Arthritis Care Res (Hoboken). 2026 Mar 9. doi: 10.1002/acr.80034. Online ahead of print.
ABSTRACT
OBJECTIVE: This study aimed to compare the performance of LFA-REAL, BILAG, and SLEDAI in detecting the clinician’s perception of improvement and worsening disease and identifying patients requiring treatment escalation.
METHODS: This was a prospective, observational study conducted at four centers. Adults with SLE were evaluated at a baseline and follow-up visit. Disease activity was assessed using SLEDAI, BILAG, and LFA-REAL at both visits by the same physician. At the follow-up visit, the Clinician’s Global Impression of Change (CGIC) was recorded to classify patients as improved, worsened, or unchanged. ROC-curve analysis was used to evaluate each instrument in detecting clinician-rated change gauged by CGIC with optimal cut-off points determined using the Youden-Index.
RESULTS: Of 163 patients enrolled, 145 (89%) completed a follow-up visit. Based on CGIC, 23% improved, 16% worsened, and 61% remained stable. For detecting CGIC-defined improvement, LFA-REAL had an AUC of 0.85 (95%CI: 0.77-0.92), compared to 0.75 (0.66-0.85) for BILAG and 0.74 (0.64-0.84) for SLEDAI. For CGIC-defined worsening, AUCs were 0.86 (0.77-0.96), 0.79 (0.67-0.90), and 0.76 (0.66-0.87). Of those with CGIC-defined worsening, 8 (35%) required treatment escalation. LFA-REAL identified a difference of ⋝10mm in all eight of these patients, compared to 4 by BILAG and 2 by SLEDAI.
CONCLUSIONS: LFA-REAL showed comparable performance relative to SLEDAI and BILAG in detecting clinician-rated change. There was a statistically significant advantage over SLEDAI for detection of improvement. These findings support its utility as a metric for disease activity in clinical practice and research.
PMID:41801044 | DOI:10.1002/acr.80034
