Research in Developmental Disabilities 35 (2014) 2812–2820

Contents lists available at ScienceDirect

Research in Developmental Disabilities

Development of traditional Chinese version of World Health

Organization Disability Assessment Schedule 2.0 36 – item(WHODAS 2.0) in Taiwan: Validity and reliability analyses

Tzu-Ying Chiu a, Chia-Feng Yen a,b,*, Cheng-Hsiu Chou c, Jin-Ding Lin d,Ai-Wen Hwang e, Hua-Fang Liao f, Wen-Chou Chi g

a Institute of Medical Science, Tzu Chi University, Hualien, Taiwanb Department of Public Health, Tzu Chi University, Hualien, Taiwanc Departments of Family Medicine, Hualien Armed Forces General Hospital, Hualien, Taiwand School of Public Health, National Defense Medical Center, Taipei, Taiwane College of Medicine, Chang Gung University, Graduate Institute of Early Intervention, Taoyuan, Taiwanf Chinese Association of Early Intervention Profession for Children with Developmental Delays, Taipei, Taiwang Department of Physical Medicine and Rehabilitation, Shuang Ho Hospital, Taipei, Taiwan

A R T I C L E I N F O

Article history:

Received 8 April 2014

Received in revised form 29 June 2014

Accepted 2 July 2014

Available online

Keywords:

WHODAS 2.0

ICF

Traditional chinese version

Disability identification

Validity and reliability

A B S T R A C T

Background: World Health Organization Disability Assessment Schedule 2.0 (WHODAS

2.0) provided a standardized method for measuring the health and disability and the

traditional Chinese version has not been developed.

Aims: To describe the process of developing the traditional Chinese version of the

WHODAS 2.0 36-item version and to evaluate the concurrent validity and test–retest

reliability of this instrument.

Methods: The study was conducted in two phases. Phase I was the process of translation of

WHODAS 2.0 36-item version. Phase II was a cross-sectional study. The participants were

307 adults who were tested the validity and reliability of draft traditional Chinese version.

Results: The reliability of Cronbach’s a and ICC in the WHODAS 2.0 traditional Chinese

version were 0.73–0.99 and 0.8–089, respectively. The content validity was good (r = 0.7–

0.76), and the concurrent validity was excellent in comparison with the WHOQOL-BREF

(p< 0.5). The construct validity, the model was explained total variance was 67.26% by the

exploratory factor analysis (EFA) and the confirmatory factor analysis (CFA) illustrated the

traditional Chinese version was good to assess disability. There was a valid and reliable

measurement scales for evaluating functioning and disability status.

Conclusion: For disability eligibility system of Taiwan government to measure the

disability, the traditional Chinese version of the WHODAS 2.0 provided valuable evidence

to design the assessment instrument.

� 2014 Elsevier Ltd. All rights reserved.

* Corresponding author at: Department of Public Health, Tzu-Chi University, Taiwan. Tel.: +886 3 856 5301x2300; fax: +886 3 856 4041.

E-mail address: mapleyeng@gmail.com (C.-F. Yen).

http://dx.doi.org/10.1016/j.ridd.2014.07.009

0891-4222/� 2014 Elsevier Ltd. All rights reserved.

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–2820 2813

1. Introduction

The disability burden issue has received much attention in developing and developed countries; however, comprehensiveestimates of the economic and social costs of disability are rare and fragmented. The World Health Survey and the GlobalBurden of Disease are based on markedly different measurement approaches and assumptions, and they yield globalprevalence estimates among the adult population of 15.6% and 19.4%, respectively. The estimates of the extra costs resultingfrom disabilities range from 11% to 69% of income in the UK, 29% to 37% in Australia, 20% to 37% in Ireland, 9% in Viet Nam,and 14% in Bosnia and Herzegovina (Gresham, Phillips, & Labi, 1980; Harwood, Rogers, Dickinson, & Ebrahim, 1994; Hobart &Thompson, 2001; Hsueh, Lee, & Hsieh, 2001; Katz, 2003; Lawton & Brody, 1969; Rockwood, Joyce, & Stolee, 1997). Mostdeveloping countries report disability prevalence rates below those reported in many developed countries because the dataare collected on a narrow set of impairments or on a medical model, which yield lower disability prevalence estimates. Agrowing number of countries are using the International Classification of Functioning, Disability and Health (ICF) frameworkwhich combined the biomedical and social models for disabilities and health conditions in their national surveys andcensuses of disability. In Taiwan, to enhance the social participation of people with disabilities, the People with DisabilitiesRights Protection Act that was promulgated in 2007 regulated that since July 2012 the disability evaluation must be based onthe ICF framework (MOHW, 2007) However, the ICF is a classification system of health and disability rather than anevaluation or measurement tool. Consequently, the World Health Organization Disability Assessment Schedule 2.0(WHODAS 2.0) was developed based on the concept of the ICF in 2001 regarding the general evaluation and measurement ofhealth conditions, disabilities, and psychometric variables (Yen et al., 2014).

