Introduction

Cystic fibrosis (CF) is an autosomal-recessive genetic disease which affects the function of various organs in a person. It is the most common life-threatening, inherited disease in caucasian and is caused by mutations in the gene encoding for the CF transmembrane conductance regulator1. Technological advancements improved the treatment of individuals with cystic fibrosis and increased their life expectancy, currently estimated at 50 years2. However, information on how adolescents and young adults cope with CF are limited3.

The productive cough, repetitive lung infections, and high number of hospitalizations associated with CF may cause a negative perception of health in individuals without CF who are unfamiliar with coping with this reality4,5,6. These factors, combined with premature mortality, physical impact, and low social acceptance of symptoms, may stigmatize individuals with CF7.

The word “stigma” defines any situation in which an individual is not fully accepted by others, with depreciation and distinction regarding a particular social level8. Also, health-related stigma occurs when the negative social experience or judgments are based on a health condition, resulting in social rejection, guilt, or devaluation9. Individuals with CF experience stigma since childhood, at home (due to physical differences from siblings caused by malnutrition) and school (due to the frequent cough and easy tiredness compared with their peers). Additionally, changes in the daily routine occur due to symptoms and the treatment needed, and these differences may lead to fear of non-acceptance and induce anxiety and concern in social interactions10.

Social withdrawal may lead to emotional disturbances, such as low self-esteem, helplessness, and depression11. Also, psychosocial issues were associated with changes in lung function in individuals with CF12. Moreover, the stigma may impair treatment adherence since the need to feel similar to healthy individuals may lead to a lack of commitment to the clinical routine. In other cases, individuals with CF hide the diagnosis from their social circle, concealing the symptoms and taking medication covertly11.

Despite the importance, few studies and instruments address the stigma in CF. In 2014, Pakhale et al.13 proposed the CF Stigma Scale to assess stigma in CF based on a scale for individuals with acquired immunodeficiency syndrome. The authors found positive correlations with other questionnaires for anxiety, depression, and symptom severity and a negative correlation with the quality of life score14.

In Brazil, a lack of instruments assessing the stigma experienced by individuals with CF is observed. Therefore, this study aimed to translate, cross-culturally adapt, and evaluate the psychometric properties of the Brazilian version of the CF Stigma Scale.

Methods

This exploratory methodological study was conducted between August 2018 and April 2020 and included the participation of translators, experts, individuals with CF, and their caregivers. Figure 1 illustrates the study flowchart.

Fig. 1
figure 1

Flowchart of the cross-cultural adaptation and validation process of the Brazilian CF Stigma Scale, 2020.

Full size image

The study was approved by the research ethics committee of the Federal University of Rio Grande do Norte and followed the Declaration of Helsinki and Resolution 466/12 of the Brazilian National Health Council (no. 96606918.3.0000.5537).

All individuals were informed about the aims, justification, risks, benefits, and procedures of the study, considering its methodology. They were also informed about the voluntary nature of participation, preservation of the anonymity of the recorded data, and the possibility of refusing to participate or drop out at any time without prejudice. After explanations, all individuals signed the informed consent form.

The CF Stigma Scale developed by Pakhale et al.13 consists of 10 items, which address different dimensions of stigma, such as shame, discrimination, and the social impact of the disease. Each item is evaluated using a Likert scale, allowing the measurement of the participants’ degree of agreement or disagreement with statements about their experience of stigma. The CF Stigma Scale has demonstrated good reliability and validity, correlating with instruments that assess quality of life, anxiety, depression, and symptom severity. For this study, the original version of the scale was translated and culturally adapted to the Brazilian context, with the aim of evaluating its applicability and psychometric properties in a sample of Brazilian patients with CF.

Translation and cross-cultural adaptation

Initially, authorization and collaboration were requested from the authors of the original CF Stigma Scale. Respecting copyright laws, Dr. Smita Pakhale (the main author) agreed and made herself available to assist in the steps.

