Translation and cross-cultural adaptation of the International Trauma Questionnaire for use in Brazilian Portuguese

ABSTRACT BACKGROUND: The most recent editions of diagnostic manuals have proposed important modifications in posttraumatic stress disorder (PTSD) criteria. The International Trauma Questionnaire (ITQ) is the gold-standard measurement for assessing PTSD and complex PTSD in accordance with the model of the 11th International Classification of Diseases (ICD-11). OBJECTIVE: The aim of this study was to adapt the ITQ for the Brazilian context. DESIGN AND SETTING: The translation and cross-cultural adaptation of the ITQ for use in Brazilian Portuguese was performed in trauma research facilities in Porto Alegre, Rio de Janeiro and Belo Horizonte, Brazil. METHODS: The adaptation followed five steps: (1) translation; (2) committee synthesis; (3) experts’ evaluation through the content validity index (CVI) and assessment of interrater agreement though kappa statistics; (4) comprehension test with clinical and community samples (n = 35); and (5) final back-translation and authors’ evaluation. RESULTS: Two independent translations were conducted. While working on a synthesis of these translations, the committee proposed changes in six items to adapt idiomatic expressions or to achieve a more accurate technical fit. Both the expert judges’ evaluation (CVI > 0.7; k > 0.55) and the pretest in the target population (mean comprehension > 3) indicated that the adapted items were adequate and comprehensible. The final back-translation was approved by the authors of the original instrument. CONCLUSION: ITQ in its Brazilian Portuguese version achieved satisfactory content validity, thus providing a tool for Brazilian research based on PTSD models of the ICD-11.


INTRODUCTION
Posttraumatic stress disorder (PTSD) has high prevalence worldwide, 1 and it is frequently diagnosed by mental health professionals. 2,3 However, controversies surround this diagnosis. The high number of symptoms, among which some are present in other mental disorders (e.g. detachment from others, sleep disturbance, concentration problems and reckless behavior), leads to high rates of comorbidities. [4][5][6] Furthermore, there are studies investigating a certain type of PTSD that is different from what is described in diagnostic manuals. When repeated exposure to trauma is associated with symptoms such as emotional dysregulation, dissociation and negative self-concept, the reaction is often described as "disorder of extreme stress not otherwise specified" or complex PTSD. 7,8 In the literature on trauma, PTSD criteria are often discussed, particularly with regard to which general symptoms of psychological distress should be understood as frequent comorbidities and not as part of the disorder; and which responses are directly related to trauma and therefore should be added to the diagnosis. [8][9][10] The most recent version of the International Classification of Diseases (ICD-11) sought to encompass current scientific knowledge and proposed a new model for PTSD, in which the basic and complex forms of PTSD were distinguished and many symptoms that were considered to relate to general distress were eliminated. PTSD is described as a reaction to trauma that includes (1) re-experience of the traumatic event (i.e. vivid intrusive memories, flashbacks or nightmares accompanied by overwhelming emotions); (2) avoidance of thoughts, memories, situations, people or activities reminiscent of the event; and (3) a state of perceived current threat in the form of hypervigilance or enhanced startle reactions to stimuli such as unexpected noises. Complex PTSD is described as a disorder that typically arises after an extreme or prolonged stressor from which escape is difficult or impossible (e.g. childhood sexual abuse, torture or prolonged domestic violence) and would comprise the sum of PTSD and more persistent symptoms of disturbances in self-organization (DSO), in three clusters: (1) affective dysregulation (e.g. self-destructive behavior, emotional anesthesia or dissociative states); (2) persistent beliefs about oneself as diminished, defeated or worthless, accompanied by feelings of shame, guilt or failure; and (3) persistent difficulties in maintaining relationships and feeling close to others. Both diagnoses require that symptoms cause significant impairment in important areas of functioning, such as social, educational or occupational. 11,12 Another important diagnostic guide, the 5 th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), provides a broader approach towards defining this phenomenon.
Instead of separating posttraumatic reactions into two conditions and reducing the overall number of symptoms (as in ICD-11), this approach does not eliminate any previous criteria and adds symptoms that are often associated with repeated exposure to trauma, to make a unique diagnosis through a new cluster of negative alterations in cognition and mood (e.g. overly negative thoughts and assumptions about oneself or the world, or difficulty in experiencing positive affect) and a new dissociative subtype. 13 Differences in the definition of posttraumatic symptoms may impact the work of clinicians and researchers in an important way. Current diagnostic differences should be investigated through empirical work, with the aim of evaluating the validity of both models in different cultures and populations, in order to achieve overall comprehension of posttraumatic reactions and to define approaches for future editions of diagnostic manuals. 14,15 Thus, it is necessary that instruments representing these new models are available in as many languages as possible. In Brazil, an adaptation of the PTSD model of the DSM-5 has already been produced, 16

OBJECTIVE
The aim of this study was to translate and culturally adapt the ITQ for use in Brazilian Portuguese, which will enable adequate investigation of the PTSD and complex PTSD models of the ICD-11 in Brazil.