WHODAS 2.0 provided a standardized method for measuring the health and disability of adults (age> 18 years) acrosscultures which contains six domains: cognition, mobility, self-care, getting alone with people, activities of daily life andsocial participation. There are various modes and versions of the WHODAS 2.0 for different purposes, including self-administration, interviewer, proxy mode, 36- item, 12-item and 12 + 24 item versions. The WHODAS 2.0 was translated intomore than 30 languages from 2001 to 2010, including the following: Albanian, Arabic, Bengali, simplified Chinese, Croatian,Czech, Danish, Dutch, English, Finnish, French, German, Greek, Hindi, Italian, Japanese, Kannada, Korean, Norwegian,Portuguese, Romanian, Russian, Serbian, Slovenian, Spanish, Sinhala, Swedish, Tamil, Thai, Turkish and Yoruba. (Cheng et al.,2012; Federici, Meloni, Mancini, Lauriola, & Olivetti Belardinelli, 2009; Kutlay et al., 2011; Tazaki, Yamaguchi, Yatsunami, &Nakane, 2014; Ustun, Kostanjsek, Chatterji, & Rehm, 2010) Globally, the WHODAS 2.0 was found to be useful for assessinghealth and disability levels in the general population through surveys and for measuring the clinical effectiveness andproductivity gains from interventions, including osteoarthritis patients, schizophrenia patients, stroke patients, chronicdiseases patients and rheumatoid arthritis patients (Almazan-Isla et al., 2014; de Pedro-Cuesta et al., 2011; Garin et al., 2010;Guilera et al., 2012; Kutlay et al., 2011; Mas-Exposito, Amador-Campos, Gomez-Benito, & Lalucat-Jo, 2012; Meesters,Verhoef, Liem, Putter, & Vliet Vlieland, 2010; Novak, Colpe, Barker, & Gfroerer, 2010; Scott et al., 2009; Sousa et al., 2010). Thismultidimensional instrument is simple to use and is a more comprehensive disability measurement (Ustun et al., 2010).

For comprehensive understanding and assessing the limitations in activity and social participation of people withdisabilities, it is must to develop the traditional Chinese Version of the WHODAS 2.0. Generally, the completed steps ofinstrument development are including forward, backward translation, reliability and validity establishment. According toBaron and Chisolm studies, the period of test–retest of WHODAS 2.0 was 7–14 days (Baron et al., 2008; Chisolm, Abrams,McArdle, Wilson, & Doyle, 2005), and the concurrent validity was administered with other known instruments, such as SF-36, SF-12, WHOQOL-BREF, LHS and FIM. Among these instruments, WHOQOL-BREF is the most widely to measure the qualityof life and to be compared with WHODAS 2.0 (Ustun et al., 2010). Therefore, the purposes of this study were to develop theWHODAS 2.0 36-item traditional Chinese version, which included describing the processes of the back translation andforward translation of the original WHODAS 2.0 36-item version, and to test the concurrent validity and the test–retestreliability of the WHODAS 2.0 36-item traditional Chinese version.

2. Methods and design

This study included two phases. Phase I was during August to December 2012, the WHODAS 2.0 was translated intotraditional Chinese by five fluent Chinese translators and one English expert translator who were native speakers oftraditional Chinese, including an OT, PT, public health professional and clinical doctor. The back translation of the traditionalChinese version was translated into English by one bilingual translator (a native speaker of English) with no previousknowledge of the WHODAS 2.0. Phase II was a cross-sectional design of data collection to test the concurrent validity and thetest–retest reliability during December 2012 to October 2013 in Taiwan (Fig. 1). The test–retest period was 7–14 days (Baronet al., 2008; Chisolm, Abrams, McArdle, Wilson, & Doyle, 2005). The interviewers were well trained of five hours inprofessional training courses before interview, and this study was approved by the Buddhist Tzu Chi General HospitalResearch Ethics Committee (IRB101-97).

2.1. Participants

In Phase I, we invited five expert native speakers of traditional Chinese for the forward translation and one native speakerof English for the backward translation. Phase II included 307 participants who were older than 18 years old and living in

[(Fig._1)TD$FIG]

Forward translation by native experts of traditional Chinese

including an OT, PT, public health expert and clinical doctor.

Back translation by one native speaker of English

Group discussion and revise to formal version

Group discussion and revise

Testing of reliability and validity

2012/08

2012/12

2013/10

Phase I Phase II

Fig. 1. Stages of developing the WHODAS 2.0 36-item traditional Chinese version.

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–28202814

elderly or disability institutions located in three cities in the northern (New Taipei City), eastern (Hualien City), and southernregions (Ciayi City) of Taiwan. The data were collected by face-to-face interviews from December 2012 to October 2013.

