Full Paper

Frank Billson, John Grigg

Frankb@eye.usyd.edu , Johng@eye.usyd.edu

Department of Clinical Ophthalmology

Tim Pye, Marcel Chaloupka, Tony Koppi

Timpye@nettl.usyd.edu.au , Mchaloup@nettl.usyd.edu.au, tony.koppi@nettle.usyd.edu.au

NeTTL, Centre for Teaching and Learning,

University of Sydney.

The University of Sydney's Faculty of Medicine has implemented a new four year Graduate Medical Program in 1997. The pedagogy underpinning this new program is that of problem based learning in small group settings. Particular emphasis in the Sydney program is given to equipping students from the beginning with skills of appraising population-based studies critically which will allow students to combine the techniques of evidenced based medicine with the acknowledged strengths of Sydney's basic and clinical sciences. (Sefton, 1994) Teaching of communication skills for a multicultural society will be given emphasis via Patient-Doctor theme sessions. Other theme sessions include the Community-Doctor theme which aims to increase the student's awareness of community issues, including promotion and prevention, and to assist them understand the different perspectives of health delivery.

The change to this new curriculum has been one of the imperatives for developing our Virtual Ophthalmology teaching program. One issue confronting all those involved in medical education is the ever increasing amount of information available to both the teacher and student and the limited time to study a particular discipline. The change to a problem based learning approach aims to empower the student with skills in problem solving and techniques to manage the avalanche of information.

Student feedback shows a desire for exposure to a broad range of clinical ophthalmic conditions in the setting of small group teaching. The difficulty in teaching a subject that has short clinical exposure is that students will have different experiences, exposure to important clinical conditions will be varied and not all students will have the opportunity to assess important ophthalmic conditions.

The virtual ophthalmology program will provide a bank of simulated cases, which cover the common cause of blindness in Australia, the important causes of acute visual loss and the impact of blindness on the community.

The program will provide the student with a forum to see a wide range of conditions. Assisting the student through practice in this virtual environment, will mean that more of the actual Clinic time can then be directed at establishing student confidence in their inter-personal relationships and all that is inherent in an encounter with people rather than treating the patient as an elusive condition. Providing students with more opportunities as the medical professional.

The program has been developed to recreate an ophthalmic patient encounter, which allows for the student to direct the inquiry process and then to formulate a plan of management for the clinical problem presented. The program focuses on three main skill areas, which are core medical practice activities. These are medical history taking, observation skills and integration of this information to formulate case management plans.

The virtual ophthalmology clinic incorporates the principles of problem based learning and is a model for computer assisted learning for all medical disciplines. The program also sits within a wider framework designated as a distributed learning environment to facilitate the learning experience for the student. A distributed learning environment is a paradigm, which is both student centred and teacher centred. The environment facilitates learning by providing flexibility in the delivery of pedagogically sound experiences. Cognitive tools are placed in the correct domain for example reference and content material reside in a library (paper or electronic), activities which may be experiments and so reside in laboratories. The classification of material allows programs to be designed which can offer course coordinators flexibility in designing courses with their own particular emphasis. The program also provides a means to update the course dependent on changing goals. Object-orientated models, which may be interdisciplinary, form part of the resources that can be accessed by academics to construct a customised course. (Tait, 1997) Technology has reached a stage where such flexibility for course design can be contemplated.

Distributed learning addresses the issues of the teachers being advisers, managers and facilitators and thus being a means by which an educational or training facility can implement an educational experience in a divergent market where the location and distribution of the teaching resources and participants (students, teachers etc.) is unimportant.

Effective teaching and learning empowers a student to take a deep approach to learning which results in the desirable outcomes of: holistic appreciation; knowledge that is active; communication and teamwork skills; problem-solving and critical thinking abilities; and lifelong learning attitudes. Whatever the subject matter, an important outcome is that the learner has an appreciation of where the subject fits into a global context. Disciplines should never be so remote and isolated that practitioners cannot see their effects from a wider perspective. (Candy, Crebert & O'Leary, 1994)

Technology can be incorporated into pedagogical strategies, and to a certain extent can substitute for tried and tested conventional practices, and can apparently achieve similar outcomes (Albanese & Mitchell, 1993), (Chambers, Mullins, Boccard & Burrows, 1992), (Piemme, 1988). The two main areas where technology offers enhanced learning capabilities are simulated or mimetic representations of the real world where experiential learning is possible, and in collaborative learning strategies which use web-based communication facilities. Technology is lacking in the area of artificial intelligence, particularly where the learner requires feedback on matters of opinion and interpretation. Technology can offer the means for learner control where students have greater flexibility and self-determination to construct their own meaning and acquire active knowledge.

