Positive Adaptations for Trauma and Healing (PATH), a Pilot Study of Group Therapy With Latino Youth

Abstract

Objective:

This study examines the acceptability and preliminary efficacy of Positive Adaptations for Trauma and Healing (PATH), a manualized treatment for Latino youth and their caregivers. PATH is a culturally adapted program that incorporates a trauma model, positive psychology, and resilience.

Method:

Latino youth (N = 16) recruited from an urban community clinic participated in PATH with their caregiver. Pre- and postintervention measures on trauma symptoms, resilience, depression, caregiver’s view of their youth’s well-being, and positive and negative emotions were gathered. Following the intake meetings (1 to 3), the families participated in 10 90-minute weekly group sessions (total of 3 groups). Caregiver groups were conducted in Spanish, and youth in English.

Results:

At pretest, 56% of the youth endorsed clinically significant symptoms on the UCLA PTSD Index (M = 34.2, SD = 11.2); the percentage dropped to 0% at posttest (M = 17.3, SD = 7.6). Youth reported pre- to posttest reductions on the Child Depression Inventory (mean difference [Mdiff] = 7.3; p = .004) and externalizing (Mdiff = 6.1; p < .001) and internalizing (Mdiff = 9.4; p < .001) behaviors on the caregiver-reported Child Behavior Checklist. Overall, there was high treatment engagement (93% attendance over 10 weeks).

Conclusion:

This novel treatment engaged a community-based Latino sample. The results suggest high acceptability and significant reduction in trauma symptoms and associated symptoms. This study included a small number of participants and results should be interpreted with caution. Future iterations will target larger number of participants to further assess feasibility.

Complex trauma refers to multiple exposures to various types of traumatic events, such as experiencing abuse and/or neglect and witnessing violence, which may occur in the caregiving system (Spinazzola et al., 2005). According to Stolbach and colleagues (2013), repeated and cumulative traumatic experiences within a comprised caregiving system contribute to affective, physiological, attentional, and behavioral impairments, as well as challenges with oneself and relational systems. In fact, cumulative childhood traumatic and stressful events (TSEs) have been linked to psychiatric diagnoses (Rutter, 2006) and impaired cognitive functioning (Liaw & Brooks-Gunn, 1994). Furthermore, evidence suggests that experiencing four or more TSEs puts an individual at severe risk for substance use and suicide in adulthood (Dube, Felitti, Dong, Giles, & Anda, 2003).

Although Latino children experience high rates of child maltreatment, Hinton, Rivera, Hofmann, Barlow, and Otto (2012) noted that there remains a critical lack of trauma treatment interventions that are optimal for traumatized ethnic minorities. For example, Suarez-Morales and colleagues (2007) noted that among community mental health organizations, very few rigorously evaluated their methods with Spanish-speaking populations. The lack of linguistic capacity to serve Spanish-speaking clients is a major issue.

Trauma Exposure Among Latino Youth

The U.S. Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, Children’s Bureau (2015) noted that in terms of maltreatment reports, Latino youth are overrepresented. The issue of trauma exposure among Latino youth is a critical issue for the mental health field, as one in every four children in the United States identifies as Latino, with this number projected to rise (Federal Interagency Forum on Child and Family Statistics, 2014). Thus, the need for mental health services among the Latino community will rise sharply in coming years.

Latinos experience unique challenges in the United States related to trauma exposure that includes disproportionate residence in urban areas with community violence (Kliewer & Lepore, 2015) as well as risky migration experiences (Alcántara, Chen, & Alegria, 2014). Chen (2010) identified that living in neighborhoods with a high rate of community violence caused youth living in these areas to experience poor psychological outcomes (anxiety and depression) and negative behavioral health risk (substance use and externalizing behaviors). Finkelhor, Shattuck, Turner, and Hamby (2014) identified that Latino youth are more likely to live in disadvantaged neighborhoods and experience community violence compared with youth of other racial/ethnic groups.

Many Latinos face unique stressors tied to the immigration process that include adjustment to a new culture, shifts in family structure and dynamics, ongoing stressors in the new country, and, for some, financial difficulties and lack of documentation to reside and work in the United States (Phipps & Degges-White, 2014). Fortuna, Porche, and Alegria (2008) noted that Latino youth who recently immigrated to the United States were at risk for poor health outcomes because of experienced traumatic events in their country of origin or during the immigration process. For example, among a community sample of school-aged immigrants (N = 1,004), 76% reported exposure to violence in their home countries (Jaycox et al., 2002). Foster (2001) identified four stages of trauma-related vulnerabilities for Latino immigrants that include premigration trauma, trauma during transit, early settlement in the country, and permanent settlement. For example, an individual fleeing Central America may have experienced trauma tied to political violence during the premigration phase. This same individual, during the physical journey, may witness the death of a companion, and during early settlement, may face exposure to racism (Yakushko, 2009). Lastly, in the fourth stage of permanent settlement, this individual may encounter difficulties with acculturation to the United States.

