A1 Journal article (refereed)
The structural validity of the Finnish version of the Disabilities of the Arm, Shoulder and Hand : a Rasch model analysis (2020)
Ikonen, J., Hulkkonen, S., Ryhänen, J., Häkkinen, A., Karppinen, J., & Repo, J. P. (2020). The structural validity of the Finnish version of the Disabilities of the Arm, Shoulder and Hand : a Rasch model analysis. Hand Therapy, 25(1), 3-10. https://doi.org/10.1177/1758998320907116
JYU authors or editors
Publication details
All authors or editors: Ikonen, J.; Hulkkonen, S.; Ryhänen, J.; Häkkinen, A.; Karppinen, J.; Repo, J. P.
Journal or series: Hand Therapy
ISSN: 1758-9983
eISSN: 1758-9991
Publication year: 2020
Volume: 25
Issue number: 1
Pages range: 3-10
Publisher: Sage Publications
Publication country: United Kingdom
Publication language: English
DOI: https://doi.org/10.1177/1758998320907116
Publication open access: Not open
Publication channel open access:
Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/74557
Abstract
The construct validity of the Disabilities of the Arm, Shoulder and Hand questionnaire (DASH) has previously been questioned. The purpose of this study was to evaluate the measurement properties of the Finnish version of the DASH for assessing disability in patients with hand complaints using Rasch Measurement Theory.
Methods
A cohort of 193 patients with typical hand and wrist complaints were recruited at a surgery outpatient clinic. The DASH scores were analysed using the Rasch model for differential item functioning, unidimensionality, fit statistics, item residual correlation, coverage/targeting and reliability.
Results
In the original DASH questionnaire, the item response thresholds were disordered for 2 of 30 of the items. The item fit was poor for 9 of 30 of the items. Unidimensionality was not supported. There was substantial residual correlation between 87 pairs of items. Item reduction (chi square 95, degrees of freedom 50, p < 0.001) and constructing two testlets led to unidimensionality (chi square 0.64, degrees of freedom 4, p = 0.96). Person separation index was 0.95. The testlets had good fit with no differential item functioning towards age or gender.
Conclusion
Unidimensionality of the original Finnish version of the DASH was not supported, meaning the questionnaire seems to gauge traits other than disability alone. Hence, the clinician must be careful when trying to measure change in patients’ scores. Item reduction or the creation of testlets did not lead to good alternatives for the original Finnish DASH. Differential item functioning showed that the original Finnish scale exhibits minor response bias by age in one item. The original Finnish DASH covers different levels of ability well among typical hand surgery patients.
Keywords: hands; wrists; surgical treatment; treatment outcomes; disabilities; measuring instruments (indicators); validation; Rasch model
Free keywords: DASH; validation; Rasch; Finnish; disability; outcome
Contributing organizations
Ministry reporting: Yes
VIRTA submission year: 2020
JUFO rating: 1
