A new measure of secondary traumatization: Psychometric properties and external validity of Secondary Traumatization Questionnaire (STQ)

Abstract

Secondary traumatization (ST) is a condition that results from prolonged exposure to traumatized individuals and which mimics symptoms of post-traumatic stress disorder (PTSD). Even though ST is conceptually linked to PTSD none of the existing measures of this construct proved to cover its manifestations in a manner that would correspond to any of the prominent models of PTSD. Thus, the present study aims to develop, psychometrically evaluate, and validate a new self-report measure of ST that would rely on DSM-5 criteria for PTSD. A total of 223 professionals (age range 22–65 years, M=38.99, SD=11.12; 74.9% females) working directly with traumatized individuals, i.e., refugees, migrants, and asylum seekers by providing different services and assistance participated in the study. They completed a newly developed 20-item Secondary Traumatization Questionnaire (STQ) relying on DSM-5 symptoms of PTSD (intrusions – five items, avoidance – two items, negative alterations in cognitions and mood (NACM) – seven items, hyperarousal – six items); a single-item measures of depression and anxiety; 12-item self-perceived quality of life (QoL) (α=.86), and Copenhagen Burnout Inventory (α=.91). STQ showed high indices of item sampling adequacy (KMO=.98), internal consistency (α=.91), and homogeneity (H5=.71). Six concurrent models of the latent composition of STQ were examined – a single factor model, four factor models (DSM-5, Dysphoria), five-factor model (Dysphoric arousal), and six-factor models (Anhedonia, Externalizing behaviors). Except for a single-factor, all other models demonstrated fair fit indices, with the four-factor DSM-5 model [χ2=317.21, p<.001, CFI=.91, TLI=.89, SRMR=.057, RMSEA=.065 (.054-.076)] being empirically the most adequate one. Factors of intrusions, avoidance, NACM, and hyperarousal were significantly related to depression (r=.22–.53, p<.01), anxiety (r=.29–.59, p<.01), QoL (r=-.21–-.39, p<.01), and burnout (r=.33–.56, p<.01). Additionally, STQ demonstrated incremental value in predicting depression [R2=.15, F(4,217)=12.27, p<.001], anxiety [R2=.17, F(4,217)=14.94, p<.001], and QoL [R2=.06, F(4,217)=4.38, p=.002] over and above burnout [R2=.17, F(1,221)=45.59, p<.001; R2=.21, F(1,221)=56.94, p<.001; R2=.15, F(1,221)=40.33, p<.001, respectively]. The psychometric quality of a newly developed STQ, comparability of its latent composition with PTSD symptom clusters, and its usefulness in the prediction of relevant mental health outcomes is discussed.