Published version: Bland JM. (2004) Teaching statistics to medical students using problem-based learning: the Australian experience. BMC Medical Education 4, 31.
Problem-based learning (PBL) is gaining popularity as a teaching method in UK medical schools, but statistics and research methods are not being included in this teaching. I discuss the advantages of PBL and the disadvantages of omitting statistics and research methods from the main teaching.
In 2003, I spent four months visiting Australian medical schools to learn about the teaching of statistics and research methods to medical students in problem-based learning curriculums. I did not find any schools where these subjects were taught in a totally integrated way.
In some schools, statistical material was integrated but taught separately, using different tutors. In one school, PBL was used only for "public health" related subjects. In some, a parallel course using more traditional techniques was given alongside the PBL teaching of other material. This model was less successful than the others.
I discuss the difficulties in implementing an integrated approach, and suggest some possible ways forward.
This project was conceived at a meeting of statisticians from medical schools, which is held every year at the Burwalls conference centre, University of Bristol. In 2002, one of the topics for discussion was problem-based learning (PBL). Several things became apparent during this discussion:
It seemed likely to me that PBL would rapidly become more widely adopted in UK medical schools, and I guessed that within ten years all the medical schools outside Oxford and Cambridge would have adopted this approach. I thought that, if this were to happen, there was real danger of statistics and research methods being marginalised in the medical curriculum. This would be bad news for myself and my colleagues, as it is difficult enough to persuade medical students of the importance of these topics and it would be even more so if it were taught outside the mainstream of the course. More importantly, in the era of evidence-based medicine these topics should be central to the medical curriculum, not on the sidelines.
I decided that this would make a very good project for my forthcoming sabbatical. I wanted to visit Australia, and PBL is widely used there. I would visit the medical schools of Australia and find out whether and how they taught statistics and research methods to their medical students in their PBL curriculums. I might then be able to bring back some ideas for my colleagues in the UK to present to their schools as PBL was implemented. I thought that I was well-suited to this role, being a professor who might carry more weight than more junior colleagues, and also being, I suspect, the only medical statistician in the UK who is a trained and experienced PBL tutor.
On a purely personal note, this project would enable my wife and me to see more of Australia than most Australians manage, and we would never outstay our welcome. By the time people were fed up with us, we would be gone. I would also get to meet a great many colleagues. To facilitate meeting them, I decided to offer presentations to anybody who wanted to hear me and sent a list of talks to the ANZSTAT email list of statisticians. I managed to give 23 presentations of various types during my four months in Australia.
I visited the following universities with medical schools: the University of Western Australia, Perth, Flinders University, Adelaide, Monash University, Melbourne, the University of Melbourne, the Australian National University, Canberra (medical school about to commence), the University of Sydney, the University of Newcastle, and the University of Queensland, Brisbane.
I omitted the University of Adelaide, the University of New South Wales, and James Cook University, Townsville. I will try to get information from them at long range, to complete the report.
Like most university teachers, I was not trained to teach and was not employed because of my prowess as a teacher. I was employed because of my knowledge of and experience in my subject, statistics, in the hope that I would use these to carry out collaborative epidemiological research. Teaching was a secondary activity. The only training I received was having been taught myself. I don't claim to any originality for what follows, but include these thoughts because they form the background to my views on PBL.
There are two sides to education: teaching and learning. Some of the traditional methods in higher education are focussed on one aspect, other on the other. Lectures, for example, are teaching, the imparting of knowledge. It has been said that the lecture is a method of transferring information from the notes of the lecturer to the notes of the student without passing through the heads of either. Certainly my own experience of attending lectures was that the material did not pass through my head in getting it down onto the page. Understanding came only on working through these notes later. Seminars, tutorials and practical classes are more focussed on learning. It is more difficult to participate in a discussion or answer questions without thinking about the material. However, I have organised practical classes where students went through the calculations as quickly as possible with as little attempt to understand the material as they could manage. Project work is almost entirely learning, as the student has to tackle an individual question without teacher intervention. We learn most from what we do, next from what we see, least from what we hear. Also, we are more likely to recall what we have learned in the context in which we learned it.
