Translation, Transcultural Adaptation, and Validation of the Brazilian Portuguese Version of the Body Image Disturbance Questionnaire for Scoliosis (BR-BIDQ-S)

Objective  The Body Image Disturbance Questionnaire for Scoliosis (BIDQ-S) for scoliosis derives from the Body Image Disturbance Questionnaire (BIDQ) with specific adaptation for scoliosis patients. Despite its significance and applicability, this instrument has never been translated into Brazilian Portuguese. The current study aimed to translate, transculturally adapt, and validate the BIDQ-S into Brazilian Portuguese. Methods  BIDQ-S was translated and culturally adapted into Brazilian Portuguese using the American Association of Orthopedic Surgeons (AAOS) criteria. The questionnaire validation relied on internal consistency and comparison with the Cobb angle, Pediatric Quality of Life Inventory (PedsQL), and Scoliosis Research Society (SRS-22). The Brazilian version (BR-)BIDQ-S validation occurred in a sample of 35 adolescents with scoliosis waiting for specialized treatment. Results  Internal consistency of the BR-BIDQ-S was 0.899 according to the Cronbach's index (i.e., virtually perfect). Although BR-BIDQ-S did not correlate with the Cobb angle, it presented correlations with the Physical, Emotional, and Social domains from the PedsQL and the Function/Activity domain from the SRS-22. Conclusion  BR-BIDQ-S was reliable in evaluating the body image of adolescents with scoliosis, presenting an internal consistency of 0,899 (virtually perfect). Moreover, similar to the original instrument, it correlated with PedsQL and SRS-22.


Introduction
Scoliosis is a lateral curvature of the spine greater than 10°a ssociated with a rotational component, while spinal asymmetry refers to deviations lower than 10°. 1 Scoliosis classification relies on etiology, curvature location, age of onset, or curvature. 1Secondary scoliosis or scoliosis of known etiology represents around 20% of all causes, while idiopathic scoliosis is the most common clinical type. 2,3Idiopathic scoliosis classification may also rely on age as infantile (0-3 years old), juvenile (4-9 years old), and adolescent (over 10 years old). 2 It is worth highlighting that scoliosis does not change the appearance of the back alone but may extend to the anterior region of the trunk, scapulae (shoulder imbalance or elevation), and pelvic and hip tilt. 3Tones et al. 4 concluded that adolescents with scoliosis can suffer a significant psychosocial impact, especially when presenting prior associated emotional factors.Among psychosocial disorders, concern about body image has become a growing problem impacting the lives and mental health of adolescents. 4Schwieger et al. 5 analyzed patients from The Bracing in Adolescent Idiopathic Scoliosis Trial (BrAIST) and reported that those with a Cobb angle !40°had worse scores on the Spinal Appearance Questionnaire (SAQ) and the Pediatric Quality of Life Inventory (PedsQOL) 4.0 Generic Scales even after 2 years of treatment follow-up.These scores reflect a significant loss of self-esteem and quality of life.
The BIDQ-S questionnaire originated from the Body Image Disturbance Questionnaire (BIDQ) proposed by Cash et al. 6 BIDQ-S is an adapted and validated BIQS version by Auerbach et al. 7 to assess body image disorders, especially in scoliosis patients.The version validated by Auerbach et al. 7 is in English, with translations into Korean, German, Turkish, and simplified Chinese.However, there is no validated version in Brazilian Portuguese.Therefore, considering the applicability and significance of this questionnaire, this study aimed to translate the BIDQ-S into Brazilian Portuguese, culturally adapt it, and validate it in this language.

Casuistry and Methods
This study transcultural adapted and validated BIDQ-S into Brazilian Portuguese using the translation and transcultural adaptation criteria proposed by Beaton et al. 8 (2007and recommended by the American Association of Orthopedic Surgeons (AAOS). 7[3] Conclusion BR-BIDQ-S was reliable in evaluating the body image of adolescents with scoliosis, presenting an internal consistency of 0,899 (virtually perfect).Moreover, similar to the original instrument, it correlated with PedsQL and SRS-22.

