Evaluating the Effectiveness of a School-Based Mental Health Training Programme: The Transformative, Resilient, Youth-Led (TRY) Gym

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The revisions to the manuscript represent significant improvements and directly address the suggestions received during the previous review. The Introduction is now more concise and systematic, providing a clearer sense of the comprehensive context for this work while retaining the same level of detail and background. The authors have articulated their theoretical framework much more clearly and have made the connections between PYD, Positive Psychology 2.0, and the mechanisms by which the program creates change much easier to follow. Additionally, the inclusion of a clearer rationale for the program components strengthens the overall argument of the paper.

The Methods section has also benefited from the revisions. The authors now provide greater clarity in their descriptions of recruitment methods, fidelity monitoring, and scale selection. They have also clarified the cultural and linguistic validation of the measures used to collect data, which enhances the study’s credibility and reproducibility. The statistical methodology is described in more detail, and the rationale for using non-parametric analyses after normality testing is now more clearly explained. Expanded explanations of post-hoc testing have increased transparency in the presentation of results.

The results are now presented with greater clarity and a stronger narrative direction than in previous drafts. The authors have addressed previous comments regarding the reporting of median values, effect sizes, and interpretation of the Friedman tests. The results section is now easier to read and is consistent with the accompanying figures and tables. The integration of quantitative and qualitative themes is much stronger, providing a more comprehensive understanding of the program’s impact on participants.

The Discussion is now more succinct and more effectively links back to the hypotheses and theoretical frameworks. The authors now discuss the implications of their findings for youth empowerment, resilience development, and co-creating mental health promotion with youth in much greater detail. The limitations section has been rewritten to provide a more balanced view of the study’s methodological limitations without undermining the research’s contributions.

The quality of the English language has also improved significantly from previous drafts. The manuscript reads more easily and is less dense in sections that previously seemed so. While minor polishing could further improve the manuscript’s flow, its current form contains all the necessary information for readers to understand the study’s findings. The revisions have fully addressed the previous reviewers’ comments, and the manuscript is now a clear, rigorous, and meaningful evaluation of a youth-developed mental health program. It adds to the body of evidence-based knowledge in the field and is therefore ready for submission.

Author Response

Responses to All Comments:

For Reviewer 1:

1. The revisions to the manuscript represent significant improvements and directly address the suggestions received during the previous review. The Introduction is now more concise and systematic, providing a clearer sense of the comprehensive context for this work while retaining the same level of detail and background. The authors have articulated their theoretical framework much more clearly and have made the connections between PYD, Positive Psychology 2.0, and the mechanisms by which the program creates change much easier to follow. Additionally, the inclusion of a clearer rationale for the program components strengthens the overall argument of the paper.

 

2. The Methods section has also benefited from the revisions. The authors now provide greater clarity in their descriptions of recruitment methods, fidelity monitoring, and scale selection. They have also clarified the cultural and linguistic validation of the measures used to collect data, which enhances the study’s credibility and reproducibility. The statistical methodology is described in more detail, and the rationale for using non-parametric analyses after normality testing is now more clearly explained. Expanded explanations of post-hoc testing have increased transparency in the presentation of results.

 

3. The results are now presented with greater clarity and a stronger narrative direction than in previous drafts. The authors have addressed previous comments regarding the reporting of median values, effect sizes, and interpretation of the Friedman tests. The results section is now easier to read and is consistent with the accompanying figures and tables. The integration of quantitative and qualitative themes is much stronger, providing a more comprehensive understanding of the program’s impact on participants.

 

4. The Discussion is now more succinct and more effectively links back to the hypotheses and theoretical frameworks. The authors now discuss the implications of their findings for youth empowerment, resilience development, and co-creating mental health promotion with youth in much greater detail. The limitations section has been rewritten to provide a more balanced view of the study’s methodological limitations without undermining the research’s contributions.

 

5. The quality of the English language has also improved significantly from previous drafts. The manuscript reads more easily and is less dense in sections that previously seemed so. While minor polishing could further improve the manuscript’s flow, its current form contains all the necessary information for readers to understand the study’s findings. The revisions have fully addressed the previous reviewers’ comments, and the manuscript is now a clear, rigorous, and meaningful evaluation of a youth-developed mental health program. It adds to the body of evidence-based knowledge in the field and is therefore ready for submission.

 

Our Responses:

• Thank you for your thorough and positive assessment. We’re grateful that the revisions clearly addressed prior feedback—particularly the streamlined Introduction, clearer articulation of the theoretical framework linking PYD and Positive Psychology 2.0, strengthened rationale for program components, and enhanced methodological transparency.

