Fryer-Edwards, Kelly PhD; Arnold, Robert M. MD; Baile, Walter MD; Tulsky, James A. MD; Petracca, Frances PhD; Back, Anthony MD
Small-group teaching formats are now commonplace in medical education, in applications as varied as problem-based learning, clinical and professional skill practice, reflective discussions, and case-based learning.1 Small-group teaching has been especially useful for teaching complex proficiencies such as clinician-patient communication.2 However, little is known about the specific teaching and learning dynamics within such small groups that make them effective.
Current resources for small-group teaching include guides for basic skills from the University of Calgary3,4 and reflective essays by master teachers.5,6 In addition to these basic skills, faculty also need descriptions of teaching practices specific to small-group teaching that address challenges such as creating a safe learning environment, working with disengaged or resistant learners, assisting learners to identify and build on their skills, and motivating learners to bring new skills into their future practice. These practices have been discussed in the psychology literature on group process7 but appear less frequently in medical education literature.
Characterizing such teaching approaches in greater detail not only can improve teaching practice, but also may contribute to research efforts. A recent Cochrane Review has noted that meta-analyses of educational interventions meant to improve learners’ communication skills are difficult to conduct, since existing reports provide only the most basic outline of what the intervention entailed, without unpacking the black box of the actual teaching that occurred.8 This missing detail also makes replicating educational interventions at other institutions or settings difficult. Using an empirical approach including direct observation of teaching, we sought to identify a set of reflective teaching practices that help make small groups effective.
In this paper, we describe an observational study of intensive small-group communication skills teaching over a three-year period. Our objectives were to identify effective teaching practices and to create a conceptual model of teaching communication skills that could be used to guide future faculty development.
The Teaching Context
Four of us (RMA, WB, AB, JAT) acted as small-group facilitators for Oncotalk, a four-day, end-of-life communication skills retreat for medical oncology fellows that was held seven times between 2002 and 2005. The retreats were held in Aspen, Colorado, during the off-seasons and were funded by the National Cancer Institute (R25#92,055). Details of Oncotalk, including pedagogical assumptions and methods, are described elsewhere.9 The faculty facilitators for this program are co-investigators on the NCI grant, and were particularly well-suited for implementing the program based on their extensive experience in teaching communication skills and formal training as small-group facilitators. We built observation and feedback on small-group teaching and daily faculty meetings into the project design to ensure consistency in teaching.
Each day of the Oncotalk retreats included a 30-minute didactic session where the facilitators described a specific communication task to the learners, followed by a three-hour practice session where the learners practiced skills that involved both cognitive and affective components. The bulk of the facilitators’ teaching occurred during the skill practice sessions, in which small groups of five learners worked with one faculty member and took turns interacting with simulated patients (SPs). Audio or video recordings were made of one small-group teaching interaction per faculty facilitator during each retreat, for a minimum of seven 3-hour observations of each of the four faculty members, and a total of 100 hours of observation. Each observation included a short debriefing interview between the faculty member and the observer (KFE) preceding and following the teaching encounter. During this 15-30 minute interview, faculty facilitators set goals for their teaching session, reflected on teaching practices and learner needs, and received feedback on their teaching. The observer, who has a background in education and qualitative research, attended daily faculty meetings, as well as all other course sessions, and interacted with learners outside of formal course sessions. Observation and interview notes were archived for data analysis. A subset of audio and video recordings were reviewed by a second analyst (FP) and partially transcribed for further analysis. A DVD of video excerpts was produced to allow for group discussion of teaching techniques, and we used the DVD and transcripts for participant review of data.
After the first two Oncotalk retreats, we developed a preliminary typology of teaching behaviors, which we was used for initial coding of transcripts and data collection in future retreats (see Table 1 for a selection of such teaching tasks).10 We transcribed and coded a complete transcript of one teaching session for each faculty facilitator to identify the teaching behaviors observed. The observer reviewed the transcript with each faculty facilitator. Through this content analysis and team discussions, we moved from a description of discrete teaching behaviors to a characterization of overarching teaching strategies used. We describe three such strategies within this paper.
The University of Washington Human Subjects Division approved these procedures. Fellows participating in Oncotalk were informed that the faculty facilitators were being observed for the purpose of improving their teaching.
