INTRODUCTION

This study presents findings from a pilot study of Developmental Couple Therapy for Complex Trauma (DCTCT; MacIntosh, 2019, 2023). The field of couple therapy requires effective couple therapy interventions for survivors of trauma. Traumas experienced at sensitive stages of development can impact the capacity of survivors to develop and maintain couple relationships and to tolerate the process of traditional approaches to couple therapy. Interventions specifically tailored to address hindered developmental self-capacities and complex trauma sequelae are essential for the success of this endeavor (MacIntosh, 2013a; MacIntosh & Ménard, 2021b).

Literature review

Survivors of childhood trauma, including physical, sexual, emotional abuse, and neglect, often struggle to develop and maintain couple relationships that are satisfying, sustainable, and emotionally and sexually intimate. Childhood trauma survivors report high levels of conflict and intimate partner violence (IPV), attachment insecurity, dissatisfaction, sexual distress, and dysfunction, as well as higher frequencies of breakups (Godbout et al., 2013; MacIntosh & Ménard, 2021a; MacIntosh & Vaillancourt-Morel, 2020). Ongoing research has generalized the work of early researchers on the impacts of childhood sexual abuse (CSA) to understand the ways in which adult relationships are affected by all forms of childhood trauma. Survivors of all forms of childhood maltreatment face challenges in their couple relationships, with higher levels of cumulative trauma associated with greater relational difficulties (Briere et al., 2008; Fergusson et al., 2013).

Until 20 years ago, providing couple therapy was considered unwise, and potentially dangerous, for those affected by complex developmental trauma. It was believed that therapy would be too destabilizing, and that survivors should engage in individual therapy as a precursor to any kind of couple intervention (MacIntosh & Johnson, 2008; Whittaker et al., 2023). However, that conventional wisdom did not address the deep and painful struggles that survivors continued to experience in their couple relationships. Partners reported that they felt left out of deep therapeutic processes that were leading to significant growth and change in their partners (Bacon & Lein, 1996; Cisse et al., 2012). Additionally, the potential of couple interventions to address both individual-level distress, including symptoms of posttraumatic stress disorder (PTSD) and complex PTSD (CPTSD), and couple-level distress, is beginning to be better understood (MacIntosh, 2019; McWey, 2022).

For this discussion, it is important to set the scene by outlining the impacts of trauma, including how these can infiltrate couple relationships and affect the process of couple therapy. PTSD comprises a series of symptoms in response to a traumatic event, such as flashbacks, nightmares, and intrusive memories of the trauma. It is also associated with avoidance of reminders of the trauma and with emotion regulation difficulties, such as dissociation, hypervigilance, and emotional flooding (American Psychiatric Association, 2013). In addition, the International Classification of Diseases (ICD) from the World Health Organization included CPTSD in the 2018 edition of the taxonomy (Cloitre et al., 2013; Maercker, 2021; Maercker et al., 2013; Van der Kolk et al., 1993). Although still subject to controversy, the concept of CPTSD is one way of making sense of the impacts of traumas experienced during sensitive periods of development, especially interpersonal traumas within the family. In addition to intrusive thoughts, arousal, and regulation difficulties, survivors of developmental trauma who are diagnosed with CPTSD may experience an array of distressing disturbances of their self-structure. These can include struggling to have a secure and solid sense of self and identifying, tolerating, and managing emotional states; significant attachment insecurity; and even hypervigilance. Additionally, survivors can avoid closeness, interpersonal comfort and contact, culminating in high anxiety about their partner, the relationship, and all aspects of their interpersonal functioning (Cloitre et al., 2011; Courtois & Ford, 2009; Ford, 2005; Ford & Courtois, 2020; Jackson et al., 2010; Krammer et al., 2015; Liotti, 2013; Maercker et al., 2013; Pearlman, 2001; Riggs, 2014).

Research on couple therapy for survivors of trauma has grown over the past 20 years. Much of this has thus far been conducted with veterans and partners of veterans, as well as with first responders (McKee et al., 2023; Rennebohm et al., 2023). Couples where one or both partners reported a history of trauma evidenced higher dropout rates, shorter courses of treatment, higher levels of individual-level symptoms, and lower levels of change in symptoms including trauma symptoms and couple satisfaction. The literature to date has not clarified why current interventions have high and premature dropout rates and low scores of symptom change (McKee et al., 2023; Whittaker et al., 2023).

Couple therapy and trauma

An important rationale for the development of DCTCT was the identification that existing models of treatment for couples dealing with the impacts of trauma did not integrate an explicit understanding of the impacts of trauma on survivor couples in terms of their unique struggles in relationships and, also, their challenges in fully engaging in the couple therapy process (MacIntosh, 2019). In the literature on couple therapy and trauma, changes in the models used and supports needed are under discussion (MacIntosh, 2023; McWey, 2022; Whittaker et al., 2023). In a recent systematic review of 10 years of research into systemic interventions for the treatment of trauma, McWey (2022) identified four studies assessing couple therapy modalities for the treatment of traumatic stress and couple distress and noted that all four studies (two involved CBCT–PTSD and two involved exposure-based paradigms) demonstrated improvements in traumatic stress symptoms, and some demonstrated improvements in couple satisfaction. Monson et al. (2022) note that disorder-specific couple therapy has the most support and the best outcomes, compared to interventions that position the partner or family as assistants or supports to the traumatized individual remaining in treatment. Examples of disorder-specific couple interventions include Cognitive-Behavioral Conjoint Therapy for PTSD, Structured Approach Therapy, Emotionally Focused Couple Therapy for PTSD, and Couple Treatment for Addiction and PTSD (Monson et al., 2022). The two approaches to couple therapy in the context of trauma that have received the most attention are Cognitive-Behavioral Conjoint Therapy for PTSD (CBCT–PTSD) and Emotionally Focused Therapy (EFT).