In traditional university education, it is the process of studying which is important, not the subject studied. For example, in the UK a degree in classics was considered a good grounding from which to run the country as a senior civil servant. This was not because a deep knowledge of Cicero was of great value. It was because the habits of mind and skills encouraged by studying, the ability to understand complex texts and precis them, to analyse texts, and to express oneself clearly, were valued. In medical education, we expect our students to learn a large amount of factual information too, but I think that the ability to organise and draw on this information in making diagnoses is more important. In recent years we have been much more keen to instil in students a critical attitude, in keeping with the principles of evidence based medicine (EBM).
There are two fallacies which I frequently come across in discussions of teaching. The first is that students learn everything we try to teach them. Thus, if we include something in the syllabus the students will leave university knowing it. Unfortunately for the teacher, most students will retain only a small fraction of what they are taught. I know that I did. I have an examination paper from my finals as a maths student. I know that I sat this paper, and it is covered with my writing as I worked out the answers, but I now have no idea what either the questions or my own answers mean. The second fallacy is that if we don't teach students something, they will never learn it. This leads to the desire to pack the syllabus with everything the students could conceivably need to know. But some of the statistical methods which I now use regularly I did not learn about as a student; they had not been invented. The computing environment in which I use those I did learn as a student is entirely different. We continue to learn what is relevant to our activities and do not retain that which is not. It is the habit of learning which is important. For example, during a discussion of the undergraduate medical syllabus, I sat between the dean of medicine, a GP, and an eminent professor of epidemiology. The question of the G-protein arose, where were students taught about the G-protein? As others at the meeting were leafing through documents trying to answer this, I asked each of my colleagues "What's a G-protein?" Neither could tell me! I found it hard to believe that this was something which had to be included in the first year of the medical degree. However, to the relief of others, the G-protein was there in the programme. But if it were not, I am confident that those students who needed this knowledge in their subsequent careers would no doubt have acquired it somehow. As it is, I am equally confident that most will soon forget it.
In the UK, the General Medical Council has stated that it is more important for future medical practitioners to have instilled into them the ability and desire for lifelong learning than to acquire a large volume of factual knowledge. "We should seek to light fires rather than to fill vessels," is their slogan. The amount of knowledge required to practice all branches of medicine is too great, and the pace of change is too fast, for students to learn all they might need to know for a career in medicine. All specialties, including general practice, now take postgraduate training, but even this will be inadequate. Doctors must practice life-long learning.
Problem-based learning in medicine began, as far as I can tell, with the work of David Sackett and his colleagues at McMaster University, Hamilton, Ontario. It began with the observation that the most respect clinicians in Hamilton were those who were able to solve problems in patient care by accessing the medical literature. Their skills enabled them to add to their knowledge as and when they needed to. To encourage the acquisition of these skills by students, the method of education should present them with problems to which they would then have to find the solutions. It is the process of finding the solution which is more important that the solution itself. This led to a method of education where students, working in small groups, are presented with a problem, usually a patient vignette, a short description of a patient presenting to a doctor.
For those unfamiliar with PBL, I have included an example drawn from the teaching at St. George's.
The students would then consider the information and decide whether there is anything about this vignette which they need to investigate. THe problem acts as a trigger to raise questions in the students' minds. Of course, there should be many things which the students have not come across before. The students list the questions which the vignette raises, apportion them between themselves, and then seek out this information before the next tutorial. When this problem is exhausted, we move on to the next.
