Psychometric evaluation of the Dutch version of the posttraumatic cognitions inventory (PTCI)

https://doi.org/10.1016/j.brat.2005.07.002Get rights and content

Abstract

Trauma-related cognitions play an increasingly prominent role in research on trauma and in clinical practice. The present study investigated the psychometric characteristics of the Dutch version of the posttraumatic cognitions inventory (PTCI) and evaluated its potential as an outcome measure. Data were collected from a treatment-seeking sample of trauma victims (n=158) and a college sample (n=178). The PTCI's three-factor structure was retained in both samples. The PTCI demonstrated high internal consistency and two-week test-retest reliability. Convergent validity was evidenced by a pattern of correlations with instruments for trauma-related cognitions, posttraumatic stress disorder, and depressive symptoms that largely met a priori expectations. Reductions in self-reported and clinician-assessed posttraumatic stress disorder symptoms were positively associated with decreased PTCI scores, indicating that the PTCI could assist treatment evaluation. It is concluded that the Dutch version of the PTCI exhibits good psychometric characteristics and has the potential to contribute to trauma-related research.

Introduction

Several cognitive formulations have been proposed to explain the development and persistence of trauma-related psychopathology (e.g., Brewin, Dalgleish, & Joseph, 1996; Ehlers & Clark, 2000; Foa & Rothbaum, 1998; Janoff-Bulman, 1992). The central role of trauma-related cognitions in these models is mirrored in clinical practice, where cognitive therapy ranks among the most effective interventions for posttraumatic stress disorder (PTSD; APA, 1994) (e.g., Ehlers et al., 2003; Marks, Lovell, Noshirvani, Livanou, & Trasher, 1998; Tarrier et al., 1999). Adequate assessment of trauma-related cognitions could benefit both domains, i.e., enable the validation and further development of cognitive models of trauma-related psychopathology, and assist the planning and evaluation of cognitive interventions to alleviate such psychopathology.

Acknowledging this theoretical and clinical need, two leading research groups collaborated to develop the posttraumatic cognitions inventory (PTCI; Foa, Ehlers, Clark, Tolin, & Orsillo, 1999). In a treatment-seeking population of PTSD patients that had been the victim of miscellaneous traumatic events, an initial psychometric evaluation of the PTCI showed promising results. The total scale and subscales demonstrated high internal consistency (α=.97, .97, .88 and .86, respectively), as well as high 1-week (P=.74, .75, .89 and .89, respectively) and 3-week (P=.85, .86, .81 and .80, respectively) test-retest reliability. Convergent validity was supported by average to high correlations with relevant subscales of the personal beliefs and reactions scale (PBRS; Resick, Schnicke, & Markway, 1991). Further, the total scale and subscale scores significantly correlated with scores on the posttraumatic stress diagnostic scale (PDS; Foa, 1995) (r=.79, .69, .57 and .78, respectively), even when controlling for depressive symptoms and state anxiety. With regard to discriminative validity, PTSD patients scored significantly higher on the total scale and the subscales than subjects without PTSD, when controlling for between-group differences in demographic characteristics, depressive symptoms and state anxiety. A discriminant function based on the PTCI scores successfully predicted PTSD diagnostic status for 86% of participants, with accurate predictions being made for 78% of the PTSD patients and 93% of the trauma victims without PTSD.

In a recent replication in a sample of motor vehicle accident survivors (Beck et al., 2004), the original three-factor structure of the PTCI was supported after four items of the ‘negative cognitions about self’-subscale were omitted. The three subscales that were thus obtained demonstrated adequate internal consistency, and two of these subscales (measuring ‘negative cognitions about self’ and ‘negative cognitions about the world’, respectively) generally satisfied expectations about their concurrent, discriminant, and discriminative validity. In contrast, the subscale measuring ‘self-blame’ demonstrated negligible concurrent validity with measures for PTSD, state anxiety, and depression, poor discriminant validity with reference to a social desirability measure (apparently attributable to perceptions of responsibility for the accident as a background variable), and did not discriminate between participants with PTSD, subsyndromal PTSD, and no PTSD. As the authors argue, the poor performance of the ‘self-blame’-subscale might result from the lower prevalence of these cognitions following motor vehicle accidents than following sexual assault, which dominated the sample of the original study (Foa et al., 1999). These findings did not essentially violate the PTCI's value in the measurement of trauma-related cognitions however, and warrant further research into its characteristics.

