CAPMH Corner (Updated Dec 2025)

Child and Adolescent Psychiatry and Mental Health (CAPMH) is the official IACAPAP Journal. The "CAPMH Corner" of the December 2025 issue of IACAPAP Bulletin summarizes the following three studies recently published in CAPMH – BRAVA: A randomized controlled trial of a brief group intervention for youth with suicidal ideation and their caregivers (Kennedy et al., 2025), Investigating the effects of a novel gamified cognitive training on adolescent mental health (Grunewald et al., 2025), and Exploring the role of resilience in selective prevention intervention for adolescents at risk of depression and anxiety in Nepal: findings from a pilot cluster randomized controlled trial (Singh et al., 2025).

Kennedy et al., (2025) discuss various interventions for adolescent suicidal behavior and introduce Building Resilience and Attachment in Vulnerable Adolescents (BRAVA) as a manualized, brief group treatment that incorporates elements of established evidence-based interventions (Hughes et al., 2023). BRAVA involves six weekly youth and caregiver groups focused on family connection to decrease suicidal behavior and increase family cohesion for youth with mild-to-moderate suicidal ideation (SI). 

The team conducts a randomized controlled trial (RCT) to assess the efficacy of BRAVA for adolescents with mild-to-moderate SI and their caregivers in reducing adolescent SI compared to Enhanced Treatment as Usual (ETU). They recruit families of adolescents aged 13–17.5 years with mild-to-moderate SI from a pediatric hospital and community-based mental health services in Ontario, Canada, between April 2021 and April 2023, and randomize eligible participants in REDCap using a 1:1 allocation ratio in parallel groups. An independent statistician generates a computer-based allocation table with randomly alternating block sizes of 4 and 6. Allocation remains concealed during eligibility screening, and study staff learn the final group assignment only after randomization. Blinding is not possible for staff or participants because intake occurs before randomization. To ensure adherence to the treatment manual, fidelity checklists are used and all intervention sessions are audio-recorded. The first 36 groups receive full fidelity coding by an independent, trained research assistant, showing 98.6% correspondence with the manual. For the remainder of the trial, 25% of sessions (38/152 groups) are randomly selected for fidelity coding, with reliability maintained at 97.8% (range 75–100%). The ETU control group receives weekly text messages and may participate in the BRAVA intervention after completing the exit assessment. 

They conduct an intake assessment, collect demographic data and mental health service history, and administer the following scales: 15-item version of SIQ-JR, Adolescent Anxiety and Avoidant Attachment Inventory – Long Form Parent Report of Youth (AAAAI), Family Assessment Device (FAD), Revised Child Anxiety and Depression Scale (RCADS), Perceived Stress Scale (PSS), Perceived Group Cohesion (PGC), and BRAVA Satisfaction (qualitative feedback on overall experience with BRAVA). They perform an intention-to-treat (ITT) analysis and use Multiple Imputation by Chained Equations (MICE) to impute outcomes with missing data at follow-up visits. Estimated marginal means for the two groups are obtained using ANCOVA, and homogeneity of variances is assessed using Levene’s test. 

Of the 278 families referred, the team enrolls 113 (40.6%), complete intake assessments, and identify 14 as ineligible. They randomize 99 families to BRAVA (n = 50) or Control (n = 49). They retain 74% of BRAVA and 79.6% of Control participants at exit, and 29.7% of BRAVA participants who complete the exit assessment also complete the 3-month follow-up. The sample averages 14.6 years, is 64% female, and 44% of European racial heritage. They conduct ITT analyses – both groups show reductions in suicidal ideation (SI) from intake to exit, with no significant group differences at exit. In sensitivity analyses without imputation, we detect a significant difference favoring BRAVA at exit, with these improvements maintained at 3-month follow-up. BRAVA participants also report significantly lower perceived stress, depression, and anxiety at post-treatment compared to controls. They report that caregiver outcomes do not differ significantly between groups, although a trend toward reduced caregiver stress is observed in BRAVA. Post-intervention SIQ-JR scores do not differ significantly between BRAVA (EMM = 40.7) and Control (EMM = 47.0; t(96) = –1.146, p = 0.261, ηg² = 0.034). Across the study, 14 youth present to the ED for psychiatric concerns (BRAVA = 6; Control = 8), with no significant difference in proportions. However, the Control group records significantly more ED visits during the pre–post period (BRAVA = 7; Control = 17, p = 0.02) and during the 3-month follow-up (BRAVA = 6; Control = 14, p = 0.04). At exit, adolescents and caregivers consistently report positive experiences with BRAVA, describing it as beneficial and highlighting skill gains such as identifying thinking traps and using validation skills. 

