Description of an examination for the objective assessment of history-taking ability.

A novel examination of the ability of final-year medical students to take a targeted history is described. One hundred and nine students were examined in 1 day. Each student interviewed three patients, each with a different problem. One of these persons was a real patient, while the other two were simulated patients. Patients were able to give consistent histories and the use of simulated patients enabled a high degree of standardization to be achieved. The examination provided a useful degree of discrimination among students; the marks scored followed a normal distribution with a mean value of 22.2 and a range from 16 to 29 out of a maximum of 30 marks. We believe that this examination is valid, reliable and practical.