Table 1 shows the participants characteristics in Phase II. There were 53.1% female and 46.9% male whose average age was47.56� 22.26 years old. The most percentage of highest educational level was senior high school (36.4%). Regarding disability,79.5% of the subjects had a disability certification in Taiwan. The most disabled type of them was intellectual disability (ID)(68.85%). The disabled severity levels were mild 24.68%, moderate 32.77%, severe 30.21% and profound 12.34% among thesamples.

2.2. Instruments

Following Phase I, the draft of the traditional Chinese version of the WHODAS 2.0 36-item scales was administered (PhaseII). We used the World Health Organization Quality of Life brief vision (WHOQOL-BREF) in the Phase II.

2.3. WHODAS 2.0 36-item version

The WHODAS 2.0 36-item version aims to measure the function of activity and participation in daily living within theprevious 30 days, including the following six domains: (1) cognition (six items): assesses communication and thinkingactivities; including concentrating, remembering, problem solving, learning and communicating, (2) getting around (fiveitems): assesses activities such as standing, moving around inside the home, getting out of the home and walking a longdistance, (3) self-care (four items): assesses hygiene, dressing, eating and staying alone., (4) getting alone with people (fiveitems): assesses interactions with other people and difficulties that might be encountered with this life domain due to ahealth condition. (5) life activities (household and school/work, eight items): assesses difficulty with day-to-day activities(i.e. those that people do on most days, including those associated with domestic responsibilities, leisure, work and school)and (6) participation (eight items): assesses social dimensions, such as community activities; barriers and hindrances in theworld around the respondent; and problems with other issues, such as maintaining personal dignity. The response of eachitems are no difficulty, mild difficulty, moderate difficulty, severe difficulty and extreme difficulty. The total score range was0–100, with a higher score indicating higher limitation in daily life. The participants were asked to answer all 36 items exceptfor the items regarding employment or studying (32 items).

2.4. WHOQOL-BREF

The WHOQOL-BREF of traditional Chinese version comprises 28 items on the quality of life within the previous 14 days,including the following four domains: physical health (including activities of daily living, dependence on medicinalsubstances and medical aids, energy and fatigue, mobility, pain and discomfort, sleep and rest and work capacity, 7 items),

Table 1

The demographic characteristics of the participants (n = 307).

Variable n % Variable n %

Gender (n = 307) Disability certification (n = 303)

Male 144 46.9 No 62 20.5

Female 163 53.1 Yes 241 79.5

Age (n = 307) 47.56� 22.26 Disability type (n = 235)

18–20 9 2.9 Intellectual 143 60.85%

21–30 90 29.3 Multiple 45 19.15%

31–40 59 19.2 Limb 32 13.62%

41–50 27 8.8 Hearing 5 2.13%

51–60 19 6.2 Vision 2 0.85%

61–70 32 10.4 Primary organs 2 0.85%

71–80 38 12.4 Balancing 1 0.43%

81–90 29 9.4 Chromosomal abnormality 1 0.43%

>91 4 1.3 Dementia 1 0.43%

ADL (n = 138) 72.06� 34.64 Autism 1 0.43%

Education (n = 291) Mentally 1 0.43%

Illiterate 41 14.1 Severity of disability (n = 235)

Elementary 75 25.8 Mild 58 24.68%

Junior 67 23.0 Moderate 77 32.77%

Senior 106 36.4 Severe 71 30.21%

Junior college 2 .7 Profound 29 12.34%

Aboriginal (n = 297) Catastrophic illness card (n = 282)

No 246 82.8 No 258 91.5

Yes 51 17.2 Yes 24 8.5

Low-income

situation (n = 294)

Have been accepted new disability

eligibility system (n = 262)

No 216 73.5 No 258 98.5

Yes 78 26.5 Yes 4 1.5

Health status (n = 290)

Hypertension 66 22.76%

Epilepsy 29 10.00%

Heart related disease 25 8.62%

Diabetes 24 8.28%

Stroke 20 6.90%

Physical related disease 12 4.14%

Renal related disease 6 2.07%

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–2820 2815

psychological health (including bodily image and appearance, negative feelings, positive feelings, self-esteem, spirituality/religion/personal beliefs, thinking, learning, memory and concentration, 6 items), social relationships (including Personalrelationships, social support and sexual activity, 3 items), environment (including financial resources, freedom, physicalsafety and security, health and social care, home environment, opportunities for acquiring new information and skills,participation in and opportunities for recreation/leisure activities, physical environment (pollution/noise/traffic/climate)and transport, 8 items), overall quality of life (2 items) and 2 items of country-specific quality of life questions. The scorerange is 0–100, with a higher score indicating a better quality of life.

2.5. Statistical analysis

All of the data were analyzed by Windows SPSS 20.0 and AMOS 20.0, and the significance level was assumed as 0.05. Thedata were excluded if the missing responses were more than 50% in every domains and the others using mean imputation, allanswers ranged from 0 to 4 of the WHODAS 2.0 (Garin et al., 2010).

