BACKGROUND AND STUDY AIMS: Guidelines for antibiotic prophylaxis of infective endocarditis associated with gastrointestinal endoscopy are not uniform. Though based on weak scientific evidence, they do represent an attempt to unify the management of antibiotic prophylaxis. We investigated whether physicians who profess to adhere to these guidelines actually do apply them correctly. MATERIALS AND METHODS: A questionnaire was sent to all 195 Swiss gastroenterologists asking whether they adhered to antibiotic prophylaxis guidelines and under what conditions did they apply antibiotic prophylaxis in gastrointestinal endoscopy. We analysed whether the Swiss gastroenterologists who claim to adhere to these guidelines actually do so in practice. RESULTS: The response rate to our questionnaire was 95%. Of the Swiss gastroenterologists, 60% correctly employ antibiotic prophylaxis guidelines in gastroscopy, as do 47% in colonoscopy. In therapeutic endoscopy, the percentage of correct antibiotic prophylaxis use depends upon the type of intervention and varies between 21 and 58%. Overuse of antibiotic prophylaxis is at least twice as frequent in colonoscopy and therapeutic endoscopy than in gastroscopy. CONCLUSIONS: Guidelines for antibiotic prophylaxis are not well applied and do not lead to uniform management despite a high degree of professed adherence to them. This phenomenon can be explained by "cognitive dissonance" and "reactance." We conclude that guidelines based on unconvincing data may even prove harmful.