The emphasis is on self-reliance, and the students are expected to provide a chair and a reporter for the session from among themselves. To ensure that the group functions, there is an academic tutor present. The tutor's role is as a facilitator rather than an imparter of information, so the tutor does not need to have specialist knowledge of the topics being discussed. Being by background a mathematician, without even "O" level GCE biology, I have had very little formal training in anatomy, biochemistry and physiology, and know little of these subjects, but I have not found this a handicap as a PBL tutor. Ignorance is strength, and I can honestly say to the students that this is a learning experience for me, too. I may be less at risk than some other tutors of turning the tutorial into a didactic teaching session.
Most institutions practising PBL do not leave the students entirely unassisted in their search for answers to the questions they generate. Traditional lectures, practicals, and tutorials may still be included, as "fixed resource provision". The lecture would then address material to which the students have been alerted by the PBL tutorial.
There are several advantages to the PBL approach. Students acquire very good skills in information retrieval. They become very independent in thought, and confident in questioning received wisdom. They learn information in a context similar to that in which they will need to recall it, the patient consultation.
I was exposed to the ideas of PBL in the early 1970s, when David Sackett came to spend a sabbatical at St. Thomas's Hospital Medical School, where I was then working. I can remember his talk on PBL very clearly, even to the case vignette being on heart disease. I remember this because I offered as my question "What is digoxin?", which I pronounced with a hard "g", as in "goat", to much hilarity. I did not know it was derived from digitalis, and in English "g" followed by "o" is usually pronounced hard.
I did not think it was feasible to implement PBL in one subject in an otherwise lecture-based curriculum, but I did begin to ask my students to read, answer questions about, and comment on, research papers from the medical literature, a process which has led to the entirely seminar-based course which I now teach.
One of the features of the McMaster course was that it was, like all North American courses, a graduate-entry course. At McMaster, they were willing to accept graduates with any major. They did not require a premed or science degree.
Other medical schools adopted the McMaster model, notably Harvard in the USA and Newcastle in Australia. From Newcastle, PBL spread throughout Australia, so that most Australian medical schools now have a PBL curriculum.
As far as I know, only four UK medical schools have adopted a PBL curriculum at the time of writing: Liverpool, Manchester, Glasgow, and St. George's. St. George's is a special case, as it has two medical courses. The five-year undergraduate entry course is taught in the traditional way, with a small "case-based learning" element added on (in which I have taught). The new four-year Graduate Entry Programme (GEP) is based on the McMaster model, a PBL course which takes graduates in any subject. We were greatly helped in setting this course up by Flinders University, and our course has considerable similarity to theirs.
In none of these courses are statistics and research methods taught as an integral part of the PBL. At St. George's, for example, we have a parallel non-PBL, seminar-based course. This attempts to match the illustrative material to the case of the week, but is taught separately from that case.
New medical schools, such as Anglia, Peninsular, and Hull-York, are being set up as PBL courses. This is particular suitable for courses based in more than one university, as in Peninsular (Plymouth and Exeter) and Hull-York. Problems can be set by teachers in either centre, and presented to students in both.
I went to Australia in search of the fully integrated teaching of statistics and research methods as part of the PBL tutorials. I did not find it anywhere. I did find three different models:
This approach was used or planned at the University of Sydney, the University of Melbourne, and the Australian National University. However, of these only the University of Sydney had actually put this into practice. The University of Melbourne and the Australian National University were about to implement what was essentially the Sydney model.
In this model, statistics and research methods are taught by PBL and the PBL triggers are integrated with the PBL problems for other parts of the course, but the material is not taught in the same tutorials or by the same tutors as anatomy, biochemistry and physiology. There will be a separate set of triggers for the statistics, etc., presented in a separate tutorial, and by separate tutors.
I asked why the main PBL tutors could not do this. As I understood it, the function of a PBL tutor is to facilitate and guide the group, not to impart knowledge. I had no problem, at least that I was aware of, in acting as PBL tutor when students were working with triggers designed to elicit questions about anatomy, biochemistry, and physiology, subjects of which I know virtually nothing. Besides, many of these tutors must routinely read journals which bristle with P values, t tests, correlation coefficients, etc. they must be familiar with the terms, if nothing else. Answers to this included:
The team at the University of Sydney were on the whole positive about their
course.
