ABSTRACT
Background
The World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) is a popular tool for evaluating functioning and disability in a range of population demographics and medical situations. However, very little is known about the WHODAS 2.0's validity and reliability, particularly when dealing with potentially life‐threatening maternal conditions (PLTCs). The aim of this study was to evaluate the validity of the WHODAS 2.0 Tigrigna version.
Methods
This cross‐sectional study was conducted in Tigray, northern Ethiopia, from December 15 to 20, 2023. Following translation and back translation, women who had experienced PLTCs during a recent pregnancy, childbirth, or postpartum period were administered the 36‐item WHODAS 2.0 in Tigrigna version 6 months after the childbirth. In total, 121 women with a history of PLTCs participated. Cronbach′s α was used to evaluate internal consistency in all six WHODAS 2.0 domains, while Spearman′s correlation coefficient was used to evaluate convergent validity. With confirmatory factor analysis, construct validity was also examined.
Results
All domain scores of the Tigrigna version of the WHODAS 2.0 indicated excellent internal consistency (α = 0.917–0.978 for 36 items and α = 0.874–0.940 for 12 items), while the Cronbach′s α coefficients for the summary score were 0.981 and 0.952 for 36 and 12 items, respectively. The convergent validity between the 36‐item and 12‐item WHODAS 2.0 showed a strong correlation between similar constructs (r = 0.909–0.981).
Conclusion
Despite the small sample limitation, the WHODAS 2.0 tool adapted to the Tigrigna version indicated an acceptable reliability and validity and therefore could be applied to women with a history of PLTCs at 6 months postpartum.
Keywords: functioning and disability, reliability, severe maternal morbidity, validity, WHODAS‐II
The World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) of the Tigrigna version has strong internal consistency (Cronbach's α coefficients = 0.98). The convergent validity between the 36‐item and 12‐item WHODAS 2.0 indicated a strong correlation (r = 0.909–0.981) within similar domains. The WHODAS 2.0 Tigrigna version is reliable and valid to measure the disability level within severe maternal morbidity patients.
Abbreviations
- CFA
confirmatory factor analysis
- CFI
comparative fit index
- CVI
content validity index
- D
domain
- ICVI
item content validity index
- KMO
Kaiser–Meyer–Olkin test
- PLTC
potentially life‐threatening conditions
- r
Spearman′s correlation coefficient
- RMSEA
root mean square error of approximation
- SCVI
scale validity index
- SD
standard deviation
- SRMR
standardized root mean residual
- TLI
Tucker Lewis index
- WHO
World Health Organization
- WHODAS‐2
World Health Organization Disability Assessment Schedule‐II
1. Introduction
Potentially life‐threatening maternal conditions (PLTCs) are the most common cause of maternal mortality worldwide, particularly in low‐and middle‐income countries [1, 2]. According to a systematic review, Ethiopia has the highest prevalence of PLTCs (17.55%) [3]. PLTC is severe maternal morbidity that includes obstetric hemorrhage, severe management indicators, hypertensive disorders, and other systemic disorders [4, 5]. In addition to mortality, PLTCs have a substantial impact on maternal morbidity, which leads to mental and physical impairments that impact everyday functioning and quality of life, some of which may be permanent or irreversible [6, 7]. Among women aged 15–44, these conditions also rank as the third leading cause of disability‐adjusted life years [8].
The World Health Organization's (WHO) report stated that 15% of the world′s population has a disability [9], but in Ethiopia the prevalence was 0.74% in 2022 [10]. This discrepancy may arise from Ethiopia′s narrower definition of disability, which focuses on difficulties in daily activities and excludes restrictions in social participation [10, 11], compared to the WHO's comprehensive definition that includes impairments, activity limitations, and participation restrictions [12]. Information about disability and functioning is becoming more important in the field of public health. Its significance as a third health indicator, alongside morbidity and mortality, is being discussed. These three health indicators supplement one another to give the health system more precise information [13]. Addressing the full spectrum of maternal morbidities and related disabilities is one of the five strategic objectives of the Sustainable Development Goal 3, which aims to end preventable maternal mortality [14].
The 2015–2020 Health Sector Transformation Plan and other national policies and guidelines in Ethiopia place a strong emphasis on enhancing maternal health by increasing access to emergency obstetric care and skilled birth attendance [15]. These guidelines, however, lack specific protocols for managing long‐term functional disabilities resulting from PLTCs and instead focus primarily on preventing mortality. In settings with a lack of resources, such as Ethiopia, where access to rehabilitation is limited, the functions of healthcare professionals, community health workers, and rehabilitation services in addressing these disabilities are still not well understood [16]. This disparity is particularly significant in low‐and middle‐income countries, where health systems deal with issues like poor infrastructure, a shortage of skilled workers, and socioeconomic barriers, all of which increase the burden of maternal morbidity and disability [17, 18].
Validating the World Health Organization Disability Assessment Schedule‐II (WHODAS 2.0) for PLTCs is an important step toward understanding the functional impact of severe maternal health conditions, particularly in postpartum women. The results of this study can influence clinical practice, guide policy development, and support initiatives aimed at improving the health and well‐being of women who suffer from severe maternal health problems. To address this, we conducted a methodological investigation that looked at the psychometric properties of the WHODAS 2.0 Tigrigna version instrument for those with PLTCs to make sure the translation and adaptation were successful. This study aimed to evaluate the validity and reliability of the WHODAS 2.0 tool in women who experienced PLTCs at 6 months postpartum.
2. Methods
2.1. Study Participants, Sampling, and Data Collection Procedure
A cross‐sectional validation study was conducted using the hospital′s registration book and medical chart to identify all women who had a history of PLTCs from June 10 up to July 10, 2023. Participants were women assessed 6 months post‐PLTCs in Tigray, northern Ethiopia. After extracting the list of study participants from the registration book, women were contacted via telephone and visited at their homes for data collection between December 15 and 20, 2023. The WHODAS 2.0 tool was administered via interviewers to 121 women who had a history of PLTCs.
