Problem-based learning
1981; Wiley; Volume: 15; Issue: 1 Linguagem: Inglês
10.1111/j.1365-2923.1981.tb02307.x
ISSN1365-2923
Tópico(s)Health Sciences Research and Education
ResumoMedical EducationVolume 15, Issue 1 p. 1-3 Free Access Problem-based learning First published: January 1981 https://doi.org/10.1111/j.1365-2923.1981.tb02307.xCitations: 3AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat We commented rather recently on the claims being made that doctors should be trained by problembased learning (Editorial, Medical Education, November 1979). This method is held to provide systematic training in solving clinical problems. A book *, by a professor of medicine and a lecturer in nursing, has now appeared which requires that further attention should be paid to this specific approach to medical education. The authors have written the book to answer two questions: Why is problem-based learning necessary? What evidence is there for its effectiveness?. The authors are the directors of the Problem-Based Learning Systems, Faculty of Health Sciences, McMaster University, Hamilton, in Canada. In going to work at McMaster in 1971, Howard Barrows was attracted primarily by the student-centred, small-group learning approach. The empirical basis of the method was derived from a study of the way in which a large number of general physicians dealt with standardized ‘patient encounters’. The aim of the method was to help medical students develop systematic approaches in their thinking about patients' problems, and to acquire both basic science and clinical information in a way ensuring retention and transfer to the actual task of the clinician. Learning from problems is regarded as a condition of human experience. Learning occurs naturally in attempts to face the problems encountered by every person each day. There is thus nothing new in advocating the use of problem-solving as a method of learning in the medical context. Medical teachers have the obligation to be interested in investigating better ways to prepare their students for the tasks they will have to perform later on as doctors. Almost all of the learning a doctor will need to accomplish in his forty or more years of professional work, after his formal medical education, will be his own responsibility. Often the evaluation and management of patients' medical problems are regarded as vocational skills. Greater academic value is assigned to the arguments for acquiring a firm knowledge base in the basic and clinical sciences; the doctor should be a scholar in the medical sciences, this ideology runs, and the mission for a medical school is to produce scholars. This argument overlooks the fact that medical problem-solving, or the clinical reasoning process, is the doctor's ‘science’. This science, previously misperceived as ‘art’, is definable, amenable to evaluation, and can be improved upon by appropriate teaching. It is of the greatest importance that the learning accomplished by medical students shoulu be of the sort that permits application to the problems they will encounter as doctors. In medical education, it can be argued, the emphasis must be on the application of knowledge. Conventional teacher-centred learning has to be contrasted with student-centred learning. In the latter, the student learns to determine what he needs to know. He becomes an active participant in the learning process. His motivation is improved thereby; what he learns is better retained, and he develops the habit of learning for his own personal or professional development, not for rewards such as examination grades which are in the gift of the teacher. Much of the learning in medical school is department-based, organized around the subject area or field in medicine of the basic sciences, such as anatomy, pharmacology, surgery, and so on. The conventional goal is for the student to gain an overall grasp of the subject area involved. A disadvantage of this concentration on specialties is that information is not integrated with relevant information from other disciplines or subject areas. It promotes memorization and understanding of a body of information in one subject area, but it does not ensure that the student will be able to apply the information in practice (which never comes packaged in accordance with a disciplinary boundary); patients do not present themselves as isolated examples of information from one discipline. In problem-based learning the student undertakes to solve a problem a patient presents, a problem in health delivery, or a research problem. His aim now is to acquire an integrated body of knowledge relating to the problem at hand, and to develop or apply problem-solving skills. Problem-based learning is ideally suited for student-centred and individualized learning. This approach allows the student to use any particular problem as a focus for the study of many different subjects, and calls on him actively to integrate this information into a method that can be applied to the problem at hand and to subsequent problems (it is erroneous to suppose that this approach devalues learning in the basic sciences, or that it overemphasizes patient management). A difficulty which faces the student is that problem-based learning is not a good way to prepare for conventional examinations. These place reliance on the recall of isolated facts and concepts. They are usually discipline- or subject-oriented. Properly, different types of examination tools are called for, to evaluate the student's ability specifically derived from problem-based learning, when not only is knowledge acquired but skills in using knowledge are achieved. The clinical reasoning process This is the most important set of abilities the doctor must possess: the cognitive processes necessary to evaluate and manage a patient's medical problems. The authors hold that the clinical reasoning process should be considered the ‘scientific method’ of clinical medicine. This cognitive process, problem-solving, is akin to related concepts of medical enquiry, such as clinical judgment and diagnostic reasoning. It is most unfortunate that the myth, often repeated, has arisen that the ‘art’ which the clinician practises is intuitive, not directly teachable, and results only from experience. Recent empirical studies of the way a doctor approaches a patient (Allal et al, 1978; Campbell, 1976; Wulff, 1980) have demonstrated that doctors are not themselves particularly aware consciously of the technical approach they use in the clinical reasoning process. It has become apparent that the clinician characteristically generates two to five hypotheses very soon after his first encountering the patient. These ideas, hunches, guesses, impressions