Review by: Reichert, Raimond (2004-07-14)

An important part of medical education is the diagnosis of illness. Ideally, the students practice on real patients. However, that is very expensive, and often difficult to realize. The cases of interest may not always be available just when they would be needed in class. Also, it is often not easy to get the patients to talk about their illness in front of students. For these and other reasons, the project Interactive Simulated Patient was initiated at the Karolinska Institute, Sweden, in 1990. This article presents the pedagogical intentions and the technological achievements.

The project aimed to develop a virtual learning environment for the diagnosis of patients. Some of the pedagocial intentions were: The system should stimulate student-activated learning methods and collaborative learning, and it should support problem-based learning. It should be designed to be as realistic as possible and to incorporate most examination and laboratory techniques clinically. To achieve these goals, the system basically consists of a huge multimedia database of patient data, with hundreds of images and video clips for each patient.

The main task for students is to come to a efficient and of course correct diagnosis, while trying to minimize the cost of the search and the diagnostic tools used for it. They have at their disposal a free-text driven patient-history function in natural language, with all patient responses in the form of video clips. They can also apply standard examination procedures, and conduct any of a huge number of laboratory tests available (more than a thousand chemical lab tests, pathology tests, and others are simulated). The system provides a feedback by showing students the correct diagnosis and how to arrive at it with minimal cost.

The quality of this learning environment is ensured by the effort of an interdisciplinary team of over 20 specialists which collected the necessary high-quality patient data over 13 years. The high degree of interactivity is not due to programming – an authoring tool was used to develop the environment – but rather to the encompassing collection of patient data and the experience of the specialists involved in putting it together. One comment in the discussion regarding the development is particularly interesting: As commercial multimedia and learning-systems developers were not able to fully understand the complex nature of the medical area itself, the project team did the technical development in-house. This demonstrates how difficult the development of highly complex interactive learning environments can be.

The results of field tests are published separately: Bergin et al (2003). Interactive Simulated Patient: Experiences with Collaborative E-Learning in Medicine. Journal of Educational Computing Research, 29(3), pp. 387–400.