Adaptation and validation of WHODAS 2 . 0 in patients with musculoskeletal pain

OBJECTIVE: To validate the Portuguese version of the World Health Organization Disability Assessment Schedule (WHODAS 2.0). METHODS: The original, 36 item version of the WHODAS 2.0, administered through an interview, was translated into Portuguese following international guidelines and tested on 9 participants from the general population. The Portuguese version was then administered to 204 patients with musculoskeletal pain. The patients’ socio-demographic and health data were collected, as were the number of sites where they were experiencing pain and the intensity of that pain. The WHODAS 2.0 was administered again by a second interviewer within three days to determine its inter-rater reliability. Construct validity was assessed according to the ability of WHODAS 2.0 to discriminate between patients with different numbers of pain sites and the correlation between WHODAS 2.0 scores and pain intensity. Internal consistency was also assessed. RESULTS: The Portuguese version of the WHODAS 2.0 is easily understood and has good internal consistency (α = 0.84), as well as, very good inter-rater reliability (ICC = 0.95). In addition, it was able to detect statistically signifi cant differences between patients with different numbers of pain sites (p < 0.01) and showed that higher levels of disability are associated with more intense pain (r = 0.44, p < 0.01), indicating that it has construct validity. CONCLUSIONS: The Portuguese version of the WHODAS 2.0 has shown to be reliable and valid when administered to patients with musculoskeletal pain. DESCRIPTORS: Pain Measurement. Musculoskeletal Pain, classifi cation. Questionnaires, utilization. Translations. Reproducibility of Results. 2 Adaptation and validation of WHODAS 2.0 Silva C et al Assessing disability provides a detailed picture of the implications a health condition has on the day to day life of an individual. This is especially relevant as different profi les of disability can be associated with similar pathologies and diagnoses.1 Thus, an assessment which verifi es in what ways health conditions affect the individual’s daily activities is necessary. The International Classification of Functioning, Disability and Health (ICF)a defi nes disability as diffi culties in functioning in one or more areas of life from the perspective of the body, the individual and society, as experienced by an individual with a specifi c health condition in interaction with contextual factors. There are a variety of instruments for assessing disability, some specific to a certain health condition, some not, such as the Functional Independence Measure or the Barthel Index, among others. The World Health organization (WHO) developed the World Health Organisation Disability Assessment Schedule (WHODAS 2.0), which assesses perceived disability associated with the health condition in the 30 days preceding its application. This instrument is divided into six domains: i) cognition; ii) mobility; iii) self-care; iv) interpersonal relationships; v) activities of daily living; and vi) participation. This instrument enables the individual’s perception of their own disability to be evaluated.14 The WHODAS 2.0 is a generic instrument, not aimed at specifi c populations or specifi c health conditions. It has been translated and evaluated for use in various languages and cultures, both the original and the adapted versions proving to be psychometrically robust.8,13,15 Three versions of the WHODAS 2.0 have been developed, differing in length and method of administration: there is a version with 36 items, one with 12 items and one with 12+24 items. The fi rst two can be self-administered, administered in an interview and to a substitute respondent. The 12+24 item version has a version that can be administered in an interview or by computer.1 The aim of this study was to validate the Portuguese version of the WHODAS 2.0.

Assessing disability provides a detailed picture of the implications a health condition has on the day to day life of an individual.This is especially relevant as different profi les of disability can be associated with similar pathologies and diagnoses. 1Thus, an assessment which verifi es in what ways health conditions affect the individual's daily activities is necessary.
The International Classification of Functioning, Disability and Health (ICF) a defi nes disability as difficulties in functioning in one or more areas of life from the perspective of the body, the individual and society, as experienced by an individual with a specifi c health condition in interaction with contextual factors.There are a variety of instruments for assessing disability, some specific to a certain health condition, some not, such as the Functional Independence Measure or the Barthel Index, among others.The World Health organization (WHO) developed the World Health Organisation Disability Assessment Schedule (WHODAS 2.0), which assesses perceived disability associated with the health condition in the 30 days preceding its application.This instrument is divided into six domains: i) cognition; ii) mobility; iii) self-care; iv) interpersonal relationships; v) activities of daily living; and vi) participation.This instrument enables the individual's perception of their own disability to be evaluated. 14The WHODAS 2.0 is a generic instrument, not aimed at specifi c populations or specifi c health conditions.It has been translated and evaluated for use in various languages and cultures, both the original and the adapted versions proving to be psychometrically robust. 8,13,15Three versions of the WHODAS 2.0 have been developed, differing in length and method of administration: there is a version with 36 items, one with 12 items and one with 12+24 items.The fi rst two can be self-administered, administered in an interview and to a substitute respondent.The 12+24 item version has a version that can be administered in an interview or by computer. 1e aim of this study was to validate the Portuguese version of the WHODAS 2.0.

