To develop objective structured clinical examination (OSCE) stations to assess the ability of physicians to address selected clinical-ethical situations, and to evaluate inter-rater agreement in these stations.Two ten-minute OSCE stations were developed using video-taped encounters between attending physicians and standardized patients. One scenario involved a daughter requesting a do-not-resuscitate (DNR) order for her competent mother without the mother's knowledge; the other involved a competent elderly woman requesting not to be re-intubated if her congestive heart failure worsened. The scenarios were evaluated using foreign medical graduates taking an OSCE. Each candidate was scored on his or her interaction with a standardized patient in the two OSCE stations by two independent observers.Eight attending physicians from the Division of General Internal Medicine at the Toronto Hospital were used to develop the OSCE stations, and 69 foreign medical graduates taking the University of Toronto Pre-Internship Program OSCE were used to evaluate the stations.The inter-rater reliability coefficients for the DNR and intubation scenarios were 0.79 (95% CI 0.69-0.87) and 0.75 (95% CI 0.62-0.84), respectively. For the DNR station, the scores of the two examiners, on a scale of 0 to 10, agreed exactly for 34 candidates (50%), within one mark for 59 candidates (87%), and within two marks for 65 candidates (96%). For the intubation station, the scores of the two examiners agreed exactly for 27 candidates (40%), within one mark for 56 candidates (84%), and within two marks for 63 candidates (94%).The authors produced ethics OSCE stations with face and content validity and satisfactory inter-rater agreement. Ethics OSCE stations may be suitable for evaluating the ability of medical students and residents to address selected clinical-ethical situations.