Crouch, Hanson, Saunders, Kilpatrick, and Resnick (2000) conducted a national examination of trauma-exposure in U.S. adolescents and found that Latino youth (N = 2,011) endorsed high rates of both physical (20.7%) and sexual (9.9%) assault. Increased exposure among youth is an important issue to consider because traumatic experiences have been linked to negative outcomes, including depression (Phipps & Degges-White, 2014) and delinquency acts (Turner, Shattuck, Finkelhor, & Hamby, 2016). In a random-digit dial telephone survey of 3,196 youth residing in California (Ages 12 to 17), Mikolajczyk, Bredehorst, Khelaifat, Maier, and Maxwell (2007) found that 10.5% of Hispanic youth experienced “moderate depression” compared with only 5.5% of White youth. A review of these investigations is important because they highlight high rates of trauma exposure and symptomatology among Latino youth.

Culturally Responsive Treatment in Health Care Settings

The National Health Care Disparities Report (Agency for Health Care Research & Quality, 2006) showed that the most important variables associated with health care disparities continue to be race, ethnicity, and socioeconomic status. Latinos have shown a high need for specialty services, but are less likely to receive them (Bridges, de Arellano, Rheingold, Danielson, & Silcott, 2010). Aggarwal, Cedeño, Guarnaccia, Kleinman, and Lewis-Fernández (2016) noted that Latino engagement in health care settings is a major public health problem.

As the Latino population continues to grow, there is a great need for culturally responsive interventions to effectively engage and treat Latino youth and their families (Falicov, 2009). Researchers have identified enduring disparities in treatment for Latinos with high rates of early dropout and missed appointments (Olfson, Cherry, & Lewis-Fernandez, 2009). Further barriers include the lack of ethnic/racial matching (Cook, McGuire, & Miranda, 2007), ability to provide services in Spanish, mistrust of the client toward the health care system, and inadequate services such as insufficient tailoring of treatment to cultural background of family and inex-perienced professionals (Alegría et al., 2008). Religion and spirituality have also been identified as potential barriers (Falicov, 2009), wherein some Latinos rely on religious leaders over a mental health provider. Furthermore, Lewis-Fernandez, Das, Alfonso, Weissman, & Olfson, 2005 found that among their Latino sample, participants reported fear of deportation as a barrier to seeking mental health care.

Research has indicated that when working with highly traumatized and ethnic minority clients, there needs to be an emphasis on techniques that address both somatization and psychological flexibility (Hinton, Pich, Hofmann, & Otto, 2013). The expression of stress through somatization is thought to be common among highly traumatized and recent immigrants. (Hinton & Lewis-Fernández, 2011). Reason being, traumatic experiences coupled with ongoing toxic stress related to the recent immigrant experience (e.g., discrimination and acculturation) causes a need to express symptomology. Hinton et al. (2013) also identified that psychological inflexibility is thought to be a key stressor and important component to target with populations dealing with posttraumatic stress disorder (PTSD). Hinton and colleagues described psychological flexibility as an adaptive processing mode. This is important for Latina/o populations, given that they navigate two languages and cultures and different social contexts. Disruption in psychological flexibility—such as experiencing multiple traumas—makes it difficult to navigate these environments (Koster, De Lissnyder, Derakshan, & De Raedt, 2011).

Hinton and colleagues (2012) elaborated upon ways to make cultural adaptations to cognitive behavioral therapy (CBT) for ethnic minority clients that have experienced numerous forms of trauma. They noted key concepts that can be summarized with four main points. Ethnic minorities tend to have a high dropout rate in therapy, which Hinton et al. noted may be in part due to prolonged exposure to the traumatic event during therapy. The first key concept is attending to culturally appropriate ways to expose the client to their trauma. A suggestion by the authors was to reduce the number of treatment session to no more than three to five therapeutic sessions that focus on the trauma exposure. Second, they placed an emphasis on emotion exposure and emotion regulation techniques, including the utilization of emotion regulation techniques from the client’s religious or cultural healing traditions. They discussed that the third way to make cultural adaptations included promoting emotional and psychological flexibility, utilizing techniques such as mindfulness. Lastly, the authors noted that they place an importance on addressing worry and generalized anxiety disorders, such as catastrophic cognitions about anxiety and PTSD symptoms, cultural syndromes, somatic symptoms, sleep-related phenomena, and culturally indicated transitional rituals. Thus, across the board, culturally informed interventions are needed to serve the needs of the greater Latino community.

Trauma-Focused Group Therapy

Group therapy has many benefits for a community-based clinic, as they allow for easier access to treatment when clinics have no capacity for new clients (Deblinger, Pollio, & Dorsey, 2016). The group format also provides a benefit in that some clients may only need a small window of therapy (e.g., 8 to 10 weeks), whereas a smaller subgroup may use the group treatment as an induction to therapy. Debelinger and colleagues further indicated that youth may experience a benefit from the group format in that therapy becomes destigmatized. Furthermore, the group format can be especially positive for ethnic minorities when clients are able to share similar cultural values, traditions, and the same language as the other group members.

