Psychometric properties of 12-item WHODAS applied through phone survey: An experience in post-crash cohort study

doi:10.21203/rs.3.rs-1532154/v1

Background: Due to more limited capability to function in daily life post injury, survivors need to be reliably assessed without extra commute. The key action is to determine the level of functioning difficulties. Having the opportunity of conducting a national post-crash traffic safety and health cohort study, we aimed to translate and assess the psychometric properties of the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) through phone surveys six month post injury.

Methods: First, having World Health Organization permission, we tested the translation validity by forward translation, expert panel, back-translation, pre-testing and cognitive interviewing, and finalizing the Persian WHODAS. Then, the psychometric study on a national cohort platform assessed the validity, reliability and applicability of 12-item WHODAS through phone surveys. We included data of 255 road traffic injury patients enrolled from the cohort study at six-month follow-up. The psychometric assessment (internal consistency reliability and stability reliability) was conducted on test-retest data of 50 patients with a 7-day time span averagely. An exploratory factor analysis tested the construct validity using extraction method of principal component factor and oblique rotation on 255 patients’ data. Regarding the multiple criteria, scree test, and factors with eigenvalues greater than 0.9 determined the number of factors retained. Data were analyzed using software STATA statistical package.

Results: The respondents were mostly male (81%) with mean age (SD) of 37.7(14.9), employed (71%) and educated (87%). The Persian version had high internal consistency reliability (Cronbach’s α=0.93) and excellent stability reliability (ICC=0.97, 95% CI: 0.92-0.98). An exploratory factor analysis retained four factors defining 86% of all variance. Factors of Self-care, Mobility, and Cognition were completely retained.

Conclusions: The translated brief WHODAS 2.0 was highly reliable, valid and responsive to be applied through phone interviews post injury.

WHODAS

disability

post-crash

psychometrics

road traffic injury

post injury

cohort study

Road traffic crashes (RTCs) result in 1.35 million deaths annually and leave 20 to 50 million people with non-fatal injuries, with many suffering a disability. The injuries account for 41.2 million years of healthy life lost [1–3], and 90% of disability-adjusted life years in Low and Middle Income Countries [4]. In Iran, road traffic injuries )RTIs( are the second leading cause of death [5, 6], and the third cause of disability adjusted life years [7]. Over 2600 person years of life lost are attributed to RTIs [8]. The people hospitalized due to RTIs mainly suffer severe injuries resulting in short- or long-term difficulties in functioning. Only with the functioning information would we recognise their right health care needs [9].

The concepts “functioning” and “disability” in public health and clinical settings represent information on how the population accomplishes its activities and participation [10]. Functioning is “generic term which includes body functions and structures, activities and participation. It indicates the positive aspects of the interaction between the individual (with a health condition) and its contextual factors (personal and environmental factors)” [11]. Disability is an umbrella term for deficiencies of body functions, activity limitations or participation restrictions [10]. The controversial issue is how to measure the functioning level, and many tools have been developed. however the tool by the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) was developed on an inclusive categories of International Classification of Functioning, Disability and Health (ICF) [11]. Although there are many tools developed to measure functioning, WHODAS 2.0 is reputable for several chief characteristics. For instance, the 11th International Classification of Disease–ICD-11 added the section “V Supplementary section for functioning assessment” to form functioning profiles and overall functioning scores [12, 13].

Tool’s psychometric properties may be affected through the method of data collection such as phone surveys comparing with in-person interviews. So it is critical to assess the tool’s psychometric properties for various data collection. Then, very little similar work has been done so far. One of the major field is assessing disability in injuries post-crash and post discharge. For such assessing, there is rare opportunity to have in-person visits, in some settings particularly in low and middle income countries (LMICs) and then the resources do not support the researches to do in-person interviews. It is critical to have robust tools to use for phone surveys. To assess this, we had a great opportunity of a cohort study. This setting helps to apply the tool and assess its psychometric properties through national cohort platform of Post-crash Traffic Safety and Health Cohort Study, Prospective Epidemiological Research Study in IrAN (PERSIAN) [14]. Then the key action was to apply reliable and valid tools in measuring the outcomes of RTIs such as functioning difficulties, a common consequence. So, the expert team chose the inclusive tool of WHODAS 2.0 to be administered to specifically RTI survivors with great care, sensitivity, and searching. As the short form of this globally approved tool has a good explanation of the full-version tool and takes five minutes on average, as well as considering the physical health conditions of RTIs interviewees, we utilized and psychometrically assessed the 12-item WHODAS 2.0, through phone interviews six months post injury. Also, to our knowledge, there are not published works in which the short version tool was used in phone surveys in RTI patients in the region. The paucity of such all-inclusive tool on creating the functioning profile of people involved in RTCs is sensible reason to conduct the translation, and to examine internal consistency, reproducibility, time consistency, and construct validity of brief WHODAS 2.0 among study population through phone surveys six months post injury. Then, it would pave the way for study of the functioning problems caused by road traffic injuries. We expected that the short version would demonstrate excellent internal consistency and test-retest reliability, defined as values of 0.80 or greater and 0.90 or greater, respectively.

