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.

Changes in Corticosteroid Use after DMARD Initiation in Patients with Rheumatoid Arthritis.

Kawai5,  Vivian K., Grijalva6,  Carlos, Arbogast6,  Patrick, Curtis4,  Jeffrey R., Solomon1,  Daniel Hal, Delzell3,  Elizabeth, Ouellet-Hellstrom2,  Rita

Brigham and Womens Hospital, Boston, MA
University of Alabama
University of Alabama–Brimingham, Birmingham, AL
Vanderbilt University, Nashville, TN
Vanderbilt University


Disease modifying anti-rheumatic drugs (DMARDs) and corticosteroids are both used to treat rheumatoid arthritis (RA). Corticosteroids have dose-related side effects, and therefore, clinicians strive to minimize their use. Biologic and non-biologic DMARD use reduces RA disease activity and thus would be expected to result in decreased use of corticosteroids. There is, however, little information about the effect of DMARD therapies on corticosteroid use. Thus, we examined the hypothesis that initiation of DMARD therapies decrease corticosteroid use.


We assembled 4 retrospective cohorts of patients with RA from 1998 to 2005 enrolled in Tennessee's Medicaid Program (TennCare), Kaiser Permanente Northern California (KPNC), Pennsylvania Pharmaceutical Assistance Contract for the Elderly (PACE), and multi-State Medicaid programs (MAX) and identified 5 mutually exclusive patient groups initiating DMARD regimens: new methotrexate (MTX) without (new MTX) or with (MTX) use of other non-biologic DMARDs in the previous year; new hydroxychloroquine and/or sulfasalazine with use of MTX in the previous year (HCQ/SSZ); new leflunomide with use of MTX in the previous year (LEF); and new TNF-a antagonists. We used McNemar's test to assess within-person differences in any use of oral corticosteroids (>= 1 prescription) during the 6 months before versus 6–12 months after starting one of the study DMARD regimens.


We identified 32476 initiators of study DMARD regimens who had a full year of baseline and one follow-up year of data available: 13% from TennCare, 18% from KPNC, 7% from PACE and 62% from MAX. In all groups, the percentage of users of corticosteroids increased in the 0–6 months before new DMARD initiation compared to the previous 6 months, likely representing worsening of the disease that led to the new DMARD regimen. In most groups, the percentage of users of corticosteroid then decreased significantly by 6 to 12 months after DMARD initiation (Figure 1). Across the 4 datasets, the absolute decrease in % of RA patients using corticosteroids ranged from 8.0% to 10.9% in new MTX users and 9.3% to 12.1% in MTX users (all P<0.05). Absolute changes ranged from 7.1% to 11.5% in group HCQ/SSZ users, 4.0% to 6.9% in group LEF users, and 3.6% to 11.5% in TNF-a antagonists (all changes <0.05 except for PACE participants who had both the lowest % corticosteroid use at DMARD initiation and the lowest absolute change).


The percentage of patients with RA receiving corticosteroids increased in the 6 months prior to the initiation of a new DMARD regimen and then decreased 6–12 months after initiation. Decreased corticosteroid use may be due to DMARD effectiveness or other reasons for disease improvement.

Figure 1. Changes in percent of users of corticosteroid after initiation of RA therapies.

To cite this abstract, please use the following information:
Kawai, Vivian K., Grijalva, Carlos, Arbogast, Patrick, Curtis, Jeffrey R., Solomon, Daniel Hal, Delzell, Elizabeth, et al; Changes in Corticosteroid Use after DMARD Initiation in Patients with Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :54
DOI: 10.1002/art.27823

Abstract Supplement

Meeting Menu