Evaluation of the effectiveness of Reframe-IT + to prevent suicidal ideation and other mental health outcomes among adolescents in Chile: 3-month, 6-month and 10-month follow-up

Abstract

Background

Schools are suitable settings to prevent suicide among adolescents. We tested the effectiveness of a school-based program to reduce suicidal ideation (SI) (Reframe-IT +). It encompasses 13 sessions (eight internet-based and five face-to-face sessions).

Methods

We conducted a cluster RCT among students aged 13–18. Twenty-one schools were randomized (Intervention Group (IG), N = 47); Control Group (CG), N = 48)). The primary outcome was SI. Secondary outcomes were depressive and anxiety symptoms, psychotic experiences, hopelessness, and emotion regulation strategies (cognitive reappraisal, and emotion suppression), problem-solving skills and behavioral activation. Outcomes were assessed at baseline, post-intervention, 3-month, 6-month, and 10-month follow-up). Trial registration: Clinical Trials NCT05229302. Registered on Registered on February 8, 2022.

Results

We found a non-significant overall intervention effect on SI across follow-up assessments, but significant overall intervention effects on depressive and anxiety symptoms and psychotic experiences. Moreover, we found time-specific reductions immediately after the intervention and close to it, particularly for depressive and anxiety symptoms, hopelessness, behavioral activation and cognitive re-appraisal. The change favoring the IG of cognitive re- appraisal was also significant at 10 months follow-up.

Conclusions

Our results suggest that Reframe-IT + reduces depressive and anxiety symptoms and psychotic experiences in the long term. Less promising results were related to suicidal ideation prevention. Further research is needed with a larger sample size.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s40359-026-05164-3.

Introduction

Common mental health disorders (CMHD), such as depression and anxiety, range from 10 to 20% in adolescents [1]. These disorders have serious consequences, including suicide-related behaviors (SRB) [2, 3]. A relevant component of SRB is suicidal ideation (SI), which has a prevalence of 15.4% to 18.4% among individuals aged 14 to 21, and it is associated with increased suicide risk [4, 5]. Approximately 38%–50% of adolescents with major depressive disorder report recurrent suicidal ideations [6, 7]. Moreover, adolescents with anxiety disorders show nearly double the risk for SI compared to non-anxious peers [8]. The treatment rates for depression and anxiety among adolescents are low, ranging from 6 to 40% [9]. Many adolescents at risk for suicide remain without the necessary support due to a lack of timely detection and intervention programs [10, 11]. Literature shows that untreated depression and anxiety during adolescence increase the likelihood of early onset of SI, its persistence, recurrence, and the risk for progression to suicide attempts [12–15]. In this context it is crucial developing accessible preventive interventions [16] and, especially SRB [9]. Schools have been considered adequate settings for implementing mental health prevention programs [17, 18]. However, the evidence about the effectiveness of school-based interventions in reducing CMHD and SRB is still limited [19–21]. Prevention interventions could face some challenges when implemented in vulnerable schools. Then, conducting research in these settings may help to understand these challenges and provide insights to overcome them. In Chile, we can find these settings in public and subsidized schools which serve a large proportion of adolescents from socioeconomically vulnerable backgrounds, who often face higher psychosocial adversity and barriers to accessing mental health services. School vulnerability is commonly estimated through the School Vulnerability Index (IVE-SINAE), developed by the National Board of School Aid and Scholarships (JUNAEB), which incorporates indicators such as household income, parental education, healthcare coverage, and participation in social support programs.

Digital technologies are potentially feasible and efficient for delivering mental health and suicide prevention programs [22], reducing costs, and increasing accessibility to support where resource limitations may hinder access to services [23]. This is particularly relevant Latin American countries where barriers to access to mental health services are high [24, 25]. Most used technologies in the context of indicated interventions include internet- or web-based interventions, smartphone applications and virtual reality [26]. Evidence suggests that digital mental health and suicide prevention programs are safe and effective [27, 28]. Most of these interventions are based on cognitive behavioral therapy (CBT) [29], using cognitive restructuring and behavioral activation techniques to reduce symptoms of depression and anxiety in adolescents [30]. To our knowledge, Reframe-IT is the only program using digital technologies tailored to be delivered in schools. It contains eight CBT internet-based modules delivered over ten weeks by school wellbeing staff, supervised by a mental health professional [31]. Preliminary studies in Australia showed that the intervention is acceptable and feasible [32]. Although reductions in SI, depression, and hopelessness among adolescents were observed [33], the effects were not significant [10], probably because the trial was underpowered.

