Psychometric and cross-cultural validity of the measure of activity performance of the hand (map-hand) across three European countries (Switzerland, Turkey and the UK)

Horton, M.; Prior, Y.; Hammond, A.; Tonga, E.; Niedermann Schneider, K.; Gilanliogullari, N.; Kjeken, I.

Abstract

Background: The MAP-Hand scale comprises 18 patient-generated items which assess hand activity among people with rheumatoid arthritis (RA). All items are scored on a 0–3 scale and are summed into a total score (0 to 54).
Objectives: The MAP-Hand is already fully validated within a UK population [1], and this study aims to assess the validity, stability and comparability across German-speaking Swiss and Turkish RA samples.
Methods: The MAP-Hand was assessed using Rasch model psychometric assessment methodology. Within the original UK validation, all items were assessed individually, then combined into four testlets (super-items) in order to account for observed local dependency among the items. This analytic strategy
was replicated, where the psychometric properties of the scale items were assessed separately within each country-specific samples, and then across a merged sample of all three countries. A differential item function (DIF) analysis was used to assess scale stability across countries.
Results: When the countries are considered separately, although slight differences are apparent, the operation of the instrument appears to be very similar across Switzerland, Turkey, and the UK. Although there is a degree of scale misfit in all countries, the scale-sample targeting and the scale reliability are both excellent
(Person Separation Index = 0.93-0.95; Cronbach’s Alpha = 0.96-0.97), and the response categories function well across all items. However, there is a large degree of inter-item dependency present, and the MAP-Hand displays apparent multidimensionality within all three countries. Additionally, some individual items display Rasch model misfit to some extent, although these items are variable
across the three countries. When the items were re-configured into four testlets, this accounted for the inter-item dependency, also improving the model fit and unidimensionality of the scale within all three countries separately, as well as when the data is merged into a single dataset. Some significant country-DIF was observed
on the ‘Dressing’ testlet (comprised of items 1-3), but this was found to have no significant effect on total person score estimates. Despite the testlet re-structuring creating a narrower scale range, the scale-sample targeting and scale reliability remain high (Person Separation Index = 0.86-0.91; Cronbach’s Alpha = 0.93).
Conclusion: It is recommended that the four-testlet structure is implemented across all three countries, as this offers a pragmatic approach to the utility and comparative scoring of the MAPHAND across the UK, Switzerland and Turkey. Although the re-structuring creates a narrower scale range and apparent reduction in the reliability, this is indicative of the initial dependency that is present, and suggests that the original reliability values are over-inflated.
REFERENCE:
[1] Prior, Y., Tennant, A., Tyson, S. et al. Measure of activity performance of the hand (MAP-Hand) questionnaire: linguistic validation, cultural adaptation and psychometric testing in people with rheumatoid arthritis in the UK. BMC Musculoskelet Disord 19, 275 (2018). https://doi.org/10.1186/s12891-018-2177-5

Citation

Horton, M., Prior, Y., Hammond, A., Tonga, E., Niedermann Schneider, K., Gilanliogullari, N., & Kjeken, I. (2023). Psychometric and cross-cultural validity of the measure of activity performance of the hand (map-hand) across three European countries (Switzerland, Turkey and the UK). Annals of the Rheumatic Diseases, 82(Suppl 1), 568.1-568. https://doi.org/10.1136/annrheumdis-2023-eular.4119

• Journal Article Type: Extended Abstract
• Acceptance Date: May 1, 2023
• Online Publication Date: May 30, 2023
• Publication Date: 2023-06
• Deposit Date: Jun 15, 2023
• Publicly Available Date: Jun 19, 2023
• Journal: Annals of the Rheumatic Diseases
• Print ISSN: 0003-4967
• Electronic ISSN: 1468-2060
• Publisher: BMJ Publishing Group
• Peer Reviewed: Peer Reviewed
• Volume: 82
• Issue: Suppl 1
• Pages: 568.1-568
• DOI: https://doi.org/10.1136/annrheumdis-2023-eular.4119
• Keywords: Rheumatology