Adaptation and validation of the Bengali version of the Chronic Urticaria Quality of Life Questionnaire (CU-Q₂oL)

Study design and data collection

This study was divided into two parts. The first part involved the adaptation and translation of the CU-Q₂oL questionnaire in Bengali using the standard methodology for translation and cross-cultural adaptation. For this purpose, we recruited two adult expert translators, two adult public health experts with previous experience in questionnaire adaptation and two independent dermatologists with more than five years of clinical experience. Among the two dermatologists, one acted as a translator, making three translators. All of them were native Bengali speakers with bilingual proficiency. They participated in this study voluntarily after signing written informed consent.

The second part involved a cross-sectional study conducted at the Dermatology out-patient department (OPD) of the College of Medicine and Sagore Dutta Hospital in Kolkata, India across three months (June to August 2021). All participants were adult (age > 18 years) chronic urticaria patients and were recruited after obtaining written informed consent. Acute urticaria (<6 weeks), urticarial vasculitis, angioedema as a solitary symptom, other skin diseases, psychiatric disorders, and any chronic disease with the potential to interfere with the quality of life formed our exclusion criteria. All patients were provided with the self-reportable CU-Q₂oL, Dermatology Life Quality Index and Urticaria Control test questionnaires and asked to fill them by themselves, or via an interview with a professional psychologist (illiterate, old-age and/or visually impaired patients) [Day 0]. We obtained the scores along with the socio-demographic and disease parameters. Relevant laboratory investigations were performed when necessary. We advised the patients to return after 2-weeks to complete the CU-Q₂oL questionnaire again (re-test) [Day 14]. The study was approved by the Institutional Review Board and conducted as per the WMA Declaration of Helsinki (updated in 2013).

We recruited 56 eligible chronic urticaria patients on day 0, while 42 patients participated in the re-test session on day 14. Thus, the final sample size of our study translated to 42 subjects.

Description of the CU-Q₂oL, Dermatology Life Quality Index and Urticaria Control test questionnaires

The original Chronic Urticaria Quality of Life (CU-Q₂oL) Questionnaire contains 10 statements and 13 questions with a total of 23 items in the Italian language. The items are categorised into six domains: pruritus (2 items), swelling (2 items), impact on life activities (6 items), sleep problems (5 items), limitations (3 items) and appearance (5 items). Each item is scored on a 5-point Likert scale (0-4) to indicate its intensities over the last 2 weeks. The total summed score is calculated and linearly transformed to a 0-100 score, with the maximum score of 100 indicating the worst HRQoL.⁵ Currently it is the only validated scale to measure specific quality of life in chronic urticaria (both inducible and spontaneous).

The Dermatology Life Quality Index is a generic HRQoL questionnaire for all dermatologic diseases to enable quick assessment in routine clinical practice. It comprises 10 questions spanning six domains-symptoms and feelings, daily activities, leisure, work and school, personal relationships, and treatment. Each question is scored on a Likert scale (0 for not at all to 3 for very much). The total score ranges from 0 to 30, with higher scores indicating poorer quality of life.⁴ The Dermatology Life Quality Index has been validated in multiple languages including the Bengali language.⁸

The Urticaria Control test is the only validated and reliable tool for assessing disease control in both chronic spontaneous and inducible urticaria, with a recall period of four weeks. It comprises four items, each being marked on a 0-4 scale. The total score is obtained by adding all the items, ranging from minimum 0 to maximum 16, with 16 points indicating complete disease control. The Urticaria Control test has been validated in multiple languages including Bengali.⁹

Translation of the CU-Q₂oL questionnaire and further modifications

The translation process is presented in Figure 1. Apart from the original Italian version we also utilised the validated English version of the questionnaire.⁶ Three expert translators translated the English version (V1.0) with previous experience in adapting health-related questionnaires from English to Bengali. After translating the questionnaire individually, the translators came to a consensus and yielded a single Bengali version of the questionnaire (V1.1). Then this questionnaire was sent to another set of three translators who translated the questionnaire into English and came to a consensus to yield a single English version (V1.2). The conceptual equivalence between the original questionnaire (V1.0) and the back-translated version (V1.2) was checked and found to be of accepted equivalence. It was checked by two health care experts with previous experience with two Bengali questionnaire adaptations. The questionnaire was found to be conceptually equivalent and the version was marked as V1.3. This questionnaire was used to conduct a cognitive interview with 10 patients suffering from chronic urticaria. Their understanding of the question, and their suggestions on addition, and modification of the questionnaire were noted for further amendment of the questionnaire. We added two synonyms of words in parenthesis for a better understanding of the questionnaire. As it was only the addition of the synonyms, keeping other questions the same as previous, the version (V1.4) was not again sent for review to the translators. This version was final for the clinical test.

