Patient self-assessment and physician's assessment of rheumatoid arthritis activity: which is more realistic in remission status? A comparison with ultrasonography

The objective of this study was to compare disease activity assessed by the patient, the physician and musculoskeletal US in patients with RA in clinical remission. We evaluated 69 patients with RA in clinical remission according to their attending rheumatologist. Tenderness and swelling in 28 joints were blindly assessed by patients and physicians. The presence of B-mode and Doppler synovitis was blindly investigated in the above joints. The DAS28 and Simplified Disease Activity Index (SDAI) were calculated. The percentage of patients in remission according to the self-derived DAS28 (26.1%) was significantly less than that according to the physician-derived DAS28 (52.2%) (P < 0.0005). There was no significant difference in the percentage of patients in remission according to the self-derived SDAI (14.5%) and the physician-derived SDAI (11.6%) (P = 0.172). We found moderate agreement between the patient-derived and physician-derived DAS28 and SDAI [intraclass correlation coefficient (ICC) = 0.620 and ICC = 0.678, respectively]. Agreement between patient and physician was better for the tender joint count (TJC; ICC = 0.509) than for the swollen joint count (SJC; ICC = 0.279). The mean (S.D.) count for B-mode synovitis [4.09 (3.25)] was significantly greater than the SJC assessed by both the patient and physician [2 (3.71) and 1.42 (2.03), respectively] (P < 0.0005 and P = 0.033, respectively). We found moderate agreement between the physician-assessed SJC and the joint count for Doppler synovitis (ICC = 0.528). Patient-assessed and physician-assessed overall RA activity showed acceptable agreement. Patient self-assessment overestimated disease activity determined by the DAS28. At the patient level, physician-assessed joint swelling showed an acceptable concordance with Doppler US synovitis.