Implementation of an Evidence-Based Psychotherapy for Trauma-Exposed Children in a Lower-Middle Income Country: the Use of Trauma-Focused Cognitive Behavioral Therapy in El Salvador

Abstract

Central American youth are at a high risk for experiencing trauma and related psychosocial problems. Despite this, few studies of evidence-based trauma-focused interventions with this population exist. The objective of this project was twofold: 1) to train providers in El Salvador in Trauma-Focused Cognitive Behavioral Therapy (TF-CBT) as part of a clinical implementation project within a non-governmental organization, and 2) to conduct program evaluation to determine the feasibility of implementation and the effectiveness of the treatment. Fifteen Salvadoran psychologists were trained in TF-CBT who then provided TF-CBT to 121 children and adolescents ages 3–18 in community-based locations. The mean number of traumas reported by youth was 4.39. Results demonstrated large effect sizes for reduction in youth-reported trauma symptoms (Cohen’s d = 2.04), depressive symptoms (Cohen’s d = 1.68), and anxiety symptoms (Cohen’s d = 1.67). Our program evaluation results suggest that it was feasible to train providers in TF-CBT, that providers were in turn able to deliver TF-CBT in community-based settings, and that TF-CBT was an effective treatment option to address trauma-related concerns for youth in El Salvador. This project is an important first step in the dissemination and implementation of evidence-based trauma-focused treatment for youth in Latin American countries.

Each year, more than half of youth around the world experience some type of physical, emotional, or sexual victimization (Hillis et al. 2016). The limited data available suggest that 30–60% of children in Latin America have experienced a traumatic event during their lifetime (Speizer et al. 2008). Latin American countries (including Central and South American countries) have a long-standing history of political instability, gang-related violence, poverty, and forced migration that has further exposed youth to potentially traumatic events (Wirtz et al. 2016). In particular, El Salvador has the highest rate of homicides in the world, with the United Nations reporting 61.8 deaths per 100,000 in 2018 (United Nations on Drugs and Crime 2019; World Population Review 2019). Approximately 60,000 gang members are present in at least 94% of the country’s municipalities, extorting and gathering intelligence on residents in order to enforce their territories’ borders (Human Rights Watch 2020). As a result, 246,000 Salvadorans were internally displaced in 2018 due to conflict and violence (World Bank Group 2019).

The broader literature illustrates that exposure to trauma increases the risk for a variety of mental and behavioral health problems in children and adolescents, including posttraumatic stress disorder (PTSD), depression, anxiety disorders, externalizing problems, and substance use disorders (Carliner et al. 2016; McLaughlin et al. 2012). In addition, trauma can also negatively affect academic achievement (Holt et al. 2007). The risks associated with poor mental health extend beyond childhood and predict decreased physical health, increased morbidity, and reduced educational, occupational, and financial achievement in adulthood (Härter et al. 2003; Kessler et al. 1997; Kessler et al. 1995; Lopez et al. 2006; Ormel et al. 2008). Effective mental health interventions for youth are therefore crucial for supporting resilient outcomes in children exposed to trauma.

Trauma-Focused Cognitive-Behavioral Therapy (TF-CBT; Cohen et al. 2017) is an evidence-based, manualized treatment protocol utilized to treat posttraumatic stress symptoms in children and adolescents ages 3–18 and their supportive and non-offending caregivers. The treatment is divided into nine treatment components summarized by the acronym PRACTICE: (1) psychoeducation, (2) parenting skills, (3) relaxation skills, (4) affective modulation skills, (5) cognitive processing skills, (6) trauma narration and processing, (7) in-vivo mastery of trauma reminders, (8) conjoint child-parent sessions, and (9) enhancing future safety and development. Although TF-CBT is typically delivered in 12–25 weekly sessions, the exact number of sessions needed to complete treatment is dependent upon how long it takes each child and caregiver to work through each of the aforementioned components. TF-CBT has robust evidence with 23 randomized trials demonstrating its effectiveness and efficacy in improving PTSD and other trauma-related symptoms across a variety of U.S. and global service settings (Cohen et al. 2017, pp. 74–80; de Arellano et al. 2014), including with youth in low- and middle-income countries like Zambia (Murray et al. 2013), the Democratic Republic of the Congo (McMullen et al. 2013; O’Callaghan et al. 2013), and Kenya and Tanzania (Dorsey et al. 2020). In the United States, TF-CBT applications have also been developed for Hispanic youth (de Arellano et al. 2012) and American Indians and Alaska Natives (BigFoot and Schmidt 2010), but the effectiveness of such applications has yet to be systematically investigated with randomized controlled trials.

