Translation and cultural adaptation of “Glasgow Children’s Benefit Inventory” into Brazilian Portuguese

Highlights • The GCBI questionnaire in tonsillectomized children obtained a Cronbach's alpha of 0.94.• The translation of the GCBI questionnaire maintained semantic equivalence.• The cross-cultural adaptation of the GCBI is conceptually and culturally adequate for the sample.


Introduction
The quality-of-life assessment tools are fundamental for the analysis of different treatments and medical conditions since they measure global changes in the health condition of patients and allow comparisons with other clinical conditions. 1 It has been demonstrated that measurement instruments of postoperative results in health, used in a systematic way, result in a better communication and decision-making between doctors and patients and increase satisfaction of patients with health care. 2 The verification of health-related quality of life of pediatric patients helps healthcare professionals to understand the impact caused by illness or treatment on the patient's life. 3Thus, it is important to have validated instruments in Brazilian Portuguese available for the health-related quality life assessment of children.
The literature offers a series of questionnaires used to measure pediatric ENT postoperative results, but most of the instruments assess the global quality of life of the patient without correlating it with an improvement or worsening after a specific intervention.The most frequently used instruments are the ''Pediatric Quality of Life Questionnaire'', the ''KINDL-R Questionnaire'' to measure the health-related quality of life in children and teenagers, the ''Child Health Questionnaire'', the ''Child Behavior Checklist'', the ''Preschool children quality of life questionnaire'' and the ''Glasgow Children's Benefit Inventory (GCBI)'', as the review carried out. 1 Among the analyzed questionnaires, the only one projected to be sensitive to changes after a medical intervention or treatment was the ''Glasgow Children Benefit Inventory'', widely used to assess the impact of different procedures (cochlear implant, 4 bone anchored prosthesis, 5 nasal endoscopic surgery, 6 adenotonsillectomy 7 ).It was originally described in English and has already been validated in several languages, such as German, 7---10 Italian, 11 Dutch, 12 Spanish, 13 Swedish, 14 Greek, 6 Turkish, 15 Russian 16 and Mandarin Chinese, 17 allowing the comparison of results obtained from ENT interventions and treatments between different countries.
This study was carried out due to the importance of measurement tools, translated and adapted into Brazilian Portuguese, for the assessment of the impact on the quality of life of children under 12 years old after pediatric ENT interventions.We chose the GCBI because it was specifically developed for this purpose and is widely used in the international literature, what allows comparisons of results. 18,19hus, the objectives of the present study were to translate and cross-culturally adapt the GCBI questionnaire.

Method
This methodological study was conducted, after the authors of the original instrument have authorized it's translation into Brazilian Portuguese, in accordance with the ethical norms and was approved by the Research Ethics Committee under the number CAAE 42075520300005327.All participants digitally signed the Free and Informed Consent Term.
The sample chosen for the application of the questionnaire was of children submitted to tonsillectomy, a common surgical procedure in pediatric population for either sleep apnea syndrome or chronic infection, alterations that significantly impact the quality of life of patients 20 The study was developed in 7-stages as proposed by Borsa et al. 21: 1) Translation into Portuguese: Translation into Portuguese by two independent bilingual translators (semantic equivalence).2) Synthesis of the translated versions: Synthesis of the two translations by the researchers.3) Evaluation of the synthesis by experts: Translated and synthetized version evaluated by a committee formed by five experts.After the independent analysis and together with the committee, an adapted version was obtained with the necessary changes for a better interpretation.4) Evaluation by the target audience: Evaluation of the translated questionnaire by 10 individuals from the target audience.Convenience sampling was selected considering patients submitted to tonsillectomy at a university hospital.A survey of patients under the following inclusion criteria was required to the hospital department in charge: children up to the age of 12 under tonsillectomy postoperative follow-up (including patients who were submitted to minor nasal procedures at the same surgical time such as adenoidectomy and/or cauterization of the inferior turbinates) between January 2019 and December 2021 with a minimum of 6 months (considering the period of time enough for surgical recovery) up to a maximum of 3 years of postoperative care (to mitigate memory bias).Exclusion criteria: patients submitted to tympanostomy with ventilation tube placement at the same surgical time to avoid confounding bias relating to postoperative satisfaction.Data from the patients' records was used for further information about the diagnosis and the intervention to which they had been submitted.
The caregivers of selected children were contacted and participated via phone call (up to 3 phone call attempts were made at different times and on different days).Sample collection of the instrument for the analysis of internal consistency was carried out through the Cronbach alpha coefficient.
The response options in the questionnaire are listed using Likert-type scales with 5 options.After the application, responses must be rescaled as follows 18 : −100 (maximum damage), −50 (moderate damage), 0 (no change), +50 (moderate benefit) and +100 (maximum benefit).The total scale and subscales are obtained by calculating the average of responses to the items.

