Fromme, H. Barrett MD, MHPE; Bhansali, Priti MD; Singhal, Geeta MD; Yudkowsky, Rachel MD, MHPE; Humphrey, Holly MD; Harris, Ilene PhD
In the seminal article “Clinical teacher effectiveness in medicine,” published in 1978, Irby1 focused on the qualities of exemplary clinical teachers in internal medicine (IM) and described the “best” clinical teachers as “enthusiastic, clear and well organized in presenting material and skillful in interactions with students/residents.” Irby later identified six domains of knowledge that excellent IM attending physicians demonstrate during teaching rounds: clinical knowledge of medicine, patients, and the context of practice and educational knowledge of learners, principles of teaching, and case-based teaching scripts.2
In a recent qualitative review of the literature, Sutkin and colleagues3 identified three main categories of characteristics of effective clinical teachers: physician, teacher, and human characteristics. They did not, however, delineate which attributes (if any) were specific to educators in the inpatient setting. Ullian and colleagues4 similarly identified the roles of teacher, person, and physician, as well as an additional role of supervisor; the qualities and skills associated with the latter role overlapped with many of those that Sutkin et al identified as teacher characteristics.
In the three decades since Irby's seminal study, the hospitalist movement has changed the face of clinical education in the inpatient setting. Many institutions have restructured their inpatient coverage by employing hospitalists, whose multiple roles include “clinical care, teaching, research, and leadership in the field of general hospital medicine.”5 As of July 2010, there were more than 30,000 hospitalists in the United States, approximately 18% of whom were employed by academic medical centers.6
Recognizing hospitalists' important educational role, researchers have endeavored to determine hospitalists' effect on trainees' inpatient education. In a survey study, Shea and colleagues7 found that more than half of responding IM programs employed hospitalists and that department chairs and residency program directors had an overall positive view of their effect on patient care and on medical education. In the University of Chicago Pritzker School of Medicine's IM program, more residents on hospitalist services than on traditional services were satisfied with their inpatient rotations, with the greatest differences in the educational domains.8 Numerous subsequent studies conducted in IM programs have replicated these findings: Hospitalists were perceived by trainees to have better knowledge and greater skill in general teaching, providing feedback, conducting rounds, and bedside teaching.9,10 They were also perceived to be more enthusiastic about teaching, more involved with patient care, and more available to learners.11,12
There have been few studies of pediatric hospitalists as educators. Reviewing more than 100 articles on hospitalists dating from 2000, Freed and Uren13 found that only 30 specifically focused on pediatric hospitalists; the majority of those addressed clinical issues. The few studies that focused on pediatric hospitalists as educators have shown that, overall, learners have been more satisfied with their teaching and have perceived pediatric hospitalists as being more accessible and knowledgeable, giving better feedback, and serving as better role models than nonhospitalist educators.14,15 A recent survey of pediatric residency program and clerkship directors did stress the importance of pediatric hospitalists as educators, reporting that hospitalists served as teaching attendings in 77% of programs surveyed.16 However, none of these studies focused specifically on the qualities and skills exhibited by exemplary pediatric hospitalist educators, and none of them were prospectively designed or conducted at more than one institution, limiting the utility of the results.
The ranks of hospitalists and their involvement in medical education have increased in recent decades. As noted above, hospitalists have been shown to demonstrate qualities and skills that make them better educators than nonhospitalists in the inpatient setting. Yet, there have been no robust studies with prospective, multi-institutional designs that focus specifically on the nature of these qualities and skills.
The purpose of this study was to fill that gap by identifying and describing the teaching qualities and skills demonstrated by exemplary pediatric hospitalists. Our goals were to identify exemplary pediatric hospitalists at multiple institutions, to elicit their perspectives about what makes them successful educators, to elicit their learners' perspectives about the qualities and skills of exemplary pediatric hospitalist educators, and to compare our results with prior literature to determine what, if any, unique qualities and skills pediatric hospitalists demonstrate that make them successful educators. By clearly defining the qualities and skills that successful pediatric hospitalist educators should demonstrate, we hope to enable hospitalists and hospitalist training programs to develop curricula and interventions targeting hospitalists' teaching skills, which will ultimately improve the education of residents and students in the inpatient setting.
