Using Clinical Experience in Discussion within Problem-Based Learning Groups

Using Clinical Experience in Discussion

within Problem-Based Learning Groups

PAUL O’NEILL*, AMANDA DUPLOCK and SARAH WILLISThe Medical School, The University of Manchester, Stopford Building, Oxford Road, Manchester,

M13 9PL, UK (*author for correspondence, Phone: +44-0161-275-7792; Fax: +44-0161-275

5584; E-mail: p.a.o’[email protected])

(Received: 27 April 2006; Accepted: 27 April 2006)

Abstract. A key principle in problem-based learning (PBL) is the student linking learning from

different sources to enrich understanding. We have explored how medical students based in a

clinical environment use clinical experience within PBL groups. We recorded the discussion of

12 third-year groups, which were meeting for the second time on a PBL case, where students

report back on the learning objectives. Discussions covering five separate PBL paper cases were

recorded. Analysis of the transcripts was based on constant comparative method using a coding

framework. The range of discussion segments of clinical experience was 2–15, with 9 of 12 groups

having at least five separate segments. Our initial coding framework covered 10 categories, of

which the most common were: a specific patient encounter (19%); an experience in the com-

munity (15%); and a personal health experience (15%). Students often used emotive phrases with

37 examples in the clinical experience segments compared with 9 from the longer non-clinical

discussion. Most clinical descriptions triggered further discussion with almost half leading to

some related medical topic. The discussion segments were subsequently coded into; �confirming’

(40); �extending’ (40); and �disconfirming’ (16) the understanding of the group for that topic.

Discussion of clinical experience encouraged students to connect to the affective aspects of

learning. It helped students to bridge between the tutorial and real clinical contexts. A clinical

experience was often a powerful pivotal point, which confirmed, extended or refuted what was

being discussed.

Key words: clinical experience, group discussions, problem-based learning

Introduction

As medical students progress through their programme, a key question ishow they integrate their learning from the early years of study, applying thisto patients with real clinical problems and developing professional expertise(Boshuizen et al., 2004). Students report difficulties in making the transitionfrom theoretical learning in the medical school to learning in clinical settingseven in schools that use problem-based learning (PBL) in the early years,which contextualises learning around (paper) clinical cases (Prince et al.,

Advances in Health Sciences Education (2006) 11:349–363 � Springer 2006

DOI 10.1007/s10459-006-9014-6

2000, 2005). In response to this, one common recent approach to curriculumdesign has been to introduce �early practical experience’ (Dornan and Bundy,2004; Littlewood et al., 2005), which was reported as helping students inlearning behavioural and basic sciences.

In 1994, we took an alternative curricular approach in deciding to con-tinue with PBL in the later clinical years of the programme with the aim ofencouraging students to link the discussions in their PBL group with clinicalexperience, so that they could start to generalise from a paper problem to realpatients with similar conditions (O’Neill et al. 2000).

Overall, we have demonstrated positive learning outcomes from theintegrated curriculum using a variety of evaluation methods (Jones et al.,2002, 2003; O’Neill et al., 2003). We have also published more specificevaluation data on some of the processes involved in undergraduate studentlearning, including a paper focussing on the ways students link between PBLand clinical experience (O’Neill et al., 2002), which is of particular relevanceto the study reported here. However, there are few other published studies onthe use of PBL in a clinical setting and how it affects student learning.Schwartz et al. (1992) have described its acceptability within a clinicalclerkship, but a later review by Foley et al. (1997) concluded there was littleevidence available and called for further work. More recently, Dammerset al. (2001) reported that using real patients within PBL groups was a po-sitive experience for the students involved, but the numbers were relativelysmall and the students had self-selected and hence their findings may lacktransferability to other settings.

In an initial questionnaire evaluation of our new clinical programme,(O’Neill et al., 2000), students reported that they were generally confident inbeing able to relate clinical experience to other knowledge. However, the studydid not explore how the students constructed these links. Subsequently, weinvestigated theways that students linked paper cases to real patients and othertypes of clinical contact and constructed a theoretical model, based on theconcept of elaboration derived from cognitive learning theory (O’Neill et al.,2002). The study used written evaluation comments from a large number ofstudents and then expanded on these comments qualitatively through focusgroups with third and fourth year undergraduate medical students. We foundthat elaboration could occur either through a subsequent clinical encounter(i.e. an event taking place outside the PBL group) or through group discussion(i.e., during a PBL group, and hence inside the group).

