Search in the Abstract Database

Search Abstracts 2012

* = Presenting author

P286. Assessing preference for deep remission in patients with Crohn's disease


W. Sandborn1, J.‑F. Colombel2, E. Louis3, R. Panaccione4, M. Yang5,6, E. Wu5, P. Mulani6, J. Chao6

1University of California, San Diego, La Jolla, United States; 2Centre Hospitalier Universitaire de Lille, Hôpital Claude Huriez, Lille, France; 3University of Liège and CHU Liège, Department of Gastroenterology, Liège, Belgium; 4University of Calgary, Director, Inflammatory Bowel Disease Clinic, Calgary, Canada; 5Analysis Group, Inc., Boston, United States; 6Abbott Laboratories, Illinois, United States



Background: To assess patient preference over 5 Crohn's disease (CD) health states, including deep remission (i.e., clinical remission [CR] with mucosal healing [MH]), CR without MH, MH without CR, moderate CD, and severe CD.

Methods: An online survey was conducted of patients with a CD diagnosis for at least 1 year and a conjoint analysis was performed to evaluate patient preference over different CD health states. Participants were asked to make preference choices between CD health states and EQ-5D health states (defined by EQ-5D item response combinations). Definitions of the CD health states were provided to help patients familiarise with the concepts. Patients were asked to make the preference choice between the randomly selected CD health state and the comparative EQ-5D health states. The selection of the subsequent EQ-5D comparators depended on the prior choices to ensure efficient estimation. Based on the data collected, logistic regression was run for each CD health state separately to estimate the probability of preferring the CD health state over various utility levels defined by EQ-5D health states. The utility value of a CD health state was defined as the probability of over 70% preference on that CD health state. A sensitivity analysis was conducted where the current survey respondents were standardised based on the age and gender distributions from the patients with CD from the 2008 Medical Expenditure Panel Survey.

Results: A total of 124 patients completed the survey. About half of the participants (before/after weighting) were female (50.8%/62.4%) with a mean (±standard deviation) age of 52.27±11.91/45.5±15.8). The majority of the participants self-reported in remission or mild disease (80.6%/85.1%) and half of them had surgical resection (51.6%/48.6%). Although nearly three quarters (72.6%/74.7%) were on prescription treatment for CD, less than half (43.5%/44.2%) experienced active symptoms in the past 4 weeks. The estimated utility values for the 5 CD health states (ie, deep remission, CR without MH, MH without CR, moderate CD, and severe CD) were 0.949, 0.767, 0.766, 0.370, and 0.019, respectively, demonstrating a higher utility score for a better/preferred health state. Similar estimates were obtained in sensitivity analysis.

Conclusions: Deep remission was clearly preferred over clinical remission only and mucosal healing only. Moderate CD and severe CD were much less preferred, with severe CD having the lowest estimated utility score.