Phase one: Translation of the 12-item WHODAS 2.0

The World Health Organization (WHO) permitted the Road Traffic Injury Research Center (RTIRC) to translate the 12-item WHODAS 2.0 into Persian language for use specially in the context of the Traffic Safety and Health Cohort Study (permission ID: 380480). Our expert team composed of members from RTIRC, and physical medicine and rehabilitation Research Center, Tabriz University of Medical Sciences. It conducted an iterative, rigid, and meticulous process of translation, according to the translation guideline recommended by the WHO in five steps: 1. forward translation; 2. Expert panel; 3. Back-translation; 4. Pre-testing and cognitive interviewing, and 5. Final version [15, 16].

Phase two: assessment of reliability and factor structure

The psychometric properties of the brief WHODAS 2.0 was evaluated at six months after crash through phone-based interviews. This study was conducted in a clinical population. The source population of the study were RTIs survivors enrolled from post-crash traffic safety and health cohort study [14]. All participants hospitalized in one of two referral trauma centers, Shohada and Imam Reza, in Tabriz, were followed up through the cohort study at six months after crash. We conducted on a representative sample of RTIs survivors 18 years and above, between 21 May 2020 and 20 March 2021. The whole hospitalized patients have been recorded in the database of Integrated Road Traffic Injury Registry System (IRTIRS) since 2019 [17]. To examine the reliability of the instrument, 50 patients were included to participate in a re-test phase. In order to conduct factor analysis, a maximum number of 21 participants per item was taken to attain the best model fit [18]. So, 255 patients were included in our sample. The selected participants were replaced by the same sex and age if they refused to participate. As inclusion criteria, participants must have been hospitalized due to RTIs six months prior to their interview date for at least 24hrs, registered, aged 18 or over, and informed consent was required. People unable to complete the questionnaire due to any disorders were excluded from the study. The study was carried out by an appropriately trained interviewer with broad experience (≥ 4years) routinely collects data from registered patients on phone calls [19]. The completeness of tools was mostly guaranteed since the interviewer filled them out.

Measures

The 12-item WHODAS 2.0

As a general measure, WHODAS 2.0 was developed based on an inclusive set of categories considered in the framework of ICF to measure disability during the preceding 30 days [11]. It can be used for epidemiological studies and other purposes. This questionnaire was researched by the WHO in many countries, where its reliability, convergent validity with other assessment instruments, constant factor structure, and other psychometric properties were analysed [20].

The short version describes 81% of the variance of the full version. Researchers can compute overall functioning scores, administer through an interview during the mean interview time of about 5 minutes [21]. The response choices for each item ranged on an ordinal scale, from zero denoting “none” to four denoting “extreme/cannot do”. The individual scores in each domain were estimated by simply adding up the results of the two relevant items [22]. Each domain contains two questions from the corresponding domain in the full version: Domain 1: Cognition (items 3 and 6 of the questionnaire); Domain 2: Mobility (items 1 and 7); Domain 3: Self-care (items 8 and 9); Domain 4: Getting along with people (items 10 and 11); Domain 5: Life activities (items 12 and 2); Domain 6: Participation (items 4 and 5) [16]. A simple way of calculating the results was used for the psychometric evaluation of the tool. The total score is a percentage as follows:

Total score = (∑ item scores/ 48) x 100

It ranged from 0 (no disability) to 48 (complete disability). A higher result indicates a higher level of disability [21]. To draw a comparison, raw total scores and subscale scores were then converted into a 0–100 scale as normalized scores, using the complex method of scoring [23]. We categorized the total disability scores according to the ICF severity ranges (no problem, 0–4; mild disability, 5–24; moderate disability, 25–49; severe/extreme disability, 50–100). We considered a disability score of greater than or equal to 25 to indicate ‘disability’ based on the WHODAS ICF [24]. When only one item from the 12 items was missing, the average of the remaining 11 items was assigned to the missing item. If more than one item was missing, the survey was rejected [25].

Socio-demographic data

Basic socio-demographic data (sex, age, marital status, education, and job) and selected crash-related information (used vehicle, counterpart vehicle, crash mechanism, and injured person’s role) were collected.

Statistical analysis

Initial analyses of the WHODAS and baseline characteristics data were performed using descriptive statistics. First an inter-item correlation matrix was visually observed. Correlations were defined as excellent (> 0.75), moderate (0.50–0.74), fair (0.25–0.49), and no meaningful correlation (< 0.25) [26]. With a sample size of 255 patients, this study has 80% power (two-sided P = 0.05) to detect a correlation coefficient of 0.2 or greater. The data were analyzed using STATA statistical package version 15 (StataCorp LLC, Texas) and adopted a significance level of 0.05.

Scale score reliability

Internal consistency reliability

Internal consistency indicates the extent to which the items in a subscale are correlated. The homogeneity was evaluated using Cronbach’s Alpha statistic [27]. A value between 0.5 and 0.75 was considered moderate while values between 0.75 and 0.9 indicate good and values greater than 0.90 specify excellent reliability [28].