The methodological procedure followed the models Guillemin et al.15 and Beaton et al.16, as well as the recommendations of the Consensus-based Standards for the Selection of Health Measurement Instruments17 and Mapi Research Institute18. The models comprise the following steps: translation; synthesis of translations; back-translation and synthesis of back-translations; review by a multiprofessional experts committee (MEC) using the Delphi method19,20; pre-test21. The last two steps were assessed using the Cohen’s kappa coefficient, according to the indices proposed by Landis and Koch22.

Two independent Brazilian translators fluent in English translated the original scale into Brazilian Portuguese. One translator was an expert in the field, familiar with the aim of the scale, and translated considering a clinical and technical perspective. The other was a non-expert, without knowledge of the scale, and represented the popular language.

A review committee (i.e., the two translators and responsible researchers) developed a consensus version of the two translations and a report with modifications and adjustments. The report and translation were sent to the authors of the original scale for analysis.

The consensus version was back translated into English to verify the content correspondence. In this step, two independent bilingual translators (native English speakers and proficient in Portuguese) with no prior contact with the scale and its aim translated the Brazilian version into English.

A new review committee (i.e., the main researcher and back-translators) developed a consensus version of the back-translations and sent it to the original authors with the report of this step. After approval, a Brazilian Portuguese professor verified the adequacy of the items of the Brazilian version to the language norms.

Cross-cultural equivalence is accomplished when the items of an instrument maintain the content validity for application in various cultural conditions16. Thus, the expert panel method was used19, comprising a MEC with eight professionals: two pulmonologists; two physical therapists; one psychologist researcher experienced in cross-cultural adaptation and one not directly involved with CF; and one mother of individual with CF and one individual with CF above 18 years16. Each pair of professionals was formed considering that one had direct involvement with CF and the other did not.

The MEC received the original scale, all versions of the translations, and reports and ensured the semantic, idiomatic, and contextual equivalences between the translation and original version15. Semantic equivalence determines multiple meanings for an item and grammatical difficulties in translation, the idiomatic verifies whether colloquial words are difficult to understand or translate, and the contextual analyzes the possibility of different meanings of words in different cultures.

Equivalences were analyzed using the Delphi method through a virtual form (Google Forms®) individually and confidentially sent via email20. The concordance was evaluated using a dichotomous scale, in which (0) indicated complete disagreement and (1) indicated complete agreement19. Each member of the MEC scored and justified the items of the Brazilian version for clarity, concordance, and adequacy to the cultural context. The approval of the items followed the scores and agreement between experts.

The acceptable validity content is not standardized due to the subjective judgment. However, this studies subjectivity can be reduced by using Cohen’s kappa coefficient21, and agreement values proposed by Landis and Koch22. After review and consensus by the MEC, the instrument was sent to the original authors for analysis.

The pre-final version was conducted online with 30 Brazilian individuals over 18 years old diagnosed with CF by the sweat test16,21, selected using non-probability (convenience) sampling. Those unable to perform or understand the procedures or who underwent lung transplantation were excluded. Individuals were recruited through disclosure on the social media of the Unidos Pela Vida institution, aimed at professionals, parents, family members, and individuals with CF. Individuals responded to the translated version and a dichotomous scale regarding their understanding of each item using Google Forms®, which was analyzed using Cohen’s kappa21,22. Last, the Brazilian version of the CF Stigma Scale was sent to the original authors for final approval of the methodological process.

Analysis of psychometric properties

After the cross-cultural adaptation, the psychometric properties of the Brazilian CF Stigma Scale were analyzed according to the methodological design of the original study by Pakhale et al.13. The sample consisted of Brazilian individuals over 18 years old diagnosed with CF by the sweat test, selected using non-probability (convenience) sampling. Individuals unable to perform or understand the procedures or with lung transplantation were excluded.

Although methodological validation studies have no specific sample size calculation, five to ten individuals are recommended for each item, resulting in a sample of 50 individuals since the scale has 10 items23. According to data from the Brazilian CF Registry, in 2020, there were a total of 6,112 individuals with CF. In Brazil, the CF population is predominantly pediatric and the proportion of adults is still less than 30% of the total. Thus, a sample of 50 individuals corresponds to almost 3% of the adult population24.