We investigated the distribution of characteristics of the participants by descriptive analysis and the score range of thedomains by the ceiling effect and floor effect. In the study, the ceiling effect was defined as the proportion of scores of 100,and the floor effect was the proportion of scores of 0 (Guilera et al., 2012; Schlote et al., 2009). The indicators of reliabilityinclude internal consistency (Cronbach’s y) and test–retest (Intra-Class Correlation Coefficient, ICC). The validityassessments include concurrent validity, content validity and construct validity. The concurrent validity was assessed bymeasuring the association of the WHODAS 2.0 and the WHOQOL-BREF; the content validity examined the relationshipamong the WHODAS 2.0 domains by Pearson’s correlation coefficient. The factor structure of the WHODAS 2.0 was examinedby exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). In the EFA, we first tested the application offactor analysis according to Kaiser Meyer Olkin (KMO) and the Bartlett’s test. Second, we identified the underlyingrelationships between the measured variables of the WHODAS 2.0. In the CFA, we confirmed the factor structure by the RootMean Square Error Approximation (RMSEA< 0.05), Comparative Fit Index (CFI> 0.9), and Goodness of Fit Index (GFI> 0.9)indicators.

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–28202816

3. Results

3.1. Phase y – forward and backward translation

After completion of the forward translation and back translation, the experts met to confirm the quality of the translation,including the clarity of the wording, to clarify the conceptual framework and to ensure the meanings of the sentences.Finally, we only revised the items 4.5 sexual activities based on cultural consideration. The minor change of item 4.5 is formsexual activities to sexual related activities. Other items of WHODAS 2.0 were same as before.

3.2. Phase II – validity and reliability of the WHODAS 2.0 36-item traditional Chinese version

3.2.1. The WHODAS 2.0 scores of samples

The average scores of each domain in the WHODAS 2.0 traditional Chinese version were as follows: 17.64 (�26.31) in lifeactivity; 16.8 (�23.99) in mobility; 16.69 (�23.06) in getting along; 15.91 (�14.74) in participation; 15.72 (�18.99) in cognition;and,12.08 (�16.75) in self care and the total summary score was 14.72 (�13.52). The highest proportion of missing data was in thegetting along domain, especially for item 4.5 (sexual activities).

3.2.2. Ceiling, floor effect and reliability

The total floor effect was 7.2% and ceiling effect was 0. The ceiling effect in every domain was less than 0.3% and flooreffect in every domain were over 20.2%. The reliability of Cronbach’s y were 0.73–0.99, and the ICC values were 0.8–0.89 intraditional Chinese version of WHODAS 2.0 (Table 2).

3.3. Validity

Tables 3 and 4 shows the positive correlation between the six domains of the WHODAS 2.0 36-item traditional Chineseversion (p< .05), and the summary score had the greatest correlation coefficient with each domain (r = 0.7–0.76), showinggood content validity. Regarding the concurrent validity, there were negative correlations between each domain in theWHODAS 2.0 36-item traditional Chinese version and the WHOQOL-BREF (p< 0.5), except in the domain of cognition in theWHODAS 2.0 and environment in the WHOQOL-BREF (p> 0.5). Table 5 shows the results of the construct validity after EFA,divided into seven factors, and the model explained the total variance was 67.26% (KMO 0.87, Bartlett’s test p< 0.05). Almostall of the items of the traditional Chinese version were attributed to their own original dimension; only three itemsduplicated others, including the following: item 6.1, Joining in community activities; item 3.4, Staying by yourself for a fewdays; and item 6.3, Living with dignity. Regarding the factor loading, the domains of cognition, mobility, self care, gettingalone with people, and life activity items were all over 0.5 (Table 5). The CFA results presented in Fig. 2 show a factor loadingrange of from 0.66 to 0.77 in the six domains of the WHODAS 2.0. The fit indicators were less accepted in CFI (0.81), GFI (0.76)and RMSEA (0.091).

Table 2

Distribution and reliability of the WHODAS 2.0 and the WHOQOL-BREF.

Domain Mean� SD Mid Range Missing

(%)

Floor 0

score n

(%)

Ceiling 100

score n (%)

Cronbach’s

a

Test–retesta

WHODAS 2.0

Do1.Cognition 15.72� 18.99 5 0–85 4 (1.3) 115 (37.5) 0 0.84 0.87

Do2.Mobility 16.8� 23.99 6.25 0–100 5 (1.6) 144 (46.9) 4 (1.3) 0.88 0.87

Do3.Self-care 12.08� 16.75 10 0–100 3 (1) 137 (44.6) 2 (0.7) 0.82 0.83

Do4.Getting along 16.69� 23.06 8.33 0–100 256 (83.4) 147 (47.9) 2 (0.7) 0.88 0.87

Do5-1.Life activities:

household

17.64� 26.31 10 0–100 5 (1.6) 147 (48.2) 10 (3.3) 0.95 0.87

Do5-2.Life activities:

work and school

task (n = 184)

