Cross-cultural adaptation of the Brazilian version of the Questionnaire on Eating and Weight Patterns-5 (QEWP-5)

Introduction: The Questionnaire on Eating and Weight Patterns-5 (QEWP-5) is a self-report instrument developed to screen individuals for binge eating disorder (BED), as defined by the DSM-5. However, this version of the instrument had not been adapted for the Brazilian population. Objective: To describe translation and cross-cultural adaptation of the QEWP-5 into Brazilian Portuguese. Methods: Translation and cross-cultural adaptation of the QEWP-5 included the following steps: forward translation, comparison of translations and a synthesis version, blind back-translations, comparison of the back translations with the original version, and a comprehensibility test. The comprehensibility test was conducted with a sample of 10 participants with BED or bulimia nervosa and 10 eating disorders experts. Additionally, a Content Validity Index (CVI-I) was calculated for each item and then averaged to produce an index for the entire scale (CVI-Ave), to assess content equivalence. Results: Some inconsistencies emerged during the process of translation and adaptation. However, the expert committee solved them by consensus. The participants of the comprehensibility test understood the Brazilian version of QEWP-5 well. Only 2 patients (20%) had doubts about items related to subjective binge eating episodes. Content equivalence analysis rated all items relevant, with CVI-I ranging from 0.8 to 1.0 and an overall CVI-Ave of 0.94. In view of the good overall assessment of the pre-final version of the instrument, additional changes were not made to the final version. Conclusion: The Brazilian version of the QEWP-5 was cross-culturally adapted and was well understood by the target population. Further studies are required to assess its psychometric properties.


Introduction
Binge-eating disorder (BED) is an eating disorder recognized in the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders, (DSM-5), 1 and characterized by recurrent episodes of binge eating (eating an unusually large amount of food associated with a sense of loss of control over eating). Additionally, there is marked distress related to these episodes. In BED, binge eating occurs at least once a week over a 3-month period and is not followed by the inappropriate compensatory behaviors seen in bulimia nervosa (BN). 1 Additionally, binge episodes must be associated with at least 3 of the following symptoms: eating more rapidly than normal, eating until feeling uncomfortable, eating large amounts of food when not physically hungry, eating alone because of embarrassment over the amount of food being consumed, or having feelings of disgust, guilt, or depression following these episodes. BED is the most common eating disorder 2 and is associated with physical, psychological, and functional impairment. 3 The definition of a binge eating episode is one of the difficulties involving diagnosis of BED. Binge eating is defined as: 1) eating in a discrete period of time (usually less than 2 hours), a quantity of food definitely larger than most people would eat under similar circumstances; and 2) a sense of lack of control (feeling that one could not stop or control what or how much one is eating). This central component of BED diagnosis, also called objective binge eating (OBE) is difficult to assess because there is no exact definition of what is considered "a quantity of food definitely larger than most people would eat," and also because the sense of lack of control is based only on one's own perception. 1,4 Several instruments have been developed to assess symptoms of eating disorders and to assess binge eating. The most widely used measures include: 1) the Eating Disorders Examination -Questionnaire (EDE-Q), 5 a self-report version of the EDE interview 6 developed to assess the frequency and severity of eating disorder behaviors and psychopathology; 2) the Binge Eating Scale (BES), 7 developed to assess binge eating severity in individuals with obesity (The BES has been adapted for Portuguese 8

and validated in obese
Brazilian women) 9 ; and 3) the Questionnaire on Eating and Weight Patterns-Revised (QEWP-R), 10 which is designed to screen individuals for BED, as diagnosed by the DSM-IV. 11 The QEWP-R has been adapted and validated for the Brazilian population. 12 As a result of changes made in the DSM-5, instruments developed to assess BED according to previous criteria needed to be updated in line with the current diagnostic criteria. The QEWP-R was therefore updated as the QEWP-5, 13 a 26-item questionnaire that includes the following modifications: 1) revision of the frequency of binge eating and compensatory behaviors; 2) revision of the threshold for inappropriate compensatory behaviors -exclusion criteria; 3) removal of some questions that were not related to the diagnostic criteria; 4) incorporation of questions to assess subjective binge eating -SBE (loss of control eating in the absence of consuming a large quantity of food); and 5) revision of the decision rules for diagnosis. 13 However, to date, the QEWP-5 has not been translated into or adapted for Portuguese. Cross-cultural adaptation of the QEWP-5, following international guidelines, is therefore essential to make available a correctly translated instrument for use in Brazilian settings.
The present study aims to describe the process of cross-cultural adaptation of the QEWP-5 for Brazilian Portuguese.

