Development and validation of the inflammatory bowel disease objective disability index (IBDODI).

INTRODUCTION: Current measuring tools for disability related to inflammatory bowel disease (IBD) rely mostly on patient-reported disability perception. It may, therefore, largely depend on patients' subjectivity, thus hindering objective evaluation. In the present study, we aimed to identify clinical plausibly objectifiable variables related to disability and to develop and validate a long-term disability index based on them. METHODS: Answering an online survey, IBD patients reported their officially recognised degree of general and work disability plus several long-term disability-related clinical variables. Responses were randomly allocated to a training or a validation dataset. Multiple logistic regression tests were performed in the training set. Variables statistically or clinically related to disability were used to model an Unweighted-IBD Objective Disability Index (U-IBDODI) -where all variables had the same weight- and a Weighted- Index (W-IBDODI), where variables were weighted according to the regression results. Both scores were subsequently validated. RESULTS: The analysis included 930 questionnaires. Patients' mean age was 41±11, 642 (65.4%) were women, and 582 (59.3%) had Crohn's Disease. The training dataset included 665 surveys. In the validation set (n=265), U-IBDODI mean values were 3.7+/-1.3 for patients with work disability and 2.3+/-1.4 for patients without (p<0.001). U-IBDODI AUROCs for predicting work and general disability were 0.839 and 0.675, respectively. The corresponding W-IBDODI values were 10.9+/-3.7 and 6.9+/-3.7 (p<0.001), 0.837 and 0.606. CONCLUSIONS: Predictive values of IBDODI scores were good for work disability and moderate for general disability. These indexes may provide a more objective evaluation of permanent disability, complementing existing indexes.