Collaborative environment for clinical reasoningand distance learning sessions

Elizabeth Medelez Ortega*, Anita Burgun, Franck Le Duff,Pierre Le Beux

Medical Informatics Laboratory, Medical School of Rennes, Rennes I, France

Received 15 December 2002; accepted 5 March 2003

KEYWORDS

Medical education;

Distance learning;

Educational models;

Patient simulation;

Computer-assisted in-

struction;

Medical informatics ap-

plications

Summary Background: The medical curriculum has changed with the adoption of thestudent-centered learning paradigm. Clinical reasoning learning (CRL) is used in orderto develop and improve students’ clinical reasoning and problem-solving skills.Purpose: We have observed that, in complement to traditional CRL sessions, studentscommonly consult resources available on the internet. Based on this observation, ourobjective is to create computer tools to coordinate CRL sessions at distance,integrating these electronic resources at every step of the reasoning process.Material and methods: In order to create the system, we elaborated an object-oriented model of a computer-supported collaborative learning environment. Theproposed system includes a local web-server to store electronic resources and arelational database to store their electronic addresses (urls ). JAVA

TM was used as theprogramming language. Results: We developed a set of cooperative platform-independent tools. This environment includes a communication tool. Multimediadata exchange is possible. Information is shared thanks to an electronic notepad andwhiteboard tools. Perspectives: This learning environment will be integrated in theFrench Virtual Medical University project, and is intended to be used for under-graduate, internships, residency or continuing medical education.– 2003 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

University education is changing on two differentaxes, the pedagogy and the technology areas.Educational methods use the student-centered

paradigm and collaborative learning; that is, stu-dents participate actively in small learning groups.The teacher is not the source of knowledge; but,his/her function is to facilitate the learning pro-cess.

In addition, information and communicationstechnologies can be used to complement student-centered approaches. Many pedagogical resourcescan be found on the internet, such as e-books, webcourses, on-line training, digital libraries, createdto support distance learning; where distance doesnot refer only to a physical separation between

*Corresponding author. Address: Laboratoire d’InformatiqueMedicale �/ Faculte de Medecine, 2, Avenue du Professeur LeonBernard, 35043 Rennes Cedex, France. Tel.: �/33-299-28-4215;fax: �/33-299-28-4160.

E-mail address: elizabeth.medelez@univ-rennes1.fr (E.Medelez Ortega).

International Journal of Medical Informatics (2003) 70, 345�/351

www.elsevier.com/locate/ijmedinf

1386-5056/03/$ - see front matter – 2003 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/S1386-5056(03)00042-X

teacher and learner but also to a technologicalseparation, e.g. computer-mediated communica-tion.

In France, medical schools have created theFrench Virtual Medical University1 project [1].The general purpose of this project is to supportinitial medical training and continuing medicaleducation and to allow the general public to haveaccess to medical information. This virtual univer-sity has foreseen the use of new educationaltechnologies, including problem-based learning(PBL) and clinical reasoning learning (CRL).

Currently, we can find research work aboutcomputer support for PBL in medicine [2�/4],collaborative reasoning [5] and teleconsultation[6]. However, specifically for the CRL method,the computer supported system has been lessstudied and developed. This paper presents acomputer supported environment, created to sup-port the CRL session at distance.

2. Clinical case based learning paradigm

The PBL curriculum is appropriate for under-graduate students. PBL [7] allows students toacquire knowledge in fundamental sciences anddevelop diagnostic reasoning skills. Students workin small groups, they analyze clinical cases, givenin paper format.

For the interns or residents, the CRL seems to bemore adequate. The role-play format, used in CRLsessions, allows students to improve their clinicalreasoning skills and practice their medical knowl-edge much like in the real situation but, withoutrisk for the patient. We describe the principalfeatures of this method.

2.1. Clinical reasoning learning

CRL [8,9] is a pedagogical method, based onclinical cases, first implemented at SherbrookeUniversity, in Canada. This method has beencreated for clerkship students, and the goal ofthis activity is to develop the problem solving skills.The clinical cases used in this kind of learning mustbe relatively complex in order to generate multipleand concurrent hypotheses, but simple enough tohave time to discuss it. The duration of every CLRsession is approximately 1 h.

For this type of learning, the most importantstep in the reasoning strategy is the earliergeneration of hypotheses. In general, the consecu-

tive and repetitive steps followed during the CRLsessions are:

. Information acquisition

. Hypotheses generation

. Data collection

. Hypotheses assessment

. Laboratory test selection

. Diagnosis selection and confirmation

. Treatment choice

. Summary

During a typical CRL session, a group of six toeight students, coordinated by a tutor, simulates aconsultation. This tutor is a medical doctor and, atthe same time, a faculty member. A student playsthe role of the patient ; he/she has previouslyreceived the clinical case script. It contains thepatient’s history, the list of his/her symptoms, theresults of physical examination and the results ofthe laboratory tests.