A general critique of exposure-based interventions is their applicability in a group setting. Deblinger, Stauffer, and Steer (2001) utilized trauma-focused cognitive behavioral therapy in a group format with sexually abused children, Ages 2.5 to 10 years, and their nonoffending mother. The intervention helped mothers overcome abuse-related distress and intrusive thoughts, and children showed improvement in body safety knowledge. Although there was no significant difference on PTSD symptoms, the authors hypothesized that assessing changes in trauma scores was compromised due to the group format and age of the children. In order to address this issue, Deblinger and colleagues (2016) suggested that individual therapy be incorporated midway through the group program to allow for trauma narratives.

The current study addresses the gap in Latino mental health research by evaluating the efficacy of a manualized group treatment for Latinos who experienced trauma. Based on Hinton and colleagues (2012), we modified existing programs to develop Positive Adaptations for Trauma and Healing (PATH), a culturally adapted treatment that was created to address functional impairments of early onset and chronic trauma. It combines an existing trauma treatment— Attachment, Self-Regulation, and Competency (ARC), developed by Kinniburgh and Blaustein (2005)—with intervention strategies created to boost resilience and positive affect (Moskowitz et al., 2014). Although the operational definition of resilience varies, in the context of trauma and for the purpose of the current study, resilience was defined as the ability of some individuals who are exposed to trauma to fare better than others in physical and mental health outcomes (Atkinson, Martin, & Rankin, 2009).

The current study employs a positive affect intervention (PAI). As reported by Moskowitz and colleagues (2017), PAIs provide respite from the distress associated with toxic stress and help motivate and sustain ongoing efforts to cope with the negative effects. Furthermore, PAIs have been shown to broaden attentional focus (Rowe, Hirsh, & Anderson, 2006) and improve social and intellectual resources (Gable, Gonzaga, & Strachman, 2006).

The purpose of this study is to develop, implement, and evaluate the use of PATH, a 10-week group therapy trauma treatment that integrates positive psychology and resilience factors into an existing trauma treatment for low-socioeconomic-status, trauma-exposed Latino youth and their Spanish-speaking caregivers. This is the first study that evaluates a treatment that combines these three modalities to reduce symptoms of complex trauma among Latino youth and their caregivers in a group setting. We hypothesized that participating in the intervention would result in a clinically significant reduction in trauma symptoms (as measured by the Trauma Symptom Checklist and UCLA PTSD Index) for these youth.

Method

Participants

Youth Ages 8 to 16 years (five adolescents, 11 school-aged) and their caregivers were recruited into the study. Participants were referred by their primary care provider or school to an urban, public safety net hospital between September 2013 and March 2014. No caregiver had beyond a high school education, and all families utilized Medi-Cal insurance, which requires a family income equivalent or below 138% of the poverty level (e.g., family of three, <$27,821). Youth Ages 8 to 16 who had experienced three or more forms of trauma (e.g., community violence, physical abuse, and witnessed domestic violence) within an impaired caregiving system were eligible to be included in the study. We operationalized the term impaired caregiving system as one or more occasions of reported abuse by the caregiver or Child Protective Services report of maltreatment, neglect, and interpersonal violence. Exclusion criteria included participants whose sole trauma exposure was sexual abuse, those unable to attend the groups with their caregivers, and those with caregivers who were not proficient in Spanish or English. The study was approved by the Human Research Protection Program (Institutional Review Board). All caregivers provided informed consent; minors provided informed assent.

Recruitment and Assessment Procedures

Participants were identified from a waitlist and contacted by the clinic intake coordinator. Upon intake, two clinicians (at least one was bilingual and bicultural, and all were clinical psychologists in training) met with the families and conducted the Trauma History Questionnaire to determine whether or not they met the inclusion criteria (three or more lifetime and different forms of trauma) and, for those eligible, offer study participation. Consents and written information about the study were provided in both Spanish and English. The project goals and objectives were explained to the participants orally in the language of their choice, as were consent procedures. The caregiver and youth then separated to complete a battery of measures. Standardized instructions were provided by clinicians who were trained to administer the measures. The measures took 30 to 45 min for the youth to complete.

Following the initial meeting, a clinician met individually with the caregiver. During this second meeting, the clinician further assessed the family’s need for treatment and provided feedback from the youth’s measures. Given the difficulty that the research has noted when engaging Latinos in therapy (Falicov, 2009; Lewis-Fernandez et al., 2005), we implemented cultural adaptations. This was done in several ways. We showed knowledge and respect for differences among Latino cultures by thoroughly assessing each caregiver’s history and the youth’s developmental history. This included understanding the caregiver’s family history, country of origin, and reasons for migration. The therapist asked specifically about traumatic events during the migration narrative, keeping in mind that these narratives can be traumatic and at times overwhelm the caregiver. Shifts in relationships among the family were assessed. Importantly, we discussed with each caregiver the meaning and the importance that the role of religion and spirituality played in their family. This in depth intake procedure typically took two or three 1-hr sessions. Participants were not provided financial compensation.