A parallel course
This approach was used at Monash University, the University of Queensland, Flinders University, and the University of Newcastle.
In this approach we have a non-PBL course which is given separately from the main PBL course. This may consist of any combination of lectures, seminars, practicals, web pages, or text handouts. Usually there is an attempt to link this to the PBLs by using examples related to the case of the week. For example, the case of the week might be asthma and the parallel course could include a critique of a paper reporting a trial of a treatment for asthma.
There are several problems with this approach to teaching statistics and research methods. As noted in the Introduction, the subject may seem peripheral to the main thrust of the medicine course. Student feedback tends to give a much lower approval to parallel courses than to the main PBL teaching. Finally, teaching is dependent on the cases chosen by the PBL teachers, who may change the cases or reorder them at little or no notice. This can make statistics teaching, which is much more dependent on the order of presentation than most subjects in the medical curriculum, extremely difficult.
Most people involved in these courses were unhappy with them, the exception
being the project in the Flinders course.
PBL used only for "public health" related subjects
This approach was used at the University of Western Australia.
This was an unusual model, found at only one university. PBL teaching had been initiated by an enthusiast, Sally Reagan, after a period spent at McMaster University. She was a member of the public health group and persuaded her colleagues to introduce PBL. However, only about one third of the course is taught this way, anatomy, biochemistry and physiology are taught traditionally.
The consequence of this approach is that the tutors are drawn from the population medicine area and so are quite happy to teach statistics, research methods and EBM. The triggers can be chosen as population-oriented problems, rather than being restricted to the patient case.
People I spoke to were very positive about this course, not surprisingly as they were the educational leaders in their institution.
I was surprised and disappointed not to find anywhere in Australia a single instance of truly integrated teaching of statistics and research methods through PBL. I have to conclude from this that there are considerable difficulties. I did, however, find an instance of unintegrated PBL teaching, at the University of Western Australia, where anatomy, biochemistry and physiology have to catch up.
What are these difficulties?
As Sir Humphrey Appleby was wont to remark, we should not see the problems but the opportunities. If we can integrate statistics and research methods into PBL, there may be great advantages.
One consequence of integration would be that the subject would not be marginalised or seen as separate. It would be just one aspect of medicine. Indeed, one of the features of a PBL course is that the distinctions between the different subjects should become indistinct.
A second advantage would be that students would be learning statistics and research methods in the contexts in which we hope that they will apply them: the interpretation of clinical data and the assessment of research evidence. The relevance of the subject should be very clear and during their professional careers they would be more likely to be able to recall and use this material when needed.
So how can we do it?
There must be an advocate for statistics and research methods at the start of the preparations for PBL, who is able to argue convincingly for the inclusion of these subjects. They must form part of the matrix of topics and problems. This is not an easy task and I have been told of great difficulty experienced at this level, as other teachers sought to exclude these subjects do that their own discipline could have more time. This is a natural human reaction, of course. I think that statistics and research methods are central to medical education, and would argue for more of them, at the expense of some anatomy if necessary.
We must get away from the idea that the problem must be a patient vignette. There are some statistical topics which can be covered quite conveniently in this way, such as measurement error, coefficients of variation, reference intervals, sensitivity and specificity, etc. After all, when my GP looks at my serum cholesterol on his practice computer, it has by the side of it a 95% reference interval. However, problems could equally be a published paper. We use these routinely in the teaching of statistics, research methods, and critical appraisal. The St. George's seminar-based course contains many examples which could easily be adapted for PBL use.
