Mixed messages in learning communication skills? Students comparing role model behaviour in clerkships with formal training

Abstract

Background: Medical students learn professional communication through formal training and in clinical practice. Physicians working in clinical practice have a powerful influence on student learning. However, they may demonstrate communication behaviours not aligning with recommendations in training programs.

Aims: This study aims to identify more precisely what differences students perceive between role model communication behaviour during clerkships and formal training.

Method: In a cross-sectional study, data were collected about physicians’ communication performance as perceived by students. Students filled out a questionnaire in four different clerkships in their fourth and fifth year.

Results: Just over half of the students reported communication similar to formal training. This was especially true for students in the later clerkships (paediatrics and primary care). Good examples were seen in providing information corresponding to patients’ needs and in shared decision making, although students often noted that in fact the doctor made the decision. Bad examples were observed in exploring cognitions and emotions, and in providing information meeting patient's pace.

Conclusions: Further study is needed on actual physician behaviour in clinical practice. From our results, we conclude that students need help in reflecting on and learning from the gap in communication patterns they observe in training versus clinical practice.

Introduction

Physicians need adequate and effective patient-centred communication skills to achieve optimal health outcomes in patients (Stewart et al. 2000; Di Blasi et al. 2001). Medical students should acquire these skills during their training, parallel to medical knowledge and other medical skills. In the past few decades, the importance of communication skills has been acknowledged and medical schools now spend much time and effort on formal communication training. Evidently, communication skills can be effectively trained (Yedidia et al. 2003) and research has shown which methods work best (Langewitz et al. 1998; Smith et al. 2007).

However, students not only learn about professional communication with patients during formal training in medical school but also in the context of clinical practice. This last context is so powerful as a learning environment that it may trigger specific learning mechanisms (Back et al. 2009; Egnew & Wilson 2010). There is ample evidence that, especially in the field of learning communication skills, role models in clinical practice play a paramount role (Thiedke et al. 2004; Weissmann et al. 2006; Egnew & Wilson 2010). In the transition to the uncertain context of daily practice, students may be inclined to focus on and act like the physicians that seem to function well in that environment. Bandura was one of the first to point out this form of social learning (Bandura 1962). In the eye of the observer, role models who have high status, prestige, power, warmth and/or sympathy enhance this form of informal learning.

Table

Practice points

During clerkships, students saw physicians using communication skills, which not always match with the skills they learned during training programmes. Especially in exploring thoughts, expectations and feelings, students often perceived bad examples. There were significant differences between the first (internal medicine and surgery) and the last two clerkships (paediatrics and primary care). Physicians educating medical students need to discuss communication behaviour in their consultations. Formal training and clinical practice need to be more in accordance.

Moreover, everyday clinical practice offers students a more complicated context than formal training and confronts them with time constraints and real patients. Recent literature shows that such context factors influence communication patterns (MacDonald & Green 2001; Bensing et al. 2003; Hawken 2005; Yamkovenko & Holton 2010; Essers et al. 2011). Although students value patient centredness and want to learn about interactions with patients, in their encounters in clinical practice, their attention is focused on history taking and time management (Small et al. 2008). Some studies even show that patient centredness decreases over time: senior students have less patient-centred attitudes than more junior students (Haidet et al. 2002; Bombeke et al. 2010). Both mechanisms – role modelling and attention focusing – have an impact on the transfer of communication skills from training into the context of clinical practice. As a consequence, communicating with patients in daily practice may seem very different for students from what they have practiced in their training sessions, and they may even not recognise the similarities.

Considering the impact clinical teachers, as role models, have on student learning during clinical work (Cruess et al. 2008; Wear & Skillicorn 2009; Helmich et al. 2010), we need a thorough insight in what communication behaviour of clinical teachers students actually observe. If we want formal training to be optimally effective, it is relevant that students see good examples by clinicians who reinforce formal training. However, if this is not the case, transfer of communication skills may be hampered (Heaven et al. 2006; Burke & Hutchins 2007). Several studies have offered insight in students’ overall impression of communication skills demonstrated by their role models (Nogueira-Martins et al. 2006; Bombeke et al. 2010; Egnew & Wilson 2010; Egnew & Wilson 2011). All studies report both positive and negative role-modelling. However, little is known about the specific communication behaviours students perceive in their role-models in clinical practice. We, therefore, decided to study to what extent the specific communication behaviours, aimed at in formal training, are recognised by students in clinical practice during clerkships. Our research questions were: to what extent do students perceive the communication behaviour of their supervisors in clinical practice as similar to what they have been taught in communication skills training? Are there differences between the different clerkships in this respect?