Individuals were recruited through dissemination on the social media of the Unidos Pela Vida institution, aimed at professionals, parents, family members, and individuals with CF. Those interested contacted the researchers and received a link for the Google Forms® with the informed consent and study forms. The individuals with CF included in the translation and cross-cultural adaptation phase also participated in this analysis.

Initially, individuals responded to the test step regarding personal and socioeconomic data. Next, four instruments were presented: Short-Form 12-Item Survey - version 2 (SF-12v2), General Anxiety Disorder 7-item scale (GAD-7), Cystic Fibrosis Quality of Life Questionnaire - Revised (CFQ-R), and the Brazilian CF Stigma Scale. The retest step was conducted after three weeks, when individuals responded to the same instruments. The test-retest period in the original study was three months. Since the study was conducted during the coronavirus disease 2019 (COVID-19) pandemic, we shortened the test-retest period of the present study to minimize potential sample losses.

The SF-12v2 is a self-report scale assessing the general quality of life, with 12 items to assess the current physical and mental health status of individuals based on their experiences over the past four weeks. It is a short version of the Short-Form 36-Item to reduce the time completion without compromising the quality of assessment. Responses are recorded on a five-point scale ranging from “all the time” to “never” or “excellent” to “poor”; a cut-off score of 50 points determines a good quality of life25.

The GAD-7 is a self-report scale of seven items assessing anxiety over the past two weeks. Each item addresses a symptom of anxiety grouped into a single domain for generalized anxiety. Responses are recorded on a four-point scale (0 = not at all; 1 = several days; 2 = more than half the days; 3 = nearly every day), and the total score ranges from 0 to 21; high scores correspond to greater severity of anxiety symptoms. Moreno et al.26 translated and validated the GAD-7 for the Brazilian population.

The CFQ-R is a self-report scale for individuals ≥ 14 years old composed of 50 questions grouped into physical, role, vitality, emotion, social, nutrition, body image, treatment, health, weight, respiratory, and digestive domains. Scores for each domain range from 0 to 100, and a 50-point cut-off indicates a good quality of life27. The CFQ-R was translated and validated into Brazilian Portuguese by Rozov et al.28.

Analysis

Data were exported to an Excel spreadsheet to verify missing values and to the SPSS version 26 for analysis. Data were presented as mean and standard deviation, with a significance level of 5%. The content validity, test-retest reliability, and convergent validity were assessed. The test-retest interval was three weeks. Also, internal consistency (Cronbach alpha coefficient) and test-retest reliability (Pearson correlation coefficient) were considered as reliability for the total score and subscales. The convergent validity was calculated using the paired t-test.

Results

Translation and cross-cultural adaptation

Adjustments during the translation and cross-cultural adaptation prevented potential misunderstanding due a literal translation. In item 4, the word “bad” would have a literal translation of “ruim” and was changed to “má” to maintain the real meaning. In item 5, the word “disgusting” is translated as “repugnante”, which is not commonly used in colloquial language and could also misrepresent the real meaning, being consensually changed to “desagradável”. Also, sentence sequences in other items were slightly changed to be better understood by Brazilian individuals with CF.

After analysis by the MEC, Cohen’s kappa coefficients were > 0.61, indicating substantial concordance and dismissing new discussions22. In the pre-test, 30 individuals with CF aged between 18 and 42 years were included, being 22 (73%) females. The education level varied from incomplete high school to higher education. Cohen’s kappa coefficient for clarity and understanding of the items ranged from 0.48 to 0.72, indicating moderate to substantial concordance22.

Analysis of psychometric properties

A total of 60 individuals with CF were interested in the study, of which 57 and 52 completed the test and retest, respectively. Only data from individuals who completed both steps were used for the analysis of psychometric properties. Table 1 shows the sample characterization.