Cognitive-Behavioral Conjoint Therapy for PTSD (CBCT–PTSD) is a 15-session manualized cognitive behavioral approach to treating couple distress and individual-level PTSD symptoms, simultaneously. Results of small-scale uncontrolled studies and a randomized-controlled trial suggest that CBCT–PTSD has a strong impact on individual-level PTSD symptoms and some impact on couple-level distress (Blount et al., 2017; Fredman et al., 2011; Monson et al., 2008, 2012; Wagner et al., 2016). Applications to couple groups are currently being explored as expansions to this original intervention model (Monson et al., 2020, 2022). Although CBCT–PTSD demonstrates impacts on individual-level PTSD symptoms and some change in couple satisfaction, the treatment is a short-term cognitive behavioral intervention designed, primarily, for survivors of trauma experienced in adulthood. Many participants have experienced military-related traumas, and CBCT–PTSD does not appear to address the symptomatology of childhood trauma survivors evidencing CPTSD.

The research program leading to DCTCT began with an examination of the process and outcomes of EFT with CSA survivors (MacIntosh & Johnson, 2008). This study enrolled ten couples in which one partner had experienced CSA; using the reliable change index, the research team identified that 50% of the participants demonstrated clinically significant improvement in couple satisfaction, and 50% of the CSA survivors demonstrated clinically significant improvement in symptoms of PTSD, over the course of treatment. Notably, couples that engaged in more disclosure and processing of traumatic material in sessions demonstrated more significant changes in couple satisfaction. Also, survivors in three of the ten couples terminated their relationships over the course of the treatment and accounted for three of the five survivors who evidenced clinically significant improvements in PTSD symptoms in the study. This may demonstrate the impacts of a distressed relationship on trauma symptoms or the ongoing benefit of couple therapy on individual-level symptomatology regardless of the couple-level outcome.

Upon later re-analysis of the close to 350 audio tapes of this study, (MacIntosh, 2013a, 2013b), it became more evident that the impacts of CPTSD may have been impacting the response of couples to the therapeutic process of EFT. For survivor couples, most of the therapeutic processes were extremely challenging. The flow of sessions was frequently interrupted by severe emotion dysregulation, making it necessary for therapists to slow or stop the therapeutic process to help survivors regulate their emotions. These moments of therapeutic detour were challenging in two main ways. First, when flooded with overwhelming shame, survivors could become unable to see, hear, or feel, anything that their partner was saying to them about their own needs and feelings. Second, these states of high dysregulation shifted the focus from couple-based intervention to an individually oriented focus around grounding and re-regulation, leaving partners feeling alienated by a disjointed process that was interrupted repeatedly. Based on the process analysis of stages and steps of EFT, these impacts were evident in that the majority of couples did not complete the stage of de-escalation, the first stage of the EFT model. Couples would initially achieve de-escalation and begin to experience lower levels of conflict and distress (Johnson & Greenman, 2006), but would quickly re-escalate over and over. For these couples, it appeared as though history was repeating itself within that negative interaction cycle. The trauma of the past and the distressing repetitive interaction pattern of the present were mingling and expanding into a conviction that feeling anything in the present is impossible (MacIntosh, 2013a, 2013b, 2016).

Following this uncontrolled study, another research team (Dalton et al., 2013) completed a randomized-controlled trial of EFT with CSA survivors and obtained similar quantitative results (MacIntosh & Johnson, 2008). However, these researchers did not perform a process analysis, making it difficult to ascertain whether similar themes, in relation to emotion dysregulation, challenges moving through de-escalation, and negative interaction cycles imbued with past trauma, were present.

To examine the challenges posed by the impacts of complex trauma on couples in therapy, a series of small qualitative studies was a logical next step. The main research question was as follows: what are the key factors that interfere with the process and outcomes of couple therapy with trauma survivors? Based on the initial process exploration of the EFT and CSA study (MacIntosh & Johnson, 2008), the impacts of deficits in emotion regulation on the process of therapy appeared to be profound. These also appeared to be reciprocally influenced by the challenges that survivors experienced in mentalizing: in a survivor holding on to their partner in their mind and being able to engage with and acknowledge their partner's unique self-experience, needs, and distress, without falling into acute states of shame and losing their capacity to empathize and to acknowledge their partner. Impacts of developmental trauma on these self-capacities in survivors appeared to limit their ability to tolerate and fully engage with the couple therapy process (MacIntosh, 2013a, 2013b).

At the heart of healthy relationships is the capacity of both partners to regulate their own emotions and coregulate with one another, producing a window of tolerance for emotional experiences that is wide enough to buffer against the inevitable challenges and conflicts of relational life (Cloitre et al., 2006). For many trauma survivors, difficulties with emotion regulation lead to an inability to manage high levels of distress; they may instead turn to self-injury and dissociation or turn to substances, sex, disordered eating, and other harmful behaviors in an attempt to distract and numb themselves (Bateman et al., 2009).

Another potential mechanism to address the difficulties of survivors in the therapeutic process and during couple distress is mentalizing. Mentalizing is, in essence, thinking about thinking, a developmental self-capacity that includes being able to flexibly and reflectively make sense of the internal mental world of ourselves and others (Fonagy, 2010; Fonagy et al., 1995; Gergely & Unoka, 2008). When one mentalizes well, they are capable of curiosity about their own mind and the minds of others, they are aware of the opaqueness of mental states, they can regulate themselves in relation to the mental and emotional states of others, and they can be tentative and playful in their awareness of the mental states of themselves and others. When one does not have the capacity to mentalize well, they tend to rigidity in their approach to mental states, they can make fast and inaccurate assumptions about themselves and others, and they find it hard to engage with others playfully and in coregulation (Luyten et al., 2012). Childhood trauma has been associated with impairments in mentalizing capacities (Allen & Fonagy, 2002; Fonagy & Target, 2006; Stein & Allen, 2007). Also, like strong emotion regulation capacities, the ability to mentalize well is foundational in the development and maintenance of strong interpersonal relationships (Fletcher & MacIntosh, 2018; Gergely & Unoka, 2008; MacIntosh, 2013a; Stein & Allen, 2007). Mentalizing is vital to the functioning of couple relationships and in engaging and benefiting from therapy.