Palavras-chave
The study occurred from September 2021 to June 2022 and is part of a larger project approved by the institutional Research Ethics Committee under number CAAE 27816320.0000.5520.All subjects agreeing to participate in the study and their guardians signed the Free and Informed Consent (ICF) and Assent forms.
We performed all stages of translation and cross-cultural validation proposed by AAOS. 8In the first stage, three bilingual Portuguese-English authors experienced in medicine freely translated (BIDQ-S), 7 from English to Brazilian Portuguese.The second stage consisted of translation harmonization and synthesis.The third stage was an adaptation based on the opinion of three experts in spine who were fluent in English and evaluated the harmonized version.The final version was the result of the second harmonization based on experts' suggestions.
The final version of the BIDQ-S for Brazilian Portuguese underwent a pre-test for cross-cultural adaptation with ten adolescents reporting a complete and sufficient understanding of the instrument.Next, the validation of this version employed 35 patients with adolescent idiopathic scoliosis who were on the waiting list for treatment, using data obtained directly from managers of the Unified Health System (SUS, for its acronym in Portuguese).All patients' contact occurred by letters, telephone calls, electronic means, through patient associations, or in person by the study team.
The research subjects were from the Hospital's Orthopedics Service.The study included all subjects identified as having adolescent idiopathic scoliosis aged from 10 to 18 years old who were awaiting treatment at a service with potential access.Patients with non-idiopathic scoliosis (congenital or neuromuscular), scoliosis lower than 10 degrees, or waiting for a second spinal procedure were excluded.

Sociodemographic Questionnaire and Radiological Information
The applied sociodemographic and clinical questionnaire asked about age, gender, self-declared ethnicity/race, weight, height, body mass index (BMI), education, and origin.Next, we collected information regarding illness duration, associated conditions, development of secondary sexual characteristics, etc.We also analyzed panoramic radiographs of the patient's spine to determine the Cobb angle of the main curvature and, as a result, the degree of scoliosis (ranging from 1 to 3).
The instrument: The Body Image Disturbance Questionnaire-Scoliosis (BIDQ-S) 6 The questionnaire had seven items.Items 1, 2, 5, 6, and 7 were subdivided into A and B. Items 1A, 2A, 3, 4, 5A, 6A, and 7A were questions with objective answers about image disturbances.Each question had five answer options, ranging from "not worried" to "extremely worried."Each answer received a score from 1 ("not worried") to 5 ("extremely worried").The final BIDQ-S score was the average of the points obtained on these questions.The higher the score, the greater the body image concern/disturbance level.As mentioned, items 1, 2, 5, 6, and 7 had subdivisions into A and B, with B being subjective characterizations (open responses) of the objective response from subdivision A. In this way, the BIDQ-S instrument presented a quantitative and qualitative analysis of the image disturbances in the patient with scoliosis. 7diatric Quality of Life Inventory (PedsQL) version for adolescent report (13-18 years old) 9,10 PedsQL has several versions, subdivided by age (5-7, 8-12, and 13-18, 18-25 years old) and by patient or parent report (for patients up to 18 years old).The questionnaire consists of four scales: physical dimension, emotional dimension, social dimension, and educational dimension, and the final scores are divided into these dimensions.The answers range from never (0) to virtually always (4), and the scores undergo a reversible transformation in percentages, i.e., 0 ¼ 100, 1 ¼ 75, 2 ¼ 50, 3 ¼ 25, and 4 ¼ 0. Therefore, when calculating the result, the higher the percentage, the better the quality of life.The dimension cannot be computed if more than 50% of items are incomplete.The result from each domain is the average of the items answered, while the final result is the average of the four separate domains. 9,10oliosis Research Society (SRS-22r) 11,12 The revised version of the SRS-22 consists of 22 questions divided into five domains, i.e., function/activity, pain, personal image/appearance, mental health, and treatment satisfaction.Answers are objective and range from 1 (worst answer) to 5 (best answer).This questionnaire is validated in Brazilian Portuguese.Data analysis evaluates subtotal and mean values from each domain.The percentage of the result is calculated, considering 100% the best possible answer. 11,12

Data Analysis
We used the Statistical Package for the Social Sciences (SPSS) software version 25.0 for Windows for data tabulation and analysis.Quantitative variables were expressed as mean AE standard deviation or median and interquartile range (IQR).Qualitative variables were shown as simple and relative frequencies.Statistical analysis consisted of independent samples t-tests, Pearson correlation, ANOVA, and Cronbach's alpha.The significance level was p < 0.05.

