BILAG-2004 INDEX: A RELIABLE TOOL FOR ASSESSMENT OF SLE ACTIVITY

C. Gordon,C. Yee , D.A. Isenberg , V. Farewell , A. Prabu , K. Sokoll , L. Teh , A. Rahman , I.N. Bruce , B. Griffiths , M. Akil , N. McHugh , D. D'Cruz , M.A. Khamashta , S. Bowman , P. Maddison , A. Zoma , E. Allen Department of Rheumatology, University of Birmingham, Birmingham, Centre for Rheumatology Research, University College London, London, MRC Biostatistic Unit, University of Cambridge, Cambridge, Department of Rheumatology, University of Leeds, Leeds, Department of Rheumatology, Blackburn Royal Infirmary, Blackburn, ARC Epidemiology Unit, University of Manchester, Manchester, Department of Rheumatology, Freeman Hospital, Newcastle upon Tyne, Department of Rheumatology, Sheffield Teaching Hospitals NHS Trust, Sheffield, Department of Rheumatology, Royal National Hospital for Rheumatic Diseases NHS Trust, Bath, Lupus Research Unit, St Thomas' Hospital, London, Department of Rheumatology, University Hospital Birmingham NHS Trust, Birmingham, Department of Rheumatology, University of Wales, Bangor, Department of Rheumatology, Lanarkshire Acute Hospitals NHS Trust, Lanarkshire, United KingdomObjectives: To test inter-rater reliability of BILAG-2004 index for assessment of SLE activityMethods: SLE patients were recruited from 11 centres. They were assessed by 2 raters separately in routine clinical practice. Disease activity was determined using BILAG-2004 index. 2 exercises were performed: changes were made to the index after the first exercise (E1) and additional training provided to raters before the second exercise (E2). E1 and E2 involved 12 and 14 raters respectively. Reliability was assessed using level of agreement, kappa statistic, intraclass correlation coefficient (ICC) and analysis of disagreement. Major disagreement was defined as score difference of A & C/D/E and B & D/E between raters, and minor disagreement was score difference of A & B or B & C between raters.Results: 97 patients (90% women, 74% Caucasian, 8% Afro-Caribbean, 13% South Asian) were recruited for E1; 97 patients (90% women, 68% Caucasian, 16% Afro-Caribbean, 11% South Asian) were recruited for E2. Mean age was 42.3 years (18.5 - 82.2) for E1 and 43.7 years (17.7 - 75) for E2. Mean disease duration was 9.4 years (0 - 32.1) for E1 and 10 years (0 - 34.8) for E2. The total ICC for E1 was 0.45 which improved to 0.64 in E2. There were 43 (44.3%) major and 28 (28.9%) minor disagreements in E1, while these were 24 (24.7%) and 34 (35.1%) respectively in E2. The low kappa and ICC despite good agreement in Eye, GIT and Constitution (in E2) was due to inactivity in these systems in most patients. Agreement, Kappa and ICC for both Reliability Exercises (E1 and E2) Agreement kappa ICC E1 E1 (weighted) E2 (weighted) E1 (weighted) E2 (weighted) E1 E2 Constitution 48.5 (69.6) 81.4 (90.2) 0.18 (0.18) 0.20 (0.31) 0.09 0.37 Skin 69.1 (80.4) 72.2 (84.3) 0.50 (0.55) 0.58 (0.65) 0.40 0.46 Neuropsy 74.2 (84.0) 91.8 (93.0) 0.36 (0.38) 0.59 (0.62) 0.34 0.60 Musculoskel 51.6 (72.4) 56.7 (77.3) 0.26 (0.35) 0.34 (0.47) 0.43 0.55 Cardioresp 90.7 (92.0) 90.7 (92.3) 0.53 (0.58) 0.44 (0.45) 0.18 0.38 GIT 96.9 (97.4) 99.0 (99.0) -0.01(-0.01) 0 (0) 0 0.08 Eye 95.9 (96.9) 99.0 (99.0) 0 (0) 0.66 (0.49) 0 0.20 Renal 96.9 (98.5) 95.9 (97.9) 0.91 (0.94) 0.80 (0.88) 0.97 0.99 Haematology 91.5 (95.7) 88.7 (94.3) 0.84 (0.85) 0.79 (0.82) 0.75 0.86 Conclusion: BILAG-2004 is a reliable index to assess SLE activity. There was improvement in agreement, kappa statistic, intraclass correlation coefficient and major disagreements from E1 to E2. The use of a well defined glossary and training of raters is essential to ensure the optimal performance of the index.Citation: Ann Rheum Dis, volume 65, supplement II, year 2006, page 63Session: SLE – clinical aspects