Methods
Permission to cross-culturally adapt the scale for Brazilian Portuguese was requested from and granted by the original authors of the QEWP-5. We began a process of symmetrical translation based on the stages proposed by Sousa & Rojjanasrirat. 14 This methodology involves the following five steps:

Forward translation
Forward translations were conducted by two independent bilingual eating disorder specialists (T1 and T2), whose native language is Brazilian Portuguese. They produced two versions (T1 and T2) of the instrument.

Comparison of the translations and synthesis version
A third eating disorder specialist with experience in translation, adaptation, and validation of scales compared the two different translations (T1 and T2) with the original version of the QEWP-5 and evaluated any semantic inconsistencies (including any linguistic or conceptual issues). After these comparisons, a merged and synthesized version of the two translations was produced (SV).
The three translated versions (T1, T2 and SV) were presented to an eating disorders expert committee (three psychiatrists, one dietitian and one psychologist).
Ambiguities and discrepancies were discussed, and consensus was achieved, with participation of all three translators. This process generated the preliminary version (PV) of the translated instrument.

Blinded back-translation
The PV was back translated into English by two other independent translators whose native language was English, but who had different profiles. The first was experienced in psychiatric terminology and the second translator was more familiar with colloquial phrases and emotional terms in English. They were blinded to the original version of the QEWP-5. This process resulted in two back-translated versions (BTL-1 and BTL-2) of the instrument.

Comparison of the back-translations
The two back-translations were compared with the original instrument. One of the developers of the original version of the QEWP-5 participated in this step, evaluating both BTL-1 and BTL-2. This step generated the pre-final version (PFV) of the QEWP-5 in Brazilian Portuguese.

Comprehensibility
The PFV was tested on 20 participants (10 patients and 10 experts) as proposed by Sousa and Rojjanasrirat. 14 The comprehensibility of the PFV was pilot tested with They were also asked to provide suggestions for items they rated as unclear. Items rated unclear by at least 20% of the participants were revised.
Next, a group of 10 eating disorder experts (who were not on the initial expert committee) were invited to evaluate the comprehensibility and relevance of the items on the scale. First, each expert rated the items as clear or unclear, and provided suggestions to make the language clearer. They then evaluated content equivalence using the following ratings: 1) not relevant; 2) unable to assess relevance; 3) relevant but needs minor alteration; 4) very relevant and succinct. 15 Items rated as unclear by at least 20% of the experts (comprehensibility evaluation) and classified as 1 or 2 on the relevance scale were revised. Finally, a Content Validity Index (CVI) was calculated for each item (CVI-I) and then averaged to produce an index for the entire scale (CVI-Ave). The minimum cutoffs for acceptability were an individual CVI-I of 0.78 or above 15 and a CVI-Ave of 0.90 or above. 16 This study was approved by the ethics committee at Instituto de Psiquiatria, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, Brazil. Written informed consent was obtained from all study participants before any of the study procedures were performed.

