IIMS 96 contents
[ IIMS 96 contents ]

Clinical medicine: Can the computer replace the patient?

Peter Devitt and Edward Palmer
University of Adelaide


The foundations of medical education have long been the learning of the art of medicine at the patient's bedside. Over the last decade various changes have threatened to loosen the bedrock on which these foundations have been laid. Changes in styles of learning, with a shift from didactic to problem based education, restricted access to patients and increasing numbers of undergraduates have all placed increasing strain on existing teaching and learning resources.

Various strategies have been adopted to meet these challenges and have included the use of surrogate patients (Stillman, 1992), a shift to community based medicine and teaching and an examination of the role the electronic medium can play (Papa, 1989; Khadra, 1995; Andrews, 1992).

Although computers have been available as an educational tool for over two decades, they have had minimal impact in medical education and particularly little impression on clinical teaching. Several reasons exist. Most teachers in medicine teach with the same style and resources with which they themselves were taught - and these clinicians were brought up in the pre-computer era. The majority of medical practitioners are not familiar with computer aided instruction and do not see a role for it in their teaching. Another reason for the limited use in medical education is that although few resources may be required to run these systems, computer aided instruction does take a considerable amount of time and effort to set up and particularly to write the programs and required content.

Computer material is available for medical education and it has been developed in many formats to fit different styles of learning (Clayden, 1988). However, perhaps the most cogent reason for the failure of computer based instruction to make a serious impact on medical education has been the perceived lack of creditability of the material and the absence of any critical evaluation of computer aided instruction as a valid learning tool.

Computer aided instruction is an integral part of the clinical component of the undergraduate curriculum in the Faculty of Medicine at the University of Adelaide. We report on the use and acceptance of this learning resource by the student body at this medical school.

Materials and methods

A computer based multimedia program was designed and built within the Faculty of Medicine. This program was first made available at the beginning of 1994, when 12 modules had been developed for the program (Devitt, 1993). The case in each module was text based and generously supplemented with clinical photographs, radiographs and video footage to produce realistic clinical scenarios. The aim of the author of each case was to produce material that would allow the users to explore clinical problems in much the same way as they would be expected to do in real life as a medical practitioner. In other words, the style of the work was problem solving, rather than simple question and answer. This style was used to be in keeping with the aims of the learning objectives of the Faculty, one of which was to promote self directed learning in a problem based environment.

The cases were delivered in a program called Medici. Medici was designed using HyperCard for the Macintosh and Toolbook for Windows on PCs. An authoring system has been developed that allows individuals who are not 'computer literate' to produce their own case material and transpose the completed work themselves directly into Medici. The authoring system exists in two formats. The most efficient system is to use a template which has been designed for Filemaker Pro. The material is written (or pasted) directly into this template and exported into Medici for instant use. The second method, designed for those people without Filemaker Pro, is a text template. The author produces the material under a series of key headings and the competed case is then transferred into the appropriate format.

Cases were written by clinicians to provide realistic clinical scenarios for the students. Users were invited to diagnose, investigate and treat 'patients' with a variety of conditions. Most cases were based on real instances and modified as necessary. The data provided in the cases in terms of physical signs, pathological tests, radiological investigations, operative specimens and drug details was genuine and taken from real case material.

A typical case put the users in the role of an intern or general practitioner and invited them to manage the case. The users would be introduced to the 'patient' who would give a story. The users were required to make some decisions about the patient's story and would then be given further information. This information might be a photograph of part of the patient's anatomy or a description of what was found on the physical examination of the patient. Further opinions would be sought from the users, and this would be followed by more information, perhaps some data from clinical investigations to study. In this way, the users would be taken through each case and provided with a score and critique on completion.

The users could select which way they wanted to manage each case, what investigations they wanted to perform, how the case was to be treated, and did not always have to move along a pre-determined course. Each case asked the users to test their skills of history taking, physical examination, diagnosis and management. At the end of the problem solving exercise the users got a score for their performance and were provided with a critique. This critique varied from a tutorial on the topic to an opinion on how the case should have best been managed. The computer recorded the progress of each user through the case.

