Experimentally reducing event centrality using a modified expressive writing intervention

Highlights

• We attempted to experimentally reduce event centrality using ACT.

• We created a modified expressive writing intervention.

• Those that followed the intervention correctly evidenced decreased event centrality.

• The results demonstrate an effective method for coping with stressful experiences.

Abstract

Event centrality, the extent to which one perceives a stressful or traumatic event as central to one's identity, has been shown to be one of the predictors of PTSD symptoms. Boals and Murrell (in press) found that an Acceptance and Commitment Therapy (ACT)-based, therapist-led treatment resulted in significant decreases in event centrality, which in turn led to decreases in PTSD symptoms. In the current study, a version of this treatment was administered using a modified expressive writing intervention. Participants were randomly assigned to learn core components about either ACT, cognitive-behavioral therapy (CBT), or baseball (control) via audio analogs. The ACT and CBT groups then attempted to apply what they learned in two subsequent expressive writing sessions, while the baseball group wrote about a neutral topic. The results revealed that participants in the ACT and CBT conditions evidenced significant decreases in event centrality, if they followed the writing instructions correctly, in comparison to the control group. However, there were no group differences in changes in PTSD symptoms. These results suggest that principles of ACT and traditional CBT expressed using a modified expressive writing intervention hold great promise to help individuals recover from stressful experiences.

Introduction

When deciding how best to treat individuals who are distressed after experiencing a stressful or traumatic event, it is helpful to consider the contextual processes that facilitate and maintain that distress. One such process, event centrality, occurs when individuals come to view stressful events as core parts of their identities and sense of selves (Berntsen & Rubin, 2006). As an example, McNally, Lasko, Macklin, and Pitman (1995) noted that when treating Vietnam veterans for posttraumatic stress disorder (PTSD), clients who came to the clinic wearing military regalia had disproportionately higher rates of PTSD and were some of the most challenging clinical cases, in comparison to Vietnam veterans not wearing regalia. The Centrality of Events Scale (CES; Berntsen & Rubin, 2006) was created to assess individual differences in event centrality. The CES includes items such as “I feel that this event has become part of my identity”, “this event permanently changed my life”, and “this event has become a reference point for the way I understand myself and the world” (Berntsen & Rubin, 2006). Because they have become integrated into the life story, events high in event centrality are repeatedly reflected upon and can alter interpretations of past and current experiences. These event memories have also been referred to as “self-defining memories” (Singer, 1995). Such construal of events gives meaning and continuity to one's self and life story (McAdams, Josselson, & Lieblich, 2006) and can influence future behavior and goals (Sutin & Robins, 2008).

Studies have consistently found that event centrality is one of the strongest and most reliable correlates of PTSD symptoms (Berntsen and Rubin, 2007, Rubin et al., 2014, Rubin et al., 2008). Event centrality continues to predict PTSD symptoms, even after controlling for a variety of other predictors, such as depression, anxiety, dissociation, and severity of the event (Berntsen and Rubin, 2007, Schuettler and Boals, 2011). Associations between event centrality and PTSD symptoms have been found in a variety of samples including college students (Boals, 2010, Rubin et al., 2008), older adults (Berntsen and Thomsen, 2005, Boals et al., 2012), Iraq war veterans (Brown, Antonius, Kramer, Root, & Hirst, 2010), those grieving the loss of a loved one (Boelen, 2012), victims of terrorism (Blix, Solberg, & Heir, 2014) and sexual assault victims (Robinaugh & McNally, 2011). In addition, event centrality has been found to predict psychological outcomes using longitudinal and prospective designs (Boals, 2014, Boals & Ruggero, in press, Boelen, 2012).

One limitation of the research exploring the relationship between event centrality and PTSD symptoms is the correlational nature of the studies conducted thus far, limiting our understanding of any potential causal links. To our knowledge, there has only been one attempt to experimentally manipulate event centrality. Boals and Murrell (in press) attempted to reduce event centrality using a modified version of Acceptance and Commitment Therapy (ACT; Hayes, Strosahl, & Wilson, 1999) in a sample of female victims of domestic violence seeking counseling at a community outreach center. ACT is a type of cognitive-behavioral therapy designed to increase individuals' abilities to be psychologically flexible, or to make contact with their thoughts and feelings in the present moment; and, depending on their values, either maintain or change their current behavior. Some components of ACT are particularly relevant to the role of event centrality in PTSD: self-as-context (a constant perspective from which to observe experiences); contact with the present moment (nonjudgmental awareness of experiences as they occur); and defusion (decrease in attachment to experiences). An ideal outcome (i.e., return to baseline or posttraumatic growth) for someone who has a high level of centrality also involves high levels of these three ACT processes. As these components were heavily emphasized in the Boals and Murrell (in press) study, that sample will be used to illustrate the overlap between centrality and ACT. Women who have survived domestic violence often are highly identified with the traumatic events they have experienced. Victimization is woven into their life stories and dealing with abuse is a core piece of their identities. They must frequently evaluate their worldviews about trauma and well-being for the very important purposes of staying alive and keeping their children safe. In ACT treatment, assisting clients to stay present to actual contingencies (as opposed to thoughts and feelings about those events) allows for learning vital information and for a broadening perspective – a chance to learn completely new perspectives, to see themselves and their worlds in all new ways. Perhaps, most important with respect to centrality, clients can accommodate the idea that they experienced the trauma of domestic violence but they experienced many other things too.

