Introduction: Treatment approaches within health care are highly variable. This is an unavoidable characteristic of clinical practice and could be characterised either by uncertainty of information or outcome or by concern for a person’s preferences or both. Accordingly, several studies have reported large variations in general dental practitioners’ (GDPs’) administration strategies of antibiotic prophylaxis in patients with specific medical conditions. Objectives: To study GDPs’ certainty in administration of antibiotic prophylaxis in patients with specific medical conditions, and secondly, to evaluate GDPs’ assessment of risk for complications if antibiotics were not administered. Methods: Postal questionnaires in combination with telephone interviews were used. A total of 101 GDPs participated. The GDPs’ were presented with three simulated cases of patients with specific medical conditions (moderate hypertension, not well-controlled type 1 diabetes, and heart valve prosthesis) for which antibiotic prophylaxis might be considered necessary when performing dental procedures (scaling, tooth removal, and root canal treatment). The GDPs were asked to mark their assessments of certainty and risk on a 100 mm VAS. Differences were analyzed with t-test. The study was supported by the Swedish Research Council (grant 521-2001-6341). Results: Generally, the GDPs presented high certainty in all their decisions independent on whether they administered antibiotic prophylaxis or not (P>0.05). However, when they assessed the risk for complication if not administering antibiotic prophylaxis, the values were much higher for GDPs that would administer antibiotics compared to those that would not (P<0.05). Conclusions: The GDPs presented high certainty in their decisions, independent on whether they administered antibiotic prophylaxis or not. This could implicate that GDPs are less susceptible to modify their behaviour and that implementation of evidence into practice would be difficult.