Methods

Setting

In the medical curriculum of the Radboud University Nijmegen Medical Centre, students follow a longitudinal training program in communication and consultations skills (C&C; see Figure 1). The medical curriculum consists of a 3-year bachelor and a 3-year master program. During the bachelor years, communication is discussed in a number of lectures and group sessions. The systematic communication skills training program intentionally starts relatively late: at the end of the third year (PCT-1), preceding the first clerkships in the fourth year. In the masters program, clerkships occur within a broader structure of ‘episodes’. Each episode includes 1–4 weeks of classroom-based preparation sessions, followed by one or more clerkships, and completed with one to two reflection weeks, and assessment. C&C skills training is extended throughout the fourth, fifth and sixth year in between, and alternating with clerkships. This schedule was chosen with the specific intention to consolidate the skills students have developed throughout the clinical period, and to teach students new skills during the full length of the program, hoping to bridge the gap between training and clinical practice better than before. For the communication skills training, students are divided into small groups of three or four, allowing every student to have their own practice turns and receive personal agenda-led feedback.

Figure 1. Medical curriculum of the Radboud University Nijmegen Medical Centre and Order of episodes and clerkships.

The program starts with simple history taking, eliciting the medical problem as well as the patient context, and continues with consultations and communication difficulties, which can be encountered in clinical practice of various specialties and in primary care. The biopsychosocial model is used for gathering person-centred context information (Borrell-Carrio et al. 2004). Students learn to ask questions about the patient perspective covering the biological dimension (the patient's own narrative of the symptoms), the psychological dimension with questions about cognitions (thoughts, attributions and expectations), emotions and behaviours; and the social dimension. The program was a development cooperation between the departments of primary care and medical psychology together with the restructuring faculty committee and coordinators of the various clerkships (all medical specialists). These specialties are also involved in the formal training sessions. However, many of the student's role models during clerkships do not participate as teachers or supervisors in C&C skills training.

Design

The study is a quantitative survey with a cross-sectional design. We collected data about physicians’ communication performance as perceived by students by means of a questionnaire. The questionnaires were presented and collected by two student research assistants at the end of a group reflection session in the final week of four different clerkships.

Participants

Participants were students having finished their clerkships in internal medicine and surgery (both in the fourth year) and paediatrics and primary care (in the fifth year). The order of the clerkships is shown in Figure 1. In three consecutive months (February to May 2010), we approached the different groups finishing their clerkships and asked to complete the questionnaires at the end of their final monthly group meeting. Participation was voluntary and anonymous. The study was conducted with permission of the institutional review board.

Questionnaires

The questionnaire contained general and detailed closed questions, asking students how frequently they observed good and bad examples of various communication behaviours in their role-models during that particular clerkship. Questions were derived from the learning objectives of the communication skills training program, using the bio-psycho-social model as a reference.

Students were first asked how often they observed senior physicians during that clerkship and then to base their answers on their observations of physicians performing their clinical work. Their observations could involve the students’ supervisor as well as other physicians on the wards or out-patient clinic.

Answers were scored on a four-point scale (very often, regularly, not so often and almost never; Appendix 1). The questionnaire was discussed by the researchers and tested by a group of communication skills teachers (psychologist and physicians). This questionnaire was then piloted in two groups of students and further adjusted using students’ comments.

Analysis

Data management and analysis was performed using SPSS 17. We computed chi-quadrate and linear-by-linear association tests for analysis.

Results

Of all 316 students in the groups, 289 questionnaires were returned (91%). The mean age of these students was 23 years. Over two-thirds of the students were female (71%) and most had a Dutch ethnic background (94%), which is representative of the Nijmegen medical student population. Ninety-three percent of the students very often or often observed consultations by physicians and only 7% not so often or almost never did. Almost all students (98%) observed more than two different physicians (19%: 2–4 doctors, 79%: 5 or more doctors; see Table 1).

