gms | German Medical Science

Neurology is rated as one of the most difficult clinical fields not only by medical students, but also by physicians specialized in other areas, which leads to a high degree of subjective uncertainty in terms of neurological issues [1], [2], [3], [4], [5]. The significant reasons behind this phenomenon, for which Jozefowicz coined the term “neurophobia” [6] in 1994, were identified in surveys as being the absence of prior knowledge concerning neuroanatomy and the generally poor quality of teaching neurological content, with no distinction being made concerning specific teaching methods [2], [4].

Team-based learning (TBL) is a carefully structured teaching and learning process that was developed primarily for business schools by Dr. Larry K. Michaelsen in the 1970’s at the University of Oklahoma [7]. Briefly summarized, TBL is an active instructional and learning strategy in permanent small groups (= teams) which, as a learner-centered method, is chiefly based on the theory of constructivism [8], [9] and social learning [10]. It is organized into three phases according to a fixed sequence:

1.

Individual preparation outside of class;

2.

In-class testing of the individual’s readiness (individual readiness assurance test [iRAT]) and the team’s readiness (team readiness assurance test [tRAT]);

3.

In-class application exercises paired with group decision-making [11].

The major points are the assignment of responsibility in phase 1 to the student for the initial exposure to the content and the presence of only one instructor during phases 2 and 3 who is well-versed in the course content, but need not be an expert in group work. Through group work and immediate instructor feedback, the participants are able to acquire an understanding of the concepts and to deepen this understanding through the application-oriented assignments [11].

In recent years, there have been reports in many publications on the positive effects of TBL in respect to learning success [12], [13], [14], [15], [16], [17], increased motivation and readiness on the part of students [18], [19], improved problem-solving skills [18], [20], and better communication processes and team-based skills [21], [22].

Since only one instructor is required to lead a total of up to 120 students [23], there is a very advantageous cost-benefit ratio, which has contributed to the prevalence of TBL not only at North American medical schools, but also at other international medical schools. In contrast, up until now there has been only one published report originating from the German-speaking countries dealing with two TBL modules in the field of biochemistry/pharmacology at the Medical University of Vienna [24].

The aim of this pilot project was the implementation of TBL in the subject of neurology for promotion of active learning processes to improve learning success and understanding of neurological diseases. In particular, the feasibility from the instructor’s perspective and the acceptance by the students who experienced the TBL approach for the first time was to be investigated. Furthermore, a second step is intended to focus on analyzing possible effects on an increase in exam-relevant knowledge.

The description of the TBL method is done based on the proposed guidelines by Haidet et al. [25] for describing TBL activities.

General Context: The TBL neurology course was offered as a series of voluntary sessions consisting of six units within the scope of the regular neurology block practicum at the Neurologische Universitätsklinik Freiburg that usually takes place during the students’ third semester of clinical study. The 90-minute units took place on a weekly basis during the second half of the 2011-12 winter semester in the evenings (5:15 – 6:45 pm) in a seminar room at the neurology clinic. The students were personally informed about TBL during lectures and seminars, through the homepage of the university’s neurology clinic, and by email via the university’s learning platform, “Campusonline”, and were requested to sign up by email. The lecture topics and their corresponding learning objectives were selected as the topics for the individual units: headache, stroke, infections of the nervous system, polyneuropathies, critical care neurology, and basal ganglia diseases.

Team Formation: The instructor formed the teams through a randomly even distribution of both sexes into five groups (corresponding to 7 participants per group). The distribution process was not made transparent to the participants; only the permanent assignment for all units without the possibility of switching was pointed out.

Phase 1: Individual Preparation: For all units – in addition to attending the lecture – the particular lecture slides (20-30 PowerPoint slides on average) along with selected guidelines of the Deutsche Gesellschaft für Neurologie (DGN/German Neurological Society) were recommended as preparation. In some cases overviews from the Deutsche Ärzteblatt (total of 3-4 pages on average) were employed. All participants were made aware of the individual preparatory materials by email 3-5 days prior to each unit. The materials were accessible to all participants in the neurology course on the university’s electronic learning platform, Campusonline. The importance of preparation for the success of TBL was pointed out to the participants during the introduction and in each email.