Based on literature suggesting that human support could improve the effectiveness of digital interventions to prevent suicide in adults [34, 35], we created five additional face-to-face modules reinforcing the content of the original program (Reframe-IT +). The protocol of this intervention is described elsewhere [36]. The pilot study significantly reduced SI and depression symptoms and improved behavioral activation, cognitive reappraisal, and problem-solving skills [37]. Additionally, in the Cluster RCT (cRCT), at post-intervention, Reframe-IT + reduced SI and symptoms of depression, anxiety, and hopelessness and showed improvement in similar skills as in the pilot study [38]. This study aimed to evaluate the effectiveness of Reframe-IT + in preventing SI and other mental health outcomes among adolescents attending vulnerable schools in Chile, three, six, and ten months after its implementation.

Methods

Participants and procedure

This study was approved by the ethical committee of Universidad de Talca (Folio 02–2012, May 12, 2021). Informed consent was obtained from all participants and their legal guardians. The authors confirm that all methods were performed following the Declaration of Helsinki.

Trial registration: ClinicalTrials.gov, Identifier: NCT05229302. Registered on February 8, 2022.

Study design, setting, participants, and eligibility criteria

This study was a cluster-randomized controlled trial. Municipal and subsidized schools are the primary providers of education for most low-income students in the Region of Maule. The sampling frame included all municipal and subsidized co-educational secondary schools in the Region of Maule with at least two grades between 1 and 3 Medio (equivalent to 9th to 11th grade in the United States). All invited schools were informed of the study's objectives and methodology. Twenty-one out of 41 eligible schools accepted the invitation to participate. After obtaining approval from school principals, the research team attended parent meetings at the start of the academic year to explain the study and obtain written informed consent for the student's participation. Upon receiving the signed consent forms, schools were randomly assigned to two groups: the Reframe-IT + intervention group and the control group. Randomization was conducted in a 1:1 ratio by an independent statistician using the Microsoft Excel RAND function. However, because an odd number of schools agreed to participate (21 schools), one additional school was allocated to the control group. The primary criterion to include students into the intervention was a score of ≥ 3 points on the Columbia-Suicide Severity Rating Scale (C-SSRS) [39], validated as a self-report tool in Chile [40]. This criterion allows including adolescents with moderate to high risk. The exclusion criteria included: 1) a score greater than 19 on the PHQ-8; 2) one or more suicide attempts in the past 30 days; 3) the presence of perceptual anomalies such as auditory or visual hallucinations; and 4) intellectual disability or communication difficulties due to language barriers. For the PHQ-8, a cut-off score of 19 was used to indicate severe depression, aligned with the PHQ-9 scoring threshold, given evidence that both scales (PHQ-8 and PHQ-9) have similar predictive validity for depressive disorders across severity levels [41]. All eligible students and their parents were invited to clinical interviews to confirm exclusion criteria and assess their motivation to join the study. Adolescents scoring ≥ 3 points on the C-SSRS and more than 19 points on the PHQ-8 were excluded. They were also invited, along with their caregivers, to discuss their results and facilitate referrals to health services. Students that dropped out during the intervention period signed a document. Assessments were conducted at the participants' schools by research assistants who were blinded to the school condition and independent from the treatment delivery process. The assessments were administered through an online survey in the schools' computer labs. The sample consisted of 51 adolescents in the intervention group and 49 in the control group. Follow-up assessments were conducted at four time points: post-intervention, 3-month, 6-month, and 10-month follow-up. Recruitment for each follow-up phase was prioritized, and different strategies were implemented to contact all participants. First, contact information for a relative was rigorously collected during the initial trial and updated at the end of the intervention. Second regular contact was maintained with schools. Third, multiple phone calls, were used for participants who could not be located through these primary channels. The baseline assessment was from July 25th, 2023, to October 2nd, 2023, and the post-intervention assessment was from October 24th, 2023, to December 27th, 2023. A 3-month follow-up was conducted between March 7 and April 8, 2024, and the 6-month follow-up occurred from June 6 to July 17, 2024. Finally, the 10-month follow-up was from September 16 to October 16, 2024 (Fig. 1 summarizes the recruitment and follow-up of schools and students).