We used a prospective qualitative design, with interviews and focus groups, to identify and describe the qualities and skills of exemplary and ideal pediatric hospitalist educators. We selected three sites with academic pediatric hospitalist programs on the basis of variability in program size, design, and geographic location: University of Chicago Pritzker School of Medicine (H.B.F.), University of Connecticut School of Medicine (P.B.), and Baylor College of Medicine (G.S.). We obtained IRB approval at each institution as well as at the University of Illinois at Chicago College of Medicine, the site of two authors (R.Y. and I.H.) who helped design the study and review and analyze the data. We conducted this study from November 2008 through January 2009.
We identified exemplary hospitalists at each institution by asking the site's chief residents and program directors to achieve consensus regarding the top 25% of pediatric hospitalists in their residency programs across all training sites (one site at the University of Chicago, two at the University of Connecticut, and three at Baylor) based on their experiences with the hospitalists and their knowledge of the hospitalists' evaluations and learner feedback. The site investigators, who are all practicing pediatric hospitalists, were excluded from consideration.
We invited by e-mail the six hospitalists identified by the chief residents and program directors to participate in 30- to 45-minute interviews. Each of the three site investigators interviewed the hospitalists associated with her residency program, using a semistructured format and asking five open-ended questions that focused on how they viewed their roles as educators on an inpatient service. The two questions relevant to this study were:
1. Describe your qualities and skills that are essential for you to be effective in your role as an educator/teacher on the inpatient service.
2. What do you view as the most important things associated with effective clinical teaching on the inpatient service?
These two questions were designed to elicit responses regarding the characteristics both of exemplary hospitalists (as identified by their programs) and ideal hospitalists (as the interviewees perceived the ideal). Each hospitalist also provided basic demographic data, which we deidentified.
Following the hospitalist interviews, the site investigator invited by e-mail the medical students and residents who had most recently worked with the selected hospitalists to participate in 30- to 45-minute focus groups (one for students and another for residents) regarding their education-related experiences with the hospitalists. A total of 18 students and 18 residents were invited across the three sites. No compensation was offered or provided. The investigator asked the focus-group session participants semistructured questions similar to those in the hospitalist interviews. We slightly altered one of the questions to reference the exemplary hospitalist: “Describe the qualities and skills that Dr. X has that make him/her effective in his/her role as an educator/teacher on the inpatient service.” The second question was not changed. We did not collect demographic data for the residents and students.
Although each site investigator conducted the interviews and focus groups independently, the investigators first discussed approaches among themselves to create as much consistency as possible, across sites, in the interview and focus-group experiences.
Using grounded theory as the general design approach for the study, we analyzed the data using the constant comparative method.17 Content units for analysis were words and phrases. We deidentified participants and assigned coding designations. Interviews and focus groups were audio recorded, transcribed, and deidentified; the audio files were then destroyed. Two coders (H.B.F. and P.B.) identified themes independently for the first two hospitalist interviews and two focus groups. After initial independent review of the first four sets of data, the two coders estimated intercoder reliability and reached consensus on themes through discussion. Prior to discussion, the two coders had 75% coder agreement; after discussion, they achieved 100% coder agreement. Coding was divided between these two individuals for subsequent data analysis. Both coders reviewed final coding of all data to ensure agreement. All focus-group and interview data were evaluated for trustworthiness by member checking (asking participants to review transcripts for accuracy and themes for effectiveness in capturing their perspectives). The themes for qualities and skills of exemplary pediatric hospitalists were triangulated using prior literature as a comparison for the data we obtained.
All six invited hospitalists participated in the interviews. The overall response rate was 78% (14/18) for invited residents and 89% (16/18) for invited medical students. Four of the six hospitalists were male, and five of the six were assistant professors. The hospitalists' ages ranged from 36 to 54 years, and their years in practice ranged from 3 to 18 years. Their average number of service weeks per year was 24.
Domains and themes
We categorized the 266 comments from the interviews and focus groups into 36 themes representing the specific qualities and skills of exemplary pediatric hospitalist educators, which we then organized into four domains: teaching skills, personal qualities, patient care skills, and role modeling. These domains and themes summarized the qualities and skills demonstrated by the pediatric hospitalists identified as exemplary educators and the qualities and skills that participants believed pediatric hospitalists should demonstrate to be ideal educators.