Test–retest reliability

Another measurement used to directly assess the scale score reliability of the tool was repeatability assessment using the test–retest method. Comparing the results of the first and second measurements examined the consistency over time. A group of 50 people was interviewed on telephone, between 21 May 2020 and 15 June 2020. All of them participated in both test and retest measurements. They did not have any treatment during that week period or changes in their medication. The average time between two measurements was 7 days (range from 6 to 8 days). Along with the criteria of the study, specific exclusion criteria for this step included having psychiatric problems affect the validity of the responses or an event experienced during the test-retest interval significantly increases the level of disability (e.g. disease, another car crash, and fall).

The normality of total scale and subscales’ distributions were tested based on the value of skewness and kurtosis. The test-retest correlation was measured by tau Kendal b. The reliability of the method was assessed by intra-class correlation coefficient (ICC). Scores of test-retest were analysed using the “single-rater type (k = 1), absolute-agreement, 2-way mixed-effects model to produce the intra-class correlation with 95% confidence intervals [28]. An ICC coefficient of > = 0.75 was considered as evidence of measurement stability. ICC between 0.4 and 0.75 indicates fair to good reliability, and ICC < 0.4 indicates poor reliability [29].

Factor structure

Exploratory factor analysis (EFA)

The exploratory factor analysis is a fundamental tool to define the optimal count of latent variables in the validation of 12-item WHODAS 2.0 completed by 255 patients six months after crash on phone calls. As the correlation matrix was not compound symmetric, we did not run the method of Maximum Likelihood. Since 5-point Likert scales, are subject to lack of normality, we extracted factors using method of principal component factor and oblique rotation on data of 255 patients, from May 2020 to March 2021. To test the assumptions of EFA, the overall KMO (Kaiser-Meyer-Olkin) to measure sampling adequacy was 0.81, and the Bartlett’s test of sphericity proved an appropriate model (X² = 4278.422 and P < 0.0001) [30]. We used multiple criteria for determining the number of factors such as cumulative percent of variance. Also, according to the scree plot, factors with eigenvalues greater than 0.9 were proper to retain [31] (Fig. 1).

Floor and ceiling effects

The floor and ceiling effects were derived from the percentage of respondents with the lowest and highest possible scale scores for individual items of the tool. Ceiling and floor effects were considered present when it is higher than 15% [32].

The research data were data for the brief-version WHODAS 2.0, and crash-related and socio-demographic characteristics. A total of 255 subjects were approached by a research assistant. The majority of participants were male (81%). The mean age (SD) was 37.7(14.9; 18-86yrs). The respondents were mostly employed (71%) and educated (87%) (Table 1). The mechanism of half of the crashes was vehicle-vehicle collision. The majority of subjects were motorcycle and car drivers. A fourth of patients were passenger or pillion passenger when they had crash (Table 1).

The average disability score for the study population (n=255) was 26.87 ± 22.1 on a scale of 0 to 100 points, with 50 percentile at 25.

Table 1

Baseline characteristics of study population (N=255)

Variables	Categories	N=255 (%)
Sex	Male	207 (81%)
Sex	Female	48 (19%)
Age	18-24yrs	54 (21%)
	25-44yrs	124 (49%)
	45-64yrs	59 (23%)
	65yrs and above	18 (7%)
Marital status	Never married	77 (30%)
	Married	165 (65%)
	Divorced/ widowed	13 (5%)
Education level	Illiterate	33 (13%)
	School education	180 (71%)
	Academic education	42 (16%)
Job	Employed	182 (71%)
Job	Unemployed	73 (29%)
Injured person’s role	Driver	130 (50.9%)
	Pedestrian	55 (21.6%)
	Passenger/ pillion passenger	69 (27.1%)
	Undefined	1 (0.4)

Phase one: Translation of the 12-item WHODAS 2.0

The forward translation was done by three qualified translators as native-like English speakers and one of them is a professor living in the USA. They were familiar with health and disability terms. The tool with 15 items and 402 words was translated into Persian language with 15 items and 380 words. Then the panel discussed on the term “emotionally affected” which takes several Persian meanings, and finalized. An independent linguist blind to the original English terms performed back translation with 15 items and 365 words. Three experts examined compatibility of the forward and back translations per each item, separately on a 4-level Likert scale. Measuring the compatibility, modified Kappa rate (>0.76) revealed that two items were completely compatible and the others’ compatibilities were acceptable, except for the item 5 lower than 0.76, due to the semantic complexity of the used word in forward translation. The expert panel’s and translators’ discussion and minor amendments were followed by a pre-test, to reach the final translation (Table 4).