Analysis of sessions revealed that key EFT process-focused interventions frequently led to overwhelming difficulties with emotional regulation in survivors, which then fueled shutdowns in mentalizing. The focus of the therapy turned to helping the trauma survivor regain the emotional regulation needed to be able to re-enter the process of couple therapy. As a result, partners were left in the wake of their partner's dysregulation, losing the momentum of the treatment (MacIntosh, 2013a, 2013b; MacIntosh & Johnson, 2008; Maheux et al., 2016). These moments of emotion dysregulation and stalled EFT processing heightened this researcher's awareness that survivors, in those moments, appeared to lose their capacity to hold their partner in mind, to hold their own mind in mind, and to be able to see the perspective of any other person, including the therapist and the partner. Through these analyses, this researcher began to question the potential role of mentalizing as a key mechanism through which emotion dysregulation leads to relationship distress and to the challenges that many trauma survivors experience in tolerating traditional approaches to couple therapy, including EFT, which does not consider mentalizing as a key construct in couple treatment (MacIntosh, 2013a, 2013b).

Baseline levels of mentalizing were examined in both traumatized and untraumatized partners, and the results suggested that survivors always struggle to mentalize in sessions, whereas partners are only “working” hard to mentalize in sessions where the challenge and intensity were higher. However, without established “cutoffs” to measure mentalizing against, clinical criteria for strong and weak mentalizing cannot be used to rate sessions. Information from this analysis could be used to compare survivors to partners at the couple and larger participant pool levels, but without a sense of clinical utility (MacIntosh & Esposito, 2017).

To dive deeper into the relationships between mentalizing, trauma survivors and their partners, and the couple therapy process, a qualitative coding method was deployed to explore the relationships between therapist interventions and mentalizing, and mentalizing and therapist responses. Couples in the 2008 study showed a decrease in unreflective mentalizing and an increase in mental exploration over the course of treatment, as well as an increase in rigid and dysregulated mentalizing across the participants and couples. Further, couples that did not increase in ratings of couple satisfaction evidenced elevations in unreflective and rigid mentalizing over the course of the therapy. Meanwhile, couples that did increase in ratings of couple satisfaction showed increased mental exploration and awareness of mental states and dysregulated mentalizing. Couples that did not improve in relation to traumatic stress symptoms had higher levels of dysregulated and externally focused mentalizing characteristics, while couples that did improve trauma symptoms had higher levels of mental exploration and awareness of mental states. In essence, couples who improved in relation to traumatic stress and couple distress also demonstrated improvements in mentalizing, whereas those who did not either deteriorated in mentalizing or did not improve (MacIntosh et al., 2019).

These process analyses revealed the implicit assumption, embedded within any model of therapy, that patients are able to engage with and tolerate the interventions and tasks that it involves. In this instance, survivors of trauma who evidenced CPTSD symptomatology were trying to navigate the challenges of developing and maintaining couple relationships and tolerating and benefiting from couple therapy. Extant approaches to couple therapy for trauma survivors assume that certain self-capacities such as emotion regulation and mentalizing are on board, allowing the couple to tolerate, participate in the tasks, and benefit from the therapy.

In essence, the therapies require that survivors can “do” the tasks of the therapy. For many survivors with complex trauma, its developmental impacts interfere with their ability to do the tasks of therapy, thus limiting their capacity for engaging, tolerating, and benefiting from the therapeutic interventions. This is the foundational rationale for the development of DCTCT. To be effective, treatments for childhood trauma survivors and their partners dealing with CPTSD must integrate an understanding of the devastating impacts of trauma on the capacity of survivors to build strong and supportive relationships and to tolerate and participate in couple therapy. An increasing understanding of interpersonal sequelae of childhood trauma is required, along with a deepening awareness of the moderating role of supportive relationships in the impacts of trauma in survivors. Practitioners have also come to understand the importance of integrating partners into the therapy process for survivors, both from the perspective of healing the individual-level trauma symptoms and in relation to couple-level distress. Couple distress itself may worsen trauma symptoms and, in fact, may be negatively impacted by focusing solely on individual treatments.

For couples dealing with the long-term impacts of trauma, couple therapy is not just an adjunct to individual treatment (MacIntosh & Ménard, 2021b; Monson et al., 2009; Taft et al., 2011). Adding a partner to the process of treatment has been found to be beneficial in several forms of mental health distress previously believed to be “individual” in nature including depression, PTSD, infertility, and substance misuse, among others (Fletcher & MacIntosh, 2018). The DCTCT model integrates an understanding of the specific challenges faced by therapists and couples in couple therapy with trauma survivors, the importance of the development of self-capacities such as mentalizing and emotion regulation, and the opportunity to work through trauma in a couple context to capitalize on the presence of the primary attachment figure. Intervening at the level of developmental self-capacities in the first stage of treatment is expected to improve the process and outcomes of couple therapy at both the individual and couple levels. Our preliminary research has culminated in the development of a model of couple therapy for complex trauma that integrates these findings hypothesizing that the primary mechanisms of action will be to improve the capacities of the couples for emotion regulation and mentalizing and to build more secure attachment in the dyad (MacIntosh, 2019).