During the eight months from February to September 1995 an increased number of case studies were made available for all undergraduates in the Faculty of Medicine at the University of Adelaide. Students had access to the material on computers based at two teaching hospitals (Royal Adelaide and Queen Elizabeth Hospitals) and in the medical school. All students were allowed access. The computers were housed in clinical studies suites and were available during office hours.

All students were invited to use the material as well as several other programs, which had been specifically developed for undergraduate use. Some of this material was commercially available, but most was developed on campus. Study of the material was voluntary, except for IVth Year students (those in their second clinical year). These students were required to undertake a minimum of six hours study on the cases during the year and this formed part of the assessment of one of the clinical attachments. Although this (surgical) attachment lasted only six weeks, the students were able to do the computer based material in their own time and at any stage during the year they chose. To gain access to the cases all students had to log on, using their student number as identification.

Cases were rotated for student use. Each case could be accessed in Medici for three months. The scope of material chosen was such that there was always a broad spectrum of clinical material for study.

All the cases were designed for students in their clinical years. Some of the management issues were aimed at the more senior students, but still provided important learning areas for the more junior undergraduates. It was anticipated that only the cases with significant amounts of anatomy or pathology would be of interest to the pre-clinical students.

Students were asked to complete questionnaires. Users of Medici and its cases were asked their opinions on the package; its ease of use and logic, and the content of each case. They were asked about the relevance and the degree of complexity of the material and the use of graphics. In a separate questionnaire students were asked their attitudes to computers, previous experience and perceived importance of computers in their undergraduate education.


An evaluation of the students' computer knowledge was undertaken to provide an accurate gauge of the target audience. It was found that 10% of students had never used a computer before and 50% of the students did not feel comfortable using a computer. Of those students who had used computers, almost all used the machines for word processing and 70% had used them for statistics and games. Less than one quarter of the students had ever used a spreadsheet and only 30% had worked on a database.

Sixty cases were used in Medici between February and October 1995. The content ranged from acute abdominal pain and its management to X-ray interpretation, and from depression to malignant melanoma (Table 1). Each case varied in length and complexity, some of which contained only three components to the large branched problems, which might contain over thirty separate components.

DisciplineNo. of casesExamples of topics
Anatomy2Jaw swelling
Gastroenterology15Peptic ulcer disease
Pancreatic cancer
General medicine2Breathlessness
General practice4Melanoma
Vascular disease
General surgery17Appendicitis
Head and neck5Thyroid
Salivary glands
Pathology4Intestinal malignancy
Radiology3Chest X-rays
Trauma2Ruptured spleen
Urology4Bladder cancer

Table 1: Material used in Medici between February and October 1995 (sixty cases)

The number of cases available for study each month varied between 8 and 31 (Table 2). The use of the computer program remained fairly constant throughout the year, as shown by the number of cases studied each month (Table 3). The dip seen in the middle of the year coincided with mid-year examinations and the mid-winter break. The sudden rise in users in September occurred when the Third Year students were informed that there was an example of an examination question for them on Medici.

MonthNo. cases

Table 2: Numbers of cases available for student use

MonthNo. cases

Table 3: Number of cases completed each month

Not unexpectedly, the greatest numbers of users came from the ranks of the Fourth Year students, where six hours of computer use was mandatory and part of the assessment. Virtually all the students completed the designated time and a substantial number did more. Table 4 shows the time that each student spend on the computer during the year. There were exactly 800 students enrolled in the Faculty of Medicine in 1995 and the distribution between the years II to VI is shown in Table 4.

YearNo. studentsAve timeNo. students using Medici

Table 4: Computer usage from Years Two to Six.
Time spent on computers shown in minutes

A rather more meaningful figure on student use of this component of computer aided learning within the Faculty can obtained by looking at the data for those students who actually used the material. Table 4 shows the heaviest use was from the Fourth Year students. This was hardly surprising, given the compulsory nature of this learning resource, however, this group of students still managed to put in an average of 79 minutes more than was required. The Fourth Year students spent an average of 11 minutes on each case.