or preliminary diagnoses are at first vague (for example, ‘something wrong with the heart’), but they serve to focus the clinician's history-taking and examination of the patient. They are based on very little data, deriving from some of the first cues presented by the patient. All the working hypotheses are usually developed within the first quarter of his clinical interview of the patient. The clinician then acquires the further information he needs to refine his early hypotheses, and for ranking them in order of likelihood. He uses ‘search’ questions for the purpose. Finally, as the picture of the patient's illness evolves in the mind of the clinician, his ‘case-building’ leads on to a formulation, from which follows the decision on the part of the clinician how he should act if he has decided on intervention. The clinical reasoning process can thus be divided into the following behaviours which Barrows & Tamblyn consider can be evaluated and learned: (1) information, perception and interpretation; (2) hypothesis generation; (3) enquiry strategy and clinical skills; (4) problem formulation; and (5) diagnostic and therapeutic decisions. The clinical reasoning process is a scientific approach to patient problems. The process has definable segments: perception of initial cues, assembling an initial concept of the problem, generating multiple hypo theses, employing an enquiry strategy, integrating the information elicited into a growing formulation, co-ordination with evaluative and therapeutic decisions. If this is a more or less accurate representation of the way clinicians proceed, how ought this knowledge to influence medical education? Barrows & Tamblyn suggest that the learning of clinical problem-solving calls for new teaching methods. They point to the disadvantages in using real patients in examinations of clinical ability. Simulated problems are to be preferred, or paper patients (the patient management problem approach, computer formats, etc.). Barrows has been a foremost pioneer in establishing ground rules for the educational use of the simulated patient; students must realize that the simulation is in fact based upon a real patient, and is faithful to the real picture that the patient presented. However, since it is a simulated patient that is deployed, there is no need for students to be concerned about making mistakes or about asking questions or performing examinations that might upset the patient. The interaction with the simulated patient can be interrupted for discussion with a group of fellow students. The simulated patient provides an educational advantage in challenging the students' interviewing, examination skills, and interpersonal abilities. ‘Time out’ discussions are possible, when the simulated patient can temporarily be set aside. Contacts with real patients, in the actual clinical setting, can be brought about when the student has already acquired the necessary basic skills in clinical reasoning and examination through his experience of the simulated formats (Ostrow, 1980). Learning resources These, fortunately for the protagonists of problem-based learning, are numerous. They include books, journals, reprints, tape-slide programmes, videotapes, audio-tapes, films, models and specimens, teachers, research workers, and they can occur in numerous places (clinics, wards, laboratories, community agencies, field studies and health projects). Self-evaluation complements self-study. The problem-solving approach calls on teachers to use new methods and responses when they see students making mistakes. Instead of simply pointing out that the student is wrong, they need to ask why he came to the conclusion he did, to help him to consider whether there. might be alternative approaches, and encourage him to draw on fellow students for suggestions; by asking the student for explanations, alternative hypotheses or formulations, the teacher accepts the role of guide. Even if the student considers he is doing well and the teacher is satisfied with his performance, the student needs evaluation tools that will tell him when he is progressing appropriately in his training, and has no undetected weaknesses that need attention. Summative evaluations are thus called for, which provide the student with immediate feedback about his progress and allow him to correct his weaknesses. Test measures must, of course, be appropriately matched with the skill being measured. It is now highly questionable that retention of facts correlates with clinical competence. Evaluation tools for problem-based learning will concentrate on process rather than content. MCQ and short-answer exams have questionable validity for this purpose. Observation of the student carrying out a patient interview and examination is to be preferred; review of the case record is appropriate; patient problem simulations, of the PMP type or the sequential management type, are to be preferred. Simulated patients provide invaluable assessment procedures. Mechanical mannequins are now far developed, and there are a variety of models that can be used. ‘There does not seem to be a part of human anatomy that cannot be authentically simulated. Virtually all aspects of human physiology can be simulated’ (Abrahamson, Wallace & Hansen, 1980). Methods such as these allow better evaluation of clinical reasoning skills; and self-assessment procedures are to be preferred. Barrows & Tamblyn have written a book that is important for the challenge it presents to conventional procedures in medical education. They raise questions which need to be faced. It is appropriate that their book is the first in a promising new series of books on medical education to be published by the Springer Company. Footnotes * Barrows, H.S. & Tamblyn, R. (1980) Problem-Based Learning: An Approach to Medical Education. New York: Springer Publishing Company. Pp. 206. $19.95. References Abrahamson, S., Wallace, P. & Hansen, S. (1980) Simulation in medical education: static and interactive manikins. Programmed Learning and Educational Technology, 17, 148. Allal, J., Gordon, M., Jason, H., Kagan, N., Loupe, M.s. & Jordon, R.D. (1978) Medical Problem Solving. Harvard University Press, Cambridge . Campbell, E.J.M. (1976) Basic science, science, and medical education. Lancet, ii, 134. Wulff, H.R. (1980) Evaluation of competence in clinical decision-making. Paper given at Nijmegen Conference of the Association for Medical Education in Europe, September 1980. In press. Ostrow, D.N. (1980) Surrogate patients in medical education. Programmed Learning and Educational Technology, 17, 82. Citing Literature Volume15, Issue1January 1981Pages 1-3 ReferencesRelatedInformation
Referência(s)