Translation and cross-cultural adaptation of WHOQOL-HIV Bref among people living with HIV/AIDS in Pakistan

Background Reliable Health-Related Quality of Life (HRQoL) assessment will be useful in identifying health issues and in identifying health care actions. Due to the lack of a psychometrically valid tool in Urdu, we aim to translate and examine the psychometric and cross-cultural adaptation of WHOQOL HIV Bref among people living with HIV/AIDS (PLWHA) in Pakistan. Methods The standard forward-backwards translation technique was used to convert English version of the WHOQOL HIV Bref into Urdu. After cognitive debriefing, final Urdu version of instrument was developed. Based on the principle of at least 5 subjects for each item, a sample of 182 patients was used using a universal random sampling technique from the Pakistan Institute of Medical Sciences, Islamabad. The Cronbach’s alpha and intra-class correlation coefficients (ICC) were estimated to assess internal validity and reliability of the translated version. Exploratory factor analysis was carried out to determine the factor structure and independent associations between the instrument domains and CD-4T-cell count were assessed using multivariable linear regression Results High Cronbach alpha 0.93 was found for all WHOQOL HIV Bref facets. The test–retest reliability demonstrated a statistically significant ICC ranged from 0.88 to 0.98 (p < 0.001). In known group validity, lower CD-4 lymphocytes count was significantly related to poor scores for all six domains (p < 0.001). Similarly, symptomatic subjects had significantly lower scores compared to asymptomatic subjects on the physical, psychological, social relationship and independence domains (p < 0.05). Statistically significant positive correlation of all six domains of instrument with CD4 cells count (p < 0.001), exhibiting patients with higher CD-4 cells will have higher mean scores of all domains. Factor analysis revealed 5 domains, including physical health, psychological health, social relationship, environmental, and spiritual health. Multivariable linear regression analysis reported; only physical, psychological health and environment health domains were found significantly associated with higher CD-4 lymphocytes count (Beta = 0.121, p < 0.001, Beta = 0.103, p = 0.002, and Beta = 0.032, p = 0.032). Conclusion Findings suggested that the Urdu version of WHOQOL HIV Bref is a psychometrically valid and culturally well-adapted HRQoL measurement tool for PLWHA in Pakistan.

AIDS (PLWHA) worldwide [1]. Despite the fall in peak of new HIV cases all over the world the low-middle-income countries (LMICs) like Pakistan are facing an escalation in the number of HIV cases [2,3]. With the advent of safe Highly Active Antiretroviral Therapy (HAART), the survival of PLWHA has been improved like normal persons [4]. However, Health-Related Quality of Life (HRQoL) of PLWHA remains substantially lower than the general population, even when the majority of HIV-positive people have viral control and are immunologically stable [4][5][6]. Nevertheless, HIV infection and its associated factors like aging, treatment adherence, social conditions, relationship problems, comorbidities, and social stigma still have a significant impact on (HRQoL), even among individuals whose viral count suppressed by ART [4,7]. Therefore, HRQoL of PLWHA has become an important focus for researchers and healthcare providers.
HRQoL and other patient-reported outcomes are extremely important for assessing PLWHA experience and perspective [8]. They reflect the satisfaction of patients and the benefit of HAART that is not necessarily covered by other endpoints. HRQoL results are commonly used in clinical trials and regulatory and reimbursement agencies have begun to request these data in their intervention assessment process [9]. Precise evaluation of HRQoL with valid measures has become crucial for improving PLWHA quality of life [10,11]. Several tools, both generic and disease-specific, are used to measure HRQoL in PLWHA [4]. Generic questionnaires such as the frequently used EQ-5D, WHOQOL, HUI, M-QOL, SF-12, and SF-36 have the advantage of enabling a comparison of HRQoL across different disease populations. However, HIV-specific instruments such as WHOQOL-HIV Bref, HIV-QL-31, MOS-HIV, ACTG-21, AIDS-HAQ, MQOL-HIV, FAHI, PROQOL-HIV, and HIV-SQUAD, have shown greater sensitivity to detect small but clinically significant differences in treatment effects because they focus more on the specific effects of the disease [4,12].
Originally, generic WHOQOL-100 item scale was developed by World Health Organization (WHO) to assess individuals' perceptions of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns [13]. Then, an abbreviated 26-item WHO-QOL Bref has been developed. Later, a 100-item HIVspecific questionnaire (WHOQOL-HIV) was developed and most recently a brief version was developed. The WHOQOL HIV Bref is considered as the most sensitive and reliable particular tool for assessing the HRQoL of PLWHA [10,14]. It contains both generic measures and HIV specific facets. In Pakistan, a validated version of the generic WHOQOL instrument is available [15]. However WHOQOL HIV bref is not validated in local settings of Pakistan. To be used in a specific country, Such an instrument must be translated and validated in the local language to position the individual HRQoL within their local context in terms of culture and values [16]. This instrument has not been translated into Urdu, although it has been translated into many other languages [12,[16][17][18][19][20][21][22][23][24][25][26]. Hence, the current study aimed to assess the psychometric properties and cultural adaptability of the Urdu version of WHOQOL-HIV Bref in PLWHA in Pakistan.