Phase 2: Individual and Team Readiness Testing (iRAT and tRAT) / Immediate Feedback: All TBL units began with a paper-based, five-minute iRAT which consisted of three multiple-choice questions on clinical presentations, diagnostics, and therapy. The same questions were then each discussed for five minutes in the teams, which were then responsible for coming up with an answer and appointing a team spokesperson (tRAT). At the signal, the teams simultaneously held up their responses on colored sheets of printed paper. A discussion among the spokespersons moderated by instructor about the different team responses then followed during which the elimination of the other possible answers needed to be actively justified. The instructor gave immediate oral feedback during and after the discussion and ended the on average 10-minute process as needed with a short summary of the basic underlying concept.

Phase 3: Application Exercises: Process of Problem Solving / The “four S’s”: The application exercises were each comprised of a clinical case closely based on real ones routinely seen in practice (“significant problem”) with one to two related multiple-choice questions. The questions pertained to appropriate decisions regarding diagnostic and therapeutic steps and were based on the key points imparted during the RAT. One application exercise per unit was given directly to all of the teams to be worked on and discussed for ten minutes (“same problem”). Afterward, just as during the tRAT, all teams shared their group-based responses at the same time when given the signal (“specific choice” and “simultaneous reporting”). The selected and eliminated answers were discussed by the teams. Again, the instructor provided immediate oral feedback after closing the discussion and concluded the on average 10-15 minute-long process as appropriate with a brief summary of the basic underlying concept.

Incentives / Peer Evaluation by Participants: Since this involved voluntary participation in an instructional course, no grading or peer evaluations by the participants took place. The motivation for attending came from pointing out the improvements on multiple levels as described in the literature (see Introduction).

At the conclusion of the last TBL unit, all participants present were asked to fill out a written evaluation on the TBL approach in general and the TBL neurology course in particular using a Lickert scale (from 1 = strongly agree to 6 = strongly disagree). In addition, an overall assessment of the TBL neurology course in the form of a grade was requested. On the backside of the evaluation there was the option to respond to open-ended statements (“I found the following to be good…” and “I found the following to be less satisfactory…”). The evaluation survey used was based on Wiener et al. [24]].

The final exam for the block practicum consisted of 40 multiple-choice questions which reflected a mix of reproduction and transfer questions, as well as case-based problem solving. 15 questions referred to the topics addressed in the TBL units and were developed after conduction of the TBL sessions and not by the TBL instructor. The remaining 25 questions referred to complementary topics, such as multiple sclerosis, dementia, vertigo, muscle diseases, epilepsy, neuro-oncology and the clinical neurological examination. All questions were internally reviewed and revised twice by the instructors of the main lectures and the lecturers from the neurological clinic. Alongside the 35 students attending the TBL neurology course, another 132 students participated in the block practicum.

All evaluation and examination data were analyzed using Microsoft Excel and the statistics software PAST, which is available online free of charge [26]. The average values and standard deviations are given descriptively. The Mann-Whitney U test was applied to determine differences in the exam results between the two groups.

35 students of the 167 participants in the neurology block practicum registered by email for the TBL neurology course (21 female, 14 male; ages 21-29 years). None of these students had taken any TBL-based course prior to this. During the first session, a five minute-long introduction to the concept and process of TBL took place. The multiple-choice questions and the applied exercises were presented on slides. The tRAT phases and the handling of applied assignments in groups were shaped from the start by lively discussions within the teams. The instructor-moderated discussions among the teams about the reasons and justifications for the answers chosen and those not chosen were also lively and at the same time respectful in relation to the other students. All questions could be answered to the satisfaction of all the participants, whereby only in a few instances was final clarification by the instructor necessary, since the teams had usually dealt with the critical points during the joint discussion. All multiple-choice questions for the RAT, as well as the applied exercises, were each solved correctly by at least one team. Due to the voluntary nature and the heavy schedules in the third semester of clinical study, not all participants attended all sessions. On average, 28 students were present (corresponding to approximately 80%, maximum 35, minimum 19 at a session before Christmas). The time required to prepare a unit added up to an average of 8 hours (review, selection and provision of suitable preparatory readings from the literature: 3-4 hours; drafting three multiple-choice questions and one application exercise: 4-5 hours).