YearNo. studentsAverage time

Table 5: Actual student use of Medici.
The time each student spent on Medici over the 9 months is shown in minutes

Less than five percent of students required any extra assistance to enable them to use Medici. The majority of students found the program simple to use and ranked the ease of use as 1.6 on a scale of 1 to 10, where 1 was the easiest (Figure 1). With respect to the complexity of the case material, most students felt the balance was appropriate with most cases not too easy nor too difficult (Figure 2). Most students found that the material within Medici was highly relevant and pertinent to their studies (Figure 3). The students gave a mean score of 8.5 out of 10 on a scale of 1-10.

Figure 1

Figure 1: Degree of ease of use of Medici

Figure 2

Figure 2: Degree of complexity of cases

Figure 3

Figure 3: The relevance of the content of Medici

In the other survey students were asked to comment on the value of computer aided instruction as a learning resource. More than half the students thought it was beneficial, with most of the other students unable to decide. Some of the specific comments showed that

Almost half of the students surveyed felt that computer aided instruction should become a compulsory component of the course, with the bulk of the remainder being undecided. The undergraduates did not want to see the computer based material replace existing course components and many thought that this learning resource might lose its appeal if made compulsory, particularly if it consumed a substantial amount of time.


Medical education is at the crossroads, and it is uncertain as to whether it can - or should - continue to rely on patient based learning. Traditionally, the teaching of medicine has been the domain of the universities, with their affiliated large metropolitan hospitals, and all their resources. These sources are no longer readily available.

Health care is increasingly expensive and many communities are not prepared to spend as much on health as they were a decade ago. The large metropolitan hospitals are considered a financial drain and resources are now being shifted to community based health care. These fiscal restraints and redistribution of the health dollar have had a major impact on medical education.

Strategies for meeting the educational needs of students have included the use of surrogate patients (Stillman, 1990), and the adaptation of video and computer material (Piemme, 1988). A larger proportion of students will spend more time away from the parent institution and will learn their trade in the community. The community will not supply all their needs and the students will still require learning resources. The universities will have to provide these resources and the electronic medium offers an ideal way of providing some of this self paced distance education.

The present study has shown that students now look to computers as a learning resource and expect to find a proportion of the course material delivered in this form. However, the observation that 10% of students had never used a computer came as a surprise, considering the supposed availability of computers and computer aided instruction in secondary schools. More recent questioning of the students suggests that this percentage of computer illiteracy is failing rapidly.

Provided the framework that contains the learning material is not unduly complex or convoluted, most students in tertiary educational establishments will use the electronic medium - even if they have had little previous experience with computers. The degree of usage will of course depend on how useful the students perceive the material to be, how relevant it is to the course and whether the content is likely to be part of previous experience with computers. The degree of usage will of course depend on how useful the students perceive the material to be, how relevant it is to the course and whether the content is likely to be part of any examination. Apart from the answers to the questionnaires giving a positive response to the use of computers and Medici in particular, the continued use of the material throughout the year by students from all years suggests that computer aided learning is seen by students as a useful resource.

This has been shown by the extensive use of Medici through the eight months, by both those students who were required to study the cases as part of their ongoing assessment, and by those students who used it without solicitation. The Fourth Year students spent an average of 79 minutes longer on the Medici cases that they were required to. The use of this learning resource would probably have been much higher if the students had unrestricted access to the machines. Unfortunately, for security reasons, the computers could only be used in working hours and there were only three machines in the two main teaching hospitals to service the needs of over 350 full time clinical students. Many students had to be turned away to allow the Fourth Year students to complete their required studies.

The type and style of material that is delivered in computer format is crucial. Not only must it be relevant, but it must be given in a format that will keep the students' attention. Medical education - as with any form of education - can be provided in an active or a passive medium. The active process will promote understanding and interest, while the passive system tends to be dull, boring, encourages rote learning and is often not retained by students. Unfortunately much of what passes for medical education is given in passive format and is based on the didactic lecture. Many of the early computer programs tended to match this style and presented tutorials on the computer screen which offered little more than electronic page turning. An alternative, but equally passive style, was to use the computer as a question and answer tool, with progress only being made when the user had chosen the correct answer.

However, medicine is not based on multiple choice questions or turning over pages in a book, but on encounters with human beings. Therefore, to get the most of the computer for teaching it is necessary to use its power to recreate the real world. Case based teaching is a powerful learning tool and can be produced well on the computer. This is learning by 'revelation' (Clayden, 1988) and apart from the high degree of realism, can also be an enjoyable experience for the users.