Ethical considerations
The study protocol was approved by the National AIDS Control Program of Pakistan (NACP) and Research Ethics Committee (REC) of ART center of Pakistan Institute of Medical Sciences (PIMS) (Approval No; 1827). Written informed consent from patients was taken after providing them with the written information as well as a verbal explanation of the study. Participants were told that study participation was voluntary and can be terminated at any time. All the information collected in this research will be confidential and only available to the research team that manages the data; Upon completion of the analysis, the contact details of the participants will be removed. All the study procedures were handled following the Helsinki Declaration ethical principles [27].

Study design and study setting
For data collection, we adopted a cross-sectional study design. The study was carried out at the ART center at PIMS hospital, which is a 947-bed hospital located in Islamabad Capital Territory (ICT), Pakistan. PIMS hospital is a teaching and referral center catering to the needs of more than 1.5 million population. ART center PIMS is the biggest center of HIV care in Pakistan having more than 4000 registered PLWH who are getting free of charge HIV treatment. Almost 15 to 20 PLWHA visit the ART center daily for their medical needs. This center was selected because of the geographic location, and it is the biggest referral center where patients from different ethnic groups and regions of Pakistan receive their ART.

Sample size
For the psychometric evaluation of questionnaires, a minimum of five subjects should be selected for each question [28]. Therefore, 155 PLWHA was recruited in the current study for the validation of the 31-item instrument in the present study. With a drop-out rate of 20% (such a high rate was considered due to stigma among PLWH), 190 subjects were approached. Participants who did not complete the questionnaire and patients with missing information (viral load, CD4 T cell count) were excluded; therefore, 182 (95.8%) subjects were included in the final analysis.

Participants
PLWHA included if they were (1) older than 18 years of age; (2) more than six months of HIV diagnosis; (3) CD-4T lymphocyte count and viral load tests not more than four weeks older; (4) regular follow-up at the ART centre and (5) able to communicate in the Urdu language. Patients who were terminally ill, visually impaired, hearing impaired, multiple comorbid, cognitively impaired, and unable to complete interviews were excluded.

Questionnaire
The questionnaire consisted of three parts, including sociodemographic information, HIV related characteristics, and the WHOQOL HIV Bref. The sociodemographic part consisted of gender, age, education, marital status, and work status. HIV related components included HIV since diagnosed, viral load, CD-4 lymphocytes count, HIV serostatus, total time on ART and mode of HIV transmission. HIV serostatus was divided into three stages Asymptomatic, symptomatic and AIDS converted similarly CD-4T lymphocytes count were also divided into three groups.
Five-point Likert scale is used to develop the ratings ranging from 1, Very poor/Not at all/Very dissatisfied/ Never to 5, Very good/Extremely/Completely/Very satisfied/Always. Out of the 31 items, 7 are negative statements, for which responses were reversely coded to ensure higher scores representing better quality of life. Each domain item contributes equally to the score of the domain. Each domain consisted of a different number of questions and the average domain score was multiplied by four for the calculation of each domain score. The total domain score ranged from 4 as the lowest score to 20 as the most effective score [29]. WHOQOL HIV user manual also provides sociodemographic and current health characteristics of PLWHA such as age, gender, marital status and route of HIV transmission [30].