Another student plays the role of secretary, he/she writes on the blackboard the important ele-ments during the session. Every student writes his/her own notes. The other students without aprevious preparation play the role of physicians.The teacher or tutor has the responsibility forstimulating, helping, correcting and coordinatingstudents. He/she selects one student for simula-tion of patient questioning. The rest of studentsobserve them, and follow the clinical reasoning.The tutor can change the interviewer during thesession.

Some universities use a variant version of CRLsessions. The secretary’s role does not exist and/orall students can ask questions to the patient, in thiscase the session becomes a collaborative diagnosisexercise. The disadvantage of this variant, that ismore difficult to follow a logic reasoning process,due to multiple and concurrent interviewers.

At the beginning, the simulated patient ex-presses the reason for the consultation, using anon-expert language. The initial interviewer beginsthe examination, he/she has to think aloud, thatmeans, he/she expresses his/her reasoning. Assoon as possible, he/she formulates the earlyhypotheses. Every question or complementaryexam helps to reinforce or discard a given hypoth-esis. It can be rejected by the tutor if thejustification is not valid. When the diagnosis hasbeen confirmed, the students can prescribe atreatment and/or schedule another consultation.Then a summary of the simulated consultation canbe written-up. Finally, students and tutor discussthe case, the problems found and missing informa-tion.1http://www.umvf.prd.fr/.

346 E. Medelez Ortega et al.

2.2. Integration of electronic resources

In order to help the students during the CRLsessions; we propose using some electronic re-sources at every step of the reasoning process, asshown in Fig. 1, [10].

During the first step, information acquisition, wecan use multimedia data to enrich the physicalexam; e.g. some real breathing or cardiac soundsor pictures of the skin can be used. These types ofphysical signs cannot be reproduced by the simu-lated patient.

Based on his/her hypotheses, the studentsearches for new symptoms, in order to confirmor reject a hypothesis. Some studies have detectederrors during the diagnostic reasoning due to lackof information or non-updated information [11].For this reason, we encourage the students to usediagnosis-support or decision-making systems forchecking against a list of possible symptoms orhaving a list of illnesses including specific symp-toms; especially, if an unusual sign is detected and/or a rare disease is evoked.

Complementary tests help to complete a non-explicit symptom or to confirm a hypothesis. Inorder to choose the appropriate exam or learn thenormal value of the test, the students can useguidelines. Multimedia data enrich the comple-mentary test, allowing the student to analyze andinterpret the laboratory results.

Medical digital libraries include useful informa-tion about drugs, diseases, research and patient

information. Students can consult them in order toupdate their knowledge, to choose the best treat-ment or to find consumer-oriented information inorder to inform patient, in the best way, about theillness, the possible treatment and other recom-mendations. Other pedagogical resources like lec-ture notes or e-books can be consulted during thewhole session.

3. Design

Based on this new form of CRL sessions, wecreated an object model as shown in Fig. 2.

Four types, or roles, of users have been identi-fied: tutor, patient, secretary and physician. Duringthe sessions, the users can communicate amongthem, in real time. The patient, who feels one ormore symptoms, has a medical record. The physi-cian interrogates the patient. He/she can ask for acomplementary exam or laboratory test, whichshould be authorized by the tutor. Exams arecomposed of multimedia data: video, sound, imageand/or text. Physicians could mark the images toindicate the problem area. These annotations canbe observed by other users. Every student canwrite his/her own notes; they are composed of thesymptoms, the hypotheses, the diagnosis and thepossible treatment, if it is necessary.

The secretary writes a summary of importantelements observed during the session; in general,we can find the same elements in the doctors’

Fig. 1 Resources integration.

Collaborative environment for clinical reasoning and distance learning sessions 347

notes. The notes and the summary can be con-sulted by all users. The tutor can evaluate themand send immediate feedback. The tutor canchange the physician interviewer or ask for parti-cipation at any moment. The physicians can consultexternal resources, like pedagogical documents,decision making systems, medical digital librariesor medical guidelines. These resources can helpstudents to enrich their notes.

To create a computer system to support CLRsessions at distance, it is necessary to provide theusers with:

1) synchronous communication2) multimedia data exchange3) electronic whiteboard and notebook4) What you see is what i see (WYSIWIS) interface5) internet and electronic resources access

In addition, the system should be executed fromany platform and it should not require sophisti-cated hardware.

4. Implementation

From these assumptions and design require-ments, we have developed a computer system,based in the computer-supported cooperative worktechnology. This system has been programmed in

JAVATM language, for its main characteristics: plat-

form independence, portability, multithread, se-curity, object-oriented, and distributed. Wedescribe briefly the main functionalities of thesystem.

