To test the interrater reliability of the revised British Isles Lupus Assessment Group 2004 (BILAG-2004) index for the assessment of systemic lupus erythematosus (SLE) activity.
Patients with SLE were recruited from 11 centers. Two physician raters separately assessed the patients' disease activity using the BILAG-2004 index in routine clinical practice. Scores ranged from A (for very active disease) to E (for inactivity). Two reliability exercises were performed. Changes were made to the index after the first exercise (E1), and additional training was provided to the raters before the second exercise (E2). E1 and E2 involved 12 and 14 raters, respectively. Interrater reliability was assessed using kappa statistics and intraclass correlation coefficients. Levels of agreement and the extent of major disagreement were also examined. Major disagreement was defined as a score difference between raters of A versus C, D, or E or B versus D or E.
For each exercise, 97 patients were recruited. In E1, the mean age of the patients was 42.3 years (range 18.5–82.2 years), 89.7% were women, and 74.2% were white, 8.2% were Afro-Caribbean, and 13.4% were South Asian, and in E2, the mean age was 43.7 years (range 17.7–75 years), 90.7% were women, and 68% were white, 15.5% were Afro-Caribbean, and 11.3% were South Asian. The mean disease duration was 9.4 years (range 0–32.1 years) for patients in E1 and 10 years (range 0–34.8 years) in E2. There was improvement in the interrater reliability and the level of agreement from E1 to E2. Further improvement was achieved after removal of poorly performing items.
The BILAG-2004 index is a reliable tool to assess SLE activity. The use of a well-defined glossary and training of raters are essential to ensure the optimal performance of the index.