To date, TF-CBT has not been systematically implemented or evaluated in any Central or South American country. This represents a significant science-to-practice gap for several reasons: 1) TF-CBT has been successfully implemented in other low- and middle-income countries (e.g., Murray et al. 2013; O’Callaghan et al. 2013); 2) TF-CBT has been shown to be effective for treating trauma and has already been applied for Hispanic populations in the U.S. (de Arellano et al. 2012); 3) trauma exposure is highly prevalent in certain Central and South American countries, such as El Salvador (United Nations on Drugs and Crime 2019; World Bank Group 2019; World Population Review 2019; Wirtz et al. 2016); and 4) access to evidence-based treatments for youth in low- and middle-income countries, including El Salvador, is limited (Wang et al. 2007). Unfortunately, El Salvador has particularly low access to mental health services due to the lack of legislation dedicated to mental health, limited expenditures on mental health, and the fact it has the the lowest number of mental health facilities and providers in Central America (World Health Organization 2011). These barriers highlight not only the need for trauma-focused treatment for Salvadoran youth, but also for creative approaches to implementation in order to increase access to care.

The present project was part of clinical implementation programming within a non-governmental organization (NGO), which provided TF-CBT at five schools and one community mental health clinic in San Salvador, El Salvador. The program focused on training Salvadoran mental health providers in TF-CBT in order to build local workforce capacity for the provision of evidence-based trauma focused treatment for youth. A secondary purpose of the project was to conduct program evaluation examining the feasibility of implementing TF-CBT in El Salvador and the potential effectiveness of the intervention with Salvadoran youth and their caregivers. The primary aim of the current paper is to describe the results of this program evaluation.

Method

Setting

The project was conducted in two bordering municipalities in El Salvador: La Libertad and San Salvador. Children and adolescents were provided treatment at one of six locations: five schools in the municipalities of Santa Tecla and San Salvador, and one community-based mental health clinic in the municipality of San Salvador.

Institutional Review Boards (IRBs) at our institution (U.S. University) and in El Salvador were consulted. The IRBs determined the project to be part of quality improvement efforts (as part of a clinical-focused service program) and thus did not require IRB approval. Nonetheless, the research team maintained close communication with local university and organizational partners and informed all stakeholders of each effort’s progress and findings on a regular basis.

Participants

One hundred thirty-one Salvadoran children and adolescents aged 3 to 18 were referred for treatment of posttraumatic stress symptoms. Inclusion criteria for the current project were as follows: 1) patient age between three and 18 years; 2) at least one remembered/reported trauma; 3) significant current posttraumatic stress symptoms, defined by a) ≥ three current PTSD-DSM-5 symptoms reported by child or caregiver during a clinical interview, and b) child’s score of ≥20 on the Child PTSD Symptom Scale-5 (CPSS-5); and 4) caregiver consent and child assent. Exclusion criteria were 1) current significant suicidal ideation; 2) substance use disorder; 2) active psychotic symptoms; 3) pervasive developmental delays; or 4) currently living with or in close contact with the perpetrator of the abuse. Ten youth were deemed ineligible for the project based on exclusion criteria: four youth were deemed to not need trauma treatment due to primary presenting concerns better explained by non-trauma-related anxiety, and six youth were experiencing ongoing trauma exposure within their home (e.g., exposure to domestic violence or were living with the perpetrator of the domestic violence at the time of screening). Youth who were experiencing ongoing trauma exposure within the home were either referred for alternative supportive services if these were available, or provided supportive services by the TF-CBT trained therapist that focused on the Enhancing Safety component of treatment (e.g., creating a safety plan). Two children were initially excluded due to active suicidal ideation. These children were referred for psychiatric services and once their suicidal ideation had remitted, they were enrolled in the project, resulting in a final sample of 121 youth who received TF-CBT treatment.

