perceived social support.

5. School climate-caring relationship (Availability of trusted person) The availability of caring persons for students will be measured by three items of caring relationship measure from the California Healthy Kids Resilience Module (Ozer & Schotland, 2011). Items include presence of an adult who "cares about me", "notices when I am not there", "who listens to me when I have something to say", each item are rating on a 4-point scale.

6. Sense of community Sense of community (SOC) will be assessed by the 8-item Brief sense of community scale (BSCS) for adolescents. Each item of the scale is graded on a 5-point Likert type scoring from 1 (Strongly disagreed) to 5 (Strongly agreed). Four domains of the scales include Needs Fulfilment (items 1 and 2), Group Membership (items 3 and 4), Influence (items 5 and 6), and Emotional Connection (items 7 and 8). The score of each domain is calculated by taking the mean of the items (Peterson, et al., 2008). Higher score indicates higher level of perceived sense of community.

   Understanding of mental health & Help seeking intention

7. Mental health literacy Mental health literacy will be measured by 9 items from the Mental Health Literacy Scale (O'Connor & Casey, 2015). The scale assesses individuals' understanding on the various aspect of mental health. In the present study, 9 items that are related to understanding of depression, anxiety, and professional help-seeking will be selected. Items are rated on a 4-point Likert Scale from 1 (very unlikely) to 4 (very likely).

8. Help seeking intention The willingness to seek help if they encounter mental health problems is assessed with the General Help-Seeking Questionnaire (Rickwood et al., 2005). It asks how likely it for students is to seek help from a list of people (such as their peers, sisters or brothers in schools) for personal or emotion problems, from 1 (Extremely unlikely) to 7 (Extremely likely).

Quality of life It will be assessed using two validated instruments designed for adolescents: the EQ-5D Youth Version (EQ-5D-Y) and the Child Health Utility 9D (CHU9D) Index.

EQ-5D-Y is a generic, child-friendly self-complete instrument measuring HRQoL in children and adolescents aged 8 to 15 years. Its design is based on the EQ-5D-3L instrument, which was developed to measure HRQoL in adults. It evaluates five dimensions of health-related quality of life: mobility, self-care, participation in usual activities, pain/discomfort, and anxiety/depression. Each dimension of EQ-5D-Y is measured across three levels of severity, from "no problems" to "having a lot of problems" (Wille et al., 2010). This instrument has been validated and used among adolescent population in Hong Kong (Lin et al., 2022), however, its value set has not yet been derived in the local context.

In addition to EQ-5D-Y, the CHU9D has been validated for its sensitivity in capturing the effectiveness of health interventions among children and adolescents, with age range from 7 to 17 years (Rowen et al., 2021). It encompasses nine dimensions: worried, sad, pain, tired, annoyed, schoolwork/homework, sleep, daily routine, and activities, each with five-level responses (Stevens, 2012). The Chinese version of CHU9D has also been validated in a school-based sample in mainland China (Yang et al., 2018). The instrument has not been validated among Hong Kong adolescents, and we can examine its reliability and validity in this evaluation.

In this evaluation, the dual approach allows us to capture a broader health dimensions relevant to the adolescent population, and both instruments will be used to estimate the quality-adjusted life-years (QALYs) of secondary students for evaluating the programme's effectiveness. Responses from the CHU9D will be transformed into QALYs weights derived from a UK general population sample using an algorithm developed by Stevens (Stevens, 2012), producing a utility value set of between 0.33 (worst health state) and 1 (best health state), and a utility score of zero denotes death. In addition, the utility weights of EQ-5D-Y will be generated and validation in this study before QALYs evaluation, using valuation methods: discrete choice experiment (DCE) and composite time trade-off (cTTO) (Roudijk et al., 2022). The DCE tasks required participants to choose between two distinct health states described by the EQ-5D-Y-3L, identified as options 'A' and 'B'. a standard 10-year TTO approach was used for states deemed better than death, and a lead-time variant (which involves living 10 years in full health followed by 10 years in an EQ-5D-Y-3L health state) was applied for states considered worse than death (Janssen et al., 2013). At the respondent's point of indifference (t), the utilities for the EQ-5D-Y-3L health states will be calculated as follows:

U = t/10 for better-than-dead states U = (t - 10)/10 for worse-than-dead states The possible range of observed utilities was - 1 to 1.

ii) Measures on process of change Psychological process

1. Mattering Mattering refers to the perception of importance, significance and being valued by others. It is measured by the General Mattering Scale (GMS) (Liu et al., 2023). It is a 5-items scales that asks questions such as "How important do you feel you are to other people?" and "How much do other people depend on you?". Each item is rated from 1 (Not at all) to 4 (A lot).

2. Hope Hope is defined as an indicator of psychological resilience and positive future orientation. The 6-item Snyder's State Hope Scale will be used here to assess how participants hold the belief in own capacity to initiate and sustain actions (agency) as well as the belief in own capacity to generate routes (pathways) to reach their goal (Snyder et al., 1996). Each item of the scale is graded on an 8-point Likert type scoring from 1 (Definitely false) to 8 (Definitely true). Two domains of the scales include Agency (items 2, 4 and 6), and pathways (items 1, 3 and 5). Subscale scores for agency or pathways are derived by adding the three even- and odd-numbered items, and the total score of State Hope Scale is the sum of all six items.

3. Resilience The two-item version of the Connor-Davidson Resilience Scale (CD-RISC2) (Ni et al., 2016) will be used to measure the ability to cope with stress and adverse effects of traumatic events.

4. Emotion Regulation The use of emotion regulation strategies: cognitive reappraisal and expressive suppression is measured with the Emotion Regulation Questionnaire (ERQ-9) (Spaapen et al., 2014). It is a 9-items questionnaire rated from 1 (Strongly disagree) to 7 (Strongly agree). Example statements are "I control my emotions by not expressing them." "When I'm faced with a stressful situation, I make myself think about it in a way that helps me stay calm."

5. Empowerment The Self-subscale of the Youth Empowerment Scale-Mental Health (YES-MH) (Konstantopoulou et al., 2024; Walker et al., 2009) consists of 7 items used to measure the extent to which students are encouraged to take an active role in managing their mental health. Sample items included "I know how to take care of my mental or emotional health".

6. Perceived Stress Stress level is measured by the 4-items Perceived Stress Scale (PSS-4) (Liu & Zhang, 2023; Vallejo et al., 2018). It consists of items such as "How often have you felt that you were unable to control the important things in your life?" and "How often have you felt that things were going your way?" and is rated with a point scale ranging from 1 to 5.

   Social process

7. Trust belief Trust belief will be measured by the General Trust Questionnaire (Yamagishi & Yamagishi, 1994). It contains 6 items that measure individuals' beliefs about honesty and trustworthiness of others. In the present study, participants will be asked to evaluate the items with references to the people in their school. Items are rated on a 5-point Likert Scale, from 1 (strongly disagree) to 5 (strongly agree). Sample items are "Most people in the school are trustworthy."

8. Engagement Students' engagement in the program will be measured by the Tiffany-Eckenrode Program Participation Scale (TEPPS) (Tiffany et al., 2012). The "Voice/Influence" 4-items subscale is selected, each item is rated from 1 (Not at all) to 5 (Very true for me). Examples are "I feel I have a lot of voice/power to influence decisions about the program." "I am very involved in the program activities."

9. Sense of Place Sense of place will be assessed by the 11-item self-report Place Subscale of Sense of Place Scale (Domingues et al, 2021). Items 1-3 measure the emotional bond whereas items 4-7 measure the continuity as a sub-dimension of place identity. Item 8 reflects the importance of the place in supporting important or desired activities and goals. Item 9 measure the place dependence. Items 10 and 11 assess the memories of the place. A 5-point rating scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used. Total score is the summation of all items from the subscale. Higher score indicates the higher level of perceived sense of place.