We have chosen to simulate the clinical encounter because, simulation software places students in the professional role (principle of authenticity) and gives them opportunities to make decisions about applying and integrating information previously learned (Junkala, 1991). Exposure to group discussions will be needed to emphasis the value judgements inherent in clinical decision making. Group settings will allow students to reflectively analyse their clinical reasoning processes. The realisation that clinical decisions are modified by the circumstances in which they occur will help to develop cognitive flexibility and adaptability. (Principle of Adaptation and Articulation)

When students use the multimedia case simulations they will have to identify and research their own learning issues and solutions to the case. The inherent ambiguity in clinical cases will become apparent when the case is examined from different perspectives and in different contexts. This experience will lead the learner to understand the complexity of clinical cases and the development of sceptical certainty and the encouragement for life long learning (Koschmann, Meyers, Feltovich & Barrows, 1994) (Principle of timelessness). The challenge in developing this type of learning environment is to facilitate imaginative immersion that enables experiences, learner control and reflective reasoning (Leyland, 1996).

Woolliscroft (Woolliscroft, Calhoun, Beauchamp, Wolf & Maxim, 1984) concluded that student performance is most reliably measured not in an observed and therefore inherently stressful situation but from an analysis of the students written review of the patient interview. The virtual ophthalmology clinic provides an environment for practice, reflection, generation and presentation of diagnostic and management data providing examiners with the opportunity to design and monitor the curriculum as well as corral and assess reviews and seminar presentations.

Mimetic worlds provide a realistic context for exploration and heuristic activities, which enable the learner to interact and model an environment in the first person. The interactivity provided in these environments can enable learners to construct schemas and test hypothesis against the environment and see the results of their actions, thus gaining experience, in relatively short time periods. The challenge in developing this type of learning environment is to facilitate imaginative immersion that enables experiences, learner control and reflective reasoning (Leyland, 1996)

The virtual ophthalmology program has been developed to complement the clinical teaching sessions and clinical exposure that the student experiences during the short time studying ophthalmology. Scientific advances in medicine have not changed the fact that doctors' core clinical skills are interpersonal (Novack, Volk, Drossman & Lipkin, 1993) (Engel, 1973). This program aims to reinforce the communication skills developed in the Patient-Doctor theme sessions and place these skills in an ophthalmic setting. This has formed a major component of the program and to achieve this we have developed a novel communication navigator (see below).

Medical students also desire integration of ophthalmology with other medical subjects, as future management of the whole patient requires broad based clinical skills. Students in the past have often struggled with obtaining the technical skills required to elicit clinical signs with the ophthalmic examination instruments. This has resulted in a reduction in the time available to interpret the findings and integrate them into management plans for the patient.

The virtual ophthalmology program provides an opportunity for students to practice interpretation of clinical information and the formulation of management plans for patients in a non-threatening environment. That doesn't have the same time constraints that are placed on the student in a clinic setting when real patients are attending for medical care so encouraging students to explore patient treatment and management options thus reinforcing general generic skills required as a medical practitioner.

The virtual ophthalmology clinic introduces the student to a simulated consulting room into which they select patients from a waiting room. The initial version of the program has recorded 13 problems. The clinical problems were derived from real patients who participated in a videoed interview. The interview was transcribed and the responses edited to match the conversation navigator icons. Four actors then played the parts of the 13 patients from the original interviews. This process provided the greatest realism while preserving patient confidentiality.

Once a patient has been selected then a problem is randomly derived from the actors' problem subgroup. These details are logged with the student's profile. The clinical encounter is divided into four areas, which are performed sequentially with the ability to go back but not forward until a section has been completed. The four areas are medical history taking, clinical examination, laboratory investigations and patient management formulation. An electronic form will be constructed in the form of a medical record. The student will have to identify what they find relevant in the history and interpret this by way of developing a list of differential diagnoses. When this has been completed the form is lodged and the student can then proceed forward to the next stage being examination. Again the students observations will be logged with an interpretation of the data. Once this has been done then an opportunity to order any investigation will be provided with the students having to justify their selections. Finally a clinical management plan is formulated. The program collates the student's record and sends a copy to the tutor.

The medical interview is the cornerstone of clinical medicine. Information gathering during the medical interview forms the basis for hypothesis generation and initial clinical problem-solving by the clinician. (Woolliscroft et al., 1984). Medical education has formalised the teaching of medical history taking since the mid 1970's (Novack et al., 1993). The new Graduate Medical Program places significant emphasis on communication skills with weekly Doctor-Patient theme sessions scheduled throughout the course. The first 14 week block covers a generic interview process which is then built on and applied to medical history taking and then to each of the medical disciplines. This program builds on the skills acquired in the first two years of the course and then applies them to the discipline of ophthalmology. To do this we have designed a novel conversation navigator.

