ORIGINAL ARTICLE Patterns of Infliximab Use Among Crohn’s Disease Patients in a Community Setting Alice R. Pressman, MS,* Susan Hutfless, MPH,* ,† Fernando Velayos, MD, ‡ Bruce Fireman,* James D. Lewis, MD, § James Allison, MD,* ,‡ Oren Abramson, MD,* and Lisa J. Herrinton, PhD* Background: Information on infliximab use in a community set- ting is important to understand patterns of medication use and to anticipate and plan for costs associated with the drug. We sought to understand predictors of initiation and discontinuation of infliximab in the community-based setting of Kaiser Permanente, Northern California, which provides integrated care to its members. Methods: The cohort study was set during 1998 –2006. Predictors of initiation were assessed among 494 Crohn’s disease (CD) patients who initiated infliximab and 2470 CD patients who did not initiate infliximab (controls). Data were obtained through linkage of com- puterized clinical information and were analyzed using logistic regression and Cox survival analysis. Results: Infliximab infusions have increased rapidly since 2001, with no evidence of leveling off. Initiators were appreciably younger than controls (P  0.001), but were similar to controls with respect to sex and race/ethnicity. The presence of at least 1 comorbidity was related to a modest increase in the risk of initiating (compared with none: 1 comorbidity, odds ratio [OR] = 1.52 with 95% confidence interval [CI] 1.16 –2.00; 2 comorbidities, OR = 1.38 with CI 0.89 –2.13). By 3 years after initiating, only 20% of patients remained on infliximab. Conclusions: In a community-based setting infliximab use has steadily increased. Age and comorbidity are associated with initia- tion, but sex and race/ethnicity are not. More information is needed to determine why, in this community-based setting, a large number of patients on infliximab discontinued their treatment. (Inflamm Bowel Dis 2008;14:1265–1272) Key Words: Crohn’s disease, infliximab, health services research, computerized medical information, variation in care A decade has passed since Targan et al 1 reported that a single infusion of infliximab was an effective short-term treatment for patients with moderate-to-severe, treatment- resistant Crohn’s disease (CD). A year later, in August of 1998, infliximab, an antitumor necrosis factor alpha mono- clonal antibody, was approved in the US for the treatment of luminal CD in patients failing conventional therapies or who were steroid-dependent. 2 Since then, the use of infliximab in the setting of CD has evolved. In 2002, results from the ACCENT I trial demonstrated that chronic use of infliximab every 8 weeks was superior to a single infusion, resulting in greater rates of remission and reduced rates of steroid depen- dence at 1 year. 3 In 2004, results from the ACCENT II trial demonstrated that chronic use of infliximab every 8 weeks was effective in patients with fistulizing CD. 4–6 Balanced with these favorable clinical trials were reports of adverse events with the medication, which included infection, con- gestive heart disease, cancers such as lymphoma, as well as problems associated with episodic therapy, such as formation of antibodies against infliximab, infusion reactions, and loss of response. 7–13 While randomized controlled trials are used to show the efficacy of a medication in homogenous groups of patients, based on strict entry criteria, strictly protocolized treatment, and resources to monitor and follow patients, they do not provide insight into how a medication is used in a community setting, where patients are heterogeneous, treatment protocols are variable, and resources to monitor and follow patients are limited. Information on infliximab use in a community setting is important to detect patterns of inappropriate use and op- portunities for quality improvement, as well as to anticipate and plan for costs associated with the drug. Although several studies have reported the utilization and outcomes of inflix- imab therapy at specialty centers, 12–15 it is notable that nearly a decade after infliximab was approved for the treatment of CD, data describing utilization patterns in a community set- ting are sparse. To this end, we sought to understand pre- scribing patterns in the community-based setting of Kaiser Permanente, Northern California, which provides integrated care to its members and records comprehensive clinical and Received for publication March 13, 2008; Accepted March 15, 2008. From the *Division of Research, Kaiser Permanente Northern California, Oakland, California, † Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, ‡ Division of Gastroenterology, Department of Internal Medicine, University of Califor- nia, San Francisco, California, § Department of Medicine, Department of Biostatistics and Epidemiology, and Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, Pennsylvania. Funded by a cooperative agreement from the Crohn’s and Colitis Foun- dation and Centers for Disease Control (U01 DP000340), and by a grant from the Kaiser Foundation Research Institute. Reprints: Lisa J. Herrinton, PhD, Division of Research, Kaiser Permanente Northern California, 2000 Broadway Ave., Oakland, CA 94612 (e-mail: lisa.herrinton@kp.org). Copyright © 2008 Crohn’s & Colitis Foundation of America, Inc. DOI 10.1002/ibd.20483 Published online 1 May 2008 in Wiley InterScience (www.interscience. wiley.com). Inflamm Bowel Dis ● Volume 14, Number 9, September 2008 1265