Cross Cultural Adaptation and Validation of Hindi Version of WHOQOL-BREF in Patients With Chronic Low Back Pain

Aim of the study was to validate Hindi version of WHOQOL-BREF in chronic low back pain patients (CLBP). In this cross-sectional study, 111 patients with CLBP were recruited. In addition to demographic information, two questionnaires Hi WHOQOL-BREF and SF-36 (Reference scale) were administered at day 0 and day 3. NRS was used for pain evaluation. Cronbach’s alpha coecient was used for scale reliability. Construct validity was analysed using Pearson correlation coecient. Conrmatory factor analysis was performed to determine the relationships between the eight domains of SF-36 and four domains of the WHOQOL-BREF. Cronbach’s alpha coecients were acceptable for all domains of both Hi WHOQOL-BREF (0.869 - 0.938) and SF-36 (0.752 - 0.943) questionnaires. Pearson correlation coecients of both instruments were partly to strongly correlate with most domains (r ≥ 0.40). Correlations for domains with similar constructs were stronger than those measuring varied constructs. Conrmatory factor analysis recommended approximately good relationships among the SF-36 and WHOQOL-BREF domains. Our study suggests that WHOQOL-BREF Hindi version is a reliable and valid tool for clinical and research use in CLBP.


Introduction
Low back pain is a common health problem worldwide with considerable impact on nancial and socioeconomic status. It is the most common cause of absenteeism and job-related disability. Disease remains as one of the top two contributors of global disability for over two decades with a total cost of around billions of dollars annually. 1 In India, nearly 60% of people suffer from low back pain in their life at some point or another. 2 Chronic pain impacts multiple life domains such as physical and mental wellbeing, social relationship and functional ability thus impacting the quality of life (QOL). 3 Assessing this health related quality of life (HRQoL) can provide an estimate of how the disease in uences people's lives and how people manage to live with chronic back pain. 4 Several questionnaires are available for assessing QoL. These questionnaires help de ne patient's disability and impairment, degree of change of condition over time and the appropriate choice of therapy. 5 World Health Organization QOL-BREF (WHOQOL-BREF) questionnaire is one of the best known instruments that has been developed for cross-cultural comparisons of Qol and is available in more than 40 languages making between countries comparison feasible. 6 Questionnaire has been extensively used to assess QoL in various conditions including bronchial asthma, visceral leshmaniasis, mental illness and many more. 7,8,9 Hindi is a widely used language in the second most populous country in the world i.e. India. 10 WHO BRIEF questionnaire is short and is easy to use in busy outpatient departments of developing countries. A generic Hindi version has been developed in India as a part of multi country initiative by WHO. 6 Though the questionnaire appears suited for Hindi population, the Hindi version has not been tested for construct in patients with chronic low back pain (CLBP). Thus, aim of this study was to perform a cross cultural adaptation and validation of Hindi (Hi) version of WHOQOL-BREF health questionnaire in patients with CLBP. Methodology Procedure Initial permission was sought from Dolores Campanario World Health Organization (Permissions Management, Reprint Rights and Licensing) to reproduce, reprint and/or re-translate WHOQOL-BREF Hindi version cross-culturally and validate it for use in CLBP patients. The study was approved by the Institute Ethics Committee, PGIMER, Chandigarh, India (INT/IEC/2021/SPL-403 dated 12th March, 2021) and registered with Central trial registry India vide no CTRI/2021/04/033034. Written informed consent was obtained from all participants before participating in the study. All methods were performed in accordance with the relevant guidelines and regulations.
Patients of either gender with CLBP duration of >3 months, aged 18-65 years who were able to read and speak native Hindi language, attending Pain Clinic in the Department of Anaesthesia, PGIMER, Chandigarh were recruited in the present study. Patients with any other pain conditions, chronic disorders, or the presence of "red ags" were excluded from the study.

Phases of the Study
The present study was performed in two phases: (a) cognitive debrie ng of Hi-WHOQOL-BREF (b) crosscultural validation of the resulting adapted Hi-WHOQOL-BREF Cognitive debrie ng In order to assess the comprehensibility of the Hindi version provided by WHO, cognitive debrie ng, a method to test and validate a questionnaire, was administered to 10 CLBP patients of either gender as per study inclusion criteria. Along with completing the Hi-WHOQOL-BREF, all the patients described understanding of the items of the scale in their own verbatim. They were also asked to suggest alternate words for the words that they found di cult to understand. The responses were summarized, including the suggestions indicated by the participants.