Phase two:

Factor structure

An exploratory factor analysis was conducted for 12 item WHODAS 2.0. We retained four factors based on the cumulative percent of variance extracted 86% of all variance, the eigenvalue greater than 0.9, and the scree plot (Fig. 1). All items showed uniqueness less than 0.2, except for the item 6 for “concentrating doing something for ten minutes” with 0.44 [33]. Accordingly, four factors with factor loadings equal or >0.3 on 255 people were retained. Items grouping on the same factor presented four factors as follows: factor 1, “social/self-activities” included items involving “Life activities”, “Participation”, and “Getting along” as in the original instrument; factors 2, 3 and 4 indicated activities of “Mobility”, “Self-care”, and “Cognition”, respectively as in the original tool. The loading matrix indicated the item 2 cross loaded onto the factors one and four, though the factor loading was negative for factor four. Each and every item of factors 2 to 4 cross-loaded onto the factor one, too. The highest loadings were considered the most representative, except for the item 6 (Table 2). Since the difference of its loadings was less than 0.1, we categorized it in factor 4 in accordance with the theory of original WHODAS. The total Cronbach’s α was 0.92 ranged from 0.91 for item 5 to 0.93 for both items of Cognition.

Table 2

Factor loading matrix for WHODAS 2.0 items (n=255)

Items (number of item)	Factor1	Factor2	Factor3	Factor4
Maintaining a friendship? (11)	0.9438
Dealing with people you do not know? (10)	0.9393
How much of a problem did you have in joining in community activities (for example, festivities, religious or other activities) in the same way as anyone else can? (4)	0.9266
How much have you been emotionally affected by your health problems? (5)	0.8671
Your day-to-day work/ school? (12)	0.8224
Taking care of your household responsibilities? (2)	0.6231			-0.5900
Concentrating on doing something for ten minutes? (6)	0.4884			0.3985
Walking a long distance such as a kilometer [or equivalent]? (7)	0.5777	0.7656
Standing for long periods such as 30 minutes? (1)	0.6050	0.7171
Washing your whole body? (8)	0.6102		0.7529
Getting dressed? (9)	0.6140		0.7500
Learning a new task, for example, learning how to get to a new place? (3)	0.3865			0.8528
Eigen values	*7.12*	*1.25*	*1.07*	*0.93*
Cumulative variance (%)	*0.59*	*0.69*	*0.78*	*0.86*

The PCA-based factor loading plot depicted clusters containing items with similar correlations of more than 0.30 with a factor (Fig. 2) [34]. There were three items separated from other clusters (Learning, Concentrating and Taking care), cross loaded onto two factors of one and four.

Scale score reliability

Floor and ceiling effects

For the total data (N=255), the floor effect values (an answer of “no problem”) ranged from 24% (item 12) to 91% (item six). All values were above 25% except for items five and 12. The ceiling effect values (an answer of “extremely large” or “cannot do”) ranged from 0.4% (item six) to 6% (item one) except for item seven with 32% ceiling effect.

Internal consistency

For the total data (N=255), the Cronbach’s α coefficients resulted in 0.93 (total scale) ranged from 0.92 (items 1, 4, 5, 10, 11, and 12) to 0.93 (items 2, 3, 6, 7, 8 and 9). The Cronbach’s α coefficients for subscales fluctuated between 0.82 (for Participation) and 0.88 (Cognition).

Internal consistency was assessed for the test-retest data (N= 50) by estimating Cronbach’s α coefficient for the whole scale (0.89), subscales, and items ranged from 0.88 for items 8 (Washing your whole body) and 9 (Getting dressed) to 0.90 for item 7 (Walking a long distance). The Cronbach’s α coefficients for individual subscales were high (Table 3).

Test–retest reliability

Consistency over time was examined by comparing the results of the first and second measurements in 50 people. The skewness-kurtosis test showed the normal distribution for some subscales 4, 5, and 6, on test data. Kendall's tau-b indicated no significant differences between test and retest responses (Table 3).

The reliability of the test-retest method was confirmed using the ICC. Time consistency became very high as ranged from 0.89 (subscale 4) to 0.99 (subscale 2) indicating good to excellent reliability and stability. For the overall result, the ICC was 0.97, demonstrating excellent test-retest reliability (Table 4). The ICC for items ranged from 0.85 to 0.97, good to excellent stability (Table 4).

The test data showed the lowest and the highest mean disability score in Cognition (3.1), and Mobility (7.0), respectively (Table 3).

Table 3

Reliability indices of the Persian version of 12-item WHODAS 2.0 in phone-based survey (n=50)

Subscales (items)	Mean (SD)		Cronbach’s alpha	ICC (95% CI)	Agreement	Kendall's tau-b^*
Subscales (items)	test	retest	Cronbach’s alpha		Agreement	Kendall's tau-b^*
Cognition (3 & 6)	3.10 (1.90)	2.86 (1.80)	0.79	0.93 (0.87-0.96)	Excellent	0.71
Mobility (1 & 7)	7.02 (2.69)	7.06 (2.62)	0.84	0.99 (0.98-0.99)	Excellent	0.91
Self-care (8 & 9)	3.64 (2.26)	3.36 (2.07)	0.77	0.92 (0.85-0.95)	Excellent	0.75
Getting along (10 & 11)	5.56 (2.34)	4.94 (2.19)	0.78	0.89 (0.78-0.94)	Good	0.75
Life activities (12 & 2)	6.68 (1.47)	6.50 (1.43)	0.75	0.94 (0.89-0.97)	Excellent	0.87
Participation (4 & 5)	5.90 (1.84)	5.68 (1.74)	0.76	0.96 (0.93-0.98)	Excellent	0.84
Total score	31.90 (9.52)	30.40 (8.68)	0.81	0.97 (0.92-0.98)	Excellent	0.82

Note: each domain has two items. ^*At significance level of 0.0001.