METHODS
The process of translating the original WHODAS 2.0 was carried out according to internationally established guidelines 3 involving the following steps:   1).The WHODAS 2.0 areas in which the participant reported the most disability were in activities of daily living, mobility and participation (Table 2).
Participants were interviewed twice at the rehabilitation clinics where they had been admitted.In the fi rst interview, demographic and health data were collected and information on the number of sites where pain was felt and the intensity of that pain, and the 36 item WHODAS 2.0 was administered.In the second interview, the WHODAS 2.0 was administered.
Demographic data were collected (age, sex, marital status, years of schooling, occupation, height and weight).The values for height and weight were self-reported, wich is a valid method of characterizing the prevalence of being overweight and obesity. 4he participants were asked whether they had been diagnosed with any of the following pathologies: high blood pressure, diabetes, arthritis, cardiovascular disease, respiratory disease, cancer, depression or other health problem.
The participants were asked to indicate on a body chart showing the whole body, where they had felt any pain in the preceding week.The number of sites with pain were counted and categorized according to Picavet et al 11 into: pain in one site; pain in two or three sites; pain in four or more sites; widespread pain (pain above and below the waist, pain on the left or right side of the body, and axial pain).
The intensity of the pain was assessed using a 10 cm visual numerical scale, showing a vertical classifi cation from zero (no pain) to ten (the worst pain imaginable).
Those participants who experienced pain in more than one site were asked to evaluate the mean intensity of the pain for the various sites over the preceding week (overall intensity).Pain assessment was carried out following international guidelines for assessing pain in the elderly.c The 36 item version of the WHODAS 2.0, translated and adapted into European Portuguese using the methodology described above, was administered in two interviews by two different interviewers, with minimum and maximum intervals of one and three days, respectively.The repetition of the WHODAS 2.0 aimed to assess the inter-evaluator reliability.
Limiting the interval between the two applications aimed to minimize the possibility of changes in the profi le of the patient's disability, which would have compromised the assessment of reliability.
The data analysis was carried out using the Statistical Package for the Social Sciences (SPSS).The sample was characterized using descriptive statistics (mean, standard deviation and frequencies).The WHODAS 2.0 scores were calculated per domain by adding the scores in each item in the domain and the total score was turned into a value between 0 and 100, as described in the WHODAS 2.0 manual.The higher the score, the greater the disability.Total scores were calculated taking into account the 36 items or taking into account all items except those referring to domain 5.2 "daily living activities" -work/school (total 32 items), when participants neither worked nor studied. 15ternal consistency was assessed using Cronbach's alpha (α), which varied between 0 and 1, with: (α < 0.6) "unacceptable"; (0.6 ≤ α < 0.7) "weak"; (0.7 ≤ α < 0.8) "reasonable"; (0.8 ≤ α < 0.9) "good"; and (α ≥ 0.9) "very good". 10Inter-evaluator reliability was assessed using the intra-class coeffi cient of correlation (ICC), which ranged between 0 and 1, considering "weak" (ICC < 0.4); "satisfactory" (0.4 ≤ ICC < 0.75); and "very good" (ICC ≥ 0.75). 12Differences in the WHODAS 2.0 score between groups with different number of pain sites were assessed using ANOVA and the Bonferroni test.
Correlation between WHODAS 2.0 and the intensity of pain was assessed using Pearson's coeffi cient of correlation.The level of signifi cance was p < 0.05.
This study was approved by the Bioethics and Medical Ethics Service of the Faculty of Medicine of the Universidade do Porto, Portugal, 2011.Participants signed a consent form.

