Expressive writing intervention promotes resilience among juvenile justice-involved youth

Highlights

• Incarcerated youth's psychiatric disorders are under-identified and under-treated.

• A writing-based mental health intervention was developed for incarcerated youth.

• The intervention was experimentally evaluated with incarcerated youth.

• Participants exhibited significant gains in resilience compared to a control group.

• Intervention delivery had good fidelity and participants reported satisfaction.

Abstract

Youth involved in child welfare and juvenile justice systems suffer from alarmingly high rates of mental health challenges. In particular, exposure to trauma (e.g., maltreatment) is one critical experience that amplifies their risk for delinquency and recidivism. Despite a profound need to address these youth's mental health needs, there is a paucity of trauma-informed and youth-centered treatments that are clinically feasible in under-resourced residential settings (e.g., juvenile detention facilities). In response to this gap, our research team collaborated with the juvenile justice subsection of a large American city's child welfare system with the goal of creating an intervention tailored to the needs of underserved system-involved youth. The resultant program, WRITE ON (Writing and Reflecting on Identity To Empower Ourselves as Narrators), leverages research on the therapeutic benefits of expressive writing to implement a brief, cost-effective intervention in youth residential settings. This paper describes intervention development and presents findings from the pilot study, which comprised a multisite experimental evaluation of youth (N = 53) residing in short-term detention facilities. This pilot study aimed to: 1) assess intervention implementation fidelity, including participant satisfaction, and 2) evaluate the mental health outcomes of youth receiving WRITE ON as compared to those in a comparison support group (CSG). Results indicated that the intervention was delivered with good fidelity, participants reported high levels of satisfaction, and WRITE ON participants exhibited significant (p < 0.01) gains in resilience compared to their counterparts in the CSG. Collectively, results suggest that a larger clinical trial investigating the effectiveness of WRITE ON with system-involved youth is warranted.

Introduction

In addition to physical, financial, and educational difficulties, youth involved in the child welfare and juvenile justice systems suffer from alarmingly high rates of mental health challenges. Studies have estimated that 40–80% of youth served by these systems experience psychiatric illnesses, representing rates greatly exceeding those of youth in community samples (Burge, 2007, Krezmien et al., 2008, Teplin et al., 2002). Disconcertingly, many of these youth do not receive adequate mental health services (Burns et al., 2004, United States Department of Justice, 2005). Numerous studies have underscored the need for more coordination between the child welfare and juvenile justice systems in order to increase youth's access to, and receipt of, mental health services (e.g., Chuang & Wells, 2010). Several major cities have responded to this call to action by merging their child welfare and juvenile justice systems. The present study reports on an expressive writing-based intervention developed and implemented in collaboration with one such merged government agency in a large American city.

The considerable mental health needs of incarcerated youth are well documented. For example, findings from a random sample of 1829 detained youth revealed that nearly two-thirds of boys and three-quarters of girls met diagnostic criteria for one or more psychiatric disorders (Teplin et al., 2002). Furthermore, the study found that 92.5% of participants had experienced at least one trauma (e.g., maltreatment, neglect), and 56.8% had been exposed to six or more traumatic events (Abram et al., 2004). The interplay between these individuals' trauma exposure, mental health problems, and delinquency is notable. Incarcerated youth with trauma histories are at significantly greater risk for suicide attempts and recidivism than those with no history of abuse (Halemba et al., 2004, Sedlak and McPherson, 2010). Moreover, a number of investigations have suggested that entry into the juvenile justice system may exacerbate pre-existing conditions (e.g., Kupers & Toch, 1999), and that youth with mental health challenges are more likely to stay incarcerated for longer periods of time (United States Congress House Committee on Government Reform and Oversight, 2004). Youth's transitions into incarceration, characterized by removal from their families and communities, are often inherently traumatizing or re-traumatizing (e.g., Adams, 2010), and their physical and mental health outcomes may worsen by mere virtue of being detained (Chesney-Lind & Pasko, 2012).

Notably, there is a pronounced disparity between the need for and the availability of mental health services for these individuals. Investigations by the United States Department of Justice (2005) revealed that juvenile detention and correctional facilities across the country fail to address the mental health needs of detained youth. Research has further elucidated the under-utilization of mental health treatments within this population, with studies finding that only one-third of incarcerated youth with documented psychiatric disorders receive treatment while incarcerated (Pumariega et al., 1999). These patterns are disquieting in light of data showing that under-treatment may contribute to the onset and maintenance of criminal behaviors (Pumariega et al., 1999, United States Congress House Committee on Government Reform) in addition to being associated with chronic physical and mental health challenges affecting youth, their families, and communities (Massoglia, 2008). As such, the mental health of incarcerated adolescents can be reasonably construed as a significant public health concern.

Research has highlighted several reasons for the juvenile justice system's failure to address this population's needs. Although there exist a handful of mental health interventions that have been identified as “promising” (e.g., multisystemic therapy; for a review, see National Mental Health Association, 2004), these practices necessitate family involvement, the employment of mental health professionals, screening for and identification of psychiatric disorders, and integrated support among multiple systems (e.g., educational and child welfare systems). The implementation of these treatments is impeded by inadequate screening for mental health and trauma-related symptoms, difficulty engaging families (Burke et al., 2014, Justice for Families, 2012), insufficient clinical resources for intervention delivery (Sedlak & Bruce, 2010), and inadequate collaboration among systems (Huang, Ryan, & Herz, 2012). Further, many extant programs have low levels of youth engagement (Sedlak & McPherson, 2010) and retention (Gaarder, Rodriguez, & Zatz, 2004). Lastly, by nature of the detention facilities in several cities, youth placed in non-secure and residential facilities must travel to secure detention facilities in order to receive medical and mental health care. Such removal from their facilities and transport to secure detention centers are effortful, disruptive, and potentially traumatic, as all youth entering secure detention centers must be strip searched upon arrival. Taken together, the literature exposes a profound need for trauma-informed and youth-centered mental health treatment that is clinically feasible in under-resourced juvenile justice-related settings. Given its brief and low-cost nature coupled with its demonstrated effectiveness, expressive writing (EW) may have utility within this population.

Though it has not been systematically implemented or evaluated in juvenile justice populations, regular written self-disclosure has been linked with improvements in psychological outcomes in clinical and nonclinical samples (e.g., Pennebaker & Chung, 2011). The first study on EW demonstrated that university students experienced lasting improvements in their mental health when instructed to write, three separate times, about the most traumatic event of their lives (Pennebaker & Beall, 1986). Since its conception, hundreds of studies have implemented the intervention with adults, and the largest and most recent meta-analysis on EW interventions (Frattaroli, 2006) revealed an overall small, yet significant, effect size (d = 0.15). Across EW studies, mental health benefits after writing seem to be greater for those who wrote about a previously undisclosed trauma (Páez, Velasco, & González, 1999), are more inhibited (Smyth, 1998), have less opportunity for confiding, or belong to a stigmatized social group (Richards, Beal, Seagal, & Pennebaker, 2000). In addition, studies (e.g., Smyth, True, & Souto, 2001) have revealed that individuals improved most when they constructed narratives and made meaning out of negative life events, rather than merely recounting them in list format. Potential mechanisms underlying these changes are the schematic restructuring of traumatic memories (Smyth et al., 2001) and desensitization to trauma-related stimuli (Lepore & Greenberg, 2002).

Although there is a paucity of expressive writing research on at-risk youth, expressive writing may be an effective form of therapy for adolescents. To our knowledge, only five expressive writing studies have been conducted with at-risk youth (Evans, 2000, Muris et al., 2002, OʼHeeron, 1993, Stice et al., 2006a, Stice et al., 2006b), and none has been conducted with juvenile justice-involved youth. The most robust evidence for the potential utility of expressive writing interventions for adolescents comes from a recent meta-analysis on this topic that is inclusive of 21 independent studies with youth (though their study is not focused on at-risk youth in particular). Specifically, Travagin, Margola, and Revenson (2015) conducted a meta-analysis evaluating the effects of EW interventions among adolescents aged 10–18 years. The overall adjusted effect size was significant yet relatively small (g = 0.13), with significant effects for the individual domains of internalizing problems, social adjustment, problem behaviors, and school participation. Gender did not moderate the effects of EW, but adolescents with high levels of emotional problems at baseline exhibited greater improvements in school participation. Moreover, in one study youth who reported the highest levels of community violence exposure displayed the greatest changes in lability (Kliewer et al., 2011). Relatedly, the use of trauma narratives, a central component of EW, is also integral to trauma-focused cognitive behavior therapy (TF-CBT), which has demonstrated effectiveness at reducing symptoms of posttraumatic stress disorder (PTSD), depression, anxiety, and behavior problems in children and adolescents participating in randomized controlled trials (e.g., Deblinger, Mannarino, Cohen, Runyon, & Steer, 2011). Moreover, adolescents who are wary of adult and authority figures may respond better to creative approaches than traditional psychotherapy (Utley & Garza, 2011).

Despite the brevity, ease of delivery, and demonstrated effectiveness of EW interventions, none has been implemented or evaluated in juvenile justice settings. An exhaustive search of the literature revealed that there is currently one manualized writing-based intervention for incarcerated youth, Writing Our Stories (Smitherman & Thompson, 2002). Although the program's evaluation comprised anecdotal data and subjectively scored tests and lacked a control comparison group, data indicated that youth who completed the program demonstrated greater emotional awareness and understanding of their placement in the juvenile justice system. Collectively, the literature suggests that written self-expression may be a compelling treatment for adolescents with histories of trauma exposure who reside in underserved settings.

In light of these considerations, our research team collaborated with the juvenile justice subsection of a large American city's child welfare system with the goal of creating an intervention tailored to the needs of underserved system-involved youth. The product of this work was the development of a writing-based intervention named WRITE ON, an acronym for Writing and Reflecting on Identity To Empower Ourselves as Narrators. Particularly novel aspects of WRITE ON include that it 1) uses a youth-centered and trauma-informed manualized curriculum, 2) is based on empirically supported therapeutic paradigms, and 3) is the first implementation of the EW paradigm in a juvenile justice-involved sample.

WRITE ON operates within a framework of trauma-informed treatment and its development was influenced by the Substance Abuse and Mental Health Services Administration's (SAMHSA) six key principles of a trauma-informed approach: safety, trustworthiness and transparency, peer support, collaboration and mutuality, empowerment, voice and choice, and cultural, historical, and gender issues (SAMHSA, 2015). WRITE ON is strength-based in that it seeks to foster youth dignity, autonomy, and resilience over and above “fixing” pathologies, an aim in line with Larson's (2000) Positive Youth Development approach to treating juvenile justice-involved youth.

In their meta-analysis on expressive writing in youth, Travagin et al. (2015) reported that larger effect sizes were found in interventions with more than three sessions than in interventions with three or fewer sessions, as well as in studies with writing sessions spaced more than one day apart as opposed to interventions with sessions on consecutive days. WRITE ON was accordingly designed to be a six-week-long group-based intervention, with sessions taking place twice a week, for 1.5 h at a time, for a total of 18 h of intervention.

Each WRITE ON intervention session followed a fixed structure, in line with findings that the most effective programs for juvenile justice-involved youth are highly structured (Altschuler, 1998). Weekly themes (i.e., emotions, self-expression, relationships, past self, present self, future self) were selected based on anecdotal data collected from interventionists involved in three programs across the country (i.e., The Beat Within, Youth Arts Alliance, Telling My Story) regarding youth's preferred writing topics. Sessions began with the collaborative creation and review of a community agreement (i.e., set of rules), emphasizing safety and respect. Next, community-building activities aimed to increase group members' comfort with and trust in each other. A choice of two writing prompts was administered to youth at each session. Writing prompts encouraged emotional and cognitive reflection in the context of narrative construction (e.g., Imagine you wrote an autobiography of your life up to this point in time, and then you opened the book to a random page. Write that page.), in accordance with studies demonstrating the beneficial effects of emotional and cognitive processing and narrative creation (e.g., Ullrich & Lutgendorf, 2002). Writing prompts were designed to be culturally relevant and sensitive (e.g., “Write about a community you are part of. How might that community help you? How might it hurt you?”). If they elected not to respond to either writing prompt, youth were offered Pongo (http://www.pongoteenwriting.org/introduction-to-writing-activities.html?) worksheets (i.e., fill-in-the-blank poems), and if they declined writing, they were provided with word searches. Following the writing activities, youth were given the opportunity to verbally share their writing with the group and receive semi-structured strength-based feedback from their peers. Youth also had the options of receiving individual written feedback from the facilitator or keeping their writing entirely private. This array of options for writing, sharing, and receiving feedback was created to foster youth empowerment and choice. Groups culminated with a collaborative closing activity (e.g., creating a group poem wherein each member contributed one line). Each participant was supplied with a journal to use between groups, though no homework was assigned.

In three separate sessions, Pennebaker's expressive writing prompt (i.e., trauma narrative; see Pennebaker & Beall, 1986) was administered as a second writing activity. Youth were given envelopes in which to deposit their trauma writing, and this writing was not shared with the group or facilitator. The spacing of these trauma prompts throughout the six weeks was based on the finding that the effects of writing are greatest when individuals have the opportunity to habituate to traumatic memories over multiple writing sessions (Smyth et al., 2001). Additionally, youth safety and trust were maintained by introducing trauma writing after four sessions focused on relationship-building and by succeeding trauma writing with group mindfulness activities. The inclusion of these activities derived from the clinical literature highlighting the importance of incorporating coping skills into trauma treatments (e.g., Salloum & Overstreet, 2012).

The present study was the first to implement and systematically evaluate a trauma-informed writing-based intervention for youth in detention employing an experimental design. An overall aim of this pilot study was to assess fidelity of intervention delivery, including participant satisfaction, in a sampling of urban detention centers. Second, we compared the mental health outcomes of youth receiving WRITE ON to those in a comparison support group (CSG) by analyzing quantitative outcome data informing positive and negative mental health outcomes. Based on previous research (Pennebaker and Chung, 2011, Travagin et al., 2015), we operationalized psychological health as encompassing a number of measurable constructs, focusing on both protective (e.g., resilience) and risky (e.g., negative affect) mental health experiences given research suggesting that mental health and illness represent distinct constructs (Keyes, 2005). We hypothesized that WRITE ON would reduce negative mental health outcomes and improve positive mental health outcomes.