Translation and cross-cultural adaptation of the Thyroidectomy-Related Voice Questionnaire (TVQ) to Brazilian Portuguese Tradução e adaptação transcultural do Thyroidectomy-Related Voice Questionnaire (TVQ)

Leandro Pernambuco 1,2  ABSTRACT Purpose: To translate and cross-culturally adapt the Thyroidectomy-Related Voice Questionnaire (TVQ) to Brazilian Portuguese. Methods: We divided the process of translation and cross-cultural adaptation into the following stages: two independent translations; synthesis of the translations; analysis by an expert committee; pretest; back-translation; final synthesis; and final version. We performed the pretest with 20 patients before or after thyroidectomy. We characterized the sample by means of descriptive analysis, and calculated the agreement between the experts by the Item Content Validity Index (I-CVI) and the Questionnaire Content Validity Index (Q-CVI). Results: Throughout the translation and cross-cultural adaptation, the instrument required operational, semantic, idiomatic, and syntactic/grammatical equivalences, especially after the analyses by the experts and the target population. The I-CVI and Q-CVI were mostly acceptable. The back-translation was equivalent to the original version. The comparison between the original, translated, and back-translated versions made possible the final version. Conclusion: We translated and adapted the TVQ to Brazilian Portuguese. The questionnaire is suitable for the next steps


INTRODUCTION
Pharyngolaryngeal symptoms are common after thyroidectomy (1)(2)(3)(4) and generally relate to the involvement of the recurrent or superior laryngeal nerve during surgery (1) .However, symptoms such as difficulties in swallowing and vocal changes can occur even in the apparent absence of nerve damage.This corresponds to a condition of multiple and sometimes undefined causes, known as post-thyroidectomy syndrome (1,3) .
Although the clinical routine usually neglects these symptoms, considering it as minor complications of thyroid disease or thyroidectomy, they significantly affect the quality of life of patients (1) .In this sense, studies recommend investigating laryngeal, vocal, and swallowing conditions before and after thyroidectomy (5) .This investigation usually involves instrumental exams that demand cost, time, and that are not always available in the services (6) .An alternative to minimize these barriers is using self-reported questionnaires as screening tools.These questionnaires can select those patients who, in fact, need more specific tests for diagnostic confirmation (6) .Some authors proposed The Thyroidectomy-Related Voice Questionnaire (TVQ) for this purpose (5) .
The TVQ consists of 20 questions that address how often general vocal complaints and representative symptoms of laryngopharyngeal reflux, vocal fold paralysis, and changes in swallowing have occurred in the last month.A cut-off point defines the need to refer the patient to more specific exams (5) .Since the publication of the TVQ, other studies have reproduced it, applying the instrument both before and after surgery (2,(6)(7)(8)(9) .
Brazil does not have a questionnaire with the characteristics of the TVQ currently under validation.In this perspective, this study translates and cross-culturally adapts the TVQ to Brazilian Portuguese aiming to start its validation process.The translation and adaptation of the TVQ to Brazilian Portuguese followed traditional international guidelines (10) and the recommendations of Pernambuco, Espelt, Magalhães Júnior, and Lima (2017) (11) .After the consent and authorization of the researcher responsible for the original instrument, we performed the following steps:

This
1) Translation: independent translations of the original version into Brazilian Portuguese by two translators, one of them being a health professional and the other not, both without knowledge about the TVQ.Thus, we had two versions: T1 and T2.
2) Synthesis of the translations: the two translators of the first stage plus two researchers from the group responsible for the validation process reached a consensus after comparing the T1 and T2 versions, evaluating semantic, idiomatic, conceptual, linguistic, contextual, and cultural discrepancies.This led to a synthesis version (T3).
3) Expert committee: we sent the T3 version by email to a committee of twenty-eight (28) professionals with expertise in the area of voice and/or swallowing.These professionals worked clinically with, or studied, patients undergoing thyroidectomy.Twenty-one (21) experts accepted the invitation.These experts judged four aspects: relevance of each item for the purpose of the instrument; feasibility of items within the Brazilian cultural context; operational changes in the items; and adequacy of the T3 version of the instrument.Furthermore, we asked the judges to justify their disapprovals and forward comments or suggestions that they considered relevant.
For aspects such as relevance and feasibility, the judges indicated, on a Likert-type scale, whether they considered the item to be: (1) irrelevant or infeasible; (2) little relevant or little feasible; (3) relevant or feasible; (4) very relevant or very feasible.Regarding the operational changes, the alternatives were 'yes' or 'no'.Finally, the experts judged whether the synthesis of the translations was adequate or inadequate.
We calculated the Item Content Validity Index (I-CVI) (12) , that is, the proportion of approvals for each aspect, to determine the level of agreement between the experts.In the case of relevance and feasibility, categories 3 and 4 corresponded to approvals, and the other categories represented disapprovals.We considered as acceptable those items with I-CVI above 0.78.Moreover, we calculated the Questionnaire Content Validity Index (Q-CVI) using the arithmetic mean of the I-CVI in each of the four aspects.The minimum acceptable value for the Q-CVI was 0.90.
We submitted both the I-CVI and Q-CVI values, as well as the comments and suggestions from experts, to analysis by two researchers in the group responsible for the validation process.These researchers reached a consensus on the necessary adjustments.Then, we sent the reviewed version of the instrument back to the same twenty-one (21) specialists from the previous step to judge whether they considered the items adequate or inadequate and write comments and/or suggestions.The answer rate was 80%.We analyzed the comments and suggestions for the I-CVI and Q-CVI, subsequently developing the pretest version (T4).
4) Pretest: we performed this step with the target population (pre-and post-thyroidectomy patients) to verify the adequacy, structure, and application of the items in a real context.The pretest included volunteers aged 18 years or over, of both genders, who were undergoing, or candidates for, partial or total thyroidectomy.We excluded from this step subjects with difficulty in understanding simple commands; with self-reported or medically recorded psychiatric, neurological, neuromuscular, or neurodegenerative alterations; with cognitive decline, noticeable or reported by a companion; with lowered level of consciousness; with previous surgeries in the head and neck; and with previous cancer treatment.The convenience sample consisted of 20 volunteers: nine (45%) prethyroidectomy patients and 10 (55%) post-thyroidectomy patients.Everyone who met the eligibility criteria signed the informed consent form and underwent a cognitive interview to confirm their understanding of the instrument items.In that interview, the researcher used the paraphrase strategy, asking a question and demanding the interviewee to repeat it immediately afterwards.In addition, for each item, the interviewer directly asked the participant if he/she had understood the question and if he/she had any suggestions for modification.During the interview, the researcher also recorded possible operational difficulties of the interviewee and nonverbal reactions such as facial expressions and impressions that evoked incomprehension or disinterest.The same researcher conducted all interviews, recording everything in writing.Finally, two researchers from the group responsible for the validation process analyzed the results and reached a consensus on the need for adjustments, inclusion, or exclusion of items, which led to the prefinal version (T5).
5. Back-translation: to assess whether the items reflected the content of the original version, we sent the T5 version for back-translation by an English-speaking consultant whose mother tongue is English but who is also fluent in Brazilian Portuguese.The translator was unaware of the TVQ.
6. Final synthesis: made by consensus by two researchers from the group responsible for the validation process.The researchers compared the original version to the back-translated and prefinal versions in terms of semantic, idiomatic, experiential, conceptual, syntactic/grammatical, and operational equivalences.This analysis led to the final version of the instrument.
The researchers analyzed the equivalences by consensus according to the following criteria (13) : (1) semantic equivalence: did the words mean the same thing?; (2) idiomatic equivalence: was it necessary to formulate an expression equivalent to colloquialisms difficult to translate?; (3) experiential equivalence: was the original item replaced by a similar item that actually exists in the target culture?; (4) conceptual equivalence: did the words/expressions with a different conceptual meaning between cultures need to be replaced or deleted?; (5) syntactic/grammatical equivalence: were adjustments related to orthographic or grammatical questions necessary?; (6) operational equivalence: did procedures inherent to the application of the instrument need to be modified.