Cross-Cultural Validation
All the recruited patients completed the Hi-WHOQOL-BREF together with socio-demographic information and other reference scales (36-Item Short Form Survey: SF-36 and Numeric Rating Scale: NRS). To assess test-retest reliability, the Hi-WHOQOL-BREF was re-administered to these patients 3 days after they completed the baseline questionnaire. E-version (Google forms) of both questionnaires were shared with the patients on Day 3 via WhatsApp. This was done to avoid unnecessary visit of patients to outpatient department during COVID pandemic and this was a suitable alternative to the paper version as well. 11 ( Figure 1) Instruments Numeric Rating Scale (NRS) Pain severity was assessed using a 0 to 100 NRS. In the survey, patients were asked, "On a scale from 0 to 100, mark/tell your level of CLBP, with 0 being none and 100 being unbearable." 12 SF-36 (36-Item Short Form Survey) It is a multicultural scale consisting of 36 questions and categorized into eight-domain pro le of scores: physical functioning (PF; 10 items), general health (GH; 5 items), role physical (i.e., role limitations due to the physical health problems, RP; 4 items), bodily pain (BP; 2 items), social functioning (SF; 2 items), vitality (VT; 4 items), role emotional (i.e., role limitations due to emotional problems, RE; 3 items), and mental health (MH; 5 items). For each domain, a score ranging from 0 to 100 was assessed with a higher score indicating better health. 13 Scoring was done using PRO-CoRE scoring software provided by QualityMetric Incorporated, LLC. Score were represented both as individual components as well as two nal domains: PCS (physical component score) and MCS (mental component score).Higher the score better the QOL. Considerable evidence has been found for the reliability of the SF-36 (Cronbach's alpha greater than 0.85, reliability coe cient greater than 0.75 and for construct validity in terms of distinguishing between groups with expected health differences. 14,15 World Health Organization QOL-BREF (WHOQOL-BREF) The questionnaire contains two items from the overall QOL and GH and 24 items of patient satisfaction that are divided into four domains: Physical health with 7 items (Domain 1), psychological health with 6 items (Domain 2), social relationships with 3 items (Domain 3) and environmental health with 8 items (Domain 4). Each item is rated on a 5-point Likert scale. Raw domain scores were transformed to a 0-100 score according to WHO guidelines. 6 Sample size calculation According to HinkinTR, the sample size required to perform exploratory factor analysis is a sample size to number of items ratio of no lower than 4:1. As Hi-HI-WHOQOL-BREF has 26 items (26*4=104), minimum of 104 patients were required. Assuming an attrition of about 5%, 111 cases were enrolled. 16

STATISTICAL ANALYSIS
The statistical analysis was carried out using IBM SPSS (Statistical Package for Social Sciences) statistical version 20. 17 All quantitative variables were estimated using measures of central location (mean) and measures of dispersion (standard deviation). Normality of data was checked by Skewness and Kurtosis. For normally distributed data, mean were compared for follow-up using the paired t-test. Internal consistency of the domains was assessed by using Cronbach's alpha coe cient for the entire scale, each construct and each factor. A value of >0.70 was considered su cient. 18 Test re-test reliability was assessed using Intra-class correlation coe cient [ICC]. To evaluate the convergence and discriminant validity, correlations among the SF-36 and the Hi-WHOQOL-BREF were examined applying Pearson correlation coe cient. It was hypothesized that those domains that are conceptually related would be more strongly correlated, but those domains in the two instruments with less in common would demonstrate weaker correlations. Therefore, we assumed moderate to high correlations (Pearson correlation coe cient signi cant at two-tailed level of signi cance (p ≤0.01 and p ≤0.05) between all domains of the Hi-WHOQOL-BREF and all domains of the SF-36. Con rmatory factor analysis (CFA) was done to examine the relationships between the eight domains of SF-36 and four domains of Hi-WHOQOL-BREF. Extent to which variance in each domain was explained by other domains in both instruments was tested. We used the goodness-of-r index (GFI), incremental t index (IFI), the comparative t index (CFI), the Tucker-Lewis Index (TLI) and the root-mean-square error of approximation (RMSEA) in order to assess model t. P<0.05 was considered as statistically signi cant

Results
Results of cognitive debrie ng were reviewed by one of the investigators (RS). No item caused di culties in comprehension for more than three participants. Hence, no modi cations were suggested and the nal translated original Hi-WHOQOL-BREF was locked down and entered for the cross-cultural validation phase in CLBP patients. These 10 patients were not enrolled in the nal sample.

