Educational research: current trends, evidence base and unanswered questions

Medical education is a rapidly developing field requiring specific research approaches to provide evidence that can effectively translate into clinical practice

Medical education is a relatively young scientific field, with its origins in the mid-1960s when papers specific to medical education were first published. Yet, by necessity, it has evolved rapidly in the past 50 years, not least because of coincident dramatic changes in the fields of education, cognitive psychology and medicine itself.

In medicine, the most obvious advances have been in the therapeutic and diagnostic armamentarium available to the clinician, but perhaps even more important have been the changes in the patient population. Increased life expectancy and ageing, associated with increases in multimorbidity and chronic disease, have had a dramatic effect. Additionally, an increased expectation of patient-centred care and patients’ own increased access to relevant medical information over the internet play a role. The latter has led to a rapid decrease in the pre-existing expertise asymmetry between the health care provider and the patient. A different type of doctor is now required.

A similar change has occurred in education. Even in the time when problem-based learning and other student-centred learning approaches were being developed, it was still the medical schools that decided which books and other resources would be made available in the library and which lectures and seminars the student had access to. Students now have a whole internet of accessible resources, presentations and lectures they can access, and much more is known about effective and efficient human learning and assessment. This requires a radical rethinking of the approach of medical schools to better adapt and educate modern students.

It is logical that medical education had to mature quickly, and in this article we outline some important current trends in medical education research of the past decades and consider how these may evolve in the future.

Redefinition of competence

Medical competence was traditionally defined as a combination of knowledge, skills and attitudes. The assumption that each of these elements could be taught and assessed separately meant that a student was deemed competent if he or she was sufficient in each of these three elements independently. The literature in medical education in the 1960s, 1970s and well into the 1980s was replete with studies aiming to establish the superiority of one method over another for teaching or assessing each of these elements; for example, trying to demonstrate that open-ended questions are better than multiple-choice questions1,2 or, in the educational context, that problem-based learning was better than lectures.3,4 Gradually, it became clear that competence cannot be taught and assessed in such a reductionist way, with recognition that the whole is more than the sum of its constituent parts.5 This has led to the notion of competencies,6,7 definitions of which are multiple. However, all agree that competencies are professional tasks that a competent student/graduate should be able to manage successfully, using at the right time the right knowledge, skills and attitudes in conjunction with metacognitive abilities such as reflection and self-regulation. This new model arose from a fundamental rethink, moving from the reductionist approach focusing on the single elements of competence and then trying to add them together to make “competence”, to a new model aimed at preserving a necessary level of integration. To illustrate this shift of thinking, the old model was much like a car factory and the new model is much like health care. In the former, the parts are simply put together to make a car; in the latter, a patient’s health is much more than just an addition of laboratory values. However, although the competencies approach seemed good in theory, it was not easy to convert it into practice.

A parallel recognition in medical education research has been that we should refer to expert performance as opposed to expertise — that the ability to demonstrate competence depends on the situation (expert performance or state) as opposed to being inherent to the individual physician (trait).8,9 This progresses the idea of competence as being more than the sum of the individual parts to an understanding of why clinical performance goes well or poorly in a given situation. Clinical reasoning and medical error research have stressed the need to think of competence as being situational and dependent on the interaction of various parts in the situation.10,11 Current medical education research draws strongly on these notions, upon which we now elaborate.

The role of theory

The nature of publications in medical education journals has changed. In the early days, it was common to see publications describing an existing educational or assessment approach at a particular medical school or an idea for a new approach or an evaluation study describing the opinions of students about their course.12 Increasingly, such publications were viewed as largely unhelpful by the major medical education journals, principally because an educational method that worked well in one context would simply not work in any other given context, especially in different countries with different cultures. This realisation was often counterintuitive to medical professionals involved in medical education who were familiar with research outcomes being replicable regardless of context. For example, if statins have a positive effect on cholesterol in a Finnish population, it can be assumed that the same would hold for an Australian population. Education, however, is much more contextual, as there are myriad factors that influence the success of an educational approach. Therefore, research shifted focus to understanding why certain approaches work and others do not. For example, in problem-based learning, specific metrics — such as two tutorial sessions per week or the number of students in a tutorial group — are in themselves unlikely to result in an effective outcome. More fundamental is the fact that problem-based learning approaches stimulate more meaningful learning, more active learning and more interactive or collaborative learning, each of which stimulates better retention of and access to the learned subject matter.13 Research that explains why certain approaches are successful can therefore be used in and adapted to various contexts. It follows that medical education research focuses less on producing replicable results than on adaptable results.14