Close

Figure 1:

Process of adaptation of the questionnaire from English to Bengali. Footnote for reviewers: The infographic was created by HM in Microsoft Word 2010® for this manuscript only

Export to PPT

Based on the feedback of our cognitive interview session, we have made slight additional modifications. The original CU-Q₂oL questionnaire contains 10 statements and 13 questions with a total of 23 items with five-point Likert-type response options. We have changed the 13 questions to statements for a uniform type of items keeping the response options the same. The logic behind converting the questions into statements is described below with an example. If a question is asked as “Do you have difficulty falling asleep?,” the answer may be-“yes, I have difficulty” or “No, I don’t have difficulty.” However, if the statement is “I have difficulty in falling asleep.,” the response may be ranging from “Not at all” to “very much.” Furthermore, the original questionnaire does not have any instruction section on how to fill the questionnaire up. Hence, we added a section of text on how to mark the responses on the questionnaire. This would help to make the questionnaire more suitable for self-administration. In addition to instructional text, a figure is also used to show how to mark the response. This would further ease the understanding of how to respond.

Assessment of the internal consistency, reliability and validity of the Bengali CU-Q₂oL

We calculated the Cronbach alpha coefficient to assess the internal consistency of the Bengali CU-Q₂oL. Based on existing literature, we considered a coefficient <0.6 as unacceptable, 0.6-0.65 undesirable, 0.65-0.7 minimally acceptable, 0.7-0.8 respectable, 0.8-0.9 excellent and >0.9 excessive.¹⁰

Test-retest reliability was assessed across 2 visits 14-day apart (D0 and D14) to evaluate the consistency of Bengali CU-Q₂oL by calculating the intra-class correlation coefficient. Intra-class correlation coefficient values <0.4, 0.4-0.75 and >0.75 indicate poor, acceptable and strong reliability, respectively.¹⁰

We used the validated Bengali Dermatology Life Quality Index and Urticaria Control test questionnaires to determine the overall HRQoL and disease control of our patients. The scores obtained were statistically correlated with the score of Bengali CU-Q₂oL (test-scale) by Spearman’s correlation to assess the validity of this scale, as construct validity of a test scale is indicated by the correlation with other standard scales. We have considered Spearman correlation coefficients of <0.3, 0.3-0.6 and >0.6 to indicate weak, moderate and strong correlations, respectively.¹⁰

Statistical analysis

Continuous variables were expressed as mean and standard deviation while categorical data were presented as percentages. Categorical data were statistically analysed by either the Binomial test or the Chi-square test, while Mann-Whitney U and Kruskal-Wallis test with the Bonferroni correction method (α = 0.05/6 = 0.0083) were used for continuous data as appropriate. The survey response obtained in the test and re-test phases were coded on Microsoft Excel 2010 (Microsoft Inc., USA) spreadsheet. We analysed data in IBM SPSS Statistics 20 (IBM, Chicago, USA) and GraphPad Prism 6.01 (GraphPad Software, CA, USA). Spearman correlation was used to obtain a correlation between the Bengali CU-Q₂oL, Dermatology Life Quality Index and Urticaria Control test scores. Internal consistency was tested by Cronbach’s alpha and test-retest reliability was tested by intra-class correlation coefficient. For all the statistical tests, a P < 0.05 was considered statistically significant.

Patient characteristics

The mean age of our patients (n = 42) was 38.41 ± 13.38 years (male: female 29:13). The mean duration of chronic urticaria was 8.53 ± 1.55 weeks. Ten (23.8%) and 3 (7.1%, 2- symptomatic dermographism,1- delayed pressure urticaria) patients had concomitant angioedema and physical urticaria respectively. All patients were receiving second-generation antihistamines as their treatment. Table 1 highlights the demographic details of our patients.

None of our patients complained of any difficulty in understanding or answering the Bengali CU-Q₂oL questionnaire. The average time taken to complete the CU-Q₂oL, Dermatology Life Quality Index and UCT questionnaires were 8-10 minutes, 5-6 minutes and 3-4 minutes respectively.