 

• We’re pleased the improved clarity in recruitment, fidelity, scale selection, validation, and statistical reporting—including non-parametric rationale and post-hoc explanations—was helpful, and that the narrative and integration of quantitative and qualitative findings now read more cohesively.

 

• We also appreciate the feedback on language quality and complete minor polishing to further improve flow before submission. Thank you for your thoughtful guidance throughout this process; please let us know if any further adjustments are recommended.

Reviewer 2 Report (Previous Reviewer 4)

Comments and Suggestions for Authors

This a thoroughly revised version of the paper that I have reviewed in the past. Overall, the introduction is much more robust that the previous version, the figures have been substantially improved and the authors have done an excellent job in presenting the reliablity of the measures. I just have a few minor comments that should be addressed.

1. Section 3.2, the authors mention that "all variables were not normally distributed...", which is not the case according to the results in Table 3. Since the sample size is 80 (above 50), the rule-of-thumb is to report the Kolmogorov- Smirnov test, rather than the Shapiro-Wilk.
2. The use of non-parametric tests is justified in the case of non-normally distributed variables.
3. Figure 3 and 4 does not make sense statistically because it discusses mean scores, whereas above the authors have selected non-parametric tests which are based on the mean ranks. Are those the mean ranks?

Author Response

Responses to All Comments:

For Reviewer 2:

1. Section 3.2, the authors mention that "all variables were not normally distributed...", which is not the case according to the results in Table 3. Since the sample size is 80 (above 50), the rule-of-thumb is to report the Kolmogorov- Smirnov test, rather than the Shapiro-Wilk.
2. The use of non-parametric tests is justified in the case of non-normally distributed variables.

• We thank the reviewer for highlighting the inconsistency in our normality assessment in Section 3.2 and for suggesting the appropriate use of the Kolmogorov-Smirnov (K-S) test given our sample size of 80. Upon re-evaluation, we acknowledge that the original reliance on the Shapiro-Wilk test was not optimal for this sample size, as the K-S test is indeed the preferred rule-of-thumb for n > 50 due to its robustness and power in detecting deviations from normality. We have revised Section 3.2 to report K-S test results for all variables, confirming that none of scales meet the normality assumption across three timepoints. This change aligns with the reviewer's recommendation and strengthens the methodological rigor. Furthermore, the use of non-parametric tests remains well-justified for our non-normally distributed variables.

3. Figure 3 and 4 does not make sense statistically because it discusses mean scores, whereas above the authors have selected non-parametric tests which are based on the mean ranks. Are those the mean ranks?

• Regarding Figure 3 and 4, we appreciate the reviewer's astute observation that presenting mean scores is statistically incongruent with our adoption of non-parametric tests, which rely on median ranks rather than arithmetic means. To address this, we have revised both figures to display medians instead of means, providing a more appropriate central tendency measure that aligns with the non-parametric framework. These updates better reflect the underlying data distribution and enhance interpretability without altering the substantive conclusions. The accompanying figure captions and text in Sections 4.2 and 4.3 have been updated accordingly to explicitly reference medians and their statistical rationale.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study is generally well-planned and systematically presented. The researchers have demonstrated a strong approach based on youth participation, especially by making students an active part of the process. In this respect, the study offers a valuable contribution not only to the field of mental health, but also disciplines such as youth development, psychoeducation and school-based interventions.

However, a more detailed explanation of some methodological and reporting points could increase the reliability and widespread impact of the study. Such clarifications will contribute to the research reaching a wider audience in academic literature and in practice.

Some suggestions for researchers in this direction are given in the attached file

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to thank the authors for allowing me to review this manuscript, which described the implementation and evaluation of the Transformative, Resilient and Youth-Led/Driven (TRY) Gym programme—a co-creative, school-based intervention designed to promote mental health among adolescents in Hong Kong. The topic is relevant, and the intervention is socially and pedagogically important, given the increasingly challenging mental health landscape facing youth across the world. The manuscript sets out a strong rationale based in the Positive Youth Development (PYD) framework and provided background context on the educational and psychological pressure upon students in Hong Kong.  However, the introduction would benefit from a broader international contextualization. The authors should consider comparing the TRY Gym program with other similar school-based mental health interventions developed in high-pressure non-Western educational systems. For example, the recent study by Laurenzi et al. (2024, https://doi.org/10.1016/j.ssmmh.2023.100289) provides valuable insights into the co-design and implementation of a school-based adolescent mental health program in Nepal and South Africa. A comparison of TRY Gym with such work could provide crucial cross-cultural depth and highlight the unique or transferable features of the Hong Kong model. Furthermore, incorporating current literature on the psychological constructs underlying TRY Gym would enhance its theoretical grounding. For example, Graziano et al. (2024, https://doi.org/10.1038/s41598-024-70836-2) examine how emotional self-efficacy interacts with empathy and gender in educational contexts, concepts that are directly relevant to the emotional competencies that TRY Gym aims to promote. The inclusion of such references could not only strengthen the theoretical argument, but also support a more globally relevant understanding of adolescent mental health promotion.