The sample had a mean age of 11.96 years (SD = 3.65) with the following demographic distribution: 47.1% female, 100% Hispanic (Spanish speaking). The most common index traumas included: traumatic loss (24.0%), witnessing domestic violence (16.5%), physical abuse (14.9%), and sexual abuse (13.2%; see Table 1 for complete information about index traumas). Youth in this sample reported experiencing an average of 4.39 traumatic events (SD = 2.30, range 1–11), with 90.7% experiencing more than one traumatic event in their lifetime. The vast majority had at least one identified caregiver involved in treatment; only six (4.6%) participated without a caregiver. Caregivers involved in treatment were primarily mothers (58.7%), followed by grandmothers (9.9%) and fathers (6.6%). All caregivers spoke Spanish as their primary language.

Table 1.

Index traumas reported by study sample

Trauma type	N (%)
Traumatic Loss / Grief	29 (24.0%)
Domestic Violence	20 (16.5%)
Physical Abuse	18 (14.9%)
Sexual Abuse	16 (13.2%)
Serious Accident	12 (9.9%)
Community Violence	9 (7.4%)
Death threat	3 (2.5%)
Assault	3 (2.5%)
Stressful Medical Procedure	2 (1.7%)
Separation	2 (1.7%)
Abandonment	2 (1.7%)
Bullying	2 (1.7%)
Other	3 (2.5%)

Measures

Spanish Version of the Child PTSD Symptom Scale for DSM-5, Self-Report

The CPSS-5 (Foa et al. 2018) is a 27-item DSM-5 adaptation of the original CPSS (Foa et al. 2001; Kataoka et al. 2003) that measures symptoms of posttraumatic stress with the first 20 questions, and related functional impairment with the final seven questions. The English version of the instrument has strong psychometric properties, including high reliability for the total score (α = 0.92), convergent and divergent validity, and diagnostic sensitivity (0.93) and specificity (0.82) when evaluated using a clinical cutoff score of 31 (Foa et al. 2018). Although there are no published psychometric studies of the Spanish language version of the CPSS-5, the Spanish language version of the original CPSS demonstrated strong psychometric properties with high reliability of the total score (α = 0.90) and moderate to good reliability for the subscales (α = 0.70 to α = 0.80; Serrano-Ibáñez et al. 2018). For the present study, the Spanish version of the CPSS-5 was reviewed by a team of bilingual psychologists and compared with the English language version of the instrument to provide item-level refinement of any notable discrepancies. Following this process, a team of monolingual Spanish speaking psychologists in El Salvador reviewed the Spanish language version of the instrument to ensure that terms were understandable and fit the specific vocabulary utilized in El Salvador. In the current project, reliability of the overall total posttraumatic stress scale (i.e., the first 20 items) was adequate (α = 0.79). The first 20 items of the scale measure posttraumatic stress symptoms are the only items included in the overall posttraumatic stress scale reliability analysis (the other seven items measure functional impairment).