The study presented here builds on the previous work by looking at waysin which students discuss and use clinical experience within PBL groups. Itdoes not rely on participants recalling instances where this kind of discussiontakes place: rather, it records what is actually taking place during a series ofPBL groups and uses these recordings as the units of analyses. Our aim was

PAUL O’NEILL AND SARAH WILLIS350

to generate hypotheses using exploratory, qualitative methods (Denzin andLincoln, 2000; Greenhalgh and Taylor, 1997).

Description of PBL in the Medical Programme

Our curriculum is designed as a spiral, with 3 distinctive phases to PBL groupwork over the whole 5-year programme. Themodel of PBL used in years 1 to 4(phases 1 and 2) is a �closed-loop’ in which, after initial discussion andformulation of hypotheses and questions regarding a paper case (Barrows,1986), students undertake private study and then return to the group to discusstheir findings and evaluate their original thoughts about the case.

In phase 1 (years 1 & 2), the groups meet three times per week in themedical school and the overall goal is for the PBL paper cases to stimulatelearning about normal form and function.

In phase 2 (years 3 & 4), the students move from the medical school to aclinical base. As we are a large medical school with over 400 graduates peryear, the students are based in 1 of 4 �Sectors’ from year 3 onwards. EachSector consists of a large teaching hospital and several affiliated districtgeneral hospitals. Within the Sector, a hospital dean and his/her team areresponsible for organising the clinical firm attachments, clinical skills labo-ratory activities, lectures, seminars, workshops and other support activitiesfor the PBL cases and curriculum.

During phase 2, each PBL group consists of approximately eight students,who meet for one hour with a tutor to discuss a paper case using a setapproach (O’Neill et al., 2002). The tutors of the groups in year 3 onwardsare predominantly NHS consultants engaged in clinical practice. The stu-dents set group learning objectives (questions) at the end of the first dis-cussion session and then meet again after 1 week for 90 minutes to discusstheir findings.

Given that the students in phase 2 onwards are working in a clinicalenvironment, the students can seek out clinical experience as well as using, forexample, their personal experience, books, articles and lectures to answer theirlearning questions. In order to encourage this, we have modified the PBLgroup process to emphasise the need to link to clinical experience (O’Neillet al., 2002). The PBL cases consist of clinical stories that combine diagnosis,investigation and management as well as the relevant basic and clinical sci-ences and the wider psychosocial aspects of clinical medicine. In phase 2, year3, the students study two core modules: �Heart, Lung and Blood’ (HLB) and�Nutrition, Metabolism and Excretion’ (NME), in which the PBL cases arederived from the core curriculum in these themes (O’Neill et al., 1999).

During each of the 14-week core modules in year 3, the students have 2clinical clerkships lasting 7 weeks each. When designing the programme, itwas decided that the concept of apprenticeship learning through longer

PROBLEM-BASED LEARNING AND CLINICAL EXPERIENCE 351

attachments would take precedence over exposure to all clinical specialities(O’Neill et al. 2000). It followed that the speciality of the clinical clerkshipwould not necessarily match with the content of the PBL case, though effortswere made to offer some relevant clinical experience (e.g. being on a Urologyclerkship during the �NME’ module).

Whilst the majority of each week in the year 3 is spent on clinicalattachments in hospitals, one day per week is set-aside for students to makeuse of community resources, which are based around a particular primarycare practice with a general practitioner tutor.

In year 4, students again study two core modules: �Families and Children’and � Mind andMovement’. In the latter, students start to use the problems ofreal patients that they have seen as the focus for discussion in their PBL groups.

In phase 3 (year 5), the approach to PBL is developed further by not havingset cases to discuss; instead the students set the agenda for their group workthrough discussing patients that they have encountered in their clinical(including community) attachments. This last year is designed as an appren-ticeship whereby students have five different attachments and learn by shad-owing doctors within each attachment. An evaluation of the outcomes and thegroup work in phase 3 has been reported elsewhere (MacPherson et al., 2001).

Methods

The study was carried out in one of the four teaching hospital Sectors whilststudents were studying the second module in year 3 of the undergraduateprogramme. In order to make efficient use of clinical resources and to matchbest the hospital attachments with the theme of the module (O’Neill et al.,2002), half of the students undertake the HLB module as their secondrotation of the year and the other half undertake the NME module. Thus,some of the PBL groups we studied were in the NME module and some werein the HLB module.

A medical student (AD) carried out the study as part of her requiredresearch project in year 4 of the programme. As preparation for the research,she was trained in qualitative observation by the other authors of the paper.This included training sessions using existing videotape recordings of PBLgroups to ensure that the study protocol could be followed and that the fieldnotes were of a high standard.