2.2. Measurements
The WHODAS 2.0 is a multifaceted, comprehensive disability assessment tool. Six domains comprise the full WHODAS 2.0 (36 items): the first domain is called “Cognition” (six items); the second is called “Mobility” (five items); the third is called “Self‐care” (four items); the fourth is called “Getting along” (five items); the fifth is called “Life activities” (eight items: four items related to activities at home and four related to activities at work or school); and the sixth is called “Participation” (eight items). When time is limited, the 12‐item short‐form version can be used to quickly evaluate general functioning and disability in clinical and research contexts. S1 through S12 were the labels assigned to the 12 WHODAS 2.0 items. Two questions from each of the six domains in the 36‐item version comprise the 12‐item WHODAS 2.0's structure: “cognition” (D1.1 = S6 and D1.4 = S3), “mobility” (D2.1 = S1 and D2.5 = S7), “self‐care” (D3.1 = S8 and D3.2 = S9), “getting along” (D4.1 = S10 and D4.2 = S11), “life activities” (D5.1 = S2 and D5.5 = S12), and “participation” (D6.1 = S4 and D6.5 = S5). The degree of difficulty encountered during the previous 30 days is represented by the five possibilities for each item response: none = “0”, mild = “1”, moderate = “2”, severe = “3”, extreme/cannot do = “4”. If the person is neither working nor enrolled in school, the 36 and 12 items are lowered to 32 and 11, respectively. A higher WHODAS 2.0 score indicates greater difficulties in daily activities. The overall score ranges from 0 to 100 (“0” = no disability, “100” = full disability) [12].
2.3. Translation and Adaptation
The form was first translated into the Tigrigna language by the corresponding author, a maternity and reproductive health specialist, while taking into account the local community′s conceptual and cultural background. Experts in midwifery, nursing, gynecology, and public health evaluated the initial draft. The second draft was then back‐translated into English by a person who speaks Tigrigna and English proficiently. A third draft was produced following a comparison of the original and back‐translated versions.
2.4. Data Quality Control
We performed the following activities to ensure the quality of the data: (1) we provided 2 days of training to the supervisors and data collectors; (2) we closely monitored the data collection process; and (3) we used an electronic mobile application (Kobo toolbox) for data collection.
2.5. Data Analysis
For data analysis, we used Stata version 17 (StataCorp LLC, College Station, United States) and SPSS version 24 (IBM, Armonk, United States). To describe the socio‐demographic characteristics of the study participants, we used descriptive statistics. The data showed a considerable deviation from the normal distribution, as indicated by the Shapiro–Wilk test p‐value of less than 0.05. Therefore, in order to assess the relationship between the 36‐item and 12‐item WHODAS 2.0, we employed Spearman's correlation coefficient. Furthermore, we employed three methods to examine the data: content validity, construct validity, and internal consistency.
2.5.1. Reliability
Internal consistency was evaluated using the Cronbach's alpha coefficient. Cronbach′s alpha coefficient was categorized as: excellent (α ≥ 0.9), good (0.8 ≤ α < 0.9), acceptable (0.7 ≤ α < 0.8), questionable (0.6 ≤ α < 0.7), poor (0.5 ≤ α < 0.6), and unacceptable (α < 0.5) [19]. To determine if individual items correlated to the same construct, the item‐deleted Cronbach′s alpha coefficients were also calculated. Item‐total correlations (correlations between each scale item and the overall score) were used to further analyze internal consistency. Spearman‐Brown scores, which are more effective at assessing the internal consistency of scales with only a couple of items, were also included in the internal consistency [20].
2.5.2. Face and Content Validity
Initially, we used various kinds of expert opinions with experience in maternal health research (gynecologists = “1”, midwives = “3”, public health = “4”, and nursing = “1”) to evaluate the face and content validity. We also attempted to make sure that the experts had a clear understanding of and expectations for tool development. During a panel discussion, the experts offered valuable input for tool enhancement. Therefore, to encourage straightforward communication, we made a face‐to‐face approach through an expert panel discussion. Initially, the experts reviewed the item part (face validity) and concluded that the tool was suitable for assessing functioning and disability in the study region (Tigray Region). In the second step, the WHODAS‐2 tool′s content validity was checked to determine if it adequately evaluated the disability issue in the local (Tigray region) context. The item content validity index (ICVI) and scale content validity index (SCVI) are two components of the content validity index (CVI), which we used to evaluate content validity. Using a four‐level item rating system (1 = not relevant, 2 = somewhat relevant, 3 = fairly relevant, and 4 = highly significant), the CVI was determined based on the evaluation of nine experts in the disciplines of public health, midwifery, nursing, and obstetrics. While a score of three or four represented agreement, a score of one or two showed disagreement. According to the study, a CVI of ≥ 0.78 is acceptable when there are nine or more experts [21].
2.5.3. Floor and Ceiling Effects
The floor and ceiling effects were evaluated by calculating the percentage of subjects who scored the lowest or highest results on the WHODAS 2.0 items. The floor or ceiling effects were detected if more than 15% of the participants gave the lowest or highest score, respectively [22].
2.5.4. Construct Validity
Construct validity refers to how successfully an instrument measures the theoretical concept that it is intended to measure. We employed Spearman correlations to examine convergent validity between the findings of the 36‐item and 12‐item WHODAS 2.0 when there was skewed data distribution. The correlation levels were categorized as very weak (0–0.19), weak (0.20–0.39), moderate (0.40–0.59), strong (0.60–0.79), and very strong (0.80–1.00) [23]. We hypothesized a correlation between the 12‐item WHODAS 2.0 and the 36‐item WHODAS 2.0 disability assessment. The degree of disability on the 36‐item WHODAS 2.0 increases with the degree of disability on the 12‐item WHODAS 2.0 [12].
