Objectives Efficacy of TNFi agents in the management of rheumatoid arthritis (RA) has been demonstrated in clinical trials. Variations with respect to patient (pt) profile, extent of physician familiarization with TNFi agents, and pt management may affect real-world outcomes. The aim of this analysis is to compare the pt profile and outcomes of RA pts treated with infliximab under routine clinical practice in biologic treatment registry sites of different sizes. Methods BioTRAC is an ongoing, prospective registry of pts initiating treatment for RA with infliximab as first biologics or treated with a biologic for <6mons. Pts with RA treated with infliximab who were enrolled between 2002-2012 were included (N=832). The number of pts enrolled in BioTRAC at each site was used as the measure for size of the registry site, resulting in the classification of 3 sizes of sites-Group A=sites enrolling 1-15pts; Group B=sites enrolling 16-35pts; Group C=sites enrolling>35pts. Total pts enrolled in each type of practice site was; Group A:n=324, Group B:n=239, Group C:n=269. Results Mean age was 55.8 yrs with 76% of pts being female and 74.2% RF+. No significant differences in demographic characteristics and RF status were observed between the subgroups. Pts seen at larger sites had significantly shorter disease duration (Group A: 11.9 years, Group B: 10.9 years, Group C: 7.5 years; P<0.001) and had been treated with a smaller number of DMARDs (2.5, 2.5, 1.6, respectively; P<0.001). Furthermore, a trend towards lower disease activity at infliximab initiation was observed in larger sites as indicated by the decreased physician global assessment (7.0 vs 6.4 vs 6.2;P<0.001), DAS28-CRP (5.6 vs 5.3 vs 5.3; P=0.023), CDAI (38.1 vs 34.5 vs 35.0; P=0.019), and SDAI (40.9 vs. 36.3 vs. 36.9;P=0.009). Significant differences were also observed with respect to pt management with a significantly greater proportion of pts in the larger sites being treated with concomitant DMARD (87.3% vs 89.5% vs 95.2%; P=0.004) and lower proportion being treated with corticosteroid (23.5% vs 22.6% vs 15.6%;P=0.044). Upon adjusting for baseline disease activity, DAS28-CRP remission (P=0.013) and clinically meaningful improvement in HAQ-DI (Δ≥0.25; P=0.025) over 24mons of treatment was significantly greater among pts seen in larger sites. Conclusions Consistent with findings from a Canadian early RA registry1, results of this real-world observational study demonstrate that significant variation in disease characteristics, pt management and outcomes exist within the BioTRAC registry based on the size of the site. A trend towards earlier infliximab initiation and improved outcomes was observed with larger enrolment sites. References 1. Harris J, et al. Improving outcomes in early RA by determining best practices: Does site size matter or is best treatment early? An analysis of the Canadian ERA cohort. CRA 2013 Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.3740