We could use a paper as a trigger for non-research methods topics. For example, a paper on an asthma trial could trigger questions about asthma as well as about randomisation. This may lead to too many questions being raised by the trigger. Another possibility would be to have such a trigger immediately following a case vignette problem on the disease in question. The patient vignette would raise the questions on anatomy, biochemistry, pharmacology, etc., associated with the disease. The following problem using a research paper would then raise only the research methodological questions. The fixed resource sessions in this week would then be devoted to these.
We may also, for variety, link research publications to the patient vignette by devices such as newspaper articles which the patient presents to the physician (e.g. reporting a trial), or say that in a patient problem the clinician has already found a Cochrane review.
Fixed resource sessions could include lectures, but I am very reluctant to suggest formal lectures in statistics and research methods for medical students. I would prefer to offer the open question and answer sessions that I do in my current seminar-based course, where students can ask me to explain anything they are unsure of. For example, if the trigger is a randomised controlled clinical trial with results presented in terms of P values, we might be expecting the trigger to lead to questions such as "why randomize?", "what does P<0.05 mean?", and "should patients be told they are in a trial?". Students should have made some attempt to answer these before the fixed resource session.
I remain optimistic that we can incorporate statistics and research methods into a fully problem-based curriculum. I think that if we do not, medicine will be the poorer for it.
I did not manage to visit all universities with medical degrees, though I did visit most. These accounts are my personal distillation of what I learned at them. I recorded a greatly varying amount of information, but wherever I went my informants were generous with their time and very willing to share their experience. The variety is entirely due to my note-taking.
The University of Western Australia has a six-year medical course with an intake of 140 students. Most of these are aged 17-18, but there are some mature students.
The curriculum is divided into two three-year parts, roughly corresponding to pre-clinical and clinical.
In the first three years, PBL is used only for part of the curriculum, those that we might think of as the public health or social aspects. These are collectively the Foundation of Clinical Practice and make up about 1/3 of the course. Anatomy, biochemistry and physiology are taught in the traditional way by the science faculty.
Typical problems deal with rural health and deprivation, aboriginal health, etc. Typical problems which bring in statistical aspects are a report on leukaemia in clusters (rates, P values, confidence intervals) and high blood pressure (intra- and inter-personal variation).
In the second three years, evidence-based medicine (EBM) is taught from start of year 4. The idea is to start with the use of synthesized evidence than move downwards to the individual study. There is a one-day EBM seminar in the first four weeks, which deals with review and meta-analysis, the basic summary statistics used, odds ratio, number needed to treat, etc. Then there are clinical PBL tutorials with the expectation that students will look at synthesized evidence. There is a new, voluntary EBM clinic. Students are encouraged to do wider searches and to present these with a critique. There are back-up question and answer sessions, which are much appreciated by the students. In the fifth year there are two-hour EBM seminars about three-weekly, criticising randomised controlled trials, etc.
Further comments from UWA are that is very important to make sure that the learning method is up front for the students, that they are familiar and happy with it. This is just as important as tutor training. There is a need continually to reinforce PBL ideas. We also need to develop PBL stimuli as the course progresses, or students become rather bored with it. Patient presentations can be used to enrich PBL experiences, as can varying the type of student response requested. An example at UWA is a court-room scenario used in the geriatrics block.
Informants: Sandra Carr, Judith Flynn, Ian Jacobs, Matthew Knuiman, Sally Reagan
Flinders has a graduate entry, 3 year course, with an intake of about 100 students, including a substantial number of fee-paying overseas students, mostly from North America.
Flinders University has a PBL programme. Flinders helped St. George's to set up the Graduate Entry Programme, and our course is largely based on theirs. There is very little statistics included in their problems. At the time the course was set up, there was no strong advocate in the school for statistics and research methods, though as the students are graduates we might expect that some or even most of them have covered these subjects already.
There is what is described as a statistics elective, consisting of statistical questions attached to the problems on their website. It is unclear how many students attempt these.