Table 4

Item reliability of the short form of WHODAS 2.0 tool

Items		ICC (95% CI)	Items in Persian Language
S1	Standing for long periods such as 30 minutes?	0.98 (0.97- 0.99)	به مدت طولانی سرپا ایستادن (مثلاً تا 30 دقیقه)؟
S2	Taking care of your household responsibilities?	0.96 (0.94- 0.98)	انجام مسئولیت‌های خود در منزل؟
S3	Learning a new task, for example, learning how to get to a new place?	0.86 (0.77- 0.93)	یادگیری کار جدید (مثل یافتن آدرس جدید)؟
S4	How much of a problem did you have in joining in community activities (for example, festivities, religious or other activities) in the same way as anyone else can?	0.93 (0.89- 0.96)	برای اینکه بتوانید مشابه سایر افراد، در فعالیت‌های جمعی (مثل جشنواره‌ها، فعالیت‌های مذهبی یا غیره) شرکت کنید چقدر مشکل داشتید؟
S5	How much have you been emotionally affected by your health problems?	0.94 (0.91- 0.97)	هیجان ]روحیه و احساس[ شما چقدر تحت تاثیر وضعیت سلامتی‌تان قرار گرفته است؟
S6	Concentrating on doing something for ten minutes?	0.93 (0.87- 0.96)	تمرکز ده دقیقه‌ای بر انجام کاری؟
S7	Walking a long distance such as a kilometre [or equivalent]?	0.97 (0.96- 0.99)	پیاده‌روی های طولانی (مثلاً بیش از یک کیلومتر)؟
S8	Washing your whole body?	0.90 (0.83- 0.94)	حمام کردن و شستن تمام بدن‌تان؟
S9	Getting dressed?	0.88 (0.79- 0.94)	لباس پوشیدن؟
S10	Dealing with people you do not know?	0.85 (0.71- 0.92)	رابطه برقرار کردن با غریبه‌ها؟
S11	Maintaining a friendship?	0.88 (0.79- 0.94)	حفظ رابطه دوستی؟
S12	Your day-to-day work/school?	0.85 (0.73- 0.92)	انجام وظایف روزانه شغلی/ تحصیلی؟
	Total score	0.97 (0.93- 0.98)

This paper developed the Persian WHODAS 2.0 by translation, then described psychometric properties of the Persian version of phone-based interviewer-administered tool for road traffic injury survivors in Tabriz, Iran, six months after crash, between 21 May 2020 and 20 March 2021. This questionnaire was translated and validated in different languages even in Persian [35, 36] but neither specifically through phone survey nor in RTIs patients. So, in this study, we translated the tool into Persian, and indicated that Persian WHODAS 2.0 is psychometrically reliable to the great extent in phone-based assessment of disability. Satisfactory internal consistency reveals the Persian WHODAS to measure the construct of disability and deliver high reliable score at item level and total score in our representative sample. The schedule is robust to changes in the same subjects over time and less prone to measurement error caused by time elapse, as ICC showed excellent reliability. Exploratory factor analysis discloses the high validity of the schedule as items hang consistently together to create the single construct of disability. The final model represented four factors of social/self-activities, cognition, mobility, and self-care, accounted for 86% of the total variance.

According to the exploratory factor analysis, the Persian WHODAS 2.0 measured disability with four components accounted for 86% of total variance and demonstrated better results compared with some studies [35, 37-39]. The factor loadings of the individual items were strong, as eleven items’ factor loadings were above 0.60 and one item showed 0.39, consistent with factor loadings in a study [37]. One factor, titled social/self-activities, contains items pertained to the subscales of Getting along, Life activities, and Participation, similar to a study in terms of the items of the extracted factor, not the number of extracted factors [35]. Three separate factors certainly distinguished cognition, mobility, and self-care subscales. The structure is quite similar to the structure found in the original English version [21]. A study indicated different groupings, attained a two-factor solution in older people from different dementia sites, with subscales of mobility, life activities, cognition, and participation in its first dominant factor and the self-care and getting along subscales in its second factor [40]. It seems higher number of sample size would help to extract the remaining three subscales separately as the same three distinct factors as in the original version. However, it is better to consider the “most appropriate” rather than the “correct” number of factors.

No measurement difference was indicated between the test and retest results. In spite of importance of using the spectrum of non-extreme in face-to-face interviews, it is not practical in phone-based surveys. However, we attained high consistency between total and subscales’ scores in phone-based surveys.

The Cronbach’s α coefficient for the Persian version indicated the ideal internal consistency and better results compared with a study among Syrian refugees (α=0.74)[37]. The Cronbach’s α coefficient was similar to several studies with comparable target groups (Ethiopian Amharic version α=0.88, in trauma patients α=0.91[35]) and different target groups (a global study [25], Brazilian study in patients with chikungunya (α=0.93) [41], in only phone-based psychometric study of WHODAS among critically ill patients (α=0.91) [26], Spanish version in depressed patients (α=0.89)[38], and Portuguese version in patients with musculoskeletal pain (α=0.84) [42]). The highest Cronbach’s alpha was related to Cognition in consistent with a study in Ethiopia [35].