Table 1  Respondent numbers and number of observed doctors

	Number of	Number of doctors observed in clerkship
Clerkship	students	0	1	2–4	5 or more
1 Internal medicine	81	1	2	19	59
3 Surgery	73	0	0	6	67
4 Paediatrics	55	1	2	5	47
7 Primary Care	80	0	0	25	55
Total	289	2	4	55	228

For analysis purposes, we combined the answers ‘very often’ and ‘regularly’ as well as the answers ‘not so often’ and ‘almost never’. In general, 54% of all students reported to have seen examples similar to those of their C&C training regularly or very often, 46% answered with not so often or almost never. Table 2 presents details per group. We found no significant difference between the groups, but there is a significant linear-by-linear association with more similar consultations towards the later clerkships in paediatrics and primary care. We found a significant difference between internal medicine and primary and community care, the latter perceived as showing more resemblance to the formal training (p < 0.05). Results are presented below under the different headings in Table 2.

Table 2  Frequency examples of communication similar to C&C training per clerkship (%)

	Very often/ regularly	Not so often/ almost never
1 Internal medicine	37 (46^†)	43 (54)
3 Surgery	35 (49)	37 (51)
4 Paediatrics	30 (58)	22 (42)
7 Primary care	48 (63)	28 (37)
Total^‡	150 (54)	132 (46)

^†χ²(Clerkship 1×Clerkship 7) = 6400 (p < 0.10); linear by linear association = 5.737 (p < 0.05).

^‡Linear by linear association = 5.106 (p < 0.05).

Information gathering

Table 3 presents the results on information gathering. The majority of the students (90%) had often seen good examples of exploring the patient perspective (the patient's narrative on their condition) during their clerkships and a minority (10%) had often seen bad examples. There were no significant differences between the groups (Table 3). However, we did find a significant linear-by-linear trend with more good examples towards the later clerkships (p < 0.000). We also found significant differences between the clerkships regarding exploration of the psychological dimension (including attributions, expectations, concerns and behaviour) and the social context. Students reported significantly more good examples in this respect during the primary care and paediatrics clerkships than during the internal and surgery clerkships (episodes 1 and 3 < episodes 4 and 7; p < 0.01). As to exploring patients’ behaviour, students observed paediatricians do this significantly more often than physicians in internal clerkships (1 < 4; p < 0.05). Exploring the patient's social context was seen significantly more frequently in surgery than in primary care (1 > 7; p < 0.05). Almost all students (91%) saw good examples of medical history taking in all clerkships.

Table 3  Frequency of examples of information gathering (%)

	Very often/ regularly	Not so often/ almost never	Significant differences between clerkships
Good examples of exploring patient's physical perspective	263 (90)	26 (10)
Bad examples of exploring patient's physical perspective	27 (10)	262 (90)
Good examples of exploring cognitions and emotions	118 (41)	171 (59)	1 < 4^†
			1 < 7^‡
			3 < 4^§
			3 < 7^‡
Bad examples of exploring cognitions and emotions	104 (37)	185 (63)	1 > 4
			1 > 7^§
			3 > 4
			3 ∼ 7
Good examples of exploring behaviour	156 (54)	133 (46)	1 > 4^§
Bad examples of exploring behaviour	61 (21)	228 (79)	1 < 3^†
			1 > 7^‡
			4 < 7^§
Good examples of exploring social context	120 (41)	169 (58)	3 < 4^†
			4 > 7^†
Bad examples of exploring social context	79 (28)	210 (72)	1 < 3^§
			1 < 7^§
Good examples of exploring medical history	262 (91)	24 (9)
Bad examples of exploring medical history	54 (19)	235 (81)	3 > 4^†
			4 < 7^†

†p < 0.01; ‡p < 0.00; §p < 0.05.

Clerkships 1: internal medicine; 3: surgery; 4: paediatrics; 7: primary and community care.

Bad examples of exploring the presented symptoms or problems largely mirror the pattern seen in the good examples. Students observed bad examples of exploring cognition and emotions more frequently in the internal and surgery clerkships than in paediatrics and primary and community care (1 > 7; p < 0.05). They also saw more bad examples of exploring the social context in internal medicine than in surgery and primary care (1 < 3 and 1 < 7; both p < 0.05). Students noticed bad examples of medical history taking in 19% of their observations. In surgery and primary care clerkships, they observed these significantly more frequently than in paediatrics (3 > 4 and 4 < 7; both p < 0.01).