However, there is an extensive small-group project, in which students practice statistical and research methods skills. The project is carried out in a community environment in rural areas by students working in groups of 3 to 5. A list of projects of interest to the communities which students visit is compiled, and students choose one. They design the instruments, which might be a random or convenience sample, with face to face or telephone interviews, often using a mixture of quantitative and qualitative methods. The locations are small country towns 300-400 Km away from Adelaide. (These distances are unimaginable in a UK context!) The studies are design in consultation with the communities, and often receive considerable publicity in local media such as radio. About 30 students visit a community at once and the programme runs three times a year. Students have to produce poster of their study which can go back to the local community, and publication is encouraged. There is AU$1,000 prize for the best in the year. Communities value the students contribution and when students talk to the stakeholders in the communities they see that their efforts are going to make a difference, which is very motivating. Studies can build upon previous projects and there is feedback.
Ethics approval can be a problem with medical students' projects. The Ethics Committee reviews proposals in a block after review by the tutor, who also ensures approval by stakeholders.
These projects are seen as fitting closely with the PBL philosophy, as students
are presented with a problem which they have to solve. Some sample projects
were:
Informants Neil Piller, David Prideaux.
The course at Monash is in a state of transition from the previous non-PBL 6-year curriculum to a new PBL 5-year curriculum. The intake is about a total of about 140 Australian and 70 Malaysian students, mostly aged 17-18.
In the new curriculum, now in its second year, statistics is not taught as part of the PBL, but as a parallel course. Integration is greatly emphasised and so biostatistics and epidemiology are taught in an integrated way. There is a substantial practical and seminar based course in year one. There is only one lecture to orient to students in week 1. Epidemiological material was provided on a CD ROM, but this was reading rather than interactive. Statistical material was given on paper and much of the epidemiology was given on paper also, as it was needed in class. Students got a block of material covering four weeks at once. The student response to the CD was discouraging, few seemed to use it.
The course is taught to 16 groups of about 14 or 15 students. This is done by having 8 groups on two different days, because of competition for tutors with other courses. Tutors are mostly epidemiologists, including Ph.D. students, about half are medically qualified but only one is statistically qualified. Most have a health sciences background, MPH, etc. It is a 2 hour tutorial, including a half hour break for coffee.
This is a summary of the course as it ran in the first year of the new curriculum:
Week 1. Opening lecture to entertain and motivate the students. Students were introduced to the idea of epidemiology using identification of HIV and AIDS (extract from the television film "And the Band Played On"), given a lecture on why statistics is important in medicine, and given a quiz on their current knowledge.
Teaching in the following weeks is seminar based, students told to do reading in advance, most don't.
Week 2. Exercise, mostly calculations based on a document on bites and stings.
The document was long.
Week 3. Rural visit. (Students got back having forgotten the start of the
statistics and epidemiology course.)
Week 4. Making measurements, external and internal rotation of hip.
Distribution and measurement. Paper on prevalence of osteoarthritis,
cross-sectional studies. This was a long paper, 14 pages of A4 print.
Week 5. Standard error, random sampling, using last week's data as a
population. A practical exercise which didn't work very well. Scenario of
occupational health rather than a full paper, concepts of "normal" and
reference ranges.
Week 6. Confidence intervals and significance tests, t distribution, measures
of association: relative risk and odds ratio, cohort studies. Paper reading
exercise. Too much statistics for one session!
Week 7. Proportions and two-way tables, case-control studies, exercise on
case-control paper.
Week 8. t tests, ethics of trials. Paper using t tests, critiques of
randomised controlled trials using abstracts.
Week 9. Diagnostic tests, paired t tests, McNemar's test.
Week 10. Outbreak epidemiology.
Week 11. Correlation and regression, association and causation. Example on
shingles.
Week 12. Non-parametric methods, exposure assessment.
Week 13. Revision.
Topics are all linked to the case of the week, though students seldom appeared to make the link.
