Comparison of the construct validity and reproducibility of four different types of patient-reported outcome measures (PROMs) in patients with rheumatoid arthritis

Abstract

Patient-reported outcome measures (PROMs) are increasingly used in the assessment of patients with rheumatoid arthritis (RA). The aim of this study was to assess the construct validity and reproducibility of four types of PROMs (numerical rating scale (NRS), visual analogue scale (VAS), verbal rating scale (VRS), and Likert scale). In addition, patients' preference for one of these PROMs was measured. Patients with stable RA and no change in pain medication or steroid medication (n = 211) were asked to complete a questionnaire directly following, and 5 days after their outpatient visit. The questionnaire measured the validity of the PROMs in four domains (pain, fatigue, experienced disease activity, and general well-being). To assess construct validity, Pearson correlation coefficients were calculated using the Disease Activity Score-3. To assess reproducibility, intraclass correlation coefficients (ICCs) were calculated. Correlation coefficients between the different PROMs within each domain were compared. There were no differences regarding construct validity between the different PROM scale types. Regarding reproducibility, VAS and NRS scored better in the domains pain and experienced disease activity. Patients chose NRS as preferred measurement method, with the VAS in second place. The four scale types of PROMs appeared to be equally valid to assess pain, fatigue, experienced disease activity, and general well-being. VAS and NRS appeared to be more reliable for measuring pain and experienced disease activity. The patients favoured NRS. In combination with other practical advantages of NRS over VAS, we do advise the NRS as the preferred scale type for PROMs.

Keywords: Outcome assessment; PROM; Psychometric properties; Rheumatoid arthritis; Self management.