Fig. 1.

Recruitment and follow-up of schools and students. Notes: IVE-SINAE = for initials in Spanish of “Índice de Vulnerabilidad Escolar—Sistema Nacional de Asignación con Equidad” (School vulnerability index—National Allocation System with Equity); SI = Suicidal Ideation; C- SSRS = Columbia–Suicide Severity Rating Scale; PHQ-9 = Patient Health Questionnaire-9; SA = Suicide Attempt

All participants in both groups received the protocol for managing suicidal ideation defined by the Chilean Ministry of Health through referrals to primary care centers. This protocol includes psychological interventions based on cognitive-behavioral therapy (CBT) and pharmacological treatment, when necessary. This constituted the treatment-as-usual (TAU) condition. Further details of the procedures can be found in the RCT protocol [36].

Sample size

The sample size calculation assumed a 55% loss to follow-up among eligible students (C-SSRS score ≥ 3). We also assumed a harmonic mean of 8 students recruited per school. The expected post-intervention SIQ scores were based on results from a pilot study under review: control group mean = 51.9 (SD = 21.4) and intervention group mean = 41.0 (SD = 18.6). We expected to recruit at least 18 schools. We used the “clustersampsi” command in Stata Software to estimate the number of clusters in two arms, using the following command: clustersampsi, samplesize mu1(X1) mu2(X2) sd1(Y1) sd2(Y2) m(Z) rho(R). Where X1 = mean in arm 1; X2 = mean in arm 2; Y1 = standard deviation (SD) in arm 1; Y2 = SD in arm 2; Z = number of children per school on average (harmonic mean) (n = 10); and R = intracluster correlation (R = 0.023). The two arms were balanced for the school's size.

Intervention and control group

Intervention group (IC). Reframe-IT + is a blended intervention adapted from the original program and described in [36]. The online component comprises eight ~ 20-min video-based modules of CBT and is administered at schools by trained psychologists (one psychologist per school). Each participant has access to their personalized web page accessed via secure login. Each module is completed in the psychologist’s presence, and participants can also access it at home 24 h a day. Additionally, the student finds a series of fact sheets covering various related topics, including managing SI, plus downloadable relaxation MP3s [10, 32]. The face-to- face component comprises five CBT sessions to 1) build a safety plan, 2) reinforce emotion recognition and practice relaxation exercises, 3) understand the relationships among emotions, thoughts, and behavior, 4) develop problem-solving skills, and 5) promote thinking restructuring. The first face-to-face session consists of building an individualized safety plan to provide students with a toolkit of strategies to help them cope with psychological distress and suicidal ideation [42, 43]. The other face-to-face sessions followed two online sessions to ensure the content was complementary. These face-to-face sessions used brief videos (1–2 min), activities, and mood checking. Each session lasted about 30 min. Sessions were generally delivered weekly during school hours, although some sessions required rescheduling due to school-related activities or absenteeism.

Control group (CG)

Students in this group were referred to primary care clinics and received the TAU defined by the Chilean Ministry of Health for this population.

Measures

The primary outcome measure for the study was the Suicidal Ideation Questionnaire (SIQ), which provided a continuous score post-intervention. The SIQ is a 15-item self-report instrument designed to assess suicidal ideation in adolescents, demonstrating high internal consistency (α = 0.93) and test–retest reliability over a four-week period (kappa = 0.72) [44]. A Spanish version of the SIQ, previously validated for adolescents in Chile [45], was used in this study. At baseline, the internal consistency of the SIQ in our sample yielded a Cronbach’s α of 0.95 and McDonald’s ω of 0.96.