Statistical analysis
Data was exported from Google Forms software, where the questionnaire was answered, to Microsoft Excel software.
The statistical analyses were performed by the software SPSS version 21.0.The results obtained were described in the sample by the median and the interquartile range while the internal consistency of items was evaluated by the Cronbach alpha coefficient.

Results
Each question was analyzed as shown in Table 1, comparing the original questionnaire (Column 1) with the synthetized translation (Column 2).After the analysis by the group of experts and target audience, the final version was generated (Column 3) and, based on it, the back translated version was made (Column 4).The changes made in the translation and cultural adaptation of the questionnaire can be found in Column 5.
The final version of the translated and adapted instrument in Brazilian Portuguese can be found in Appendix 2.
For the collection of responses obtained in the questionnaire, a phone call attempt was made to 99 people responsible for the operated children (up to 3 attempts were made on different days and at different times).Among the 99 people contacted, 69 (69.6%) accepted the phone invitation to participate in the study, 29 (29.2%) did not answer the phone call and 1 (1.1%) refused to participate.The Free and Informed Consent Term and the adapted questionnaire were sent via a digital message app for the self-application of those 69 participants who agreed to be part of the study.
A total of 28 (40.5%)responded questionnaires were returned, which were used for the data analysis.In this sample, there was information from responsible people for 17 male children and 11 female children submitted to tonsillectomy.The ages of patients varied from 2-years old to 7-years old on the date of the surgery with an average of 4-years old (Fig. 1).
Considering the 24 questions, a total Cronbach alpha coefficient of 0.94 was found, what suggests a high reliability of the items.
In the sample, it was also found in the result of the questionnaire general score a median of +66.7 (responses with  In the result of the subareas evaluated by the questionnaire, positive results were also observed, what suggests a postoperative improvement.In the psychosocial evaluation, a median of +70 was found (responses with interquartile range between +5 and +100), in the physical health evaluation a median of +100 was obtained (responses with an interquartile range between +70 and +100), in the analysis of behavior a median of +71.42 was found (responses with an interquartile range between +23.21 and +98.21), in the subarea of vitality a median of +83.33 was observed (responses with an interquartile range between +50 and +100) (Fig. 2).

