real-world database study showed that within two years over 75% of CD patients had atleast one indicator of inadequate response during maintenance therapy with biologics. Doseescalation, flares, CD-related urgent care, and augmentation with non-biologic treatmentwere the major indicators of inadequate response.Table 1. Indicators of Inadequate Response to Biologics During the Maintenance PhaseAmong Patients With CD

Note: Natalizumab was not included in the analyses due to low sample size. *Comparisonsbetween adalimumab, certolizumab and infliximab treated groups with chi-square test.

Sa1120

Indicators of Inadequate Response to Anti-Tumor Necrosis Factor Therapies inPatients With Ulcerative Colitis From Real-World Practice SettingsHind T. Hatoum, Haridarshan Patel, Swu-Jane Lin, Reema Mody

Background: The study objective was to determine the incidence of inadequate response toinduction (I-phase) and maintenance (M-phase) treatment with anti-tumor necrosis factortherapies (anti-TNFs) in ulcerative colitis (UC) patients in real-world practice settings.Methods: A retrospective cohort study was performed using the US Clinformatics administra-tive health claims database (2005-2012). Patients with UC were identified as having thefollowing: ≥2 claims for UC (ICD-9-556.xx); ≥1 claim for an anti-TNF [adalimumab (ADA)or infliximab (INF)]; ≥6 months of plan enrollment prior to the first diagnosis of UC (indexdate); and ≥1 year follow-up from the first anti-TNF (index drug). Non anti-TNF UCtreatments included 5-ASA, oral or injectable corticosteroids, enteral budesonide, or immuno-modulators. The indicators used as a proxy for assessing inadequate response during the I-phase were switching to another anti-TNF or augmentation with any new non anti-TNFtreatment. Additional indicators of dose escalation, flares, UC-related surgery, and UC-related urgent care (hospitalization or emergency room visit) were included in the M-phase.Patients who received labeled doses of anti-TNFs without any encounters of UC-relatedurgent care (during the I-phase) and with ≥2 claims for anti-TNF were included in the M-phase. Bivariate statistical analyses were performed using chi-square test. Median time toand rates of inadequate response were estimated using the Kaplan-Meier Method. Results:A total of 587 UC patients met the study inclusion criteria, 51.8% of whom were male; themean (SD) age of patients was 42.8± 13.4 years and follow-up was 2.75±1.45 years. A totalof 22.3% (n=131) and 74.5% (n=437) of patients had received ADA and INF, respectively,as index drug. Mean (SD) duration from the initial UC diagnosis to first anti-TNF was471±476 days. After diagnosis, 89.8% (n=527) of patients received ≥1 new non anti-TNFtreatment within 6 months prior to index drug. During the I-phase, 5.3% (n=31) of patientshad at least one indicator of inadequate response. Of 338 patients included in the M-phase,77.8% (n=263) had at least one indicator of inadequate response (Table 1). The overallinadequate response rates did not differ significantly between ADA and INF. Estimatedcumulative rates of inadequate response in the M-phase were 66.2% and 77.7% in first andsecond year, respectively, with a median time to inadequate response of 197 days and nosignificant difference between the two anti-TNFs. Conclusion: This large-scale real-worldassessment of index anti-TNFs for the treatment of UC demonstrated that over three-quartersof patients on maintenance therapy had at least one indicator of inadequate response withintwo years, with dose escalation being the most commonly observed indicator.Table 1. Indicators of Inadequate Response to Anti-TNFs during the Maintenance PhaseAmong Patients with Ulcerative Colitis

*Comparisons between adalimumab and infliximab using Chi-square test.

S-205 AGA Abstracts

Sa1121

Predictive Factors of the Course of Crohn's Disease - Can We Treat inAnticipation?Francisca Dias de Castro, Joana Magalhães, Pedro Carvalho, Maria J. Moreira, José Cotter