Group Structure

Based on age of the identified youth and time of referral, three groups were carried out over a 9-month period, one group running at a time. The groups were structured in a way that the youth participated with other youth of a similar age. Caregivers participated at times with their youth and at times in a separate group with the other caregivers. Although the information and structure of the groups remained the same at all ages, the group leaders chose examples of the various tasks and descriptors in accordance with the developmental stage and cognitive abilities of the various groups. Each group consisted of a weekly 90-min session for 10 weeks. Treatment was delivered by supervised psychology doctoral interns and postdoctoral trainees.

Intervention

PATH is a 10-week, 90-min-per-session group therapy model that targets trauma-exposed youth who have endured multiple traumatic experiences. PATH integrates positive psychology, resilience factors, and an existing trauma framework (see Table 1). Intervention integrity was managed via (a) manualized instruction each week, (b) weekly supervision to discuss process and needed augmentations to the intervention, and (c) the lead clinician sitting in all sessions to ensure intervention integrity.

Table 1.

Treatment Manual for PATH

Week	Trauma component	Positive affect	Resilience	Group format
1	Psychoed of trauma	Emotional literacy	Social competence	Separate
2	Attachment and attunement	Noticing positive emotions	Social competence, critical consciousness, and problem solving	Conjoint
3	Consistent response and routines and rituals	Amplifying and gratitude	Social competence	Conjoint
4	Affect management and consistent response	Mindfulness	Self-acceptance (awareness of one’s abilities)	Separate
5	Affect identification and modulation	Mindfulness of breath	Problem solving	Separate
6	Caregiver consistent response	Mindfulness	Social competence and problem solving	Conjoint
7	Affect identification, modulation, and expression	Acts of kindness and strengths	Sense of self	Separate
8	Trauma narrative functions	Positive reappraisal	Critical consciousness	Conjoint
9	Self-development and identity	Noticing strengths and setting attainable goals	Social competence and sense of purpose of the future	Separate
10	Wrap up			Conjoint

Note. “Group format” refers to whether the session was “Separate,” meaning the parents met separately with the lead clinician and the youth met with another clinician, or “Conjoint,” indicating that the youth and caregivers were together with the two clinicians. PATH = Positive Adaptations for Trauma and Healing.

The integrated model utilizes interventions from the ARC trauma treatment (Kinniburgh & Blaustein, 2005). The first building block is to target attachment-focused interventions through four phase-oriented principles: (a) strengthening rituals and routines, (b) assisting the caregiver in attending to the child’s affect, (c) improving attunement between caregiver and child, and (d) increasing the amount of praise given by the caregiver. The second building block focuses on improving the child’s self-regulation. The third block builds the child’s executive functions and developmental competencies.

Consistent with Zolkoski and Bullock’s (2012) conceptualization of resilience theory, the interventions focuses on five capacities: social competence, including empathy, caring, flexibility, communication skills, and a sense of humor; problem-solving skills; critical consciousness, meaning the ability to connect emotions and behaviors to an earlier and possibly traumatic experience; autonomy; and a sense of purpose, which rests on belief in a positive future. These five theoretical components were operationalized and interwoven into the treatment manual weekly. Lastly, seven positive affect skills (Moskowitz et al., 2012) were adopted for PATH and delivered chronologically. These skills were noticing positive events, amplifying, gratitude, mindfulness, acts of kindness, positive reappraisal, and setting attainable goals. Six of the 10 weeks were split sessions, meaning that the caregivers met with a therapist while the youth met their own therapist; each group facilitator taught the skills in the first half of the 90-min session, which were then practiced in vivo for the second half of the session or utilized the following week. In these cases, groups would split into a youth group and a caregiver group to learn the skill so that they would be able to participate in the intervention the following week.

For example, we introduced the gratitude intervention in separate groups, asking participants to write out individually their own gratitude to the other family member in the group (e.g., caregiver would write the gratitude letter to their youth, and vice versa). The next week the participants read their letters to each other aloud in a multifamily group format. The following is an example of the gratitude intervention, which was read with all members present. The quotes provided have been translated from Spanish to English. A mother stated, “You know, my son, sometimes I think that if you were not with me, all my happiness would go away—I am grateful that you love me because I know you do.”

Following the large group interventions, we also processed the letters that were read in the smaller break-out groups. An excerpt from a grandmother is as follows:

After I lost my son to meth, I never actually knew if I could be a parent. I always felt like it was my fault [that her son was addicted to meth]. Listening to [child’s name] lets me know that I am doing something right as a parent.