Table 1 Characteristics of the sample participating in the validation stage of the CF stigma scale for Brazil, 2020 (n = 52).
Full size table

The analysis of psychometric properties and reliability measures are presented in Table 2. The Brazilian CF Stigma Scale scores positively correlated with GAD-7 and negatively with SF-12v2 scores. The Brazilian CF Stigma Scale also demonstrated adequate internal consistency (α = 0.836), strong test-retest correlation (r = 0.886, p < 0.0001), and no significant difference between items responses in the test and retest (p = 0.3).

Table 2 Analysis of psychometric properties of the Brazilian CF stigma scale (n = 52).
Full size table

Table 3 shows the correlation of the Brazilian CF Stigma Scale scores with SF-12v2 and CFQ-R domains. The Brazilian CF Stigma Scale was negatively correlated with the mental domain of the SF-12v2 and domains physical, role, vitality, emotion, social, body image, treatment, and health of the CFQ-R.

Table 3 Correlations of the CF stigma scale scores with the SF-12v2 and CFQ-R domains (n = 52).
Full size table

Discussion

The present study demonstrated that the Brazilian CF Stigma Scale was adequately adapted and cross-culturally validated, with substantial content validity, high level of internal consistency, convergent validity (moderate and significant correlations with the other instruments), and strong test-retest reliability. These results corroborated those found by Pakhale et al.13. The profile of translators and inclusion of individuals with or related to CF in the MEC may have provided robustness to the translation and cross-cultural adaptation, favoring the semantic, conceptual, and contextual equivalences29,30,31,32,33.

Psychometric properties demonstrated that the Brazilian CF Stigma Scale may be used nationwide since it can provide valid and reliable information about the experience of stigma. When compared with the SF-12v2 and CFQ-R domains, the Brazilian CF Stigma Scale scores showed moderate negative correlations with mental (SF-12v2), motion, and health (CFQ-R) domains, and a strong negative correlation with the social domain (CFQ-R), whereas the other correlations were not significant. These results suggested that mental state, emotions, health, and especially social interactions could influence the perception of stigma.

The selection of a non-probabilistic sample should be considered when interpreting our results. However, this fact does not undermine the contribution of this study, which is the translation, cross-cultural adaptation and validation of the CF Stigma Scale for Brazilian individuals with CF. Moreover, despite being large, the sample did not comprise individuals from diverse Brazilian regions, socioeconomic classes, and educational levels. However, once the Brazilian CF Stigma Scale is validated for the national territory, professionals may use it to assess whether individuals with CF from different regions and specific contexts are experiencing any stigma. Moreover, this scale may be compared with other instruments assessing factors that may trigger stigma and impact treatment adherence, as hypothesized by Berge and Patterson11.

Considering that data for the analysis of psychometric properties were collected during the COVID-19 pandemic, social isolation may have influenced the scores of the instruments included in the study since they assess quality of life and anxiety. However, this influence possibly did not affect the validation and evaluation of psychometric properties since individuals were already in social isolation during data collection in the first step. Another limitation of the study was the exclusion of some instruments used by Pakhale et al.13. For example, the Centre for Epidemiologic Studies Depression scale could not be used because its Brazilian Portuguese translation was not accessible.

Although the test-retest period of the present study (three weeks) was shorter than in the original study (three months), results were possibly not affected. The GAD-7 and CFQ-R responses refer to the last two weeks, while SF-12v2 is based on the experiences over the past four weeks. Considering the CF Stigma Scale, the stigma associated with chronic respiratory diseases typically does not have an acute onset; it tends to develop gradually over time. Stigma intensifies as symptoms become more evident and the limitations imposed by the condition become notable to others. Therefore, stigma is more often a phenomenon that builds over time rather than arising suddenly. Thus, the experience of stigma is unlikely to change over three weeks.

In conclusion, the CF Stigma Scale was translated and adapted for the Brazilian context and demonstrated favorable psychometric properties for use in the Brazilian population. It is easy to apply and cost-effective and should be used in future studies to assess whether the experience of stigma is present in different socioeconomic and educational contexts or in younger individuals. Also, further studies should explore potential relationships between the Brazilian CF Stigma Scale and respiratory (e.g., lung function) or health-related outcomes (e.g., treatment adherence).