Due to constraints with the timing of the student’s research project andthe year 3 modules, all observations took place during weeks 9–11 of thesecond module in year 3 of the programme. The student briefed each PBLgroup together with their tutor about the general nature of the study usingdelayed informed consent (Boter et al., 2003). This approach was adoptedbecause informing the students and tutors about the precise research ques-tions at the time of recruitment may have led the participants to alter their


behaviour significantly. Delayed informed consent entailed the participantsagreeing to take part in a study of PBL, with the understanding that theywould be informed about the research questions at the end of the period ofdata collection. The University ethics committee approved the studyincluding the consent procedure.

Once participants had given their consent, each group was observed andtheir discussion recorded during their second group meeting considering aPBL case. The researcher (AD) recorded detailed field notes, which were usedin the later analysis. The recordings were transcribed and imported intoNVivo software for ease of subsequent analysis.

Before any detailed analysis, the research team considered the kinds ofexamples of clinical experience that might be introduced during a PBL group.We wanted to allow not only direct clinical experience gained in a hospital orgeneral practice, but also personal health experience and indirect observationof medicine seen in dramatised settings such as popular medical televisiondramas. An attempt was made to capture the frames of reference for thestudy by constructing a series of initial codes (see Table I). We allowed for an�other’ category of clinical experience so that our coding of any data wouldnot be overly prescriptive, but grounded in the conversations of the medicalstudent discussants.

Following this preliminary sensitisation to the kinds of discussion ofclinical experience that might be present in the data, each author then readthe transcripts and decided on the segments that contained discussion ofpossible relevant experience. Using the initial coding framework, the

Table I. Initial Codes of (clinical) experience and proportion within the clinical experience dis-

cussion segments

Code Definition Proportion (%)

Clinic Experience gained in hospital clinics 3

Consultant Experience with a specific consultant 13

General

Practitioner

Any clinical experience gained at the

community placement

15

Patients Any reference to a specific patient 19

Personal A personal experience related to a clinical scenario 15

Tutor Any clinical experience encouraged by the tutor 8

Media Reference to a clinical scenario on television, film, media 7

Wards Any clinical experience from the wards 6

Work A clinical experience from work or work

experience outside the University

3

Other An unclassifiable experience (lack of

sufficient detail available)

11


transcripts were encoded in NVivo and decontextualised for subsequentfurther detailed analysis (Denzin and Lincoln, 2000). We then evolved asecondary coding (category) framework for the individual segments ofdiscussion of clinical experience using a constant comparative method,including refinement with negative or discrepant instances (Pope et al.,2000). In order to help with presentation of the summarised data, we haveprovided simple counts in the results, but, in keeping with qualitative dataanalysis, these have not been tested statistically (Bickman and Rog, 1998;Pope et al., 2000).

Results

General

We observed and recorded the second (1.5 h) PBL meeting of twelve groupsfrom year 3 of the programme with seven groups from one core clinicalmodule and five from the other core clinical module. In these twelve groups,we observed discussion of one of five separate PBL paper cases: Haematuria,urinary incontinence or renal impairment (NME); and somatisation or leu-kaemia (HLB).

In 5 groups, there were more than 11 segments of discussion of clinicalexperience, but in 3 groups it was fewer than 5 segments. There was a similarvariation (range 47 s to 12 min 30 s) in the amount of time spend in dis-cussing clinical experience with two groups spending more than 10 minuteson this aspect, but in one group it took less than one minute. For theremaining groups, it was between 1 and 10 minutes. The clinical discussionsegments were classified into the 11 initial codes (Table I), of which the mostcommon reference was to an encounter with a specific patient (19%), fol-lowed by an experience in the community (15% – �at my general practitio-ners’ [GP]), and a personal health experience (15%). A reference to the mediawas found in 7% of discussion segments.

In their clinical attachments, students work in pairs and this was apparentin their discussion, where a student frequently added further detail to theclinical account that had been begun by their partner. Students often usedemotive phrases (e.g. �and it was the most awful thing’, �distressing’, �amaz-ing’, �impressive’) in their descriptions, with 37 examples of this kind oflanguage compared with only 9 from the much longer segments of non-clinical discussion.

Student 1: ‘‘And cyclosporin as well can cause chronic renal failure?’’Student 2: ‘‘Oh I’ve seen that.. Well I didn’t see it actually happen’’Student 1: ‘‘Well what do you mean?’’Student 2: ‘‘No, no, I saw somebody in the hospice die after he’d been on dialysis for

20 years. And then he got pneumonia and he decided that he did not want


to die like that. So he stopped his dialysis. And it was the most awful thingI had ever seen. Someone dying’’.