The study confirmed that the scale was consistent over time representing excellent test-retest reliability as ICC fell in a range from 0.89 (Getting along) to 0.99 (Mobility) which is in line with a study in people with and without hand injuries (ICC value of 0.88) [43]. A study in patients with Kashin-Beck disease affirmed a good repeatability of the 12-item WHODAS 2.0 [44]. We reached higher time consistency at total and subscale levels compared with a study in Polish society (ICC value of 0.91, ranged from 0.50 (Self-care) to 0.97 (Mobility)) [16]. Both studies showed the highest ICC for the subscale of Mobility with the items of “Standing for long periods such as 30 minutes?” and “Walking a long distance such as a kilometre [or equivalent]?” It could be explained by the fact that these activities are of physical activities rarely affected by any actions through the time between test and retest. Generally, the results might differ regarding the mode of administration (in-person interviews versus our phone-based interview) and characteristics of the study population such as age, health status, country of origin) [45]. Moreover, the quality of translation definitely influence the understanding of questions by interviewees.

Data showed floor effects but no ceiling effects, except for one item (item seven). It was almost in line with baseline result of a study in critically ill patients [26] as well as a study of polish society [16]. Overall, study setting and way of using a tool might bring about floor and ceiling effects in data. In this study, the tool was used through phone interviews. However, the content of the tool was examined by asking the opinion of interviewees about the clarity and comprehensibility [46].

This tool is short, and have less complex questions, so it is a proper choice to use in such prospective cohort study with various questionnaires and examinations. The tool was completed via phone calls. This technique is essential for a couple of reasons. It takes around five minutes to complete it on phone calls [47]. It enabled data collection without unnecessary travel of patients to the study site/ hospital or travel of interviewers to the place of patients’ residence, due to patients’ movement restrictions. Consecutively it helped to avoid their exposure to pathogens in the current situation of the Pandemic COVID-19. Also, in-person interview is not an easy-to-access opportunity for every researcher/ participant. Our data were attained by one assistant interviewer with more than four years of experience in running phone interviews to fill out questionnaires determined in the IRTIRS. So, lack of such skills would also lead to different results in phone-based interviews. Since conducting telephone interviews requires special communication skills, it is necessary to repeat the research on the basis of phone-interview conducted by different interviewers. On the other hand, some patients such as elderly patients and those with hearing problems may have problem with phone survey. So it is recommended to reassess the tool in specific target population such as elderly people.

In this study, subjects were recruited at six months after crash. It is important to ensure that their function was partially stable and patients have got recovered from their minor injuries causing slight physical restrictions [19].

Strengths and Limitations

To the best of our knowledge, this is the first study in which the psychometric properties of the Persian version of the 12-item WHODAS 2.0 have been evaluated through phone interviews in road traffic injury patients above 18 years in the region and even the region. This study enables assessing functioning, and disability in injured population unable to attend the study site and conducting studies to compare results between different settings.

The age limitation in the tool assessment is defined owing to some feasibility issues. So, the psychometric assessment of brief WHODAS in teenagers is recommended. We plan to extend the research by developing and administrating the tool to the population with Azeri language, the most common language in the northwest Iran. Since it is necessary to use the valid and reliable tool for people whose mother language is Azeri, and there are difficulties in running the Persian version for aged and illiterate people. Regarding the effect of various communication skills, we recommend running phone-based survey by different interviewers.

The potential sources of bias might be considered when making inferences based on our findings. This study as an exploratory cross-sectional one, six months after crash, never provides evidence on the longitudinal performance of the tool for short or long term timing after crash. Moreover, this was a cohort study involving patients with road traffic injuries hospitalized in the only two referral hospitals in Northwest Iran and registered. Thus, it mainly represents the patients needed to be hospitalized post-crash. Conducting the research on the basis of phone interviews in the same and different contexts but in larger sample size is needed. We suggest using a longitudinal design to evaluate responsiveness to change and test–retest reliability.

The current study translated and examined the psychometric properties of the 12-item WHODAS 2.0 in a study population experienced road traffic crashes and injuries approximately six months prior to study time and hospitalized in one of two referral trauma centres located in the city of Tabriz, Iran. We realized that the disability assessment tool is short and easy-to-use, and it is suitable to apply in phone-based interviews during follow-ups of injured patients above 18 years.

WHODAS 2.0: World Health Organization Disability Assessment Schedule 2.0; RTIs: road traffic injuries; ICF: International Classification of Functioning, Disability and Health; ICD-11: 11^th International Classification of Disease; LMICs: low and middle income countries; PERSIAN: Prospective Epidemiological Research Study in IrAN; RTIRC: Road Traffic Injury Research Center; IRTIRS: Integrated Road Traffic Injury Registry System; ICC: intra-class correlation coefficient; EFA: Exploratory factor analysis

Ethical Approval and Consent to participate

The study was evaluated and approved by the Ethics Committee of the Road Traffic Injury Research Center, Tabriz University of Medical Sciences (cohort ethics code: IR.TBZMED.REC.1398.543). An informed consent was required from all participants.

