Background Systemic lupus erythematosus (SLE) has a broad spectrum of clinical manifestations and laboratory abnormalities, posing formidable diagnostic challenges.
Objectives To compare clinical outcomes and healthcare resource utilization between patients (pts) with SLE diagnosed early versus late.
Methods Patients aged 18-64 years with diagnosed SLE were identified from the MarketScan® Commercial Claims and Encounters database, between 1/2000-6/2010. Confirmed diagnosis required >1 claim for rheumatologist visits with a diagnostic code for SLE (ICD-9 code: 710.0x) and, in some cases, a requirement for ≥1 claim for a typical SLE medication (1). SLE onset date was defined during the baseline period based on antinuclear antibody tests and presence of clinical manifestations characteristic of SLE according to American College of Rheumatology Classification Criteria. Patients were stratified into Early or Late Diagnosis groups based on time between probable SLE onset and diagnosis (<6 or ≥6 months, respectively). Groups of early and late diagnosis pts were matched 1:1 using propensity score based on age, gender, year of SLE diagnosis, geographic region, and health plan type. The post-SLE diagnosis period was used to compare outcomes between early and late diagnosis groups. Poisson regression was used to compare SLE flare rates, and all-cause and SLE-related inpatient and outpatient visits. Results were stratified at ≤3, ≤6 and >6 months post-SLE diagnosis to evaluate effect modification.
Results 9,713 pts fulfilled the study criteria; pts in the Early Diagnosis and Late Diagnosis group had a median SLE onset time of 2 and 10 months prior to diagnosis, respectively. Mean follow-up after SLE diagnosis was 28 months (SD 21 months, Median: 21 months). After matching, there were 4,274 pts in each group for analysis, and the Late Diagnosis group had more hypertension, kidney disease and osteopenia/osteoporosis. During the post-SLE diagnosis period, pts in the Early Diagnosis group had significantly lower rates of flares than patients in the Late Diagnosis group (any severity RR=0.95 [95%CI 0.94-0.96]; mild, RR=0.95 [0.93-0.96]; moderate, RR=0.93 [0.91-0.96]; severe, RR=0.83 [0.78-0.89]). Pts in the Early Diagnosis group had significantly lower all-cause and SLE-related hospitalizations (all-cause, RR=0.72 [0.68-0.77]; SLE-related, RR=0.76 [0.69-0.83]) and outpatient visits (all-cause, RR=0.77 [0.76-0.77]; SLE-related RR=0.87 [0.85-0.88]) compared to pts in the Late Diagnosis group. The lower all-cause and SLE-related hospitalization rate among pts in the Early versus Late Diagnosis group was more evident over time: all-cause, RR=0.93 [0.82-1.05]; SLE-related, RR=0.98 [0.84-1.15] at ≤3 months; all-cause, RR=0.82 [0.73-0.91]; SLE-related, RR=0.89 [0.76-1.03] at ≤6 months; all-cause, RR=0.65 [0.60-0.71]; SLE-related, RR=0.65 [0.57-0.74] for the remaining months after diagnosis.
Conclusions Early diagnosis of SLE was associated with better clinical outcomes and reduced resource utilization. This finding needs to be further explored within the context of background disease activity.
Engel-Nitz, et al.; Distinguishing Levels of Disease Flare in Patients with SLE Using Administrative Claims Data. [abstract]. Arthritis Rheum 2010;62 Suppl 10:764 DOI: 10.1002/art.28532
Disclosure of Interest C. Korves Grant/Research support from: GSK, F. Laliberte Grant/Research support from: GSK, E. Suthoff Grant/Research support from: GSK, R. Wei Grant/Research support from: GSK, A. Oglesby Employee of: GSK, G. Dennis Employee of: Human Genome Sciences, M. Duh Grant/Research support from: GSK