2.5.5. Confirmatory Factor Analysis (CFA)
Exploratory or CFA can be used to assess construct validity [24]. Exploratory factor analysis is employed when a researcher wants to discover the pattern of answers; in these circumstances, the structure of factors is data driven, whereas CFA starts with a hypothesis about how many factors there are or which items load on which domain or factors there are [25]. We performed the CFA since functional disability has a known dimension. The sample size for this study was 121 participants. While the Kaiser–Meyer–Olkin (KMO) index (> 0.5) and Bartlett′s test of sphericity (p < 0.05) were used to assess sample adequacy and factorability, we acknowledge that the relatively small sample size may limit the statistical power and stability of the factor structure, potentially affecting the precision of model fit indices like root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), Tucker Lewis Index (TLI), and Comparative Fit Index (CFI). The acceptable model fit was determined at RMSEA < 0.08, SRMR < 0.08, TLI > 0.90, and the CFI > 0.90) [26, 27].
3. Results
3.1. Characteristics of the Study Participants
In total, the study included 121 participants who had experienced potentially life‐threatening maternal conditions during recent childbirth. The average age was 28.95 (±6.62) years. Most of the study respondents were 115 (95.04%) married, 88 (72.73%) housewives, and 80 (66.12%) lived in urban areas (Table 1).
Table 1.
Socio‐demographic characteristics of the study participants in Tigray, Northern Ethiopia, 2024.
| Variables | Categories | Data |
|---|---|---|
| Age (years) | Mean age | 28.95 ± 6.62 |
| Educational status | Did not attend school | 19 (15.70) |
| Primary (Grades 1–8) | 19 (15.70) | |
| Secondary (Grades 9–12) | 55 (45.50) | |
| Diploma and above | 28 (23.10) | |
| Place of residence | Rural | 41 (33.90) |
| Urban | 80 (66.10) | |
| A woman's main occupation | House wife | 88 (72.73) |
| Governmental employed | 20 (16.53) | |
| Merchant | 13 (10.74) | |
| Marital status | Single | 6 (5.00) |
| Married | 115 (95.00) |
Note: Data are presented as mean ± standard deviation and frequency (percentage).
3.2. Face and Content Validity
To determine face and content validity, we consulted different types of experts. Experts followed structured steps to ensure clarity and orientation on validation expectations. They assessed the tool′s face validity, reviewing individual items for relevance and clarity. After evaluation, they concluded the instrument was suitable for assessing disability and functioning within the Tigray region's specific context. In the panel discussion process, the experts translated two WHODAS 2.0 items synonymously: “Getting out of your home” and “maintaining friendships” to “Move out of the yard/corridor” and “continuing friendships,” respectively. The second step, content validity, was ensured by providing an answer to the questions of whether the WHODAS‐2 tool was appropriate for evaluating the disability issue in the Tigray setting. The results of the content validity index from the nine participating experts were I‐CVI = 0.78–1.0 and S‐CVI = 0.96.
3.3. WHODAS 2.0 Domain Scores
The overall means (SDs) of the 32‐item and 36‐item WHODAS 2.0 were 34.60 (26.28) and 32.16 (24.96), respectively. The minimum disability means score was 12.65 in the ‘work/school activities' domain, and the maximum score was 41.78 in the ‘life activity/home' domain. The mean score for the 12‐item WHODAS 2.0 varied between 25.00 and 77.88 (Table 2).
Table 2.
Descriptive analysis of the 36‐item and 12‐item WHODAS 2.0 scores.
| 36‐item WHODAS 2.0 score | 12‐item WHODAS 2.0 score | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Domain | Mean | SD | Median | IQR | Min–Max | Mean | SD | Median | IQR | Min–Max |
| Domain 1 | 31.06 | 25.81 | 25.00 | 45.83 | 0–87.50 | 25.00 | 11.18 | 31.25 | 18.75 | 0–43.75 |
| Domain 2 | 39.87 | 25.70 | 40.00 | 37.50 | 0–95.00 | 50.00 | 22.36 | 62.50 | 37.50 | 0–87.50 |
| Domain 3 | 27.89 | 27.42 | 25.00 | 50.00 | 0–100 | 58.17 | 34.63 | 75.00 | 62.50 | 0–100 |
| Domain 4 | 30.57 | 29.74 | 25.00 | 50.00 | 0–100 | 58.00 | 33.34 | 75.00 | 62.50 | 0–87.50 |
| Domain 5.1 | 41.78 | 31.80 | 25.00 | 43.75 | 0–100 | 77.88 | 27.68 | 100.00 | 50.00 | 25–100 |
| Domain 5.2 | 12.65 | 25.93 | 0.00 | 0.00 | 0–93.75 | 62.50 | 21.50 | 75.00 | 25.00 | 25–100 |
| Domain 6 | 36.23 | 28.58 | 25.00 | 50.00 | 0–93.75 | 62.50 | 20.91 | 75.00 | 25.00 | 25–100 |
| Overall‐32 item | 34.60 | 26.28 | 25.00 | 52.73 | 0–86.71 | 36.36 | 27.65 | 27.27 | 51.14 | 0–93.18 |
| Overall‐36 item | 32.16 | 24.96 | 22.22 | 47.91 | 0–84.72 | 66.17 | 25.08 | 78.40 | 47.73 | 15.95–93.18 |
Abbreviations: IQR, interquartile range; SD, standard deviation; WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
3.4. Reliability
Each of the following domains' Cronbach's α values showed excellent internal consistency in the WHODAS‐2.0 Tigrigna version: cognition (D1 = 0.965), mobility (D2 = 0.961), self‐care (D3 = 0.925), getting along (D4 = 0.965), housework (D5.1–4 = 0.978), work or school activities (D5.5–8 = 0.917), and participation (D6 = 0.976). The 36‐item WHODAS 2.0 had an overall internal consistency score of 0.981 (Table 3).
Table 3.