10.29� 16.35 0 0–78.57 0 107 (58.2) 0 0.99 0.89

Do6.Participation 15.91� 14.74 12.5 0–100 30 (9.7) 62 (20.2) 1 (0.3) 0.73 0.89

Summary score 14.72� 13.52 11.32 0–66.98 263 (85.4) 22 (7.2) 0 0.91 0.80

WHOQOL-BREF

Physical 67.58� 16.15 67.86 7.14–100 18 (5.9) 0 8 (2.6) 0.76 X

Psychological 68� 16.63 70.83 8.33–100 11 (3.6) 0 10 (3.3) 0.74 X

Social Relationships 68.28� 20.05 75 0–100 221 (72) 6 (2) 30 (9.9) 0.82 X

Environment 71.70� 13.50 71.88 15.63–100 6 (1.9) 0 11 (3.6) 0.73 X

a Concurrent validity were 307 participants and test retest were 100 participants in the study.

Table 4

The correlation of the WHODAS 2.0 and the WHOQOL-BREF (n = 307).

WHOQOL-BREF WHODAS 2.0

Do1.

Cognition

Do2.

Mobility

Do3.

Self-care

Do4. Getting

along

Do5. Life

activities

Do6.

Participation

Summary

score

Physical �.18*** �.53*** �.33*** �.33*** �.39*** �.35*** �.46***

Psychological �.15*** �.34*** �.22*** �.34*** �.30*** �.22*** �.34***

Social Relationships �.17*** �.31*** �.25*** �.37*** �.26*** �.20*** �.33***

Environment �.11 �.28*** �.12* �.29*** �.21*** �.26*** �.27***

* p< .05.

*** p< .001.

Table 3

Correlation of the WHODAS 2.0 (n = 307).

WHODAS 2.0 Do1.

Cognition

Do2.

Mobility

Do3.

Self-care

Do4. Getting

along

Do5. Life

activities

Do6.

Participation

Summary

score

Do1.Cognition 1 .37*** .52*** .60*** .43*** .39*** .76***

Do2.Mobility 1 .62*** .44*** .61*** .36*** .73***

Do3.Self-care 1 .51*** .57*** .45*** .76***

Do4.Getting along 1 .44*** .47*** .76***

Do5.Life activities 1 .40*** .75***

Do6.Participation 1 .70***

Summary score 1

*** p< .001.

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–2820 2817

4. Discussion

4.1. Missing values in the WHODAS 2.0 36-item traditional Chinese version

The highest proportion of missing domain was in Domain 4, getting along of our study. Many studies illustrated that theproportion of missing domains was varied and the most of them showed that the trait of highest missing proportion wasitem 4.5, sexual activity (Federici et al., 2009; Garin et al., 2010; Guilera et al., 2012; Schlote et al., 2009; Zhao et al., 2012). Thepossible causes were the sexual activity was lacking because of living type and the issue of sex is private for Asia culture.

4.2. Ceiling and floor effect

The top three domains with the floor effect (the proportion of scores of 0) were ‘‘work and school task’’, ‘‘household’’ and‘‘getting along’’. The result differed from those of other studies; studies in Spain and Italy showed that the main floor effectdomains were those of self care, getting around, or getting alone with people (Federici et al., 2009; Garin et al., 2010; Guileraet al., 2012). It might be the occupational training classes provided by some institutions for our participants, the floor effect indomain 5: work and school tasks were higher than in other studies. In above studies, all of them get the higher proportion of 0score, it might be that the instrument of WHODAS 2.0 was not sensitive to detect the difference in highly functioning ofpeople or subjects in the study did the excellent performance in daily life. In the present study, the summary score of Chineseversion was 7.0% that means the summary scores of WHODAS 2.0 traditional Chinese version get little floor effect, it is alsoshowed that people with disabilities get a floor effect in every domains, but in summary scores of WHODAS 2.0 traditionalChinese version will integrate peoples performance of dairy life. It proved that WHODAS 2.0 traditional Chinese version cancombine the varied aspects of function.

4.3. Internal consistency and test–retest reliability

The reliability of Cronbach’s a in the WHODAS 2.0 traditional Chinese version was 0.73–0.99, and the results were similarwith those of other studies (Federici et al., 2009; Garin et al., 2010; Guilera et al., 2012; Kutlay et al., 2011; Mas-Expositoet al., 2012; Meesters et al., 2010; Schlote et al., 2009; Wolf et al., 2012), which confirmed that the WHODAS 2.0 was a validand reliable measure even in the WHODAS 2.0 traditional Chinese version. The WHODAS 2.0 traditional Chinese versiontest–retest reliability was 0.83–0.89, which indicated that the WHODAS 2.0 traditional Chinese version shows higherstability over time in every domain. The ICC value was better and had more consistency than was found in other studies. Astudy in Spain and seven European studies have ICC ranges of 0.63–0.88 and 0.20–0.69, respectively (Garin et al., 2010;Guilera et al., 2012). Few studies have examined the ICC of WHODAS 2.0 from the perspective of establishing reliability andvalidity, and this step in developing the WHODAS 2.0 traditional Chinese version was essential.

