A systematic literature review of factor analytic and mixture models of ICD-11 PTSD and CPTSD using the International Trauma Questionnaire

Highlights

• Review of latent structure of International Trauma Questionnaire.

• Correlated six-factor and two-factor second-order most commonly supported symptom structures.

• Distinct classes representing PTSD and CPTSD symptom profiles have been reported in clinical and non-clinical samples.

• Validity PTSD and CPTSD supported when tested in conjunction with other variables.

• More research needed on populations with high rates of trauma exposure.

Abstract

The 11th version of the International Classification of Diseases (ICD-11; WHO, 2018) describes two distinct trauma related disorders, Posttraumatic Stress Disorder (PTSD) and Complex PTSD (CPTSD). This review aims to summarise and synthesize evidence from factor analytic and mixture modelling studies that have investigated the latent structure of the International Trauma Questionnaire. A systematic search of PsycInfo, Web of Science, Scopus and Pubmed databases was conducted to identify relevant articles. Thirty-three studies met the inclusion criteria for this systematic review. The latent structure of the ITQ was best represented by two models; a correlated six-factor model (Re-experiencing, Avoidance, Threat, Affect Dysregulation, Negative Self Concept, and Disturbed Relationships) and a two-factor second-order model (PTSD and Disturbances in Self-Organization). Mixture model studies consistently identified distinct classes representing those displaying PTSD and CPTSD symptoms. Numerous studies demonstrated support for the factorial and discriminant validity of PTSD and CPTSD when analysed in conjunction with other variables. Overall, support was found for the conceptual coherence of PTSD and CPTSD as empirically distinguishable disorders, as measured by the ITQ. The available evidence demonstrates that the ITQ is a valid measure of ICD-11 PTSD and CPTSD. Recommendations for future research are included.

Introduction

The 11th version of the International Classification of Diseases (ICD-11; World Health Organisation, 2018) describes two trauma-related disorders: Posttraumatic Stress Disorder (PTSD) and Complex PTSD (CPTSD). Diagnosis of PTSD requires (1) re-experiencing in the here and now (RE), (2) avoidance of traumatic reminders (AV) and (3) sense of current threat (TH). CPTSD includes these core PTSD symptom clusters in addition to the symptom clusters of (1) affective dysregulation (AD), (2) negative self-concept (NSC), and (3) disturbances in relationships (DR), collectively referred to as “Disturbances in Self-Organisation” (DSO; Maercker et al., 2013). Both disorders require traumatic exposure and evidence of functional impairment for diagnosis, and the two conditions are distinguished on the basis of their symptom presentation. Type of traumatic exposure is considered a risk factor rather than a prerequisite for a differential diagnosis (Cloitre, 2020; Hyland, Murphy et al., 2017), however, CPTSD was theorized to occur more commonly following trauma exposure that was prolonged, repeated, interpersonal in nature, and inescapable (Cloitre, Garvert, Brewin, Bryant, & Maercker, 2013).

The International Trauma Questionnaire (ITQ; Cloitre, Roberts, Bisson, & Brewin, 2015) was developed as a self-report measure for the assessment of ICD-11 PTSD and CPTSD diagnoses. The development of the PTSD items was based on the work of Brewin, Lanius, Novac, Schnyder, and Galea (2009) and the selection of DSO items was based on results from DSM-IV field trials which investigated the most frequently reported CPTSD symptoms (Van der Kolk, Roth, Pelcovitz, Sunday, & Spinazzola, 2005) and results from an expert opinion survey where clinicians were asked to identify the most common and impairing CPTSD symptoms (Cloitre et al., 2011). The selection of items for the finalised 12-item ITQ was based on results from Item Response Models which assessed the performance of each of the individual symptom indicators (Cloitre, Shevlin et al., 2018). Research demonstrated support for the convergent and discriminant validity of a preliminary 23-item version of the ITQ (Hyland, Shevlin, Brewin et al., 2017; Karatzias et al., 2016). In keeping with the WHO’s organising principle for the ICD-11 of maximizing clinical utility via a focus on a small number of core symptoms for each disorder (Reed, 2010), a finalised 12-item version of the ITQ was developed whereby each symptom cluster was measured by two items (Cloitre, Shevlin et al., 2018). Additionally, the ITQ screens for a respondent’s index trauma event, how long ago, the event occurred, and evidence of functional impairment associated with the PTSD and DSO symptoms. An adapted version of the ITQ has been developed for use in children and adolescents (ITQ-CA; Cloitre, Bisson et al., 2018), with research demonstrating support for the psychometric properties of this measure (e.g. Bruckmann, Haselgruber, Sölva, & Lueger-Schuster, 2020; Haselgruber, Sölva, & Lueger-Schuster, 2020b; Kazlauskas et al., 2020; Sölva, Haselgruber, & Lueger-Schuster, 2020). The ITQ has been validated and translated for use in twenty-five languages (International Trauma Consortium, n.d.) including Arabic (Vallières et al., 2018), Chinese (Ho et al., 2019) and Lithuanian (Kazlauskas, Gegieckaite, Hyland, Zelviene, & Cloitre, 2018). Given that the ICD-11 is the classification system used worldwide to described mental health disorders, that the ITQ is the only available self-report measure specifically designed to measure these diagnoses, and that the ITQ is frequently used in both clinical services and epidemiological research, summarising existing evidence on the validity of ITQ as a measure of ICD-11 PTSD and CPTSD is an important research endeavour.