Methods

The manuscript's methods section is the weakest part, and needs marked improvement. The authors discuss an embedded mixed-methods design, however critical details needed to evaluate the scientific rigor of the study are missing or undeveloped. First, no power analysis is provided and no statistical or theoretical justification for sample size is provided. As a result, it is not possible to evaluate if the quantitative component is adequately powered to detect meaningful change at the three measurement points. Second, the study is conducted without a control group and is presented with no randomization which significantly limits causal claims and strength. The authors do not address or recognize the limitations anywhere in the manuscript. Third, there is a serious lack of detail related to the psychometric validation of the measurements used. The authors use multiple self-report scales but do not report whether they have been validated for use with the Hong Kong adolescent population. No details are offered regarding cultural validation procedures or linguistically appropriate adaptations, leading one to consider whether the measures are appropriate and reliable enough. Fourth, attrition is reported (down from 104 participants to 80) but no dropout analysis is offered. We do not know if students who dropped out were systematically different from those that were retained, or how the dropout may impact the findings. Fifth, the manuscript mentions implementation fidelity (that facilitators were trained and a manual developed), but no detail on how fidelity was monitored across multiple schools is given. Did facilitators use the same facilitator scripts? Were there observation checklists or fidelity logs? No mention is made of any of these procedures. Finally, with respect to the qualitative component, there is insufficient detail on how interviews were conducted, coded and analyzed. There is no mention regarding the methodology used for the development of themes or the determination of saturation, and without reporting intercoder reliability or transparency in the development of thematic categories the qualitative aspects of the paper lack support.

Results

The results section describes the findings in a mostly straightforward and coherent way. Non-parametric Friedman tests are used which is appropriate considering the sample size and non-normal distribution. However, p-values are reported without effect sizes; as such, readers are limited in their ability to assess the magnitude of the results. Reporting effect sizes, e.g., Kendall's W or eta-squared, would provide readers with a clearer practical orientation to the statistically significant differences identified. No correction for multiple testing is mentioned in this section, and thus, the authors do not acknowledge the risk of Type I error. The qualitative results are presented in terms of themes developed from interviews, but are outlined to be generally under-developed, with too few illustrative quotes. Ultimately, this weakens the impact of the qualitative results in the manuscript, in that the thematic analysis is shallow, and unattributed to quotes made by research participants. If qualitative results are to be retained in the final version, they would benefit from elaboration with a richer analysis and obligation to use verbatim quotes that provide insight into the adolescents' lived experiences of the intervention.

Discussion

The discussion section reiterates the main findings, but fails to provide a critical analysis or to integrate the study into broader academic discussions. In particular, the authors do not address the methodological limitations mentioned above. There is no reflection on the lack of a control group, the high dropout rate, the reliance on self-report or the limited cultural validation of the instruments. This lack of self-criticism compromises the scientific integrity of the work. I strongly recommend the inclusion of a subsection explicitly outlining these limitations and explaining how they affect the interpretation and generalizability of the results.

The discussion should also draw comparisons with similar international interventions to contextualize the TRY Gym model and identify points of convergence or innovation. For example, the study by Redman et al. (2024, https://doi.org/10.1186/s13643-024-02482-8) provides a systematic review protocol of creative bibliotherapy interventions targeting adolescent mental health — an approach that shares TRY Gym’s interest in expressive and participatory methods. Reflecting on how TRY Gym aligns with or differs from such programs would enrich the discussion and support possible avenues for international transferability or adaptation.

Finally, the manuscript does not include a section on the practical implications of the TRY Gym program. This is a major shortcoming. The authors should address how schools might implement this program, what training is required, what institutional support is needed, and how the intervention might be extended or modified to other cultural or educational contexts. Without this information, the value of the study as a model for practice is diminished.

Figures and Tables A

lthough all tables are functional, the reader would have better accessibility if graphical representations accompany the text, perhaps featuring line graphs to depict differences across T1, T2 and T3. The figures, where provided, are clear, however, could also include captions to interpret them, not just repeat what's already in the figure.

 

 

 

Comments on the Quality of English Language

The English is mostly comprehensible but would benefit from careful language editing to improve clarity and academic tone. Several grammatical issues, such as incorrect verb tenses and awkward phrasing, detract from the manuscript’s readability. A thorough proofread by a native or fluent English speaker is advised prior to resubmission.