The model of DCTCT (MacIntosh, 2019, 2023) developed and piloted in this study is a manualized, evidence-based, 40-session couple treatment based on this hypothesis that developing stronger capacities for emotion regulation and mentalizing and building secure attachment within the couple will lead to improvements in individual-level traumatic stress symptomatology and couple-level distress. The first of four stages consists of psychoeducation about concepts such as attachment, emotion regulation, and mentalizing to help couples learn about these developmental self-capacities. Many trauma survivors and their partners have let us know that unstructured approaches that move quickly into emotional processing and reflective perspective taking have resulted in overwhelming and frightening emotional states. Thus, the first stage of DCTCT is explicitly cognitive and educational, focused on trauma and relationships writ large; it discourages disclosures of traumatic material until the couple is better prepared. This structured, cognitive focus does not avoid the realities of the impacts of trauma but, rather, scaffolds the couple into the therapy through containment with cognitive content to assist the couple in building a shared foundation for the work that will follow.

The second stage of DCTCT focuses on building skills in emotion regulation and mentalizing capacities. This is accomplished through process-level experiences and explicit exercises, activities, and experiential scaffolding of new capacities by the therapist. In this stage, therapists model and facilitate emotion regulation as the couple moves into a deeper understanding of their individual and relational challenges in the context of their traumatic histories. These sessions can help them imagine that they may begin to build the capacity to navigate the daily struggles and conflicts of family life. Additionally, therapists adopt an explicitly mentalizing stance with couples, as a form of modeling. Therapists encourage mentalizing using questions and prompts that will facilitate the thinking about mental states necessary in the development of mentalizing capacities. For instance, questions beginning with—can you imagine what they might be thinking or feeling? or, I'm finding myself making the assumption that you are feeling X, can you help me understand what you are feeling right now?

In the first two stages of therapy, these concepts and self-capacities are introduced and elaborated through a conscious, intentional, and structured process. This can be seen as preparation for the challenging work of the third stage as the couple moves into working through trauma and its relational impact. The third stage of DCTCT involves working at the edge of where trauma and couple life intersect. This can include helping couples disclose and process traumatic experiences together in a dyadic context, which can build empathy and mentalizing capacities, exploring, and working on challenges with sexuality, an area of couple life that can be particularly challenging for trauma survivors (MacIntosh & Vaillancourt-Morel, 2020), and, ultimately, working toward developing earned attachment security. Then, the fourth and final stage of the DCTCT model focuses on consolidation of skills and capacities, and new attachment priming relational behaviors as the couple shifts away from processing traumatic memories and into making these new ways of being a part of daily life.

A fundamental focus of the DCTCT program of research is on ensuring access to trauma-informed treatment for couple distress and complex trauma symptoms to couples from equity-seeking groups including lesbian, gay, bisexual, transgender, intersex, queer or questioning, asexual (LGBTQA+), Black, Indigenous, and people of color (BIPOC), refugees, and forced migrants, as well as those dealing with significant socioeconomic stressors. Its time-limited nature provides accessibility and opportunity to trauma survivor couples who would otherwise not have access to trauma-informed couple interventions, also facilitating the training of new practitioners. Therapists are trained in utilizing gender-neutral language, experiences of oppression and marginalization are framed as present-tense traumas and contextualized within the larger picture of the trauma that the couple is dealing with, and all of the materials used for handouts and exercises have been vetted for cultural sensitivity by members of the marginalized communities. Therapists working on the team represent the communities that we serve, including the principal investigator (PI), who is a queer, gender nonconforming psychologist. Most recently, our team has piloted a virtual couple group format of DCTCT for LGBTQ+, BIPOC, Indigenous, and other historically marginalized groups and feedback was sought from all participants to allow for the adaptation of materials, resources, and the manual for therapists and the couple workbook to be culturally sensitive and representative of the communities to which we seek to provide care (MacIntosh & McNamee, submitted). For trauma survivor couples, barriers to accessing appropriate treatment include the lack of effective treatment options to address their needs and a shortage of therapists trained in trauma-informed approaches to couple therapy. For couples in distress, financial constraints, a scarcity of well-trained clinicians, cohesiveness with values and beliefs, and issues related to identity—gender identity, sexuality, culture, race, ethnicity, and others—all conspire to limit access to couple interventions that may address the specific needs of trauma survivor couples from equity-seeking groups.

DCTCT is still in its infancy and this implementation pilot study is a first step in establishing its role as a key contributor to the field of couple interventions for trauma survivors.

METHOD

Research design and objectives

The pilot study explored the application of the treatment manual with a group of seven couples, sequentially enrolled to examine implementation of the interventions across stages and couples (MacIntosh, 2019). The objectives of this pilot study were to assess the implementation of the DCTCT treatment manual (MacIntosh, 2019) and to assess outcomes. The study was reviewed and approved by the McGill University Research Ethics Board II.

Context

Research took place at the McGill University Couple and Family Therapy Clinic, a training and research clinic integrated in the master's in science applied (MScA) in Couple and Family Therapy at McGill University in Montreal, Quebec, Canada. Couples were enrolled sequentially and provided with DCTCT by the PI, with start of therapy staggered between couples over a period of 3 years. Sessions were observed by MScA interns who had been trained by this author in DCTCT through both coursework and internship supervision, who provided feedback and commentary on the psychotherapy process and implementation fidelity. The pilot study began in person the year before the Covid-19 pandemic and continued online throughout the pandemic, with a brief interruption at the outset of lockdown measures as university activities were moved online.