Establishing the validity of the ITQ is a critical element in the larger, on-going process of evaluating the validity of the ICD-11’s new descriptions of PTSD and CPTSD. Much of the existing literature has focused on testing the validity of the ITQ as a measure of PTSD and CPTSD by means of two analytical procedures: confirmatory factor analysis (CFA) and latent class/ profile analysis (LCA/LPA). Factor analysis is a statistical technique whereby continuous latent variables (i.e. factors) are used to explain the common content of observed variables (Lubke & Muthén, 2005), and thus tests if responses to the ITQ can be explained by a set of continuous latent variables described in the WHO’s model of PTSD and CPTSD (i.e., PTSD and DSO symptoms). On the other hand, mixture models utilise categorical latent variables to assign individuals into homogeneous groups, or latent classes, based on their responses to observed categorical (LCA) or continuous (LPA) symptom indicators (Nylund, Asparouhov, & Muthén, 2007). Therefore, mixture models are used to test if responses to the ITQ can be explained by a categorical latent variable (i.e., belonging to a PTSD or CPTSD class). These methodological approaches test the factorial and discriminant validity of the ITQ, respectively.

Given that the ITQ was developed with the intention to enhance understanding of the “…nature, predictors, course, treatment and outcomes of PTSD and CPTSD” (Cloitre, Shevlin et al., 2018, p17), it is imperative to synthesise the extant evidence base regarding the validity of this measure. Brewin et al. (2017) provided a comprehensive review of the validity and applicability of ICD-11 PTSD and CPTSD symptom proposals, however, given that research has evolved since then and with the release of the 12-item ITQ in 2018, there is a need for an updated synthesis of research investigating the latent structure of the ITQ. Furthermore, there is a plethora of factor analytic studies investigating the latent structure of PTSD, as per DSM definitions, with a systematic literature review by Armour, Műllerová, and Elhai (2016) highlighting the various factor analytic models identified within the DSM literature. Given that the ICD-11 description of PTSD is markedly narrower to that of DSM-IV and DSM-5 (Maercker et al., 2013) and with the inclusion of the new diagnosis of CPTSD in ICD-11, it is imperative to summarize findings from research investigating the latent structure of ICD-11 PTSD and CPTSD in a similar manner to what has been done for DSM. Therefore, the goal of this systematic review was to collate and synthesise all studies conducted to date on the latent structure of the ITQ using CFA and LCA/ LPA approaches. We aimed to address two questions: (1) what is the optimal factor structure of the ITQ, and (2) how many classes best represent responses to the ITQ? In addition, we intended to investigate variation in these findings in relation to age and other socio-demographic or clinical characteristics. Four electronic databases (PsycINFO, Web of Science, Scopus and Pubmed) were searched using a series of search terms created to reflect the study aims. This study was conducted in adherence with Preferred Reporting items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Liberati, Tetzlaff, Altman, & The PRISMA Group, 2009) and the quality of each individual study was assessed using a novel quality assessment tool created for studies employing factor analytic and mixture modelling methodologies.