Another key aspect of the PATH treatment includes a trauma narrative and positive reappraisal. The first exert is from a 10-year-old boy initially referred for diagnostic clarification, possibly attention-deficit hyperactivity disorder. His mother indicated at intake that she was assaulted by the boy’s father—which was the last time the family saw the father. The mother indicated during group that the family had a hard time discussing their past experience of domestic violence. The boy said, “I saw my parents fighting; I was scared my dad was going to hit or kill my mom. I was so scared that I wanted to ignore it. … Whenever my parents fought, I always thought it was my fault.”

His mother stated, “I didn’t realize everything he saw; we as parents don’t always realize how we can affect our children.” She turned to the boy and said, “[The boy’s name], it was not your fault that your father and I fought, it was issues between us.”

In the assessment phase, we found that caregivers placed a great deal of importance on both religious and spiritual worldviews, so we incorporated prayer into the opening activity when the caregivers met without the youth. The prayer, which was a way of the caregivers centering themselves and verbally stating intention for the group meeting, was led weekly by a different caregiver. In order to facilitate cohesion, we asked individuals that had an easier time speaking in a group setting to go first, making it easier for the other members to share later on in the group.

Measures Completed by the Clinician, Caregivers, and Youth

Measures were administered to youth and caregivers prior to and upon completion of treatment. All youth were proficient and completed measures in English; caregivers were given the measures in Spanish. The clinicians involved met individually with each family to administer pre- and posttreatment measures. After completion of treatment, clinicians provided feedback to partici- pants regarding changes in symptoms, need for further treatment, and additional resources, as appropriate. Measures were completed by the youth, caregivers, and clinicians.

Measures completed by the clinician.

The Child and Adolescent Needs and Strengths (CANS) Comprehensive Assessment.

The CANS is a multipurpose tool, completed solely for recruitment purposes (i.e., to sort out those participants that meet inclusion criteria) by the clinician after collecting data from the participants (Lyon, 2009). It was developed to support assessment and treatment planning, and for the monitoring of service outcomes. The intake clinician determined whether the caregivers fit the inclusion criteria of “impaired,” as noted by physical or mental health difficulties. A score of “2” or higher identified the caregiving system as “impaired.”

Trauma History Questionnaire (THQ).

The THQ is a 24-item measure that examines potentially traumatic events (Green, 1996). For endorsed items, respondents are asked to provide the frequency of the event as well as their age at the time of the event. The intake clinician assessed both the caregiver and youth to determine if the youth had experienced more than three different types of traumatic events.

Measures completed by the caregivers: The Child Behavior Check List (CBCL).

The CBCL is a standardized measure completed by the caretaker who spends the most time with the child (Achenbach & Rescorla, 2001). The CBCL yields scores on internalizing, externalizing, and total problems as well as scores on Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM–IV–TR; American Psychiatric Association, 2000) scales and is one of the most widely used outcome measures (Nakamura, Ebesutani, Bernstein, & Chorpita, 2009). The raw scores convert to T scores, with a score between 50 and 65 indicating the individual is at risk for the disorder (e.g., externalizing behavior), and a score above 65 indicates clinical significance. Parents completed forms in Spanish—which was normed in Puerto Rico and Latin America. Internal consistency alpha values range from .63 to .79. Criterion validity is supported by multiple regressions, odds ratios, and discriminant analyses all at p < .01. Pre- and posttest Cronbach’s alpha values for this study were .97 and .96, respectively.

Measures completed by the youth.

The Child Depression Inventory: Short version (CDI:S)

The CDI:S is a 10-item, self-rated scale suitable for youth Ages 7 to 17 (Kovacs, 2003). It was developed to provide a psychometrically sound way to quickly screen for depressive symptoms. Higher scores indicate more depressive symptomatology. The CDI has demonstrated good psychometric properties (Allgaier et al., 2012), for example, a Cronbach’s alpha of .86. Concurrent and discriminant validity were adequate (Kovacs & Mehler, 2009). The pre- and posttest Cronbach’s alphas for this study were .75 to .89, respectively.

The Individual Protective Factors Index (IPFI).

The IPFI is a 71-item self-report instrument, designed to assess resilience in three domains: (a) social bonding (school, family, and prosocial norms); (b) personal competence (self-concept, self-control, and positive outlook); and (c) social competence (confidence and cooperation; Springer & Phillips, 1997). Validated with 2,416 adolescents from 15 U.S. sites, it is internally reliable (α= .93) and has good construct validity (Springer & Phillips, 1997). The pre- and posttest Cronbach’s alphas for this study were .90 to .93.

The Trauma Symptom Checklist for Children (TSCC)

The TSCC is a 54-item questionnaire (Briere, 1989). The survey is used to assess youth’s self-reports of trauma-related symptoms, including anger, anxiety, depression, dissociation, and posttraumatic stress. Each of the subscales contains a raw score which is converted to a T score. A T score between 50 and 65 on a subscale indicates the individual is at risk for the disorder (e.g., anxiety), and a score above 65 indicates clinical significance. The TSCC has been normed with a population of ethnically and economically diverse children (Ohan, Myers, & Collett, 2002). Studies using the TSCC indicate that it is reliable, with alpha values in the mid-to high 80s for all scales except the Sexual Concerns scale, which has alpha values in the high 60s and low 70s. It also has convergent and predictive validity in samples of traumatized and nontraumatized youth (Lanktree et al., 2008). Pre- and posttest Cronbach’s alphas for this study were .93 to .88, respectively.