Student 4: ‘‘Is ciclosporin an immuno-suppressant?’’Student 6: ‘‘Yeah.’’

Most descriptions of a clinical experience triggered further discussion, withalmost half leading to some related medical topic (46/96), and in a further20% it initiated the discussion of another clinical incident:

Student 3 ‘‘ ... He was coming up on his sixth cycle of chemotherapy, so it was like achance at the last’’.

Tutor: ‘‘Do you all understand the drugs that he was on and why he was on them?’’Student 8: ‘‘At X I’ve seen quite a lot of haematology patients. They have like this,

they’ve got, on the ward I usually work on, they’ve got the specialisedisolation units for them, for when they’re neutropenic. They’re alwayslike really clued into their illness. Cos it’s a big kind of part of theirtreatment, the interaction between the doctor and patient.’’

Detailed secondary analysis

Following on from the initial analysis using the primary codes, we re-examined the de-contexualised segments in a more detailed analysis. Therewere three main categories that emerged, which best explained how clinicalexperience was being used within the PBL group meetings.

Confirming

The category of �confirming’ is where students in the groups used adescription of some clinical experience to confirm what was being discussedwithin the group at that point. These segments were often short. Out of 96discussion segments, 40 were categorised as being �confirming’ of the dis-cussion topic. One example was where the group were discussing the man-agement of people with anxiety:

Student 6: ‘‘Drugs only ever take some of it away and people that I’vespoken to say drugs only ever really take the edge off it, theydo not really cure you’’.

Or in discussion of senile purpura:

Student 3: ‘‘I’ve met a lady who had it, she had lots of investigations and they justdecided as you get older, your skin gets thinner , so it had this purplepattern...’’

Student 7: ‘‘Like little bruises’’

Sometimes the segment was longer, but seemed to agree with the overallconclusion of the group. In the example below, an experience is brought


forward as unusual and confirming that most bladder tumours are detectedon review:

Student 1: ‘‘Yeah, eighty percent are confined to be bladder mucosa And twentypercent penetrate the muscle. And that carries like a 50% mortality. But Ithink that a lot of them are picked up just kind of...’’

Student 2: ‘‘Just on review’’Student 3: ‘‘Yeah, just on review. I mean, like, cos a lot of them, it depends where it

is, cos a lot of them get.. there’s another guy who had loss of frequency,cos just had like a bit of tumour that kept flopping over the top of hisurethra or the thing that, you know, the bottom of the bladder. And thatis why her first noticed, so it was really small. But because of the symp-toms, he managed to pick it up early. So I think it depends on where it is.’’

In this category and also in the other categories described below, studentsused examples of simulated experience that they had come across in themedia to influence a group discussion. In this example, the students werediscussing the genetics of blood disorders (Leukaemia case):

Student 2: ‘‘This was on ER last week by the way’’Tutor: ‘‘Oh was it?’’Student 1: ‘‘Oh yeah, the woman had it, yeah’’Student 2: ‘‘And then the dad had a blood test and they were like �this is not your

son’ and he was like �yes he is’’’Student 1: ‘‘Because he wasn’t a carrier, but his son had thallasaemia’’

Extending

We found that the category of �extending’ was as common as �confirming’with 40 instances from the 96 discussion segments identified. The categoryencompassed a discussion of an experience that broadened or extended thegroups thinking around a problem through illustration with a particularexperience. In the first example, the group are discussing the presentation oftransitional cell carcinoma and having agreed on the male: female ratio, goon to discuss the usual age of presentation:

Student 1: ‘‘Yeah. And again, it tends to be slightly older, in your sort of fifties, youcan get this more in sort of sixties and seventies, it’s quite a lot older...’’

Student 2: ‘‘It can happen in the forties’’Student 1: ‘‘Oh, well that’s rare...’’Student 3: ‘‘The guy I saw was forty’’.Student 2: ‘‘I’d go with forty six’’.Student 3: Forty five/forty six and all that.Student 4: ‘‘Really?’’Student 3: ‘‘Yeah’’.Student 3: ‘‘Okay’’


Experience also broadened a group’s view away from a simple biomedicalmodel and towards a view of how illness impacts on people: In this example,a student expresses admiration for a patient who is coping with dialysis,whilst looking after her family and working

Student 1: ‘‘I remember going to a clinic and chatting to some people on dialysis andI asked this woman who had to be on dialysis and she was sort of like fourhours, she had to come into hospital for four hours, three, three times aweek or something. And she had a full time job, she had three kids, I said‘‘How do you cope with all that?’’

