Background

Excellence in medical education requires adaptation of the curriculum to meet the changing needs of students. In 1997, Problem based leaning (PBL) was introduced to the Sydney Medical Program (SMP), providing a long-established form of student-centred teaching within the medical curriculum. However, increasing student numbers (from 142 Year 1 students in 1997 to 332 in 2016) and limited teaching resources, rendered this model of teaching unsustainable. With a lack of standardisation across cohorts, extensive face-to-face PBL group meeting time, and limited individual accountability for contribution to group work, student satisfaction with the PBL model had decreased in recent years, and a suitable replacement was sought in Team-based learning (TBL) [1].

First introduced to medical education in 2001, TBL has gained popularity as a resource efficient and student-centred teaching pedagogy. Although TBL remains student centred, it is highly structured, with core design elements, and specific steps [2]. TBL involves multiple small ‘teams’ of five to seven students in one large class room of up to 100 students. Potential advantages of TBL include increased engagement of students in learning, deeper understanding of concepts, and a sense of responsibility towards teammates [1,2,3,4]. Additionally, when compared to PBL, TBL maintains the advantages of small group teaching and learning, without requiring such large numbers of teachers [2]. In a recent systematic review on the use of TBL within the health professions, Reimschiesele and colleagues reported that although learner reaction to TBL has been mixed, students generally favour the active learning style of TBL, with multiple opportunities for peer learning, problem solving and feedback [5]. However, there is a gap in literature reporting on the implementation of TBL across entire academic years of the curriculum, involving multiple, large student cohorts. Further to this, we believe that this is one of the first research papers reporting on the large-scale replacement of PBL with TBL across two student cohorts; and providing a direct comparison on the resource and cost implications for such change. Our study provides information and outcomes relevant for educators and curriculum designers working within health professional education.

Following an extensive pilot study in 2016 [1], TBL replaced PBL in the SMP in 2017. This study sought to explore student perceptions of TBL during its first year of iteration in Years 1 and 2 at a large medical school, and to consider resource implications for the change from PBL. Specifically, our research questions were:

What are students’ perception of their experience of TBL?

What are the resource implications when implementing TBL?

Methods

Sampling and participants

In 2017, all Year 1 and Year 2 medical students (n = 625) medical students participated in the TBL sessions as part of their required teaching activities. Students included Year 1 (n = 275) and Year 2 (n = 350).

Content of the TBL sessions

The teaching blocks and number of TBLs completed at the time of data collection (July 2017) are displayed in Table 1. Year 1 teaching blocks included: Orientation and Foundations (3 TBLs), Musculoskeletal sciences (4 TBLs), and Respiratory sciences (4 TBLs). Year 2 teaching blocks included: Neurosciences (6 TBLs), Endocrine, Nutrition, Sexual Health, HIV (6 TBLs), and Renal-Urology (4 TBLs).

Table 1 Number of TBLs within each teaching blocks at the time of the study
Full size table

TBL course design

The TBL classes were held approximately once per week across each teaching block on main university campus. Each TBL session was 2.5 h in duration. In Year 1, five TBL classes were run simultaneously. In Year 2, six TBL classes were run simultaneously. Each class consisted of 55 to 60 students, with 11 to 12 student teams per class. Students were allocated to teams to ensure an even distribution of age, gender, and science background. Teams consisted of either five or six students, and teams remained together for the entire year. Each class was facilitated by a team with expertise in the given discipline: one clinician, one basic scientist, and one medical registrar. All student TBL activities (pre-class and class) are summarised in Table 2. We followed Michaelsen and Sweet’s recommendation of three distinct phases to establish the TBL process [6].

Table 2 TBL pre-class and class activity schedule
Full size table

1) The ‘preparatory phase’, where students were allocated compulsory readings or/and pre-recorded lectures prior to class. Additionally, each TBL was aligned with the curriculum course content, so that lectures, laboratory and clinical based teaching were provided prior to the TBL.

2) The ‘readiness assurance phase’, where the Individual Readiness Assurance Test (IRAT) was implemented as an online pre-class activity, delivered via Blackboard Learn (Learning Management System). The Team Readiness Assurance Test (TRAT) was delivered in class, followed by immediate feedback and clarification by the facilitators.

3) The ‘application phase’, where clinical problem-solving activities were undertaken by students in their teams.