Descriptive statistics and reliability of the WHODAS 2.0 Tigrigna version.
| Domain (D) | Cronbach's α | Item‐total correlation | Cronbach's α if item deleted |
|---|---|---|---|
| Cognition (D1) | 0.965 | — | — |
| Item D1.1 | — | 0.904 | 0.956 |
| Item D1.2 | — | 0.933 | 0.953 |
| Item D1.3 | — | 0.907 | 0.956 |
| Item D1.4 | — | 0.893 | 0.957 |
| Item D1.5 | — | 0.871 | 0.960 |
| Item D1.6 | — | 0.816 | 0.964 |
| Mobility (D2) | 0.961 | — | — |
| Item D2.1 | — | 0.923 | 0.946 |
| Item D2.2 | — | 0.870 | 0.955 |
| Item D2.3 | — | 0.886 | 0.952 |
| Item D2.4 | — | 0.897 | 0.951 |
| Item D2.5 | — | 0.882 | 0.954 |
| Self‐care (D3) | 0.925 | — | — |
| Item D3.1 | — | 0.913 | 0.872 |
| Item D3.2 | — | 0.835 | 0.902 |
| Item D3.3 | — | 0.796 | 0.921 |
| Item D3.4 | — | 0.807 | 0.909 |
| Getting along (D4) | 0.965 | — | — |
| Item D4.1 | — | 0.933 | 0.953 |
| Item D4.2 | — | 0.921 | 0.954 |
| Item D4.3 | — | 0.939 | 0.951 |
| Item D4.4 | — | 0.889 | 0.962 |
| Item D4.5 | — | 0.879 | 0.964 |
| Housework (D5.1‐4) | 0.978 | — | — |
| Item D5.1 | — | 0.954 | 0.968 |
| Item D5.2 | — | 0.936 | 0.973 |
| Item D5.3 | — | 0.934 | 0.974 |
| Item D5.4 | — | 0.954 | 0.968 |
| Work/school activities (D5.5‐8)a | 0.917 | — | — |
| Item D5.5 | — | 0.817 | 0.889 |
| Item D5.6 | — | 0.675 | 0.915 |
| Item D5.7 | — | 0.862 | 0.875 |
| Item D5.8 | — | 0.898 | 0.860 |
| Participation(D6) | 0.976 | — | — |
| Item D6.1 | — | 0.921 | 0.972 |
| Item D6.2 | — | 0.900 | 0.973 |
| Item D6.3 | — | 0.868 | 0.975 |
| Item D6.4 | — | 0.897 | 0.973 |
| Item D6.5 | — | 0.926 | 0.972 |
| Item D6.6 | — | 0.934 | 0.971 |
| Item D6.7 | — | 0.925 | 0.972 |
| Item D6.8 | — | 0.870 | 0.975 |
| Total score (32‐Items) | 0.990 | — | — |
| Total score (36‐Items) | 0.981 | — | — |
Abbreviations: α, Cronbach's alpha; D, domain; WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
Domain 5.5‐8 work/school activity was tested in 26 participants.
The 12‐item WHODAS 2.0 and the 11‐item WHODAS (ignoring work/school activity) had Cronbach's alpha coefficients of 0.952 and 0.973, respectively. Internal consistency of the 12 items within the domain was evaluated using the Spearman‐Brown coefficient (split half), which was 0.916 for cognitive, 0.929 for mobility, 0.926 for self‐care, 0.874 for life activities, 0.914 for participation, and 0.940 for getting along with others.
3.5. Floor and Ceiling Effect
Floor effects were found in all domains: 57%, 40%, 55%, 56%, 57%, 47%, and 45% in Domain 1, Domain 2, Domain 3, Domain 4, Domain 5a, Domain 5b, and Domain 6, respectively, whereas ceiling effects were not present in any domain and ranged from 0.0% to 0.88%.
3.6. Construct Validity
We present the correlations between the short‐form (12‐item) and full‐form (36‐item) WHODAS 2.0 assessments across various domains of functioning. The correlations demonstrated the degree of agreement between the two versions in measuring different aspects of functioning and disability, including cognition, self‐care, getting around, getting along with others, life activities, and participation. Overall, the results indicated strong correlations across most domains, suggesting that the short‐form WHODAS 2.0 is a valid alternative to the full version for assessing disability. High correlations were observed in the domains of cognition (D1), mobility (D2), self‐care (D3), getting along (D4), life activity (D5a), and participation (D6) with coefficients exceeding 0.909, indicating that both versions effectively capture this aspect of functioning. The total WHODAS 2.0 scores revealed a high correlation of 0.990, further supporting the validity of the short‐form as a comprehensive tool for evaluating overall functioning (Table 4).
Table 4.
Correlations of the full‐form (36‐item) and short‐form (12‐item) WHODAS 2.0.
| 36‐item WHODAS 2.0 (full version) | ||||||||
|---|---|---|---|---|---|---|---|---|
| 12‐items WHODAS 2.0 (short form) | Understanding and communicating (D1.1–1.6) | Getting around (D2.1–2.5) | Self‐care (D3.1–3.4) | Getting along (D4.1–4.5) | Life activities in | Participation (D6.1–6.8) | Total WHODAS 2.0 score | |
| House (D5.1–5.4) | Work/school (D5.5–5.8) | |||||||
| S3 and S6 (cognition) | 0.970 | 0.826 | 0.771 | 0.822 | 0.812 | 0.346 | 0.852 | 0.916 |
| S1 and S7 (Mobility) | 0.727 | 0.972 | 0.810 | 0.767 | 0.836 | 0.406 | 0.827 | 0.897 |
| S8 and S9 (Self‐care) | 0.739 | 0.840 | 0.970 | 0.780 | 0.817 | 0.445 | 0.804 | 0.873 |
| S10 and S11 (Getting along) | 0.827 | 0.785 | 0.748 | 0.981 | 0.798 | 0.445 | 0.804 | 0.879 |
| S2 (Life‐activities) | 0.768 | 0.829 | 0.785 | 0.858 | 0.966 | 0.539 | 0.839 | 0.890 |
| S12 (Life‐activities) | 0.598 | 0.680 | 0.539 | 0.793 | 0.796 | 0.909 | 0.909 | 0.776 |
| S4 and S5 (Participation) | 0.797 | 0.847 | 0.820 | 0.835 | 0.858 | 0.434 | 0.971 | 0.929 |
| Overall D12 | 0.883 | 0.936 | 0.887 | 0.912 | 0.911 | 0.723 | 0.925 | 0.990 |
Note: Values in bold represent the correlations between each short‐form domain and its corresponding domain in the full 36‐item WHODAS 2.0.
