Background and study aim: Little is known about how gastroenterologists perceive the appropriateness of colonoscopies they perform. The objective of this study was to compare the appropriateness and necessity of colonoscopies as assessed by an expert panel and by the gastroenterologists performing the colonoscopies. Methods: This observational study included 21 centers in 11 countries. Patients referred for colonoscopy were consecutively included. Appropriateness and necessity of colonoscopies were independently rated on a 9-point scale by the gastroenterologists performing them and by an expert panel using a validated method (RAND). The differences between the ratings from the two groups were examined. Results: 6004 patients were included in the study. Comparisons of ratings were possible for 5381 (89.6%) patients. The gastroenterologists' mean appropriateness rating was 7.2 ± 1.7, and the panel's mean appropriateness rating was 5.4±2.3 (P<0.001). The percentages of indications rated inappropriate, uncertain, appropriate, and necessary were 4.1 23.8%, 14.2%, and 58.0% for the gastroenterologists and 27.2%, 26.7%, 25.0%, and 21.1% for the panel, respectively. Agreement between the two groups' ratings was poor (28.8%, kappa=0.11). Differences between the two groups' ratings decreased with increasing patient age, decreasing health status, and decreasing expertise level of the referring physician. However, the gastroenterologists produced consistently higher ratings. Conclusions: Compared with an expert panel, gastroenterologists tend to overestimate the appropriateness of colonoscopies they perform. Except for well-delineated reasons, participating gastroenterologists weighed patient characteristics differently from the panel when judging appropriateness. Ways to increase the prospective use of appropriateness criteria in order to improve appropriateness and reduce overuse of colonoscopies should be examined further.