Data collection and analysis

Questionnaire

A questionnaire was distributed to all students following completion of the TBLs listed in Table 1. The questionnaire was anonymous, and distributed by administrative staff, along with a participant information statement (which also explained who was involved in the research), following the last TBL class. The questionnaire included 14 closed items, using a five point Likert-scale, ranging from 1 ‘strongly disagree’, to 5 ‘strongly agree’; and open-ended questions. The questionnaire was adapted from a previously validated questionnaire designed by Thompson and colleagues (2009) [7]. The questions used considered the team processes and dynamics of TBL, for example, “Team members encouraged one another to express their opinions”. Additionally, the questions considered the outcomes of the TBL design, for example, “Completion of the prescribed out-of-class preparation assisted in my learning”. Quantitative data were analysed using descriptive statistics. Thematic analysis was used to code and categorise qualitative data into themes, building an understanding of students’ expereince throughout the year. A portion of the data was read by the first author and analysed to identify themes. Following negotation of meaning with the third and fourth authors, a coding framework was developed and applied to the full data set [8].

Ethics approval

Ethics approval was gained from the University of Sydney Human Research Ethics Committee. Approval project number: 2016/136.

Results

In total, 490/625 (78%) of students responded to the questionnaire, including: 232/275 (84%) Year 1 students and 258/350 (74%) Year 2 students. Of the Year 1 respondents, 119 (51%) were male; 98 (42%) were female; and 15 students (6%) did not respond to this question. Of the Year 2 respondents, 122 (49%) were male; 121 (45%) were female; and 15 (6%) did not respond to this question.

Student responses to closed items are provided in Fig. 1 (Year 1) and Fig. 2 (Year 2). The distribution of responses across the 5 point Likert scale was similar between the first and second year students. Responses to items 1–5 indicate that the small group dynamics worked well, with most students making an effort to participate, listen, and encourage others in group discussion. Responses to item 7 suggest that assigned pre-readings were not always completed, with only 57% of Year 1 respondents, and 46% of Year 2 respondents agreeing or strongly agreeing with the statement (item 7), “Students did the pre-readings prior to class”. Notably, only 17% of Year 1 students, and 27% of Year 2 students agreed or strongly agreed with the statement (item 13), “Competitiveness between groups enhanced my learning”, since team scores were not announced.

Fig. 1
figure 1

Year 1 student responses to closed items regarding their experience of TBL

(N = 232)

Full size image
Fig. 2
figure 2

Year 2 student responses to closed items regarding their experience of TBL (N = 258)

Full size image

Students’ responses to open ended questions are illustrated in Tables 3 and 4, categorised as ‘best features of TBL’ and ‘suggestions for improvement/most difficult features of TBL’. Students found the positive aspects of TBL to be the interactive discussion with facilitators who were experts in their field; provision of a clinical context by clinicians; tutors circulating the room to answer questions; the tests (IRAT and TRAT), with the explanations from tutors that followed; peer discussion in small groups (5–6 students); and the interactions with teams in one large room. Suggested improvements included better alignment of pre-reading tasks with the TBL case, increased opportunity for clinical reasoning, and greater feedback on the mechanistic flow charts. Students also suggested a shorter duration of classes (2 hours) and an earlier release of the IRAT to be completed prior to class.

Table 3 Students’ perceptions of Team-based learning regarding the ‘best features’
Full size table
Table 4 Students’ perceptions of Team-based learning regarding the ‘suggested improvements/most difficult features’
Full size table

Discussion

We sought to explore students’ perceptions of their experience in TBL during the first complete implementation of TBL in Years 1 and 2 of the medical curriculum. Results indicate that generally students found their learning experience in TBL to be positive. Students found the positive aspects to be their interactive discussion with experts as facilitators, the presence of clinicians, the tests (IRAT and TRAT) with the explanations from facilitators that followed, peer discussion and problem-solving in small groups, and the inter-team discussion. Suggested improvements included better alignment of pre-reading tasks with the TBL case, increased opportunity for clinical reasoning and interactions with the facilitators; an earlier release of the IRAT; and shorter duration of in-class time. We applied the three phases of TBL: 1) The ‘preparatory phase’, 2) The ‘readiness assurance phase’, and 3) The ‘application phase’, as a framework to discuss our findings. In addition, we consider the resource implications for TBL facilitation.

The ‘preparatory phase’

Designated preparation for essential knowledge acquisition for TBL was designed to shift the burden of learning content during class. However, students’ compliance with completion of assigned pre-reading and preparation was poorer than expected. Only 57% of Year 1 respondents, and 46% of Year 2 respondents agreed or strongly agreed with the statement (item 7) “Students did the pre-readings prior to class”. This is a reduction from the 68% of students who were in agreement with this statement in the 2016 pilot study [1]. Failure to prepare impacts on the teams’ learning and performance, and is an important area to address [9]. The recommended practice in TBL is to announce the TRAT scores to the class [10], however, this step was not included in the 2017 TBL implementation. Perhaps the lack of student preparation indicates that the announcement of team scores may have helped motivate students to complete pre-readings and attend class better prepared to engage and contribute to their teams. Additionally, qualitative feedback from students indicated that there was a lack of alignment of the test with the pre-reading material. It is recommended that preparation material be identified after the multiple choice questions have been written to ensure relevance of the material, and completion of assigned preparation [2].

