Psychometric properties of the Trauma Relevant Assumptions Scale

Abstract

This article describes the psychometric properties of a novel questionnaire, i.e. the Trauma Relevant Assumptions Scale (TRAS). The added value of the TRAS over previous trauma relevant belief questionnaires is that the TRAS enables measuring valence and rigidity of beliefs simultaneously. Both aspects are thought to be predictive of the development of chronic PTSD symptoms. For the exploratory factor analysis, the TRAS was administered to 309 adult volunteers. Principal components analysis yielded two factors: Assumptions about Self and Assumptions about the World. The two-factor structure was confirmed in a sample of 185 traumatized individuals. The TRAS seems to be a valid and reliable instrument, which is strongly related to post-trauma symptoms and has good discriminative validity. Apart from research settings, the TRAS may also be suitable in therapeutic settings to identify the severity of dysfunctional assumptions, and to assess the progress in change from negative assumptions to more positive assumptions.

Introduction

One of the puzzles surrounding post-traumatic stress disorder (PTSD) is to discover which factors predict the development of chronic PTSD (Ehlers & Steil, 1995), and explaining why these factors are predictive. In addition to variables like dissociation and trauma severity (e.g. McNally, Bryant, & Ehlers, 2003; Ozer, Best, Lipsey, & Weiss, 2003), pre- and post-trauma beliefs are also thought to be predictive of the development of chronic PTSD symptoms (e.g. Calhoun & Resick, 1993; Ehlers & Clark, 2000; Foa & Riggs, 1993; Horowitz & Reidbord, 1992; Janoff-Bulman, 1992). The hypothesis states that dysfunctional beliefs with respect to the self, others or the world may disturb successful post-trauma processing. Moreover, it is stated that not only the degree of negativity/positivity of the held assumptions is crucial for the development of post-traumatic symptoms, but also the degree of flexibility-rigidity. Although several questionnaires are available to assess pre- and post-trauma cognitions (e.g. Foa, Ehlers, Clark, Tolin, & Orsillo, 1999; Janoff-Bulman, 1992), no questionnaire exists that can successfully assess the degree of flexibility-rigidity of beliefs. The aim of the present study was to develop such a questionnaire.

Foa, Zinbarg, and Rothbaum (1992), Foa, Riggs, Dancu, and Rothbaum (1993), Foa et al. (1999) proposed a curvilinear relation between beliefs and PTSD symptoms. PTSD may develop when rigid negative schemas (“The world is never safe,” “I am always incompetent”) are confirmed, or when rigid positive schemas (“The world is always safe,” “I am never incompetent”) are violated. Thus, Foa et al. (1999) suggest that the presence of rigid concepts about self and the world (positive or negative) renders individuals vulnerable to develop PTSD. In contrast, people with more flexible beliefs about safety (“The world is sometimes safe and sometimes dangerous”) will be most likely to recover after a traumatic event. Although it can be argued that people with negative schemas already suffer from some kind of psychopathology, Foa and Riggs (1993) suggest that these negative schemas may have resided in long-term memory with high activation thresholds and therefore may not have resulted in severe psychopathology. Although several studies indeed showed a relation between negative beliefs and post-trauma symptoms (Ali, Dunmore, Clark, & Ehlers, 2002; Dunmore et al., 1999, Dunmore et al., 2001), the beliefs were only assessed after the traumatic experience and the degree of flexibility-rigidity of beliefs was not assessed.

Currently, several questionnaires are available that measure trauma related beliefs. These include the World Assumptions Scale (WAS; Janoff-Bulman, 1992), the Post-traumatic Cognitions Inventory (PTCI; Foa et al., 1999) and the Personal Beliefs and Reactions Scale (PBRS; Resick et al., 1991). These questionnaires have been shown to be reliable and valid in assessing beliefs and change in strength of beliefs, which has appeared to be related to PTSD status (e.g. Foa & Rauch, 2004). However, a disadvantage of these questionnaires is that several beliefs refer to a traumatic experience (all e.g., 10 out of 36 items of the PTCI). These items have to be changed or removed from the questionnaire if the beliefs are assessed prior to the traumatic event. Moreover, the existing belief questionnaires are not able to measure the flexibility-rigidity dimension of the beliefs. Although participants can rate to what degree they hold certain beliefs, this does not necessarily indicate the degree of flexibility-rigidity. For example, rating an item like “People can’t be trusted” (item 7 in the PTCI) on a scale from 1 (totally disagree) to 7 (totally agree), gives no information about the rigidity of this belief, because the wording of the items does not refer to a rigid state. We propose that rigidity can better be measured when the items refer to a rigid statement. This can be accomplished by adding the word “always” or “never” to the item (e.g. “The world is never a good place” versus “The world is always a good place” or “People can never be trusted” versus “People can always be trusted”). Finally, a further disadvantage of current belief questionnaires is that they focussed on negative beliefs, excluding positive beliefs. A person who does not agree with a negative belief does not necessarily hold a positive belief.

The present study describes the psychometric properties of a novel questionnaire enabling to assess beliefs in terms of negativity/positivity and flexibility-rigidity. The content of the items was based on existing belief and assumption questionnaires (e.g. PTCI, WAS). However, the items were reworded, such that the beliefs were split into two rigid and opposing (positive versus negative) assumptions, which were placed on either side of a Visual Analogue Scale (VAS). By this means, the rating scale of the TRAS ranges from dysfunctional to functional to dysfunctional again. That is from a rigid negative belief, to a flexible belief to a rigid positive belief. On the other hand, the rating scales of current questionnaires range from dysfunctional to functional, that is from agreement with a negative belief to disagreement with a negative belief. The rating scale of the TRAS enables researchers to finally test whether pre-trauma beliefs and post-trauma symptomatology are related curvilinearly as suggested by Foa et al., 1992, Foa et al., 1993, Foa et al., 1999.

This article describes two studies. The first study assesses the psychometric properties of the Trauma Relevant Assumptions Scale (TRAS), in adult volunteers. Because many items of the TRAS are based on the PTCI, we hypothesize that the TRAS will consist of two subscales similar to the PTCI: self and world. The Self-scale of the Miskimins Self-Goal Other (MSGO) Discrepancy Scale was included in order to assess the convergent validity of the expected subscale Self. Moreover, it will be tested whether scoring negatively in a rigid way is indeed strongly related to a high level of general psychopathology (see Foa & Riggs, 1993). Finally, it will be tested whether scoring rigid negative is related to neuroticism. This relation may be expected because neuroticism refers to a temperamental sensitivity to negative stimuli, emotional instability and maladjustment (Goldberg, 1992).

The second study investigates whether (1) the psychometric properties as obtained in the first study are confirmed in a traumatized sample, (2) the assumptions of the TRAS are not only related to PTSD, but also to other common post-trauma symptoms (depression, state anxiety and general psychopathology), and (3) whether the TRAS can discriminate between traumatized participants with and without current PTSD.