Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Assessing Systemic Lupus Erythematosus Disease Severity and Disease Flares: Development of a Claims-Based Algorithm.

Garris2,  Cindy P., Jhingran2,  Priti M., Engel-Nitz4,  Nicole M., Riedel5,  Aylin A., Goldberg4,  George, Bass1,  Damon L., Dennis3,  Gregory J.

GlaxoSmithKline, King of Prussia, PA
GlaxoSmithKline, Research Triangle Park, NC
Human Genome Sciences, Rockville, MD
i3 Innovus, Eden Prarie, MN
i3 Innovus, Eden Prarie, NC


Disease severity and disease flares are based primarily on clinical presentation in patients with systemic lupus erythematosus (SLE). Disease severity is not clearly defined but physicians often describe it as a relationship between disease activity and the organs involved. In clinical trials, disease activity and disease flares are usually assessed by composite scores which integrate both clinical and serologic variables. The ability to describe and evaluate populations of SLE patients in real-world settings has been limited due to the lack of validated methodology to identify disease severity or flares in administrative claims data.


To develop and validate algorithms to identify SLE disease severity and the occurrence and severity of SLE flares using administrative claims.


This retrospective, observational study (BLM001HO) used medical and pharmacy data (1/1/2004 to 12/31/2008) from a large managed care health plan. Patients meeting the following criteria were included in the analysis: at least 3 years of continuous enrollment in the health plan (1 year pre-index baseline period and 2 years follow up period); were between the ages of 18 and 64; and evidence of SLE. Evidence of SLE during the study period was based upon a combination of diagnosis code (ICD-9 code 710.0x), visits to rheumatologists, and filled prescriptions for lupus treatment medications (corticosteroids, antimalarials, and immunosuppressives).


A total of 2990 SLE patients were identified in the claims database (92% female, mean age 44 years). Initial algorithms for identifying SLE disease severity and flare severity were developed based upon the literature and clinical consultation with 2 rheumatologists and 1 internal medicine physician. Medical and pharmacy claims for a random sample of SLE patients were reviewed to understand the patients' profile of health services utilization, conditions diagnosed, use of services, types of providers visited, and medication use. Following evaluation of these patient profiles, the initial algorithms were further refined. The resulting disease severity and flare algorithms were applied to the SLE patients identified in the claims database.

SLE Disease Severity Algorithm

MildNot Moderate or High Severity
ModerateModerate Rx OR Moderate Medical Condition
Rx• Oral corticosteroid dose >=7.5 mg/day to <60 mg/day
 • Immunosuppressive agent (excluding cyclophosphamide)
Medical Condition:Cardiorespiratory: myocarditis, pericarditis, pleurisy/pleural effusion, vasculitis (excluding aortitis)
 Constitutional: hepatitis (non-viral)
 Gastrointestinal: acute pancreatitis, lupus enterititis/colitis
 Hematology: hemolytic anemia
 Musculoskeletal: ischemic necrosis of bone
 Neuropsychiatric: demyelinating syndrome/acute inflammatory demyelinating polyradiculoneuropathy, mononeuropathy/polyneuropathy, myelopathy, pseudotumor cerebri, seizure
HighIntensive RX OR Severe Medical Condition
Rx:• Oral corticosteroid dose >=60 mg/day
 • Cyclophosphamide
 • Rituximab
Medical Condition:Cardiorespiratory: aortitis, arterial/ venous thrombosis, cardiac tamponade, pulmonary hemorrhage, stroke/TIA
 Gastrointestinal: intestinal pseudo-obstruction
 Neuropsychiatric: acute confusional state/psychosis, aseptic meningitis, cranial neuropathy
 Ophthalmic: optic neuritis
 Renal: end stage renal disease (ESRD)
SLE Flare Severity Algorithm 
RX Initiation• Hydroxychloroquine or other antimalarial
• Oral corticosteroid with prednisone-equivalent dose of <=7.5 mg/day 
 • Non-immunosuppressive therapy (NSAIDS, androgens)
RX Initiation• Oral corticosteroid with prednisone-equivalent dose >7.5 mg/day & <= 40 mg/day
 • Immunosuppressive therapy (excluding cyclophosphamide)
Medical• ER visit with primary diagnosis of SLE (710.0x) or SLE-related condition
 • Office visit for new SLE-related condition*
RX Initiation• Oral corticosteroid with prednisone-equivalent dose >40 mg/day
 • Cyclophosphamide
Medical• Hospital stay with primary diagnosis of SLE (710.0x) or SLE-related condition*
*SLE-related conditions: disease severity conditions above or any of following: arthritis/arthralgia, dry eye/tear film insufficiency, rash, low white blood cell count (leukopenia, neutropenia, lymphocytopenia), lymph node enlargement, myalgia/myositis, urticaria

Applying the algorithm, 26% of patients were classified as mild, 52% were moderate and 22% had high disease severity over the 2 year follow-up period. Over the 2-year period, 71% of patients had at least one mild flare, 86% had at least one moderate flare and 20% had one or more severe flares.


This study has identified an algorithm that may be used to identify SLE disease severity and the occurrence and severity of SLE flares, using a combination of medical and pharmacy administrative claims. Applying this clinically-based algorithm to other claims-based studies of patients with SLE may enable an improved evaluation of SLE patients in real-world populations.

To cite this abstract, please use the following information:
Garris, Cindy P., Jhingran, Priti M., Engel-Nitz, Nicole M., Riedel, Aylin A., Goldberg, George, Bass, Damon L., et al; Assessing Systemic Lupus Erythematosus Disease Severity and Disease Flares: Development of a Claims-Based Algorithm. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :757
DOI: 10.1002/art.28525

Abstract Supplement

Meeting Menu