Background and Aim: the course of Crohn's disease (CD) is highly variable and difficult topredict on the basis of information at the time of diagnosis. However it would be useful tocategorize patients at the onset of disease in low and high risk on the basis that treatingsevere CD with "top-down" strategy might change the natural history of CD. The aim ofthis study was to evaluate factors at first presentation that might predict the severity ofCD. Methods: retrospective, single-center study including 146 patients with CD diagnosedbetween June 1983 and December 2011. We defined aggressive CD as need for more than2 steroids courses required after diagnosis, need for surgery after diagnosis (except surgeryfor perianal disease or ileocecal resection as first choice of treatment) and need of admissionfor flare after diagnosis. The variables studied at diagnosis were smoking status, familyhistory of inflammatory bowel disease, extraintestinal manifestations, Montreal Classification(age, location of CD, involvement of the upper gastrointestinal tract, phenotype, perianaldisease), need for corticosteroids, admission and surgery on the first flare. Statistical analysiswas performed with SPSS vs 18.0 and a p value of less than 0,05 was considered statisticallysignificant. Results: 146 patients with CD were included, 55,5% female, with mean age37,6±11,9 years and mean follow-up of 86±60 months. 80 patients presented at follow-upwith non aggressive CD (55%) and 66 (45%) with aggressive CD. Independent factorspresent at diagnosis and significantly associated with aggressive CD were smoking status(p=0,022), stenotic and penetrating phenotype (p<0,001) and corticosteroids and admissionon the first flare (p<0,001). Given the results of the univariate analysis, in our cohort, thepresence at diagnosis in an individual patient of 2 or more out of the 4 factors was associatedwith a high risk of aggressive disease with a accuracy of 0,73 (sensitivity of 88%, a specificityof 61%, a positive predictive value of 65% and a negative predictive value of 86%). Conclu-sions: at diagnosis of Crohn's disease predictors of subsequent aggressive course are theinitial requirement for corticosteroids and admission, smoking status and penetrating andstenotic disease. The combination of 2 or more variables is highly associated with aggressivedisease and may be a useful way to make therapeutic decisions at diagnosis.

Sa1122

Patients With Ulcerative Colitis Are More Concerned About Complications ofTheir Disease Than Side Effects of MedicationsCorey A. Siegel, Kimberly D. Thompson, Danielle Walls, Susan J. Connor, Sabrina K.Stewart, Meenakshi Bewtra, David T. Rubin, Geri L. Baumblatt, Stefan D. Holubar, Astrid-Jane Greenup, Alexandra Sechi, Afaf Girgis, Jan Gollins

Background and aim: Patients with ulcerative colitis (UC) are often fearful and uncertainabout how their disease will affect their future. Better understanding these fears can helpproviders communicate more effectively with patients and facilitate more informed sharedmedical decisions. The aim of our study was to learn which aspects of UC care are mostconcerning to patients. Methods: Two patient focus groups were conducted at Dartmouth-Hitchcock to obtain qualitative data about patients' fears regarding UC treatment options.These focus groups (1) informed the development of a patient survey and (2) providedaudio to be used with affect trace technology. Next, a web-based survey was sent to UCpatients throughout the United States (US) and Australia (AUS). Within the survey, audioclips were played and respondents showed their strength of agreement or disagreement withrecorded statements in real-time using their computer mouse (affect trace). Qualitativemethods were used to analyze the focus group transcripts, standard quantitative analysisfor the survey results, and cluster analysis was performed on the affect trace response tothe audio clips. Results: 460 patients with ulcerative colitis (370 US, 90 AUS) respondedto the online survey. 53% of the respondents were women, mean age 49 (range 19 to 81).Patients represented 45 US states, and a single AUS state. US and AUS results were similar,and therefore presented together. The majority of patients (95%) rated their current healthas moderate or good. Most (92%) had UC for more than 2 years. Overall, the most concerningaspects of UC were the risk of colon cancer (CRC) (37%) and possible need for an ostomy(29%). When stratified by exposure to immunomodulators or biologics, those with prioror current exposure were most worried about needing an ostomy. In contrast, those whohad not received these therapies were most worried about CRC. Only 14% of patientsindicated that possible side effects from medications was their biggest concern. When askedwhat treatments they were very concerned about, 84% included surgery, followed by biologics(65%), steroids (63%), methotrexate (58%) and immunomodulators (45%). On clusteranalysis of the moment-to-moment affect trace, the most divergence in opinion centered onthe appropriate timing for colectomy and fear and uncertainty about their future with UC.Conclusion: Patients with UC are more fearful of developing CRC or having an ostomy thanthey are of side effects from medications. Of the medications, they are equally fearful ofsteroids and biologics, and less so of immunomodulators. To facilitate informed treatmentdecisions for patients with UC, in addition to discussing the benefits and risks of medications,it is also important to discuss the best strategies for decreasing the risk of colectomy and CRC.

Sa1123

What Patients Want to Know When Making Decisions About Their Treatmentfor Ulcerative Colitis: A Mixed Methods Approach Using Affect TraceTechnologyKimberly D. Thompson, Susan J. Connor, Danielle Walls, Jan Gollins, Sabrina K. Stewart,Meenakshi Bewtra, Geri L. Baumblatt, Stefan D. Holubar, Astrid-Jane Greenup, AlexandraSechi, Afaf Girgis, David T. Rubin, Corey A. Siegel

Background and Aim: Decision making for patients with ulcerative colitis (UC) involvesweighing tradeoffs of benefits and risks of medical and surgical interventions. To improveprovider communication, we aimed to understand what aspects of UC and its treatment aremost important to patients and how they would like to be informed. Methods: Two patientfocus groups were conducted to learn about patients' information needs regarding UC

AG

AA

bst

ract

s