The secondary outcome was measured using the Patient Health Questionnaire-8 (PHQ-8), which assesses eight of the nine criteria for depressive disorders based on the DSM-5 [46], with responses ranging from 0 (not at all) to 3 (nearly every day). The ninth criterion, related to suicidal ideation, was excluded as this was already assessed via the Columbia-Suicide Severity Rating Scale (C-SSRS) for inclusion criteria and through the SIQ as the primary outcome. We wanted to have a measure of depression severity and avoid collinearity in the analyses. Total scores can range from 0 to 24 [47]. Borghero et al. [48] reported that the PHQ-9, widely used in Chilean adolescents had an internal consistency of 0.78, with sensitivity and specificity of 86.2% and 82.9%, respectively. In this study, the PHQ-8 demonstrated Cronbach’s α of 0.86 and McDonald’s ω of 0.86 at baseline. Additional measures included the Generalized Anxiety Disorder Scale (GAD-7) [49], the Beck Hopelessness Scale (BHS) [50], the Community Assessment of Psychic Experiences-Positive (CAPE-P15) [51] and the Short Form of the Social Problem-Solving Inventory-Revised (SPSI-R Short Form) [52], a 25-item scale evaluating social problem-solving skills. The Emotion Regulation Questionnaire for Children and Adolescents (ERQ-CA) [53] was also employed, assessing two emotion regulation styles: Cognitive Reappraisal (CR) which consists of redefining a potential emotion-eliciting situation in such a way that its emotional impact is changed, and Expressive Suppression (ES), which involves inhibiting ongoing emotional expression. The ERQ-CA has been validated in Chilean adolescents [54]. Lastly, the Cognitive Behavioral Therapy Skills Questionnaire (CBT-SQ) [55] was used, a 16-item scale assessing two core CBT skills: cognitive restructuring (CR) and behavioral activation (BA), which measure changes in cognitive style and avoidance behavior.

Statistical analyses

Firstly, we used descriptive statistics to assess balance across arms at baseline (Table 1). We registered and presented percentages of attendance (App. 1, Suppl. Table 1) and the frequencies of each session implementation, which needed to be rescheduled (App. 1, Suppl. Table 2).

Table 1.

Demographics with all students at risk included in the trial

Variables	Control (n = 49)		Intervention (n = 51)
Sex	n	%	n	%
Male	12	24.5	22	43.1
Female	37	75.5	29	56.9
Psychological/psychiatric treatment
Never	14	28.6	20	39.2
In the past	18	36.7	17	33.3
Currently	17	34.7	14	27.5
Psychological/psychiatric treatment in family
Never	4	8.2	8	15.7
In the past	17	34.7	13	25.5
Currently	13	26.5	12	23.5
I do not know	15	30.6	18	35.3
History of suicide attempts in family
No	7	14.3	12	23.5
Yes	15	30.6	15	29.4
I do not know	27	55.1	24	47.1
History of suicide in family
No	19	38.8	30	58.8
Yes	11	22.5	8	15.7
I do not know	19	38.8	13	25.5
	n	Mean (SD	n	Mean (SD
Age	49	15.4 (1.0)	51	15.3 (1.1)
SIQ	49	51.4 (16.9)	51	47.4 (17.5)
CSSRS	49	5 (1.5)	51	4.5 (1.3)
PHQ-8	49	13.9 (2.9)	51	13.0 (2.8)
GAD-7	49	13.2 (4.5)	51	12.0 (3.6)
SPSI-R	49	47.2 (7.8)	51	45 (11.8)
BHS	49	7.4 (4.4)	51	8.2 (4.6)
CBT Questionnaire
Behavioral activation	49	18.5 (5.8)	51	18.4 (5.9)
Cognitive restructuring	49	21.8 (5.7)	51	20.5 (5.5)
ERQ-CA
Cognitive reappraisal	49	18.9 (3.7)	51	18.6 (4.6)
Expressive suppression	49	14.2 (2.9)	51	13.9 (3.4)
CAPE-15	49	39.0 (8.9)	51	34.3 (8.4)

The primary between-group analysis was carried out on an intention-to-treat basis for post- intervention, 3, 6, and 10-month follow-up on SIQ scores. We used linear mixed-effects model analysis to compare the intervention and control groups regarding the change in outcome measures from baseline to post-intervention and follow-ups. We included age, sex, and the baseline score in each model. Schools were included as a random factor for each outcome measure to account for possible school effects. Additionally, we conducted an interaction analysis between the group and time, allowing a random effect on time due to the difference between time intervals (Table 2). We repeated the analyses with secondary and additional outcomes (Table 2).

Table 2.