Consent for publication

All authors have read and announced their consent for publication of this work.

Availability of supporting data

Data are available upon reasonable request. The interested researchers may contact Dr. Homayoun Sadeghi-Bazargani (PI). The applications will be reviewed upon approval by the research council and the regional ethics committee of Tabriz University of Medical Sciences. The study website is https://cohortsafety.tbzmed.ac.ir/.

Competing interest

Not applicable.

Funding

Not applicable.

Authors’ contributions

HSB conceived and designed the study, HSB and NS conducted the analyses and, HSB, NS interpreted the results. HSB and NS drafted the manuscript. MAJ, MF, and SBH contributed in the critical drafting and revising of the manuscript for important intellectual content. All authors read and approved the final manuscript.

Acknowledgements

The authors would like to thank the participants in the study.

Palmera-Suárez, R., et al., Disability related to road traffic crashes among adults in Spain. Gaceta Sanitaria, 2015. 29: p. 43-48.
Peden, M., McGee K, and Krug E, Injury: A leading cause of the global burden of disease, 2000. 2002, World Health Organization: Geneva.
World Health Organization, Global Status Report on Road Safety: time for action. 2009 Geneva.
Abegaz, T. and S. Gebremedhin, Magnitude of road traffic accident related injuries and fatalities in Ethiopia. PLOS ONE, 2019. 14(1): p. e0202240.
Bazargan-Hejazi, S., et al., The Burden of Road Traffic Injuries in Iran and 15 Surrounding Countries: 1990-2016. Arch Iran Med, 2018. 21(12): p. 556-565.
Azami-Aghdash, S., et al., Barriers to and Facilitators of Road Traffic Injuries Prevention in Iran; A Qualitative Study. Bull Emerg Trauma, 2019. 7(4): p. 390-398.
Institute for Health Metrics and Evaluation. 2010; Available from:http://ghdx.healthdata.org/record/ihme-data/gbd-2010-results-cause-1990-2010.
Saadat, S., et al., The Most Important Causes of Death in Iranian Population; a Retrospective Cohort Study. Emerg (Tehran), 2015. 3(1): p. 16-21.
Stucki, G. and J. Bickenbach, Functioning: the third health indicator in the health system and the key indicator for rehabilitation. Eur J Phys Rehabil Med, 2017. 53(1): p. 134-138.
Castro, S., et al., The World Health Organization Disability Assessment Schedule 2 (WHODAS 2.0): remarks on the need to revise the WHODAS. Cadernos de Saúde Pública, 2019. 35.
World Health Organization. International Classification of Functioning, Disability and Health (ICF). 2021 [cited202124/05/2021]; Available from:https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health.
World Health Organization. WHO Disability Assessment Schedule 2.0 (WHODAS 2.0). 2018; Available from:https://www.who.int/classifications/icf/whodasii/en/, retrieved at 29.03.2020.
Papadakaki, M., et al., Psychological distress and physical disability in patients sustaining severe injuries in road traffic crashes: Results from a one-year cohort study from three European countries. Injury, 2017. 48(2): p. 297-306.
Homayoun Sadeghi-Bazargani, et al., PERSIAN Traffic Safety and Health Cohort: A Study Protocol on Post-Crash Mental and Physical Health Consequences. Journal of Injury Prevention, 2022.
World Health Organization. Process of translation and adaptation of instruments. [cited 2021 17 May 2021]; Available from:http://Www.Who.Int/Substance_Abuse/Research_Tools/Translation/En/.
Ćwirlej-Sozańska, A., et al., Psychometric properties and validation of the polish version of the 12-item WHODAS 2.0. BMC Public Health, 2020. 20(1): p. 1203.
Sadeghi-Bazargani, H., et al., Developing a National Integrated Road Traffic Injury Registry System: A Conceptual Model for a Multidisciplinary Setting. J Multidiscip Healthc, 2020. 13: p. 983-996.
Lynn, M.R., Determination and quantification of content validity. Nurs Res, 1986. 35(6): p. 382-5.
Marin, S., et al., The protocol for validating phone interview tools on post-discharge consequences of road traffic injuries. Journal of Injury and Violence Research, 2020. 12(3).
T. Bedirhan Üstün, et al., Chapter 58: The World Health Surveys, in Health Systems Performance Assessment. 2003, World Health Organization: Geneva. p. 797-808.
Ustun, T.B., et al., Measuring health and disability : manual for WHO Disability Assessment Schedule (WHODAS 2.0) / edited by T.B. Üstün, N. Kostanjsek, S. Chatterji, J.Rehm. 2010, World Health Organization: Geneva.
Silva, A.G., et al., Self-Reported Disability: Association With Lower Extremity Performance and Other Determinants in Older Adults Attending Primary Care. Physical Therapy, 2015. 95(12): p. 1628-1637.
Wolf, A.C., et al., 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. J Rehabil Med, 2012. 44(9): p. 747-55.
Lee, H.H., et al., Is WHODAS 2.0 Useful for Colorectal Cancer Survivors? Annals of rehabilitation medicine, 2017. 41(4): p. 667-676.
