Domains of effective teaching process students perspectives in two medical schools

Abstract

Background: There has been little systematic investigation of student belief about the characteristics of the optimum process for clinical bedside teaching.

Aims: The intent was to identify the most important characteristics of the bedside teaching experience from the perspective of two groups of students, one from Oman and the other from Canada.

Method: Students were asked to complete a questionnaire about their concept of the ideal process of bedside teaching. Their answers were analyzed using factor analysis.

Results: Answers provided by the students was consistent with the presence of six domains. These corresponded to Preparation, Introduction, Experience, Summary, Explanation, and Conclusion. ‘Preparation’ involves consideration of the patient and the knowledge level of the learners, ‘Introduction’ involves effective communication, and ‘Experience’ means the need for the students to be actively involved in the history and physical examination. The remaining three domains deal with the need to provide a summary and elaboration as well as advice and feedback. These 6 factors explained 60% of the total variance.

Conclusions: While these areas still need to be defined more closely, the application of these six domains to the structure of the bedside teaching experience is likely to result in improved student learning.

Introduction

Teaching in the presence of the patient is perhaps the most significant method used to learn clinical medicine. This method which is usually referred to as ‘bedside teaching’is a fluid entity by itself, occurring as a learning experience whenever students, teachers and patients interact (Gale & Gale 2006) and so can take many forms (Hartley et al. 2003; Beckman 2004). The definition that suites our research is the one used by Nair et al. (1997) which basically states that bedside teaching (BST) occurs ‘when a clinician takes a group of learners to the bedside of a patient, listens to the history, elicits physical signs, makes a provisional diagnosis and decides on the best diagnostic and therapeutic options’. This rich and complex experience has been the focus of many articles based on personal experience or opinion of experienced clinicians and educators (Kroenke & Omori 1997), and some experimental studies have attempted to define it in more objective terms (Ramani 2003). Such studies have looked at a variety of components such as feedback or instruction in clinical procedures, and furthermore, attempts have been made to provide models that describe the cognitive processes and structuring that lead to learning and development of expertise (Ramani et al. 2003; Janicick & Fletcher 2003; Cox 1993). There have been rather few studies, however, that have explored the specific process of bedside teaching from the perspective of the learner. Nair et al. (1997) reported that students uniformly felt that bedside teaching was the best way to learn clinical medicine, and about half of his student sample felt that there was insufficient exposure to this approach. Much more recently, Williams et al. (2008) used a qualitative approach and reached similar conclusions, but were able also to identify some distinct barriers to wider implementation of bedside teaching and to suggest some strategies for overcoming these problems. We have previously described the characteristics of students’ perceptions of the ideal bedside teacher, based on information from one medical school (AlWeshahi et al. 2007). In this paper we discuss the characteristics of the ideal bedside teaching process, using data from two different medical schools. A clear definition of the student perceptions of what the bedside teaching process should entail may lead to a more effective utilization of this process as a powerful tool for learning clinical medicine.

Other research in the area of clinical teaching and teaching in general has enabled us to identify many of the issues that might be germane to the process of effective bedside teaching. Some of the more important issues in this regard are shown in Table 1. This list is not exhaustive, but it covers the major points identified in the literature of medical education, and mutatis mutandis is consistent with the education literature in general.

Table

Practice points

This study identified the most important characteristics of the bedside teaching experience based on student views of the process. Answers by the students were consistent with the presence of six domains: Preparation, Introduction, Experience, Summary, Explanation, and Conclusion. The application of these six domains to the structure of the bedside teaching experience is likely to result in improved student learning.

Table 1.  Important factors in the process of effective clinical teaching

Interaction/communication

• Attention to the comfort, privacy and dignity of the patient  including consenting (Howe & Anderson 2003)

• Proper introduction of learners/patient (Schwenk & Whitman 1987)

• Clear communication (Van de Wiel & Boshuizen 1999)

Appropriate pedagogy

• A student-centred approach (Wolpaw et al. 2003)

• Creation of an environment conducive to learning (Hutchinson 2003)

• Articulation of clear learning objectives (Miffin & Price 1997)

• Appropriate organization of the learning experience (Harris 1998)

• The linking of new learning to old knowledge (Schmidt et al. 1990)

• Encouragement of critical thinking (Maudsley & Strivens 2000)

