Refine
Year of publication
- 2019 (4) (remove)
Document Type
Language
- English (4) (remove)
Has Fulltext
- no (4)
Is part of the Bibliography
- no (4)
Keywords
- Rasch model (2)
- Functional Independence Measure (1)
- ICF (1)
- SF-36 (1)
- WHODAS 2.0 (1)
- activities of daily living (1)
- assessment instruments (1)
- disabilities of the arm, shoulder and hand (1)
- health status (1)
- outcome assessment (1)
Background
Limitations in upper limb functioning are common in Musculoskeletal disorders and the Disabilities of the Arm, Shoulder and Hand scale (DASH) has gained widespread use in this context. However, various concerns have been raised about its construct validity and so this study seeks to examine this and other psychometric aspects of both the DASH and QuickDASH from a modern test theory perspective.
Methods
Participants in the study were eligible if they had a confirmed diagnosis of Rheumatoid Arthritis (RA). They were mailed a questionnaire booklet which included the DASH. Construct validity was examined by fit to the Rasch measurement model. The degree of precision of both the DASH and QuickDASH were considered through their Standard Error of Measurement (SEM).
Results
Three hundred and thirty-seven subjects with confirmed RA took part, with a mean age of 62.0 years (SD12.1); 73.6% (n = 252) were female. The median standardized score on the DASH was 33 (IQR 17.5–55.0). Significant misfit of the DASH and QuickDASH was observed but, after accommodating local dependency among items in a two-testlet solution, satisfactory fit was obtained, supporting the unidimensionality of the total sets and the sufficiency of the raw (ordinal or standardized) scores.
Conclusion
Having accommodated local response dependency in the DASH and QuickDASH item sets, their total scores are shown to be valid, given they satisfy the Rasch model assumptions. The Rasch transformation should be used whenever all items are used to calculate a change score, or to apply parametric statistics within an RA population.
Significance and innovations
Most previous modern psychometric analyses of both the DASH and QuickDASH have failed to fully address the effect of a breach of the local independence assumption upon construct validity.
Accommodating this problem by creating ‘super items’ or testlets, removes this effect and shows that both versions of the scale are valid and unidimensional, as applied with a bi-factor equivalent solution to an RA population.
The Standard Error of Measurement of a scale can be biased by failing to take into account the local dependency in the data which inflates reliability and thus making the SEM appear better (i.e. smaller) than the true value without bias.
Background: The Extended Barthel Index (EBI), consisting of the original Barthel Index plus 6 cognitive items, provides a tool to monitor patients’ outcomes in rehabilitation. Whether the EBI provides a unidimensional metric, thus can be reported as a valid sum-score, remains to be examined.
Objective: To examine whether the EBI can be reported as unidimensional interval-scaled metric for neurological and musculoskeletal rehabilitation.
Methods: Rasch analysis of a calibration sample of 800 cases from neurological or musculoskeletal rehabilitation in 2016 in Switzerland.
Results: In the baseline analysis no fit to the Rasch Model was achieved. When accommodating local dependencies with a testlet approach satisfactory fit to the Rasch Model was achieved, and an interval scale transformation table was created.
Conclusion: The results support the reporting of adapted EBI total scores for both rehabilitation groups by applying the interval scaled transformation table presented in this study.
Abstract
Objective: Since the 1990s the Functional Independence Measure (FIM™) was believed to measure 2 different constructs, represented by its motor and cognitive subscales. The practice of reporting FIM™ total scores, together with recent developments in the understanding of the influence of locally dependent items on fit to the Rasch model, raises the question of whether the FIM™ 18-item version can be reported as a unidimensional interval-scaled metric.
Design: Rasch analysis of the FIM™ using testlet approaches to accommodate local response dependency.
Patients: A calibration sample containing 946 cases of data from 11,103 patients undergoing neurological or musculoskeletal rehabilitation in Switzerland in 2016.