Estimated means and SDs on outcome variables for groups

Variables and groups	Baseline, mean (SD)	Post-intervention, mean (SD)	3 months, mean (SD)	6 months, mean (SD)	10 months, mean (SD)	Overall intervention effect Global AMD (p value)
SIQ
CG	51.4 (16.9)	39.7 (18.1)	37.2 (21.2)	34.5 (21.1)	33.5 (19.9)
IG	47.3 (17.5)	32.0 (13.0)	27.8 (15.5)	28.9 (16.1)	26.4 (14.3)
AMD		−3.789	−6.259	−3.137	−3.536	−4.8 (0.069)
PHQ-8
CG	13.9 (2.9)	10.8 (4.6)	9.9 (4.9)	9.4 (5.0)	9.1 (4.3)
IG	13.0 (2.8)	7.3 (4.2)	6.3 (3.6)	7.0 (4.9)	6.5 (3.9)
AMD		−2.645**	−2.583**	−1.561	−1.862	−1.2 (0.015)
GAD-7
CG	13.2 (4.5)	10.6 (5.3)	9.6 (5.6)	9.0 (4.9)	8.3 (5.0)
IG	11.9 (3.6)	7.2 (4.2)	6.5 (4.3)	7.1 (4.6)	6.6 (4.2)
AMD		−1.862**	−2.137**	−1.136*	−0.872	−1.6 (0.021)
BHS
CG	7.4 (4.4)	6.9 (4.6)	6.2 (5.1)	5.5 (4.6)	3.9 (4.2)
IG	8.1 (4.6)	3.8 (4.2)	4.1 (4.9)	4.0 (4.2)	3.3 (4.1)
AMD		−3.903**	−2.913**	−2.337*	−1.504	−0.6 (0.404)
CAPE-15
CG	39.0 (8.9)	29.7 (9.2)	28.9 (11.0)	27.0 (10.1)	25.1 (9.0)
IG	34.3 (8.4)	26.8 (8.6)	25.6 (7.8)	26.0 (9.7)	23.4 (8.3)
AMD		1.898	1.451	2.380	2.537	−3.2 (0.026)
SPSI-R
CG	41.1 (7.8)	44.0 (13.7)	41.2 (17.6)	37.5 (17.8)	39.0 (12.1)
IG	45.0 (11.7)	41.0 (11.7)	37.0 (13.0)	38.8 (13.3)	40.4 (9.6)
AMD		−0.293	−2.616	3.194	3.491	−0.4 (0.818)
CBT-BA
CG	18.5 (5.8)	18.6 (6.2)	19.8 (6.2)	19.2 (6.5)	21.3 (5.9)
IG	18.4 (5.9)	21.4 (5.3)	20.8 (6.5)	20.0 (6.1)	22.1 (6.0)
AMD		2.960*	1.278	1.270	1.470	0.6 (0.510)
CBT-CR
CG	21.8 (5.7)	21.0 (4.9)	21.9 (6.3)	21.0 (6.8)	21.9 (6.0)
IG	20.5 (5.5)	21.2 (5.9)	19.8 (6.1)	22.0 (6.2)	22.0 (6.7)
AMD		1.561	−0.745	2.153	1.657	−1.1 (0.220)
ERQ-CR
CG	18.9 (3.7)	19.9 (4.5)	20.0 (4.7)	20.0 (4.0)	20.2 (3.5)
IG	18.6 (4.6)	22.1 (4.0)	21.0 (5.4)	22.0 (4.7)	22.6 (4.1)
AMD		2.271*	1.074	2.176*	2.745*	1.1 (0.064)
ERQ-ES
CG	14.2 (2.9)	13.7 (3.2)	13.0 (3.0)	13.3 (2.7)	12.2 (3.2)
IG	13.9 (3.4)	13.0 (3.3)	12.3 (3.6)	12.7 (3.0)	12.1 (3.4)
AMD		−0.447	−0.522	−0.125	0.425	−0.3 (0.490)

^* p ≤ 0.05; ** p ≤ 0.001. All analyses were adjusted for age and sex, and clustering by school

CG Control group, IG Intervention group, AMD Adjusted Mean Difference, SIQ Suicidal Ideation Questionnaire, PHQ-8 Patient Health Questionnaire-8, GAD-7 Generalized Anxiety Disorder Scale, BHS Beck Hopelessness Scale, CAPE-15 The Community Assessment of Psychic Experiences-Positive scale, SPSI-R Social Problem-Solving Inventory-Revised, CBT-BA Cognitive Behavioral Therapy Skills Questionnaire-Behavioral activation, CBT-CR Cognitive Behavioral Therapy Skills Questionnaire-Cognitive restructuring, ERQ-CR The Emotion Regulation Questionnaire for Children and Adolescent-Cognitive Reappraisal, ERQ-ES The Emotion Regulation Questionnaire for Children and Adolescent- Expressive Suppression

We presented the adjusted mean difference (AMD) at each time of post-intervention and follow- ups (Table 2).