It is clear that such research cannot take place without a strong theoretical foundation. Theoretical foundations in research are important for the formulation of the correct research questions, the analysis of results and the validity of the conclusions drawn. In the complex world of medical education, theory serves as a lens to view the results, predict why success was or was not achieved and plan the next steps.14,15 To give a sports analogy, theory provides expectations regarding the “game” — the rules and boundaries of play — so that the interactions between the players make sense. Without this, one will not be able to fully appreciate what is happening or why. An additional illustration can be found in medical research. Until the 1980s, peptic ulcer was seen as a disorder resulting from the overproduction of acid and so the dominant theoretical lens that underpinned related research and management was a disorder of gastric pH. Only after Marshall demonstrated the important role that Helicobacter pylori (Hp) plays in the pathogenesis of peptic ulcer16 was the theoretical lens shifted to an infectious one (“from pH to Hp”). Whereas old research into peptic ulcer focused on acid production and what leads to excessive acid (eg, stress and non-steroidal anti-inflammatory drugs), current research now considers testing for and treating H. pylori with antibiotics.

Medical education has therefore become rich in explanatory theories such as cognitivist17 and cognitive load theory,18,19 transformative education theory,20 self-regulated learning,21 situated cognition,10 complexity and systems theory,22 and social agency.23

Inclusion of more diverse methodologies

It was logical, then, that more diverse methodologies were required, as there were many research questions that simply could not be answered with traditional causal comparative designs (of which the randomised controlled trial is but one example). Even more research questions could not be answered with quantitative research. In the early 1990s, this led to a wide uptake of qualitative research methodologies in medical education research.24 Focus groups, structured interviews, critical incident techniques and similar methodologies became popular approaches to achieve a better understanding of why certain education methods or assessment approaches were so successful. Our impression is that, currently, more qualitative than quantitative research seems to be published in medical education, but the pendulum may well swing back and forth.

Inclusion of more diverse epistemologies

It became clear that certain phenomena in medical education could not be studied well from a purely logical positivist perspective, and gradually more publications began employing constructivist epistemologies.

Logical positivist perspectives are helpful when the object of study is predictable and tangible and exists regardless of whether it is perceived; gravity, for example, exists even if there is nobody there to observe it. Other phenomena, however, need to be constructed or else they do not exist. Communication is such a phenomenon. Its physical substrate is simply soundwaves or moving air, which dissipate as they hit the tympanic membrane. It is well known that the way this meaning is constructed is largely dependent on previous memories and experiences and cultural upbringing.25 In order to study communication, more inclusive perspectives, such as constructivist ones, will therefore be needed.26 It follows that medical education literature contains both quantitative and qualitative studies employing either logical positivist or constructivist epistemologies, making the field very broad and unfortunately also increasingly difficult to access for a more general readership.

A final important development is the inclusion of the notion of complexity, which is associated with a certain level of unpredictability. Much as a clinician cannot predict exactly what he or she is going to say, for example, 2 minutes and 30 seconds into a consultation, a teacher cannot approach education in an algorithmic or precisely predictable way. Yet, both the expert clinician and the expert teacher are able to manage unpredictable situations well. Just as there is no single approach which suits all individual patients, there is no single approach which suits all educational situations.11,27 Thus, we believe that research that seeks to find the single best educational approach will generally be inconclusive. In response, the medical educational research agenda has focused on understanding why methods work, what new educational or assessment approaches bring to the table and, more recently, how to programmatically align the different methods to build an optimal curriculum or assessment program.

In this article, we have pointed out a number of research gaps. We also see the need to connect more strongly between education and clinical practice (a medical education equivalent to the Framingham Study, using new methodologies, epistemologies and theories), medical errors and the study of expert performance in older physicians.

Conclusion

Medical education is quickly becoming a well established scientific field in its own right, with its own theoretical frameworks, its own vocabulary and its own rich array of international journals. That in itself is a positive development; the international community is growing rapidly — especially with the “awakening” of Asia. It comes with a risk, however, that medical education terms and research become inaccessible to the everyday clinician and teacher. No research can be useful if it does not find its way to actual practice. Of this the medical education community is fully aware, and certain measures have been put in place. The first is the establishment of journals with a translational remit (such as The Clinical Teacher), in which experienced medical education researchers describe important findings in a more generally accessible language. The second is a call to attention for the broker role of medical educationalists in universities and teaching hospitals in many countries around the world, aiming to instil these modern paradigms into teaching staff and, through this, convey the outcomes of medical education research. The third and perhaps most important development is initiatives like this from the MJA to provide a forum for medical educators in which they can explain the important developments in their field with relevance to the practice of medicine and medical education.