For instance, in the Introduction, the sentence "Adolescence is a developmental stage that begins with puberty and ends at the onset of adulthood [1]. During this period, individuals undergo significant neural, hormonal, and interpersonal changes [2]" is clear but could be made more engaging with academic transitions. Conversely, the passage beginning "In the context of Hong Kong, the most recent Youth Development Blueprint..." contains multiple issues in verb agreement, punctuation, and run-on structure. The sentence: "from 2014 to 2024, 228 reported death of students was caused by suicide, from the lowest 9 cases in 2014 to the highest 32 cases in 2023" should be corrected to: "From 2014 to 2024, 228 student deaths were reported as suicide cases, ranging from 9 in 2014 to a peak of 32 in 2023."

Similarly, the expression "there was 28 students died for suicide" is grammatically incorrect and should be rewritten as: "There were 28 student suicides in 2024, representing a slight decline from the previous year but still the second highest on record." The phrase "planned committing suicide" in "1.7% secondary students, and 2.8% primary students planned committing suicide" should be revised to: "planned to commit suicide."

In addition, the use of commas instead of conjunctions or semicolons results in several run-on sentences, as seen in: "the study involved around 257 000 primary students and 173 000 secondary students. According to Yeong [20] from Hong Kong Suicide Press Database (HKSPD), there were 11 committing suicide press reports of people who aged from 11 to 16..." which could be rewritten as: "The study involved approximately 257,000 primary and 173,000 secondary students. According to Yeong [20] from the Hong Kong Suicide Press Database (HKSPD), there were 11 reported press cases of suicide among individuals aged 11 to 16 between January and mid-April 2025."

Throughout the manuscript, phrases such as "the study revealed 1.7% secondary students" would benefit from more formal structure: "The study revealed that 1.7% of secondary students..." Also, subject-verb agreement issues appear frequently, e.g., "mental disorder is affecting 14% youth aged from 10 to 19" should be revised to: "Mental disorders affect 14% of youth aged 10 to 19."

A more formal and precise tone would significantly strengthen the impact and clarity of this important work.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thanks so much for the opportunity to review! This paper presents a timely and important intervention study that addresses adolescent mental health in Hong Kong through the innovative, youth-led, school-based initiative known as TRY Gym. The integration of PYD principles with a mixed-methods evaluation strategy seems reasonable. Despite these commendable aspects, though, the manuscript requires significant development in several key areas.

• Key Areas for Development - Theoretical depth: Needs more;
• Methodological transparency: Crucial for credibility; and
• Engagement with systemic dimensions of youth mental health: This feels underdeveloped.

Intro and Theoretical Framing

The intro outlines the context of adolescent mental health nicely in Hong Kong and articulates the relevance of PYD principles. Yet, the theoretical framing would benefit from a more profound critical engagement with the structural factors that shape mental health trajectories in East Asian contexts. Sure, the reliance on PYD and PP 2.0 is appropriate, but it lacks sufficient originality, unfortunately. The authors miss an important opportunity to critically interrogate how systemic educational pressures and cultural norms surrounding academic achievement and family obligation intersect with youth wellbeing. Integrating perspectives from the sociology of education and/or (cross-)cultural psychology could significantly strengthen the framework, offering a more nuanced understanding of these complex interactions. There's a great deal of research along these domains, so the authors should explore a bit more. This study could be framed as being more "unique."

Mixed-Methods Approach and Youth Leadership

The study's adoption of a mixed-methods approach, combining quantitative outcomes with qualitative insights, and its incorporation of a youth-led design aligned with principles of empowerment and agency, are SUPER. However, the manuscript does not sufficiently address the constraints that institutional hierarchies may impose on authentic youth leadership. There is very, very minimal discussion about how power asymmetries between adult facilitators and youth participants were navigated, particularly within school settings where authority is deeply entrenched. A more reflexive stance on these dynamics would undoubtedly strengthen the credibility of the participatory claims.

Methodological Concerns - Participant exclusion: HUGE ISSUE. 24 participants (nearly 1/4 of sample) excluded. No clear justification;

No sensitivity analysis limits confidence;

Need analysis comparing excluded vs. included to rule out systematic bias;

Recruitment bias: Teacher/social worker referral likely introduced selection bias;

Potentially favored students already identified as needing support; maybe other.

• Could exclude more marginalized youth less visible to school personnel.

Internal consistency of measures - Motivational competence (α=.681) - Marginal at best.

Emotional competence (α=.780) - Also marginal.