The authors note that high satisfaction and strong group cohesion suggest caregivers benefit from BRAVA despite no significant effects on attachment, family functioning, or perceived stress. They acknowledge several limitations: limited generalizability, especially to youth with more acute mental health concerns and SI; potential courtesy bias in qualitative findings; absence of 3-month follow-up data for the ETU group and high BRAVA attrition; the decision not to blind assessors or collect 3-month data from controls for ethical and safety reasons; uncertainties about replicating effects with less-experienced clinicians despite the mitigating value of the co-facilitator model and manualized approach; and the fact that although SI improves for BRAVA participants, 59.5% remain in the clinical range, suggesting BRAVA may be best suited to a stepped-care model where some adolescents require additional individualized treatment. They recommend future research assessing longer-term outcomes and examining the added value of BRAVA features such as rolling entry and separate youth–caregiver groups. They also propose evaluating the feasibility of implementing BRAVA in diverse populations and settings. 
 

Grunewald et al. (2025) underscore adolescence as a critical period for mental health vulnerability and highlight the potential of app-based or gamified cognitive training. They note that 90% of Australian adolescents play digital games regularly, averaging about 98 minutes per day and suggest that gamification can provide ecologically valid content that reflects adolescents’ real-world environments and experiences. They conduct a study to evaluate the effectiveness of a novel gamified app-based affective control training program – the Social Brain Train (SBT), and to examine whether gamification improves training uptake and outcomes. SBT includes three components: (1) a gamified affective dual n-back task that awards points for performance, (2) a cognitive interpretation bias modification (CBM-I) task in which participants resolve ambiguous social 'puzzles' positively, and (3) a psychoeducation component featuring unlockable badges and brief facts about mental health and the brain. They compare SBT with a non-gamified affective control training (AffeCT) in adolescents aged 13–16 years. They plan for a moderated effect size and recruit 253 participants, excluding 96 for likely fraudulent responses. Eligible adolescents who complete baseline measures are randomized to SBT (n = 77) or AffeCT (n = 67) using computer-generated, age-stratified block randomization with mixed block sizes 

The authors use a brief questionnaire to collect demographic information. They assess adolescents using the Generalized Anxiety Disorder-7 Scale (GAD-7), Patient Health Questionnaire–Adolescent (PHQ-A), a modified Child Anxiety Life Interference Scale (CALIS), the reappraisal subscale of the Emotion Regulation Questionnaire–Child and Adolescent version (ERQ-CA), the Repetitive Thinking Questionnaire (RTQ-10), the Online and Offline Social Sensitivity Scale (O2S3), and the Health and Social Risk Questionnaire (HSRQ). Interpretation bias is measured at baseline and post-training using the Scrambled Sentences Task (SST) and affective control using 2-back task. They administer 12 self-paced training sessions of either SBT or AffeCT over 15 days, sending reminders twice daily at 8 a.m. and 5 p.m. They analyze outcomes using general linear models, linear mixed models, and correlation analyses. 

The team finds no significant baseline differences between groups on demographic or clinical variables, except that the SBT group reports lower self-rated emotion regulation at baseline than the AffeCT group (F(1,140)=4.33, p=.039). They also observe no group differences in app helpfulness (F(1,96)=0.02, p=.896), ease of use (F(1,96)=0.18, p=.669), or likeability (F(1,96)=0.91, p=.342). They note that the total minutes spent training did not differ significantly across the two training groups. Participants assigned to SBT training, however, did engage in more training sessions than participants assigned to AffeCT training. Additionally, all participants showed improvements in affective control performance and a reduction in interpretation bias and rumination from baseline to post-training. The observed reduction in rumination persisted at 1-month follow-up. 

The authors acknowledge the limitations of the study including lower-than-expected training compliance (51%, and only 36% in the AffeCT group), which may reduce power to detect effects; the absence of a placebo non-affective control condition, making it difficult to rule out placebo effects; the short one-month follow-up period; and the need to identify and exclude fraudulent participants due to online recruitment. They recommend future trials include placebo-training controls, ensure adequate power, extend follow-up, and refine fraud-detection procedures. 
 

Singh et al. (2025) highlight the mental health risks faced by adolescents in LMICs and discuss preventive interventions, focusing on the ALIVE (improving adolescent mental health by reducing the impact of poverty) selective prevention program (developed within a conceptual framework linking poverty, self-regulation, and adolescent depression and anxiety (Lund et al., 2023). The authors hypothesize that self-regulation and resilience are positively correlated, and that resilience is negatively correlated with depressive and anxiety symptoms. They conduct a study to examine the sensitivity to change in internal and external resilience from baseline to 6 and 12 months across the ALIVE intervention arms (self-regulation, economic, combined, and control). This study is nested within the ALIVE parallel four-arm pilot cluster randomized controlled trial (cRCT). 