26 participants (13 female, 13 male) took part in the evaluation. The average age of those evaluating was 24.6 years (21 to 29 years of age); at the time of the evaluation the participants were on average beginning their ninth semester of specialized study (7^th to 13^th semester). The results of the written evaluation regarding the TBL approach in general and the TBL neurology course in particular are presented in Figures 1 [Fig. 1] and 2 [Fig. 2]. In addition, 24 open-ended responses were submitted: 18 completing the statement “I found the following to be good…” and 6 responding to “I found the following to be less satisfactory….” (For a summary, see Figure 3 [Fig. 3]).

The group of TBL participants achieved a significantly higher overall point total (33.7 vs. 32.5 points; p= 0.016) in comparison to the group which did not participate in the voluntary TBL course. For the pool of questions on the non-TBL topics, there is no significant difference between the two groups (20. 7 vs. 20.3; p= 0.358), while for the question pool dealing with the TBL topics, a significantly better result can be seen for the TBL participants (13.0 vs. 12.2 points; p= 0.006). Additionally, in contrast to the group of non-TBL participants (0.813 vs. 0.812 points per question; p=0.827), a significantly better test performance concerning TBL topics was displayed by the TBL participants compared to non-TBL topics (0.867 vs. 0.828 points per question; p= 0.045) (see Table 1 [Tab. 1]).

In the search for a teaching method that can optimally impart to students the subject of neurology, which is already perceived to be complex, and at the same time be simply realized without requiring extra personnel, TBL appeared at first glance to be the ideal solution. Due to the unclear reasons for the current near absence of this method in medical education in German-speaking regions, this pilot study was intended to supply data on the feasibility and acceptance of the method.

In regard to feasibility, two essential tasks resulted for the TBL instructor: preparation and conduction of the individual sessions. The selection of the preparatory readings was made with particular care: these were supposed to be as short as possible, while still allowing for structured processing of the particular topics. Along with the recommended attendance of the main lecture, the slides presented during the lectures could be reviewed on the learning platform. Fortunately, the Gesellschaft für Neurologie (DGN/German Neurological Society) has made many guidelines and their brief summaries available as pathways online [http://www.dgn.org/leitlinien.html], and these could also be used as part of the preparation. By uploading the material to the central learning platform, it was possible to establish a secure and reliable way of reaching all participants. A second main focus for the preparation was the drafting of questions and application exercises. Here, the estimation of a question’s degree of difficulty proved to be critical: multiple-choice questions were almost always correctly solved during tRAT; however despite this, led to critical discussions within and between the teams. Finding the ideal level as perceived by the instructor first occurred in full starting with the third unit, which also resulted in a positive assessment of the multiple-choice questions in connection with facilitation of the learning process on the final evaluation. The relatively long period of time, about eight hours, for preparing a unit appears at first to be a deterrent. However, this is only the case for the very first holding of a session, since not only the preparatory readings, but also all of the questions and cases that prove to be meaningful can be used again, should the module be repeated.

The realization of the TBL units was without complication from the perspective of the instructor, and in the student evaluations the role of the instructor was rated positively without exception, even though the instructor had no prior experience with group-based teaching methods. This corresponds to the conclusion in [11] that subject knowledge is required of TBL instructors, whereas expertise in group processes is not necessarily. Despite this, in addition to autodidactic study of basic TBL literature [7] and the published TBL applications [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], attendance of a TBL workshop (held by Prof. Wiener at the 2011 GMA conference) proved to be very helpful due to the lack of expertise on-site and made it possible to experience important TBL principles as a kind of self-test. Moreover, the art of moderating discussions, which is not usually a component of other teaching methods, was demonstrated. Training other TBL instructors on-site through live demonstrations with discussions and sharing appears thoroughly feasible now following the first TBL series.