The present study investigated the psychometric properties of the Dutch version of the PTCI. As such, it offers the first published findings obtained with a translation of this instrument. In addition, the PTCI's potential as an outcome measure was evaluated by investigating its sensitivity to self-reported and clinician-assessed reductions in PTSD symptomatology.

Section snippets

Participants

At the University of Amsterdam (UvA) participants (n=157) were referred to the outpatient clinic of the department of clinical psychology by general practitioners, company medical officers, and victim assistance agencies. Also, participants were recruited from hospital emergency departments, or were self-referred following media coverage. At Leiden University (LU) data were collected from women who were living in a protected home following severe sexual or physical domestic violence (n=28). All

Results

Of the 185 participants recruited at the UvA and LU, 107 participants (57.8%) satisfied DSM-IV (APA, 1994) diagnostic criteria for PTSD. Fifty-one participants (27.6%) who did not meet either the avoidance criterion or the hyperarousal criterion of this diagnosis were given a subthreshold diagnosis of PTSD (Blanchard, Hickling, Taylor, Loos, & Gerardi, 1994). These participants are collectively referred to as the treatment-seeking sample (n=158). Twenty-seven participants (14.6%) did not have

Discussion

The present study comprised a psychometric evaluation of the PTCI, and thereby replicates earlier studies that introduced this instrument to the field of trauma research (Foa et al., 1999; Beck et al., 2004). The three-factor structure proposed by Foa et al. replicated remarkably well among two samples of treatment-seeking PTSD patients and students. Also, the PTCI total scale and its three subscales demonstrated high internal consistency in both samples. These findings further support the

Acknowledgements

This study was conducted as part of a project funded by the Netherlands Organization for Health Research and Development (ZON).

References (28)

  • B.S. Dansky et al.

    The trauma constellation identification scale: A measure of the psychological impact of a stressful life event

    Journal of Traumatic Stress

    (1990)
  • A. Ehlers et al.

    A randomized controlled trial of cognitive therapy, a self-help booklet, and repeated assessments as early interventions for posttraumatic stress disorder

    Archives of General Psychiatry

    (2003)
  • M.B. First et al.

    Structured clinical interview for DSM-IV axis I disorders

    (1996)
  • E.B. Foa

    The posttraumatic diagnostic scale (PDS) manual

    (1995)
  • Cited by (55)

    • Past adversity and obsessive-compulsive symptoms: The mediating roles of posttraumatic cognitions and posttraumatic stress symptoms

      2021, Journal of Obsessive-Compulsive and Related Disorders
      Citation Excerpt :

      PTCI items are rated on a scale from 1 (Totally disagree) to 7 (Totally agree), and the score for each scale is derived as the mean of its items. The PTCI has exhibited adequate-to-good reliability and validity (e.g., Beck et al., 2004; Foa et al., 1999; Van Emmerik et al., 2006), although the Self-Blame subscale has exhibited infirmities with respect to concurrent and discriminant validity (e.g., Beck et al., 2004). In this study, the Self-Blame, Negative Cognitions about Self, and Negative Cognitions about the World scales demonstrated adequate to excellent internal reliability, producing Cronbach's alphas of .75, .94, and 0.84, respectively.

    • A psychometric evaluation of the Posttraumatic Cognitions Inventory with Veterans seeking treatment following military trauma exposure

      2018, Journal of Affective Disorders
      Citation Excerpt :

      Their initial investigation found excellent internal consistency reliability, high convergent validity with PTSD symptom severity, and good discriminative validity to distinguish between those with and without PTSD following trauma. Since its introduction, research examining the PTCI has included participants with a wide variety of trauma histories (e.g. Daie-Gabai et al., 2011), including physical and sexual assault (Foa et al., 1999; Van Emmerik et al., 2006) and motor vehicle accidents (e.g. Beck et al., 2004), for which the interpersonal nature of the events may vary. To date, PTCI validation studies have supported three-factor structures and identified acceptable model fit for samples who have experienced assault.

    View all citing articles on Scopus
    View full text