Assessment is integrated. There are two mid-semester tests, MCQ and short answer, covering the whole course. At the end of the year there is an OSCE, totally integrated with a case-based structure.
The organiser, statistician Rory Wolfe, commented "If you lose the students early they will ignore the subject and concentrate on anatomy, etc. Integrated assessment allows this. There is too much work in the course to make it worth their while, if they take a strictly practical view. We are trying to teach material not core to students' interests."
Plans for a change of the running order of cases the second time the course is run are causing problems. This is currently under discussion. The order of presentation of topics in statistics is difficult to change and it seems likely that unless the team are going to rewrite most of the statistics course the notional link with the case of the week will be broken. They will be running with the same course in the second year.
In the first run, this course did not get as good feedback from students as other parts of the first year. David Goddard, a tutor on the course, told me that when they reached standard error and confidence intervals, his students rebelled. David just could not find the right words to help them. He then produced his own handout. Things improved and this handout was widely copied. He thought that this course was a moderate education experience, many students were not enthused.
What will happen in later years of the 5 year programme? There are four themes running through the curriculum, and statistics and research methods come under Population Health. In year 2 this includes knowledge management (where you get information from and whether you believe it, done by the Institute of Health Services Research at Monash), health economics, and health promotion. In 3rd year they will study critical appraisal, linked to the case of the week. In the 4th year the students will study preventive medicine and health service management.
David Goddard is a dedicated teacher with a lot of good ideas. He challenges the students by asking whether they want their careers to be as leaders or followers of opinion. David has many ideas for involving students in the classroom, such as a bingo game where small groups of students compete for a little prize. This uses numerical questions and gets collaboration going, bringing out the competitive spirit. Another is a crossword against the clock. He likes to challenge them with questions like "How would the world be worse off without P values?"! Well, how would it?
Informants: David Goddard, Rory Wolfe.
The University of Melbourne has a six-year programme with an intake of 270 students, 2/3 aged 17-18, 1/3 graduates. Some graduates can skip a year, the advanced medical science year in semesters 6 and 7, when students can choose from a great variety of subjects to study in depth, laboratory subjects, anthropology, medical humanities, etc.
A PBL course has been running for about five years. In the first 5 semesters, basic medical sciences are taught via PBL tutorials, using two tutorials per problem. At present public-health-related areas, including statistics, research methods, and evidence-based medicine, are not taught as part of the PBL programme, and consistently get lower student ratings than the PBL subjects. This course is currently undergoing considerable revision.
The new plan is to embed these topics in the PBLs, integrated with them but not taught by the same tutors. There will be a separate tutorial for each problem, where additional material related to the problem will be presented to the students. This second tutorial will be taught by "community oriented" tutors, from public health, general practice, etc. These sessions will be devoted to teasing out non-biological science issues. There will also be separate fixed resource sessions (e.g. lectures).
In semesters 8-12, much teaching is problem-based, but not PBL in the true sense, sessions are more tutor-oriented. Tutors in this clinical period have reported that students who have been through the PBL programme are much more independent in their thinking than students taught under the old curriculum.
Informants: John Carlin, Sue Elliott, Steve Farish, Terry Nolan
The Australian National University currently has no medical degree, but some students from the University of Sydney are taught at Canberra Hospital during their clinical period. However, the University is about to embark on a new medical degree, and the intake of Sydney students will be phased out as the course takes its first students in 2004.
The new course will be PBL, only the first two years have been planned. They will be using individual patient problems, with EBM in the clinical skills theme.
Informant: Bruce Shadbolt
The University of Sydney has a four-year course with an intake of 240 students, all graduates. The course is PBL, with three tutorials per problem. Problems are mostly individual cases, mostly set in a GPs surgery. The course is highly web-supported. All case material is presented on an intranet via terminals in the tutorial rooms. These include many still images with voice-over. A lot of resources are provided throughout the week. Apart from search facilities, there is a set of learning topics, a 1-2 page overview on the web as a kick start. There are up to 6 lectures and other theme sessions (labs, communication skills, etc.) but no dissection. The web-based nature of the course makes continual revision a relatively straightforward task.
