TY - JOUR
T1 - Psychometric properties and cross-language equivalence of the revised Bristol Rheumatoid Arthritis Fatigue and the Rheumatoid Arthritis Impact of Disease scales in rheumatoid arthritis
AU - Oude Voshaar, Martijn A.H.
AU - Bode, Christina
AU - Hewlett, Sarah
AU - Kirwan, John
AU - Gossec, Laure
AU - van de Laar, Mart A.F.J.
N1 - Springer deal
PY - 2019/9/1
Y1 - 2019/9/1
N2 - Objective: To assess psychometric properties and cross-language measurement equivalence of six versions of the Bristol Rheumatoid Arthritis Fatigue Scale (BRAF-MDQ) and the Rheumatoid Arthritis Impact of Disease Score (RAID in rheumatoid arthritis (RA). Methods: Both questionnaires were completed by French (n = 206), German (n = 206), Dutch (n = 317), Spanish (n = 157), Swedish (n = 170) and UK (n = 210) RA patients. The presence of cross-language differential item functioning (DIF) was examined using the generalized partial credit model. The impact of DIF on the item and total scores was examined by comparing DIF unadjusted and DIF adjusted expected item and scale scores. IRT-based methods were used to assess psychometric properties of the instruments. Results: 11 of the 20 BRAF-MDQ (55%) and 4 of the 7 RAID items (57%) exhibited significant DIF in at least one of the six countries. The mean number of items with DIF per country was 2.6 for BRAF-MDQ and 1.1 for RAID. However, the impact of DIF on the total RAID and BRAF-MDQ scores, as well as the BRAF subscales, was found to be negligible at the group level. Only for the BRAF physical subscale was there evidence of minor DIF. Marginal reliabilities of BRAF-MDQ (0.93) and RAID (0.89) were excellent, and precise scores could be obtained across the spectrum of disease impact and fatigue scores measured by these PROMs. Conclusion: This study supports the cross-language measurement equivalence of BRAF-MDQ and RAID and provides further support for the psychometric properties of these measures in RA.
AB - Objective: To assess psychometric properties and cross-language measurement equivalence of six versions of the Bristol Rheumatoid Arthritis Fatigue Scale (BRAF-MDQ) and the Rheumatoid Arthritis Impact of Disease Score (RAID in rheumatoid arthritis (RA). Methods: Both questionnaires were completed by French (n = 206), German (n = 206), Dutch (n = 317), Spanish (n = 157), Swedish (n = 170) and UK (n = 210) RA patients. The presence of cross-language differential item functioning (DIF) was examined using the generalized partial credit model. The impact of DIF on the item and total scores was examined by comparing DIF unadjusted and DIF adjusted expected item and scale scores. IRT-based methods were used to assess psychometric properties of the instruments. Results: 11 of the 20 BRAF-MDQ (55%) and 4 of the 7 RAID items (57%) exhibited significant DIF in at least one of the six countries. The mean number of items with DIF per country was 2.6 for BRAF-MDQ and 1.1 for RAID. However, the impact of DIF on the total RAID and BRAF-MDQ scores, as well as the BRAF subscales, was found to be negligible at the group level. Only for the BRAF physical subscale was there evidence of minor DIF. Marginal reliabilities of BRAF-MDQ (0.93) and RAID (0.89) were excellent, and precise scores could be obtained across the spectrum of disease impact and fatigue scores measured by these PROMs. Conclusion: This study supports the cross-language measurement equivalence of BRAF-MDQ and RAID and provides further support for the psychometric properties of these measures in RA.
KW - UT-Hybrid-D
KW - Disease impact
KW - Fatigue
KW - Item response theory
KW - Patient reported outcome
KW - Reliability
KW - Rheumatoid arthritis
KW - Validity
KW - Differential item functioning
UR - http://www.scopus.com/inward/record.url?scp=85065136280&partnerID=8YFLogxK
U2 - 10.1007/s11136-019-02188-8
DO - 10.1007/s11136-019-02188-8
M3 - Article
AN - SCOPUS:85065136280
VL - 28
SP - 2543
EP - 2552
JO - Quality of life research
JF - Quality of life research
SN - 0962-9343
IS - 9
ER -