WHOQOL HIV Bref instrument translation
WHO was contacted regarding permission to conduct cross-cultural validation of WHOQOL HIV Bref in the Urdu language (the national language of Pakistan); the English version of the WHOQOL HIV Bref instrument was translated to Urdu by standardized forward-backwards translation procedure ( Fig. 1) explained in 5 steps by Beaton's guidelines [31]. Expert committee (Comprised of health professionals, language professionals, methodologists and translators) was responsible for the consolidation of original questionnaire and all versions of translation (T1, T2, T12, BT1, BT2). The original English version was independently forward translated into Urdu by a bilingual physician (T1) and a non-medical linguist (T2). Later, the authors reconciled the two translations in order to produce a single version of the forward translation (T12). Later the Urdu version (T12) was sent to native expert bilingual physician (BT1) and a native independent bilingual liguist (BT2) for backward translation.
Translators were asked to report any sort of difficulty during translation. After that authors compared the forward backwards translations and amended the questionnaire to produce preliminary translated Urdu version. Upon ensuring consistency in the translated version and the original English version, the preliminary version of Urdu was pre-tested for cognitive debriefing on 30 conveniently chosen PLWHAs at the PIMS hospital. Data were analyzed and Cronbach's alpha for each domain was shown to be between 0.78 and 0.92. Relevant changes to a few items have been made Based on the feedback of the respondents to submit the final version of WHOQOL HIV Bref in Urdu. Expert committee ensured the semantic, idiomatic, experiential and conceptual equivalence of translated version to the original questionnaire.

Data collection procedure
In PIMS ART center, PLWHA regularly visits for ART follow up (All medicines provided to patients were free of cost from the Government of Pakistan) in a separate counseling room. A simple random sampling technique was used to collect data from PLWHA. Data were collected from July to Sep 2019. Eligible patients participated in the study after written consent under the guidance of a trained investigator. After describing the objectives of the study and nature of research study participants were interviewed face-to-face. Respondents were asked to provide sociodemographic information and rate 31 statements of quality of life instrument describing specific behaviors related to HIV care during the past two weeks. Their clinical data was retrived from their medical recods. They were asked to fill the questionnaire for the collection of sociodemographic information and HIV characteristics. The completed questionnaires were checked by the study investigator and soring were performed. Furthermore, incomplete questionnaires were excluded from the study. Fifty four subjects were asked to complete the WHOQOL HIV Bref questionnaire again after the gap of two weeks for test retest reliability analysis.

Statistical analysis
Statistical analysis was performed using Statistical Package for Social Sciences version 24 ® (SPSS v24, SPSS Inc., Chicago, IL, USA). Categorical parameters were calculated as frequencies and percentages, while numerical variables were reported as means and standard deviations. The reliability of the WHOQOL HIV Bref was measured by Cronbach's α coefficient. Cronbach's α > 0.70 indicates a good consistency [32]. For test-retest analysis, an Intraclass Correlation Coefficient (ICC) was used with an ICC > 0.75; demonstrating high test-retest reliability [33].
Known group analysis was used to assess how well the Urdu version of the WHOQOL HIV Bref instrument differentiates between patients living with HIV concerning their HIV stages and CD-4T cell count. One-way Analysis of Variance (ANOVA) was performed to assess known-group validity. Post hoc tests were conducted to look at critical contrasts in domain scores among the three CD-4T-cell count /mm 3 (good CD-4T cell count (≥ 500), medium CD-4T cell count (200-499) and poor CD-4T cell count (≤ 200). It was hypothesized that HIVsymptomatic members and PLWHA with lower CD4 T Cell counts would have lower HRQoL scores mostly. Convergent validity was determined by measuring average variance extract (AVE), and composite reliability (CR) by using smart PLS inbuild command for all facets with minimal acceptable value of 0.5 and value > 0.7 indicates excellent validity [34,35]. Statistical examinations of all HRQoL items were conducted by utilizing inverted item scores, and a p value of < 0.05 (two-tailed test) was considered as a basis of measurable centrality for all analyses [36]. The independent associations between the domains and CD-4T-cell count were assessed using multivariable linear regression [37] . The construct validity was evaluated by Exploratory Factor Analysis (EFA) by considering Kaiser-Meyer Olkin and Bartlett's test of sphericity. The EFA validity was tried by extracting components via principal component analysis, taken after Varimax rotation with Kaisers' normalization [38].

