Background:Patients with RA are at increased risk of infection compared with the general population, but it is unclear whether this is due to the underlying disease or to immunosuppressive medications used to manage the disease.1Some biologic DMARDs (bDMARDs) have been associated with an increased risk of serious infection.2A large cohort study found no increased risk of serious infection in patients initiating abatacept compared with patients initiating other bDMARDs.3It is clinically important to identify which patients are at a higher risk of infections at the time of initiating treatment with a bDMARD. However, studies that assess risk factors for infection and derive corresponding risk scores of infection, especially at the time of bDMARD treatment initiation, are lacking or based on too few patients.4Objectives:To identify the risk factors for serious infections among patients with RA initiating treatment with a bDMARD in a real-world observational setting.Methods:The Truven MarketScan®Commercial and Supplemental Medicare databases were used to identify patients diagnosed with RA who initiated treatment with a bDMARD between January 2007 and December 2015. Patients were followed from treatment initiation until the occurrence of a serious infection requiring hospitalisation, the end of enrolment or 31 December 2015, whichever came first. The Cox proportional hazards model was used to estimate the hazard ratios (HRs) of serious infection associated with baseline risk factors including demographics, the presence of co-morbidities, prior hospitalised infections and medications. An infection risk score was developed using the independent risk factors found to be significant in the model.Results:The study cohort included 84,308 patients initiating treatment with a bDMARD, mainly etanercept (36.7%), adalimumab (29.3%), infliximab (12.4%), rituximab (7.3%) and abatacept (6.8%). During a mean follow-up of 6.6 months, 1724 patients were hospitalised for a serious infection (incidence rate 3.7/100 persons per year). The baseline risk factors significantly and independently associated with serious infections were age, prior hospitalisation for infection, hypertension, diabetes, lymphoma, asthma, chronic obstructive pulmonary disease, cardiovascular disease, other autoimmune disease, corticosteroid use and antibiotic use. The infection risk score, with a possible range of 0 to 15, had a mean (SD) value of 2.6 (1.9) with range 0–12.5. The HR (95% CI) of serious infection was 1.43 (1.40–1.45) for every unit increase in the risk score. Relative to patients with a score of 0, the HR (95% CI) of serious infection for a risk score of 5 was 5.9 (5.3–6.5), and for a risk score of 10 was 34.5 (28.5–41.6).Conclusion:In this large, real-world cohort of patients with RA who were initiating treatment with a bDMARD, several patient characteristics were found to independently predict the subsequent risk of serious infection. The risk score, based on easily available patient characteristics, can be a simple and useful tool for the clinician to identify patients at higher risk of infection at the time of bDMARD initiation for the treatment of RA.References:[1]Doran MF, et al.Arthritis Rheum2002;46:2287–93.[2]Singh JA.Curr Rheumatol Rep2016;18:61.[3]Montastruc F, et al.Semin Arthritis Rheum2019;48:1053–8.[4]Jani M, et al.Curr Opin Rheumatol2019;31:285–92.Acknowledgments:Joanna Wright (editorial assistance, Caudex; funding: Bristol-Myers Squibb).Disclosure of Interests:Samy Suissa Grant/research support from: AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Novartis, Consultant of: AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Novartis (advisory board meetings), Speakers bureau: AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Novartis, Paul Brassard: None declared, Alyssa L. Dominique Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, Teresa Simon Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, Marie Hudson: None declared