UCLA Posttraumatic Stress Disorder Index.

The UCLA PTSD Index (Steinberg, Brymer, Decker, & Pynoos, 2004) is a 22-item measure assessing for PTSD according to the DSM–IV–TR (American Psychiatric Association, 2000). Children endorse the frequency of their symptoms in the past month on a Likert scale (0 = none and 4 = most). Raw scores of 38 and above are considered clinically significant. A total score cutoff of 38 has a sensitivity of 0.93 and specificity of 0.87 in detecting PTSD (Rodriguez, Steinberg, Saltzman, & Pynoos, 2001). Reports on internal consistency have found Cronbach’s alpha to fall around 0.90 (Roussos et al., 2005). Pre- and posttest Cronbach’s alphas for this study were .85 to .88, respectively.

Modified Positive and Negative Affect Scale (PANAS-X).

The PANAS-X (Watson & Clark, 1994) was designed to assess two independent constructs: positive and negative emotion. It was modified for this study and comprised five negative (NA) and five positive (PA) mood terms that were developmentally appropriate. The youth were asked to rate, on a five-point scale (0 = not at all and 4 = extremely), the extent to which they had experienced each mood in the prior 7 days. The five PA items were “happy,” “proud,” “love,” “excited,” and “grateful”; NA items included “sad,” “nervous,” “lonely,” “guilty,” and “mad.” Scores on both scales range from 0 to 20. For the original study, reliability and validity reported by Watson, Clark, and Tellegen (1988) was moderately good. The PANAS has strong validity with such measures of depression and state anxiety. Pre- and posttest Cronbach’s alphas for this study were .75 to .68, respectively.

Perceived Stress Scale (PSS).

The PSS is a four-item measure that assesses the youth’s perception how they have managed stress over the past month (Cohen, Kamarck, & Mermelstein, 1983). The PSS has demonstrated adequate reliability and an alpha reliability coefficient for .60 (Cohen & Williamson, 1988). Pre- and posttest Cronbach’s alphas for this study were .61 to .27, respectively.

Skills Practice Scale (SPS).

The skills practice scale has 10 items and assesses the youth’s perceived ability to manage stress over the prior week using positive skills taught in the PATH program, such as gratitude and positive reappraisal (Caponigro, Moran, Kring, & Moskowitz, 2014). Responses ranged from 0 (never) to 4 (very often) during the past month. Pre- and posttest Cronbach’s alphas for this study were .78 to .8.

Analyses

We summarized demographic characteristics of the sample using means and standard deviations for continuous measures and frequency counts or percentages for categorical variables. We also examined change in symptoms from pretreatment (baseline) to post treatment using paired t tests.

Sixteen youth Ages 8 to 16 years (M = 11.1, SD = 1.98) and their caregivers participated in the study. All self-identified as “Latino,” with origins from various Latin American countries (see Table 2). Sixty-six percent of the youth who participated in the study were female. Thirty-eight percent of the youth had experienced homelessness prior to the study. The mean number of lifetime traumatic types reported by youth was 5.2 (SD = 2.29). The reported traumatic types included physical and sexual abuse, medical trauma, and witnessing domestic violence, among others. All measures were gathered with the exception of two participants who did not complete the mPANASc, PSS, and SPS due to time constraints.

Table 2.

Demographic, Clinical Characteristics of Youth and Caregiving System

Family member	Mean or # (%)a	SD
Youth gender (n = 16)
Male	6 (38%)
Female	10 (63%)
Age (mean)	11.1	±1.98
Ethnicity (%)
Mexican	10 (62.5%)
Mexican/Salvadorian	2 (12.5%)
Honduran	1 (6%)
Mexican/Nicaraguan	1 (6%)
Nicaraguan	1 (6%)
Nicaraguan/Honduran	1 (6%)
Religion/Spirituality (%)
Catholic	9 (56%)
Christian Evangelical	3 (19%)
Christian Pentecostal	2 (12%)
Spiritual Religious	2 (12%)
Number of trauma types	5.2
Single-parent home	13 (81.3%)
Average number of children per household (mean)	3.1	±1.69
Experienced domestic violence within the caregiving system	12 (75%)
Experienced homelessness	6 (37.5)
Undocumented	2 (12%)
Fathers (n = 15)
Living with youth	2 (13%)
Deported following youth’s birth	10 (66%)
Not residing with youth due to deportation	6 (40%)
Mothers (n = 14)
Living with youth	12 (86%)
Deported following youth’s birth	1 (7%)
Not residing with youth due to deportation	1 (7%)
Undocumented	7 (50%)
U-Visab	4 (28.6%)

^aRaw data and percent.