• Active engagement of the students

• Knowing the learner (Vaughn & Baker 2001)

Specific objectives

• Instruction in history-taking

• Instruction in procedures for physical examination

• Discussion of management plans

• Evidence-based approach (Gruppen et al. 2005)

• Holistic approach (Mauksch 2005)

• Compare and contrast cases (Klayman & Brown 1993)

Coaching

• Appropriate discussion

• Immediate, constructive and concise feedback  (Kilminster & Jolly 2000)

• Illustration of appropriate professional behaviour and social  modeling (Wright et al. 1998)

• Direct observation of learner's skills (Holmboe 2004)

Conclusion

• Opportunity for reflection (Branch 2002)

• Appropriate summary of the learning (Cox 1993)

• Discussion of the practical utility of what has been  learned (Torre et al. 2003)

There are several potential sources of information about the process of bedside teaching, but the opinion of medical students has proved to be useful and representative; this was the source of information for our previous conclusions about the characteristics of the ideal bedside teacher. In the present work, where we have looked at the process of bedside teaching rather than the behaviour of the instructor, we were interested in extending our sample to include a North American medical school. We thus report here information from students from two different countries.

Methods

Context of study

The study was conducted at the College of Medicine and Health Sciences of Sultan Qaboos University, Oman, and at the Faculty of Medicine and Dentistry of University of Alberta, Canada. The former has a three year clinical curriculum as part of a seven year program. Students are recruited directly from high school; the pre-clinical curriculum uses a variety of different teaching strategies and the clinical teaching involves clerkships that are similar to those used in many other countries. The University of Alberta has a two-year preclinical curriculum, which uses both lectures and problem-based learning, followed by a series of clerkships that take two more years. In other words it is a typical North American program, which is mostly graduate entry. In both schools, clinical teaching takes place in in-patient, ambulatory and rural sites. The principal teaching method in Oman is bedside teaching which is conducted in a protected time, usually for an hour, with a group of students led by a clinician. The clinical service for the patient is conducted at other times, so here the bedside process is entirely focused on teaching and learning. At the University of Alberta, bedside teaching is usually conducted in the context of providing clinical service to the patient.

Instrument

A 47-item questionnaire was developed based on the issues identified by the literature review. Each item was designed to measure one or more of those identified issues. A theoretical organizational model for bedside teaching suggested by Cox (1993) was used to organize the items in the instrument. The items were constructed specifically for this questionnaire, since no appropriate instrument was already available. The only demographic variable collected was the gender of the individual completing the questionnaire. With the exception of Item 27 which was concerned with the duration of the session and Item 46 which asked the student to reflect on the appropriateness of the questionnaire to elicit information about bedside teaching, the remaining 44 items were all concerned with student perceptions of different domains of the process of bedside teaching itself. Students were asked to respond using a five-point Likert scale (5 = strongly agree, 4 = agree, 3 = neither agree nor disagree, 2 = disagree, 1 = strongly disagree). In order to provide validation for the instrument, the questionnaire was reviewed independently by three established international experts of whom two suggested minor changes in the phrasing. The questionnaire was piloted on a group of 10 students who were encouraged to make open-ended comments on the questions. No substantive suggestions were made by the students and thus no further modifications were made to the questionnaire.

The students received letters from the researchers explaining all aspects of the study, that they would be permitted to see the results, and explicitly guaranteeing personal anonymity. The project was approved by the Ethics Committee at the College of Medicine and Health Sciences of Sultan Qaboos University and the Health Research Ethics Board of the University of Alberta.

Subjects

In Oman, one hundred and seventy four final year medical students in the academic years 2003/2004 and 2004/2005 received the questionnaire. The questionnaires were delivered by the student representative of the respective academic year, who volunteered to distribute and collect the completed questionnaires. In Canada, eighty seven final year medical students in the academic year 2006/2007 received the questionnaire which was distributed and collected by the investigators. Thus, both of groups of students were in their final medical school year and were at about the same stage of learning.

Statistical analysis

The scores assigned to the items on the five-point Likert scale were recorded and the mean value and standard error was determined. The combined scores from both groups were used in this study, since we were interested in the common features rather than the differences.

We used the subjects-to-variables ratio (SVT) rule suggested by Bryant & Yarnold (1995) to decide on the adequacy of the number of cases for principal component analysis. Barlett's sphericity test was used to determine whether there was a correlation between items. The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy was also employed. These two tests were used to determine whether factor analysis was an appropriate tool.

