Invariance of the Construct of Posttraumatic Stress Disorder: A Systematic Review

Abstract

We conducted a systematic review of studies that have evaluated invariance of the construct of posttraumatic stress disorder (PTSD) to summarize their conclusions related to invariance/noninvariance and sources of noninvariance. In November 2017, we searched Pubmed, PSYCINFO, PILOTS Web of Science, CINAHL, Medline, and Psychological and Behavioral Science Collection for abstracts and articles with these inclusionary criteria: peer-reviewed, including DSM-IV or DSM-5 PTSD invariance as a main study aim, use of multigroup confirmatory factor analyses, and use of an independent PTSD instrument or module. In total, 45 articles out of 1,169 initially identified abstracts met inclusion criteria. Research assistants then followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to complete a secondary search and independently extract data. Results indicated that DSM-IV dysphoric arousal and DSM-5 hybrid model factors demonstrated the most stability; sources of instability were some intrusion (distress to trauma cues), dysphoria/numbing (traumatic amnesia, foreshortened future, emotional numbness, detachment), and arousal (hypervigilance) items. The PTSD Checklist and PTSD Reaction Index were most often used to assess PTSD in studies investigating its invariance; however, these measures demonstrated partial conceptual equivalence of PTSD across subgroups. Instead, clinician-administered measures demonstrated more conceptual equivalence across subgroups. Age, gender, cultural/linguistic factors, and sample diversity had the least moderating effect on PTSD's symptom structure. Our review demonstrates the need to examine invariance of the PTSD construct following recommended guidelines for each empirical and clinical trial study to draw meaningful multigroup comparative conclusions.Clinicians and researchers use diverse instruments to measure the psychological construct of posttraumatic stress disorder (PTSD). Scores obtained from these instruments are compared across subgroups, using statistical tests that depend on-yet assume-empirically established conceptual equivalence of the construct of PTSD. From a latent variable model perspective, participants with the same true PTSD factor (latent) score interpret and/or respond to PTSD-related items in a conceptually similar manner (i.e., same probability of observed scores), and PTSD items similarly capture the latent symptomology of PTSD across subgroups (Bauer, 2017;Dimitrov, 2010;

Posttraumatic Stress Disorder Symptoms and Reckless Behaviors: A Network Analysis Approach

“…Another important note is that the original PRBQ was developed with the idea that a latent construct of reckless behaviors underlies the different types (i.e., levels) as indicated by the different PRBQ items. In line with this notion, there is evidence for a unidimensional construct of reckless behaviors as measured by the PRBQ (Contractor, Weiss, et al., 2019) as well as other measures (Shaw, Wagner, Arnett, & Aber, 1992; Weiss et al., 2018). The results of the current study also provide further evidence of the potential unified nature of the reckless behaviors construct as indicated by the large number of strong intercommunity associations between the PRBQ items.…”

“…The PRBQ (Contractor, Weiss, et al., 2019) is a 16‐item self‐report measure developed to assess PTSD Criterion E2. The current study utilized the first 14 items of the PRBQ, which assess the extent to which an individual has engaged in specific reckless behaviors in the past month, including problematic alcohol use, drug use, gambling, technology use, disordered eating behaviors; impulsive or risky sexual behaviors; illegal behaviors (e.g., arson, burglary, illegal prostitution); reckless spending; physically or verbally aggressive behaviors; property destruction (e.g., deliberately smashing dishes, breaking furniture, or wrecking someone's cellphone, car, home, or other personal belongings); reckless driving (e.g., road rage, excessive speeding); deliberate nonsuicidal self‐harm; and suicidal behaviors.…”

“…The response options range from 0 ( never ) to 4 ( very frequently ), and the 14 items can be summed for a total score. The PRBQ has demonstrated excellent internal consistency and good convergent and construct validity (Contractor, Weiss, et al., 2019). In the current study, the Cronbach's alpha value was .94.…”

“…To date, the most heterogeneous model of PTSD with good empirical support has been the seven‐factor hybrid model (Armour et al., 2015), which consists of intrusions, avoidance, negative affect, anhedonia, externalizing behaviors, anxious arousal, and dysphoric arousal symptom clusters (see Table 1). In several factor analytic studies, the hybrid model has outperformed other models, including the four‐factor DSM‐5 model (e.g., Contractor et al., 2018; Ito, Takebayashi, Suzuki, & Horikoshi, 2019), and several studies have shown the hybrid model's factors to have differential associations with external variables (e.g., Liu, Wang, Cao, Qing, & Armour, 2016; Zelazny & Simms, 2015) and substantial construct equivalence across compared subgroups (Contractor, Caldas, et al., 2019). Utilizing the seven‐factor hybrid model of PTSD rather than the 20 individual PTSD items reduces the number of parameters to be estimated in the network, which may increase the accuracy and stability of the estimated network when sample sizes are small (Epskamp, Borsboom, & Fried, 2018; Epskamp, Kruis, & Marsman, 2017).…”

An exploration of the DSM-5 posttraumatic stress disorder symptom latent variable network

“…The model has demonstrated best model fit across different age groups, cultures and trauma types, as previously reviewed (e.g. Armour et al, 2016; Contractor, Caldas, Dolan, Natesan, & Weiss, 2019 ). Also, the hybrid model demonstrated longitudinal invariance across time (Wang et al, 2017 ).…”