Recruitment and participants

Participants were recruited for the study via local public and private mental health facilities and practitioners, with a particular focus on recruiting participants from equity-seeking and traditionally under-served groups. To meet inclusion criteria, couples were required to be living together for a minimum of 1 year, one partner had to meet the diagnostic criteria for CPTSD according to the International Trauma Questionnaire (ITQ), and couples had to score in the distressed range on the Dyadic Adjustment Scale (DAS). Additionally, couples had to agree for sessions to be videotaped for research and training purposes. A total of 13 couples met the inclusion criteria. Two couples were excluded before beginning treatment due to a conflict of interest with the research team in one case, and uncontrolled psychotic symptoms in one partner in a second case. A third couple was referred out for specialized treatment after three sessions due to severe IPV requiring stabilization and de-escalation that exceeded the limits of the pilot study. Three couples who were enrolled and began treatment dropped out due to uncontrollable external circumstances (an immigration crisis in one couple and two couples lost to follow-up during the service interruption caused by the pandemic). Seven couples completed treatment and all outcome measures (see Table 1 for demographic characteristics).

Measures and data collection

Psychometric measures were chosen for their utility, reliability, and validity in measuring the constructs of interest. Data were collected by the PI at baseline and upon completion of the treatment.

Baseline trauma exposure

The Life Events Checklist (LEC-17; Gray et al., 2004) is a measure of lifetime trauma exposure including childhood interpersonal traumas, natural disasters, exposure to violence and abuse in adulthood, and other life-threatening events. This measure was administered once at baseline to provide an estimate of the frequency and severity of traumatic life events.

PTSD and complex trauma symptoms

PTSD and CPTSD symptoms were measured using the ITQ-11 (Cloitre et al., 2021). The ITQ is a brief, straightforward measure of core symptoms of PTSD and CPTSD based on the ICD-11 that was designed for research and clinical progress tracking. The validity of this measure has been confirmed through confirmatory factor analyses and it has demonstrated strong psychometric properties, with Chronbach's α of 0.93. The ITQ-11 does not have a clinical cutoff score, with higher scores reflecting higher levels of distress. It has categorical scoring methods for the two diagnostic categories of PTSD and CPTSD.

Couple satisfaction

Relationship satisfaction was measured using the DAS (Spanier, 1976). The DAS is a widely used self-report index of global couple satisfaction. Studies have confirmed the validity of the DAS as a measure of couple satisfaction and it is widely used, allowing for comparisons with other research (MacIntosh & Butters, 2014). A reliability score of 0.96 (Cronbach's α) has been reported by the author (Spanier, 1976). A Likert-type scale of 5–6 points is used to assess agreements and frequency of events. Higher scores reflect high satisfaction. The DAS has a cutoff score of 98 for a clinically distressed couple and 114 for a couple that is clinically satisfied.

Emotion regulation

Emotion regulation capacities were measured using the Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004), a 36-item scale assessing emotion regulation and dysregulation. It is a frequently used measure of emotion regulation representing six distinct factors demonstrating average to strong reliability (0.93 internal consistency, 0.80 Cronbach's α for subscales). Higher scores on the DERS reflect higher levels of dysregulation, with a score of 70 or higher being reflective of clinically problematic dysregulation.

Attachment

The Experiences in Close Relationships Revised (ECR-12-R; Brennan et al., 1998) Scale was used to assess changes in attachment representations over the course of treatment. It is a 12-item measure of attachment in romantic relationships derived from the longer 36-item measure (Parker et al., 2011). The measure has two subscales. The first, attachment avoidance, represents fear of connection and closeness, a need for distance and self-reliance, and difficulties with sharing. The second, attachment anxiety, represents fear of rejection and abandonment and acute distress on separation. Psychometric properties have been found to be strong (coefficient α of 0.91 and 0.94 for the anxiety and avoidance subscales, respectively). The ECR-12-R does not have clinical cutoffs as it is designed to measure levels of the two subscales, with higher scores reflecting higher levels of avoidance and anxiety.

Mentalizing

Mentalizing capacities were measured using the Reflective Functioning Questionnaire (RFQ, Fonagy et al., 2016), which is a 54-item self-report measure of mentalizing capacity developed by the core researchers in the field of mentalizing. Psychometric analyses demonstrate strong reliability and validity (Luyten et al., 2019). The Cronbach's α of the RFQ-c (concreteness) and RFQ-u (uncertainty) subscales were 0.69 and 0.7, respectively. This study focused on changes in poor mentalizing such as rigidity and concreteness in thinking about mental states.

RESULTS

Over the course of 3 years, including Covid-19-related interruptions, seven couples completed an average of 30 sessions of DCTCT (range 20–40) in a staggered-entry format. Each couple was offered treatment over the course of one academic year, with the PI, MacIntosh. Over the course of the 3 years, couples began therapy in sequence to allow for ongoing implementation and adaptation checks. Results of the LEC indicated that this was a group of participants that had experienced a high level of multiple traumas, with an average of 5.2 reported lifetime criterion A traumas (range 2–14), and there were no couples where any partner did not report a history of at least two traumatic events.

A paired-samples t-test was performed to assess change in trauma symptoms over the course of DCTCT as measured by the full-scale ITQ-11. The results indicated a statistically significant reduction in trauma symptoms from pre- (M = 56.50, SD = 27.78) to post- (M = 29.35, SD = 25.03) DCTCT treatment (t [13] = 3.242, p > 0.003). The effect size for the difference between pre- and posttrauma symptoms was calculated using Cohen's d, resulting in a value of 0.87, which is considered a strong effect. Additionally, from a diagnostic perspective, six of seven participants who had been diagnosed with CPTSD upon intake no longer met criteria for a diagnosis of CPTSD or PTSD at the end of treatment and one participant who had been diagnosed with PTSD upon intake no longer met the criteria for PTSD at the end of treatment (see Table 2 for pre- and postmeasures).