Results
Translation, cross-cultural adaptation  With relation to item 9, for example, the translators disagreed on how to translate the expression "during the times" (in Portuguese "nas ocasiões" or "nas vezes"). In the final version, the final consensus was to use "nas vezes." In items 11 and 23, the expression "feeling disgusted with yourself" was initially translated as "sentir repugnância por si mesmo." However, in the SV this expression was changed to "sentir repulsa por si mesmo," because the Portuguese version of DSM-5 uses the word "repulsa" to describe one of the symptoms associated with binge eating.
There was also disagreement on how to translate to the word "upset" in items 13 and 25. The Portuguese word chosen in the final version was "perturbaram," because it was considered that this expression best describes the distress associated with binge eating.
Item 26 about parents' silhouettes was the subject of some debate. The consensus was that the Portuguese translations used in T1 and T2 versions for "If you have no knowledge of your biological father and/or mother, don't circle anything for that" was difficult to understand (in Portuguese, "Se você não conhece seu pai e/ou sua mãe biológicos não circule aquele que não conhece").
In the SV, this sentence was therefore changed for the following Portuguese expression "se você não conhece seu pai e/ou mãe biológicos, não circule nada para esse pai e/ou mãe, isto é, circule apenas para o pai e/ou mãe biológicos que você conhece." However, since the silhouettes were introduced in QEWP-5 for research purposes only and are not a diagnostic item, they can be omitted without prejudice.
An expert committee evaluated and compared the SV with the original version of QEWP-5. This group suggested some changes to address inconsistencies. Table 2 shows a summary of the items modified after the expert panel and

Comprehensibility
The PFV was pilot tested on ten patients (two men and eight women) diagnosed with BED (n = 7) and BN Also, two patients asked if they could describe more than one episode of SBE.
A group of ten experts in eating disorders (five psychiatrists, two nutritionists, two psychologists, and one nurse) was invited to evaluate the instructions,

Discussion
The QEWP-5 13  The process of cross-cultural adaptation of instruments needs to follow rigorous and standardized guidelines to generate a reliable translated instrument. 14 This is an essential procedure that enables comparison of results obtained from samples with different cultural backgrounds. 18 Although there is no consensus on the best methodological approach, international guidelines on this process do agree that symmetrical translation should be conducted, following a "road map" comprising forward translations, back translations, experts' panel, and pre-testing. 14,19,20 Unlike the QEWP-R, 17  week over the last 3 months," 1 rather than the DSM-IV-TR 21 criterion of "at least two binge days a week for 6 months." 11 Another change made in the QEWP-5 was to alter the threshold for inappropriate compensatory behaviors. In the QEWP-R, 17 the threshold for misuse in terms of compensatory behaviors was "taking more than twice the recommended dose of medications to avoid weight gain." In contrast, in the QEWP-5, 13 taking more than the recommended dose of diuretics, obesity drugs, or laxatives is considered misuse.
Another important change in the QEWP-5 was inclusion of questions to assess SBE. SBE describes episodes in which eating is out of control, but the amount of food is not considered unusually large. 6 There is evidence showing that SBE can cause marked distress and impairment to individuals who experience it, similar to OBE. 22 The 11th edition of the International

Classification of Diseases for Mortality and Morbidity
Statistics (ICD-11) therefore included both OBE and SBE in the diagnostic criteria for BED. 23,24 Therefore, the QEWP-5 can also potentially be used to assess BED according to ICD-11 criteria.
It is important to highlight that the main doubts raised in the comprehensibility test were related to items assessing SBE. One possible explanation is that the expression "an amount of food not considered unusually large" is ambiguous. We therefore consider that the problem lies in the definition of SBE itself and not specifically with the question asked in the QEWP-5. Along the same lines, Mitchell et al. 25 have commented that it is difficult to distinguish OBE from SBE in individuals with BED, especially when self-report instruments (like the QEWP-5) are used. The level of agreement between self-report instruments and clinical interviews for assessment of OBE and SBE tends to be low. 25 In a study that compared the EDE interview with the EDE-Q for assessment of the features of eating disorders in patients with BED, Grilo et al. 26 found that SBE frequencies assessed with the EDE and the EDE-Q were not significantly correlated and that the magnitude of the difference between them was large.
These authors concluded that the EDE-Q may therefore underestimate SBE frequency. 26 The present study has some limitations.
First, although the sample size analyzed for the comprehensibility test was that recommended by the guideline followed when conducting the cross-cultural adaptation, 14

Conclusion
The Brazilian Portuguese version of the QEWP-5 was correctly adapted. Items were well understood by the target population. This version is available for