The simulation of a medical interview process is difficult. We have developed a unique conversation navigator which will students to obtain medical histories in a nonlinear process as close as possible to a real patient encounter. To do this we have deconstructed the patient history-taking process to find suitable icons, which replicate the normal spoken exchange between doctor and patient. While there should be little intended difference in meaning between questions displayed textually and an iconic representation of the same question, there will be some ambiguity caused by individual student's personal associative interpretations of those icons. As well as ambiguities caused by contextual changes as the questions apply to different symptoms and scenarios. However a moments reflection will show that succinctness of question and response are variables between all individuals and to some extent the conversation navigator replicates this reality.

How then can we justify their use? Iconic representations cannot be imbued with the specificity of its 'languaged' counterpart. However the iconic device is free from some of the disadvantages of textual representation that justify icon use as an alternative. Questions presented in text would have to be accessible to the students as an exhaustive list covering every conceivable situation within the simulation. They would have to be specific to each scenario, ie patient and disease.

The student may see a list of questions as a prescriptive solution to the diagnostic process giving little credence to individual learning styles and cognitive processes. A hierarchal list suggests that the author is expert and the student may feel less autonomy which is desirable for developing higher order thinking skills and be encouraged to memorise (usually for the sake of passing exams) rather than theorise. If a student can be encouraged to exert a cognitive effort in an attempt at 'understanding' it is more likely that some learning might occur. No two people model (internalise) or manifest (externalise) 'knowledge' identically. Likewise, no two medical conditions or patients are identical in terms of their treatment.

The language in which textual questions are expressed is not the students (it is the authors), and therefore liable in some way to diminish the sense of ownership and immersion. The student in choosing to select an icon has to employ more cognitive effort in regards to the contextual or temporal appropriateness of the question. If the student choses to select questions from a list randomly there would be a debilitating cognitive effort in remembering which questions had already been asked and the equally debilitating cognitive drain of switching from the 'drama' to an inconsistent navigational operation.

Inconsistent navigational aids mean that the face to face conversation simulation is constantly disrupted. The complexities of the actions such as moving the mouse to a menu bar, clicking a pull down menu, scrolling down the list, deciding and choosing a question interrupts the flow of interaction between "doctor and patient". Taking one from the state of immersion out to an extrinsic navigational operation. Because the iconic navigator exists within the domain of the virtual patient, ie it will not be a feature of a control bar, and the icon palette will float proximate to the patient and be movable by the student. The promise for the is a Natural Language Interface (NLI), that would allow spoken interrogation of the virtual patient, however research into Artificial Intelligence (AI) has not yet achieved a reliable engine for this purpose.

Icons are 'generic' and non- hierarchical, so that the student may continue on the basis of prior information, including patient responses, to pursue a line of questioning further (eg. tell-me-more button). Having exhausted that avenue, the student may then create another path in the conversation. In this way the student progressively builds an internal model of the patient, contrasting and comparing their prior medical knowledge against the information presented.

With this model the student will have increased opportunities to approach the clinical examination in the context of problem. So often because of time constraints the examination is taught in isolation (eg feel that lump rather than this patient complains of feeling a lump) and as spot diagnoses rather than in the context of the patient's concerns and perceptions. Having already elicited a history a context for the examination will be identified and alert the student to the possibility of certain findings, which then need to be confirmed or refuted.

The clinical examination module requires students to observe ocular structures and compare them to normal or abnormal processes, which they have, access to; to better understand the interface between variants of normal and disease.

Implication of a particular disease manifesting at different stages of a persons life and in the context of other illness will require different strategies for management. This can only occur if doctors have a holistic view of the patient. Disease management requires that patients', age, gender, work or lack of it, cultural background and previous illness and medications form the setting in which individual treatments are formulated. They signify a doctor's internal processes, ie. how they synthesis prior knowledge with observed symptoms in a unique context. Internal or cognitive process are multi-layered, interrelated and can occur both systematically and spontaneously in response to each unique situation.

The virtual ophthalmology clinic addresses a number of issues in medical education including the limited time constraints, the variability in clinical experience, integration of skills of history taking with medical science using ophthalmology as the model. These features form the basis for integration of clinical information into a relevant management plan for the patient.

The conversation navigator provides a more immersive and realistic experience in the role of doctor as interviewer. By providing a more realistic experience and minimising interface obstructions allows the student to concentrate on problem solving with minimal distraction.

The learning environment also provides opportunities for assessment by allowing collation of student reviews and also the ability to observe the questioning of peers. Students will also be able to provide support to each other and teachers via electronic discussion groups.

The program by providing the student with the opportunity to make decisions and be assisted in self-learning has another consequence. It allows the teacher to be more available for the important task of sharing in the group discussions and assisting in the decisions that are qualitative; particularly concerning ethics or assumptions about a patients needs.

(c) Frank Billson, John Grigg, Tim Pye, Marcel Chaloupka, Tony Koppi

The author(s) assign to ASCILITE and educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The author(s) also grant a non-exclusive licence to ASCILITE to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the ASCILITE 97 conference papers, and for the documents to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the authors.