A paired-samples t test was performed to assess change in emotion regulation capacities from pre- to post-DCTCT treatment as measured by the DERS. Results indicated statistically significant reductions in difficulties with emotion regulation from pre (M = 100.14, SD = 24.93) to post (M = 88.57, SD = 22.06) treatment (t [13] = 2.103, p > 0.024). The effect size for the difference between pre- and posttrauma symptoms was calculated using Cohen's d, resulting in a value of 0.57, which is considered a medium to strong effect.

Paired-samples t tests were performed to assess change in attachment-related anxiety and avoidance from pre- to post-DCTCT treatment as measured by the ECR. Results indicated statistically significant reductions in attachment-related anxiety from pre (M = 3.39, SD = 2.46) to post (M = 2.72, SD = 2.08) treatment (t [13] = 2.10, p > 0.024). The effect size for the difference between pre- and posttrauma symptoms was calculated using Cohen's d, resulting in a value of 0.53, which is considered a medium strength effect. Results indicated no statistically significant changes in attachment avoidance from pre (M = 2.32, SD = 1.14) to post (M = 2.44, SD = 1.03) DCTCT treatment (t [13] = .71, p > 0.24).

Paired-samples t tests were performed to assess change in couple satisfaction from pre- to post-DCTCT as measured by the DAS. Results indicated that there were no statistically significant changes in couple satisfaction as measured by the DAS from pre (M = 98.42, SD = 15.34) to post (M = 100.14, SD = 26.14) DCTCT treatment, (t [13] = −0.38, p > 0.36). In this study, one of the seven couples made the decision to separate over the course of the treatment. Given that previous studies have identified outcomes on the DAS as bifurcated, we calculated the Reliable Change Index (RCI) on individual DAS scores to identify clinically significant change across treatment, by couple. The RCI for the DAS has been defined as a change of 1 SD (SD = 15) from the normative sample, in either direction, or moving in or out of the clinical range (Jacobson & Truax, 1991; Spanier, 1976). Using the RCI, four couples demonstrated clinically significant positive change, moving in the direction of satisfaction more than 1 SD or moving out of the clinical range. Three couples, save one partner, including the couple that made the decision to separate, demonstrated clinically significant change in the negative direction, moving from an already distressed position (>98 full scale) on the DAS to more than 1 SD lower (see Table 3).

Paired-samples t tests were performed to assess change in mentalizing from pre- to post-DCTCT treatment as measured by the RFQ. Results indicated no statistically significant changes in mentalizing as measured by the RFQ from pre (M = 31.35, SD = 20.46) to post (M = 26.62, SD = 18.62) DCTCT treatment, (t [13] = 1.18, p > 0.12) (Table 4).

DISCUSSION

In this pilot study of DCTCT, seven couples dealing with the impacts of childhood trauma and CPTSD showed statistically significant improvements in overall trauma symptoms, emotion regulation capacities, and reductions in attachment-related anxiety. These findings concur with the hypotheses that DCTCT will intervene at the level of individual trauma symptoms and emotion regulation capacities. These findings partially concur with the hypothesis that DCTCT will assist couples in building more secure attachment.

It is clinically significant that the diagnostic status of CPTSD changed from seven participants who met criteria at intake compared to only one participant who met criteria for a diagnosis at the termination of therapy. Changes in couple satisfaction, on the DAS, were not statistically significant based on paired-samples t tests. Four couples demonstrated clinically significant positive change on the RCI, whereas three others, save one partner, demonstrated clinically significant negative change. One of the couples separated over the course of treatment. The four couples that completed the most sessions in the treatment were those that demonstrated clinically significant improvements in couple satisfaction using the RCI, whereas the one that separated and those that completed the fewest sessions demonstrated clinically significant deteriorations in couple satisfaction. However, regardless of the direction of change on the DAS, improvements in relation to trauma symptoms were maintained.

The original EFT and CSA study had a similar bifurcation of the sample in relation to couple satisfaction, with three couples ending their relationships over the course of treatment (MacIntosh & Johnson, 2008). Interestingly, a similar-sized pilot study of CBCT for PTSD by Monson et al. (2004) found a similar pattern. This finding, across more than one study addressing traumatic stress in couples, begs the following questions: are these treatments that address traumatic distress at the individual and couple level having a more significant impact on the individual level and, therefore, need to be reconceptualized as individual treatments or revised to address concurrent couple distress more effectively? And, are the couples that are recruited for studies exploring clinical couple work with trauma survivors more likely to be those that will have poorer outcomes in relation to couple satisfaction than traumatic stress symptoms? Another related question is whether couples dealing with significant complex trauma may be more likely to separate in response to treatment of their traumatic stress symptoms, as a positive outcome of treatment? We have, in other discussions, visited the question of whether positive changes in couple satisfaction are, by definition, the determination of therapeutic benefit (MacIntosh & Butters, 2014). Certainly, in the case of this study, the one couple that separated was showing a high level of conflict that was experienced by one partner as extremely emotionally abusive, conflict that was not disclosed on intake or in early sessions. For this couple, the separation was described as a positive outcome and one that was facilitated by the process of therapy bringing these patterns to light, and helping the partner who was identifying as a victim of abuse to gain clarity about her needs and the impacts of these experiences.