Student 8: ‘‘Was she waiting for a transplant?’’Student 1: ‘‘Yeah, but I mean how, how do you mange a full time job and three kid?

It’s quite Amazing’’.Student 6: ‘‘Impressive isn’t it?’’

Students placed weight on their observations of doctors in clinical practiceand any associated teaching or discussion. Hence one student told the groupthat when she had seen her consultant breaking bad news, the doctor had aparticular approach:

Student: ‘‘Dr Y used to say an abnormality, then a lump, then a tumour, thencancer’’

Disconfirming

Less frequent was where students used experience to �disconfirm’ the conceptsthat the group was developing. We identified 16 instances of �disconfirming’from the 96 discussion segments. In this segment, the students are reaching aconclusion that patients with renal failure present with multiple symptoms.However a student describes a patient with no symptoms apart from backpain, which was not on the list just described.

Student 2: ‘‘Yeah, in the early stages it, it can be asymptomatic or you get toxicmetabolites accumulating in the blood ...so, malaise, loss of energy, lossof appetite, insomnia, itching, nausea, vomiting and diarrhoea, parast-haesia due to polyneuropathy, restless leg syndrome, tetany, peripheralpulmonary oedema, anaemia and...’’

Student 3: ‘‘Well, when we were talking to that guy at the GPs, he hadn’t got anysymptoms apart from back pain. (talking together). In the investigationof back pain they did blood tests and they found the high calcium and hehad really high urea and creatinine...’’

As with other examples, the introduction of clinical experience works tocontextualise the discussion into real medical practice. Introducing a patientencounter shifts the style of the PBL discussion away from talk that consistsof replicating the formalised language of a medical textbook (‘‘It can be


asymptomatic or you get toxic metabolites accumulating in the blood’’) andtowards a more conversational style (‘‘that guy at the GPs’’).

However, the introduction of a disconfirming experience could be apowerful lever moving the group to develop a confusing or incorrectconceptualisation of a particular condition. In the example, the students arediscussing chronic renal failure and calcium and phosphate metabolism.

Student 1: ‘‘In fact with saw a guy at the GPs who’d had a renal transplant’’Student 8: ‘‘Yeah, I remember him’’Student 1: ‘‘Yeah, he had hyper.., hyperparathyroidism’’Student 3: ‘‘Was that secondary to kidney disease?’’Student 1: ‘‘No, no’’

In fact, secondary hyperparathyroidism is one of the complications of renalfailure.

The use of clinical experience could also mislead the group if it not fullyexplored in the subsequent discussion or if the student/group lacks theunderpinning knowledge to counter the explicit or implicit inference fromwhat is being said:

Student 6: ‘‘Yeah, just saying that the GPs were showing us some of the notes on thecomputer from the patient who came in. I can’t remember what hispresenting complaints were but pretty non-specific. Investigated and heturned out to have acute myeloid leukaemia. If I remember rightly was hea Jehovah’s Witness? I think he was dead within a week to ten days.Because he wouldn’t have any blood transfusion or anything.’’

Student 2 (chairperson): ‘‘Shall we get back to the diagnosis of it?’’

In this instance, the chairperson is closing down the discussion of clinicalexperience and preventing the group from unpacking the inference.

From this detailed analysis of the segments discussing experience, wederived a model of how these were used in confirming, extending ordisconfirming the construct that students had around a particular aspect ofthe problem under discussion (Figure 1).

Discussion

Clinical experience was often introduced in a PBL discussion group by stu-dents using affective terms and appeared to help in bridging between thepaper cases and real clinical contexts. About half of the discussions about aclinical experience linked into some other clinical incident or medical area. Inmany cases, the introduction of a clinical experience was a pivotal point in adiscussion that confirmed, extended or refuted (disconfirmed) what was beingdescribed from other sources of information. Thus, discussion of an experi-ence can be seen as being used not only as another information resource,but also in broadening constructs around clinical problems including the


psycho-social aspects of medicine and also the emotions involved. In thisway, it was powerful and has implications as to the development of clinicalexpertise.

There are a number of limitations to the study that need to be highlighted.The process of observing and recording the groups may have significantlyaltered the process of discussion. We attempted to mitigate this in a numberof ways. The observer was a medical student so the group members shouldhave felt more comfortable with her presence compared to a more senior staffmember. We also used a consent procedure (Boter et al., 2003), which in-volved delaying informing the group of the detailed research questions untilall observations were complete. This could be criticised as not being properinformed consent, but to do otherwise would have probably strongly influ-enced the nature of the group discussion.