A Reflective, Process-Oriented Teaching Framework
The discrete teaching skills we observed included a variety of established practices, such as goal setting,11 eliciting feedback from SPs during skill practice sessions,12 and giving feedback.13,14 In addition to these basic teaching skills, our analysis identified a set of teaching practices that provide a reflective, process-oriented framework to extend these skills for small-group teaching: identifying a “learning edge,” proposing and testing hypotheses, and calibrating learners’ self-assessments. These teaching practices were used by faculty throughout their teaching sessions, including while setting up, conducting, and wrapping up the skill practice sessions. We describe each of the three reflective teaching practices below and offer a conceptual model that illustrates the iterative relationship between the practices.
Identifying a learning edge
A dominant theme among the faculty members was a focus on working with learners to identify their own “learning edge.” As defined by one faculty member, the learning edge is the place where the learner can work that will be challenging, but not overwhelming. Identifying a learning edge typically involves negotiating a commitment by a learner to work on a discrete step of a larger communication task that is relevant to the learner’s goals. Identifying a learning edge is a process-oriented perspective on goal setting that shifts the emphasis from a list of goals to a process of engagement; faculty referred to this process as “titrating” the experience to the learner. Analogous to finding an appropriate dose of morphine to provide pain relief for specific patients, faculty saw different learners needing different levels of challenge in order to learn most effectively.
During goal setting, facilitators would employ this strategy by nudging learners along to be specific in setting their learning goals. For example, in a practice skill session where the focus was to practice breaking bad news, each learner was asked to identify specifically where he or she was challenged and wanted practice. At the outset of a different teaching session, the facilitator asked each learner in turn, “Is there something about bad news discussions that is especially difficult for you? Is there something you would like to try today?” Another facilitator simply stated directly, “What I’d like you to think about is what would be the challenge for you.” Working with SPs rather than actual patients, with opportunities to take time out and get feedback, learners were encouraged to work on the parts of the patient interactions that were challenging for them. Implicit in the set up of each session was a sense of safety and an encouragement of risk taking.
In the middle of skill practice, facilitators would occasionally observe a learner working smoothly through an interaction with an SP. When this occurred, the facilitator would call a time-out and discuss retooling the learning objective or the SP interaction so that the learner could work more closely to her learning edge. This facilitator–learner interaction illustrates a common approach:
Facilitator: This discussion is going along really smoothly. Does it feel that way to you?
Learner: Yes, patients who respond like this are easy for me.
Facilitator: What kinds of patients are hard for you with these transitions conversations?
Learner: When they don’t accept that there are no other chemotherapy options that make any sense at all.
Facilitator: So, let’s try that. [To SP] Why don’t we roll back to before the statement about the current chemotherapy not working, and you can respond differently this time? Okay?
At the conclusion of skill practice, facilitators continued to push learners along toward the learning edge. For example, one facilitator asked, “What’s the take-home point for you?” to which the learner replied, “I wasn’t prepared for this conversation, and the emotion caught me off guard.” The facilitator then followed up by encouraging the learner: “That sounds like a good insight for you. Perhaps next time we can work on how you can prepare yourself prior to an encounter.” On other occasions, facilitators used the concept of identifying and working on the learning edge during their summary remarks for the whole session. One such comment a facilitator made to a group of learners was, “I think today we got to an important place for many of us: can we find a way to just be there in the face of inconsolable sadness?”
Faculty differed in their approaches to identifying the learning edge: although all four asked learners to self-reflect on where the edge would be, some facilitators also offered their own suggestion of where the learner’s edge might be. One facilitator commented, “One of the things I’d invite you to try is slowing down a bit. Because you are talking very fast, did you notice that?” By making observations in the form of suggestions of skills to try, facilitators worked with learners’ ability to self-assess. Another facilitator offered it this way: “What I want to do is add one more goal for you, which is what we talked about yesterday….” Learners were not always comfortable, which faculty took as a clue that they were working in the right places, as this exchange demonstrates:
Learner: It didn’t feel natural to me, and that really bothered me.
Facilitator: That’s exactly where you should be. I fully expect a lot of these things to feel awkward. The important thing is, how did the patient feel?
The decision to use a direct approach was more a matter of each facilitator’s style and comfort with that approach, rather than something that emerged over the course of the week. Whether given at the outset or at the end of the week, direct comments were offered respectfully, and always as an invitation or suggestion, tying back to learners’ specific behaviors or previous insights, which maintained the sense of learner-centered work even with active faculty participation.