Those couples who attended more sessions did demonstrate more gains in couple satisfaction over the course of treatment. Given that stages one and two focus on psychoeducation and skills building and dyadic processing work does not begin until stage three, it makes intuitive sense that the most significant outcome findings would be those related to emotion regulation and trauma symptoms. One potential explanation for this is that, whereas couples were offered a 40-session treatment over the course of one academic year, with the understanding that this would be sufficient to complete all stages of DCTCT, this expectation was not typically realized. Due to high levels of chaos and crisis in many of these couples’ lives, as well as interruptions related to the Covid-19 pandemic, the average number of sessions completed by couples was 30, with a range from 20 to 40. In essence, we must ask the following question: do we need to continue to revise the DCTCT protocol to be more responsive to the specific needs of our complexly traumatized participants to make attending the full 40 sessions possible? This would require a change to our internship structure and a change of therapists mid-stream for couples who are not able to complete the 40 sessions in one academic year due to some of the reasons, related to complex trauma and marginalized identity status, that brought them to the treatment in the first place.

Another ongoing issue in couple relationship research is the validity and reliability of measures of couple satisfaction. Although the DAS is the most frequently utilized measure in the field, it is very outdated in terms of couple norms and values (1976), was designed around a norm primarily of White heterosexual middle-class couples, is long at 36 items, and has not held up in terms of reliability and validity over time, demonstrating high levels of measurement error and reduced power to detect differences between groups due to high levels of error variance (Funk & Rogge, 2007). However, researchers continue to utilize this measure as it is the most frequently used and so allows for comparison among research findings (MacIntosh & Butters, 2014). An alternative measure has been identified by Monson et al. (2020) that is shorter, normed on a very large sample, appears to have greater precision and convergent validity to other measures, and strong construct validity. The team will be making this change in protocols within the McGill Couple and Family Therapy Clinic to ensure that future research has a stronger measure of couple satisfaction.

As DCTCT is a manualized treatment, the stages completed by each couple can be identified. Based on the therapeutic process outlined in the treatment manual, most couples completed Stages 1 and 2. Three of the seven couples completed Stage 3 and only one completed Stage 4. Given that psychoeducation and skills building in emotion regulation and mentalizing capacities form the focus of the first two stages of treatment, no significant change in couple satisfaction would be expected in couples that only completed the first two stages of treatment without completing Stage 3. It makes sense that the most significant impacts of the completion of Stages 1 and 2 would be improvements in trauma symptoms and emotion regulation capacities.

The one hypothesis that was not founded was that DCTCT would help couples develop improvements in mentalizing capacities and that these would contribute to improvements in couple satisfaction. Intuitively, changes in mentalizing capacities would be most significant after the completion of Stage 2 and the explicit focus on skills building, which the majority of couples did complete. However, the RFQ data did not bear this out. One interpretation would suggest that mentalizing is not the mechanism of action it was hypothesized to be and that it will be important to consider alternatives to our understanding of one of the three key hypothesized mechanisms of action of DCTCT. However, it is possible that there are two other potential ways to interpret this; the first is issue related to the use of a self-report survey measurement of mentalizing and is represented by the following question (Müller et al., 2022; Talia et al., 2019): if someone is a poor mentalizer, can they be expected to accurately rate their own mentalizing? The challenge is to measure a construct that must be reflected upon to be able to offer self-assessment. The RFQ has been critiqued for these reasons, and experts concur that mentalizing is a very challenging construct to self-rate, given the difficulty of reflecting about reflection. Unfortunately, there does not yet exist an observational, qualitative measure of this construct. I am working with international colleagues to address this concern by adapting a qualitative observational measure of mentalizing for the couple context, which will gauge change over time through the assessment of transcripts of DCTCT sessions based on an observational method developed by Talia et al. (2015). This will allow for the evaluation of mentalizing based on observable signs of reflection, using an established and validated approach to scoring each session. This pilot study will be re-analyzed and future studies will utilize this method as couples move through each stage of DCTCT to assess observational change over time and the therapeutic process.

The second issue is the unclear time frame for changing and developing mentalizing capacities in adults. A growing literature identifies mentalizing as a core feature of psychotherapy with traumatized clients and in the context of Borderline Personality Disorder, and yet, this research assesses mentalizing as an outcome variable while no available clinical or empirical research sets clear and measurable time frames for change and growth through therapy (Bateman et al., 2009; Fischer-Kern et al., 2013; Nijssens et al., 2012; Toth et al., 2008). The one mentalizing-based approach to couple treatment does not appear to have included any empirical research (Nyberg & Hertzmann, 2014). Future research could begin to address this by tracking mentalizing in all sessions using a validated scoring measure.

Similarly, these results identified statistically significant reductions in attachment-related anxiety but not attachment-related avoidance. There certainly may be an issue of bifurcation in these findings, as well, given the small sample size. However, the ECR-R does not have a mechanism for calculating reliable change. Given that the majority of couples only completed Stages 1 and 2, it does make sense that attachment-related anxiety was reduced through psychoeducation about the impacts of trauma on couple relationships and other aspects of interpersonal functioning. As survivors feel more able to stand close to topics that have been avoided and learn to tolerate and engage with subjects and emotions that might have been triggering of anxiety in the past, in the controlled and containing environment of DCTCT, they may shift away from anxious responses in relation to attachment.

Meanwhile, work on skills building and psychoeducation, which would have the largest impact on emotion regulation and trauma symptoms, should be “preserved” in response to treatment regardless of the status of the couple relationships. Significant change was observed in both diagnostic status and reductions in trauma symptoms across couples regardless of the status of the couple at the end of treatment, providing evidence of this. In fact, one couple terminated their relationship. These outcomes impact the overall mean levels of couple satisfaction and may be related to the very open inclusion criteria for this study, which opened the door to couples that would not likely have been included in many controlled studies. The study population included couples with very severe CPTSD and couple distress. In fact, the majority of couple therapy research protocols for couples dealing with trauma exclude couples where both partners are trauma survivors or where both partners meet criteria for a diagnosis of PTSD.