EBM activities are included in the cases, specifically given this title. There are EBM add-ons to some of the cases and tutors are asked to point out to the students that there is an EBM activity for this case. Clicking the EBM box then brings up the related EBM activity.
In later years there is an "old-fashioned" clinical epidemiology course, using lectures and tutorials with EBM tutors. The sessions are linked to the case of the week.
Assessment at the end of the first year is formative only, at the end of the second year it is summative. The assessments include both written and practical tests, small clinical problem scenarios and multiple choice questions. A journal article is given to students two weeks ahead of the assessment.
One pre-registration task at the end of the course is to ask a patient "What do you really want to know?" and get an evidence-based answer, which is marked by the patient!
Informants: Alex Barrett, Jill Gordon, Greg Ryan, Lindel Travino
The University of Newcastle has a mixed intake of graduate and non-graduate students.
The University of Newcastle was the first medical school in Australia and one of the first in the world to adopt PBL. Currently, they have two PBL tutorials per week with a separate biweekly public health tutorial. In the first year the public health tutorials include elements of study design, in the second year they cover basic biostatistics and decision making, sensitivity and specificity, etc. The public health trigger material is matched to the case of the week.
Informants: John Attia, Julia Byles, Bob Gibbard, Michael Hensley, Alison Koschel
The University of Queensland in Brisbane has a graduate entry course, which uses problem based learning. Evidence-based medicine is taught in a parallel stream to the PBL course, using material linked to PBL cases. There are 240 students per year, and they hare taught in EBM tutorials by about 10 tutors. Tutors are from public health or EBM backgrounds. There are about 25 students per tutorial group, though not all attend.
It was assumed at the outset that being graduates the students would be familiar with statistical concepts, but this proved not to be the case. Thus teaching on P values, etc., had to be included in the tutorials. Paul Glasziou has the students work on the material in pairs or triplets and wanders around gathering problems which they raise to add to a cache on the board, then goes through them. Each tutor does it in their own way.
The general view is that there is great rigidity in approach in the course. They tried to get EBM integrated into PBL in the initial planning, but failed. There were many objections to non-biological content and were unable to get material included. At one stage there was to be one lecture on EBM as the entire course! This led to the development of the present parallel course.
A typical example: in a case on deep vein thrombosis, the EBM content was to be the inclusion of a couple of abstracts on aspirin and on flight socks as prophylaxis during long-distance air journeys. This was vetoed on the grounds that providing these abstracts would not be PBL but didactic teaching! Of course, the abstracts are not the answer to the question on whether the a patient should take aspirin or wear flight socks, but the triggers for questions about the acceptability of the research evidence.
Informants: Paul Glasziou, Marie Louise Dick
I want to thank a lot of people who have helped make this project possible. First, the many people in Australian universities who provided me with the essentials of life, such as computers, beds, and food, particularly Max Bulsara, David Prideaux, Jenny McCulloch, Andrew Forbes, Judy Simpson, Caro Badcock, Kerrie Mengersen, Denise Schultz, and Paul Glasziou, and old friends who provided the same, Susan Button, Marilyn Chalkley, and Hugh Sadler. I must thank the many medical educators who provided me with information about their courses, as acknowledged above. Next, I thank those at St. George's Hospital Medical School who enabled this trip to take place, Robert Boyd, Principal, and Ross Anderson, my longtime head of department and friend, both of whom supported it even though I then resigned my post to go to York. Finally, I am greatly indebted to Janet Peacock, fellow statistician, friend, and right-hand person, who took over my teaching and leadership of our little statistical group while I was away having fun.
Back to Some full length papers and talks.
Back to Martin Bland's Home Page.
This page maintained by Martin Bland
Last updated: 12 January 2005.