Psychometric evaluation of the Rheumatoid Arthritis Symptom Questionnaire (RASQ) in an observational study.

ABSTRACT

OBJECTIVE: To describe the psychometric performance of the scores produced by the Rheumatoid Arthritis Symptom Questionnaire (RASQ), a new patient-reported outcome (PRO) questionnaire developed to assess the signs and symptoms of rheumatoid arthritis (RA). METHODS: Adult subjects with clinically confirmed RA completed a set of questionnaires (including the RASQ) at an initial study visit (Day 1), and then completed the RASQ and the Patient Global Impression of Change (PGI-C) on their own on Day 8. Demographic and health data were summarized using descriptive statistics, and psychometric analyses were conducted, including acceptability, item and scale distribution, reliability (internal consistency and test-re-test reliability), and construct-related validity (convergent validity and known-groups methods). RESULTS: In total, 200 subjects (females = 61.5%; white = 72.0%; and age [mean] = 60.7 years) with RA were recruited across the US and included in the analysis. There were no missing data recorded for the RASQ, and scores were well distributed for both timepoints. The RASQ Total Symptom Score surpassed the threshold (α ≥ 0.70) for internal consistency at Day 1 (α = 0.967) and test-re-test score reliability (intra-class correlation coefficient [ICC] > 0.70) (ICC = 0.960). Convergent validity analyses demonstrated that the RASQ items and Total Symptom Score had high correlations (convergent validity) with other PRO questionnaires. Known-groups methods demonstrated that the RASQ (Total Symptom Score and all single items) can differentiate between clinically distinct groups. CONCLUSIONS: The RASQ is capable of producing psychometrically sound scores when administered to adults with RA.