Table 5

The exploratory factor analysis (EFA) of the WHODAS 2.0 traditional Chinese version (n = 307).

1 2 3 4 5 6 7

D4.5 Sexual activities 0.82 0.09 0.18 0.25 0.19 0.07 0.00

D4.4 Making new friends 0.80 0.10 0.14 0.26 �0.01 0.06 0.00

D4.3 Getting along with people close to 0.75 0.15 0.17 0.11 0.08 0.11 0.10

D4.2 Maintaining a friendship 0.75 0.20 0.14 0.24 0.13 0.11 0.06

D4.1 Dealing with strangers 0.54 0.05 0.06 0.37 0.22 0.05 0.04

D6.1 Joining in community activities 0.38 0.28 0.29 0.23 0.30 0.17 0.03

D5.3 Do all needed household work 0.13 0.89 0.23 0.11 0.17 0.10 0.00

D5.2 Do important household tasks well 0.18 0.88 0.22 0.10 0.15 0.10 �0.03

D5.1 Household responsibilities 0.13 0.85 0.18 0.14 0.13 0.10 �0.03

D5.4 Household work performed as quickly as needed 0.14 0.82 0.28 0.17 0.09 0.07 �0.05

D2.5 Walking a long distance 0.14 0.22 0.79 0.17 0.07 �0.02 0.06

D2.1 Standing for long periods 0.18 0.22 0.76 0.10 0.22 0.03 0.04

D2.3 Moving around inside home 0.07 0.22 0.71 0.08 0.39 �0.01 �0.08

D2.2 Standing up from sitting 0.20 0.22 0.66 �0.07 0.34 0.09 0.01

D2.4 Getting out of home 0.28 0.30 0.61 0.17 0.18 0.13 0.01

D1.3 Problem-solving 0.16 0.08 0.09 0.79 �0.06 0.11 0.02

D1.5 Understanding 0.07 0.19 �0.12 0.77 0.26 0.15 0.12

D1.6 Conversation 0.27 0.08 0.02 0.66 0.24 0.04 0.27

D1.2 Remembering to do important things 0.15 0.09 0.19 0.65 �0.02 0.08 �0.38

D1.4 Learning a new task 0.38 0.14 0.15 0.61 0.15 0.06 �0.09

D3.4 Staying by self for a few days 0.22 0.14 0.38 0.53 0.01 0.16 0.13

D1.1 Concentration 0.26 0.06 0.27 0.52 �0.01 0.16 �0.49

D3.2 Getting dressed 0.13 0.18 0.29 0.13 0.83 0.12 �0.09

D3.3 Eating 0.17 0.11 0.23 0.11 0.82 0.03 0.00

D3.1 Washing your whole body 0.21 0.27 0.37 0.14 0.71 0.07 �0.05

D6.7 Health affects family 0.03 0.19 �0.09 0.17 0.06 0.78 0.07

D6.6 Health affects family finances 0.00 0.15 �0.11 0.15 0.10 0.78 0.06

D6.5 Health affects one’s emotions 0.10 �0.09 0.35 0.13 �0.19 0.60 0.07

D6.8 Doing things for relaxation or pleasure 0.42 0.18 0.23 �0.03 0.18 0.46 0.00

D6.4 Health affects time consumption 0.37 0.02 0.06 �0.06 0.03 0.43 �0.16

D6.2 Because of environmental barriers 0.17 0.01 0.16 0.12 0.22 0.41 0.27

D6.3 Others affects one’s dignity 0.19 �0.07 0.16 0.12 �0.15 0.27 0.72

Accumulated variance by factor 12.46 11.62 11.61 11.53 8.74 7.64 3.66

Total R2: 67.26%.

KMO: 0.87, x2: 6447.66***.

*** p< .001

[(Fig._2)TD$FIG]

Fig. 2. The confirmatory factor analysis (CFA) of the traditional Chinese version of the WHODAS 2.0 (n = 307) Fit indicators: GRI = .76; CFI = .81

RMSEA = .091.

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–28202818

;

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–2820 2819

4.4. Construct validity

4.4.1. Concurrent validity

The WHODAS 2.0 traditional Chinese version had a strong negative correlation with the WHOQOL-BREF in every domainand in the summary score except the domain of cognition and environment, which showed that the WHODAS 2.0 has goodconstruct validity with the WHOQOL-BREF. Compared with other studies, there is a negative correlation (r =�0.54) betweenthe WHOQOL-BREF and WHODAS 2.0 summary scores in the Japanese version. (Tazaki et al., 2014) In a multi-country study,the summary score of the WHODAS 2.0 was only negatively correlated with the social relationships (r =�0.14) andenvironment (r =�0.15) domains of the WHOQOL-BREF (Schmidt et al., 2010). In a previous study, the relationship betweenthe WHODAS 2.0 summary score and the WHOQOL-BREF subscale was discussed primarily, but the relationship betweeneach domains were also important. In this study, the relationship of the subscale and overall scale between the WHODAS 2.0and the WHOQOL-BREF was examined. The results also pointed out there were no relationship between cognition andenvironment. It’s indicated that the instrument of WHODAS 2.0 traditional Chinese version were valid and valuable inpractice but not comprehensive to cover the meaning of WHOQOL-BREF.

