The principles of problem-based learning are more important than the method

Problem-based learning (PBL) has been part of Australian medical education since 1978, when the method formulated at Canada’s McMaster University in the late 1960s was adopted by the University of Newcastle.1 Our own school at the University of Melbourne adopted PBL in 1999. From its inception, PBL has usually been implemented as part of broader curriculum reform involving changes to the way the curriculum is conceptualised, structured and delivered. Accordingly, the introduction of PBL to established medical schools was frequently a highly charged social, pedagogical and political process. By the time newer Australian medical schools came to devise their curricula, however, PBL had become “mainstream”, a key part of the medical educational landscape and explicitly endorsed by the Australian Medical Council (AMC).2 The design and implementation of PBL has continued to evolve, to the point that debate now centres on the details of the approach, rather than its justification. It is therefore timely to reflect on how and why PBL has come to occupy such a central place in Australian medical (and other professional) education, and (re)consider its impact and value as a teaching and learning approach.

A particularly useful definition of PBL is “a way of constructing and teaching courses using problems as the stimulus and focus for student activity”.3 Thus PBL is not about simply adding problems to a traditional lecture-based course as a trigger for student learning; it also involves (re)constructing the medical course around these problems and methods. In order to enable students to effectively adopt the desired learning approach, courses need to be integrated (by system rather than by discipline) and designed to support the expected learning outcomes of each tutorial. The word “based” therefore carries significant weight in the PBL acronym.

There are many theories of adult learning, including behaviourist, cognitivist, humanist, social learning and constructivist perspectives, all of which draw heavily on more generic psychological approaches. Although early outlines of PBL were fundamentally pragmatic (Barrows and Tamblyn, for example, present relatively little underlying theory4), it has come to be informed by all of these theories, and most meaningfully by constructivism, which proposes that adults make meaning of their learning through an internal construction of reality.5 A practical extension of this theory was proposed by Knowles, who described the adult learner as self-directed, drawing on the knowledge and skills of others, learning from effective role models, problem-oriented, and motivated by factors such as self-esteem and recognition.6 In common to most theoretical perspectives is that adult learning should be active, self-directed, collaborative and contextualised.

At first, PBL encountered wide scepticism7-9 and, anecdotally, considerable resistance within some faculties. Many objected to the manner of its introduction and, not without some justification, the uncritical zeal with which it was implemented. For example, Wittert and Nelson argued that: “With far less evidence than the standards of evidence-based medicine demand, [PBL] was widely introduced, in a revolutionary manner, as the predominant mode of learning in medical programs”.10 From this perspective, the introduction of PBL was seen as being based more on enthusiastically adopted ideology than pedagogical evidence.11

It’s true that PBL was an expression of a fundamental shift in educational philosophy, based on sound principles but without evidence of enhanced outcomes. However, education is a social enterprise with an essentially qualitative and interpretive nature,12 so randomised controlled trials are generally impractical and methodologically inappropriate.13 Arguably, no greater evidence underlies traditional didactic approaches such as lecture-based curricula, however “self-evident” their superiority may be to their proponents. Faced with a perceived need for change, the medical education community could have done considerably worse than to develop a new method of teaching based on core educational principles and strategic alignment with desired outcomes, which PBL came to represent.

Notwithstanding the methodological minefield of evaluating the effectiveness of PBL, it is reasonable to expect to find some evidence of improved outcomes for such a major curricular upheaval. Hence, the more compelling arguments against PBL have challenged its empirical and evidential basis.14-15 Current research on the effectiveness of PBL has been well documented.16-18 Early research took advantage of the “natural experiment” in many American medical schools that were running two curricula (ie, “traditional” and PBL) in parallel. PBL appeared to be superior in terms of developing the so-called “soft” yet increasingly important skills of medical practice, such as communication, teamwork and the capacity to deal with uncertainty. Students who had come through a PBL curriculum generally felt more satisfied with their education. On the other hand, traditional course graduates seemed to have superior basic science knowledge, at least as evidenced by their certification exams.

The “knowledge issue” is all-important although clouded by simplistic claims. Although PBL students did gain slightly less knowledge, they also seemed to remember more of their acquired knowledge.18 Others found no significant difference in actual knowledge (assessed objectively) of PBL graduates, but a significant decrease in self-perceived knowledge when compared with graduates of traditional curricula.19

Much of the PBL literature is concerned with how best to prepare tutors (and students) for the effective implementation of PBL.20 Major recurring themes include approaches to tutor training and support, facilitation techniques and managing group dynamics, and the issue of tutors’ content expertise required for effective facilitation. A common conclusion is that any problems with PBL are more likely due to poor implementation than any inherent shortcoming. As PBL relies on active engagement, collaborative participation and facilitative rather than didactic teaching, its success requires considerable training, monitoring and oversight. The task of curriculum designers and course coordinators is to continuously monitor the course for such problems and intervene as necessary.