A ‘little Jiffy’ (Kaiser 1970) factor analysis was performed to identify any significant factors, which were selected on the basis of Kaiser–Guttman criteria and Cattell's scree plot.

Results

One hundred and seventy four students in Oman and eighty seven students in Canada were contacted; the response rate was 75% and 97% respectively making the total number of respondents 214 students from both schools of whom 48.5% were male and 51.5% were female. Cronbach's α was 0.89 for all items.

Barlett's sphericity test was highly significant (0.00), which indicates that there is a correlation between items. The Kaiser-Meyer-Olkin measure of sampling adequacy (KMO) was 0.805 which suggests that the degree of common variance among the variables is meritorious. The results of these 2 tests justified the use of factor analysis.

The number of factors that were accepted was based on 3 criteria; Cattell's scree plot, an eigenvalue of more than 1 and whether items could be attributed to a theoretical cluster. Twenty one of the 44 items were excluded either because they were not discriminating or they were ambiguous. Accordingly, the subjects-to-variables ratio (SVT) was 5.42. We used principal Component Analysis with Varimax as a rotation method on 24 items and that revealed 6 factors, which correspond to the hypothetical domains of Preparation, Introduction, Experience, Summary, Explanation, and Conclusion. Figures in bold type represent items that loaded into each of the six factors. Those factors explained 60% of the total variance. Cronbach's α for each of the six factors was 0.58, 0.66, 0.65, 0.73, 0.75 and 0.78 respectively (Table 2). In addition the mean score provided by the students and standard error for each item are also shown in Table 2.

Table 2.  Items responses and factor analysis

	Components
Items	Mean ± SE	1-Preparation	2-Introduction	3-Experience	4-Summary	5-Explaination	6-Conclusion
I would like my teacher to know my background and abilities in terms of knowledge and skills before the BST session	3.97 ± 0.06	0.70	0.02	0.09	0.17	−0.14	0.01
I would like the teacher to inform the patient about the BST session beforehand	4.35 ± 0.05	0.551	0.31	−0.19	0.11	0.33	0.15
I would like the teacher to ask the patient's agreement to the physical examination to be carried out on him/her	4.33 ± 0.05	0.50	0.48	−0.16	0.02	0.14	0.27
The teacher should be part of the patient's management team	4.01 ± 0.06	0.50	−0.10	0.25	0.13	0.09	0.36
I would like the teacher to introduce me to the patient by my name	4.13 ± 0.05	0.08	0.78	0.02	0.09	0.14	0.10
I would like the teacher to introduce the patient to me	4.16 ± 0.05	−0.04	0.83	0.12	0.15	−0.04	0.08
I prefer to communicate in simple language to any questions asked by the patient	4.31 ± 0.04	0.30	0.45	0.32	0.11	0.34	−0.23
I would like the teacher to allow me to present the case	4.16 ± 0.05	0.05	−0.06	0.60	0.17	0.28	0.30
I would like the teacher to guide me as I elicit the physical signs	4.30 ± 0.05	−0.16	0.02	0.61	0.26	0.33	0.06
I would like the teacher to allow me to think aloud and describe what I feel during the physical examination	4.22 ± 0.05	0.00	0.13	0.57	0.32	0.00	0.32
I would like the teacher to observe me taking history from the patient	3.64 ± 0.08	0.10	0.06	0.70	−0.05	−0.04	0.01
I would like the teacher to point out explicitly what we learnt from the patient encounter	4.37 ± 0.04	0.06	0.13	0.08	0.61	0.04	0.32
I would like the teacher to summarize the history and physical examination findings and organize them logically for us	4.37 ± 0.05	0.17	0.13	0.13	0.80	0.13	0.05
I would like the teacher to help me summarize what I have learned	4.40 ± 0.04	0.15	0.07	0.13	0.75	0.20	0.23
I would like the teacher to correct any misconceptions or doubts I have	4.43 ± 0.05	0.25	−0.06	0.36	0.13	0.65	0.21
I would like the teacher to encourage me to think critically about the patient encounter	4.36 ± 0.05	0.25	0.11	0.24	−0.00	0.60	0.33
I would like the teacher to explain the pathophysiological basis of the patient's problem(s)	4.27 ± 0.05	−0.15	0.14	0.05	0.23	0.72	0.17
I would like the teacher to discuss how to investigate and manage such a patient	4.53 ± 0.04	−0.01	0.06	−0.02	0.28	0.79	013
I would like the teacher to give me immediate feedback about my performance	4.28 ± 0.05	0.19	0.18	0.33	0.23	0.04	0.58
I would like the teacher to tell me about his/her experience(s) with similar patients	4.28 ± 0.05	0.11	0.26	0.35	0.03	0.19	0.55
I would like the teacher to help me to identify the key learning outcomes from this BST experience	4.39 ± 0.04	0.08	0.12	0.07	0.23	0.16	0.77
I would like the teacher to conclude with guidelines for encounters with future patients	4.33 ± 0.04	0.15	0.09	−0.06	0.08	0.27	0.73
I would like the teacher to help me to devise a flowchart, working plan or algorithm for future patient encounters	4.16 ± 0.06	−0.05	−0.24	0.19	0.29	0.15	0.83
Total variance explained by each factor: (%)		7.4	9.2	9.9	9.5	11.2	12.8

