Among other significant findings, Elstein, Shulman, & Sprafka (1978) demonstrated that diagnostic expertise is content-specific, tending to be limited to clinical problems with which a physician has had substantial experience. They recommended use of a highly structured, problem-oriented curriculum in which students systematically learn about common problems, so that medical schools could guarantee students’ competence to diagnose and treat those problems. Clinical training at most medical schools is far less organized than this: available clinical facilities dictate the patient problems with which students becomefamiliar. In order to increase control over the training environment and the learning outcomes that result, a growing number of schools now use standardized patients (SPs) to teach and test key clinical skills. SPs are non-physicians trained to accurately and consistently portray patients in simulated clinical situations. Students interact with SPs as though they are interviewing, examining, or counseling real patients. This paper discusses the variety of methods in which schools supplement traditional instruction with SP-basedprograms and reviews recent studies of the psychometric characteristics of SP-based tests. © 1990, SAGE PUBLICATIONS, INC. All rights reserved.