Patient characteristics
Sociodemographic and health-related data of participants is given in Table 1. Of the total 182 patients, most respondents were males (n = 134, 73.6%), more than half (n = 104, 57.1%) fall in the age category of 25-50 years, and almost half of individuals (n = 90, 49.5%) were married or in a relationship. The majority of participants were illiterate (56%) but were able to understand and converse in the Urdu language. No PLWH had AIDS and 81% were asymptomatic. The majority of participants (53%) infected with HIV through intravenous drug use, while (16.5%) having sex with a male partner is the reason why they got the infection.

WHOQOL HIV Bref scale
The WHOQOL HIV-Bref domain scores were calculated following the user's manual of the WHOQOL-HIV instrument [30]. Questions showing the lowest scores were activities of daily living, sleep and rest, participation in leisure activities, meaningful life and social inclusion. Across the domains, physical health and psychological health showed the highest score; social relationships showed average scores while the level of independence showed the least count. The majority of items have skewness and kurtosis coefficients within the range of − 1.00 to 1.00. Only five items have a coefficient higher than 1.00, but still within the acceptable range of − 2.00 to 2.00 [39]. An item with the highest kurtosis (5.3) was the one measuring friends' social support accompanied by negative feelings (4.5) ( Table 2).

Reliability test
Reliability analysis revealed excellent reliability of the translated Urdu version of the scale with Cronbach's alpha value 0.934. All the six subscales exhibited remarkable internal consistencies ranging from Cronbach's alpha value of 0.931 for the physical health domain to 0.934 for the social relationship domain (Table 3).

Test re-test analysis
The test-retest reliability analysis was performed on 54 PLWH after two weeks gap. ICC value > 0.75 (p < 0.001) was considered significant [40]. Findings revealed statistically significant (p < 0.001) ICC value for all six domains  (Table 3).

Known group validity
Results revealed remarkable differences in the domain's mean scores stratified by CD4 T cell level. The results indicate patients with higher CD4 T cell count had higher mean scores of physical health, level of independence, psychological health, social relationship, environment health and spiritual health (p < 0.001) compared to patients with CD4 T cell level 200-499 cells/mm 3 and less than 200 cells/mm 3 . Post hoc analysis also demonstrated a statistically significant variation in mean scores for all six domains except spirituality domain; mean count did not vary substantially among patients with CD4 T-cell count less than 200 and patients with CD4 T-cell count 200-499 cells/mm 3 (Table 4). Five out of the six domain scores discriminated significantly among disease stages (asymptomatic and symptomatic). Asymptomatic patients had higher scores acros all subdomains except for spirtuality compared to symptomatic patients (Table 5).

Convergent validity
Results revealed that based on the Composite reliability (CR) value, all facets show good convergent validity as CR value is higher than 0.7 except for the social relationship domain 0.664 (0.458-0.755). Other parameters Average Variance Extracted (AVE) also demonstrates that all the facets exhibit acceptable convergent validity (Table 6).

Factor validity
Twenty-nine items of the WHOQOL HIV Bref instrument were analyzed for factor analysis. Kaiser-Meyer   Olkin and Bartlett's test of sphericity results were significantly acceptable. The analysis was acceptable as factor analysis of six domains with Eigen value above one represented 62.1% of the variance. First domain attributed to 34.5% of the variance. Furthermore, all six domains contain three or more items except for the fifth and sixth domains which include 2 and 1 items respectively. As recommended by Bryant et al. every domain should contain a minimum of three items; therefore, the model was modified to five domains [41]. Factor analysis of 5 domains accounted for 61% of the total variance. All unique areas were divided into a few aspects; however, similar regions were gathered, and stacked onto comparable components. For example, all items of the level of independence and physical health loaded into factor 1.
All factors with details are given in Table 8.