Discussion

According to situated learning theory, the development of clinical expertise depends more on specific experiences in a realistic setting, and less on generic problem-solving skills and general knowledge (Perkins & Salomon 1989). Therefore, it is important to maximize the learning from those realistic experiences such as bedside teaching. In bedside teaching, the primary participants are a teacher, the students and a patient, and the interaction of these three will play the major role in determining the success of the process. The characteristics of the teacher have been described in other communications, but the behaviour of the teacher and the way they conduct the learning that happens at the bedside may be different from their characteristics. For example, characteristics such as empathy for student and patient, approachability and so on are important, but do not tell us how the bedside teaching should actually be conducted. The principal component analysis suggested that the data could be assigned to six domains, and these are immediately recognizable as discrete components of the bedside teaching process.

The first (Preparation) includes factors that are concerned with the preliminary setting for the teaching and involves both preparing the patient for the encounter and taking the trouble to determine the initial levels of knowledge of the students. The second domain (Introduction) is concerned with introduction of the patient to students and vice-versa and communication in very simple language around the patient's bed. The third domain (Experience) is concerned with the clinical experience at the bedside which may involve, but is not limited to, presentation of the case by a student, taking a history and eliciting physical signs. It involves the teacher's observation of the learning of his/her students. The fourth domain (Summary) is concerned with the process by which there is an opportunity to summarize the learning which has just occurred. The fifth domain (Explanation) is concerned with an explanation and elaboration of what happened during the patient encounter. The last domain (Conclusion) identifies feedback and a future working knowledge as a last step in the bedside teaching process. Some of these domains were explicitly recognized by other researchers or alluded to in some models of clinical teaching (Janicick & Fletcher 2003; Cox 1993). Each of these is discussed in more detail below.

The domain we have called ‘preparation’ involves the instructor understanding the prior knowledge and skills of the student and understanding the status of the patient. Many admit that one of the chronic problems of clinical education is teaching at inappropriate level, so that, for example, a third year undergraduate student becomes exposed to instruction at the level of a sub-specialty resident. This is a particular problem when there is a heterogeneous group of learners at the bedside, and the students clearly felt that the teacher must understand in advance the audience for whom the learning has been arranged. This is a reflection of ample research by constructivists in terms of building new knowledge on prior knowledge by developing a better understanding of the learner (Vaughn & Baker 2001).

The issue of patient consent for teaching is polemic, with some authors favoring a mandatory written consent (Hartley et al. 2003) and others a less formal verbal consent, which must still be accompanied by a full explanation of the purpose and the procedures (Howe & Anderson 2003). Clearly the preparation phase must also involve ensuring that the patient is ready for the subsequent interaction with the learners, and this carries with it the idea that the physician conducting the bedside teaching should be in contact with the patient and familiar with their illness. This will occur when the instructor is part of the management team, as shown by the response to this question.