RESULTS
The overall Cronbach's alpha value for the 36 item version of the WHODAS 2.0 was 0.84, indicating good internal consistency.Due to non-applicable responses in domain 5 "activities of daily living" by the rest of the participants, 65 participants were considered in this analysis.The Cronbach's alpha value decreased with the removal of most of the items, indicating their importance in evaluating disability.Table 3 shows the results of the Cronbach's alpha test for all WHODAS 2.0 domains and the value obtained when each of these domains was removed.
The ICC was > 0.80 for all areas of the WHODAS 2.0, indicating good inter-evaluator reliability.The ICC was 0.95, taking into account the total WHODAS 2.0 score, which also indicates good inter-evaluator reliability (Table 4).
Construct validity was assessed by comparing the participants' level of disability with the different number of pain points and by assessing the link between disability and intensity of the pain, as the literature indicates that disability increases with the number of points at which pain is felt and with the intensity of that pain. 5,9,14Statistically signifi cant differences were found in the total WHODAS 2.0 scores between participants with widespread pain (mean = 35.7;SD = 13.3) and those with pain in only one spot (mean = 24.1;SD = 28.4,p = 0.009) and in two or three sites (mean = 29.0;SD = 14.1; p = 0.003), indicating greater incapacity in participants with widespread pain.The intensity of pain in the preceding week was 5.9 (SD = 2.0), verifying a statistically signifi cant correlation between pain intensity and the WHODAS 2.0 (total and domain scores) indicating greater disability in participants with more intense pain (Table 4).

DISCUSSION
The 36 item Portuguese version of the WHODAS 2.0, administered in interview form, was deemed to be equivalent to the original version and easy to understand.The results of the psychometric properties evaluation showed the instrument has good internal consistency (α = 0.84).However, in studies by Baron et al, 2 Garin et al 7 and Ustun et al, 15 the alpha value was above 0.95.This difference may be explained by the small number of participants considered in this analysis (n = 65), due to non-applicable responses in domain 5 "activities of daily living: home and work or school" by the rest of the participants.A large number of participants reported not working or going to school and a large number of male participants reported not doing housework.The ICC for inter-evaluator reliability was 0.95, indicating good inter-evaluator reliability.This fi gure was higher than that obtained by Schlote et al, 13 who reported ICC of 0.67.However, the results of Schlote et al 13 may be explained by the long interval (six months) between the two WHODAS 2.0.The WHODAS 2.0 has been shown capable of fi nding logical relationships supported in the literature.In this study, individuals with widespread pain had lower functionality than those with pain in one spot or in two or three spots.More intense pain also seemed to be associated with lower functionality.Both fi ndings are in concordance with the results of previous studies 5,6,9,14 and suggest that the instrument has construct validity.
The results indicate that the 36 item Portuguese version of the WHODAS 2, administered in interview form, is valid and reliable.These results are in agreement with those presented by a variety of studies carried out using the original version on patients with infl ammatory arthritis, knee osteoarthritis, chronic disease, osteoarthritis and strokes. 2,7,8,13e of the limitations of this study is that an disability measure validated in Portuguese, which would have allowed these results to be compared with those obtained by applying the WHODAS 2.0, was not used.One of the instruments used in the process of validating the WHODAS 2.0 in other countries is the health status questionnaire (SF 36).Although the SF36 and the WHODAS 2.0 assess different aspects of related concepts, as the former assesses quality of life related to health and the latter limitations to day to day activities and restrictions to participation, studies show that a moderate association exists between the scores from the two instruments.Therefore, using this instrument would have enabled criterion validity to be assessed, making the validation of this instrument more robust.Not using patients with diverse pathologies limits the possibilities of generalizing the results.Thus, in the future, it is necessary to assess this instrument for use in patients with other types of health conditions.

Table 1 .
Characterization of the sample used to validate the WHODAS 2.0, 2011.

Table 2 .
Mean score and standard deviation for the sample for each of the WHODAS 2.0 domains and the total mean score, 2011.The total score was calculated without the values for area 5.2, as 63% of the participants reported they were not working. b

Table 3 .
Cronbach's alpha results by WHODAS 2.0 domain and for the total score, 2011.

Table 4 .
Results for inter-evaluator reliability and correlation between WHODAS 2.0 and the intensity of pain experienced per domain and for the total score, 2011.World Health Organization Disability Assessment Schedule 2.0 a Some participants reported not doing housework and/or working b p < 0.01