Safety Planning and Cognitive Behavioral Therapy for Adolescent Suicide Prevention in Mozambique: A Hybrid Effectiveness/Implementation Cluster Randomized Trial

More than 75% of suicide deaths occur in low-and middle-income countries (LMICs) and almost 90% of adolescents who die by suicide live in LMICs. Globally, suicide is the fourth leading cause of death for youth and young adults aged 15-29. The African WHO region has the highest age-standardized suicide rates in the world, with countries like Mozambique having suicide rates almost three times the global average. Six of the top 10 countries by suicide rates in the world are in the African region. Despite this, there are few to no evidence-based adolescent suicide prevention packages developed for, and tested in, the African context. This is an urgent need to safeguard the well-being of youth and young adults.

Data from the investigative team suggest that 15-25% of high school students in Mozambique are experiencing current suicidal ideation (SI), ~40% of those with ideation have past month suicidal behavior (SB), and 9% have had a lifetime suicide attempt. Approximately 50% of youth expressing SI or SB also express clinically significant symptoms of depression, anxiety, and/or PTSD. To address this problem, the proposed work builds upon innovations in the field and a decades-long partnership between the University of Washington and government partners in Mozambique. First, while the Suicide Safety Planning Intervention (SPI) has evidence in high-income contexts, investigators have conducted what the investigators believe to be the first pilot of SPI among adolescents 12-19 in the African region and found it feasible for delivery by non-specialists and acceptable by adolescents. Second, the investigative team adapted and demonstrated the feasibility, acceptability, and initial effectiveness of the Common Elements Treatment Approach (CETA-MZ), a transdiagnostic CBT-based therapy delivered by non-specialists in Mozambique. For young adults aged 18-24, CETA-MZ was shown to decrease depressive symptoms by 80% and reduce suicidal ideation from 20% to <3% by visit five. The results of this work have led to the national scale-up of CETA-MZ in HIV/AIDS settings in Mozambique. The investigators believe that these two evidence-based practices, both with demonstrated feasibility in Mozambique, have the potential to be powerful interventions to prevent adolescent suicidal behavior. Yet, a recent meta-analysis found lower comorbidity of psychiatric disorders and suicidal behavior in LMICs (~50%) compared to high-income countries (~90%). Therefore, it is possible that applying CETA-MZ to address psychiatric symptoms may not lead to significant decreases in suicidal behavior above and beyond SPI alone. For these reasons, the investigators propose to test both SPI alone and the integration of SPI into CETA-MZ - adapted for adolescent suicide prevention - to create a transdiagnostic CBT intervention for suicide (TCBT-S) delivered by non-specialists in Mozambican secondary schools. Due to limited resources for mental health in the African region the investigators aim to test whether the potential gains in effectiveness with the more resource intensive TCBT-S justify its scale-up versus the brief SPI intervention. This study will also generate evidence on costs, implementation determinants, and potential mechanisms of intervention effects to optimize intervention components and implementation strategies for future scale-up, if effective. The specific aims are to:

Specific Aim 1: Test the effectiveness of SPI and TCBT-S for decreasing suicidal behaviors, compared to Enhanced Usual Care (EUC). Using a three-arm parallel cluster RCT the investigators will randomize 7 secondary schools each to EUC, SPI, and TCBT-S (21 schools total) to evaluate effects on suicidal behaviors (primary) and suicidal ideation/depressive symptoms (secondary). EUC will involve screening and active referral to government youth friendly mental health service programs. Exploratory analyses will examine mechanisms of intervention effects.

Specific Aim 2: Assess implementation outcomes and determinants (barriers/facilitators) to EUC, SPI, and TCBT-S implementation using the RE-AIM evaluation framework. RE-AIM domains will be populated separately and compared across arms. The investigators will conduct a sequential quantitative to qualitative explanatory analysis - organized around the Consolidated Framework for Implementation Research - among individuals and clusters with highest/lowest effectiveness/implementation outcomes to explore determinants of implementation.

Specific Aim 3: Estimate the cost and cost-effectiveness of SPI and TCBT-S, compared to EUC. The investigators will conduct micro-costing and time-and-motion observation separately for each study arm to estimate incremental costs of implementing each intervention. The investigators will construct a Markov model parameterized with cost and trial outcomes data to project budget impact and cost-effectiveness for scale-up to provincial and national levels.