DIFFERENCES IN THE USE OF 2ND-LINE AGENTS AND PREDNISONE FOR TREATMENT OF RHEUMATOID-ARTHRITIS BY RHEUMATOLOGISTS AND NON-RHEUMATOLOGISTS

Objective. To compare the use of methotrexate (MTX), intramuscular (im ) gold, hydroxychloroquine, and prednisone for rheumatoid arthritis (R A) treatment among patients managed by rheumatologists and nonrheumato logists. Methods. Multiple regression analysis to estimate the likelih ood of starting treatment and response to treatment for patients manag ed by rheumatologists and nonrheumatologists. All regression analyses were adjusted for patient demographic and clinical characteristics. Re sults. Therapy with all agents studied was initiated more frequently f or patients with RA with at least some contact with rheumatologists du ring the year than for those managed strictly by nonrheumatologists. T he adjusted odds ratios for starts on these medications ranged from 1. 14 for im gold to 15.11 for MTX for patients managed by rheumatologist s compared to those managed by nonrheumatologists. However, due to the low frequency of initiation of treatment with most of these drugs for patients managed strictly by nonrheumatologists, only the odds ratio for prednisone reached statistical significance (OR = 2.94, p = 0.0082 ). In the year after initiation of therapy with these agents, patients managed by rheumatologists experienced better response to treatment t han those managed by nonrheumatologists. These differences were statis tically significant for MTX (p = 0.0447) and nearly significant for im gold (p = 0.0597). Conclusion. These results provide evidence of syst ematic differences in the propensity of rheumatologists and nonrheumat ologists to initiate therapy with these antirheumatic drugs. If the ob served differences in initial response to treatment translate into sub stantial differences in longterm outcomes, then these results suggest that the welfare of patients with RA may be jeopardized by the current trend toward primary care and restricted access to rheumatologists.