An item analysis according to the Rasch model of the German 12-item WHO Disability Assessment Schedule (WHODAS 2.0)

Age-related differences in psychometric properties of World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0): a retrospective analysis

The aim of this study was to detect a possible age-related differential item functioning (DIF) of the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) among people with musculoskeletal disorders. This was a cross-sectional study of 1739 consecutive patients with chronic musculoskeletal pain at a university clinic. Difficulty and discrimination parameters were calculated for each of 12 items of the WHODAS 2.0 using item response theory. The presence of DIF was assessed both numerically and graphically. When considering differences in both difficulty and discrimination, a statistically significant DIF was observed for all the items except for the item 'washing' (P values <0.001). The DIF was mostly uniform. Items 'standing', 'household responsibilities', 'learning a new task', 'emotional affectedness', 'concentrating', 'washing' and 'dressing' showed better precision among older patients. Items 'walking', 'dealing with strangers', 'maintaining a friendship' and 'day-to-day work' were more informative among younger patients. Difficulty and discrimination parameters were alike for all 12 items. The DIF was mostly uniform. These results can be taken into account when applying the WHODAS 2.0 to people with musculoskeletal complaints, especially when the studied groups are predominated by people of certain age.

Dimensionality and invariance of ADL, IADL, BI-M2/WG-SS, and GALI in large surveys in France (2008-2014) and implications for measuring disability in epidemiology

BACKGROUND

The epidemiological investigation and surveillance of disability requires well-constructed, invariant, and, if possible, exchangeable measures. However, the current or recommended measures have not been thoroughly investigated with respect to these issues. Here we examined the dimensional structure and invariance of four measures across sociodemographic groups: Activities of Daily Living (ADL), Instrumental Activities of Daily Living (IADL), Budapest Initiative Mark 2 (BI-M2) and Washington Group on Disability Statistics Short Set (WG-SS), and Global Activity Limitation Indicator (GALI).

METHODS

We used data from three large nationwide representative surveys conducted in France between 2008 and 2014. The surveys included these four measures and classical and modern approaches (correlations, principal component analysis, Rasch modeling) were used to assess their dimensional structure as well as their invariance through differential item functioning (DIF) for sociodemographic characteristics. Polytomous logistic regression models were used to assess gradients in health inequalities associated with these measures.

RESULTS

For many items of ADL, IADL, and BI-M2/WG-SS, we consistently observed disordered response thresholds, rejection of unidimensionality, and DIF evidence for sociodemographic characteristics across the survey samples. Health inequality gradients were erratic. In addition, it was impossible to identify a common continuum for GALI, ADL, IADL, and BI-M2/WG-SS or their constituent items.

CONCLUSION

This study warns against the current practice of investigating disability in epidemiology using measures that are unsuitable for epidemiological use, incommensurable, and inadequate regarding the basic requisites of dimensionality and invariance. Developing invariant measures and equating them along a common continuum to enlarge the common bases of measurement should therefore be a priority.

WHODAS measurement properties for women with dysmenorrhea

BACKGROUND

There is an association of dysmenorrhea with human functioning and disability. However, no patient-reported outcome measure has been developed to assess this construct in women with dysmenorrhea. WHODAS 2.0 has been recognized as an important generic patient-reported outcome information of physical function and disability. Thus, the aim of this study was to assess the measurement properties of the WHODAS 2.0 in women with dysmenorrhea.

METHODS

This is an online and cross-sectional study conducted with Brazilian women aged 14 to 42 years with self-report of dysmenorrhea in the last three months. According to COSMIN, structural validity was evaluated by exploratory and confirmatory factor analysis; internal consistency by Cronbach's Alpha; measurement invariance by multigroup confirmatory factor analysis between geographic regions of Brazil; and construct validity by correlating WHODAS 2.0 to the Numerical Rating Scale for pain severity.