RESULTS
Chart 1 shows the versions of the TVQ items at each stage of the translation and cross-cultural adaptation process.
It is noteworthy that, during the synthesis of the translations, the authors proposed an operational equivalence to the target culture.It consisted in transforming the questionnaire into an interview, modifying the statements of the instrument and the structure of the items from affirmative statements to direct interrogative sentences.
Still in the synthesis stage, the authors also decided by consensus to insert the pronoun "you" at the beginning of all items after their transformation into direct interrogative sentences.However, in the next step, one of the experts suggested removing the pronoun to make the instrument less informal.Hence, the authors decided to exclude the pronoun from all items.
Table 1 shows the results of the decisions of the expert committee on the I-CVI and Q-CVI.Regarding relevance, only items 17 and 18 were below acceptable levels.However, the translation of these items was semantically mistaken.Therefore, the authors decided to adjust the items (and not to exclude them) according to the suggestions of experts.
Regarding the feasibility of the items, only one of them was below the acceptable level, showing that the experts agreed that the TVQ items are mostly feasible in the Brazilian culture.The Q-CVI was acceptable for both relevance and feasibility.Operational equivalence was also widely accepted, as only two experts did not agree with the proposed change.
Regarding the adequacy of the synthesis of the translations, three items were below the acceptable level, as well as the Q-CVI.The authors then changed these items and, after the second round of expert judgment, the I-CVI and Q-CVI became adequate.
The most substantial changes in the items occurred after consulting the experts (Chart 1).During the pretest, there was a constant need to remember the response categories to the interviewees.Thus, the authors recommend that the interviewer repeat the answer possibilities at the end of each question.After the pretest, item five (5) became item ten (10).With that, items six (6), seven (7), eight (8), nine (9), and ten (10) became items five (5), six (6), seven (7), eight (8), and nine (9), respectively.
The back-translation corresponded to the content of the original version, differing only in relation to the sentence structure.Thus, the translated version of the TVQ, adapted to Brazilian Portuguese (Appendix 1), proved to be equivalent to the original version.

DISCUSSION
The methods of the present study enabled the translation and cross-cultural adaptation of the TVQ instrument to the Brazilian culture, making it suitable for the next steps in the validation process.After validation, the Brazilian Portuguese version of the TVQ can be reproduced to track, evaluate, and monitor general vocal complaints and representative symptoms of laryngopharyngeal reflux, vocal fold paralysis, and changes in swallowing in patients undergoing, or candidates for, thyroidectomy (8).Incorporating an instrument such as TVQ in the clinical routine may favor the creation of an operational flow in the services, and more assertively direct therapeutic procedures before and after thyroidectomy.
This study of translation and cross-cultural adaptation showed the relevance of each step for the improvement of the instrument in its Brazilian Portuguese version.Two individuals with different profiles translated the instrument; experts judged the items and translations; representatives of the target population were able to give their opinion; and the back-translation corresponded to the original version, thus enabling the final synthesis.The present study highlights the collaboration of members of the expert committee and target population, as their comments and suggestions encouraged the authors to discuss and adapt the instrument.
The I-CVI and Q-CVI values were mostly adequate.However, inadequate items went through reviews and adjustments according to needs.The fact that experts consider most items relevant to the questionnaire and feasible in the Brazilian culture ratified the need for the instrument.
In this study, the first need was to suggest an operational equivalence in relation to the format of the items.The proposal to transform them into interrogative sentences and apply the instrument as an interview considered the fact that most Brazilians prefer that the assessment instruments be applied in this way, mainly due to personal preference and reading difficulties (14).In Brazil, there are limitations to the wide use of self-applied instruments, considering the low educational level of part of the population and the consequent difficulty in understanding (15).Most experts agreed with this equivalence, so the authors maintained it.
In addition to operational equivalences, semantic, idiomatic, and syntactic/grammatical equivalences of some items were necessary.Among them, the predominant one was the semantic equivalence, necessary to ensure that the words or expressions of the original version have the same meaning.
The idea of validating the TVQ in Brazil was initially due to the need to have a standardized, valid, and reliable instrument that would allow to know thyroidectomy-related vocal and swallowing symptoms.Although several services already carry out this survey, their instruments are not robust from a psychometric point of view.Therefore, the process of validating the Brazilian Portuguese version of the TVQ will continue from the translated and adapted version presented in this study.

CONCLUSION
After all the methodological steps, this study translated and adapted the TVQ to Brazilian Portuguese, making the instrument suitable for the next validation steps.
is a validation study restricted to the stage of translation and cross-cultural adaptation.The study was approved by the Human Research Ethics Committee of the Health Sciences Center of the Federal University of Paraiba, Brazil, through Opinion No. 2,190,942/2017, under Resolution 466/2012 of the National Health Council (NHC).All study participants signed the Informed Consent Form (ICF).

Chart 1 .
Versions obtained during the translation and cross-cultural adaptation of the Thyroidectomy Voice-Related Questionnaire (TVQ) to Brazilian Portuguese

Table 1 .
Item Content Validity Index (I-CVI) and Questionnaire Content Validity Index (Q-CVI) according to the relevance and feasibility of the items, operational equivalence and synthesis of translations in the two rounds Corresponds to the suggestion to transform the affirmative declarative sentences of the original version into direct interrogative sentences *