Discussion
The current study focused on the appraisal of reliability, validity and cross-cultural adaptation of the Urdu version of the WHOQOL HIV Bref instrument. Findings demonstrate that WHOQOL HIV Bref is a highly reliable and validated tool in the Urdu language. This is the first study that systematically translated and validated the 31-item instrument in Pakistan. Conceptually our study highlighted the high reliability of the original instrument developed by WHO. The none response rate in our study was less (2.6%) displayed good acceptability by PLWHA. Overall, Cronbach's alpha value is greater than 0.93 which indicates excellent internal consistency [42]. The percentage of the lowest and highest possible floor and ceiling scores of all domains in each domain was within an acceptable range, i.e. less than 10% [43]. Individual domains and each question's internal consistency was more than 0.9, which is highly acceptable, and these findings are in line with the Iranian, Brazilian and Chinese studies [12,18,25]. The skewness and kurtosis value of almost all the items fall within the acceptable range (− 1.00 to 1.00) which are consistent with the findings of the study conducted among Taiwanese and older Portuguese PLWHA [20,21]. Reliability coefficients (ICC values) of the Urdu version of the WHOQOL HIV Bref instrument ranging from 0.872 to 0.992; these findings are consistent with psychometric evaluation studies conducted in China and Malaysia [12,16,23]. The time interval was 12 to 16 days between the first and second ratings as proposed by Anne Anastasi and Susana Urbina [44]. The two weeks gap was selected to minimize the memory effect if the period is too short and that genuine differences in scores are not likely to have occurred if the period is too long [44].
Known group comparison shows the Urdu version of WHOQOL HIV Bref is a flexible tool for measuring the HIV care-related quality of life, as the questionnaire significantly differentiates btween patients with different levels of CD4 T cells; the higher the CD4 T cell levels, the better quality of life. These results are comparable to the findings of a study conducted in Chinese PLWH [12]. Of note, the difference in social relationship and spirituality facets (larger standard deviation) among PLWH with different HIV stages, indicates that though AIDS is a nonmortal disease at present, the level of affected emotional distress remains the same even at entirely different stage of HIV. This eplicates the reality that, despite HIV/AIDS is presently a treatable syndrome, however, does not represent a loss of life sentence. The participants have difficulty coping with the disease and managing related emotional distress regardless of their medical stage.
In addition, the social relationship factor may face similar challenges, and therefore interventional studies aimed at improving emotional and social well-being are needed to clarify this aspect. Along with other difficulties limited financial resources; the social aspect mainly affects PLWHA care and this is of particular importance in developing countries. Our study also showed that asymptomatic patients had better health-related outcomes compared to symptomatic patients.  an excellent ability to discriminate against the HIV stages [12,23,24]. Factor analysis (FA) of WHOQOL-HIV Bref showed a five-factor model structure which is in line with the Portuguese and Malaysian HIV patients [21,23]. This is found to be maximum about imbricate constructs between the factors due to the difference in perceptions and interpretations grounded in the Pakistani muslim culture. In addition to the Exploratory Factor Analysis (EFA), Canavarro et al. performed a WHOQOL-HIV Bref (CFA) jointly and found that each of the five-factor model and the original six-domain model worked reasonably well. [21]. Furthermore, multiple regression analyses from Peltzer et al. explained that only four domains (spirituality, psychological, surroundings, and level of independence) majorly predicted overall HRQoL [45]. Different studies also suggested that the model could be improved by changing some questions in a construct [21,24]. WHOQOL Group Instruments have been identified as the best tools in terms of simplicity and scope of response [46]. Our findings suggest that WHOQOL HIV enables doctors and patients both to discuss more sensitive questions and to focus discussions on patient-induced non-medical issues. Our study has shown that WHO-QOL-BREF is a sensitive tool for detecting PLWHA health variations in disease stages, viral load suppression and CD 4T lymphocytes count.

Limitations
This research has limitations that need to be noted. First, Pakistan is a country of great diversity and ethnicity. There are more than 180,000 patients of HIV but only 22,947 people are receiving the ARV free of cost from the government established ART centers. Thus, the results of this study do not apply to those who are not receiving therapy. Seconed, most of the investigation populace included were intravenous drug users who were in a lamentable environment or incapable of sticking to the treatment rules.
Third, all study patients were outpatients and who might already be in a good physical state and none of the study participants had AIDS. Therefore, a discriminative property of instrument was tested between the symptomatic and asymptomatic groups only. For this reason, it is recommended that longitudinal studies should be conducted to further differentiate between different clinical stages of HIV infection. Finally, we focused only on the traditional psychometric properties of the Urdu version of the WHOQOL. Future studies are needed to examine the measurement error and clinimetric features (e.g., clinical validity and utility) of this rating scale.

Conclusion
Current findings indicate that the Urdu version of WHO-QOL HIV Bref is a psychometrically valid and culturally adapted HRQoL assessment tool for PLWHA in Pakistan. ART centres in Pakistan could use this tool to evaluate the HRQoL of HIV patients and develop tailored interventions to improve health outcomes.