Üstün, T.B., et al., Developing the World Health Organization Disability Assessment Schedule 2.0. Bulletin of the World Health Organization, 2010. 88(11): p. 815-823.
Higgins, A., et al., The psychometric properties and minimal clinically important difference for disability assessment using WHODAS 2.0 in critically ill patients. Critical Care and Resuscitation, 2021. 23: p. 103-112.
Andresen, E.M., Criteria for assessing the tools of disability outcomes research. Archives of physical medicine and rehabilitation, 2000. 81: p. S15-S20.
Koo, T.K. and M.Y. Li, A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med, 2016. 15(2): p. 155-63.
Cicchetti, D.V., Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological assessment, 1994. 6(4): p. 284.
Watkins, M.W., Exploratory Factor Analysis: A Guide to Best Practice. Journal of Black Psychology, 2018. 44(3): p. 219-246.
Rencher, A.C. and C. WF, Methods of Multivariate Analysis. 2012, New Jersey: Wiley.
Terwee, C.B., et al., Quality criteria were proposed for measurement properties of health status questionnaires. Journal of clinical epidemiology, 2007. 60(1): p. 34-42.
MacCallum, R.C., et al., Sample Size in Factor Analysis: The Role of Model Error. Multivariate Behav Res, 2001. 36(4): p. 611-37.
Tavakol, M. and A. Wetzel, Factor Analysis: a means for theory and instrument development in support of construct validity. International journal of medical education, 2020. 11: p. 245-247.
Abedzadeh–kalahroudi, M., et al., Psychometric properties of the world health organization disability assessment schedule II -12 Item (WHODAS II) in trauma patients. Injury, 2016. 47(5): p. 1104-1108.
Rajeziesfahani, S., et al., Validity of the 36-item Persian (Farsi) version of the world health organization disability assessment schedule (WHODAS) 2.0. International Journal of Mental Health, 2019. 48(1): p. 14-39.
Akhtar, A., et al., Exploratory and confirmatory factor analysis of the 12-item Arabic World Health Organization Disability Assessment Schedule (WHODAS 2.0) as a screening tool for Syrian refugees. BJPsych Open, 2021. 7(6): p. e190.
Luciano, J.V., et al., Psychometric properties of the twelve item World Health Organization Disability Assessment Schedule II (WHO-DAS II) in Spanish primary care patients with a first major depressive episode. J Affect Disord, 2010. 121(1-2): p. 52-8.
Saltychev, M., et al., A study of the psychometric properties of 12-item World Health Organization Disability Assessment Schedule 2.0 in a large population of people with chronic musculoskeletal pain. Clinical Rehabilitation, 2017. 31(2): p. 262-272.
Sousa, R.M., et al., 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. Int J Methods Psychiatr Res, 2010. 19(1): p. 1-17.
Jéssica, A., et al., Psychometric properties of WHODAS for use in patients with chikungunya in Brazil. Fisioterapia e Pesquisa, 2019. 26.
Silva, C., et al., Adaptation and validation of WHODAS 2.0 in patients with musculoskeletal pain. Rev Saude Publica, 2013. 47(4): p. 752-8.
Marom, B.S., et al., Cross-cultural adaptation of the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) for Hebrew-speaking subjects with and without hand injury. Disabil Rehabil, 2017. 39(12): p. 1155-1161.
Younus, M.I., et al., Reliability and validity of the 12-item WHODAS 2.0 in patients with Kashin-Beck disease. Rheumatol Int, 2017. 37(9): p. 1567-1573.
Kirchberger, I., et al., Feasibility and psychometric properties of the German 12-item WHO Disability Assessment Schedule (WHODAS 2.0) in a population-based sample of patients with myocardial infarction from the MONICA/KORA myocardial infarction registry. Population Health Metrics, 2014. 12(1): p. 27.
Anjos, D.B.M.D., et al., Reliability and construct validity of the Instrument to Measure the Impact of Valve Heart Disease on the Patient's Daily Life. Revista latino-americana de enfermagem, 2016. 24: p. e2730-e2730.
White, M.C., et al., Measurement of patient reported disability using WHODAS 2.0 before and after surgical intervention in Madagascar. BMC health services research, 2018. 18(1): p. 305-305.

No competing interests reported.

Psychometric properties of 12-item WHODAS applied through phone survey: An experience in post-crash cohort study

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Phase one: Translation of the 12-item WHODAS 2.0

Phase two: assessment of reliability and factor structure

Measures

The 12-item WHODAS 2.0

Socio-demographic data

Statistical analysis

Scale score reliability

Internal consistency reliability

Test–retest reliability

Factor structure

Floor and ceiling effects

Results

Phase one: Translation of the 12-item WHODAS 2.0

Phase two:

Factor structure

Scale score reliability

Floor and ceiling effects

Internal consistency

Test–retest reliability

Discussion

Strengths and Limitations

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1