4.4.2. Factor analysis – EFA and CFA

The factor structure in our EFA explained 67.26% of total variance that better than other language versions (Federici et al.,2009; Posl et al., 2007). The domains were divided into 7 factors in the traditional Chinese version that different with originalstructure, but the 7th factor was only one item ‘‘D6.3 Others affects one’s dignity’’ and it contributed the explanation ofmodel only 3.66% (Table 5) that would be attributed the property of our samples. Most of them were intellectual disabilitythat difficult to recognize the meaning of dignity for their life. The factor loading for every item were greater than 0.4 in CFAthat means the structure was acceptable (Nunnally, 1978). Our study confirmed that the WHODAS 2.0 traditional Chineseversion is suitable for evaluating disability status.

5. Conclusion and limitations

The traditional Chinese version of WHODAS 2.0 was the first to be developed and systemic examined in the present studyand the results indicated that it is a valid and reliable instrument to evaluate functioning and disability for people withdisability. For disability eligibility system of Taiwan government, the results provided valuable evidence to design theassessment instrument. However, the high internal consistency and ICC of this traditional Chinese version might be valid fora population with ID or living in institutions. Except for extrapolation, we should evaluate the long-term effectiveness andefficiency of the traditional Chinese version of the WHODAS 2.0 in clinical practice and welfare services in the future.

Acknowledgments

This research was financially supported by the Ministry of Science and Technology (NSC 101-2314-B-320-003) in Taiwan.We also thank the institutional staff members who supported and oversaw the measurement.

References

Almazan-Isla, J., Comın-Comın, M., Damian, J., Alcalde-Cabero, E., Ruiz, C., Franco, E., et al. (2014). Analysis of disability using WHODAS 2.0 among the middle-agedand elderly in Cinco Villas, Spain. Disability and Health Journal, 7(1), 78–87.

Baron, M., Schieir, O., Hudson, M., Steele, R., Kolahi, S., Berkson, L., et al. (2008). The clinimetric properties of the World Health Organization Disability AssessmentSchedule II in early inflammatory arthritis. Arthritis and Rheumatism (Arthritis Care & Research), 59(3), 382–390.

Cheng, H. G., Huang, Y. Q., Liu, Z., Zhang, M., Lee, S., Shen, Y., et al. (2012). Disability associated with mental disorders in metropolitan China: An application of thequantile regression approach. Psychiatry Research, 199(3), 212–219.

Chisolm, T. H., Abrams, H. B., McArdle, R., Wilson, R. H., & Doyle, P. J. (2005). The WHO-DAS II: Psychometric properties in the measurement of functional healthstatus in adults with acquired hearing loss. Trends in Amplification, 9(3), 111–126.

de Pedro-Cuesta, J., Alberquilla, A., Virues-Ortega, J., Carmona, M., Alcalde-Cabero, E., Bosca, G., et al. (2011). ICF disability measured by WHO-DAS II in threecommunity diagnostic groups in Madrid, Spain. Gaceta Sanitaria, 25(Suppl. 2), 21–28.

Federici, S., Meloni, F., Mancini, A., Lauriola, M., & Olivetti Belardinelli, M. (2009). World Health Organisation Disability Assessment Schedule II: Contribution to theItalian validation. Disability and Rehabilitation, 31(7), 553–564.

Garin, O., Ayuso-Mateos, J. L., Almansa, J., Nieto, M., Chatterji, S., Vilagut, G., et al. (2010). Validation of the ‘‘World Health Organization Disability AssessmentSchedule, WHODAS-2’’ in patients with chronic diseases. Health Qual Life Outcomes, 8, 51.

Gresham, G. E., Phillips, T. F., & Labi, M. L. (1980). ADL status in stroke: Relative merits of three standard indexes. Archives of Physical Medicine and Rehabilitation,61(8), 355–358.

Guilera, G., Gomez-Benito, J., Pino, O., Rojo, J. E., Cuesta, M. J., Martinez-Aran, A., et al. (2012). Utility of the World Health Organization Disability AssessmentSchedule II in schizophrenia. Schizophrenia Research, 138(2–3), 240–247.

Harwood, R. H., Rogers, A., Dickinson, E., & Ebrahim, S. (1994). Measuring handicap: The London Handicap Scale, a new outcome measure for chronic disease.Quality in Health Care, 3(1), 11–16.