Spanish Form of the Revised Child Anxiety and Depression Scale – Short Version

The RCADS-SV (Ebesutani et al. 2012) is a 25-item measure of broad anxiety and depression symptoms. The measure has demonstrated a stable 2-factor structure with adequate reliability of subscales across both school-based (α: 0.80–0.86) and clinical (α: 0.79–0.91) samples. Additionally, scores were convergent with domain-specific diagnosis in a group of clinically referred individuals (Ebesutani et al. 2012). The instrument’s Spanish translation was based on items from the full-length parent-report examination conducted by Park et al. (2016). Numerous other translations of the parent- and child-report formats of both the full-length and 25-item RCADS are freely available at the original author’s lab website, which also offers norm-based scoring spreadsheets (https://www.childfirst.ucla.edu/resources/). The instrument was tailored for Spanish used in El Salvador using a similar process as described above. Reliability of the subscales in the current sample was similar to the original instrument development study (Anxiety α = 0.79; Depression α = 0.72).

Procedure

Potential participants included 131 consecutive referrals for treatment between February to June 2019 at five schools and one community-based mental health clinic. Five psychologists saw participants at a community mental health clinic and 10 psychologists provided school-based treatment. Children and caregivers completed an assessment prior to beginning treatment, which was standard practice for the implementing agency. The assessment included the use of clinical interviews and administration of questionnaires (i.e., CPSS-5 and RCADS-SV) with both the child and caregiver. The same questionnaires were also administered at the time of treatment termination (within one week of the last treatment session).

Child Safety Protocol

A safety protocol that was appropriate for the local Salvadoran context and informed by key stakeholders in El Salvador (i.e., schools, implementing NGO, and psychologists) was developed and implemented. Training materials for psychologists implementing TF-CBT were developed and tailored to the treatment setting (school vs. clinic), including steps for assessing the severity of suicidal/homicidal risk, seeking consultation, working with caregivers for safety planning, and identifying appropriate community resources. Country standards and legal requirements for mandated reporting were followed in cases of current reported or suspected child abuse.

Treatment Training, Supervision, and Consultation

An initial three-day TF-CBT training was provided for 15 psychologists in January 2019 via a learning collaborative model (Bunger et al. 2016). The training was led by a bilingual clinical psychologist (second author), who is licensed in the United States and has conducted prior TF-CBT trainings in Spanish in collaboration with a Nationally Certified TF-CBT bilingual trainer (last author). A follow-up two-day training was held four months later in May 2019 as providers transitioned to implementing the second phase of treatment (i.e., trauma exposures and family sessions) with participants. TF-CBT trainings were conducted in-person, in Spanish, were interactive, and included several strategies to increase and maintain engagement, including demonstrations, roleplays, activities, and raffles to win therapy resources. Psychologists received all materials and handouts in Spanish. During the trainings, the TF-CBT trainers also assessed clinicians’ baseline knowledge and skills related to evidence-based interventions. This information helped the implementation team to tailor trainings and consultation calls by providing additional opportunities to review and highlight the key mechanisms of change for TF-CBT.

Treatment was provided by 15 psychologists who were licensed to provide mental health services in El Salvador and had completed the equivalent of a bachelor’s degree in psychology (requirements for a psychologist in El Salvador). The psychologists were employed by a local organization that provided mental health services via a community-based mental health clinic in San Salvador and at local schools. Over the span of the project, psychologists participated in 14 bi-weekly group consultation calls (February – August 2019). The psychologists were split into two groups for consultation calls (one group of seven and another group of eight) in order to make calls more efficient and comprehensive. Consultation calls involved review of treatment progress through the TF-CBT components, behavioral rehearsal of techniques, issues of risk/safety, specific challenges/barriers to implementing TF-CBT in the schools and clinic contexts, and self-care (please see Orengo-Aguayo et al. 2020 for a more in-depth description of the training procedures).

Intervention

See the Introduction for a detailed description of Trauma-Focused Cognitive-Behavioral Therapy (TF-CBT). For the current project, psychologists began delivering TF-CBT in March 2019 and completed treatment in August 2019. Each psychologist dedicated 16 h per week to the project (two workdays). TF-CBT was delivered in Spanish and incorporated the use of materials translated and compiled by our team (e.g., example agendas, worksheets, informational child and caregiver handouts). Psychologists delivered weekly individual sessions with children and caregivers, along with conjoint child-caregiver sessions. For children who participated without a caregiver, conjoint child-caregiver sessions were not included in the treatment.