RESULTS

One thousand three hundred and eighty-seven women (24.7 ± 6.5 years) with dysmenorrhea participated in the study. WHODAS 2.0 presented a single factor by exploratory factor analysis and adequate model by confirmatory factor analysis (CFI = 0.924, TLI = 0.900, RMSEA = 0.038), excellent internal consistence (α = 0.892) for all items and an invariancy across geographic regions (ΔCFI ≤ 0.01 and ΔRMSEA < 0.015). Correlation between WHODAS 2.0 and numerical rating scale was positive and moderate (r = 0.337).

CONCLUSION

WHODAS 2.0 has a valid structure to assess functioning and disability related to dysmenorrhea in women.

Measuring disability in amyotrophic lateral sclerosis/motor neuron disease: the WHODAS 2.0-36, WHODAS 2.0-32, and WHODAS 2.0-12

Aim: To investigate whether the World Health Organization Disability Assessment Schedule 2.0 (WHODAS) can provide interval level measurement of disability in Amyotrophic Lateral Sclerosis (ALS), allowing parametric analyses. Methods: Data on the WHODAS 12, 32, and 36-item versions, from 1120 patients studied at one or more time points, were fit to the Rasch model and comparisons made against ALSFRS-R, King's staging, and mortality. Trajectory modeling was undertaken for a newly diagnosed (≤6 months) cohort of 454 individuals. Results: Total scores for WHODAS 32 and 36-item versions can be converted to interval level measurement suitable for individual clinical use, and the 12-item WHODAS total for group use. The 36-item version is shown to be equivalent to the 32-item version. Expected correlations were seen with King's staging, ALSFRS-R, and EQ-5D-5L. Trajectory analysis of disability (WHODAS 2.0) showed three clearly demarcated groups with differences in King's staging, depressive symptomatology and mortality, but not age. Conclusions: The WHODAS 2.0 is a brief patient reported outcome measure which can be used to measure disability in ALS. Provided the patient answers all 36 (32 if not working) items, the conversion table produces an interval level estimate for parametric analyses. The different trajectories demonstrated from diagnosis support the concept of a prodromal period, and suggest the WHODAS 2.0 could be used for surveillance of at risk populations, such as those with genetic predisposition.

Psychometric evaluation of the Patient Assessment of Chronic Illness Care instrument using Item analysis according to the Rasch model

RATIONALE, AIMS AND OBJECTIVES

The Patient Assessment of Chronic Illness Care (PACIC) instrument assesses patient care in alignment with the chronic care model. The aim of the study was to comprehensively validate the PACIC using the Rasch model. A special focus was placed on the investigation of local dependence (LD), differential item functioning (DIF) and targeting.

METHOD

This secondary analysis utilized data of 760 patients with a diagnosis of diabetes who had participated in the Swiss CoDiab-VD cohort study. The psychometric properties of the French PACIC-version were evaluated using the Rasch model. DIF was investigated in relation to age, gender, education, year of recruitment into the CoDiab-VD cohort study, type of diabetes and whether patients got an injectable antidiabetic drug or not.

RESULTS

The initial analysis of the PACIC revealed poor fit to the Rasch model (χ² -p < 0.001) with response dependency being the most prominent problem. After combining the items into two testlets (testlet 1: Items 1-11; testlet 2: Items 12-20), good overall model fit was found (χ² -p = 0.77) as well as good reliability (Person Separation Index = 0.85) and targeting. DIF with regard to whether patients got an injectable antidiabetic drug or not was found for testlet 2. However, the size of this DIF was regarded as not being substantial.

CONCLUSION

The PACIC is a well-targeted, reliable unidimensional instrument to assess patient care in alignment with the chronic care model in patients with diabetes. It is free of substantial DIF. The PACIC-20 sum score can hence be used in clinical practice for individual diagnostic. For evaluation purposes like assessment of change or group evaluations, the usage of the interval-scale level person parameters is recommended as it permits using parametric statistical analyses and provides a more accurate picture about the actual amount of change.