Results
Most subjects were female, around 14 years old, mixed race, attending elementary level II, from the state capital, and with an average Cobb angle above 50 degrees (►Table 1).
The BIDQ-S had the worst scores related to questions 1 ,2 ,3, and 7, in that order, demonstrating concern about appearance, difficulty stopping thinking about appearance, sadness associated with scoliosis, and the need to avoid situations because of scoliosis.There was no difference in the responses of girls compared with boys (►Table 2).The internal consistency of the BIDQ-S in the Brazilian Portuguese version was 0.899 per Cronbach's alpha coefficient.This internal consistency is virtually perfect according to the Lands and Koch criteria. 13here was no significant correlation between BIDQ-S and the Cobb angle (r ¼ 0.312 with p ¼ 0.094) using Pearson's correlation.Regarding health-related quality of life, measured by PedsQL, the lowest scores were in the Emotional Dimension domain.The PedsQL did not correlate with the Cobb angle (r ¼ -0.259 with p 0.168); there was a correlation between the BIDQ-S and the Physical Dimension, Emotional Dimension, and Social Dimension domains of the PedsQL (►Table 3).
There was a significant correlation between the BIDQ-S and the SRS-22 in the function/activity dimension.The SRS-22 was also related to the Cobb angle in the function/ activity and pain dimensions through Pearson correlation (►Table 4).