Abbreviation: WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
A correlation matrix between the six domains and the overall score of the 36‐item WHODAS 2.0 was developed to assess the validity of domain selection. The correlation level was strong in each case and statistically significant at p < 0.001. Very strong correlations were observed between the domains and the overall domain score of functioning and disability (r = 0.871–0.944). Strong correlations were also observed between the domains of functioning and disability (r = 0.729–0.862) (Table 5).
Table 5.
Correlations of the WHODAS 2.0 with all the domains and their summary scores.
| Domains | D1 | D2 | D3 | D4 | D5 | D6 | Overall score |
|---|---|---|---|---|---|---|---|
| D1: Cognition | 1 | — | — | — | — | — | — |
| D2: Mobility | 0.807 | 1 | — | — | — | — | — |
| D3: Self‐care | 0.729 | 0.835 | 1 | — | — | — | — |
| D4: Getting along | 0.840 | 0.813 | 0.761 | 1 | — | — | — |
| D5: Life activities | 0.793 | 0.855 | 0.810 | 0.836 | 1 | — | — |
| D6: Participation | 0.837 | 0.857 | 0.822 | 0.837 | 0.862 | 1 | — |
| Overall score | 0.907 | 0.936 | 0.871 | 0.905 | 0.915 | 0.944 | 1 |
Note: Values in bold represent the correlations between each domain and overall score.
Abbreviations: D, domain; WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
We investigated the WHODAS 2.0's theoretical validity by analyzing its internal structure, the correlations between the items in the domain and the domain itself, and the correlations between the domain's items and the overall disability score. Results showed that the domain result and the overall result had strong correlation (r = 0.817–0.966 and r = 0.629–0.927, respectively). All statistical analysis results were significant at p < 0.01 (Table 6).
Table 6.
Correlations between WHODAS 2.0 items, domains, and the overall results.
| Domain | Domain items | Domain results | Overall results |
|---|---|---|---|
| Domain 1: Cognition | Item 1 | 0.927 | 0.919 |
| Item 2 | 0.953 | 0.892 | |
| Item 3 | 0.939 | 0.858 | |
| Item 4 | 0.935 | 0.848 | |
| Item 5 | 0.916 | 0.797 | |
| Item 6 | 0.885 | 0.783 | |
| Domain 2: Mobility | Item 1 | 0.949 | 0.871 |
| Item 2 | 0.930 | 0.885 | |
| Item 3 | 0.935 | 0.898 | |
| Item 4 | 0.939 | 0.861 | |
| Item 5 | 0.931 | 0.867 | |
| Domain 3: Self‐care | Item 1 | 0.956 | 0.864 |
| Item 2 | 0.930 | 0.829 | |
| Item 3 | 0.875 | 0.759 | |
| Item 4 | 0.911 | 0.798 | |
| Domain 4: Getting along | Item 1 | 0.945 | 0.863 |
| Item 2 | 0.943 | 0.839 | |
| Item 3 | 0.944 | 0.859 | |
| Item 4 | 0.939 | 0.842 | |
| Item 5 | 0.919 | 0.877 | |
| Domain 5a: Life activity | Item 1 | 0.966 | 0.890 |
| Item 2 | 0.963 | 0.893 | |
| Item 3 | 0.963 | 0.881 | |
| Item 4 | 0.963 | 0.882 | |
| Domain 5b: Life activitya | Item 5 | 0.909 | 0.776 |
| Item 6 | 0.817 | 0.629 | |
| Item 7 | 0.892 | 0.778 | |
| Item 8 | 0.929 | 0.717 | |
| Domain 6: Participation | Item 1 | 0.902 | 0.881 |
| Item 2 | 0.894 | 0.816 | |
| Item 3 | 0.865 | 0.838 | |
| Item 4 | 0.904 | 0.844 | |
| Item 5 | 0.934 | 0.887 | |
| Item 6 | 0.937 | 0.927 | |
| Item 7 | 0.944 | 0.925 | |
| Item 8 | 0.906 | 0.849 |
Abbreviation: WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
Domain 5b: Life activity number of participants = 26.
3.7. CFA
The KMO test, which measures sample adequacy, and Bartlett's test of sphericity were used to check whether the data was suitable for factor analysis. KMO = 0.939, Bartlett's test of sphericity < 0.001, and a total variance of 81.70% were the results shown. CFA was conducted using both a one‐factor structure (short form) and a six‐factor structure (full version). For both models, the goodness‐of‐fit indices fell outside the acceptable range. For instance, the one‐factor model exhibited the following fit indices: RMSEA = 0.300, SRMR = 0.084, CFI = 0.682 and TLI = 0.604. Factor loadings in the one‐factor structure ranged from 0.62 (D2.1) to 0.90 (D5.1) (Figure 1), while those in the six‐factor structure ranged from 0.58 (D3.4) to 0.99 (D3.1) (Figure 2).
Figure 1.
A one‐factor structure diagram of the confirmatory factor analysis for WHODAS 2.0 in participants with PLTC. PLTC, potentially life‐threatening conditions; WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
Figure 2.
Six‐factor structure diagram of the confirmatory factor analysis for WHODAS 2.0 in participants with PLTC. PLTC, potentially life‐threatening conditions; WHODAS‐2, World Health Organization Disability Assessment Schedule‐II.