Sensitivity power analysis

To quantify the power to detect the smallest between-group effect, a sensitivity power analysis was conducted using the realized post-intervention sample (n = 48 CG; n = 47 IG). A post-only cluster-ANCOVA framework was applied, with school as the clustering unit and baseline–post correlation (r = 0.36) used to adjust the residual variance. The observed school-level intraclass correlation coefficient (ICC) at post was approximately zero, resulting in negligible design effects.

Results

Demographic features

Twenty-one schools participated in this trial (11 in the Intervention Group and 10 in the Control Group). Figure 1 shows the flowchart of the study. Students at suicidal risk (C- SRRS score ≥ 3) was 20.7% in the CG and 18.8% in IG. A similar number of students in both groups were excluded because they had a high severity of depressive symptoms. 224 students and their parents/caregivers were interviewed: 123 in the Intervention Group and 101 students in the Control Group. After this step, more (n = 72) students in IG did not participate than in CG (n = 52), and the main reason was that students or parents refused to participate, especially in the IG.

Before the intervention, 100 secondary school students participated (47 in the IG and 48 in the CG). The number of students enrolled in both groups was comparable (See Table 1) regarding age, symptomatology, and psychological skills. Regarding students who entered the trial, there were more girls than boys in the control group. In both groups, most of the students had a history of receiving psychological or psychiatric treatment in the past or currently. The history of suicide attempts in the family was similar in both groups (29.4% in the IG and 30.6% in CG); however, the history of suicide in the family was slightly higher in the control group.

Program implementation

92.2% of students completed all sessions of the program (Attendance percentages are shown in App. 1, Suppl. Table 1). Four students did not complete the program at different times: one dropped out in the third session, another in the seventh session, one in the eighth session, and another in the tenth session. The time range to complete the sessions was between 13 and 19 weeks. Delays in this time were related to several reasons, such as students’ absenteeism and out-of-planning school activities (anniversaries, national holidays, and Chilean voting). Although there was slightly more student absenteeism when the sessions were online than when they were face-to-face (Suppl. Table 2), 100% of these two types of sessions were successfully rescheduled.

Intervention effectiveness at post-intervention and follow-up

Table 2 shows the overall intervention effect across follow-up assessments and time-specific results of primary and secondary outcomes. The general effectiveness of the intervention in the primary outcome, suicidal ideation, showed a reduction in the IG compared with the CG, although not significant.

Regarding the overall intervention effect across follow-up assessments on secondary outcomes, we found a significant reduction in depressive symptoms (AMD = −1.2; CI = −2.41 to 0.71), anxiety symptoms (AMD = −1.6; CI = −2.87 to 1.89), and psychotic experiences (AMD = −3.2; CI = −4.83 to −3.15). We did not find a significant overall intervention effect across follow-up assessments on the emotional regulation strategies; however, there was an increase in the cognitive re-appraisal skill (AMD IC) in the IG that was close to significance.

Regarding the time-specific effect, the intervention seems to have an effect immediately after the intervention and close to it (3- and 6 months), particularly for depressive and anxiety symptoms, hopelessness, behavioral activation and cognitive reappraisal. The change favoring the IG of cognitive reappraisal was also significant at the 10-month follow-up.

The sensitivity power analysis indicated that, given the realized post-intervention sample size (n = 48 CG; n = 47 IG), the observed baseline–post correlation (r = 0.36), and an ICC ≈ 0 at the school level, the study had 80% power (α = 0.05) to detect a between-group difference of approximately 8.5 SIQ points (Cohen’s d ≈ 0.57) at post. Effects smaller than this threshold would likely remain undetected.

Attrition

We completed a 10-month follow-up with the assessment of 72.5% in the IG and 71.4% in the CG. In most of the variables, there were no differences in baseline features between students who ended the follow-up versus those who dropped out from the study, except for suicidal ideation assessed with the CSSRS in the CG and cognitive restructuring in both groups (App. 2, Table 2).