Identifying a learning edge was described by one facilitator as the primary goal of his teaching. He explained that “the learning edge is the place where learners will really be engaged and learning will happen.” Faculty used this language and the rationale explicitly during the skill practice sessions. The practice sessions were continually framed by the facilitators as opportunities to work on areas that learners struggled with in practice, which served to invite learners into the session and ask them to step up and experiment with skills that would be most useful to them. Learners responded to this concept and to the expectation that they would take risks and work on areas that were more challenging for them during these sessions.
Proposing and testing hypotheses
During skill practice sessions, faculty facilitators made observations, formed hypotheses, and tested these hypotheses explicitly in the course of their interactions with learners. They formulated hypotheses on multiple issues, including barriers or facilitators to learning for specific learners, learners’ emotional responses to patient interactions, group dynamics and roles, and teaching strategies based on hypotheses about where the learners might be stuck. For example, one facilitator reflected during a debriefing session:
This learner is having a hard time identifying patient emotion. I wonder if he is seeing it and is afraid to go there, or is not attuned to the emotional data he is getting from the patient. In our next practice session, I am going to just ask him about what emotional data he sees in every patient seen by the group and see what comes up.
Proposing and testing hypotheses was the practice that helped faculty identify barriers, both emotional and cognitive, that needed to be addressed before learners could achieve their goals or reach a higher level of skill development. These barriers were seen as important opportunities for learning.
Facilitators made specific observations throughout the small-group sessions, often referring to notes they took during learners’ encounters with SPs. For each learner, they compiled their observations of learner goals, the specific behaviors used by learners during skill practice, and the lessons the learners stated they were taking away from each skill practice session. During their debriefing interviews, the faculty noted that tracking goals and take-home lessons for each learner helped them prepare for each session, recalling where a learner was working previously. Having notes about learners’ specific behaviors helped the faculty give targeted feedback during the skill practice sessions. Rather than give feedback on all behavior and skills they observed, facilitators tied their feedback to learner goals and to the areas that the facilitator felt would help the learner move to the next level of skill development. Hypothesis formation helped the faculty identify where these areas might be for each learner, and also helped faculty attend to what each learner might be ready to hear. For example, one learner was challenged to be silent in the face of patients’ emotion. The facilitator hypothesized during debriefing that the learner’s discomfort had to do with her own personal grief experience; he also believed that it was too vulnerable a point to make within the small group. Instead, the facilitator chose to focus on the observed behavior itself (e.g., a tendency to fill the silences with more medical information) rather than share his guess about the emotional barrier for this learner.
Facilitators tested their hypotheses with learners in a variety of ways. One facilitator stated his behavioral observation directly to the learner: “I noticed that when the patient came right back at you after the bad news, you sat back in your chair and crossed your arms. Did you notice that?” After describing the behavior this facilitator went further to propose a hypothesis to the learner: “I imagine there is so much coming at you right now that you are just putting some stuff on hold, is that right?” Facilitators also used probing questions to explore their hypotheses with learners: “I was wondering how you are feeling about this. Where do you think the patient is, and where are you?” The facilitator in this exchange did not make his own hypothesis explicit with the learner, but it was the formulation of a hypothesis that the learner was finding the patient’s emotional response difficult that led to the teaching intervention of asking specific probing questions of the learner.
Faculty facilitators also proposed their hypotheses directly to the learner during a practice session. For example, “This is reminiscent of the encounter from this morning. You noticed her emotion but you made a decision to hold off responding to it until after you discussed the medical information. I am just wondering, was that a cognitive decision on your part, or more of a ‘I don’t want to go there’ decision?” During this exchange, the learner acknowledged that he did not know where to go with the family member’s intense sorrow of losing her loved one. The group was able to brainstorm possible responses to the family member, and the facilitator was able to offer an important teaching point by observing what happened to the conversation when the learner did not want to follow where the person was going.
Whether individual facilitators’ hypotheses were on target or not, each commented that they gained important feedback about where the learner was and what teaching strategies were going to be the most useful. Even when they posed their hypotheses and were proven wrong, the facilitators gained insight into the learner’s needs. For faculty, developing the skills of observation allowed them to reflect back specific behaviors they saw within the encounter; shared with respect and sensitivity, these specific reflections allowed learners to achieve greater skill development.