We observed almost all of the groups from one particular year based inone teaching hospital Sector, so this means that the results cannot be readilygeneralised to how experience might be discussed by student groups later inthe programme or at other sites. The results are also confounded to somedegree by the timetabling necessity of recording PBL tutorials on five sepa-rate cases. Consequently, we could not explore whether the nature of the PBLcase was a significant influence on stimulating discussion of experiences. Wedid find that the proportion of time spent by groups talking about their

Direct clinical experience of student

Discussion with clinical teachers Other experience

Personal ‘Media’

Community (GP)

Hospital

Confirmation ofConstruct

Extension ofConstruct

Disconfirmation of Construct

Discussion of experience leads to:

Elaboration INSIDE the Group through student experiences

brought for discussion

Figure 1. Model of how PBL groups introduce and use experience from different setting

to help in development of their constructs.


experiences varied two-fold across the cases, but cannot go further withoutmore research. However, being able to analyse across the PBL cases, therebyweakening influence of a particular case, offsets this drawback.

The initial coding framework was agreed after discussion betweenthe authors. Our intention was to focus mainly on the experiences theundergraduate programme provided through clinical attachments. However,it became clear in our preliminary work that students often introducedexperiences from other settings. Hence, we included vicarious experiencesfrom the media and personal history within our coding and, as judged by theconversation around these in the PBL groups, these were viewed as validtopics for discussion. This included personal experience (�It was a girl I knew atprimary school. My little brother’s childminder’s daughter’) and also referencesto the media (�I remember seeing on ‘‘Richard and Judy’’ or something like that.They were saying women are supposed to like sort of tense for like a minute a daytheir pelvic floor to try to increase the strength. Supposed to, like you’re trying tohold urine in’). These references to the media, whilst not personal experience,are important as other studies have reported that the portrayal of illness andits management may be misleading (Diem et al., 1996).

Almost one in five of the descriptions of an experience concerned a specificpatient, indicating that it is a powerful influence on students thinking, whichthey bring to PBL groups. These encounters – as well as personal experiencediscussed in the groups – are often highly emotional, and hence more likely tobe remembered and brought up in discussion of a relevant case. This hasparallels to that described by Kenny and Beagan (2004), emphasising theimportance of discouraging �student detachment from the messiness of realpatients’ lives and emotions’ in cases designed for PBL.

Interestingly, incidents from community attachments also appearedfrequently, even though these occupy only one day per week during themodules. Students almost invariably referred to this as �at my GPs’ rather thanby name, which is in contrast to how they described hospital experience (�wewere with Dr X’). There are a number of possible reasons for this. Firstly,students have the hospital environment in common; so referring to a doctor byname is likely to be meaningful for other members of the group. Secondly, thehospital doctor could be viewed as being particularly influential, which wehave illustrated in the results. The third reason is that �my GPs’ is a simplelabel for a complex set of experiences including the general practitioner, thepractice staff, practice and local environment as well as the patients. We haveexplored this in a previous study (Silverstone et al., 2001) suggesting thatshort labels should not be taken at face value in programme evaluation.

The relative time spent in discussing experience was small when comparedto the length of the tutorial. Yet, the majority of groups had five or moreseparate instances in each groupmeeting, whichmeans that the pattern was for


short, but frequent discussions. Our more detailed analysis showed anunderlying schema could explain these based on categories of confirming,extending, or disconfirming the construct of that aspect of the problem thatwas being discussed. The influence of these examples was commonly reinforcedby the use of emotive terms, which were supported by the clinical partner of thestudent. More recently, others have looked at how PBL groups use knowledgein their group discussions early in an undergraduate medical curriculum. In apilot study, Visschers-Pleijers et al. (2004) used the coding system of VanBoxtel (2000) to analyse group interaction and found that mostly groups en-gaged in �co-construction’ indicating that understanding was being formulatedbetween group members as distinct from �elaboration’, where a group memberthought through aloud his or her reasoning. In our study, most of the dis-cussion segments were apparently �co-constructions’, though it would be ofinterest to use Van Boxtel’s (2000) approach prospectively in a future study. Ina subsequent paper, Visschers-Pleijers et al. (2005) have reported that PBLgroups spent most time on �cumulative reasoning’, in which the students en-gage in �automatic consensus’. They suggest that more emphasis should beplaced on promoting discussion around contradictory information, a role thatclinical experience appeared to fulfil for our students.