Intervention Fidelity

Psychologists completed weekly, online surveys documenting TF-CBT sessions with youth and caregivers. For each session, psychologists reported the following: 1) TF-CBT component(s) covered, 2) techniques/activities used, 3) skills taught, 4) length of session, 5) barriers/challenges encountered, and 6) plans for the next session. Feedback was regularly provided to psychologists regarding their adherence to the treatment protocol (based on psychologists’ self-report). In addition, psychologists presented their TF-CBT cases during group consultation calls and were provided with feedback to help ensure adherence to the treatment model.

Cultural Applications

Cultural applications were incorporated throughout the project. For instance, training and materials were all tailored and delivered in Spanish by bilingual and bicultural trainers. Safety protocols were co-developed with local experts’ knowledgeable of child protection regulations and laws. TF-CBT components were tailored to fit the local context by using relevant cultural examples (e.g., teaching parenting skills that incorporated cultural norms of “respeto” and “familismo”). Trainers took the time to build a strong connection with the therapists (e.g., providing praise for accomplishing project goals, being available to consult via WhatsApp, checking in on well-being and family life during consultation calls, offering longer breaks during training to connect with peers, and participating in culturally appropriate exchanges of affect such as hugs or high-fives after accomplishing project goals).

Results

Data Analysis

Descriptive statistics were generated detailing frequencies of demographic characteristics, trauma exposure, and treatment completion among participants. Changes in CPSS-5 and RCADS-SV scores from pre- to post-treatment were evaluated with paired sample t-tests for treatment completers.

Treatment Training and Intervention Fidelity

All psychologists attended both the initial and follow-up TF-CBT trainings and biweekly consultation calls. Review of psychologists’ weekly self-report of intervention fidelity demonstrated that psychologists consistently followed the treatment protocol and implemented appropriate components of the model. However, at the beginning of the project, several psychologists began to implement the treatment components out of order. By reviewing the fidelity measures and through regular consultation calls, project trainers were able to correct this error and assist the psychologists with adherence to the treatment protocol.

Treatment Completion and Number of Sessions Attended

One-hundred four children out of 121 enrolled in the project (86.0%) successfully completed all components of TF-CBT. Seven children (5.8%) who began treatment were unable to complete treatment because the family moved or the child withdrew from his/her school. Two children (1.7%) did not complete TF-CBT due to a medical issue that required prolonged hospitalization. One child (0.8%) was unable to complete treatment due to ongoing exposure to trauma within the home. Finally, seven children (5.3%) dropped out of treatment for unknown reasons. The mean number of sessions attended by the youth who completed treatment was 14.30 (SD = 3.15, range 8–21) and caregivers attended a mean number of 6.34 sessions (SD = 3.57, range 1–16).

Clinical Outcomes

Table 2 provides a summary of pre-post change on clinical outcomes of interest for children who competed treatment. Mean pre- and post-treatment CPSS-5 self-report scores were 44.39 (SD = 13.04) and 9.80 (SD = 10.28) respectively. This difference was clinically and statistically significant, t(93) = 19.78, p < .001, Cohen’s d = 2.04. Mean pre- and post-treatment CPSS-5 caregiver-report scores were 45.44 (SD = 14.94) and 7.51 (SD = 7.51) respectively. This difference was also clinically and statistically significant, t(93) = 21.60, p < .001, Cohen’s d = 2.23. Mean pre- and post-treatment RCADS-SV depression subscale self-report t-scores were 62.72 (SD = 12.65) and 39.18 (SD = 8.74) respectively. This difference was clinically and statistically significant, t(93) = 16.28, p < .001, Cohen’s d = 1.68. Mean pre- and post-treatment RCADS-SV depression subscale caregiver-report t-scores were 73.90 (SD = 10.10) and 45.67 (SD = 8.16) respectively. This difference was clinically and statistically significant, t(93) = 22.07, p < .001, Cohen’s d = 2.28. Mean pre- and post-treatment RCADS-SV anxiety subscale self-report t-scores were 62.10 (SD = 11.86) and 39.51 (SD = 10.20) respectively. This difference was clinically and statistically significant, t(93) = 16.19, p < .001, Cohen’s d = 1.67. Mean pre- and post-treatment RCADS-SV anxiety subscale caregiver-report t-scores were 70.80 (SD = 12.41) and 44.17 (SD = 9.10) respectively. This difference was also clinically and statistically significant, t(93) = 18.55, p < .001, Cohen’s d = 1.91.