   Mental health awareness

10. Mental health stigma The Peer Mental Health Stigmatization Scale-Revised (PMHSS-R) (Nearchou et al., 2021) has 11 statements which are rated on a 5-point Likert scale, 1 (Disagree completely) to 5 (Agree completely).

11. Help seeking attitudes The Self-Stigma of Seeking Help (SSOSH) scale consists of 10 items that measure attitudes toward and intent to seek psychological help (Vogel et al., 2006). It is rated on a 5-point scale. Examples are "I would feel okay about myself if I made the choice to seek professional help" and "Seeking psychological help would make me feel less intelligent".

i) Intervention costs The intervention is assumed to be in a steady operational state, implying that existing staff are utilized, and no additional training costs are incurred. Costs associated with the intervention mainly include staffing costs and intervention direct costs. For the staffing costs, wages for the school or research staffs who coordinate and implement the intervention will be calculated in line with the assessing cost-effectiveness approach, with a 30% on-cost adjustment to account for superannuation, leave entitlements, and other employee benefits. Intervention direct costs include the costs for materials directly used in the intervention, such as wellness hubs/clubs building, educational materials, digital resources, and other consumables.

ii) Cost savings The intervention is projected to yield cost savings through two primary avenues. First, parental productivity gains: By improving students' mental wellness, the intervention is expected to reduce the number of lost workdays for parents, which occur due to their children's mental health-related absences. Second, children productivity gains: The intervention aims to reduce absenteeism in students, thereby enhancing their academic productivity. The number of school days gained will be quantified and valued.

8. Analysis Statistical analyses For the pilot schools, pre-and post-intervention analyses will be conducted to evaluate the effectiveness of the intervention on students' mental wellness outcomes, including flourishing, quality of life, mental distress, understating of mental health and help-seeking intention, and social environment. A series of mixed effect models will be employed to account for both fixed (e.g., time) and random effects (e.g., school clusters). In addition, subgroup analyses will be conducted to further evaluate the effectiveness of the intervention across different groups (e.g., school banding, participants' completion rate, and participants' baseline mental health status, facilitating a more comprehensive understanding of its impacts and potential refinements of the programme. Mixed-effect models with interaction terms and stratified analyses will be employed.

For the RCT, an intention-to-treat (ITT) approach will be conducted first to maintain the validity of the findings despite potential dropouts or non-adherence. The primary comparative analyses will include Analysis of covariance (ANCOVA) to adjust for baseline differences between the intervention and waitlist control groups at post-intervention and follow-up. To handle the clustered nature of the data, mixed effects models, for example, generalized estimating equations (GEE), will be employed to investigate the intervention effects on main outcomes and the change process over time and across different schools. To explore potential mechanisms underlying the intervention (i.e., our hypothesized process of change in the Logic Model), a series of structural equation modeling (SEM) will be used to examine the mediating effects of psychosocial process and mental health awareness to achieve the intended outcomes throughout the intervention.

Inclusion Criteria:

• student in Hong Kong secondary school
• Studying at Form 1 to Form 3 at the time of recruitment
• Competence in comprehending written Chinese or English
• Competence in speaking Cantonese or English
• Written consent from students and their legal guardian

Exclusion Criteria:

• Not studying at Form 1 to 3 in Hong Kong secondary school
• Incompetence in comprehending written Chinese or English
• Incompetence in speaking Cantonese or English
• No written consent from students or their legal guardian

The overall results of masked IPD can be shared upon request made to the study principal investigator.

The IPD and supporting information will be available from beginning the 1st year and ending 3 years after the publication of results.

A formal non-commercial study proposal must be made in writing to the study principal investigator and the detailed use and analysis of the IPD have to be described.