Calibrating learners’ self-assessments
In the third reflective practice that facilitators employed, they deliberately and explicitly calibrated each learner’s self-assessment. Calibration refers to a process of focusing the discussion and feedback on those issues that are closest to the learning edge for each learner, including assumptions about what the learner is ready for and what will help him or her move forward, as described in the section above. Calibrating involved using an extended think-aloud technique with learners in order to elicit their self-assessments, values, and beliefs, then using these insights to foster further reflection. This practice is a kind of mentored reflection that teachers and learners share. That is, while continually probing for learners’ insights and self-assessments, facilitators found ways to extend learners’ reflections by adding their own observations or continuing with more questions to foster reflection. Common probes they used asked learners to think about what their own internal processes were during an interaction. Two facilitators’ questions are illustrative: “What is it about this case that makes you uncomfortable?” and “What’s the discomfort you are feeling?” Reflections on the skills learners used and their impact on the conversation were also explored and reinforced. Calibration in some cases involved facilitators reframing an experience or a take-home lesson for the learner into something he could work with in the future. For example:
Facilitator: So, looking back at the whole conversation, what do you think?
Learner: I’ve identified a weakness.
Facilitator: And did you pick up ways to work with that?
Facilitator: I think you did really well. Some of the work here is the insight. You can really change these things, when we know what motivates us to make certain choices within an interaction.
The facilitators used the skill of calibration to help keep learners focused on their learning goals by asking for self-assessments of how the current encounter was playing out. For example, one facilitator asked during a time-out:
Facilitator: What’s your sense of what’s going on with this patient?
Learner: I think she’s denying what’s going on.
Facilitator: One of your goals was listening. How do you think you did there?
Learner: I feel like I was talking a lot.
Facilitator: What made you feel that?
Learner: There is just so much information I was trying to get out about treatment options and our next steps. I just didn’t want her to think there was nothing more we could do.
Facilitator: What is it like for you to be quiet?
When calibrating learners’ self-assessments, facilitators reinforced the skills or insights that would help them to improve. They also highlighted learners’ strengths:
Facilitator: I feel like actually, this is a strength of yours that I don’t want to see you curtail. To me, you are expressive in a very warm way, with your hands. I mean, I remember that somebody commented, yesterday, that after you started to use your hand more, that you became [interrupted]
Learner: Much better connection …
Facilitator: It was like you were more open, that you were opening up a little bit and so I think that’s a skill not everybody has. I don’t have the same facility with using my hands, in that way, to draw people in. And so I would want you to see that as a strength that you can use to different degrees, depending on how the patient reacts, rather than you feeling like it’s something you’ve just gotta curtail. So, I think that’s a great example of one of the strengths that you actually have.
In the most recent retreats, facilitators began to prioritize discussions about strengths as building blocks for acquiring new skills and taking risks. One gave these instructions: “For today, I want you to think about your strengths and think about how they can help you try something new today.” They also used the group and their own observations to help learners identify their strengths, naming specific skills or qualities: “Use your instincts, your authenticity. The times when you used it, it was really powerful.”
In wrap-up or closing sessions, facilitators continued to ask probing questions to prompt learners’ self-assessment: “What are you doing that is new for you? That you wouldn’t have done five days ago?” These kinds of probes again required specific and in-the-moment observations on the part of the facilitator. The skill of calibrating self-assessment became the engine that pushed learning forward that the facilitators used to assist learners in consolidating the skills they practiced during the session.
The three teaching practices we describe here characterize a process-oriented approach to teaching that is iterative and dynamic. They are reflective practices, in that the facilitator is continually reflecting upon the learner’s needs, specific observations within the skill practice session, their own interactions with the learner, and what it will take to help the learning experience be fruitful. By proposing and testing hypotheses, facilitators could get closer to identifying a learning edge for learners. Calibrating self-assessments also involved hypothesis testing, consolidating learning and pushing learners to a new learning edge. Working at the learning edge can be slightly uncomfortable, so collaborating with learners to experience success and identify their own strengths (calibrating self-assessments) reinforces learners’ ability to take risks and push their skill development further. Working dynamically with each of these teaching practices moved facilitators and learners toward more insight into their teaching and learning behaviors, respectively (see Figure 1).