Reacting to patients, reflecting feelings, providing information and shared decision making

The results are listed in Table 4. In reacting to patient's cues, 72% of the students reported to have frequently seen good examples. However, 28% of the students also often observed bad examples. Significantly more good examples were seen in the primary care than in the surgery clerkship (3 < 4; p < 0.05). General practitioners as well as surgeons showed significantly more bad examples of reacting to patient cues than paediatricians (4 < 7; p < 0.05 and 3 > 4; p < 0.01). Both in the primary care and surgery clerkships, students observed more bad examples of reacting to patient cues than in the internal medicine clerkship (1 < 3 and 1 < 7; both p < 0.01). We also found significant differences between the specialties in observed good examples of reflecting patients’ feelings: primary care physicians gave them more often than internal physicians (1 < 7; p < 0.05) and paediatricians (4 < 7; p < 0.05).

Table 4  Reacting to patient, reflecting feelings, providing information and shared decision making (%)

	Very often/ regularly	Not so often/ almost never	Significant differences between clerkships
Good examples of reacting to patient cues	209 (72)	80 (28)	3 < 4*
Bad examples of reacting to patient cues	81 (28)	208 (72)	1 < 3^†
			1 < 7^†
			3 > 4^†
			4 < 7*
Good examples of reflecting feelings	194 (67)	95 (33)	1 < 7*
			4 < 7*
Bad examples of reflecting feelings	63 (22)	226 (78)	1 < 3^†
			1 < 7^†
Providing information meeting patient's needs	271 (94)	18 (6)
Providing information meeting patient's pace	211 (73)	78 (27)
Coming to a shared decision on management	229 (75)	60 (25)
In fact the doctor took decision on management	174 (60)	115 (40)

*p < 0.05; †p < 0.01.

Clerkships 1: internal medicine; 3: surgery; 4: paediatrics; 7: primary and community care.

In providing information corresponding to the patients’ needs, 94% of the students often or very often saw good examples. However, in providing the information adapted to the patients’ pace, good examples were seen less often (74% vs. 94%). On shared decision making, we found that 79% of the students often or very often saw good examples during their clerkships. However, 60% of the students also often noted that in fact the doctor made the decision. There were no significant differences between clerkships.

Discussion

As can be seen from our data, during clerkships, students observe different doctors who may be considered as role models. The most striking result is that in almost half of the consultations, students feel their role models are not showing the kind of communication we teach them in formal training (Table 2). Especially in exploring thoughts, expectations and feelings, students did not often see good examples, and often saw bad ones. The differences between the clerkships are also striking, with significant differences between the first (internal medicine and surgery) and the last two clerkships (paediatrics and primary care) in exploring thoughts, expectations and feelings. This seems to confirm earlier findings and reflects the attention that historically has been given to good communication in these specialties (Byrne & Long 1978; Pantell et al. 1980; Egnew & Wilson 2010).

Although differences may indeed be caused by the specialty of the physicians who were observed (and the contextual influences therein), differences may also be explained by the sequence of the clerkships, as we found a significant linear-by-linear trend with more good examples towards the later clerkships (p < 0.00). This may illustrate students’ lack of experience and their focus on taking a correct history in the earlier clerkships. In addition, students may experience higher stress levels during the first clerkships. They may need all their attention to survive during their first experience of clinical practice. Their novice status may thus explain the order effect in students’ perceptions. In the later clerkships, when students have more experience, they are perhaps more open to other aspects and more able to recognise and appreciate communication skills such as questions about attributions and concerns than in the first clerkships. Perhaps internists and surgeons do ask explorative questions, but students do not notice them because they are overly focused on the medical content and thus miss communication matters.

All students will observe various physicians and health professionals, with varying communication behaviour, throughout their clinical training. Thus, every student will have to deal with the differences between classroom and clinical practice. Given these differences and considering the complexity of the skills and the behaviour needed in communicating with patients, it seems obvious that students need help in observing and learning from the communication patterns their role models show. In addition, there is a serious reason for concern because students may not always see what there is to see. Raghoebar-Krieger et al. (2001) showed this for medical skills but it seems reasonable to assume that this is also true for communication skills.

In order to learn from their experiences, the need for debriefing and for discussion throughout the clerkships seems high, considering the impact of this form of learning (Egnew & Wilson 2011) and contextual influences on communication (Bensing et al. 2003, Essers et al. 2011). However, in a recent study, Wouda and van de Wiel (2012) found that there is no substantial increase in student communication levels from clerkships. They conclude that most physicians seem to believe that students have already reached a satisfactory level and put no effort in to improve this. Improving student communication levels, however, requires faculty development concerning practicing and teaching communication skills (Perron et al. 2009).