Participants received either treatment-as-usual from the community outreach center (control group) or treatment-as-usual plus four 1-h long ACT sessions conducted by a trained therapist (experimental group). The results revealed that participants in the experimental condition evidenced significant decreases in event centrality, whereas the control group evidenced no change in event centrality from pre- to post-intervention. In addition, the experimental group evidenced significant decreases in PTSD symptoms and depression. A mediation analysis revealed the effect of condition on PTSD symptoms was mediated by decreases in event centrality. Thus the results of this one study suggest that four sessions of ACT led by a trained therapist can result in reductions in event centrality for a stressful event.

Although there is an abundance of empirical evidence to support the efficacy of research supported treatments for PTSD, such as prolonged exposure and cognitive processing therapy, these treatments are not widely used in clinical practice (Becker, Zayfert, & Anderson, 2004). This example of the inability of science to influence real-world practice and improve patient care is unfortunate. Cook, Schnurr, & Foa (2004) identified a number of potential barriers to the use of exposure therapy in clinical settings. Some of the potential barriers for patients include a reluctance or inability to engage in trauma memories, belief that treatment will exacerbate symptoms and distress, and a desire to drop out of treatment. In fact, premature termination is a risk factor for any treatment that requires ongoing interaction between trained clinicians and clients. Across varied settings, rates of premature termination range from 19.7 percent (Swift & Greenberg, 2012) to 40 percent (Pekarik & Wierzbicki, 1986). In the aforementioned study using a therapist-led ACT intervention, 25% of participants failed to complete the entire intervention (Boals & Murrell, in press). To address premature termination and related issues we considered modified expressive writing as an exposure analog. Expressive writing about a stressful event may be less threatening to some individuals than seeing a therapist, and thus may overcome some of the aforementioned barriers to therapy.

Expressive writing intervention involves an individual writing about their “deepest thoughts and feelings” about a stressful event, typically two to four writing sessions, each session lasting 20 min. Over 150 studies and three meta-analyses (Frattaroli, 2006, Frisina et al., 2004, Smyth, 1998) have found the expressive writing intervention leads to improved mental and physical health benefits. We hypothesized, given the results of Boals and Murrell (in press) that the benefits of expressive writing exposure could be bolstered through a modification-an additional experimental manipulation that targets event centrality. Thus participants in this study listened to a brief analog of ACT therapy components very similar to the work conducted in the therapy sessions (i.e., self-as-context, contact with the present moment, and defusion). Then they were asked to apply the concepts they heard throughout an expressive writing intervention.

In order to adequately assess effects of the ACT analog over and above exposure from the writing alone, it was necessary to include an active control condition. Given that ACT is a type of cognitive-behavioral therapy (CBT), there are numerous overlapping concepts between ACT and traditional CBT. However, traditional CBT does not have the same emphasis on self-as-context or defusion. From an ACT perspective, changing or reducing negative thoughts and feelings about a stressful event or trauma is not the goal. While a cognitive shift may occur during or post ACT, the change in thinking is more likely about how one considers thoughts and feelings generally, as opposed to changing specific thoughts or feelings. Further, decreases in elicitation and arousal in the presence of stressors will likely occur, but those are not the main desired outcomes from an ACT perspective as they are in some traditional CBT models. Rather, a critical outcome of the ACT model is that acceptance of unchangeable thoughts, feelings, and situations allows an individual to see those experiences as part of their lives, but not as the sum total of their identities. The goal of treatment is to change the power and literality of undesired private events. This outcome is, in summary, reducing the centrality of the negative events. The emphasis on changing or reducing thoughts about negative events in CBT is less focused on decentralizing, and yet is a gold-standard in treating stress and trauma. Given that traditional CBT has both some overlap with, and yet some features unique from, ACT, a CBT analog seemed fitting for comparison.

The purpose of the current study was to test whether the expressive writing intervention, modified by either an ACT analog or a CBT analog, could effectively reduce levels of event centrality. The analogs were 10-minute audio scripts that described typical ACT and CBT therapy work and included a related exercise. We then asked participants to apply the concepts in the audio scripts to help them cope with a stressful event during expressive writing sessions. As a control, we also created an audio script on the history of baseball and then asked participants in the control group to write about a neutral topic. This neutral manipulation and neutral writing combination was used as we needed to see what would happen to centrality levels over time when participants had no experimental interactions with their stressful or traumatic event. This was important for two reasons. First, it is fairly common for journaling and other related writing exercises (e.g., storyboard, values lifeline) to be used in ACT to address self-as-context and defusion (Hayes et al., 1999). Second, given that expressive writing has been shown to lower distress scores in other studies (Frattaroli, 2006), we hypothesized that the exposure provided in both the ACT and CBT groups would lead to some reductions in stress that would not be seen in the baseball group. Since centrality is most relevant in the context of stress and trauma, and the scores on PTSD symptoms and event centrality are positively correlated (Berntsen & Rubin, 2006), this neutral group allowed us to examine the clear and clinically meaningful effects of our experimental manipulation. Given the specific, intentional emphases on self-as-context and defusion in the ACT analog, we further hypothesized that the decreases in event centrality would be larger in the ACT group.