OBJECTIVES: Relative preference weights for outcomes of anticoagulant therapy can be used to inform quantitative benefit-risk analyses. Whether patients with cardiovascular diseases (CVD) and the general population have different preferences for benefits and risks of anticoagulant therapy is unknown. Using a Discrete Choice Experiment (DCE), we elicited and compared anticoagulant treatment outcomes preferences between patients and the general population. METHODS: A sample of patients with CVD and a general US population sample were selected from online panels. A DCE questionnaire was designed and administered to elicit preferences for benefits and risks. Seven attributes described hypothetical treatments randomly labeled “new drug,” “old drug,” or “no drug”: non-fatal stroke, non-fatal myocardial infarction (MI), cardiovascular death, minor bleeding, major bleeding, bleeding death, and need for therapeutic monitoring.  We estimated preference weights and maximum acceptable risks.   RESULTS: A total of 341 patients and 352 individuals from the general population completed the questionnaire. On average, patients perceived a 1% increased risk of a fatal bleeding equivalent to a 2% increase in non-fatal MI, a 3% increase in non-fatal stroke, a 3% increase in cardiovascular death, a 6% increase in major bleeding, or a 16% increase in minor bleeding. As compared to the patients, the general population had similar preferences except that they perceived a 3% increase in non-fatal MI or a 13% increase in minor bleeding equivalent to a 1% increase in risk of bleeding death.  Patients were less likely to choose “no drug” (odds ratio, 0.72; 95% confidence interval, 0.61- 0.84) or “old drug” (odds ratio, 0.86; 95% confidence interval, 0.81-0.93) than “new drug.”  The general population sample was indifferent to the drug labels.   CONCLUSIONS: Patients and the general population had similar relative preferences for anticoagulant treatment outcomes but were more likely to choose “new drug,” irrespective of its relative benefits and risks.