Previous researchers have looked at how students from PBL and con-ventional curricula use knowledge in problem solving, but these were basedon the early use of PBL, not in the later clinical years (Patel et al., 1991) andfocussed on how bioscience knowledge might be used in clinical reasoning(Patel et al., 1988). Our findings suggest that students in a clinical environ-ment do not use clinical experience to illustrate bioscience principles, butrather to shape the construct they hold of different aspects of clinical prob-lems. This has implications for the development of clinical expertise in that attimes, students used experience to confirm the typicality of the paper case(prototype) and also to extend their understanding of the different ways inwhich a clinical condition might be seen (exemplars) (Norman, 2000). Fur-thermore, experiences positively influenced how students viewed the widerpsychosocial aspects of medicine.

Experience could also disconfirm the understanding that the group wasconstructing during their discussion. This might be beneficial, but a concernwould be that false or misleading premises could be introduced during theclosing group stage on the basis of experience, which carries great weightbecause of the affective element described by the student. We believe thatthere is a key role for the tutor in ensuring that groups do not prematurelyclose discussion without fully exploring the implications of what the expe-rience described. Engagement of the clinical teachers in supporting the stu-dents learning through PBL is important, but Dornan et al. (2005) found that


clinicians had difficulty in adjusting to their role in a curriculum using PBL ina clinical environment.

In conclusion, we consider that using PBL in a clinical environment offerscurriculum planners a tool to encourage students to integrate experiencesgained in different settings and to mitigate against clinical experiences indifferent attachments being a series of unconnected events (Remmen et al.,1998). Future studies should examine how discussion of clinical experience inPBL groups can be promoted and whether different approaches to discussionmight affect how students acquire understanding of clinical problems as theyproceed through the programme and how this influences the integration oftheir learning and development of diagnostic competence (Schmidt et al.,1996).

Acknowledgements

We thank the tutors and students who participated in this study.

References

Barrows, H.S. (1986). A taxonomy of Problem-based learning methods. Medical Education 20: 481–486.

Bickman, L. & Rog, D.J., (eds.), (1998). Handbook of Applied Social Research Methods California: Sage

Publications.

Boshhuizen, H.P.A., Bromme, R. & Gruber, H., (eds.), (2004). Professional Learning: Gaps and Transi-

tions on the Way from Novice to Expert Netherlands: Kluwer Academic Publishers.

Boter, H., Van Delden, J.J.M., De Haan, RJ & Rinkel, G.J.E. (2003). Modified informed consent pro-

cedure: consent to postponed information. British Medical Journal 327: 284–285.

Dammers, J., Spencer, J. & Thomas, M. (2001). Using real patients in problem-based learning: students’

comments on the value of using real, as opposed to paper cases, in a problem-based learning module in

general practice. Medical Education 35: 27–34.

Denzin, N.K. & Lincoln, Y.S., (eds.), (2000). Handbook of Qualitative Research California: Sage Publi-

cations.

Diem, S.J., Lantos, J.D. & Tulsky, J.A. (1996). Cardiopulmonary resuscitation on television – miracles

and misinformation. New England Journal of Medicine 334: 1578–1582.

Dornan, T. & Bundy, C. (2004). What can experience add to early medical education? Consensus survey.

British Medical Journal 329: 834–7.

Dornan, T., Scherpbier, A., King, N. & Boshuizen, H. (2005). Clinical teachers and problem-based

learning: a phenomenological study. Medical Education 39: 163–170.

Foley, R.P., Polson, A.L. & Vance, J.M. (1997). Review of the literature on PBL in the clinical setting.

Teaching & Learning in Medicine 9: 4–9.

Greenhalgh, T. & Taylor, R (1997). How to read a paper: papers that go beyond numbers (qualitative

research). British Medical Journal 315: 740–743.

Jones, A., McArdle, P.J. & O’Neill, P.A. (2002). Perceptions of how well graduates are prepared for the

role of a pre-registration house officer: A comparison of the outcomes of a traditional and a Problem-

Based Learning Course at a UK Medical School. Medical Education 36: 16–25.

Jones, A., Willis, S. & O’Neill, P.A. (2003). A Qualitative Study of the Attitudes to Teamwork of

Graduates from a Traditional and an Integrated Undergraduate Medical Course. Advances in Health

Sciences Education 8: 139–148.


Kenny, N.P. & Beagan, B.L. (2004). The patient as text: a challenge for problem based learning. Medical

Education 38: 1071–1079.