4. Discussion
In both public health research and clinical practice, health scales have gained significant popularity [28]. According to the WHO, evaluating health requires consideration of functional, psychological, and social dimensions, reflecting its complex and multifaceted nature. Health is difficult to evaluate since it is multidimensional and dependent on both internal and environmental influences. To evaluate health with a comprehensive and holistic approach, it is necessary to develop scales that assess different aspects of health conditions. Furthermore, for any research tool to be effective, it must demonstrate high validity and reliability in its measurements. If scales from one culture are used in another, adaptations are needed because a reliable measure of suffering in one culture may not be significant or understandable in another [29].
With certain limitations, the WHODAS 2.0 Tigrigna version confirmed acceptable scale properties. The Cronbach's alpha values of the total score and all the domains had excellent internal consistency (alpha ≥ 0.90), except for the self‐care (D3) and work/school activities (D5b), which still maintained good internal consistency with an alpha value of > 0.860. It also satisfies Nunnally's requirement, according to which Cronbach's alpha for a good scale must be more than 0.70 [19]. We verified that the 36‐item WHODAS 2.0 had good reliability in our study of women with PLTC. Compared with the other domains, the “self‐care” and work/school activity domains have slightly lower reliability values, which is not unique to our study [30, 31]. The internal consistency reliability of the WHODAS 2.0 from the general population and specific patient populations worldwide showed similar results, as the lower alpha values in the cited article were found for self‐care and work/school activity [32, 33, 34]. The reasons for lower internal consistency in the “self‐care” and work/school activities domain could be due to the small number of items in the domain. WHODAS 2.0 has fewer items in self‐care (4 items) and work/school (4 items) compared to other domains (e.g., mobility has 5, cognition has 6, participation has 8). Cronbach's alpha is sensitive to scale length, and shorter scales often yield lower reliability estimates [35].
The strong correlation (r = 0.990) between the 36‐item WHODAS 2.0 and the total score of the 12‐item suggests excellent convergent validity. Thus, indicating that the short form version effectively captures the overall disability construct measured by the full version. In the specific domains, it also indicates that there were strong correlations (cognition: r = 0.970, mobility: r = 0.972, self‐care: r = 0.970, getting along with: r = 0.981, life activities: r = 0.966, and participation: r = 0.971) between short version items and their corresponding full form domains, indicating strong convergent validity. This study aligns with other studies conducted by WHO that show the 12 items capture around 80% of the variance in the full version [12].
The results support the divergent validity of the short‐form WHODAS 2.0, with low correlations between theoretically distinct domains compared to strong within‐domain associations. This aligns with the instrument's design, which aims to differentiate between distinct functional areas while maintaining interconnected overall disability measure [12, 36]. These findings support the utility of the short‐form WHODAS 2.0 for both domain‐specific and overall disability assessments in clinical and research settings [36]. In confirmatory factor analysis, strong factor loadings show that the observed variables are adequately represented by latent factors. However, poor model fit (such as high RMSEA and low CFI/TLI) suggests that the overall model does not adequately represent the data. The small sample size may be the cause of this poor model fit. A small sample size can lead to unstable estimates and inflated factor loadings while also negatively impacting fit indices [37].
Although the KMO index (KMO = 0.939) and Bartlett's test of sphericity (< 0.001) indicated sufficient sample adequacy for factorability, these measures do not fully compensate for the limitations imposed by a small sample size, which may affect the stability of the factor structure. With this limitation, our results indicate positive findings about the validity and reliability of the Tigrigna version of the WHODAS 2.0. Therefore, we suggest that the Tigrigna version of the WHODAS 2.0 is valid for measuring functioning and disability of PLTCs in Tigray, northern Ethiopia. Validating the WHODAS 2.0 tool for PLTCs has been performed, which improves its use in analyzing long‐term functional disability, guiding future research on postpartum recovery, improving maternal health frameworks, guiding public health policies, and informing targeted clinical interventions.
Concerning the study's limitations, CFA analysis is more sample dependent to confirm the validity of unresolved subscales and enable the development of a valid scoring structure, as the one we applied in this study could be sample dependent, especially owing to the floor effect in every item and a consequence of the low response rate in some categories. Another limitation is the absence of a test–retest assessment; the current study was cross‐sectional.
5. Conclusion
This study evaluated the validity and reliability of the Tigrigna version of WHODAS 2.0 in assessing disability among women who experienced PLTC in Tigray, Ethiopia. The findings indicate that the tool has excellent internal consistency for most domains (Cronbach's α > 0.90), strong convergent validity (high correlation between 36‐item and 12‐item versions, r = 0.990), and acceptable content validity (I‐CVI = 0.78–1.0, S‐CVI = 0.96). However, confirmatory factor analysis (CFA) revealed poor model fit, likely due to the small sample size, which may have led to unstable estimates and inflated factor loadings. Despite this limitation, the Tigrigna WHODAS 2.0 effectively captures functional disability in women with PLTCs. Its adoption in clinical practice and research can enhance postpartum care and disability monitoring. Further studies with larger samples and longitudinal designs are needed to optimize its use.
Author Contributions
Fitiwi Tinsae Baykemagn: conceptualization, investigation, writing–original draft, methodology, validation, visualization, formal analysis, project administration, data curation, supervision, resources, funding acquisition, software, writing – review and editing. Girmatsion Fisseha Abreha: conceptualization, methodology, writing – review and editing. Yibrah Berhe Zelelow: conceptualization, writing–review and editing, validation, methodology. Alemayehu Bayray Kahsay: conceptualization, methodology, writing – review and editing.
Funding
The authors have nothing to report.