Medici meets many of these needs. The program is simple and intuitive to use, without a frightening arrange of complex screens. The content is case based and the material delivered in a logical sequence, matching the revelation of clinical problem solving. The users are required to think and make decisions, and act the part of a medical practitioner. The problem must not be so long that the users become bored. The Medici cases were designed to be completed in less than twenty minutes - and the students worked through most cases in well under this time.

If computers are to be used in clinical education, there are several requirements that need to be met. First of all, the promoter of this resource should not attempt to replace the clinical educational process, but rather to supplement it. Nowhere has it been shown that any learning resource is as effective as the learning and experience of bedside medicine. However, the field of medicine is broad and no student can hope to see all there is to be seen during undergraduate training. Therefore a key role for this type of technology is to act as an adjuvant to clinical learning and to provide material that may not be otherwise readily available. With the shift away from large metropolitan hospitals to community based teaching, students will have less access to a wide range of disease processes and will rely more heavily on resources such as the computer to broaden their horizons.

There is little doubt that with today's technology and its rapid progression, the clinician as teacher could be replaced by the computer. However, whether that would be to the advantage or detriment of medical education has yet to be determined. It has yet to be shown that computer based education is superior to more conventional learning resources, both in terms of what is learnt and the speed at which the information is learnt. So the question to be posed and answered is, should the computer replace the clinician? That doyen of medicine, Sir William Osler, stated that 'to study the phenomenon of disease without books is to sail an uncharted sea, while to study books without patients is not to go to sea at all.' What Osler implied about learning from books is just as pertinent to any other learning resource, and the best medicine will always be learnt at the bedside under the tutelage of the clinician.


Andrews, P. V., Schwarz, J. and Helme, R. D. (1992). Students can learn medicine with computers. Med J Aust. 157, 693- 95.

Clayden, G. S. and Wilson, B. (1988). Computer assisted learning in medical education. Med Educ. 22, 456-467.

Devitt, P. G. ( 1993). Medici - a Problem Oriented Learning Program for Medical Education. ASCILITE Conference Proceedings, 194-200. Dec.

Khadra, M. H., Guinea, A. L. and Hill, D. A. (1995). The acceptance of computer assisted learning by medical students. Aust NZ J. Surg. 65, 610-612.

Papa, M. and Meyer, S. (1989). A computer assisted learning tool designed to improve clinical problem solving skills. Ann Emerg Med. 18, 269-273.

Piemme, T. E. (1988). Computer assisted learning and evaluation in medicine. JAMA 260, 367-372.

Stillman, P. L., Regan, M. B., Philbin, M. and Haley, H. L. (1990). Results of a survey on the use of standardized patients to teach and evaluate clinical skills. Acad-Med. 65, 288-292.

Stillman P. L., Regan, M. B., Haley, H. L., Norcini, J. J., Friedman, M. and Sutnick, A. L. (1992). The use of a patient note to evaluate clinical skills of first-year residents who are graduates of foreign medical schools. Acad-Med. 67: S57-9.

Contact: Dr Peter Devitt
Department of Surgery
Royal Adelaide Hospital
Adelaide SA 5000
Tel: 08 222 5516 Fax: 08 232 3471
Email: pdevitt@medicine.adelaide.edu.au

Please cite as: Devitt, P. G. and Palmer, E. (1996). Clinical medicine: Can the computer replace the patient? In C. McBeath and R. Atkinson (Eds), Proceedings of the Third International Interactive Multimedia Symposium, 107-112. Perth, Western Australia, 21-25 January. Promaco Conventions. http://www.aset.org.au/confs/iims/1996/ad/devitt.html

[ IIMS 96 contents ] [ IIMS Main ] [ ASET home ]
This URL: http://www.aset.org.au/confs/iims/1996/ad/devitt.html
© 1996 Promaco Conventions. Reproduced by permission. Last revision: 14 Jan 2004. Editor: Roger Atkinson
Previous URL 4 Dec 2000 to 30 Sep 2002: http://cleo.murdoch.edu.au/gen/aset/confs/iims/96/ad/devitt.html