Development of the United Kingdom Evaluation of Daily Activities Questionnaire in Rheumatoid Arthritis using Rasch Analysis

Hammond, Alison; Tennant, Alan; Tyson, Sarah; Nordenskiold, Ulla; Gill, Rachel

Abstract

Background/Purpose:
The Evaluation of Daily Activity Questionnaire (EDAQ) is a patient reported measure of activity/activity limitations in Rheumatoid Arthritis (RA) (Nordenskiold et al 1996). The UK version includes 138 items in 14 activity of daily living and participation domains (Eating/Drinking; Personal Care; Dressing; Bathing; Cooking; Moving Indoors; House Cleaning; Laundry; Moving and Transfers; Communication; Moving Outdoors; Gardening/Household Maintenance; Caring; and Hobbies/Leisure/ Social Activities). Each domain is split into two sections: one (A) which scores without aids, alternate methods or help; and another (B) which scores items with aids or alternate methods. All items are scored on a 0–3 scale (no difficulty to unable to do). Our aim was to identify if the UK-EDAQ fits the Rasch model in an RA cohort.
Methods:
Participants were recruited from Rheumatology clinics. Data from each domain were assessed for initial unidimensionality by a Confirmatory Factor Analysis (CFA). Each domain was analysed separately for sections A and B (where section B overrides section A when applicable). A Root Mean Square Error of Approximation (RMSEA) of 0.10 and below (mediocre fit) was considered adequate as prepepartion for Rasch analysis. Rasch analysis involved testing stochastic ordering (fit); local independence (response dependency and unidimensionality) and properties of invariance across groups (gender, age and employment status: differential item functioning (DIF). RUMM2030 software was used.
Results:
383 people with RA participated: 286 women and 97 men; average age 60.38 (SD 11.18) years; RA duration 13.2 years SD 10.72; 118 (31%) were employed. Average pain (10 point VAS) = 4.99 (SD 2.59) and fatigue 5.61 (SD 2.53). CFA of the 14 domain sections (A&B) indicated unidimensionality after adjustment for local dependency (correlated errors) within each domain. All domains achieved RMSEA < 0.10. All domains satisfied Rasch model expectations after adjustments (by the use of testlets) for local dependency (Chi-square p>0.05). Virtually all items had ordered thresholds, and where this was not the case, the disordering was often marginal. Differential Item Functioning (DIF) by age, gender and employment status was also largely absent, with a few notable exceptions. For example, in the domain 'gardening and household maintenance' the item 'climbing ladders' showed significant DIF by age, with older people showing greater problems with this at any given level of the trait (ANOVA p<0.05). Most domains showed a Person Separation Index reliability (PSI) consistent with individual use, even after adjustment for local dependency (PSI range 0.75–0.94). Where values were lower, this was largely a consequence of skewed data and the presence of a substantive floor effect which affects the PSI; whereas classical alpha values remained high. All domains supported strict unidimensionality.
Conclusion:
Analysis of the UK- EDAQ support a 14 domain, two component structure (self care and mobility) where each domain, and both components, (independently for sections A and B) satisfy Rasch model requirements after adjustment for local dependency.

Citation

Hammond, A., Tennant, A., Tyson, S., Nordenskiold, U., & Gill, R. (2011). Development of the United Kingdom Evaluation of Daily Activities Questionnaire in Rheumatoid Arthritis using Rasch Analysis. Arthritis Care and Research, 63,

• Journal Article Type: Extended Abstract
• Conference Name: Abstracts of the American College of Rheumatology/Association of Rheumatology Health Professionals Annual Scientific Meeting
• Conference Location: Chicago, Illinois
• Publication Date: 2011-11
• Deposit Date: Oct 23, 2023
• Journal: Arthritis & Rheumatism-arthritis Care & Research
• Print ISSN: 2151-464X
• Publisher: Wiley
• Peer Reviewed: Peer Reviewed
• Volume: 63