Ethics Statement
Ethical approval was obtained from the Institutional Review Board (IRB) of Mekelle University, college of health sciences (approval reference: MU‐IRB 2129/2023), which holds authority for ethical approval of health research in the Tigray region. The IRB approval explicitly covered pre‐consent access to patient contact information for this study, as well as the subsequent medical record review, and was obtained prior to commencing the review. The study accessed hospital registration books and medical charts at Lemlem Karl hospital, Ayder Comprehensive Specialized Hospital, Mekelle hospital, Wukro hospital, Adigrat hospital, Abi‐adi hospital, Adwa hospital, St Mary hospital, Aksum Comprehensive Specialized Hospital, and Suhul hospital in the Tigray region. These hospitals are independent entities directly affiliated with and under the oversight of the Tigray Regional Health Bureau. To ensure proper authorization for data access, a written supporting letter was obtained from the TRHB permitting the study and data access across affiliated hospitals, and institutional permission was also obtained from the administration of each participating hospital prior to data collection.
Consent
Verbal informed consent was obtained from each participant. The consent information was read aloud, questions were addressed, and verbal agreement was documented on the consent form before proceeding. Written informed consent was not required, per approval from the IRB, in light of literacy challenges in Ethiopia, including lower female literacy rates (around 48% for women aged 15–49). This method was deemed appropriate to respect participant autonomy without imposing barriers related to reading or signing.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgments
We would like to thank the MU‐HU‐NMBU‐ICP‐V project, study participants, data collectors, and experts (who participated in the translations, evaluated the semantic equivalence, and evaluated the face and content validity of the tool).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request. When a reasonable request is made, the corresponding author can provide the dataset used and/or analyzed in this study.
References
- 1.“Maternal Mortality ,” World Health Organization, last modified April 7, 2025, https://www.who.int/news-room/fact-sheets/detail/maternal-mortality.
- 2. Mekonnen W. and Gebremariam A., “Causes of Maternal Death in Ethiopia Between 1990 and 2016: Systematic Review With Meta‐Analysis,” Ethiopian Journal of Health Development 32, no. 4 (2018): 225–242, 10.4314/ejhd.v32i4. [DOI] [Google Scholar]
- 3. Baykemagn F. T., Abreha G. F., Zelelow Y. B., Berhe A. K., and Kahsay A. B., “Global Burden of Potentially Life‐Threatening Maternal Conditions: A Systematic Review and Meta‐Analysis,” BMC Pregnancy and Childbirth 24, no. 1 (2024): 11, 10.1186/s12884-023-06199-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Say L., Souza J. P., and Pattinson R. C., “Maternal Near Miss: Towards a Standard Tool for Monitoring Quality of Maternal Health Care,” Best Practice & Research Clinical Obstetrics & Gynaecology 23, no. 3 (2009): 287–296, 10.1016/j.bpobgyn.2009.01.007. [DOI] [PubMed] [Google Scholar]
- 5. Baykemagn F. T., Abreha G. F., Zelelow Y. B., and Kahsay A. B., “Determinants of Potential Life‐Threatening Maternal Conditions (PLTCS) in Tigray, Northern Ethiopia: A Case–Control Study,” Scientific Reports 14 (2024): 30022, 10.1038/s41598-024-81415-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Filippi V., Ganaba R., Baggaley R. F., et al., “Health of Women After Severe Obstetric Complications in Burkina Faso: A Longitudinal Study,” Lancet 370, no. 9595 (2007): 1329–1337, 10.1016/S0140-6736(07)61574-8. [DOI] [PubMed] [Google Scholar]
- 7. Soma‐Pillay P., Suleman F. E., Makin J. D., and Pattinson R. C., “Cerebral White Matter Lesions After Pre‐Eclampsia,” Pregnancy Hypertension: An International Journal of Women's Cardiovascular Health 8 (2017): 15–20, 10.1016/j.preghy.2017.02.001. [DOI] [PubMed] [Google Scholar]
- 8. Koblinsky M., Chowdhury M. E., Moran A., and Ronsmans C., “Maternal Morbidity and Disability and Their Consequences: Neglected Agenda in Maternal Health,” Journal of Health, Population, and Nutrition 30, no. 2 (2012): 124–130, 10.3329/jhpn.v30i2.11294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Carroll A., “Book Review: World Report on Disability,” Irish Medical Journal 105, no. 5 (2012): 158. [Google Scholar]
- 10. Tenaw Z., Gari T., and Gebretsadik A., “The Burden of Disabilities in Sidama National Regional State, Ethiopia: A Cross‐Sectional, Descriptive Study,” PLoS One 18, no. 7 (2023): e0288763, 10.1371/journal.pone.0288763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Chala M. B., Mekonnen S., Andargie G., et al., “Prevalence of Disability and Associated Factors in Dabat Health and Demographic Surveillance System Site, Northwest Ethiopia,” BMC Public Health 17, no. 1 (2017): 762, 10.1186/s12889-017-4763-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Üstün T. B., Kostanjsek N., Chatterji S., and Rehm J., Measuring Health and Disability: Manual for WHO Disability Assessment Schedule (World Health Organization, 2010). [Google Scholar]
- 13. Flaherty L., Zimmerman D., and Hansen T. H., “Further Serological Analysis of the Qa Antigens: Analysis of an Anti‐H‐2.28 Serum,” Immunogenetics 6, no. 1 (1978): 245–251, 10.1007/BF01563914. [DOI] [Google Scholar]
- 14. World Health Organization , Health in 2015: From MDGs, Millennium Development Goals to SDGs, Sustainable Development Goals (World Health Organization, 2015). [Google Scholar]
- 15. Federal Ministry of Health , The Federal Democratic Republic of Ethiopia Health Sector Transformation Plan: 2015/16–2019/20 (Federal Ministry of Health, 2015). [Google Scholar]