Discussion
The Brazilian Portuguese version of the BIDQ-S had an internal consistency of 0.899 according to Cronbach's alpha coefficient, considered virtually perfect. 13The worst scores were in the domains concerned with appearance, difficulty stopping thinking about appearance, sadness associated with scoliosis, and the need to avoid situations due to scoliosis.Despite not correlating with the Cobb angle, the BIDQ-S questionnaire significantly correlated with the participants' health-related quality of life in the Physical Dimension, Emotional Dimension, and Social Dimension domains of the PedsQL.The BIDQ-S correlated only with the Function/Activity Dimension of the SRS-22.
Our internal consistency is similar to that obtained in equivalent studies described previously and to the original BIDQ and BIDQ-s versions, confirming the validity of the Brazilian Portuguese version of the BIDQ-S.The original BIDQ questionnaire, a generic instrument originating a version for scoliosis created by Cash et al., 6 in 2004, presented a Cronbach's alpha coefficient of 0.89. 13The Australian version of the BIDQ presented a Cronbach's alpha of 0.92. 14The adapted version and validated for scoliosis (BIDQ-S), presented by Auerbach et al., 7 2014, had a coefficient of 0.82. 7The validation of the BIDQ-S questionnaire in its Korean, Turkish, German, and Simplified Chinese had Cronbach's alpha of 0.88, 15 0.88, 16 , 0.87, 17 , and 0.877 17 , respectively.
Our study did not find a significant correlation between the Brazilian Portuguese version of the BIDQ-S and the Cobb angle, even when analyzing patients with and without surgical indication as subgroups.The BIDQ-S validation study from Auerbach et al. 6   (N ¼ 100) also did not observe a significant correlation between the BIDQ-S questionnaire and the Cobb angle. 17The same study, in a subanalysis, grouped patients with Cobb !40°a nd < 40°, observing that those with a Cobb angle !40°had a higher (worse) C-BIDQ-S result than subjects with a Cobb <40°. 17The Korean validation study by Bae et al. 15 (N ¼ 113) indicated that patients with larger Cobb angles tend to present worse body image. 15The German validation study from Wetterkamp et al. 17 ncluding 259 patients was the single one with a significant correlation between the Cobb angle and G-BIDQ-S. 17Wetterkamp et al. 17 hypothesized that the study by Auerbach et al. 7 did not find a significant correlation between the Cobb angle and the BIDQ-S due to a smaller sample size. 17his can be extrapolated to our study, raising the possibility of a lack of correlation due to the small sample size.
The mean BIDQ-S score in our study was 2.05 (AE 0.84), and it was calculated by the average points from questions 1 to 7, ranging from 1 (best possible scenario) to 5 (worst possible scenario).Compared with other studies, the German version had a mean score of 2.04 (AE 0.76), including 2.26 AE 0.84 in the surgical subgroup and 1.83 AE 0.60 in the non-surgical subgroup. 17For the Korean version, the mean score was 2.4 AE 0.8, with the score from the brace user subgroup being significantly higher in comparison with the observation subgroup; moreover, the subgroup requiring surgery had a significantly higher score than the brace subgroup. 15The mean score of the original validation study from Auerbach et al. 7 was 1.50 AE 0.49, with significant differences between the surgical (1.57), non-surgical (1.45), and control (1.06) subgroups.Our study found the highest BIDQ-S scores in questions 1, 2, and 3, respectively, consistent with the Turkish validation study, with an overall mean value of 2.03 and the highest scores in questions 1 (2.93 AE 1.267), 2 (2.40 AE 1.07), and 3 (2.11AE 1.12). 16ur study showed a significant negative correlation between BIDQ-S and SRS-22 in the function/activity dimension.When correlating BIDQ-S with SRS-22, the German validation study found a moderate negative correlation, specifically in the personal image/appearance domain from SRS-22 (À0.74). 17he Korean version of the BIDQ-S found a significant correlation only for the appearance domain of SRS-22 (p < 0.001, r ¼ À651). 15Auerbach et al. 7 also found a significant correlation between BIDQ-S and SRS-22 since BIDQ-S correlated with the total SRS-22 (À0.72), the activity (À0.53), pain (À0.53), image (À0.60), and mental (À0.50) domains scores, all with p < 0.001.We believe cultural differences may explain the inconsistency in the correlation between BIDQ-S and SRS-22 in our study and the remaining literature.
Our study found a significant negative correlation between BIDQ-S and PedsQL in the physical, emotional, and social dimensions.Bauer et al. 18 found a strong relationship between BIDQ-S, SRS-22, and PedsQL questionnaires.Their study also highlighted that the mental health domain of SRS-22r correlated well with the emotional domain of PedsQL, suggesting that mental health issues interfere more with personal image than the level of deformity alone. 18Therefore, these authors emphasize that they chose to apply the BIDQ-S as a body image questionnaire in patients with idiopathic scoliosis as it has a good correlation with other existing questionnaires and fewer questions, consequently requiring a lower application time. 18ur study has some limitations.The number of patients in our sample, although similar to the one in the original version, may have been relatively small to provide power to all analyses.Our patients were recruited from a clinical-surgical outpatient clinic; therefore, most underwent surgery as the most recommended treatment and usually presented larger Cobb angles.
As it occurred in a specialized center for scoliosis treatment, our results may not reflect the same characteristics of the scoliosis population, reducing the external validity of the study.Most of our sample consisted of girls and, although there were no significant differences between genders, this may have given some bias to the work.Despite these limitations, we could validate BIDQ-S in its Brazilian Portuguese version and carry out fundamental secondary analyses confirmed in similar studies.

Conclusion
Our study demonstrated that the Brazilian Portuguese version of the BIDQ-S (BR-BIDQ-S) is a reliable instrument for assessing adolescent body image, with an internal validity of 0.899 (virtually perfect), which is similar to the internal consistency of the original instrument.Furthermore, BR-BIDQ-S correlated with the quality of life of adolescents measured by PedsQL and with the Function/Activity Domain of SRS-22, reinforcing its validity.
(N ¼ 49) also found no correlation between the Cobb angle of the main curvature and the BIDQ-S questionnaire scores.The validation study of the BIDQ-S in its Simplified Chinese version, by Bao et al. apud Wetterkamp et al.17

Table 1
Sociodemographic of the sample

Table 2
Body Image Disturbance Questionnaire for Scoliosis (BIDQ-S) scores per questions and participants' genders

Table 3
Pediatric Quality of Life Inventory (PedsQL) score and its correlation with the Body Image Disturbance Questionnaire for Scoliosis (BIDQ-S) and Cobb angle

Table 4
Scoliosis Research Society (SRS-22) score and its correlation with the Body Image Disturbance Questionnaire for Scoliosis (BIDQ-S) and Cobb angle Questionnaire comparison used the Pearson correlation.