The ‘readiness assurance phase’

The tests (IRAT and TRAT) were well received by students, with the majority of students (71% of Year 1 students, and 80% of Year 2 students) agreeing or strongly agreeing that the “individual and team tests assisted in my learning”. In order to reduce in-class time, the IRAT was held prior to class, made available to students 1 day prior. However, a number of students requested that the IRAT be released earlier. Literature suggests that taking the IRAT online prior to class adds value to learning [11]. Additionally, there is evidence to suggest that repeated testing of students encourages retrieval of new knowledge, and helps in recall of knowledge at a later date [12]. Opportunities for critical reflection are needed to allow students to make judgements on required modification to their existing knowledge [13]. Our results suggest that reflection occurred when students compared their understanding to that of their team members during the TRAT, through discussion to agree on an answer. Notably, 94% of Year 1 students, and 95% of Year 2 students strongly agreed or agreed that “all team members made an effort to participate in discussion”, suggesting that the small group size enhanced team dynamics. Unlike the TBL format in 2016, the TRAT results were not released to the class. Therefore, it is not surprising that only 17% of Year 1 students, and 27% of Year 2 students agreed or strongly agreed with the statement (item 13), “Competitiveness between groups enhanced my learning”, compared to 69% (agreeing or strongly agreeing) in the pilot study [1]. However, current TBL literature and research suggests that friendly competition is an important component to the TBL process [1,2,3,4], indicating that this is an element of the TBL implementation requiring further consideration.

The ‘application phase’

The quality of clinical problem-solving activities plays a key role in engaging students in small group collaboration [14]. Importantly, students generally felt satisfied with the problem-solving activities within TBL, with 81% of Year 1 students, and 83% of Year 2 students strongly agreeing or agreeing that “problem solving allowed me to develop my clinical reasoning skills”. Students appreciated the expertise of facilitators, and the clinical context provided by consultants and registrars. The importance of providing clinical relevance to medical teaching is frequently highlighted in medical education literature [15]. However, students indicated a need for increased direction and discussion regarding the pathophysiology flowchart. Literature suggests that immediate feedback may assist student groups to work more independently throughout the problem-solving activities [16, 17]. Further to this, only 66% of Year 1 students and 71% of Year 2 students agreed or strongly agreed with the statement (item 10), “I received useful and timely feedback from the tutor”. In our 2016 pilot study of TBL, 80% of students had agreed or strongly agreed with this statement [1], which may reflect the extensive teaching experience of the 2016 facilitators, who were all senior consultants.

Resource implications

Although the key reasons for implementing TBL were to provide greater standardisation in teaching and an improved educational experience to students, it is also important to consider the resource efficiencies of TBL compared with PBL. In 2017, there were 625 Years 1 and 2 medical students enrolled in the medical program. Using the TBL format, with approximately 60 students per class, we implemented 11 TBL classes per week, requiring a total of 33 facilitators (3 facilitators per TBL class). The previous PBL format, with 10 students per PBL, required 62 facilitators (one facilitator per PBL group) each week. By implementing TBL, we have reduced our facilitator requirements by approximately half, and at the same time provided clinical expertise in each classroom. This has obvious cost saving implications. It should be noted, however, that the preparation required for each TBL module was time consuming, and required academic expertise. Tasks include preparation of multiple choice questions and answers, development of patient cases and clinical problem-solving activities, and Powerpoint slide presentations to guide feedback and clarification of concepts.

Study limitations

Data collection occurred after completion of three blocks of Year 1 teaching, and three blocks of Year 2 teaching. It is possible that students’ perceptions at that point in time differed from their perceptions at the end of the academic year. The findings from our study may not be generalisable to other universities. A limitation is that although the questions within our questionnaire were based on a validated questionnaire designed to measure the quality of team processes [7], our questionnaire itself was not validated prior to use.

Conclusion

Our findings demonstrate that TBL, as a replacement for PBL in Years 1 and 2 of the medical curriculum, provided a standardised approach to small group learning on a large scale, and also provided resource efficiencies. Interactions and discussion with experts as facilitators, the presence of clinicians in every class; the repeated testing (IRAT and TRAT) with feedback and explanation; small group peer discussion and problem-solving were all aspects of TBL experience that the students found positive. However, improvements are needed in terms of better alignment of pre-reading tasks with the IRAT, and greater facilitator interaction during the problem-solving activities. Additionally, consideration should be given to reducing class time, an earlier release of the IRAT, and provision of TRAT scores to the class.