Discussion

We conducted a cluster RCT to evaluate the effectiveness of a CBT-based intervention encompassing online and face-to-face sessions to reduce suicidal ideation among adolescents from secondary schools in Chile. The intervention (Reframe-IT +) reduced SI, but the overall intervention effect across follow-up assessments was non-significant. A post-hoc sensitivity power analysis suggested that, with the achieved sample size and observed variance parameters, the study had adequate power to detect between-group differences of approximately 8–9 SIQ points (d ≈ 0.6). Thus, while the sample was appropriate for assessing feasibility and moderate- to-large effects, smaller intervention effects could have remained undetected. We additionally tested the overall intervention effect across follow-up assessments on the secondary outcomes. We found significant reductions in depressive and anxiety symptoms and psychotic experiences in the IG compared to the CG. We additionally found time-specific effects on depressive symptoms at post-intervention and 3-month follow-up. Anxiety symptoms and hopelessness were significantly lower in the intervention group (IG) at post-intervention, 3- and 6-month follow-up. Regarding emotion regulation strategies, behavioral activation was significantly better in the IG only at post-intervention, and cognitive reappraisal was higher in the IG at post- intervention, 6-, and 10-month follow-up.

To the best of our knowledge, there are no other studies that have tested a blended indicated school-based intervention to reduce suicide among adolescents, such as our study, and few other studies have explored the benefit of guided and self-guided interventions to prevent suicidal ideation among adolescents and adults. In the case of adolescents, Robinson et al. [32], in a quasi-experimental study exploring the efficacy of Reframe-IT without the in-person component we used in Chile, found a reduction in suicidal ideation, depressive symptoms, and hopelessness at post-intervention. These reductions were observed at post-intervention mirroring our results [37, 56]. Later, Hetrick et al. [10], using the same intervention using RCT design found non-significant reductions in SI, depressive symptoms, and hopelessness in a small sample of adolescents in Australia at post-intervention and the 22-week follow-up compared with the control group (treatment as usual). Our findings are consistent with this latter study regarding the absence of significant effects on suicidal ideation. We initially reported a significant effect on suicidal ideation at the post-test without controlling for repeated measures after the intervention [56]. Even though our study had a larger sample size than previous studies with Reframe-IT [10] it is probably still underpowered to find significant between-group differences [57].

Additionally, differences in socioeconomic vulnerability, school context, and baseline psychosocial adversity between Australian and Chilean adolescents may also partly explain discrepancies in intervention effects. We observed similar results regarding interventions using a technological component among young adults. For example, Heuschen et al. [58] did not find a significant reduction in SI using an 8-session guided internet-based intervention based on training behavioral activation among individuals with depression (Mean age = 24.5 years) [58]. These results suggest that to increase the effect on suicidal ideation, interventions may need to include other suicidal-preventive components in a more comprehensive approach [19, 59]. This may reflect the fact that suicidal ideation is influenced by multiple interpersonal, contextual, and behavioral factors that are not necessarily modified through depression-focused CBT components alone. Therefore, future adaptations of the intervention may benefit from incorporating more suicide-specific components, including family-based strategies [60], bullying prevention [61], substance use interventions [62], and more personalized approaches according to adolescents’ psychosocial and psychopathological profiles [63].

We found a significant impact on depressive and anxiety symptoms and psychotic experiences. Others have also found positive results of digital mental health interventions on depression and anxiety among adolescents [64, 65]. One potential explanation may be related to the evidence that even though there is an established relationship between suicidal ideation and depression [11, 66], changes in depressive symptoms could be independent of changes in suicidal ideation when using a CBT-based intervention [67]. In addition, these positive effects could be related to the addition of the in-person component, which promoted more specific depression-related protective skills. This component was highly valued by students [37], reinforced, and allowed more practice of the skills promoted by the online sessions. It also could have enhanced the engagement of adolescents in the program, as shown in both attendance and attrition rates. When comparing with Hetrick et al. [10], we found a slightly higher attendance rate (92% vs 87.8% completed 100% of sessions), and lower attrition rates in the intervention group at similar follow-up times (IG: 17.6% at 24 weeks vs IG: 30.77% at 22 weeks). Concerning the difference in the absenteeism between online and face-to-face sessions, it is important to mention that the former are also delivered at schools in presence of a psychologist. As we mentioned, 100% of both digital and face-to-face sessions were successfully rescheduled. Therefore, it is probable that the absenteeism was not accounted for the format of sessions but for other external issues.

