Background Fatigue is highly prevalent, disabling, and remains under-reported and undertreated in rheumatoid arthritis (RA). PROMIS measures were developed to assess common symptoms and impacts of health domains relevant to chronic conditions. PROMIS offers item banks for computer adaptive testing (CAT), paper-based short forms (SF) and single item scales. The PROMIS fatigue item bank assesses both experience (from mild tiredness to an overwhelming, sustained sense of exhaustion) and impact on physical, mental and social activities. However, it is unclear whether generic fatigue instruments are sensitive enough to assess the symptoms and impact of fatigue in RA clinical trials or to guide decision-making in clinical care.
Objectives To evaluate the performance of three forms of PROMIS fatigue measures in RA patients seen in an academic inflammatory arthritis center.
Methods Data are from the first 139 RA patients enrolled in an ongoing study integrating PROs into routine care at an arthritis center. Three PROMIS fatigue measures (CAT [5-8 items], 7-item SF and the single fatigue item from the Global Health scale (0-5)) were administered with other legacy measures. Routine clinical measures of disease activity were recorded. Spearman correlations were used to evaluate convergent validity and pairwise comparisons were used to examine evidence of a dose response relationship across disease activity levels.
Results Participants were mostly female (81%) and white (85%) with a mean (SD) age of 56 (13) and RA duration of 12 (9) yrs. Overall mean (SD) t-scores for the PROMIS CAT and SF were 54.1 (10.2) and 53.6 (8.9); the single item was 2.6 (.9). PROMIS measures correlated highly with a fatigue VAS (.78-.87), moderately with disease activity indicators including pain (.60-.68), patient global (.66-.72), physical function (.53-.69) and CDAI (.54-.65), and weakly with MD global (.36-.39). There was evidence of a dose-response relationship in relation to disease activity levels (see table).
Conclusions Our study contributes evidence of the feasibility and construct validity of using generic PROMIS fatigue instruments in RA. While the PROMIS Fatigue CAT was most strongly associated with legacy measures and offers greatest precision, the SF and single item performed adequately and are useful when paper or briefer versions are needed. Evaluation of responsiveness and validity across diverse RA patient groups and settings (trials vs clinical care, group vs. individual level change) is needed.
Disclosure of Interest None declared