Limitations

A small sample of seven couples or 14 individuals can increase the risk of Type II error. However, given that the effect sizes obtained for the statistically significant changes in trauma symptoms, emotion regulation, and attachment-related anxiety were medium to strong, this concern is mitigated to some degree. Type I errors are possible because of the large influence of any one couple on couple satisfaction ratings. Also, it is very difficult to generalize from such small samples and even more difficult to perform any multivariate analyses of relationships, such as that between pretreatment demographic variables and outcomes of treatment. Additionally, the Covid-19 pandemic led to some breaks in treatment and the loss of couples to these breaks, which had an impact on the completion rate, the ability of our team to offer longer treatment time frames due to the university shutting down all human subjects research-related activities for almost 4 months, bringing the academic year to an end, and resulting in some of our participants leaving and moving. Future studies will need to draw on larger samples to allow for more complex analyses, to level out the potential for bifurcation of results due to couples who improve and couples who do not, being quite different, and so research can begin to identify variables associated with these outcomes. It will be important to have large enough participant groups that we can be able to compare couples that improve in relation to couple satisfaction versus those that do not.

Another limitation of this study is that this author, the developer of DCTCT, was the primary therapist for all couples. In one sense, this represents a benefit to a pilot study, in that all couples were receiving treatment that had high fidelity to the manual. Any success or failure of the treatment could be attributed to the person of the therapist rather than the modality of treatment. Some controls were put in place to protect the validity of the study. All sessions were observed by members of the research team and interns in the MScA in Couple and Family Therapy program at McGill University. Following each session, team members and interns provided feedback and debriefed the session with the therapist. These discussions brought all sessions back to the treatment manual and the manualized interventions and were important fidelity checks. Additionally, an additional check for fidelity included that all sessions were transcribed by research assistants, whose role was to identify themes and the stage of treatment of that session.

On the one hand, commitments to access and inclusion for underserved groups including BIPOC, Indigenous, LGBTQA+, forced migrants, those experiencing severe mental health distress, and those dealing with financial distress, represent the core values of the clinic and research team. On the other hand, the level of diversity among study participants may have made it difficult to make generalizations or perform any statistical analyses on data derived from such a small and very diverse sample. Again, a larger sample size would be necessary to allow us to examine relationships between identity-related variables and treatment processes and outcomes.

Although mentalizing is a key hypothesized mechanism of change within the DCTCT intervention model, the method of measurement used in this study was limited by its sole reliance upon self-reflection by participants, who may have limited capacities for self-reflection and therefore for mentalizing. Ongoing work at applying an observational method of measuring mentalizing in individual therapy to the couple therapy context will be important in remediating this measurement problem.

A significant limitation of the study is that most participants did not complete the full treatment protocol within the allotted time frame and, therefore, the majority were not able to complete the full 40 sessions outlined in the treatment protocol. It is not clear whether they would have completed the full protocol in the intended 40 sessions. However, in developing further understanding of some of the challenges of the clinic's diverse participant pool in attending regularly, it will be important for us to put strategies in place to increase both attendance and retention. The pandemic had a significant impact on attendance and retention but, also, the couples with whom we were working were highly traumatized couples and many of them were living in precarious situations financially, physically, and vocationally. The majority of missed sessions were related to this precarity—for example, needing to pick up extra shifts, needing to attend medical visits for complex health issues, and inability to obtain resources for travel to the clinic. On the one hand, these are the couples that our clinic and DCTCT has been developed to assist. On the other hand, this makes fidelity to the model and time-limited treatment challenging.

Future directions

Given these limitations, the team has begun planning the development of a more robust evidence base for DCTCT. Although 40 sessions might be sufficient to complete the stages of the model, statistical evidence for this hypothesis is not available, especially because only one couple was able to complete 40 sessions within the 8-month time frame originally proposed. However, providing 40 or more sessions across multiple academic years within a free community Couple and Family Therapy clinic may be challenging. Therefore, it will be important to consider ways of providing DCTCT that will address these concerns, increase the overall participant pool, and maintain capacity to provide long-term, pro bono, treatment to trauma survivor couples dealing with CPTSD.

During the Covid-19 pandemic, the team received a grant to develop and provide a virtual couple group format for Stages 1 and 2 of DCTCT across ten, two hour sessions using a videoconferencing platform (MacIntosh et al., 2023). Findings from the Covid groups were consistent with findings from the pilot in clarifying that 20 hours of therapy was not enough to help couples experience significant change in couple satisfaction. And yet, with the groups, feedback was positive and the majority of couples sought a referral to continue on to  Stage 3 with MScA interns within the clinic. Observationally, these couples appeared to be ready to engage with the tasks of Stage 3, were able to attend regularly, and seemed to be working hard in sessions and outside of sessions to fully benefit from the treatment.

With a focus on accessibility and advancing the evidence base for DCTCT, the next step in this program of research will be to develop and implement a sequential care model. The team is currently preparing for a further uncontrolled study to pilot this sequential care model. Stages  1 and 2 will again be provided in a virtual group format over ten, two hour sessions. Couples will then be assigned to an MScA intern, under supervision with a trained DCTCT supervisor, for a further 20 sessions to complete Stage 3 in person. Following this, couples will return to the virtual group format to complete a further ten, two hour sessions of consolidation.

DCTCT is a novel form of couple therapy designed explicitly to address individual-level CPTSD symptomatology and couple-level distress in trauma survivors and their partners. This pilot study demonstrates that participants in DCTCT show strong improvements in CPTSD diagnostic status, overall trauma symptoms, emotion regulation, and attachment-related anxiety. An approach to observational coding of mentalizing, currently in development, will provide the team with the tools necessary to accurately assess change over time, something that was not possible in this pilot study. Additionally, ensuring that couples can complete the full model through the development of a sequential care approach to the intervention will ensure that couples also experience the benefits of DCTCT on their couple satisfaction and well-being.