The study includes four arms: a self-regulation intervention, an economic intervention, a combined intervention, and a control group. Adolescents aged 13–15 years are recruited from public secondary schools in Budhanilkantha municipality and adjacent wards of Kathmandu, areas selected for their high vulnerability to urban poverty and suitability for school-based recruitment. Of 21 public schools, 11 meet eligibility criteria (≥100 students in grades 6–8 and at least 1 km apart). Eight eligible schools are randomly selected and then randomized equally (1:1:1:1), with two schools per arm, using computer-generated pseudo-randomization. They conduct assessments using the Patient Health Questionnaire–Adolescent (PHQ-A), the Generalized Anxiety Disorder scale (GAD-7), an eight-item poverty screening tool adapted from the Multidimensional Poverty Index for the ALIVE study, the 17-item Child and Youth Resilience Measure–Revised (CYRM-R), and the 10-item Rugged Resilience Measure (RRM). They analyze the data using descriptive statistics, bivariate associations, and linear mixed models with individual-level random effects for longitudinal resilience outcomes, applying maximum likelihood estimation. They also conduct pairwise comparisons, mediation analyses, independent samples t-tests, χ²-tests, and logistic regression. 

The team recruits 229 adolescents (13–15 year olds), deliver 20 group-based sessions across the three intervention arms between May and September 2024. Two groups per cluster are formed in each arm, with 10–15 adolescents per group. Sessions last 90 minutes in the self-regulation and economic arms and 2 hours 15 minutes in the combined arm. All sessions take place in schools and are delivered by two facilitators per group using a cascading capacity-building model. Participants meeting risk thresholds for depression, anxiety, or suicidality are referred to psychosocial counsellors. At baseline, the combined arm shows higher external resilience than the control group (p = 0.018). Females report higher depressive (r = 0.16, p = 0.019) and anxiety symptoms (r = 0.14, p = 0.036) and lower internal resilience (r = –0.13, p = 0.047) than males. Difficulty in self-regulation correlates positively with depressive and anxiety symptoms (r = 0.38, p < 0.001) and negatively with internal resilience (r = –0.18, p = 0.006). Both internal and external resilience correlate negatively with depressive and anxiety symptoms. Across intervention arms, no statistically significant sensitivity-to-change effects in resilience are detected. However, expected directional improvements appear: external resilience increases over time for males in the self-regulation and economic arms and for females in the self-regulation arm; internal resilience shows positive trends for males in the economic and combined arms and for both genders in the self-regulation arm. Mediation analyses showed no significant intervention effects on mental health outcomes through resilience. Internal resilience at 6 months predicts lower anxiety at 12 months, and for males across all arms, higher internal resilience is associated with lower depression. 

The authors conclude that the study underscores the need to directly target resilience in intervention design, account for ecological and contextual factors, and include longer follow-up periods to assess the sustainability of intervention effects. They acknowledge several limitations, including – limited generalizability; a relatively small sample that leaves analyses underpowered; possible social desirability bias; only three assessment time points, which may miss later-emerging intervention effects; variation in baseline external resilience between groups; potential confounders (such as prior mental health conditions or psychotropic medication use; and variations in session timing across intervention arms may influence engagement and outcomes, underscoring the need for standardized implementation or systematic evaluation of timing effects in future research. They recommend that a future, larger ALIVE trial with adequate power should test the proposed mediation model with resilience as a key process and mediator, while also exploring additional mechanisms of change and addressing stigma and group dynamics. They emphasize the importance of further research examining how resilience processes interact with social determinants of mental health to shape adolescent outcomes. 
 

REFERENCES:  

• Hughes, J. L., Horowitz, L. M., Ackerman, J. P. et al. (2023) Suicide in young people: screening, risk assessment, and intervention. BMJ. 381. https://doi.org/10.1136/bmj-2022-070630.   
• Grunewald, K., Minihan, S., Andrews, J.L. et al. Investigating the effects of a novel gamified cognitive training on adolescent mental health. Child Adolesc Psychiatry Ment Health 19, 72 (2025). https://doi.org/10.1186/s13034-025-00917-1.  
• Kennedy, A., Gray, C., Sheridan, N. et al. BRAVA: A randomized controlled trial of a brief group intervention for youth with suicidal ideation and their caregivers. Child Adolesc Psychiatry Ment Health 19, 78 (2025). https://doi.org/10.1186/s13034-025-00941-1.  
• Lund, C., Jordans, M. J. D., Garman, E., et al. (2023) Strengthening self-regulation and reducing poverty to prevent adolescent depression and anxiety: rationale, approach and methods of the ALIVE interdisciplinary research collaboration in colombia, Nepal and South Africa. Epidemiol Psychiatr Sci 32, e69. 
• Singh, R., Khanal, P., Tol, W.A. et al. Exploring the role of resilience in selective prevention intervention for adolescents at risk of depression and anxiety in Nepal: findings from a pilot cluster randomized controlled trial. Child Adolesc Psychiatry Ment Health 19, 103 (2025). https://doi.org/10.1186/s13034-025-00964-8.  
   

IACAPAP Bulletin, Issue 80