Implications for medical education and further research

For communication skills training, it seems important to further explore to what extent the possible differences in communication patterns between specialties can be related to contextual factors. It is also necessary to study the possible influence of the order of the clerkships on the observed communication behaviour in the various specialties. In order to triangulate our findings, further study should focus on the actual physician role model communication behaviour, e.g., through rating by experts. This will be helpful if we want to help students see what there is to see and to prepare them for the variety of communication behaviour they will encounter.

In medical education, debriefing needs to be incorporated into clerkships. This requires faculty time and reflective skills, but students will be able to profit more from their practice education.

Limitations

The results of this cross-sectional study, performed in one institution among medical students may be difficult to generalise. However, there is no reason to believe that perceptions of students in other medical centres or in other health professions will differ much if no explicit attention is paid to communication during clinical practice.

Strengths

This study adds to the existing literature on role model learning by looking at the communication behaviour students actually perceived of their role models in relation to the goals set in formal communication training. Furthermore, differences between specialties were studied. This study again highlights the risk that students might be influenced in an unwanted direction concerning their communication skills.

Conclusion and recommendations

Looking at our results, we recommend that teaching and supervising physicians pay attention to communication and debrief explicitly on communication performance during clinical practice. Students need help in observing communication patterns in order to learn from their experiences. Physicians need to be aware of the influence they have as a role model and of the communication skills they are supposed to show. Formal training and clinical practice need to be in accordance with each other for medical students to acquire and maintain adequate communication skills. Physicians taking part in educating medical students in clinical practice need to be able to demonstrate adequate communication behaviour, point students to adequate and relevant communication behaviour in consultations and also explicitly discuss bad examples. Although it is a challenge, good faculty development programs should be available to address this issue.

Ethical approval

The study was conducted with permission of the institutional review board. Participation was voluntary and anonymous. Ethical approval was not required for this study.

Acknowledgements

Special thanks go to Professor Marcy Rosenbaum for bringing up the idea and giving the initial suggestions for literature and to Siem Meijerink and Floor Kuijs who helped gathering the data and put them all into SPSS. And of course, we want to thank the students who filled out the questionnaires.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article. This study was supported by Radboud University Nijmegen Medical Centre.

Appendix 1: Questionnaire

Table

Appendix 1 Questionnaire
Gender	0 Male	0 Female
Age	yrs
	Dutch	Other	Mixed
Ethnic-cultural identity: I primarily feel	0	0	0
Clerkship number
During clerkships you practiced on patient-centred communication skills in the consultation and communication training program (C&C). We would like to know what examples of different aspects of the C&C training program you saw during clerkships. When answering the questions, think back of what you learned in the C&C communication training program. Please mark the answer to the following questions according to your experiences.
	Very often	Regularly	Not so often	Almost never
1. How often were you able to watch your supervising doctor's consultations during the clerkship?	0	0	0	0
2. In general, how often did you see examples of communication similar to C&C training, in the consultations you observed?	0	0	0	0
From here on we will ask for some more details. Which of the following aspects of patient centred communication did you see?
3. I saw good examples of exploring the patient's perspective concerning the symptoms	0	0	0	0
4. I saw good examples of exploring the patient's cognitions and emotions	0	0	0	0
5. I saw good examples of exploring the patient's behavior	0	0	0	0
6. I saw good examples of exploring the patient's social context	0	0	0	0
7. I saw bad examples of exploring the patient's perspective concerning the symptoms	0	0	0	0
8. I saw bad examples of exploring the patient's cognitions and emotions	0	0	0	0
9. I saw bad examples of exploring the patient's behavior	0	0	0	0
10. I saw bad examples of exploring the patient's social context	0	0	0	0
11. I saw good examples of medical history taking	0	0	0	0
12. I saw bad examples of medical history taking	0	0	0	0
13. I saw good examples of reacting to patient cues	0	0	0	0
14. I saw good examples of reflecting patient's feelings	0	0	0	0
15. I saw bad examples of reacting to patient cues	0	0	0	0
16. I saw bad examples of reflecting patient's feelings	0	0	0	0
17. I saw good examples of providing information meeting the patient's needs	0	0	0	0
18. I saw good examples of providing information meeting the patient's pace	0	0	0	0
19. I saw good examples of shared decision making concerning management of the problem	0	0	0	0
20. I saw examples where in fact the doctor took the decision for the patient concerning management of the problem	0	0	0	0
	More than 5	2–5	1–2	None
21. I saw examples of a number of different doctors	0	0	0	0
22. What conclusion do you draw from your observations?