At the time this pilot study was conducted, no group-based learning methods were offered as part of the Freiburg curriculum covering clinical study, except for the use of problem-oriented learning (POL) in two subjects (pharmacology and dermatology). Achieving sufficient recruitment of volunteers for TBL by following a good information strategy was therefore initially important for the pilot study. Through the combination of different communication channels (personal announcements, website, and detailed invitations via email), it was possible to acquire approximately 20% of the possible 167 candidates. Despite the lack of experience with TBL, there were no problems with implementing the individual steps after explaining the principles during the introduction – a result also of the clear structure of TBL. As can be seen in the evaluations, many of the TBL participants spend little time learning with their fellow students. This makes the high acceptance of the TBL approach among the participants and their desire for more TBL courses that much more surprising. This supports the basic demands for more active and team-based learning strategies which have been expressed specifically in connection with recent studies of the so-called Millennial Generation [27].

The possibility to impart concepts and principles through the use of questions and subsequent discussions is one of the strengths of TBL and very important especially for the subject of neurology. During the RAT, special value was placed on the communication of important principles (e.g. time factor in treating acute stroke, pathophysiology of intracranial pressure and clinical consequences, etc.), which then could be concretely applied in the application exercises. In addition, during the discussions it was possible to discover any problems or concepts that were not understood and to then address these as part of the feedback. As a result, a percentage of the participants were able to recognize and eliminate errors in reasoning (see Figure 2 [Fig. 2]).

Indication of specifically improved understanding of the topics handled is seen in the better exam performance of the TBL participants on the TBL-related questions. This is consistent with the enhanced performance on TBL-taught content by test-takers described in the literature [12], [13], [14], [15], [16], [17]. What is remarkable about the improvements seen in this pilot project is the level of statistical significance, for within the meaning of an intention-to-treat analysis with an average attendance rate of 80%, it also includes the analysis of a not insubstantial percentage of students not participating in the modules. However, the analysis of the effects of TBL participation on the exam results also has critical limitations: among all students taking the exam, the TBL participants were highly motivated students – ultimately demonstrated by their voluntary participation in TBL – with an interest in a group-based learning method. The transferability of these results to a group with varying motivational levels regarding participation in a group-based teaching method or to the subject of neurology itself is thus not taken as a given. Another limitation comes into existence through comparison of the results of one group with structured intervention to another without intervention (performance bias). Voluntary participation with an average absence rate of 20% and the lack of checks on the extent of preparation can be cited as mitigating this effect. In addition, the preparatory materials were freely accessible to all the participants in the neurology course and were downloaded by an average of 30 to 40 non-TBL participants. (Numbers of downloads are registered on the learning platform).

Since this pilot study was not designed to be a demonstration of effectiveness, the extent of the TBL effect, in particular on complex skills such as problem solving, in comparison to other established teaching methods should be analyzed by a controlled study.

Finally, an important point arising from the evaluation results should be addressed: a predominant majority of the TBL participants stated on the final evaluation that TBL had further increased their interest in the subject of neurology. Considering the increasing need for neurologists as a result of demographic change and the growing shortage of continuing education instructors in neurology, TBL can also be viewed as a small building block in the effort to combat the threatened lack of neurologists [http://www.dgn.org/images/stories/PM_DGN_Kampagne_Nerven_behalten.pdf].

In summary, the pilot study described here shows a good degree of feasibility for the TBL approach in the clinical subject neurology. The relatively intense amount of preparation at first, which then tapers off, is justified by the expected effectiveness concerning an improved understanding of neurology and the resulting higher level of interest in the subject. In general, there is a very advantageous cost-benefit ratio similar to that of a lecture. The acceptance on the part of a group of students without experience in team-oriented methods was very high, so that in the case of appropriate preparation, transferability to other pre-clinical and clinical subjects can be made without difficulty.

The effectiveness in regard to measurably better results in clinical reasoning and problem solving should be analyzed within the context of controlled studies. Due to the positive resonance, the TBL neurology course will continue for now on a voluntary basis in Freiburg.

J.B. wishes to thank the 12er Rat (Council of 12) of the University of Freiburg and the University of Freiburg for support from the innovation fund within the scope of the innovative studies competition, Projektwettbewerb: Innovatives Studium.

The author declare that he has no competing interests.