Discussion
In this article we presented the translation and cultural adaptation into Brazilian Portuguese of the questionnaire ''Glasgow Children's Benefit Inventory''.We applied the questionnaire with those responsible for children submitted to tonsillectomy under 12 years old in a retrospective way so that the measurement of changes in the quality of life could be more sensitive.
Although there are various methods of translation and cultural adaptation, we chose the one proposed by Borsa et al. 21The method encompasses the internationally known concept of back translation and details the process of cultural adaptation and has been used by many authors for both translation and cultural adaptation. 22,23t is important to make it clear that the translation process is the first stage of the cultural adaptation process.After translating from the original language to the target language, we observe the characteristics of the instrument that have semantic equivalence and at the same time adapt, linguistic and culturally, to a different contexto. 24Two independent translators made two versions of the questionnaire (Stage 1) that could be compared and discussed in the elaboration of a synthetized version (Stage 2) by the researchers.That mitigates the risks of linguistic, psychological, cultural as well as theoretical and practical understanding biases. 25he cultural adaptation that resulted from the evaluation of the translated version when compared with the original version by five experts, both independently and together (Stage 3) demonstrated appropriate results of understanding of the text during the application with the target audience (Sage 4).The terminology of the instrument was understandable in all items of the questionnaire and did not arise any question from the target audience.
After the back translation (Stage 5) of the final version a small difference in the expressions used was observed when compared to the original version.Changes in expressions and grammar structures of some items were made based on the need to obtain semantic, idiomatic, experimental, and conceptual equivalence, which was nearly total between the two versions.Such changes can be observed in Column 5 of Table 1.
The final version of the instrument turned out to be conceptually and culturally appropriate for the target audience.
We observed the questionnaire is retrospectively applied and there may be a memory bias.The author of the original instrument justifies this choice in the form of application of the questionnaire with the purpose of obtaining a higher sensitivity in the detection of changes when directing questions about the variation observed in the health condition due to the intervention, instead of taking pre and postoperative actions and subtracting one from the Other. 19ven though the authors 19 described in the original instrument that it is appropriate for a broad pediatric age group, we suggest it should be applied with caution in children under 2-years old due to the possible difficulty of interpretation of some items for infants.In this study, questionnaires were not applied in children under two years old.
During the application for the assessment of children submitted to tonsillectomy, the participation rate of respondents to the questionnaire was of 40.5%, similar to the participation rate of 38% 19 in the original study.The analysis of responses showed high reliability of the items, evidenced by the Cronbach alpha coefficient of 0.94, similar to the original article, which obtained a Cronbach alpha of 0.92 19 and similar to the translations and cultural adaptations into Arabic 4 0.9 and German 8 0.84.
We suggest the continuity of the instrument analysis, an increase in the sample for application of psychometric tests and validation of the questionnaire.

Conclusion
The translation and cross-cultural adaptation conducted made it possible the creation of a Brazilian version of the Glasgow Children's Benefit Inventory for the assessment of children's quality of life after ENT interventions.The final version of the adapted instrument, named Avaliac ¸ão de Glasgow dos Benefícios à Crianc ¸a is semantically, conceptually, and culturally equivalent to the original and applicable to the Brazilian children.Its application in children submitted to tonsillectomy showed high reliability of the items.

Figure 2
Figure 2 Glasgow Children's Benefits Inventory Total score and subscales (psychosocial, physical health, behavior, and vitality) for patients submitted to tonsillectomy (n = 24).The box plots show the interquartile ranges, the thicker line shows the median.Values higher than 1.5 times the interquartile range are shown as circles.interquartile range between +48.43 and +95.31), what suggests an improvement in the postoperative quality of life.In the result of the subareas evaluated by the questionnaire, positive results were also observed, what suggests a postoperative improvement.In the psychosocial evaluation, a median of +70 was found (responses with interquartile range between +5 and +100), in the physical health evaluation a median of +100 was obtained (responses with an interquartile range between +70 and +100), in the analysis of behavior a median of +71.42 was found (responses with an interquartile range between +23.21 and +98.21), in the subarea of vitality a median of +83.33 was observed (responses with an interquartile range between +50 and +100) (Fig.2).

5 )
Back translation: The synthetized and reviewed version used with the target audience was back translated into the original language (English) by two new translators.

Table 1
Linguistic adaptations of the GCBI to the Portuguese language spoken in Brazil.
1. Has your child's operation made his/her overall life better or worse? 1.A operac ¸ão de seu/sua filho (a) melhorou ou piorou a vida dele (a) em geral? 1.A cirurgia de seu/sua filho (a) melhorou ou piorou a vida dele (a) em geral? 1. Has your child's operation improved or worsen their overall life?