METHOD
The ITQ is a gold-standard open-access tool for investigating PTSD and complex PTSD as defined by the ICD-11. In the present study, the ITQ in its final 12-item version, which was recently finalized after extensive empirical work, was used. 17

Committee synthesis -A committee of academics compared
the two translated versions with the original ITQ to certify that all items expressed the same ideas, in order to achieve semantic, idiomatic and conceptual equivalences. The committee was composed of undergraduate, master's and doctoral students. Items were chosen from either of the translated versions and, whenever necessary, were changed and refined by the committee. This step generated a unified version of the ITQ in Brazilian Portuguese.

RESULTS
In making the synthesis from the two translated versions, the committee proposed minor changes regarding the use of idiomatic expressions or with the aim of achieving a more accurate technical fit in four items of the PTSD questionnaire (items 2, 3, 4 and 5) and in two items of the DSO questionnaire (items 1 and 5).
1. In item 2 of the PTSD questionnaire, the term "intensas" (= intense) was chosen over "poderosas" (the literal translation of "powerful", which had been chosen by both translators) to describe trauma memories, because the term "poderosas" has a positive connotation in Brazilian Portuguese.
2. The best translation for the word "reminders", present in items 3 and 4 of the PTSD questionnaire, was extensively discussed.
Both translators suggested terms that could be related to memories ("lembranças" and "lembretes"). This was considered by the committee to be inadequate because of item 4, which refers to external reminders of the trauma. The term "pistas" (= triggers or cues) was chosen.
3. Item 5 of the PTSD questionnaire contained the term "superalert" to describe hyperarousal symptoms. Both translators chose the literal term "super-alertas", but the committee decided that "hiperalertas" (= hyper-alert) would be a better technical fit for this expression in Brazilian Portuguese. 4. In item 1 of the DSO questionnaire, the term "upset" was translated as "triste" (= sad) or "abalado" (= shaken). Neither of these was considered appropriate by the committee, who defined the word "chateado" (another possible translation of "upset") as more adequate. 5. In item 2 of the DSO questionnaire, the expression "cut off " was translated as "afastado" (= away) or "isolado" (= isolated), but was rewritten as "desconectado" (= detached) by the committee.
The unified version created through the committee's synthesis was evaluated by the three judges in different Brazilian districts using the CVI. The results from the expert judges' evaluation showed that all items were considered adequate (> 0.7), as seen in

DISCUSSION
This study reports on the cross-cultural adaptation of ITQ for use in Brazilian Portuguese. The literature on psychometrics indicates that the adaptation process is important because it provides more than just a literal translation. However, there is no technical agreement on how to conduct this process in a reliable and objective manner. 25,26 The steps of the present study were an attempt to cover the methodological guidelines and linguistic specificities regarding both quantitative and qualitative criteria. Items 3 and 4 of the PTSD questionnaire, which contain the terms internal and external reminders of trauma, were extensively discussed and may be considered to be an exam- Most studies on the ITQ have been performed using its original version or without any report of cross-cultural adaptation steps.
Only in the study that reports on data from the Arabic version of the ITQ were interviews about content validity conducted with therapists. 21 Although satisfactory qualitative evidence was found, the authors of the Arabic version suggested that the ITQ would be best administrated with the assistance of a trained professional among illiterate and poorly literate individuals. Our results did not indicate that assistance would be needed when using the ITQ in the Brazilian context, since the 35 people used to test the version did not present difficulties in comprehending any items of either questionnaire (and not even when the scores of the participants with lower education levels were evaluated separately). The aim of the ITQ is that it should be a brief self-report screening instrument for PTSD and complex PTSD, and our results indicate that this goal was achieved in the Brazilian version of the instrument.
Some important limitations of our study need to be addressed.
Although we made an attempt to minimize the influence of regional speech patterns by conducting the study in three different cities, it was not possible to represent the entire Brazilian population with this sample. Also, further validation of the ITQ is needed (especially regarding construct validity) in studies with larger samples, in order to establish whether the items replicate the model of the ICD-11 properly. More robust psychometric studies are currently being conducted to achieve these goals. These findings enable initial research using the ICD-11 model for PTSD and complex PTSD in Brazil. Future studies should focus on advancing knowledge regarding the nature, predictors, course and outcomes of these disorders in the Brazilian population. This approach is likely to contribute to the discussion of these diagnoses in an overall manner.

CONCLUSION
The Brazilian Portuguese version of the ITQ was translated and culturally adapted to its context, and it exhibited satisfactory content validity.