Within the small groups that worked together intensively over the space of four days, learners were able to try practices that were quite new to them. If a new skill or strategy did not work the first or second time, each learner was able to reflect on what was difficult and elicit suggestions and support from her or his peers and faculty member. Taking greater risks was associated with the facilitator’s ability to gently encourage learners while pushing them, creating the sense that the facilitator was on the learner’s side and increasing trust within the group. Clearly, small groups that are formed for one-time teaching purposes will not go as far in their risk-taking as would groups existing over longer periods. However, we anticipate the same basic process-oriented teaching practices would enhance both short-term and long-term small groups or other clinical teaching activities by keeping the teaching learner-focused and responsive. Indeed, we use elements from this teaching model in other aspects of clinical teaching, from bedside teaching to faculty development.15,16 For example, when rounding on patients with fellows, one attending takes a moment to ask the fellow for her specific learning goals with the particular patient-interaction, framed specifically around what would push her most in her professional growth. For another faculty member, the practice of testing hypotheses with learners contributed to his ability to be open with fellows. Checking back with residents and fellows to find out what they are taking away from a challenging patient encounter is also a useful teaching strategy used regularly in clinical teaching.
The conceptual model of teaching suggested by these teaching practices cannot be simply categorized as faculty-directed or learner-centered. It is learner-centered in the sense that the teaching strategies employed are calibrated to learners’ specific needs, and those needs drive where the work occurs within the session. However, the approach also requires significant direction from the facilitator, who must actively work to adapt and move the session toward the learning edge for participants. We thus think of this approach as relationship-centered, in the same sense that a patient-physician encounter can be relationship-centered.17 The learning experience that occurs within the small group is highly dependent on what both the teacher and the learner bring to the experience.
A common feature between the three teaching practices we have described is that each involves high levels of observation, reflection, and ongoing modification on the part of the facilitator. The ability to reflect-in-action and reflect-on-action has been described by Schön and others as a critical skill within reflective practice.18,19 Reflective practice allows an educator to be responsive to the specific context and immediate needs of a given situation. Teaching communication skills asks teachers to be highly responsive to the unique needs of each learner. Approaching teaching as a reflective practice requires facilitators to be active co-constructors of a practice session, based on the learner’s specific needs.20,21 In our experience with Oncotalk, after seven retreats, the teaching remained fresh and challenging for the faculty, as each set of learners and each teaching session required unique responses from the faculty. Keeping the faculty working on their own learning edge helped maintain their attention and focus within the sessions.
Reflective teaching practices were easily observed in our project because we built the presence of an observer directly into the project. However, it was apparent that the presence of an observer and the occasion of a debriefing session simply made explicit facilitators’ thoughts. The teaching practices we observed evolved and developed over time; teaching skills are dynamic and, much as clinicians do not just arrive at being good doctors, good teachers must develop through practice, feedback, and experiencing success. We developed a worksheet for use in later retreats to operationalize the ongoing reflective process of the retreats (see Table 2). How these specific practices of observation, journaling, and peer debriefing work to cultivate reflective teaching practices should be the focus of future study.
In interviews and debriefing sessions with the faculty, it was clear that the overarching framework we described here informed the judgments they made on a moment-by-moment basis within the small groups. Their teaching decisions were made based on judgments about what would keep the learners and the group maximally engaged, knowing they would learn most effectively if activated. The focus of the learning, in turn, was to become better communicators through enhanced self-efficacy as well as skill development. The theory of change within this conceptual model of teaching is that by pushing themselves and experiencing success, learners developed confidence and are able to take their skills to other novel and challenging situations. Increased self-efficacy reinforces learners’ abilities to bring skills into clinical practice settings, improve their clinical practices, and continue to develop as lifelong learners (see Figure 1).22
With this program, we have developed an empirically based and theoretically informed model of teaching that provides educators with a higher level of teaching skills beyond the basics. Reflection and reflective practice are often discussed in the literature, but rarely operationalized such that other teachers can take up the practices. In this paper, we suggest three teaching skills that illustrate learner-centered, reflective teaching practices that provide a process-oriented framework for teaching skills with both cognitive and affective components.
Our research suggests that future work with teaching, faculty development, and educational interventions should attend to reflective practices. Articulating the nuanced skills of effective teaching will guide educational interventions and faculty development efforts as we gain a better understanding of what experienced teachers do.
Oncotalk is funded by a grant from the National Cancer Institute (R25 92,055). Support for the additional study of teaching practices was provided by The Greenwall Foundation.
The authors would like to thank all of the Oncotalk fellows for their active engagement in the learning process, and the actors who help bring the skill practice sessions to life.