Background: Situations where practical therapeutic decisions differ from guidelines in the management of patients with Crohn's disease (CD) have been described through opinion surveys. The feasibility of actually documenting these situations using an explicit approach has not been examined. Objective: The aim of this study was to evaluate the feasibility of a retrospective application of appropriateness criteria to a population of CD patients. Methods: Medical records of a cohort of patients diagnosed with CD were systematically reviewed. We used appropriateness criteria for treatment of CD that had been developed by the European Panel on the Appropriateness of Crohn's Disease Therapy (EPACT). First we evaluated the level of precision of the elements abstracted from medical records needed in order to be able to apply these criteria. We then assessed the appropriateness of treatment for different CD categories. Only participants with at least one physician encounter during the preceding 6 months were included. Results: 260 patient medical records were reviewed on site at 22 gastroenterologists' offices over a 2-month period in 2005. 116 (44%) patients were excluded because they had not had at least one medical visit at their referred gastroenterologist during the preceding 6 months. Medical records for 8 additional patients (3%) were not accessible. 136 (53%) medical records including 148 encounters were available for analysis. Overall, elements necessary to determine the appropriateness of treatment were available in 94% (139/148) of encounters. These elements were available in more than 90% of cases for all CD categories except for mild-moderate luminal active CD where 66% were available. Among those with all necessary elements available, 18% of treatments were judged as appropriate, 29% inappropriate, 38% uncertain according to the EPACT criteria, and for the other 15%, appropriateness had not been rated by the EPACT panel. Conclusions: The information necessary to assess the appropriateness of treatment of major types of CD was generally both present and precise in medical records. Therefore, in addition to the intended prospective use of these criteria, retrospective evaluation of the appropriateness of CD treatment using medical records is also feasible with the EPACT criteria. [Ed.]