^bThe U nonimmigrant status (U visa) is set aside for victims of certain crimes who have suffered mental or physical abuse and are helpful to law enforcement or government officials in the investigation or prosecution of criminal activity.

Results

There was a 93% attendance rate for the families, and none dropped out of treatment, meaning that all participants attended seven or more sessions and the final session. Simultaneous to group treatment, three youth and their caregivers approached the lead clinician for one individual therapy session. These were not planned, but each family at the outset of treatment was told that this was possible if there were acute needs such as harm to self and others. The three reasons for individual sessions included a disclosure of sexual abuse, a parent–child meeting to reveal self-harm to a caregiver, and the disclosure of substance use from a teen to her caregiver. Lastly, acceptability was inferred based upon the high rate of attendance.

At the time of the intake, 56% of the participants enrolled in the study endorsed clinically significant symptoms on the UCLA PTSD Index; at posttest, no client reported clinically significant symptoms (see Table 3). On the TSCC, the youth endorsed clinically and statistically significant reductions on the five subscales. Furthermore, the youth reported a statistically significant decrease in depressive symptoms as captured by the CDI. There were also statistically significant reductions in the youth’s externalizing and internalizing behaviors as reported by the caregivers on the CBCL. Lastly, although there was no change in a measure of resilience (IPFI), SPS, and positive emotions, there was a statistically significant increase in negative emotions and PSS. Of note, although there were four clients engaging in self-harm behaviors (superficial cutting) at intake, none were endorsing self-harm at posttreatment.

Table 3.

Pre- and Posttest Scores and Paired T-Test Results

Measure	N	Pretest M (SD)	Posttest M (SD)	Paired t test (df)	p value	95% CI	Hedge’s g
Youth report
CDI	16	49.1 (7.7)	41.8 (8.5)	t(15) = 3.4	.004	[2.8, 11.9]	.75
IPFI	16	239 (18.2)	249.6 (19.3)	t(15) = −2	.066	[−21.9, .8]	−.75
mPANASc
Negative emotion	14	12 (4.7)	16.1 (4.9)	t(15) = −2.9	<.05	[−7.2, −1]	−.75
Positive emotion	14	14.6 (4.3)	16.2 (3.1)	t(15) = −1.09	.29	[−4.9, 1.6]	−.42
PSS	14	4.1 (2.3)	6 (1.6)	t(15) = −2.2	<.05	[−3.8, .01]	−.93
SPS	14	23.4 (9)	28.1 (6.9)	t(15) = −1.8	.09	[−10.5, .9]	−.58
TSCC	16
Anger		48.6 (8.5)	41.1 (5.7)	t(15) = 4.2	.001	[3.7, 11.2]	.99
Anxiety		52.1 (9.4)	43.7 (8.7)	t(15) = 3.9	.001	[3.8, 13.1]	.93
Depression		45.4 (6.8)	37.8 (2.8)	t(15) = 4.4	<.001	[4, 11.4]	1.28
Dissociation		52.6 (10.2)	43.1 (7)	t(15) = 3.9	<.05	[4.3, 14.6]	1.05
PTS		52.6 (11.1)	41.8 (5.5)	t(15) = 4.4	.001	[5.5, 16.1]	1.12
UCLA PTSDI	16	34.2 (11.2)	17.3 (7.6)	t(15) = 6.9	<.001	[11.6, 22.1]	1.71
Caregiver report
CBCL	16
Internalizing		64.4 (9.9)	55 (12)	t(15) = 5.2	<.001	[5.6, 13.3]	.85
Externalizing		58.3 (11.4)	52.2 (12.4)	t(15) = 5.4	<.001	[3.7, 8.7]	.51
Total		63 (11)	53.8 (13.6)	t(15) = 5.8	<.001	[5.8, 12.7]	.74

Note. df = degrees of freedom; CI = confidence interval; CDI = Child’s Depression Inventory: Short version; IPFI = Individual Protective Factors Index; mPANASc = Modified Positive and Negative Affect Scale; PSS = Perceived Stress Scale; SPS = Skills Practice Scale; TSCC = Trauma Symptom Checklist for Children; UCLA PTSD Index = UCLA Posttraumatic Stress Disorder Index; CBCL = Child Behavior Checklist.

Discussion

The results of our study demonstrated positive support of our novel group therapy intervention. Specifically, we found statistically significant improvements on PTSD symptoms, depressed mood, and anxiety symptoms among our sample of trauma-exposed Latino youth. The average number of types of traumatic experiences among our sample was five, which was consistent with previous research (Briggs et al., 2013). Among our sample, trauma experiences included witnessing chronic domestic violence and interpersonal violence, physical abuse and neglect, high rates of homelessness, and ruptures in attachment at critical ages.

