Related Experiment Videos
Validity of retrospective disease activity assessment in systemic lupus erythematosus
A Arce-Salinas1, M H Cardiel, J Guzmán
1Department of Immunology and Rheumatology, Instituto Nacional de la Nutricón Salvador Zubirán, Mexico City, Mexico.
Objective:
To evaluate the validity of retrospective disease activity assessment derived from clinical charts.
Methods:
We prospectively evaluated 37 patients with systemic lupus erythematosus (SLE) in 90 visits using the SLE Disease Activity Index (SLEDAI), the Mexican SLEDAI (Mex-SLEDAI), and the Lupus Activity Criteria Count (LACC) indices. Routine clinical observations were written by rheumatologists blind to index scores. These notes were reviewed 2 years later to obtain retrospective index scores and their validity was assessed using prospective scores as the standard. Statistical analysis was by Spearman's rank correlation coefficient (rs), Wilcoxon matched pairs test, kappa statistic, and intraclass correlation coefficient (ri). We calculated the sensitivity and specificity of retrospective indices to detect active disease.
Results:
Median retrospective scores were lower in all indices: SLEDAI (4 VS 2, p =0.004, RS = 0.68, ri = 0.30); Mex-SLEDAI (2 vs 1, p < 0.0003, rs = 0.79, ri = 0.31); and LACC (1 vs 1, p = 0.007, rs = 0.65, ri = 0.21). Used to detect active SLE, the retrospective SLEDAI had a sensitivity of 0.68 and a specificity of 0.86; corresponding values for the Mex-SLEDAI were 0.72 and 0.91, and for the LACC, 0.77 and 0.76.
Conclusion:
Retrospective disease activity indices tended to provide lower scores than prospective evaluations. They often missed patients with mildly active disease, but when positive they were good predictors of disease activity.