Table 2.

Changes in clinical outcomes reported by youth and caregivers

	Pretest		Posttest
Outcome	M	SD	M	SD	t	df	p	d
CPSS-5 (youth)	44.39	13.04	9.80	10.28	19.78	93	<.00001	2.04
CPSS-5 (caregiver)	45.44	14.94	7.51	7.51	21.60	93	<.00001	2.23
RCADS-SV Depression Subscale (youth)	62.72	12.65	39.18	8.74	16.28	93	<.00001	1.68
RCADS-SV Depression Subscale (caregiver)	73.90	10.10	45.67	8.16	22.07	93	<.00001	2.28
RCADS-SV Anxiety Subscale (youth)	62.10	11.86	39.51	10.20	16.19	93	<.00001	1.67
RCADS-SV Anxiety Subscale (caregiver)	70.80	12.41	44.17	9.10	18.55	93	<.00001	1.91

CPSS-5 = Child Posttraumatic Stress Scale for DSM-5. RCADS-SV = Revised Children’s Anxiety and Depression Scale-Short Version. Cohen’s d = 0.2 represents a small effect size, d = 0.5 represents a medium effect size, d = 0.8 represents a large effect size

Discussion

To our knowledge, this project is the first documented implementation of Trauma-Focused Cognitive Behavioral Therapy in a Central American country. The results supported: 1) the need for trauma-focused services in countries like El Salvador (131 youth were referred for treatment within four months of the project start date, resulting in a wait-list due to the need surpassing the project’s resources); 2) the feasibility of training local providers in TF-CBT (with all materials, training, and support provided in Spanish); 3) the feasibility of implementing TF-CBT not just in a community clinic, but also within local schools (thus increasing access to mental health services for underserved youth); 4) the feasibility of delivering all TF-CBT components within a six- to seven-month window (86% of youth completed all TF-CBT components and the average number of completed sessions was comparable to studies of TF-CBT delivered in community-based settings in the United States and Canada (Cohen et al. 2011; Konanur et al. 2015); and 5) the effectiveness of TF-CBT in reducing PTSD, depression, and anxiety symptoms in trauma-exposed youth in El Salvador. Effect sizes for pre- to post-treatment reductions in PTSD symptoms were in the large range (Child Report: Cohen’s d = 2.04; Caregiver Report: Cohen’s d = 2.23) and were comparable or superior to those found in studies conducted in the United States (Rubin et al. 2017) and in other low-and-middle-income countries (O’Callaghan et al. 2013). Large and statistically significant pre- to post-treatment effects were also found for anxiety and depressive symptoms, as reported by youth and caregivers.

This project also highlights the significant childhood adversities experienced by children in El Salvador. Most children (90.7%) reported exposure to more than one traumatic event in their lifetime. With regard to the type of traumatic event experienced, most children reported an index trauma of traumatic loss, followed by witnessing domestic violence, experiencing physical abuse, and experiencing sexual abuse. This trauma profile is consistent with childhood adversities experienced in other low- and middle-income countries (Dorsey et al. 2020), and is expected in light of the significant exposure to armed conflict, gang-related violence, and homicide rates of many Central American countries (United Nations on Drugs and Crime 2019; World Population Review 2019). The high caregiver participation in treatment is also worth noting. Caregiver participation is an important component of TF-CBT, but also a simultaneously challenging issue in child mental health treatment in terms of effectively engaging caregivers, particularly in a school-based environment (Gopalan et al. 2010). In this project, 95% of children had a caregiver involved in treatment.