- 16. Chan G. J., Daniel J., Getnet M., et al., “Gaps in Maternal, Newborn, and Child Health Research: A Scoping Review of 72 Years in Ethiopia,” Journal of Global Health Reports 5 (2021): e2021033, 10.29392/001c.22125. [DOI] [Google Scholar]
- 17. Birhane B. M., Assefa Y., Belay D. M., et al., “Interventions to Improve the Quality of Maternal Care in Ethiopia: A Scoping Review,” Frontiers in Global Women's Health 5 (2024): 1289835, 10.3389/fgwh.2024.1289835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Yoseph A., Teklesilasie W., Guillen‐Grima F., and Astatkie A., “Individual‐ and Community‐Level Determinants of Maternal Health Service Utilization in Southern Ethiopia: A Multilevel Analysis,” Women's Health 19 (2023): 17455057231218195, 10.1177/17455057231218195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. George D. and Mallery P., IBM SPSS Statistics 23 Step by Step (Routledge, 2016), 10.4324/9781315545899. [DOI] [Google Scholar]
- 20. Eisinga R., Grotenhuis M., and Pelzer B., “The Reliability of a Two‐Item Scale: Pearson, Cronbach, Or Spearman‐Brown?,” International Journal of Public Health 58, no. 4 (2013): 637–642, 10.1007/s00038-012-0416-3. [DOI] [PubMed] [Google Scholar]
- 21. Yusoff M. S. B., “Abc of Content Validation and Content Validity Index Calculation,” Education in Medicine Journal 11, no. 2 (2019): 49–54, 10.21315/eimj2019.11.2.6. [DOI] [Google Scholar]
- 22. Lim C. R., Harris K., Dawson J., Beard D. J., Fitzpatrick R., and Price A. J., “Floor and Ceiling Effects in the OHS: An Analysis of the NHS PROMs Data Set,” BMJ Open 5, no. 7 (2015): e007765, 10.1136/bmjopen-2015-007765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Papageorgiou S. N., “On Correlation Coefficients and Their Interpretation,” Journal of Orthodontics 49, no. 3 (2022): 359–361, 10.1177/14653125221076142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Schreiber J. B., Nora A., Stage F. K., Barlow E. A., and King J., “Reporting Structural Equation Modeling and Confirmatory Factor Analysis Results: A Review,” Journal of Educational Research 99, no. 6 (2006): 323–338, 10.3200/JOER.99.6.323-338. [DOI] [Google Scholar]
- 25. Newsom J. T., Shaw B. A., August K. J., and Strath S. J., “Physical Activity‐Related Social Control and Social Support in Older Adults: Cognitive and Emotional Pathways to Physical Activity,” Journal of health psychology 23, no. 11 (2018): 1389–1404, 10.1177/1359105316656768. [DOI] [PubMed] [Google Scholar]
- 26. Kline R. B. and Little T. D., Principles and Practice of Structural Equation Modeling, 3rd ed. (Guilford Press, 2011). [Google Scholar]
- 27. Tabachnick B. G. and Fidell L. S., Using Multivariate Statistics, 6th ed. (Pearson, 2013). [Google Scholar]
- 28. McDowell I., Measuring Health: A Guide to Rating Scales and Questionnaires (Oxford University Press, 2006), 10.1093/acprof:oso/9780195165678.001.0001. [DOI] [Google Scholar]
- 29. Üstün T. B., Chatterji S., Kostanjsek N., et al., “Developing the World Health Organization Disability Assessment Schedule 2.0,” Bulletin of the World Health Organization 88, no. 11 (2010): 815–823, 10.2471/BLT.09.067231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Park S. H., Demetriou E. A., Pepper K. L., et al., “Validation of the 36‐Item and 12‐Item Self‐Report World Health Organization Disability Assessment Schedule II (WHODAS‐II) in Individuals With Autism Spectrum Disorder,” Autism Research 12, no. 7 (2019): 1101–1111, 10.1002/aur.2115. [DOI] [PubMed] [Google Scholar]
- 31. Silveira C., Parpinelli M. A., Pacagnella R. C., et al., “Validation of the 36‐Item Version of the WHO Disability Assessment Schedule 2.0 (WHODAS 2.0) for Assessing Women's Disability and Functioning Associated With Maternal Morbidity,” Revista Brasileira de Ginecologia e Obstetrícia/RBGO Gynecology and Obstetrics 39, no. 2 (2017): 044–052, 10.1055/s-0037-1598599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Aslan Kunt D. and Dereboy F., “Validity and Reliability of the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) in Turkish Psychiatry Patients and Healthy Controls,” Turk Psikiyatri Dergisi 29, no. 4 (2018): 248–257. [PubMed] [Google Scholar]
- 33. Yen C.‐F., Hwang A.‐W., Liou T.‐H., et al., “Validity and Reliability of the Functioning Disability Evaluation Scale‐Adult Version Based on the WHODAS 2.0: 36 Items,” Journal of the Formosan Medical Association 113, no. 11 (2014): 839–849, 10.1016/j.jfma.2014.08.008. [DOI] [PubMed] [Google Scholar]
- 34. Barreto M. C. A., Moraleida F. R. J., Graminha C. V., Leite C. F., Castro S. S., and Nunes A. C. L., “Functioning in the Fibromyalgia Syndrome: Validity and Reliability of the WHODAS 2.0,” Advances in Rheumatology 61, no. 1 (2021): 58, 10.1186/s42358-021-00216-1. [DOI] [PubMed] [Google Scholar]
- 35. Tavakol M. and Dennick R., “Making Sense of Cronbach's Alpha,” International Journal of Medical Education 2 (2011): 53–55, 10.5116/ijme.4dfb.8dfd. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Federici S., Bracalenti M., Meloni F., and Luciano J. V., “World Health Organization Disability Assessment Schedule 2.0: An International Systematic Review,” Disability and Rehabilitation 39, no. 23 (2017): 2347–2380, 10.1080/09638288.2016.1223177. [DOI] [PubMed] [Google Scholar]
- 37. Wolf E. J., Harrington K. M., Clark S. L., and Miller M. W., “Sample Size Requirements for Structural Equation Models: An Evaluation of Power, Bias, and Solution Propriety,” Educational and Psychological Measurement 73, no. 6 (2013): 913–934, 10.1177/0013164413495237. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request. When a reasonable request is made, the corresponding author can provide the dataset used and/or analyzed in this study.