Strengths and limitations

One of the strengths of our study is that we included longer follow-up assessments, which allowed us to test the maintenance of the effect of the intervention [68]. This is relevant considering the few studies exploring the long-term effectiveness of blended interventions [69].

However, the study has some limitations. First, we had a small sample size, which may have reduced the precision and stability of the estimated intervention effects. [70]. As we have explained elsewhere [56], recruiting schools and participants was difficult. Schools usually have a high workload, especially after the pandemic, increasing the chances of rejecting invitations to conduct research in their facilities and foreseeing more work for the school staff. Parents, main caregivers, and adolescents may also have reasons to reject participation. Even though the interviewer was trained to communicate the objectives of the study clearly; a lack of understanding of the benefits and purpose of the intervention can also reduce interest [71]. Trust issues with researchers, the intervention, and previous negative experiences can further discourage participation [72]. Additionally, participants may be less inclined to engage if they do not see immediate benefits or have other competing health or personal priorities [72]. Social influence and skepticism from family can also significantly diminish interest in the intervention group. In addition, because this was an indicated preventive intervention, subjects with severe depressive symptoms (PHQ-8 score > 19) were excluded, which reduced the sample size even further. Although this seems to be the right approach, considering safety reasons, it does not solve the relevant issue that individuals with a higher risk for suicide are often excluded from the interventions, which should be addressed by future research [73]. Second, our results may not be generalized to all adolescents with different socio-economic backgrounds because we included only public schools. Third, we used self-report questionnaires. This may overestimate the prevalence of symptoms and influence the responses due to memory biases. However, we used valid and reliable measures, and the scores of the outcomes and mediating factors were similar between groups at baseline, reducing the measurement bias effect. Finally, other factors undoubtedly influence the symptoms of adolescents, such as family and social issues and substance use, which were not considered in the assessments or the intervention implemented.

More research is needed to consider some of these factors to be included in the preventive interventions among adolescents using a more comprehensive approach.

Some of the study’s limitations could be addressed by incorporating higher-risk populations into preventive strategies [74], an approach that is particularly relevant for adolescents presenting with severe depressive symptoms [11, 75]. This strategy may enhance the external validity and real-world applicability of the intervention, as these adolescents constitute a substantial proportion of the population in educational settings and are at elevated risk for maladaptive behaviors [76], as well as suicidal ideation and suicidal behaviors [77]. Another potentially suitable option is simultaneously delivering interventions for both adolescents and their caregivers, which has been associated with longer-term effects [78, 79].

Conclusions

Overall, the intervention seems to have some benefits in reducing common mental health symptoms, as shown in other studies [64], but less promising results in the prevention of suicidal ideation. Given the evidence that school-based prevention programs have a limited impact on suicidal behavior [73], schools should adopt best practices for suicide intervention to support students with suicidal thoughts or behaviors [80], and they can also provide other preventive interventions for bullying and family relationships to enhance results. A promising avenue is measuring outcomes determined by young people, incorporating participants with higher depressive symptom levels, and delivering interventions targeting both adolescents and caregivers.

Authors’ contributions

DN, JR and JG conceived the study. DN and JG supervised the whole project. DN and DM oversaw data collection. DN and JG analyzed the data and edited the final version. JR developed the original Reframe-IT intervention. All authors wrote the manuscript.

Funding

This work was funded by the Agencia Nacional de Investigación y Desarrollo (ANID), through grant FONDECYT 1210093, by ANID—Millennium Science Initiative Program— NCS2021_081 and by the Programa de Investigación Asociativa (PIA) en Ciencias Cognitivas (RU-158–2019), Centro de Investigación en Ciencias Cognitivas (CICC), Faculty of Psychology, Universidad de Talca, Chile. JG received funding from ANID/FONDECYT REGULAR Nº 1261819 and Center for the Well-Being and Development of Adolescents and Children in the Digital Age (BAND) (CIN250046). Moreover, the work was supported by a National Health and Medical Research Council Investigator Grant (APP2008460) and a University of Melbourne Dame Kate Campbell Fellowship.

Data availability

Data used on this study is available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Scientific Ethics Committee of the University of Talca (1210093; 05/12/2021).

Consent for publication

Not applicable.

Competing interest

The authors declare no competing interests.