The domain we have described as ‘introduction’ involves the need for easy communication between learner and patient and this entails both politeness and communication in terms that the patient can understand. The third domain of characteristics, ‘experience’, stresses the need for active involvement of the student. Bedside teaching needs to involve the student taking the history, conducting the physical examination and being coached while these are happening. Many reports have described the decline in direct observation of learner behaviour during clinical education, despite the fact that this is a critical part of their learning (Aldeen & Gisondi 2006; Burdick & Schoffstall 1995). Students and junior doctors often report that observation by instructors during the process of obtaining a history or conducting a physical examination was rare or absent. This can lead to the trainees being unaware of their own deficiencies (Ericsson 2004). The students in our study emphasized the importance of direct supervision, in agreement with the findings of Howley & Wilson (2004).

Teaching at the bedside can produce an overwhelming amount of information, and students are not always able to discern for themselves what matters are important and what is of less relevance and interest. While giving a summary of the learning immediately after it happens is widely accepted part of the conclusion of a lecture, this is less well established in other instructional formats. Helping the students to identify the important things that they have learned is an important part of the process.

The ‘explanation’ domain involves dealing with any questions that the students may have and then encouraging the students to think critically about what happened, with a view to developing a clinical reasoning schema. Critical thinking as a potent stimulator of learning has been studied extensively and recommended elsewhere in many higher education fields (Maudsley & Strivens 2000). The process takes the student into the details of different aspects of the patient problem that might include everything from pathophysiology to aspects of ethics, collaborative practice or social support, as well key issues that surround investigation and management.

In the conclusion, students require feedback as an integral part of the process. The preferred form of feedback seems to be of the ‘on-action’ rather than ‘in-action’ type (Branch & Paranjape 2002) in that the students sought feedback after the interactive learning with the patient had occurred (Chambers & Wall 2000). Furthermore, this last domain deals with creating ‘cognitive prostheses’ (defined as affordances for overcoming cognitive load limitations) either in terms of practical take-home messages, flowcharts, specific short notes or even mental notes. This helps the students to remember and connect current information to previous learning. This part of the process looks toward the development of knowledge and skills that can be applied in future situations.

That the students appreciated receiving a specific summary of the learning might suggest that they have fewer characteristics of ‘adult learners’, who are suppose to be independent, experienced, internally motivated, and problem oriented. This ‘dependence’ is obvious in many of the responses which suggest that to a considerable extent the medical students want the teacher to guide and help. However, in agreement with other findings it is equally likely that that the process of learning is similar in all groups, but that the configuration of learner, context, and process has qualitative and quantitative components that may be different in learners at different stages of development (Merriam & Caffarella 1999). The students need for assistance with a summary is probably not a demonstration of their inability to function as ‘dult learners’ but rather is a reflection of the learning vector theory of the dependence of graduating medical students on their teachers (Paukert & Richards 2000). That the process is student centred in the affective domain rather than teacher directed is clear from the students interest in formulating the summary themselves and receiving guidance during the process.

The identifications of these domains provides a guide based on evidence from the students, who are rarely asked about their expectations, as to the process of an ideal bedside teaching experience, but the list of domains identified is certainly not exhaustive and mostly represents in-hospital bedside teaching; other environments may modify the suggested domains. For example, in a walk-in clinic, the ‘reparation’phase may not be feasible, or in the operating room ‘introduction’ which is based on a conscious patient may also be unrealistic. How these domains produce learning, whether they work collectively, separately or sequentially, the conceptualization of each domain with possible sub-domains, and their operationalization remains to be investigated. Because the process of bedside teaching is a dynamic one, the way in which the teacher and students elect to implement these domains will vary depending on the learning environment.

It seems reasonable that attention to these domains will result in an improved learning experience at the bedside, in the same way that attention to process in lectures or in other instructional modalities such as problem-based learning results in an improved learning experience. Additional refinement of the instrument, examination of a broader sample of students and further information about the learning environment in which bedside teaching occurs will enable us to devise a more precise guide to the characteristics of an effective bedside teaching process. The opinions of the students may assist in the development of a model of bedside teaching that is experimentally-based, although it is premature to suppose that the outline suggested here represents an established model of the bedside teaching process.

Acknowledgements

We thank the students who participated in this study, Dr Dwight Harley for expert statistical advice and College of Medicine and Health Sciences in the Sultanate of Oman for financial support for Dr AlWeshahi.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Additional information

Notes on contributors

Yousef Alweshahi

Dr YOUSEF ALWESHAHI is an Emergency physician who is on a study leave in medical education.

David Cook

Dr DAVID COOK is the Vargo Distinguished Teaching Professor in the Division of Studies in Medical Education (DSME).