This project provides further evidence that TF-CBT is an effective treatment for trauma-exposed youth in low- and middle-income countries and adds to the literature by documenting the first successful and effective implementation of TF-CBT in a Spanish-speaking, Central American country. It should be noted that most children endorsed more than one traumatic experience in their lifetime, documenting the need for timely access to culturally- and linguistically-competent trauma-focused, evidence-based treatments to reduce risk of future mental health problems. The results also provide some support for the use of TF-CBT with Spanish-speaking, immigrant families from Central America within the United States.

Implications

Researchers in the U.S. seeking to implement treatment programs in low-resourced settings, like the one described in this paper, should assess clinicians’ baseline knowledge and skills related to evidence-based interventions. This information can help implementation teams tailor trainings by providing additional opportunities to review and highlight the key mechanisms of change for a particular treatment. For clinicians interested in providing trauma-informed services with TF-CBT in low-resourced settings, we strongly recommend obtaining regular supervision by an experienced TF-CBT clinician, as we did in the current project, so that one may build competence. It is also likely that supervision and consultation would need to be conducted remotely, given those clinicians with extensive TF-CBT experience are mostly located in the U.S. We also recommend clinicians in low-resourced settings seek the support of local community organizations or schools with existing connections to families to facilitate the referral process and enhance treatment retention. We found that in the current project, this support was important to the success of the project. Clinicians and implementation teams should also evaluate the need for cultural and linguistic adaptations of trauma symptom measures for their Central American clients, given potential variability in trauma types and levels of exposure compared to traumas in the U.S.

Limitations

Despite the promising positive findings, several limitations should be noted. First, this was not a controlled trial. These results are part of a project whose primary purpose was not rigorous research, but rather the training and implementation of TF-CBT amongst local providers. As such, inclusion of a comparison group or waitlist control group was not possible within this project. The absence of a control group raises the question as to whether the effects are attributable to TF-CBT components or to non-specific factors such as the passage of time, provider contact, or treatment expectations. Factors related to the therapist-client relationship were not examined in this project. There is a possibility that these factors may have influenced outcomes beyond treatment model effects. Additionally, evaluators were the treating therapists, which may have introduced a potential source of bias. Finally, the small sample size limits the generalizability of these findings to other populations.

The project, however, has several strengths. This project was implemented under “real-world” conditions, with actual community providers, within local schools and a community mental health clinic, with a community sample. This approach helped to develop local capacity within the country to provide TF-CBT beyond the duration of this project. The delivery of TF-CBT within local schools increased access to mental health services to children who may otherwise have never received it, likely contributing to the significant treatment completion rate.

Future Directions

Future studies should focus on experimental designs that include a control group, randomization, independent assessors blinded to condition, and comparison to other trauma-focused interventions to determine the most effective intervention for this population. A randomized controlled trial would allow for closer monitoring of fidelity and systematic recording of implementation strategies, which are critical considering the lack of resources and trained professionals in low- and middle-income countries. This project is an important first step in the dissemination and implementation of evidence-based trauma focused treatment for youth in Central and South American countries.

Funding

This work was supported by a United States Agency for International Development grant (AID-519-C-13-00002). The funding agency had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit this paper for publication.

Compliance with Ethical Standards

Disclosure of Interest

Drs. Cohen and Mannarino receive royalties for copyrighted TF-CBT books and materials. All other authors declare that they have no conflicts to report.

Ethical Standards and Informed Consent

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation [institutional and national] and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was not obtained from patients, only consent for treatment.