Results: Baseline analysis and the traditional testlet approach showed no fit with the Rasch model. When items were grouped into 2 testlets, fit to the Rasch model was achieved, indicating unidimensionality across all 18 items. A transformation table to convert FIM™ raw ordinal scores to the corresponding Rasch interval scaled values was created.
Conclusion: This study provides evidence that FIM™ total scores represent a unidimensional set of items, supporting their use in clinical practice and outcome reporting when applying the respective transformation table. This provides a basis for standardized reporting of functioning.
Lay Abstract
The aim of this study was to look in detail at the FIM™, an assessment tool often used for patients undergoing rehabilitation. Some users report the FIM™ as 2 scores: one related to motor tasks, the other to cognitive tasks; others recommend reporting it as a single score including both motor and cognitive tasks. This study explored whether it is statistically meaningful to sum all the points into a single FIM™ total score. The results support the current practice of summing the points into a single total score for patients undergoing musculo-skeletal and neurological rehabilitation. The results also allowed an interval scale to be derived from the FIM™, enabling a broad range of calculations to be made using the FIM™ score, such as calculating the change in FIM™ outcomes from the time a patient is admitted to a rehabilitation clinic until their discharge.
Background: The International Classification of Functioning, Disability and Health is the international standard for describing and monitoring functioning. While the categories, the units of the classification, were not designed with measurement in mind, the hierarchical structure of the classification lends itself to the possibility of summating categories into some higher order domain. Focusing on the chapters of d4 Mobility, d5 Self-Care and d6 Domestic Life, this study seeks to ascertain if qualifiers rating of categories (0-No problem to 4-Complete problem) within those chapters can be summated, and whether such derived measurement is consistent with estimates obtained from well-known instruments which purport to measure the same constructs.
Methods: The current study applies secondary analysis to data previously collected in the context of validating Core Sets for stroke, rheumatoid arthritis, and osteoarthritis. Data included qualifier-based ratings of the categories in the Core Sets, and the physical functioning sub-scale of the Short-Form 36, and the World Health Organization Disability Assessment Schedule 2.0. To examine qualifier-comparator scale item agreement Kappa statistics were used. To identify whether appropriate gradients of the comparator scales were observed across qualifier levels, an Independent Sample Median Test of the ordinal scores was deployed. To investigate the internal validity of the summated ICF categories, the Rasch model was applied.
Results: Data from 2,927 subjects from Europe, Australasia, Middle East and South America were available for analysis; 36.3% had experienced a stroke, 35.8% osteoarthritis, and 27.9% had rheumatoid arthritis. The items from the Short-Form 36 could not match directly the qualifier categories as the former had only 3 response options. The Kappa between World Health Organization Disability Assessment Schedule 2.0 items and categories was low. For all qualifiers, a significant (<0.001) overall gradient was observed across the comparator scales. Only in few of the World Health Organization Disability Assessment Schedule 2.0 items could no discrete level be detected. The aggregation of the qualifiers at the Chapter and higher order levels mostly revealed fit to the Rasch model. Almost all ICF qualifiers showed ordered thresholds suggesting that the current structure and response options of the qualifiers worked as intended.
Conclusions: The findings of this study provide supporting evidence for the use of the professionally rated categories and associated qualifiers to measure functioning.
Implication for Rehabilitation
- This study provides evidence that functioning data can be collected directly with the International Classification of Functioning, Disability and Health (ICF) by using the ICF categories as items and the ICF qualifiers as rating scale.
- The findings of this study show the aggregated ratings of ICF categories from the chapters d4 Mobility, d5 Self-care, and d6 Domestic life capture a broader spectrum of the construct than the corresponding summated items from the SF36-Physical Function sub-scale and the corresponding items of the World Health Organization Disability Assessment Schedule 2.0.
- This study illustrates the potential of building quantitative measurement by aggregating ICF categories and their qualifier ratings into meaningful domains.