Rheumatology Advance Access published online on September 24, 2009
Rheumatology, doi:10.1093/rheumatology/kep183
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Numerical scoring for the Classic BILAG index
1MRC Biostatistics Unit, University of Cambridge, Cambridge, 2Rheumatology Research Group, University of Birmingham, Birmingham, 3Centre for Rheumatology, University College London, London, 4Department of Rheumatology, Royal Blackburn Hospital, Blackburn, 5Department of Rheumatology, Freeman Hospital, Newcastle-upon-Tyne, 6ARC Epidemiology Unit, University of Manchester, Manchester, 7Department of Rheumatology, North West Wales NHS Trust, Bangor, 8Department of Rheumatology, Sheffield Teaching Hospitals NHS Trust, Sheffield, 9Department of Rheumatology, Royal National Hospital for Rheumatic Diseases NHS Trust, Bath, 10Lupus Research Unit, St Thomas’ Hospital, London and 11Department of Rheumatology, University of Wales, Bangor, UK.
Correspondence to:
Caroline Gordon, Rheumatology Research Group, School of Immunity and Infection, College of Medical and Dental Sciences, The Medical School, University of Birmingham, Birmingham B15 2TT, UK. Email: p.c.gordon{at}bham.ac.uk
 |
Abstract |
|---|
Objective. To develop an additive numerical scoring scheme for the Classic BILAG index.
Methods. SLE patients were recruited into this multi-centre cross-sectional study. At every assessment, data were collected on disease activity and therapy. Logistic regression was used to model an increase in therapy, as an indicator of active disease, by the Classic BILAG score in eight systems. As both indicate inactivity, scores of D and E were set to 0 and used as the baseline in the fitted model. The coefficients from the fitted model were used to determine the numerical values for Grades A, B and C. Different scoring schemes were then compared using receiver operating characteristic (ROC) curves. Validation analysis was performed using assessments from a single centre.
Results. There were 1510 assessments from 369 SLE patients. The currently used coding scheme (A = 9, B = 3, C = 1 and D/E = 0) did not fit the data well. The regression model suggested three possible numerical scoring schemes: (i) A = 11, B = 6, C = 1 and D/E = 0; (ii) A = 12, B = 6, C = 1 and D/E = 0; and (iii) A = 11, B = 7, C = 1 and D/E = 0. These schemes produced comparable ROC curves. Based on this, A = 12, B = 6, C = 1 and D/E = 0 seemed a reasonable and practical choice. The validation analysis suggested that although the A = 12, B = 6, C = 1 and D/E = 0 coding is still reasonable, a scheme with slightly less weighting for B, such as A = 12, B = 5, C = 1 and D/E = 0, may be more appropriate.
Conclusions. A reasonable additive numerical scoring scheme based on treatment decision for the Classic BILAG index is A = 12, B = 5, C = 1, D = 0 and E = 0.
KEY WORDS: SLE, Outcome measures, Disease activity, BILAG, Statistics, Global score, Regression model, Treatment decision
Submitted 23 October 2008;
revised version accepted 1 June 2009.
CiteULike 
Connotea 
Del.icio.us What's this?