α=.681 is typically unacceptable. These values are near or below common thresholds and need careful discussion. This nuance must be addressed for the analysis to be meaningful.

Results and Implications

The results are presented generally clearly, including non-significant findings, and the integration of quantitative and qualitative data effectively strengthens the analytic narrative. Nevertheless, the implications of the findings would benefit from more direct engagement with the systemic context in which these outcomes occur. The authors mention the three-tiered support system but stop short of examining how TRY Gym might either reinforce or challenge existing institutional logics. In particular, the exclusive focus on individual coping and resilience could be interpreted as aligning with neoliberal educational discourses that frame mental health as a personal responsibility.

Conclusions

The conclusions are largely aligned with the data but do not adequately engage with broader structural determinants of adolescent mental health. This is very disappointing. The discussion gives insufficient attention to how Hong Kong's competitive academic environment and prevailing policy frameworks shape psychological outcomes. The authors might consider how programs like TRY Gym could evolve from merely fostering adaptation to advocating for systemic change, helping students not only cope with but also critically respond to systemic sources of distress.

Other Points for Review

• Socioeconomic diversity: Lightly touched (housing status), but no analysis of intersecting factors (class, gender, migration background) on engagement/outcomes. Is this intentional or an oversight?
• Social workers during interviews: Their presence may have shaped youth disclosure. This potential source of bias isn't addressed. More methodological reflexivity is needed.
• Theoretical advancement: PP 2.0 with PYD is competently executed but offers limited novel framework. It's more of a synthesis. Future work could incorporate youth-generated conceptions of wellbeing, especially those beyond Western psychological constructs.

Final Thoughts

Although the paper has potential, several major aspects require substantial revision before this manuscript could be reconsidered, as outlined above. I wish the authors the best of luck. YOU CAN DO THIS!!!

Comments on the Quality of English Language

The manuscript is generally “readable” but contains numerous grammatical inconsistencies and awkward phrasing. Expressions such as “encouraging students to socialise with their peers/mentees with these skills” are unclear or overly literal. Professional copyediting would considerably improve clarity and flow.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This is an insightful paper presenting the TRY Gym programme in Hong Kong and showing evidence in its favour using a sample of 80 participants. Please find below some feedback for improving the paper.

1. I would have expected the introductory section to begin with the problem that TRY Gym programme aims to address concretely and why this programme is particularly needed, besides from an introduction of the developmental challenges in adolescence
2. Figure 1 is a bit blurry and becomes hard to read. Please consider reformatting or pixellating accordingly.
3. Section 2.3.4, spelling error: Edinburgh
4. Mean substitution can be tricky if there are not many data points, especially considering that the substitute mean might bias the data toward extreme responses. I am wondering whether any data screening was conducted before mean substitution to verify that there are no outlying, extreme cases that could inadvertedly lead to a biased mean. Also, if I am not mistaken, mean substitution requires that the data missing patterns be Missing Completely At Random (MCAR). Has this assumption been verified?
5. Is there a power analysis to support the statistical power of the applied statistical methods?
6. line 496: If I am not mistaken, it should be that the variables were not normally distributed, not the data per se?
7. Beyond the Shapiro-Wilk test, did the authors screen the variables using the Q-Q plots? The Shapiro-Wilk test has several limitations and it might be the case that the Q-Q plots might show reasonably normally distributed variables
8. If the variables were not normally distributed, as the authors claimed, then it does not make sense to be comparing means over time. Additionally, I am wondering whether the authors could report the full test applied in each case along with the accopanying effect size to be able to guage the extent of the mean differences.
9. The sample is quite varied in its demographic background, which can actually lead to significant differences in the key outcomes due to the confounding by those demographic factors catalogued in table 2
10. In the mean score figures 3 and 4, it would be a nice addition to add the standard error of the mean in order to show the degree of uncertainty in the point estimates
11. line 556: if the table reports descriptive statistics, then why are the authors reporting the degrees of freedom?
12. I feel that inside the results' section, there are ambiguous findings especially due to the use of different types of tests. 
13. Inside the section describing the qualitative findings, I would appreciate a description of the sample's characteristics. This would be very helpful in placing the qual findings in light of the sample's background.
14. The qualitative findings: It is always a very good practice to discuss a bit on how the trustworthiness of the thematic analysis was achieved. In other words, how was credibility achieved?
15. Inside the discussion's section, I would appreciate a more thorough comparison of the findings with past evidence coming from previous programmes that had similar goals as the TRY Gym programme.
16. I think that the limitations' section requires further expansion. Particularly, the lack of a control group precludes any definite answers to some questions about the effectiveness of the programme.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf