A Comparative Effectiveness Study of a Shortened Trauma Recovery Empowerment Model and an Attachment-Informed Adaptation

Abstract

Relational challenges often associated with interpersonal trauma may hinder survivors’ abilities to fully benefit from group therapy. This quasi-experimental study compared clinical outcomes of a 16-week version of the Trauma Recovery and Empowerment Model (TREM) for women and an attachment-informed adaptation (ATREM). Prior findings of TREM are extended by demonstrating that both group conditions can facilitate comparable clinical outcomes, creating options for group facilitators. Only ATREM resulted in statistically significant improvement in individual attachment avoidance, and it trended toward a slightly higher completion rate. This study provides insight into the emerging concept of group attachment.

Keywords

trauma group therapy attachment

High prevalence of violence against women and girls has been well documented (Fallot, McHugo, Harris, & Xie, 2011). Nearly 20% of women report a history of rape, 22% report intimate partner violence, and 37.6% report childhood maltreatment (Breiding, 2015; Kim, Wildeman, Jonson-Reid, & Drake, 2017). When sexual victimization begins in childhood, there is nearly a 50% chance of sexual revictimization (Walker, Freud, Ellis, Fraine, & Wilson, 2019). Data indicate that 90% of clients who receive mental health and/or substance use treatment in public behavioral health care settings have experienced a traumatic event (Substance Abuse and Mental Health Services Administration, n.d.). Adults with histories of interpersonal trauma frequently experience chronic health conditions and enduring mental health issues related to mood, anxiety, substance use, eating disorders, posttraumatic stress disorder (PTSD), affect dysregulation, limited self-understanding, and persistent relational challenges (Devries et al., 2013; Hillberg, Hamilton-Giachritsis, & Dixon, 2011; Spohn, Wright, & Peterson, 2017). Furthermore, PTSD, depression, and substance use often co-occur among women with interpersonal trauma histories, potentially exacerbating negative outcomes (Zinzow et al., 2011).

The Trauma Recovery and Empowerment Model (TREM) is an empirically supported, manualized group therapy approach designed to concurrently address interpersonal trauma, mental health, and substance use disorders among women. TREM involves a 24- to 33-week curriculum for women who have survived trauma and struggle with mental health and/or substance use disorders (Harris & Anglin, 1998). The model has three phases: empowerment, trauma recovery, and advanced trauma recovery issues. Each week covers a designated topic, such as emotional boundaries, self-soothing, sexual abuse, and decision making, along with specific goals, guiding questions, and relevant exercises. TREM is rooted in feminist and relational theories with a present-oriented focus on enhancement of skills associated with trauma recovery. TREM facilitates recovery through cognitive restructuring, psychoeducation, and peer support (Harris & Anglin, 1998). Favorable, though somewhat inconsistent, outcomes have been previously demonstrated (Fallot et al., 2011). This article presents a rationale for enhancing TREM with more explicit attention to relationships, informed by attachment theory. It also includes a description of this adaptation, the Attachment-Informed Trauma Recovery and Empowerment Model (ATREM), and results of a pilot study comparing outcomes of ATREM with a 16-week version of TREM.

Research on TREM

The majority of investigations on the effectiveness of TREM were funded through the Substance Abuse and Mental Health Service Administration (SAMHSA)-sponsored Women, Co-Occurring Disorders, and Violence Study (WCDVS), predominantly conducted in urban outpatient or residential settings (McHugo, Kammerer, et al., 2005). In several studies, TREM was associated with statistically significant, or approaching significant, improvements in anxiety (Fallot et al., 2011), drug and alcohol problem severity (Cocozza et al., 2005; Fallot et al., 2011), drug abstinence (Amaro, Dai, Arévalo, Acevedo, Matsumoto, Nieves, & Prado, 2007), PTSD symptoms (Amaro, Dai et al., 2007; Toussaint, VanDeMark, Bornemann, & Graeber, 2007), and mental health symptom severity (Amaro, Dai et al., 2007; Toussaint et al., 2007). However, other studies found no statistically significant improvements in depression (Fallot et al., 2011), PTSD (Fallot et al., 2011), or substance use severity (Amaro, Dai et al., 2007; Morrissey et al., 2005). More recent studies of TREM found statistically significant improvements in anxiety (Karatzias, Ferguson, Gullone, & Cosgrove, 2016) and PTSD (Paquin, Kivlighan, & Drogosz, 2013), but not in depression (Karatzias et al., 2016). Early and more recent TREM studies report high attrition rates and discuss the substantial barriers to treatment completion experienced by survivors of interpersonal trauma (Chouliara et al., 2017; McHugo, Kammerer, et al., 2005). In addition, recent TREM studies include greater focus on relationships (Cihlar, 2014; Karatzias et al., 2016; Paquin et al., 2013) and recognition that interventions that explicitly target interpersonal difficulties may be needed to enhance outcomes for survivors of interpersonal trauma (Karatzias et al., 2016). Similarly, social support has been identified as a key contributor to healing and resilience following trauma (Afifi & MacMillan, 2011; Burton, Cooper, Feeny, & Zoellner, 2015; Herman, 1997; Pearlman & Courtois, 2005), suggesting that attention to relational needs in group therapy is essential to facilitate positive outcomes.

Individual and Group Attachment Styles

To explicitly target interpersonal dynamics, the current comparative effectiveness study sought to enhance outcomes of TREM by drawing upon strategies informed by attachment theory (Bowlby, 1988). According to attachment theory, implicit mental schemas about the nature and worth of self and availability and supportiveness of others, known as internal working models (IWMs), along with methods of emotion regulation, evolve out of the quality of a child’s early interactions with caregivers. The sense of interpersonal security or insecurity that develops is generalized beyond the original dyad and continues to guide and influence attachment-related affect, ideas, perceptions, expectations, and behaviors throughout a person’s life (Bowlby, 1982). Secure attachments are characterized by trust in self and others, adaptive emotion regulation strategies, and balance between intimacy and autonomy. Insecure attachments reflect high attachment anxiety and preoccupation with potential rejection and abandonment and/or high attachment avoidance and dismissiveness of closeness and dependence. Changes in attachment style can be accounted for by positive or negative close relational experiences that are meaningfully incongruent with past relationships (Mikulincer & Shaver, 2010).

Attachment theory has been extended to include patterns of behavior in group interactions based on early experiences with families and other social/cultural groups (Smith, Murphy, & Coats, 1999). Seeking out connections to groups for closeness, security, and belonging is considered as innate a function as a child seeking out a primary caregiver for protection (Markin & Marmarosh, 2010). Group attachment anxiety depicts the degree of preoccupation with acceptance by valued groups and conformist behavior to protect against rejection, while group attachment avoidance describes the extent that group affiliation and dependence are perceived as undesirable and the extent that relational disengagement is maintained (Boccato & Capozza, 2011; Smith et al., 1999). However, group attachment style is not equivalent to individual attachment style (Smith et al., 1999). For example, a group member with insecure individual/secure group attachment styles may struggle to connect with the group leader and not want to befriend individual members outside of the group but may keep the group on task and offer insightful and compassionate feedback during group interactions. In contrast, another group member may feel secure talking with one group member but fear and withdraw from group-level interactions. Group attachment style is also not equivalent to the traditional conceptualization of group cohesiveness from the seminal work of Yalom (Smith et al., 1999; Yalom & Leszcz, 2005). Attention to both group and individual attachment patterns extends the work of Yalom (Yalom & Leszcz, 2005) by providing a more nuanced understanding of individual differences in group interactions (Marmarosh, Markin, & Speigel, 2013). Numerous studies have found that individual and group attachment styles are differentially associated with group processes and outcomes, including experiences of social support in group interactions (e.g., Boccato & Capozza, 2011; Levy, Ellison, Scott, & Bernecker, 2011; Marmarosh et al., 2013; Saunders & Edelson, 1999; Smith et al., 1999; Taylor, Rietzschel, Danquah, & Berry, 2015). Hence, knowing each member’s individual and group attachment styles allows for an additional dimension in understanding seemingly contradictory or complex relational patterns and offers expanded pathways to facilitate positive mental health outcomes and supportive relationships in the group.

Social Support in Group Therapy

Although group therapy is a modality that provides opportunities for mutual aid to develop within socially supportive relationships, mere participation does not necessarily build connections or result in helpful outcomes for all individuals (Boccato & Capozza, 2011; Lundqvist, Hansson, & Svedin, 2009; Shulman, 2015). Numerous factors can impede the formation of helping connections in group therapy (Shulman, 2015), especially prior interpersonal trauma (Herman, 1997). Survivors of trauma often struggle to connect with others and report reduced perceptions of social support from a variety of sources, regardless of their availability (Burton et al., 2015; Muller, Gragtmans, & Baker, 2008). In a prior TREM study, members who did not complete treatment expressed this struggle to connect with other group members as resulting from a sense of separateness, lack of trust, and fear of judgment (Chouliara et al., 2017). Group leaders must establish a sense of safety for mutual aid to occur (Shulman, 2015); thus, attachment theory can guide this process (Marmarosh et al., 2013). Attachment research has shown differential reactions to social support in cohesive groups such that attachment avoidance has been associated with high rates of attrition attributed to feeling threatened by closeness and interdependence (Mikulincer & Shaver, 2010). Awareness of both individual and group attachment styles can help to expand clinical understanding of each member’s group interactions and experiences.

Theoretical Frameworks Associated with Attachment-Informed Group Therapy

In attachment-informed group therapy, facilitators encourage exploration of in-the-moment behaviors as manifestations of implicit attachment patterns that are activated during group interactions (Marmarosh et al., 2013). Facilitator and member feedback on here-and-now group processes may facilitate reductions in attachment anxiety and avoidance through relational experiences that provide new perspectives of self and other (Bowlby, 1988; Flores, 2010; Marmarosh et al., 2013). Experiential processing, grounded in psychodynamic theory, has been encouraged in attachment-informed therapy to promote more secure attachments (Magnavita & Anchin, 2013; Wallin, 2015). Integrating psychodynamic approaches and cognitive behavioral strategies in attachment-informed group therapy may create particularly effective practice models (Field, 2014; Magnavita & Anchin, 2013; Marmarosh, 2015). Based on an integrative framework, it was hypothesized that ATREM would be more effective than TREM in decreasing attachment anxiety and avoidance, substance use, depression, anxiety, and PTSD symptoms and in increasing perceived social support and emotion regulation capacities.

Method

Participant Inclusion and Recruitment

Participants were women receiving or seeking services at a community behavioral health agency, victim services agency, or residential substance use treatment program who met the following criteria: (a) age 18 or older; (b) a history of interpersonal trauma; (c) diagnosed with a Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) mental health and/or substance use condition or a client at a nondiagnostic service agency; (d) in early or longer remission, as defined by DSM-V for substance use disorders; (e) willing to complete pre- and postintervention questionnaires and sign an informed consent form; and (f) enroll by the third session with the intention to complete the 16-week group with an attendance rate of at least 60% of the sessions. Exclusion criteria were the following: (a) no payment plan to the agency for prior services rendered; (b) active psychosis severe enough to interfere with participating in the group; and (c) arrival for group under the influence of drugs or alcohol twice. No women were excluded from the study for these reasons.

Participants were recruited from three nonprofit agencies located in a county with suburban and rural communities. Each agency used their usual methods of disseminating information about group therapy, including all-staff emails, flyers posted in waiting areas, announcements in department meetings, and word-of-mouth. Nearly all of the participants were therapist-referred to the TREM group facilitator who served as the agency referral coordinator. The three people who referred themselves to the group discussed their participation with their primary therapists prior to referral. Upon referral, the coordinator contacted the potential participants to discuss enrollment and confirm with a yes/no question whether interpersonal trauma (i.e., physical, sexual, or emotional abuse, assault, or childhood neglect) had been experienced at any point in their lives. Upon identification of trauma history and with agreement to join an upcoming TREM group, voluntary participation in the research study was described in detail. Clients were informed that the choice of nonparticipation would not result in any disruption in service. All clients who were contacted and invited to join an ATREM/TREM group agreed to be part of the study, creating an initial sample of 129 participants who signed informed consent forms and completed preintervention questionnaires. Written informed consent was obtained and study activities proceeded in accordance with the guidelines approved by the Institutional Review Board of the University of Pennsylvania.

Due to anticipated attrition, nine to 16 participants were accepted into a group before recruitment was closed. As a quasi-experimental study, placement in the experimental treatment condition (ATREM) or the comparison treatment condition (TREM) was determined by which treatment condition was starting next and fit the participant’s schedule. ATREM and TREM alternated order of implementation with staggered start dates.

Interventions

TREM

TREM was designed for approximately eight to 10 women and two to three group facilitators, with weekly meetings of 75 min (Harris & Anglin, 1998). The present study followed agency policy of weekly 90-min groups with a 10- to 15-min break, two co-facilitators, and eight to 12 participants per group. Because TREM is considered modifiable and has been effectively condensed (Finkelstein et al., 2004; Toussaint et al., 2007), a newly manualized 16-week version was created. The first author used clinical experience with TREM, discussion with a TREM co-facilitator, anecdotal reports from prior TREM participants, and attendance at a TREM training as guides for combining related topics and reducing redundancy in sessions.

ATREM

Informed by empirical and theoretical literature (Allen, 2013; Flores, 2010; Fonagy & Bateman, 2006; Magnavita & Anchin, 2013; Marmarosh, 2015; Marmarosh et al., 2013; Tasca, Ritchie, & Balfour, 2011), an explicit attachment-based conceptual framework was integrated into the TREM curriculum. ATREM included the same 16 topics, with the same order and structure as the modified TREM curriculum, but was configured to create three open weeks to add new attachment-based information and activities in the 16-week time frame. Attachment-based activities were integrated into the other 13 weeks with experiential and cognitive activities involving imagery, art, fables, group meditation, transitional objects, body tapping, and written and verbal feedback to make relational feelings and implicit IWMs explicit. In TREM, as is common in group therapy, some acknowledgment of group dynamics likely occurs as a natural part of group work, but group relational processing is not guided by explicit attention to attachment, as in ATREM. Whereas TREM prioritizes content, ATREM integrates more processing by pausing the content and initiating in-the-moment exploration of relational dynamics and facilitating dyadic and group connections. Intense relational moments are not seen as detours to be resolved to resume a focus on content, but as fundamental areas of work. While the value of relationships and mutual aid underlies the TREM curriculum, the dominant focus of TREM involves a cognitive understanding of members’ past and present relationships outside of the group rather than a more experiential approach which prioritizes emerging group dynamics in ATREM. This experiential approach facilitates reflective functioning by inquiring about in-the-moment emotional and physical responses to experiences of support and conflict in the group; encouraging members to hypothesize about underlying motivations, intentions, and meanings of other members’ actions; and facilitating direct exploration of such hypotheses between members. Intentionally seeking out and deepening these relational moments goes beyond the scope of the TREM curriculum and training and is a key expansion represented in ATREM. ATREM co-facilitators were supported in this relational processing by having prior knowledge of each member’s individual and group attachment styles based on preintervention responses to the attachment scales. In addition, the ATREM manual contained information and examples for effectively engaging in attachment-informed experiential processing, as well as detailed weekly instructions to standardize implementation of this novel adaptation.

Facilitators and Training

The first author completed TREM training in March 2015 and facilitated four TREM groups prior to the start of the study; she also facilitated other types of groups for almost 20 years. This author was a co-facilitator for ATREM, and she trained three clinicians to facilitate ATREM and six clinicians to facilitate the 16-week version of TREM. All facilitators were Caucasian females. Each group had at least one licensed master’s level social worker or counselor with over 5 years of clinical practice experience, including experience related to trauma and to group therapy. To control for potential contamination between conditions, different co-facilitators implemented ATREM and the shortened version of TREM. Facilitators reviewed the original TREM manual and the 16-week version for their respective study conditions. They participated in training sessions that involved reviewing each week’s curriculum in the modified manuals and the fidelity checklist, engaging in role plays, and discussing concerns. Three therapists had previously facilitated TREM groups, so the focus entailed learning differences between TREM as usual and the specific requirements for this study. A list of commonly used stabilization skills was also given to both group conditions and discussed for use in situations of dysregulation. The first author was available, as needed, for support throughout the study.

Data Collection

The preintervention questionnaire was administered in person by the first author with participants in a private agency space, unless other arrangements were requested. The postintervention questionnaire was completed at the final group session with co-facilitators, unless other arrangements were requested. Data were collected from May 2015-April 2016.

Measures

The present study used a variety of measures to assess clinical outcomes associated with shortened and modified versions of TREM. The same measures used in the WCDVS were used to assess mental health symptoms, trauma histories, substance use, and posttraumatic symptoms to facilitate comparisons with prior TREM studies. Measures of attachment, social support, and emotional regulation were added in the current study to explore relational outcomes within and between treatment conditions.

The Relationship Scale Questionnaire (RSQ) measures individual attachment in close relationships (Griffin & Bartholomew, 1994). Thirty items are rated on a 5-point scale to assess attachment-related feelings, expectations, and motivations toward relationships. Higher scores reflect greater attachment insecurity. Prior studies have demonstrated high convergent validity (Scharfe & Cole, 2006) and good internal consistency for attachment anxiety (α =.83-.86) and attachment avoidance (Fraley, Roisman, Booth-LaForce, Owen, & Holland, 2013; Kirchmann et al., 2012; Kurdek, 2002). Based on Kurdek’s analysis and recommendations of the developers (Bartholomew, n.d.), this study adopted 13 prompts that loaded most adequately onto the subscales, resulting in five items for attachment anxiety and eight items for attachment avoidance. Internal consistency estimates in the present study were acceptable to good for pre- and postintervention, varying between .62 and .85.

The Social Group Attachment Scale (SGAS) measures group attachment style with 25 items that are rated on a 7-point scale, with higher scores indicating greater group attachment insecurity (Smith et al., 1999). Consistent with other studies (Keating et al., 2014; Smith et al., 1999), 19 of the 25 items loaded adequately on either the anxiety subscale (10 prompts) or avoidance subscale (nine prompts); these 19 items were used. Similar to prior research (Keating et al., 2014), Cronbach’s alphas were acceptable to good for the SGAS at pre- and postintervention, varying between .73 and .87.

The Social Provisions Scale (SPS) is a 24-item measure of six types of perceived social support: reliable alliance, guidance, attachment, social integration, reassurance of worth, and opportunity to provide nurturance (Cutrona & Russell, 1987). Respondents indicate the degree of perceived support in their social relationships on a 4-point scale, with higher scores indicating greater perceived social support. Convergent and divergent validity have been supported (Gottlieb & Bergen, 2010). Internal consistency for the full scale has been shown to be excellent (Gottlieb & Bergen, 2010), and in the present study, it was acceptable to good (α = .87, preintervention; α = .76, postintervention).

The Difficulties in Emotional Regulation Scale (DERS) is a 36-item scale that assesses difficulty regulating emotions during times of distress across six dimensions: nonacceptance of emotional responses, difficulty engaging in goal-directed behavior, lack of emotional awareness, lack of emotional clarity, difficulties controlling impulsive behaviors, and limited access to effective emotion regulation strategies (Gratz & Roemer, 2004). Each item is rated on a 5-point scale, with higher scores reflecting greater emotional dysregulation. The DERS has demonstrated adequate construct and predictive validity (Gratz & Roemer, 2004) and good to excellent internal consistency among clinical and nonclinical samples (Fowler et al., 2014; Gratz & Roemer, 2004). Good test–retest reliability has been found over 4- and 8-week periods (Gratz & Roemer, 2004). In the present study, good internal consistency was found pre- and postintervention (α = .84 and .85, respectively).

The Brief Symptom Inventory–18 (BSI-18) measures psychological distress related to depression, anxiety, and somatization (Derogatis, 2001). Respondents are asked to rate how much they have been bothered by 18 different symptoms in the past week on a 5-point scale, with higher scores indicating greater distress. The present study used the six depression and six anxiety items. Factorial and concurrent validity have been previously demonstrated (Derogatis, 2001), as well as acceptable to good internal consistency (Rehan, Antfolk, Johansson, & Santtila, 2017; Wang et al., 2010). The current study showed good to excellent internal consistency for depression and anxiety at pre- and postintervention, varying from .87-.97.

The PTSD Symptom Scale (PSS) is a 17-item scale that assesses posttraumatic diagnostic criteria and symptom severity related to reexperiencing, avoidance, and arousal symptom clusters (Foa, Riggs, Dancu, & Rothbaum, 1993). Each item is rated on a 4-point scale, with higher scores indicating greater symptom severity. In accordance with the WCDVS, the present study summed the PSS as a measure of symptom severity, not as a diagnostic tool (Fallot et al., 2011). Convergent and concurrent validity have been demonstrated by prior research (Powers, Gillihan, Rosenfield, Jerud, & Foa, 2012). In TREM studies, 1-week test–retest reliability (Fallot et al., 2011) and internal consistency (α = .90) were acceptable to excellent (Amaro, Dai et al., 2007). In the present study, Cronbach’s alphas were .89 (pre) and .97 (post), indicating very good to excellent internal consistency.

Substance use was assessed using the Addiction Severity Index (ASI), a widely used and validated measure (McLellan, Luborsky, Woody, & O’Brien, 1980). Respondents were asked to report the number of days of use of 13 substances in the past 30 days and number of years of use in their lifetimes. Prior TREM research has reported good 1 week test–retest reliabilities for alcohol (α = .82) and drug subscales (Fallot et al., 2011), good internal consistency for the alcohol severity subscale (Amaro, Dai et al., 2007), and acceptable internal consistency for the drug severity subscale (Amaro, Dai et al., 2007). As with prior TREM studies, ASI scores were converted to a 0-1 scale, with higher numbers signifying greater severity of substance use.

The Life Stressor Checklist–Revised (LSC-R) assesses lifetime experience of highly stressful and/or traumatic events (Wolfe, Kimerling, Wilson, & Keane, 1997). The WCDVS created a modified version of LSC-R containing 30 dichotomous items, with follow-up questions for “yes” responses (McHugo, Caspi, et al., 2005). Results are reported as sums (Fallot et al., 2011). One-week test–retest reliability demonstrated intraclass correlation coefficients ranging from .77-.88 (McHugo, Caspi, et al., 2005) for the full scale and subscales. For the present study, 15 items focusing on interpersonal abuse were used with a “yes” or “no” prompt and no follow-up questions.

Prior TREM studies (Fallot et al., 2011; Toussaint et al., 2007) guided the decision to gather the following sociodemographic information: age, race/ethnicity, education, employment status, and relationship status with study-designed questions.

A facilitator-report fidelity checklist was created by the first author for this study. One facilitator in each group, excluding the first author, marked “yes” or “no” to whether weekly discussion questions and activities in the TREM manual were addressed. The checklist for ATREM included the same questions as TREM along with prompts for attachment items. The fidelity checklists verified that both ATREM and TREM followed their curricula closely (95.15% and 93.64%, respectively), with no statistically significant differences between the treatment groups (z = 1.01; p = .13).

Data Analysis

All data analyses were conducted using SPSS software. Given the potential for selection bias with quasi-experimental designs, data analysis began by assessing sociodemographic and clinical differences between participants in each study condition. Chi-square tests and independent t tests were used for categorical and continuous variables, respectively. The two study conditions included participants in TREM or ATREM groups held at three agencies. An outpatient behavioral health facility conducted four ATREM groups and two TREM groups. A residential substance use treatment facility conducted two TREM groups and, for logistical reasons, did not conduct any ATREM groups. An outpatient victim services agency conducted one ATREM and one TREM group. As with some previous TREM studies (Amaro, Chernoff, Brown, Arévalo, & Gatz, 2007; Fallot et al., 2011), participants from different settings were combined to create two group conditions. For the present study, ATREM data from two agencies were combined to make one ATREM group condition, and TREM data from the three agencies were combined. Analyses found no statistically significant sociodemographic or baseline clinical differences between ATREM and TREM, as displayed in Table 1.

Table 1.

Descriptive and Clinical Demographics at Baseline.

Characteristic	ATREM (n = 37)	TREM (n = 32)	p
Characteristic	M (SD)	M (SD)	p
Age	43.1 (11.1)	41.6 (13.4)	.61
Life Stressor Checklist^a	6.8 (2.8)	7.8 (3.7)	.24
	n (%)	n (%)	p
Race/Ethnicity			.66
White	35 (94.6)	29 (90.6)
African American, Latina, Other	2 (5.4)	3 (9.4)
Education			.1
HS graduate or less	17 (45.9)	21 (65.6)
Higher than HS graduate^b	20 (54.1)	11 (34.4)
Employment			.98
Working	8 (21.6)	7 (21.9)
Not working^c	29 (78.4)	25 (78.1)
Relationship			.42
Married/Significant other	15 (40.5)	10 (31.3)
Not in a relationship^d	22 (59.5)	22 (68.8)
Attachment
RSQ–Individual
Anxiety	3.4 (1.1)	3.4 (1.1)	.92
Avoidance	3.2 (0.6)	3.3 (0.6)	.47
SGAS–Group
Anxiety	4.4 (1.2)	4.5 (1.2)	.55
Avoidance	3.9 (1.3)	4.3 (1.3)	.15
Social support
SPS total	72.6 (11.4)	67.9 (14.1)	.14
Emotional regulation
DERS total	117.5 (23.0)	116.7 (28.5)	.9
Mental health/Substance use
BSI
Depression	2.2 (1.0)	2.1 (1.1)	.59
Anxiety	2.2 (1.0)	2.2 (1.0)	.85
PSS total	31.6 (10.9)	31.6 (11.5)	.98
ASI
Recent	0.0 (0.0)	0.0 (0.0)	.8

Note. No significant differences between ATREM and TREM were detected. ATREM = Attachment-Informed Trauma Recovery and Empowerment Model; TREM = Trauma Recovery Empowerment Model; HS = high school; RSQ = Relationship Scale Questionnaire; SGAS = Social Group Attachment Scale; SPS = Social Provisions Scale; DERS = Difficulties in Emotional Regulation Scale; BSI = Brief Symptom Inventory; PSS = Posttraumatic stress disorder Symptom Scale; ASI = Addiction Severity Index.

Yes/No count of exposure to 15 various stressors/trauma experiences.

Post-HS/training, some college, college graduate, graduate level.

Not working, caregiver, not working due to disability.

Divorced/separated, single (never married), widowed.

Analyses were conducted to determine whether there were any sociodemographic differences associated with study completion. Study completion was defined as attending 62.5% of the group sessions and completing the postintervention questionnaire. Based on facilitators’ reports, 69 participants (53.49%) completed the study, which falls in the mid-range of prior retention and follow-up rates in TREM studies (Amaro, Chernoff et al., 2007; Cihlar, 2014; Fallot et al., 2011; Toussaint et al., 2007). There were no statistically significant differences in sociodemographic characteristics related to study completion. The completion rate for ATREM (57.8%) was higher than TREM (49.2%), but the difference was not statistically significant (z = 0.97; p = .33). Reported findings are based on the 69 participants who completed the study (n = 32 in TREM and n = 37 in ATREM). No participant payments or incentives were offered.

The central aim of the study, to compare the effectiveness of TREM and ATREM, was pursued with independent t tests and linear regression analyses with group condition (ATREM or TREM) as the independent variable and individual and group attachment, emotion regulation, PTSD symptoms, depression, anxiety, substance use, and perceived social support as dependent variables, similar to analyses in prior TREM studies (Fallot et al., 2011; Morrissey et al., 2005; Toussaint et al., 2007). Given the absence of statistically significant sociodemographic and clinical differences between participants in the group conditions at postintervention, only t-test results are reported to reduce redundancy. In addition to evaluating whether ATREM was more effective than TREM, an assessment of change over time within each group condition was conducted using paired t tests to determine whether being in ATREM or TREM resulted in improvement from pre- to postintervention. A participant’s data were excluded for any scale/subscale for which she did not respond to one or more questions. In total, 10 of the 20 scales/subscales had 5% or less participant data exclusion, with a range of 0%-5.4% among SGAS anxiety and avoidance, DERS, and BSI depression and anxiety. The remaining 10 scales/subscales had 6.3-15.6% of participant data excluded, which included RSQ anxiety and avoidance, SPS, PSS, and ASI. SPS for TREM was at the high end of the range (15.6%). As the exclusion rate increases, caution in data interpretation should be considered.

Results

Sample Characteristics

Participants had a mean age of 42.41 years (SD = 12.15) and were predominantly Caucasian (92.8%). Most participants had a high school diploma/General Educational Development certificate (GED) or less education (55.1%), were not working (78.3%), mainly due to disability (49.3%), and were not presently in a relationship (63.8%). On average, participants were exposed to 7.33 of 15 traumatic events (SD = 3.27), based on baseline responses to LSC-R.

Clinical Intervention Outcomes

From pre- to postintervention, both ATREM and TREM were associated with statistically significant reductions in individual and group attachment anxiety, group attachment avoidance, perceived social support, emotion regulation capacities, psychological distress, depression, anxiety, and PTSD symptom severity. However, only ATREM was associated with statistically significant reductions in individual attachment avoidance. Neither group condition demonstrated statistically significant differences in substance use, as displayed in Table 2.

Table 2.

Mean Differences for Within-Group Pre- to Postintervention Changes.

Scale	ATREM (n = 37) post–pre				TREM (n = 32) post–pre
Scale	M_diff (SD)	t(df)	p	Cohen’s d	M_diff (SD)	t(df)	p	Cohen’s d
Attachment
RSQ–Individual
Anxiety	−0.3 (0.9)	t(33) = −2.3	.01*	0.4	−0.4 (0.9)	t(28) = −2.1	.03*	0.39
Avoidance	−0.3 (0.7)	t(33) = −2.7	.00**	0.46	−0.3 (0.9)	t(28) = −1.6	.06	0.3
SGAS–Group
Anxiety	−0.8 (1.3)	t(34) = −3.4	.00**	0.58	−0.9 (1.7)	t(30) = −3.0	.00**	0.53
Avoidance	−0.5 (1.2)	t(35) = −2.3	.01*	0.41	−0.9 (1.5)	t(30) = −3.3	.00**	0.6
Social support
SPS	3.3 (9.0)	t(33) = 2.1	.02*	0.38	5.5 (13.4)	t(26) = 2.1	.02*	0.42
Emotional regulation
DERS	−18.0 (23.8)	t(36) = −4.6	.00**	0.76	−18.8 (26.4)	t(31) = −4.0	.00**	0.72
Mental health/Substance use
BSI
Depression	−0.4 (0.8)	t(35) = −3.2	.00**	0.54	−0.6 (1.3)	t(30) = −2.6	.00**	0.47
Anxiety	−0.4 (0.8)	t(35) = −2.9	.00**	0.48	−0.6 (1.0)	t(31) = −3.5	.00**	0.61
PSS	−4.7 (11.4)	t(32) = −2.4	.01*	0.41	−6.8 (15.0)	t(27) = −2.4	.01*	0.48
ASI
Recent	0.0 (0.1)	t(27) = 0.62	.73	0.16	0.0 (0.0)	t(29) = −1.6	.07	0.33

Note. ATREM = Attachment-Informed Trauma Recovery and Empowerment Model; TREM = Trauma Recovery Empowerment Model; RSQ = Relationship Scale Questionnaire; SGAS = Social Group Attachment Scale; SPS = Social Provisions Scale; DERS = Difficulties in Emotional Regulation Scale; BSI = Brief Symptom Inventory; PSS = Posttraumatic stress disorder Symptom Scale; ASI = Addiction Severity Index.

p < .05. **p < .01.

No statistically significant differences were found between ATREM and TREM for any of the clinical outcomes from pre- to postintervention, as displayed in Table 3.

Table 3.

Mean Differences Between ATREM and TREM in Pre- to Postintervention Changes.

Scale	ATREM_diff-TREM_diff
Scale	M_d_iff (SE)	t(df)	p	Cohen’s d
Attachment
RSQ–Individual
Anxiety	0.0 (0.2)	t(61) = 0.1	.53	0.02
Avoidance	0.0 (0.2)	t(61) = −0.2	.44	0.04
SGAS–Group
Anxiety	0.2 (0.4)	t(64) = 0.4	.66	0.11
Avoidance	0.4 (0.3)	t(65) = 1.2	.88	0.29
Social support
SPS	−2.2 (3.0)	t(43.5) = −0.7	.76	0.2
Emotional regulation
DERS	0.8 (6.0)	t(67) = 0.1	.56	0.03
Mental health/Substance use
BSI
Depression	0.2 (0.3)	t(49.1) = 0.6	.72	0.15
Anxiety	0.2 (0.2)	t(66) = 0.8	.78	0.19
PSS	2.2 (3.4)	t(59) = 0.6	.74	0.16
ASI recent	0.0 (0.0)	t(60) = 1.5	.93	0.4

Note. No significant differences between ATREM and TREM on any of the scales. ATREM = Attachment-Informed Trauma Recovery and Empowerment Model; TREM = Trauma Recovery Empowerment Model; RSQ = Relationship Scale Questionnaire; SGAS = Social Group Attachment Scale; SPS = Social Provisions Scale; DERS = Difficulties in Emotional Regulation Scale; BSI = Brief Symptom Inventory; PSS = Posttraumatic stress disorder Symptom Scale; ASI = Addiction Severity Index.

The effect sizes for within- and between-group change were small to medium as displayed in Tables 2 and 3.

Discussion and Implications

This pilot study developed and tested the effectiveness of a shortened version of TREM and a novel attachment-infused version of it, ATREM. To our knowledge, this is the first study to develop a modified curriculum integrating attachment-based concepts and strategies with TREM. ATREM was associated with statistically significant improvement in individual and group attachment, perceived social support, emotion regulation, depression, anxiety, and PTSD. There was no statistically significant improvement in substance use, which may be attributable to the treatment settings and lower levels of use. These results were also found with TREM, but only ATREM demonstrated a statistically significant decrease in individual attachment avoidance from pre- to postintervention. This ATREM finding is especially noteworthy given that individual attachment avoidance is considered difficult to modify (Marmarosh et al., 2013). Prior research has shown that individuals with high attachment avoidance experience some of the strongest challenges in engaging and remaining in treatment, resulting in reduced clinical gains and making accuracy of attuned responses particularly important during initial and relationally intense sessions (Marmarosh et al., 2013). While the gains observed in this study are important, no statistically significant differences were found between ATREM and TREM for any of the clinical outcomes.

The comparable findings of ATREM and TREM for between- and within-group change, along with the additional gains for ATREM with individual attachment avoidance, suggest that ATREM may be a viable treatment alternative to the 16-week version of TREM. Furthermore, the statistically nonsignificant differences in effectiveness between ATREM and TREM need to be considered in light of the high standard that was set for detecting change. Prior intervention studies on attachment or TREM typically lacked a control/comparison group or involved a wait list/treatment-as-usual control group rather than including comparison with another treatment (Fallot et al., 2011; Fonagy et al., 1996; Kinley & Reyno, 2013). Comparing a treatment group and a control group involves greater potential to detect change than comparing two treatment groups, especially when one treatment is empirically supported and the other is an adaptation that retains the essential framework of the comparison condition. Despite ATREM being a novel, untested adaptation of TREM, the higher standard of a comparative effectiveness design was chosen because a treatment-to-treatment comparison provides more relevant insights for clinical practice. Comparable findings of ATREM and TREM can be seen as advantageous in that they provide options for clinicians with diverse clinical abilities and theoretical training and for clients with diverse presenting concerns. In addition, the findings for both TREM and ATREM suggest that the full course of treatment in the traditional model may not be necessary for positive change given that the 16-week versions piloted in this study demonstrated statistically significant improvements in the clinical domains assessed in the WCDVS studies, as well as in attachment and social support. Reductions in attachment anxiety and avoidance are often thought to require long-term therapy (Strauss, Mestel, & Kirchmann, 2011), but this study suggests that short-term group therapies such as TREM and ATREM can be helpful. For ATREM, the short-term format with psychodynamic modifications demonstrates that the longer duration of treatment often associated with psychodynamic formats (Knight, 2006) may not always be necessary for positive gains. Finally, challenges with retention in trauma group therapy (Chouliara et al., 2017; McHugo, Kammerer, et al., 2005) suggest that completion of treatment may be more attainable with shorter time frames.

The current study extends attachment knowledge by assessing both group and individual attachment. Knowledge of group attachment styles provides an additional perspective for understanding individual behavior in group settings and may offer an underutilized avenue for facilitating positive change. Group attachment research is in its infancy (Marmarosh, 2015), but it has shown associations between decreases in group attachment insecurity and increases in security in intimate relationships outside of the group context, suggesting that a focus on enhancing group attachment can have broad impact in participants’ lives (Keating et al., 2014). Awareness of both group and individual attachment styles may be especially relevant with people who have experienced trauma. They often develop both types of fearful attachment styles, which can negatively affect the achievement of therapeutic gains (Marmarosh et al., 2013). Group members with dual fearful attachment styles may experience no buffer of support against feared revictimization in a group if they do not feel safe relying on the group, any specific member, or the leader when discussions feel emotionally unsafe or overwhelming (Markin & Marmarosh, 2010). Marmarosh and colleagues (2013) suggest that a secure base grounded in empathy from the outset of treatment may be especially important to assist them with managing and “integrat[ing] their contradictory impulses to merge and withdraw” (p. 167) in the group context.

Knowing group members’ attachment styles prior to the first group session, as with ATREM, can provide valuable information for tuning in to possible relational needs. Preparatory attunement to possible relational reactions carries potential to enhance initial and ongoing interactions by helping clinicians to be sensitive listeners (Shulman, 2015). The likelihood of accurate and timely attunement and responsiveness is increased by prior attachment knowledge, in part, because clinicians are primed to recognize clients’ relational styles (Marmarosh et al., 2013). Attuned and responsive clinicians help create and maintain a secure base from which members feel respected in their style of group engagement and supported in repairing relational ruptures, which may enhance group completion and clinical gains.

Limitations

Discussion of potential benefits needs to be tempered by the limitations of the study. Most importantly, the quasi-experimental design does not control for unmeasured baseline differences, and the small sample size may have had limited statistical power to detect differences. In addition, the relational nature of both ATREM and TREM may have contributed to the challenge of detecting a differential treatment effect. Relationships between members and with co-facilitators are fundamental to the healing process in both conditions, despite differing approaches to working with relationships. Given this overlap, some dilution of distinctiveness was inevitable because most clinicians, regardless of their theoretical orientations, value the therapeutic alliance (Lambert & Barley, 2001) and strive to create a secure base described by attachment theory (Bowlby, 1988). Protocol distinctions may have been facilitated by expanding reflective functioning in ATREM through more mentalization-based activities, which are considered fundamental in attachment therapies (Marmarosh et al., 2013). These approaches are challenging to manualize because they are “more abstract and unstructured” (Field, 2014, p. 21) than cognitive behavioral approaches. Nonetheless, in recognition that manuals can enhance the delivery of protocol-specific elements (Tasca, Balfour, Ritchie, & Bissada, 2006), a detailed ATREM manual was used. Fidelity to the manual and appropriate use of more abstract approaches could be enhanced by training and supervision that incorporate videotaping of group sessions (Marmarosh, 2015). The fidelity checklist used in the current study was a practical and feasible way to monitor the delivery of the psychoeducational content but was limited in capturing experiential processing.

Clinical Implications

Attachment histories and manifestations inevitably enter the therapeutic space of any group therapy approach. The present study demonstrates the feasibility and value of intentionally using attachment-based strategies to enhance assessment and interventions in group therapy. Attachment perspectives and strategies can assist clinicians in working with and through complicated relational dynamics by using these interactions as key elements of treatment. Furthermore, an awareness of group attachment style can create an additional in-road for healing from trauma. Attachment-informed clinicians may facilitate the development of each participant’s sense of a secure base through differential engagement during whole-group and dyadic activities, depending on each participant’s interpersonal strengths and needs. For example, attachment-informed facilitators can support the development of group cohesion for members with high attachment anxiety by eliciting constructive feedback and perceptions from other group members (Gallagher et al., 2014; Marmarosh et al., 2013). This information may reduce attachment anxiety–driven assumptions or misperceptions of being disliked or abandoned. For a group member experiencing high attachment avoidance, the development of group cohesion can feel threatening. In this context, facilitators may need to titrate group intensity by initially focusing on symptoms over emotions and using skills-based, educational approaches (Marmarosh et al., 2013; Muller, 2009). During challenging discussions about group dynamics, members experiencing high attachment anxiety may benefit from calming and focusing strategies, such as mindful breathing and sharing feelings with a partner prior to engaging with the whole group. In these situations, members experiencing high attachment avoidance may initially benefit from strategies that energize the thinking process, such as brainstorming or problem solving as a whole group and later benefit from support to experience, rather than block, here-and-now feelings. Hence, attachment-informed facilitators know when it is best to “strike when the iron is hot or cold” (Marmarosh et al., 2013, p. 114).

Future Research

Recent recommendations have been made for more attachment-oriented studies focusing on integrated attachment–cognitive behavioral therapy approaches (Taylor et al., 2015), similar to the ATREM protocol. The statistically significant within-group changes associated with ATREM suggest that this new protocol carries promise and warrants further examination with modifications to the study design, including future research with 6- and/or 12-month follow-up to determine whether significant differences between ATREM and TREM emerge over time, as has been found with other attachment interventions (Kilmann et al., 1999). It is possible that the full extent of attachment-based changes requires more time to manifest as new insights are absorbed and practiced outside of the group setting. Higher powered studies could examine potential mediators to clarify linkages between reductions in attachment anxiety and avoidance and clinical outcomes in group therapy. Examining ways in which facilitators’ attachment styles affect interactions with group members, interactions between co-facilitators, and outcomes would also be valuable, especially with participants experiencing greater trauma-related challenges (Degnan, Seymour, Hyde, Harris, & Berry, 2016).

Given the high attrition rate in prior TREM research (Amaro, Dai et al., 2007) and the nonsignificantly higher completion rate of ATREM in this study, it would be informative to examine the role of attachment-based strategies in group completion in future research. In addition, an assessment of group members’ concurrent involvement in individual therapy, use of prescribed psychiatric medication, and/or engagement with other psychosocial resources would strengthen future studies by facilitating examination of group participation gains in relation to other services. Finally, further research that examines group attachment style may offer additional intervention resources to enhance individual and relational well-being (Marmarosh, 2015).

Conclusion

Interpersonal trauma often involves feeling isolated, abandoned, and psychologically alone during and after intolerably distressing experiences (Allen, 2013; Herman, 1997). Such experiences often require interpersonal repairs that enhance feelings of emotional connection in important relationships. Group therapy provides multiple opportunities for and forms of relational connection, making this type of treatment well-suited to assisting people who have experienced trauma. However, group members’ isolation and fears often constrain relational openness in informal and formal settings, including therapeutic settings, suggesting that focused efforts may be required to facilitate healing interpersonal connections (Lundqvist et al., 2009). The creation of ATREM, through the integration of attachment-based strategies with TREM, was designed to provide implicit and explicit support to enhance such healing connections and the ability of participants to fully benefit from group involvement. Both ATREM and the 16-week version of TREM showed that significant clinical gains can be achieved in a shortened time frame. Given ATREM’s results in this pilot study, future research that is higher powered is warranted to determine whether clinical domains can be enhanced beyond prior TREM findings. ATREM’s integration of cognitive, behavioral and psychodynamic elements equips therapists with multidimensional treatment strategies and individualized relational knowledge to facilitate more attuned and responsive interactions. ATREM provides a novel adaptation of TREM that offers integrative strategies and attachment insights to assist women in recovering from trauma.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Melanie Masin-Moyer, DSW, LCSW, has been in clinical practice for more than 20 years in nonprofit community behavioral health care. She has worked with clients across the life span in individual, couples, family, and group therapy modalities. Most recently, she devotes her clinical practice to group therapy with women who have experienced trauma. She completed her clinical doctorate in social work in May 2017 at the University of Pennsylvania where she is now a full-time lecturer. Her research interests include attachment, mental health, women’s trauma, and group therapy.

Malitta Engstrom, PhD, is an associate professor at the University of Pennsylvania School of Social Policy & Practice, as well as a faculty fellow at the Ortner Center on Violence & Abuse in Relationships and a senior fellow at the Leonard Davis Institute of Health Economics. With an emphasis on women and families, her research aims to advance understanding of and services to address intersecting concerns related to substance use, criminal justice system involvement, victimization, HIV, and mental health. The National Institute on Drug Abuse, John A. Hartford Foundation, Penn University Research Foundation, and other organizations have funded her research.

Phyllis Solomon, PhD, is the Kenneth L. M. Pray professor and associate dean for research in the School of Social Policy & Practice, professor of social work in psychiatry at the Perelman School of Medicine, faculty fellow at the Ortner Center on Violence & Abuse in Relationships, and senior fellow at the Center for Public Health Initiatives, University of Pennsylvania. Her primary research focus is on services, service systems, and service interventions for persons with severe mental illness and their families, including issues of family violence.

References

Afifi

T. O.

MacMillan

H. L.

(2011). Resilience following child maltreatment: A review of protective factors. Canadian Journal of Psychiatry, 56, 266-272.

Allen

(2013). Treating attachment trauma with plain old therapy. Journal of Trauma & Dissociation, 14, 367-374.

Amaro

Chernoff

Brown

Arévalo

Gatz

(2007). Does integrated trauma-informed substance abuse treatment increase treatment retention? Journal of Community Psychology, 35, 845-862.

Amaro

Dai

Arévalo

Acevedo

Matsumoto

Nieves

Prado

(2007). Effects of integrated trauma treatment on outcomes in a racially/ethnically diverse sample of women in urban community-based substance abuse treatment. Journal of Urban Health, 84, 508-522.

Bartholomew

(n.d.). Self-report attachment measures. Retrieved from http://members.psyc.sfu.ca/labs/kim_bartholomew/attachment/self

Boccato

Capozza

(2011). Attachment styles and social groups: Review of a decade. Testing, Psychometrics, Methodology in Applied Psychology, 18, 19-30.

Bowlby

(1982). Attachment and loss: Attachment (2nd ed., vol. 1). New York: Basic Books.

Bowlby

(1988). A secure base: Parent-child attachment and healthy human development. New York: Basic Books.

Breiding

M. J.

(2015). Prevalence and characteristics of sexual violence, stalking, and intimate partner violence victimization-national intimate partner and sexual violence survey, United States, 2011. American Journal of Public Health, 105(4), e11-e12.

10.

Burton

M. S.

Cooper

A. A.

Feeny

N. C.

Zoellner

L. A.

(2015). The enhancement of natural resilience in trauma interventions. Journal of Contemporary Psychotherapy, 45, 193-204.

11.

Chouliara

Karatzias

Gullone

Ferguson

Cosgrove

Draucker

C. B.

(2017). Therapeutic change in group therapy for interpersonal trauma: A relational framework for research and clinical practice. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517696860

12.

Cihlar

B. E.

(2014). The trauma recovery and empowerment model: A trauma-informed treatment program for female offenders in the community (Doctoral dissertation). Available from ProQuest Dissertations and Theses database. (3608215). Retrieved from https://pqdtopen.proquest.com/pubnum/3608215.html

13.

Cocozza

J. J.

Jackson

E. W.

Hennigan

Morrissey

J. P.

Reed

B. G.

Fallot

Banks

(2005). Outcomes for women with co-occurring disorders and trauma: Program-level effects. Journal of Substance Abuse Treatment, 28, 109-119.doi:10.1016/j.jsat.2004.08.010

14.

Cutrona

C. E.

Russell

D. W.

(1987). The provisions of social relationships and adaptation to stress. Advances in Personal Relationships, 1, 37-67.

15.

Degnan

Seymour-Hyde

Harris

Berry

(2016). The role of therapist attachment in alliance and outcome: A systematic literature review. Clinical Psychology & Psychotherapy, 23, 47-65.

16.

Derogatis

L. R.

(2001). BSI 18, brief symptom inventory 18: Administration, scoring and procedures manual. Minneapolis, MN: NCS Pearson.

17.

Devries

K. M.

Mak

J. Y.

Bacchus

L. J.

Child

J. C.

Falder

Petzold

, et al. (2013). Intimate partner violence and incident depressive symptoms and suicide attempts: A systematic review of longitudinal studies. PLoS Medicine, 10(5), e1001439. doi:10.1371/journal.pmed.1001439

18.

Fallot

R. D.

McHugo

G. J.

Harris

Xie

(2011). The trauma recovery and empowerment model: A quasi-experimental effectiveness study. Journal of Dual Diagnosis, 7, 74-89.

19.

Field

T. A.

(2014). Integrating left-brain and right-brain: The neuroscience of effective counseling. The Professional Counselor, 4, 19-27.

20.

Finkelstein

VandeMark

Fallot

Brown

Cadiz

Heckman

(2004). Enhancing substance abuse recovery through integrated trauma treatment. Retrieved from https://www.samhsa.gov/sites/default/files/wcdvs-article.pdf

21.

Flores

P. J.

(2010). Group psychotherapy and neuro-plasticity: An attachment theory perspective. International Journal of Group Psychotherapy, 60, 546-570.

22.

Foa

E. B.

Riggs

D. S.

Dancu

C. V.

Rothbaum

B. O.

(1993). Reliability and validity of a brief instrument for assessing post-traumatic stress disorder. Journal of Traumatic Stress, 6, 459-473.

23.

Fonagy

Bateman

A. W.

(2006). Mechanisms of change in mentalization-based treatment of BPD. Journal of Clinical Psychology, 62, 411-430. doi:10.1002/jclp.20241

24.

Fonagy

Leigh

Steele

Kennedy

Mattoon

, et al. (1996). The relation of attachment status, psychiatric classification, and response to psychotherapy. Journal of Consulting and Clinical Psychology, 64, 22-31. doi:10.1037/0022-006X.64.1.22

25.

Fowler

J. C.

Charak

Elhai

J. D.

Allen

J. G.

Frueh

B. C.

Oldham

J. M.

(2014). Construct validity and factor structure of the difficulties in emotion regulation scale among adults with severe mental illness. Journal of Psychiatric Research, 58, 175-180.

26.

Fraley

R. C.

Roisman

G. I.

Booth-LaForce

Owen

M. T.

Holland

A. S.

(2013). Interpersonal and genetic origins of adult attachment styles: A longitudinal study from infancy to early adulthood. Journal of Personality and Social Psychology, 104, 817-838. doi:10.1037/a0031435

27.

Gallagher

M. E.

Tasca

G. A.

Ritchie

Balfour

Maxwell

Bissada

(2014). Interpersonal learning is associated with improved self-esteem in group psychotherapy for women with binge eating disorder. Psychotherapy, 51, 66-77. doi:10.1037/a0031098

28.

Gottlieb

B. H.

Bergen

A. E.

(2010). Social support concepts and measures. Journal of Psychosomatic Research, 69, 511-520. doi:10.1016/j.jpsychores.2009.10.001

29.

Gratz

K. L.

Roemer

(2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54.

30.

Griffin

D. W.

Bartholomew

(1994). The metaphysics of measurement: The case of adult attachment. In Bartholomew

Perlman

(Eds.), Advances in personal relationships: Attachment processes in adulthood (pp. 17-52, vol. 5). London, England: Jessica Kingsley.

31.

Harris

Anglin

(1998). Trauma recovery and empowerment: A clinician’s guide for working with women in groups. New York: Simon and Schuster.

32.

Herman

J. L.

(1997). Trauma and recovery. New York: Basic books.

33.

Hillberg

Hamilton-Giachritsis

Dixon

(2011). Review of meta-analyses on the association between child sexual abuse and adult mental health difficulties: A systematic approach. Trauma, Violence, & Abuse, 12, 38-49. doi:10.1177/1524838010386812

34.

Karatzias

Ferguson

Gullone

Cosgrove

(2016). Group psychotherapy for female adult survivors of interpersonal psychological trauma: A preliminary study in Scotland. Journal of Mental Health, 25, 512-519.

35.

Keating

Tasca

G. A.

Gick

Ritchie

Balfour

Bissada

(2014). Change in attachment to the therapy group generalizes to change in individual attachment among women with binge eating disorder. Psychotherapy, 51, 78-87. doi:10.1037/a0031099

36.

Kilmann

P. R.

Laughlin

J. E.

Carranza

L. V.

Downer

J. T.

Major

Parnell

M. M.

(1999). Effects of an attachment-focused group preventive intervention on insecure women. Group Dynamics: Theory, Research, and Practice, 3, 138-147. doi:10.1037/1089-2699.3.2.138

37.

Kim

Wildeman

Jonson-Reid

Drake

(2017). Lifetime prevalence of investigating child maltreatment among US children. American Journal of Public Health, 107, 274-280. doi:10.2105/AJPH.2016.303545

38.

Kinley

J. L.

Reyno

S. M.

(2013). Attachment style changes following intensive short-term group psychotherapy. International Journal of Group Psychotherapy, 63, 53-75.

39.

Kirchmann

Steyer

Mayer

Joraschky

Schreiber-Willnow

Strauss

(2012). Effects of adult inpatient group psychotherapy on attachment characteristics: An observational study comparing routine care to an untreated comparison group. Psychotherapy Research, 22, 95-114.

40.

Knight

(2006). Groups for individuals with traumatic histories: Practice considerations for social workers. Social Work, 51, 20-30.

41.

Kurdek

L. A.

(2002). On being insecure about the assessment of attachment styles. Journal of Social and Personal Relationships, 19, 811-834.

42.

Lambert

M. J.

Barley

D. E.

(2001). Research summary on the therapeutic relationship and psychotherapy outcome. Psychotherapy, 38, 357-361. doi:10.1037/0033-3204.38.4.357

43.

Levy

K. N.

Ellison

W. D.

Scott

L. N.

Bernecker

S. L.

(2011). Attachment style. Journal of Clinical Psychology, 67, 193-203.

44.

Lundqvist

Hansson

Svedin

C. G.

(2009). Group therapy for women sexually abused as children: Social interaction, adjustment, and relationships before and after group therapy. Psychoanalytic Social Work, 16, 158-175. doi:10.1080/15228870903200335

45.

Magnavita

J. J.

Anchin

J. C.

(2013). Unifying psychotherapy: Principles, methods, and evidence from clinical science. New York: Springer.

46.

Markin

R. D.

Marmarosh

(2010). Application of adult attachment theory to group member transference and the group therapy process. Psychotherapy, 47, 111-121. doi:10.1037/a0018840

47.

Marmarosh

C. L.

(2015). Emphasizing the complexity of the relationship: The next decade of attachment-based psychotherapy research. Psychotherapy, 52, 12-18. doi:10.1037/a0036504

48.

Marmarosh

C. L.

Markin

R. D.

Speigel

E. B.

(2013). Attachment in group psychotherapy. Washington, DC: American Psychological Association.

49.

McHugo

G. J.

Caspi

Kammerer

Mazelis

Jackson

Russell

, et al. (2005). The assessment of trauma history in women with co-occurring substance abuse and mental disorders and a history of interpersonal violence. Journal of Behavioral Health Services & Research, 32, 113-127.

50.

McHugo

G. J.

Kammerer

Jackson

E. W.

Markoff

L. S.

Gatz

Larson

M. J.

, et al. (2005). Women, co-occurring disorders, and violence study: Evaluation design and study population. Journal of Substance Abuse Treatment, 28, 91-107. doi:10.1016/j.jsat.2004.08.009

51.

McLellan

A. T.

Luborsky

Woody

G. E.

O’Brien

C. P.

(1980). An improved diagnostic evaluation instrument for substance abuse patients: The Addiction Severity Index. Journal of Nervous and Mental Disease, 168, 26-33.

52.

Mikulincer

Shaver

P. R.

(2010). Attachment in adulthood: Structure, dynamics, and change. New York: Guilford Press.

53.

Morrissey

J. P.

Ellis

A. R.

Gatz

Amaro

Reed

B. G.

Savage

, et al. (2005). Outcomes for women with co-occurring disorders and trauma: Program and person-level effects. Journal of Substance Abuse Treatment, 28, 121-133.

54.

Muller

R. T.

(2009). Trauma and dismissing (avoidant) attachment: Intervention strategies in individual psychotherapy. Psychotherapy, 46, 68-81. doi:10.1037/a0015135

55.

Muller

R. T.

Gragtmans

Baker

(2008). Childhood physical abuse, attachment, and adult social support: Test of a mediational model. Canadian Journal of Behavioural Science / Revue anadienne des sciences du comportement, 40, 80-89. doi:10.1037/0008-400X.40.2.80

56.

Paquin

J. D.

Kivlighan

D. M.

Drogosz

L. M.

(2013). Person-group fit, group climate, and outcomes in a sample of incarcerated women participating in trauma recovery groups. Group Dynamics: Theory, Research, and Practice, 17, 95-109. doi:10.1037/a0032702

57.

Pearlman

L. A.

Courtois

C. A

. (2005). Clinical applications of the attachment framework: Relational treatment of complex trauma. Journal of Traumatic Stress, 18, 449-459.

58.

Powers

M. B.

Gillihan

S. J.

Rosenfield

Jerud

A. B.

Foa

E. B.

(2012). Reliability and validity of the PDS and PSS-I among participants with PTSD and alcohol dependence. Journal of Anxiety Disorders, 26, 617-623. doi:10.1016/j.janxdis.2012.02.013

59.

Rehan

Antfolk

Johansson

Santtila

(2017). Do single experiences of childhood abuse increase psychopathology symptoms in adulthood? Journal of Interpersonal Violence, 34, 1021-1038.

60.

Saunders

E. A.

Edelson

J. A.

(1999). Attachment style, traumatic bonding, and developing relational capacities in a long-term trauma group for women. International Journal of Group Psychotherapy, 49, 465-485.

61.

Scharfe

Cole

(2006). Stability and change of attachment representations during emerging adulthood: An examination of mediators and moderators of change. Personal Relationships, 13, 363-374.

62.

Shulman

(2015). The skills of helping individuals, families, groups, and communities (8th ed.). Boston, MA: Cengage Learning.

63.

Smith

E. R.

Murphy

Coats

(1999). Attachment to groups: Theory and management. Journal of Personality and Social Psychology, 77, 94-110. doi:10.1037/0022-3514.77.1.94

64.

Spohn

Wright

E. M.

Peterson

J. C.

(2017). Rape and mental health outcomes among women: Examining the moderating effects of "healthy" fear levels. Violence Against Women, 23, 1055-1075. doi:10.1177/1077801216655625

65.

Strauss

B. M.

Mestel

Kirchmann

H. A.

(2011). Changes of attachment status among women with personality disorders undergoing inpatient treatment. Counselling and Psychotherapy Research, 11, 275-283.

66.

Substance Abuse and Mental Health Services Administration. (n.d.). Trauma. Retrieved from http://www.integration.samhsa.gov/clinical-practice/trauma

67.

Tasca

G. A.

Balfour

Ritchie

Bissada

(2006). Developmental changes in group climate in two types of group therapy for binge-eating disorder: A growth curve analysis. Psychotherapy Research, 16, 499-514.

68.

Tasca

G. A.

Ritchie

Balfour

(2011). Implications of attachment theory and research for the assessment and treatment of eating disorders. Psychotherapy, 48, 249-259. doi:10.1037/a0022423

69.

Taylor

Rietzschel

Danquah

Berry

(2015). Changes in attachment representations during psychological therapy. Psychotherapy Research, 25, 222-238.

70.

Toussaint

D. W.

VanDeMark

N. R.

Bornemann

Graeber

C. J.

(2007). Modifications to the Trauma Recovery and Empowerment Model (TREM) for substance-abusing women with histories of violence: Outcomes and lessons learned at a Colorado substance abuse treatment center. Journal of Community Psychology, 35, 879-894.

71.

Walker

H. E.

Freud

J. S.

Ellis

R. A.

Fraine

S. M.

Wilson

L. C.

(2019). The prevalence of sexual revictimization: A meta-analytic review. Trauma, Violence, & Abuse, 20, 67-80. doi:10.1177/1524838017692364

72.

Wallin

D. J.

(2015). Attachment in psychotherapy. New York: Guilford Press.

73.

Wang

Kelly

B. C.

Booth

B. M.

Falck

R. S.

Leukefeld

Carlson

R. G.

(2010). Examining factorial structure and measurement invariance of the Brief Symptom Inventory (BSI)-18 among drug users. Addictive Behaviors, 35, 23-29. doi:10.1016/j.addbeh.2009.08.003

74.

Wolfe

Kimerling

Wilson

Keane

T. M.

(1997). Assessing psychological trauma and PTSD. In Keane

IJWTM

(Ed.), Gender issues in the assessment of Posttraumatic Stress Disorder (pp. 192-238). New York: Guilford.

75.

Yalom

I. D.

Leszcz

(2005). The theory and practice of group psychotherapy (5th ed.). New York: Basic Books.

76.

Zinzow

H. M.

Amstadter

A. B.

McCauley

J. L.

Ruggiero

K. J.

Resnick

H. S.

Kilpatrick

D. G.

(2011). Self-rated health in relation to rape and mental health disorders in a national sample of college women. Journal of American College Health, 59, 588-594.