During my time at medical school, I enjoyed (at times) a curriculum delivered through the traditional model. As the name suggests, this is an approach experienced by the majority of doctors to date. The traditional model was first implemented by the American Medical College Association and American Academy of Medicine in 1894 (Barr, 2010) and has been used by the majority of medical schools. It traditionally consists of didactic lectures in the initial years covering the basic sciences followed by clinical years, where students learn clinical medicine while attending hospital placements.
A few years after my graduation I found myself teaching at a university which had fully adopted the use of problem based learning (PBL) in the delivery of their curriculum. PBL is a philosophy of teaching that has increasingly been used in medical education over the past 40 years. It has rapidly been replaced or supplemented in medical education as opposed to the traditional model. PBL seeks to promote a more integrated and active approach to learning right from the first year with less reliance on didactic lectures.
Having been involved in these two different approaches to medical education, I was interested to explore what the evidence was for and against each. For the purposes of this blog, I have looked at four specific areas. These include student attitudes, academic achievement, the academic process of learning and clinical functioning and skills.
Student attitudes to PBL have been highly featured in studies and many show that there is a clear favourability towards this philosophy of teaching. Blumberg and Eckenfel (1988) found that students in a problem based preclinical curriculum rated this three times higher than those in the a traditional group in terms of what they expect to experience, what they would like, and what they actually experienced. Heale et al (1988) found physicians in the problem-solving sessions rated a Continuing Medical Education short course higher compared to others who attended traditional lectures and large-group sessions. Vernon and Black (1993) performed a Meta analysis on 12 studies that looked at attitudes and towards PBL and found PBL was favored in some way by all studies. PBL appears to be preferred by the majority of students at a range of academic levels. However, Trappler (2006) found that converting a conventional curriculum to a problem based learning model for part of a psychopathology course did not show complete favourability. Students preferred the conventional lectures given by experts, rather than PBL groups run by mentors and not experts. They did however show preference towards PBL small group sessions run by experts
Academic achievement is an important factor to assess. Vernon and Blake (1993) compared a number of studies and found that those, which could be compared, showed a significant trend favouring traditional teaching methods. However, it was felt this might not be reliable. When looking at the heterogeneity of the studies there was significant variation that could not be accounted for by chance alone. Interestingly, they found that there was significant geographical variation across the United States such that New Mexico showed consistently negative effects and Michigan State showed consistently positive. Other studies have shown that the traditional method may show a slightly better outcome when assessing academic achievement. Schmidt et al (1987) looked at the same progress test taken among students in six different Universities in the Netherlands and found that those taught by a traditional approach showed slightly better outcomes. Baca et al (1990) compared performances of medical students in two separate tracks, one PBL the other a traditional model. Baca et al found that PBL students scored slightly lower in the National Board of Medical Examiners (NBME) examinations. Dochy et al (2003) conducted a meta analysis comparing 43 studies and found that when considering the effect of PBL on the knowledge of students the combined effect size is slightly negative.
It is important in medical education to enable people to continue life long learning, to overcome problems and fill in knowledge gaps. Coles (1990) and Entwistle (1983) found that PBL students would place more emphasis on understanding and meaning compared to just rote learning, seen more in those taught by a traditional approach. Students on a PBL course also place more focus on using resources such as the library and online sources rather than those taught in a traditional approach (Rankin, 1992). Students taught by a traditional model place more emphasis on the resources supplied by the faculty itself. It has also been shown that students who learn through a process of problem solving, are more likely to use this spontaneously to solve new problems in the future compared with those taught in a traditional way (Bransford et al, 1989).
Clinical competence is an important aspect in medical education and has been measured in studies comparing PBL and traditional methods. The traditional model focuses acquisition of clinical competence in the final years of a program with hospital placements. In a PBL course it may be more integrated early on. There are however, only a few studies that look at clinical competence gained in undergraduate PBL courses. Vernon and Blake (1993) compared some of these studies and found that students obtained better clinical functioning in a PBL setting compared to a traditional approach. This was statistically significant, however there was still significant heterogeneity amongst studies and for conclusive results to be made 110 studies would have to be compared, rather that the 16 samples they were able to use. They also found that in contrast to the NBME I giving better results in the traditional model, PBL students score slightly higher in NBME II and federation licensing examination which related more on clinical functioning than basic sciences.
On reflection, this evidence has indicated to me that PBL is a very valuable approach and it has a number of benefits. The traditional model in which I was taught has provided a good level of academic education. However, it may not have supported me as well as a PBL course in other areas of medical education such as academic process, clinical functioning and satisfaction. On reflection and current recommendations are for a hybridisation of the PBL and traditional approach to be used (Albanese, 2010) and I would support this view in light of the evidence.