Background There have been fundamental changes in the treatment of juvenile idiopathic arthritis (JIA) over the last decade, including the introduction and widespread use of biologic DMARDs. However, the utilization of DMARDs in clinical practice has not been well-studied.

Objectives We characterized DMARD utilization in the treatment of JIA in clinical practice in the United States on a national level and determined patient factors associated with medication use.

Methods The CARRA Registry is an observational longitudinal data capture study that encompasses all major pediatric rheumatic diseases and more than 50 active CARRA clinical sites that represent most pediatric rheumatology centers in the US. We analyzed cross-sectional baseline enrollment data for all children with prevalent JIA from May 2010 through May 2011. Current and prior medications were combined to analyze “ever use”. Non-biologic DMARDs comprised methotrexate, leflunomide, and sulfasalazine. We used parsimonious backward stepwise logistic regression models to calculate odds ratios (OR) to estimate associations between clinical patient factors and medication use.

Results We identified 2,748 children with JIA with a median disease duration of 3.9 years (interquartile range 1.8 – 7.2). Overall, 2,023 (74%) had ever received a non-biologic DMARD and 1,246 (45%) had ever received a biologic DMARD. Among 2,502 children without systemic arthritis, methotrexate use was most strongly associated with uveitis (OR 5.2; 95% confidence interval [3.6-7.6]), cyclic citrullinated peptide antibodies (OR 4.5 [1.7-12]), and extended oligoarthritis (OR 4.1 [2.5-6.6] compared to persistent oligoarthritis). Among children without systemic arthritis, biologic DMARD use was most strongly associated with rheumatoid factor positive (RF+) polyarthritis (OR 4.3 [2.9-6.6]), psoriatic arthritis (OR 3.0 [2.0-4.4]), and uveitis (OR 2.8 [2.1-3.7]). Only 8% of these biologic DMARD users had not ever received non-biologic DMARDs; this medication usage pattern was most strongly associated with enthesitis-related arthritis (OR 3.2 [1.9-5.4] compared to all other categories of JIA). Among 246 children with systemic arthritis, 160 (65%) ever received a biologic DMARD; TNF inhibitor use was associated with polyarthritis (OR 2.5 [3.8-16]) while IL-1 inhibitor use was not.

Conclusions Approximately three-quarters of all children with JIA in the CARRA Registry received non-biologic DMARDs. Nearly one-half received biologic DMARDs, and their use was strongly associated with RF+ polyarthritis, psoriatic arthritis, uveitis, and systemic arthritis.

Disclosure of Interest T. Beukelman Grant/Research support from: Pfizer, Consultant for: Novartis, S. Ringold: None Declared, T. Davis: None Declared, E. Morgan DeWitt: None Declared, C. Pelajo: None Declared, P. Weiss: None Declared, Y. Kimura Consultant for: Novartis, Genentech