Positive psychology interventions have come under scrutiny in terms of their cultural relevance and adaptability to youth (Dawood, 2013). Yet we found preliminary evidence that utilizing the interventions between the dyads produced positive shifts in the youth’s well-being as well as in the relationship among the families involved in the groups. Informal clinician reports on the group process suggest that utilizing the gratitude intervention contributed to engagement between the families and also increased the level of comfort with the group process, evidenced by increased communication about traumatic experiences among group members.

Both the UCLA PTSD Index and the TSCC PTS index measured trauma symptoms. Both measures showed a statistically significant decrease in trauma symptoms from pre- to posttest. The UCLA measure asks youth about their symptoms in relation to one to two traumatic experiences, although the clinician never actually knows if the reported experiences are the most significant trauma faced by the client/participant. The TSCC, in contrast, identifies global symptoms as they relate to posttraumatic stress without identifying an experience. The potential reason for the slight discrepancy between pre- and posttests scores in these two measures may be that the UCLA PTSD Index specifies a traumatic experience, whereas the TSCC PTS Index asks about symptoms, thus resulting in different outcome scores, particularly in participants who have experienced multiple different types of trauma.

The PANAS-X measures state levels of emotional responding. There was an increase in perceived stress as measured by the PSS (although potentially not reliable in our sample based on the posttest Cronbach’s alpha, presumably caused by a small number of items) and negative emotion endorsed by the youth in the current study, coupled with no shift in positive emotion. Although contrary to our hypothesis, the findings align with trauma theory. Presumably, the increase in “perceived stress” and identification of negative emotion indicates that the youth increased their own capacity to perceive and tolerate stress in their environment and incorporate a broader range of affect. An increase in the ability to name negative emotions may indicate that the youth experienced a positive shift in their well-being, possibly no longer avoiding the dysregulation that may come from recognizing the presence of negative emotion. The levels of negative emotions and perceived stress were not coupled with clinical levels of trauma or depression (or other negative outcomes). Thus, we do not consider them to be an indicator of negative effects of the intervention, but instead a possible indicator that the intervention worked. It may be that with continued use of the skills learned in the PATH program, negative emotion and perceived stress declined, whereas positive emotion increased over time. Future research should extend the follow-up period to capture these potential shifts.

It is notable that at intake only 56% of youth identified clinically significant symptoms on the UCLA PTSD Index. To provide a comparison, only 22% of Gresson et al.’s (2011) sample fell in the clinical significant range on the UCLA PTSD Index, and these were youth participating in treatment with the National Child Traumatic Stress Network, with more than half of the sample residing in foster care (54%) at the time of data collection. This underscores the fact that the PATH sample was a highly traumatized sample. Furthermore, some youth disclosed traumas that neither the caregivers nor the clinicians knew about during group therapy.

Finally, to facilitate the sharing of the narratives among our participants, clinicians (typically during supervision) requested that a more outgoing youth member of the group initiate the process of sharing. The youth would be the first in the group to read their letter and translate it if needed; other group members followed this format. Although Deblinger et al. (2016) noted that reading a trauma narrative aloud in a group format can be challenging, we addressed this issue by gradually exposing the dyads to sharing verbally in a group format over the course of 10 weeks (see Table 1), with the culmination being the sharing of their trauma narrative. We gauged effectiveness of this method via high group cohesion and attendance in the group as well as 100% sharing of trauma narratives.

An important strength of this study is that each family remained in treatment, which runs counter to an ongoing challenge in the mental health field of engaging Latinos in health care settings (Aggarwal et al., 2016). In the current study, the first author was bilingual and bicultural, and also tailored the intake assessment to gather information about the family’s interpersonal relationships tied to their community of origin as well as the impact of the cultural values and traditions of their community of origin on family dynamics. We hypothesize that these cultural nuances contributed to retention in the intervention.

During the intake assessment, many families identified exposure to social, cultural, and political violence in both their home country and the United States as contributing to their request for therapy. When interviewing the caregivers, we elicited immigration experiences to assess for potential trauma narratives. Given our knowledge of cultural differences within the Latino community, we assessed for intergenerational conflicts and ever-changing enculturation patterns. Lastly, we asked about the roles that religion and spirituality played within the family system and the possible impact that trauma has had on these worldviews. We hypothesize that attending to these cultural nuances was a key component of participant retention in the intervention. Multiple measures noted improvements for the youth (see Table 3). Although these positive shifts were promising, they require further evaluation to assess long-term shifts in positive functioning within the caregiving system (e.g., family, school, community). Also, there was no control group in the current study, thus making it impossible to gauge whether positive outcomes were specific to our sample. Lastly, the primary investigator was also a clinician involved in all three caregiver groups. It should be noted that the clinician was both bicultural and bilingual; his cultural background and Spanish language capacity were likely impactful in the effectiveness of the intervention, perhaps even partially explaining the high rates of success and retention (i.e., demand characteristics and expectancy effects). Relatedly, this therapeutic relationship with the participants may have affected how participants answered the questionnaires due to demand characteristics and expectancy effects.