Quality of life scores

In our patients the total CU-Q₂oL score ranged from 22.6 to 74.8, mean ± SD being 48.8 ± 16.5 (95% confidence interval 43.7-53.9) and median (IQR) being 50.4 (32.2-57.4) [Table 3, Figure 2]. Domain 4 (sleep problems) was most severely affected (56.48 ± 23.74), followed by domain 1 (pruritus) [55.91 ± 19.33], while domain 2 (swelling) [35.95 ± 17.39] was least affected [Figure 3]. Inter-domain analysis revealed a statistically significant difference between domain 1 and domains 2, 3 and 5; and domain 2 and domains 3,4 and 6 (all P ’s <0.05, K-W test with Bonferroni correction). Overall, females showed a higher score compared to males (49.77 ± 13.86 vs. 48.43 ± 17.81, P = 0.8). Interestingly, both age (r_s = -0.4, P = 0.003) and disease duration (r_s = -0.3, P = 0.02) showed a statistically significant negative correlation with total CU-Q₂oL score.

Close

Figure 2:

Frequency distribution of overall score obtained from (a) Bengali DLQI, (b) Bengali CU-Q₂oL and (c) Bengali UCT

Export to PPT

Close

Figure 3:

Individual domain-wise scores (in percentage) of the Bengali CU-Q₂oL questionnaire

Export to PPT

The overall mean ± SD Dermatology Life Quality Index score of our subjects was 11.24 ± 5.41 (95% confidence interval 9.5-12.9; range 2-20), males showing a higher score than females, although statistically comparable (P = 0.7) [Table 3 and Figure 2]. Most patients (71.4%, 30/42) reported ‘very large’ effect on their quality of life, followed by ‘small effect’ (19%, 8/42) and a ‘moderate effect’ (9.5%, 4/42).

The overall Urticaria Control test mean ± SD score of our patients was 10.21 ± 3.5 (95% confidence interval 9.1-11.3; range 3-16), comparable across genders [Table 3 and Figure 2]. Interestingly, equal proportion of patients (50%, 21/42) had poorly controlled (Urticaria Control test < 12) and well-controlled urticaria (Urticaria Control test ≥ 12) each.

Internal consistency and reliability of the Bengali CU-Q₂oL questionnaire

The internal consistency of the questionnaire and test-re-test reliability according to the domain is highlighted in Table 4. We noted an excellent overall internal consistency (Cronbach’s alpha = 0.93) of the questionnaire and nearly complete agreement (intra-class correlation coefficient = 0.91) between the test and re-test score. Hence, the adapted Bengali CU-Q₂oL questionnaire has good psychometric properties for possible further application in Bengali-speaking chronic urticaria patients to assess disease burden.

Validity of the Bengali CU-Q₂oL questionnaire

We found a significant positive correlation (r_s = 0.53, P = 0.0002) between the overall scores of Bengali Dermatology Life Quality Index and Bengali CU-Q₂oL scales [Figure 4a]. The positive correlation establishes the validity of the Bengali CU-Q₂oL to determine the chronic urticaria-specific quality of life. In contrast, we found a significant negative correlation (r_s = -0.48, P = 0.0007) between the Bengali Urticaria Control test (a higher score indicates well-controlled disease) and Bengali CU-Q₂oL scores [Figure 4b]. We have also correlated the domain wise score of Bengali CU-Q₂oL with total Dermatology Life Quality Index and Urticaria Control test scores. Except Domain 1 and 2, all other domains showed a statistically significant positive correlation with the total Bengali Dermatology Life Quality Index score and a negative correlation with the overall Bengali Urticaria Control test score [Table 5].

Close

Figure 4:

Correlation of overall score obtained from: (a) Bengali DLQI and, Bengali CU-Q₂oL questionnaires, and (b) Bengali UCT and Bengali CU-Q₂oL questionnaires

Export to PPT

Limitations

Our study was limited by its small sample size, and observational design, therefore lacking the ability to design responsiveness, e.g., before and after therapy, and lack of objective criteria such as laboratory investigations to assess the variations in the quality of life. Another important limitation was the use of rescue antihistamines in some patients in-between the test and re-test sessions, although ideally the test-retest reliability analysis should be conducted in patients without any variation in disease activity. We also failed to correlate individual domains of the CU-Q₂oL questionnaire with individual domains of the Dermatology Life Quality Index scale, as their psychometric properties are different.