Hobart, J. C., & Thompson, A. J. (2001). The five item Barthel index. Journal of Neurology, Neurosurgery & Psychiatry, 71(2), 225–230.Hsueh, I. P., Lee, M. M., & Hsieh, C. L. (2001). Psychometric characteristics of the Barthel activities of daily living index in stroke patients. Journal of the Formosan

Medical Association, 100(8), 526–532.Katz, P. P. (2003). Measures of adult general functional status: The Barthel Index, Katz Index of Activities of Daily Living, Health Assessment Questionnaire (HAQ),

MACTAR Patient Preference Disability Questionnaire, and Modified Health Assessment Questionnaire (MHAQ). Arthritis Care & Research, 49(S5), S15–S27.Kutlay, S., Kucukdeveci, A. A., Elhan, A. H., Oztuna, D., Koc, N., & Tennant, A. (2011). Validation of the World Health Organization disability assessment schedule II

(WHODAS-II) in patients with osteoarthritis. Rheumatology International, 31(3), 339–346.

T.-Y. Chiu et al. / Research in Developmental Disabilities 35 (2014) 2812–28202820

Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist, 9(3), 179–186.Mas-Exposito, L., Amador-Campos, J. A., Gomez-Benito, J., & Lalucat-Jo, L. (2012). The World Health Organization Short Disability Assessment Schedule: A

validation study in patients with schizophrenia. Comprehensive Psychiatry, 53(2), 208–216.Meesters, J. J., Verhoef, J., Liem, I. S., Putter, H., & Vliet Vlieland, T. P. (2010). Validity and responsiveness of the World Health Organization Disability Assessment

Schedule II to assess disability in rheumatoid arthritis patients. Rheumatology, 49(2), 326–333.MOHW (2007). People with disabilities rights protection act.Novak, S. P., Colpe, L. J., Barker, P. R., & Gfroerer, J. C. (2010). Development of a brief mental health impairment scale using a nationally representative sample in the

USA. International Journal of Methods in Psychiatric Research, 19(Suppl. 1), 49–60.Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.Posl, M., Cieza, A., & Stucki, G. (2007). Psychometric properties of the WHODASII in rehabilitation patients. Quality of Life Research, 16(9), 1521–1531.Rockwood, K., Joyce, B., & Stolee, P. (1997). Use of goal attainment scaling in measuring clinically important change in cognitive rehabilitation patients. Journal of

Clinical Epidemiology, 50(5), 581–588.Schlote, A., Richter, M., Wunderlich, M. T., Poppendick, U., Moller, C., Schwelm, K., et al. (2009). WHODAS II with people after stroke and their relatives. Disability

and Rehabilitation, 31(11), 855–864.Schmidt, S., Power, M., Green, A., Lucas-Carrasco, R., Eser, E., Dragomirecka, E., et al. (2010). Self and proxy rating of quality of life in adults with intellectual

disabilities: Results from the DISQOL study. Research in Developmental Disabilities, 31(5), 1015–1026.Scott, K. M., Von Korff, M., Alonso, J., Angermeyer, M. C., Bromet, E., Fayyad, J., et al. (2009). Mental-physical co-morbidity and its relationship with disability:

Results from the World Mental Health Surveys. Psychological Medicine, 39(1), 33–43.Sousa, R. M., Dewey, M. E., Acosta, D., Jotheeswaran, A. T., Castro-Costa, E., Ferri, C. P., et al. (2010). Measuring disability across cultures – the psychometric

properties of the WHODAS II in older people from seven low- and middle-income countries. The 10/66 Dementia Research Group population-based survey.International Journal of Methods in Psychiatric Research, 19(1), 1–17.

Tazaki, M., Yamaguchi, T., Yatsunami, M., & Nakane, Y. (2014). Measuring functional health among the elderly: Development of the Japanese version of the WorldHealth Organization Disability Assessment Schedule II. International Journal of Rehabilitation Research, 37(1), 48–53.

Ustun, T. B., Kostanjsek, N., Chatterji, S., & Rehm, J. (2010). Measuring health and disability: Manual for WHO disability assessment schedule. WHO.Wolf, A. C., Tate, R. L., Lannin, N. A., Middleton, J., Lane-Brown, A., & Cameron, I. D. (2012). The World Health Organization Disability Assessment Scale, WHODAS II:

Reliability and validity in the measurement of activity and participation in a spinal cord injury population. Journal of Rehabilitation Medicine, 44(9), 747–755.Yen, C. F., Chiu, T. Y., Liou, T. H., Liao, H. F., Li, Y. S., Liang, C. C., et al. (2014). Does the planned long-term care policy in Taiwan meet the needs of people with

disabilities? Health Policy, 116(1), 95–104.Zhao, H. P., Liu, Y., Li, H. L., Ma, L., Zhang, Y. J., & Wang, J. (2012). Activity limitation and participation restrictions of breast cancer patients receiving chemotherapy:

Psychometric properties and validation of the Chinese version of the WHODAS 2.0. Quality of Life Research, 22(4), 897–906.