More crucial are the debates over the role and nature of self-directed learning in PBL.21 This is probably the most challenging innovation associated with PBL, and certainly the most likely to be misinterpreted. This remains a major source of error, confusion and angst for students, tutors, curriculum designers and medical professionals alike. Attempts to clarify the concept have produced only limited success. While some proponents of PBL made vague claims about its student-centredness, many critics misguidedly — and possibly wilfully — characterised PBL as students having to “teach themselves”.9 The reality is quite different. The nature of PBL learning is typically constrained by the group, by the tutor (explicitly or subtly), and by the prospect of summative assessment, usually defined by faculty-derived learning objectives.22 Candy points out that “there are important constraints on the extent to which people can or should strive to be self-directed, particularly in learning formal or technical bodies of knowledge”.23 Taylor and Miflin seem right to suggest that self-direction is more appropriately conceived as a goal or outcome of PBL, rather than one of its learning processes.24

Ultimately, there is a crucial difference between adopting PBL as a strictly prescribed teaching method, and embracing its essential philosophy by constructing a curriculum on this foundation. Some schools measure the success of their adoption of PBL by how faithfully the original method is adhered to; others draw inspiration and guidance from the underlying principles of the approach and accept that its implementation will vary between year levels or even groups within the same year. But where there is active, collaborative, contextual and facilitated learning of core content triggered by carefully designed problems for group study, there is PBL. In this view, acceptance of its spirit defines PBL more than absolute adherence to a particular prescription.25

This pragmatic view of PBL appears to be becoming dominant in (at least Western) medical education literature. Even Taylor and Miflin, in their definitive guide on the topic, admitted that a certain vagueness in the concept of PBL was inevitable, and felt that continuing reviews of existing definitions (and presumably further calls for homogeneity) were futile.24

Increasingly at medical education conferences, discussions about which implementation of PBL is most authentic or how many lectures are permissible in a hybrid lecture–PBL curriculum occur in parallel with other discussions on the “post-PBL era”. Such reflective evaluation and revision is necessary for any curriculum renewal, however perplexing this may be for late adopters.

Yet the need for a definitive template for implementation continues to manifest itself through more recent additions to the method landscape of small-group student-centred learning methods, such as TBL (team-based learning), SBL (scenario-based learning), IBL (inquiry-based learning) and POGIL (process-oriented guided inquiry learning). Whether such developments demonstrate vigorous innovation and exploration of pedagogy, or the aggrandisement of essentially minor variations, is open to debate.

In many schools, an alternative response has been to embrace the broader notion of CBL (case-based learning). For many, traditional PBL has, at best, served its purpose.26 Even the AMC’s 2009 version of its standards for accrediting medical schools moved from PBL to CBL.27 In this approach, cases are essentially valued for their contextualising, exemplifying, integrating and even motivational functions.28 The cases are normally not expected to carry or deliver the core curriculum content, or to provide a framework for self-directed learning. As a result, a case-based approach is less doctrinaire and more faculty-directed than traditional PBL. Significantly, even McMaster University has revised its program to be considerably more directed than its original version of PBL.29

At the University of Melbourne, a curriculum renewal process that began in 2009 as part of our shift to Masters- rather than Bachelor-level entry-to-practice professional degrees provided the opportunity to examine the role of PBL in the new curriculum, as well as to review the way PBL had been conceptualised and implemented. Our major concerns were that a somewhat rigid adherence to the method had lost sight of its intent; the inefficiency of having tutors (implicitly or otherwise) strongly discouraged from offering guidance to the group; and the tendency of students to prefer solving the “diagnostic challenge” rather than gaining a deeper understanding of the enabling biosciences. Meanwhile, neurocognitive evidence has emerged that students cannot learn deeply when their brains are consumed with problem solving.9 In reconceptualising PBL as “case-supported learning” (and thus regrettably contributing “CSL” to the acronymic pantheon), we have de-emphasised the clinical aspects of the case to allow closer engagement with its underlying bioscience; our tutors are framed as facilitators with the role of helping students “navigate” towards core knowledge; and general learning areas are communicated to the students ahead of each CSL cycle to allow them to focus their learning for the coming week. Thus, rather than basing a week’s unguided learning around diagnostic challenges in this first and only preclinical year of the Doctor of Medicine course, the cases explicitly support deeper learning of the biomedical sciences. But the fundamental goal of each CSL tutorial remains active, collaborative and facilitated learning, and it is the tutor’s role to enable this.

Traditional PBL in the clinical years continues to be viewed with some suspicion by clinical teachers, as it may supplant the didactic tutorial of the clinician’s choosing. In our experience, presenting clinical students with overly structured paper-based cases diminishes their focus on interviewing and examining real patients in real settings. Many Australian medical schools have moved to the CBL model for their clinical years, where students find people with specified conditions to present to their colleagues for thorough, self-directed analysis. While connecting students better with real patients, most of these methods contain the essence of Barrows and Tamblyn’s original model,4 based around students considering clinical problems in small groups with the clinical teacher as a guide rather than a guru.

Medical education that is both effective and efficient is too important to be hampered by quibbles over pedagogical purity. From our perspective, it is most useful to consider variations of PBL as reflecting a harmonious family of educational methods,5 all founded on an educational philosophy that values active, small-group, collaborative and contextualised learning over traditional, didactic, predominantly lecture-based approaches. While details will differ according to local circumstances and requirements, the underlying principles are shared.

So we would argue that PBL — which began as an expression of an ideology — has become a necessity. Despite the resistance and scepticism of the past 40 years, very few medical educators would now advocate a return to a fully didactic and lecture-based curriculum. The experience of PBL has convinced most that some form of active discovery within small groups improves the learning experience of medical students. In this sense, PBL has been a necessary philosophical shift, breaking the stranglehold that didactic, information-focused methods of teaching have traditionally had on professional education.