Littlewood, S., Ypinazar, V., Margolis, S.A., Scherpbier, A., Spencer, J. & Dornan, T. (2005). Early

practical experience and the social responsiveness of clinical education: systematic review. British

Medical Journal 331: 387–391.

MacPherson, R., Jones, A., Whitehouse, C.R. & O’Neill, P.A. (2001). Small Group Learning in the Final

Year of a Medical Degree: A Quantitative and Qualitative Evaluation. Medical Teacher 23: 494–502.

Norman, G.R. (2000). The epistemology of clinical reasoning: perspectives from philosophy, psychology

and neuroscience. Academic Medicine 75: S127–S133.

O’Neill, A.P., Metcalfe, D. & David, T.J. (1999). The core content of the undergraduate curriculum in

Manchester. Medical Education 33: 121–129.

O’Neill, P.A., Baxter, C.M. & Morris, J. (2000). Evaluation of a new problem-based learning course in a

clinical environment. Medical Education 34: 222–230.

O’Neill, P.A., Willis, S. & Jones, A. (2002). A Model of How Students link Problem-Based Learning with

Clinical Experience Through Elaboration. Academic Medicine 77: 552–561.

O’Neill, P.A., Jones, A., Willis, S.C. & McArdle, P.J. (2003). Does a New Undergraduate Curriculum

Based on �Tomorrow’s Doctors’ Prepare House Officers Better for their First Post? A Qualitative study

of the Views of Pre-Registration House Officers Using Critical Incidents. Medical Education 37: 1100–

1108.

Patel, V.L., Groen, G.J. & Scott, H.M. (1988). Biomedical knowledge in explanations of clinical problems

by medical students. Medical Education 22: 398–406.

Patel, V.L., Groen, G.J. & Norman, G.R. (1991). Effects of conventional and problem-based curricula on

problem solving. Academic Medicine 66: 380–389.

Pope, C., Ziebland, S. & Mays, N. (2000). Analysing qualitative data. British Medical Journal 320: 114–

116.

Prince, K.J.A.H., Van de Wiel, M.W.J., Scherpbier, A.J.J.A., Van der Vleuten, C.P.M. & Boshuizen,

H.P.A. (2000). A qualitative analysis of the transition from theory to practice in undergraduate training

in a PBL-medical school. Advances in Health Science Education 5: 105–116.

Prince, K.J.A.H., Boshuizen, H.P.A., Van derVleuten, C.P.M. & Scherpbier, A.J.J.A (2005). Students

opinions about their preparation for clinical practice. Medical Education 39: 704–712.

Remmen, R., Denekens, J. & Scherpbier, A.J.J.A., et al. (1998). Evaluation of skills training during

clerkships using student focus groups. Medical Teacher 20: 428–432.

Schmidt, H.G., Machiels-Bongaerts, M., Hermans, H., Ten Cate, T.J., Venekamp, R. & Boshuizen,

H.P.A. (1996). The development of diagnostic competence: comparison of a problem-based, an inte-

grated, and a conventional medical curriculum. Academic Medicine 71: 658–664.

Schwartz, R.W., Donnelly, M.B., Nash, P.P., Johnson, S.B., Young, B. & Griffin, W.O. (1992). Problem-

Based Learning: an effective educational method for a surgical clerkship. Journal of Surgical Research

53: 326–330.

Silverstone, Z., Whitehouse, C.R., Willis, S., McArdle, P.J., Jones, A. & O’Neill, P.A. (2001). Students’

Conceptual Model of a Good Community Attachment (�A Good G.P’) in a Problem-Based Learning

Curriculum. Medical Education 35: 946–956.

Van Boxtel, C. (2000). Collaborative concept learning: collaborative learning tasks, student interaction,

and the learning of physics concepts. Enshede, Printpartners Ipskamp.

Visschers-Pleijers, A.J.S.F., Dolmans, D.H.J.M., Wolfhagen, I.H.A.P. & Van Der Vleuten, C.P.M. (2004).

Explorations of a method to analyze group interactions in problem-based learning. Medical Teacher 26:

471–478.

Visschers-Pleijers, A.J.S.F., Dolmans, D.H.J.M., De Lang, B.A., Wolfhagen, I.H.A.P. & Van der Vleuten,

C.P.M. (2005). Analysis of verbal interactions in tutorial groups: a process study.Medical Education 40:

129–137.


Using Clinical Experience in Discussion within Problem-Based Learning Groups

Documents

Transcript of Using Clinical Experience in Discussion within Problem-Based Learning Groups