4.1. Synchronous communication

We have created a textual chat line system thatallows the users to communicate and simulate aconsultation at distance, see Fig. 3. We decided touse a textual communication instead of audio orvideo communication, because it did not requirespecial hardware and it is relatively fast even witha modem connection. This application has a simpleinterface and it uses representative icons for eachrole. If a member of the group arrives later, he/shecan read the previous messages, using the logbutton. The tutor can send personal messagesduring the consultation.

4.2. Multimedia data support

In order to enrich laboratory tests or physicalexaminations, we use multimedia documents. Theyare physically stored on a web-server and theiraddresses (urls) in a relational database. Thepatient, who must send the laboratory test results,can search for the appropriate document using a

Fig. 2 Object model.

348 E. Medelez Ortega et al.

simple interface, see Fig. 4. He/she can select theclinical case; the list of documents associated tothis case can be updated. All the documents storedin the database can be consulted. For each docu-ment, we have a document description and we cansee images or listen to sounds. Currently, thesystem cannot use video data.

These multimedia documents are sent to all theusers. Some modifications can be made over

images, e.g. users can zoom in/out of them, writeor draw annotations over the images.

4.3. Electronic whiteboard and notebook

The physicians’ notes and the summary typed bythe secretary are accessible by other users. Theinformation written with these electronic tools areupdated regularly. The tutor can evaluate and send

Fig. 3 Teleconsultation tool.

Fig. 4 Access to multimedia database.

Collaborative environment for clinical reasoning and distance learning sessions 349

a message for correcting the diagnosis process ofevery student.

Each student can compare his/her notes with thenotes written by other physicians or with thesecretary’s summary. This comparison could allowthe students to make a self-evaluation and learnfrom their mistakes.

The annotations made on the complementarytest images can be consulted by users. The tutorcan send feedback to the student, if it is necessary.

4.4. WYSIWIS interface

Our system uses a simple type of WYSIWIS inter-face. The communication tool, electronic white-board and notebook use this kind of interface. Thatis, all users can see the same information in almostreal-time. We do not use multi-cursors or colla-borative text tools, because these applications candistract users.

However some users’ actions must be keptconfidential. For example, all personal commentsof the tutor or access to the complementary testresults database are not available for all users.

4.5. Electronic resources access

External resources can be used by studentsduring the CRL sessions. The University of Rennes1 has developed a pedagogical network [12]. Itincludes some electronic books, commented clin-ical cases, teaching files, medical images, multi-ple-choice questionnaires, laboratory testguidelines and a computer-assisted diagnostic sys-tem. Other sites like MedLine2 and MedLinePlus3

are recommended by the teachers, because theycontain useful information for health professionalsand consumers.

In order to simplify the search of Frenchresources on the internet, students can useDoc’CisMef [13], a catalogue and index ofFrench-speaking medical sites. Other resourcesfor medical training [14] will be available at theserver of the French Virtual Medical University.

All these resources can be consulted using a webbrowser.

5. Discussion

The clinical cases are frequently used in medicaltraining by many medical schools. The number

of clinical cases available on the internet isincreasing. Unfortunately, for the most part ofthem, clinical signs and laboratory test results arecompletely revealed at the beginning of thesession. Furthermore, students cannot interactwith the patient or apply their clinical reasoningskills. In the CRL method, the case informationis gradually unfolded. In this case, students mustask questions to the simulated patient in orderto acquire information, like in a classical consulta-tion and they can improve their questioningstrategies.

In general, students do their training courses inhospitals, sometimes in remote areas. They musttravel to attend CRL sessions. For solving thedistance constraint, we have designed a computersystem for CLR sessions, not only for internship orresidency students, but also for continuing medicaleducation. Even if our system was created fordistance learning, it can be used in face-to-faceCRL sessions. In this case, the communication toolis not necessary, but the learning group can use theother applications, especially, the multimedia dataexchange.

There are some variants of CRL sessions, forexample sometimes the secretary’s role is notrequired, or there are several students who askthe patient questions at the same time. Our systemis flexible and allows simulation of the differentvariants of CRL session, and the tutor must becapable of letting students communicate or inter-rupting them, if it is necessary.

Another variant is the utilization of real com-plementary test results. For example, a real X-raypicture can be used during the simulation, and canbe compared with a normal one. An advantage ofthe simulated patient is the representation of thesymptoms of the illness. For this reason, we useimages and sounds during the CRL sessions. Stu-dents can analyze images or sounds for determiningif they are normal or not. We intend to integrate, inthe near future, video data to represent othertypes of exams, such as echography, angiography orendoscopy.

In our system, although images and sounds canbe used to simulate inspection and auscultation,the students cannot practice a whole medicalexamination, including palpation. Perhaps in thenear future, virtual reality can be integratedinto the system and it will allow students tomake the simulations more realistic. We alsointend to add medical semantic web [15] toolsfor reaching resources in the French Virtual MedicalUniversity.