Reliability a) Internal Consistency
Internal consistency of the clinical measure of Hi-WHOQOL-BREF was assessed for total and sub-scales by using Cronbach's-α. Value obtained in the study ranged between 0.753 to 0.971 for items (Table 3) and 0.869 -0.938 for domains ( Table 2) for Hi-WHOQOL-BREF. Value obtained for SF-36 ranged between 0.752 -0.943 for domains. (Table 2) The results show good internal coherence between these speci c items within each domain in population studied. b) Test-Retest Reliability between Baseline and Day 3 All the patients screened at baseline were re-administered the Hi-WHOQOL-BREF after 3 days. ICC was calculated from the data of patients who responded for both observations. Retest was successfully administered after three days in 93.6% patients. Seven patients did not respond at day 3. Any patient responding thereafter was not included. Test re-test was found to be good for all four components of Hi-WHOQOL-BREF (ICC 0.768 -0.883) ( Table 2). Signi cant difference between Day 0 and Day 3 was observed only in two items; item-15 of domain 1 (PH) with p-value of .000 and item 5 of domain 2 (PS) with p-value of .023. All other items were comparable. Item wise description of Hi-WHOQOL-BREF ICC is presented in Table-3. c) Bland-Altman Plot A scatter plot was created for total baseline and total retest scores for all above mentioned scales and plotted against difference of two set of scores. The plots for Hi-WHOQOL-BREF showed good test retest reliability. (Figure 2      Con rmatory factor analysis showed that approximately all three models were tted well and similarly.
The WHOQOL-100 allows detailed assessment of each individual facet relating to QOL. In certain instances however, the WHOQOL-100 may be too lengthy for practical use. The WHOQOL-BREF Field Trial Version has therefore been developed to provide a short form of QOL assessment questionnaire that looks at domain level pro les. WHOQOL-BREF has been widely translated and validated in several languages and in many chronic disorders. 19 Hindi version used in the present study was obtained from WHO organisation online. Though there are many published QOL measures, there is still a lack of consensus amongst researchers about its de nition as seen in the choice of items for their instruments while assessing QOL. SF-36 is a su cient measure of health status and functioning of patients with LBP 29 and hence we decided to use it as a reference scale. However, it is a more objective questionnaire when compared to WHOQOL-BREF and requires more time to ll. On the other hand, WHOQOL-BREF is a completely subjective questionnaire.
WHO-QOL assessment scale is a valuable tool for patients with CLBP where prognosis is likely to involve only partial recovery or remission. Countries where no validated QoL measures currently exist can be con dent that data yielded by work involving the WHOQOL assessments will be genuinely sensitive to their setting. This study shall make multi-centre QoL research possible and comparable.
One of the advantages of this study is the use of a sample from one of the largest government hospitals in North India for the validation of the Hindi version of a patient satisfaction questionnaire. This sample may overcome any cultural and environmental factors that could lead to differences in the instrument's ability to measure an object of interest. The high response rate to the questionnaire can be explained by its design, which takes into consideration the de ciencies observed in other questionnaires, such as a large number of questions, including all the activities performed in daily living of a common person, which may discourage participants from completing the questionnaire. .

STUDY LIMITATIONS
The present study has certain limitations. Even though explained, an intention to aggravate ones condition for the above reason and thus creating a test retest reliability bias cannot be ruled out. Also, investigators did not undergo any formal training to conduct cognitive debrie ng. Another limitation was the short test-retest time interval (3 days). This was selected assuming no change in disease state within such a short span of time. However, the chance of the memory effect for an observed effect of good testretest reliability cannot be ruled out. Finally, the generalizability of the Hi-WHOQOL-BREF to other musculoskeletal disorders apart from CLBP cannot be assured from the results of the present study. Average of test and retest scores of Hi-WHOQOL-BREF is plotted against the difference of these two scores. Note: Continuous horizontal centre line is of mean difference and the line above and below represent limits of agreement (the mean difference ± 1.96 times the SD of the differences).