2http://www.ncbi.nlm.nih.gov/PubMed/.3http://www.medlineplus.gov/.

350 E. Medelez Ortega et al.

6. Conclusion

In this paper, we described a learning environ-ment, based on computer-supported collaborativework technology. It was designed to simulate a CRLsession at distance and contains a synchronouscommunication tool, it allows multimedia dataexchange; electronic whiteboard and notes allowstudents to share information, and electronicresources on the internet can be consulted.

The utilization of pedagogical resources providesrelevant information to medical students. It isimportant for medical student to learn how tolook for medical information in order to improve orupdate their medical knowledge. Also, the utiliza-tion of multimedia data allows students to repre-sent a more realistic consultation. Otherfunctionalities, such as video or virtual realitysupport could be integrated into this environmentin the future.

References

[1] P. Le Beux, F. Le Duff, A. Fresnel, Y. Berland, R. Beuscart,A. Burgun, J.M. Brunetaud, G. Chatellier, S. Darmoni, R.Duvauferrier, M. Fieschi, P. Gillois, F. Guille, F. Kohler, D.Pagonis, B. Pouliquen, G. Soula, J. Weber, The FrenchVirtual Medical University, Stud. Health Technol. Inform. 77(2000) 554�/562.

[2] A. Rendas, P.R. Pinto, T. Gamboa, A computer simulationdesigned for problem-based learning, Med. Educ. 33 (1)(1999) 47�/54.

[3] D.M. Kennedy, N. Eizenberg, G.E. Kennedy, An evaluation ofinteractive multimedia designed to support problem-basedlearning in medicine, Proceedings of the EdMedia (WorldConference on Educational Multimedia, Hypermedia andTelecommunications), (2000) 508�/513.

[4] J. Conway, R. Sharkey, Integrating on campus problembased learning and practice based learning: issues and

challenges in using computer mediated communication,Nurse Educ. Today 22 (7) (2002) 552�/562.

[5] M. Quintero, A. Aguilera, M. Abraham, H. Villegas, G.Montilla, B. Solaiman, Medical decision-making and colla-borative reasoning, Proceedings of the Second IEEE Inter-national Symposium on Bioinformatics and Bioengineering(BIBE’01), 161�/165.

[6] E.V. Kopsacheilis, I. Kamilatos, M.G. Strintzis, L. Makris,Design of CSCW applications for medical teleconsultationand remote diagnosis support, Med. Inform. (Lond.) 22 (2)(1997) 121�/132.

[7] A. Hasman, H.P.A. Boshuizen, Medical informatics andproblem-based learning, Methods Inform. Med. 40 (2)(2001) 78�/82.

[8] M. Chamberland, Les Seances d’Apprentissage du Raisonne-ment Clinique (ARC) un exemple d’activite pedagogiquecontextualisee adapte aux stages cliniques en Medecine,Ann. Med. Interne. (Paris) 149 (8) (1998) 479�/484.

[9] J. Corcos, Clinical reasoning learning sessions, Ann. Acad.Med. Singapore 30 (4) (2001) 353�/355.

[10] E. Medelez, A. Burgun, F. Le Duff, P. Le Beux, Integration ofelectronic resources and communication technologies dur-ing clinical reasoning learning sessions, Stud. HealthTechnol. Inform. 90 (2002) 107�/111.

[11] J.B. Paolaggi, J. Coste, Le raisonnement medical de lascience a la pratique clinique, Estem, Paris, 2001.

[12] A. Fresnel, P. Jarno, A. Burgun, D. Delamarre, P. Denier, M.Cleret, C. Courtin, L.P. Seka, B. Pouliquen, L. Cleran, C.Riou, F. Le Duff, H. Lesaux, R. Duvauferrier, P. Le Beux, Afirst evaluation of a pedagogical network for medicalstudents at the University Hospital of Rennes, Med. Inform.(Lond.) 23 (3) (1998) 253�/264.

[13] S.J. Darmoni, B. Thirion, J.P. Leroy, M. Douyere, B.Lacoste, C. Godard, I. Rigolle, M. Brisou, S. Videau, E.Goupy, J. Piot, M. Quere, S. Ouazir, H. Abdulrab, Doc’CIS-MEF: a search tool based on ‘encapsulated’ MeSH the-saurus, Med. Inform. 10 (Pt 1) (2001) 314�/318.

[14] J.M. Brunetaud, S. Darmoni, N. Souf, E. Dufresne, R.Beuscart, A resource server for medical training, MethodsInform. Med. 41 (2) (2002) 177�/182.

[15] M. Cuggia, F. Mougin, P. Le Beux, Indexing method of digitalaudiovisual medical resources with semantic web integra-tion, Stud. Health Technol. Inform. 95 (2003) 715�/720.

Collaborative environment for clinical reasoning and distance learning sessions 351