Because our analysis demonstrated many similarities in the frequency of comments in response to the two questions (“exemplary” versus “ideal” hospitalist educators), Table 1 and the Appendix combine data for responses to both questions. Any noteworthy variations in responses between the questions and among respondent groups (hospitalists, residents, students) are described below. In our presentation of results, we focus on the main themes in each domain and illustrate them with representative comments. The Appendix provides a tabular compendium of all themes in each domain, with descriptions of each theme and typical comments from the interviews and focus groups.
Qualities and skills of exemplary hospitalist educators
Domain 1: Teaching skills.
Comments by all three participant groups were most frequently classified as relating to teaching skills and accounted for 55% of all responses (147/266). We classified 63% of respondent comments in this domain (92/147) into six main themes.
The first theme in this domain, patient/work-centered teaching, relates to hospitalists' ability to find teachable moments and salient teaching points within the context and time constraints of patient care, team dynamics, and learner need (“tailors … teaching points to what is going on,” “incorporates teaching into whatever he is doing”). The technique-related themes of setting expectations and giving feedback were mentioned nearly as frequently. A fourth theme, stimulation of learning and problem-solving, focuses on hospitalists' actively engaging learners in a results-oriented thought process (“helps them think through the process,” “leads through clinical reasoning”). Learner climate (“supportive educational environment”) and learner involvement (“enables learners to be actively involved in the care of the patient”) rounded out the top six teaching skills themes.
Overall, this domain elicited more comments from residents and medical students than from hospitalists. Student comments referred more frequently to an ideal hospitalist, whereas resident comments referred more often to the actual exemplary hospitalists. Students also contributed more comments than the other participant groups did in the themes of giving feedback, patient/work-centered learning, and learner climate.
There were interesting differences between qualities and skills participants cited for exemplary hospitalists and those they cited for ideal hospitalists. Learners viewed time management and encouraging autonomy as essential skills in the ideal hospitalist, but they did not mention them when characterizing their own exemplary hospitalist educators. Correspondingly, the hospitalists never mentioned time management and rarely mentioned encouraging autonomy as essential characteristics.
Domain 2: Personal qualities.
Personal qualities, those qualities of character that influence behavior and functioning, accounted for 23% of all comments (62/266). Results in this domain varied by participant group.
Hospitalists provided the majority of comments classified as self-reflection/insight. They felt that not only did they themselves “take … feedback to heart” and “admit errors and improve” but also that they expected ideal hospitalists both to “reflect on how [they] learn” and recognize that it is “okay to say [you] don't know.”
Medical students frequently mentioned empathy/caring, and they were the only group that made comments about supportiveness, such as “supportive of the group” and “encouraging.” Students did not mention motivation/dedication (of hospitalists toward teaching and patient care), but hospitalists commented that they needed to have “perseverance … to figure something out” and the dedication to “try a different tactic or a little bit harder” with difficult learners.
Residents commented most frequently about the qualities of hospitalists' respect for learners, approachability, professional behavior toward colleagues and learners, and motivation/dedication. All three participant groups referred to personal qualities more often when discussing exemplary hospitalists than when describing ideal hospitalists.
Domain 3: Patient care skills.
Patient care skills accounted for only a small percentage of the respondent comments (12%; 32/266). In this domain, more than one-third of comments (12/32) related to knowledge acquisition, or hospitalists' ability to keep up-to-date about current practices and obtain knowledge that they do not already possess (“know where to get the information,” “read the literature”). Medical students, residents, and hospitalists all viewed knowledge acquisition as integral to hospitalists' ability to be effective educators. Other frequently mentioned themes were systems knowledge (“knowing how things function,” “how to negotiate the system”) and advocacy for patient health and success after discharge (“[making sure] the transition from being an inpatient to an outpatient is really smooth [and] the kid doesn't get lost”).
Domain 4: Role modeling.
Initially, we included role modeling comments within the other three domains because these comments referred to role modeling of skills or qualities. However, role modeling is a unique metacognitive skill that may include the educator making his or her thoughts and actions transparent to learners. We defined it as a separate domain to emphasize its relevance to each of the other domains and draw attention to the need for hospitalists to actively prepare to role model. The three themes within this domain reflect the themes previously mentioned for the other domains.
Comments concerning role modeling were most often attributed to an ideal hospitalist educator rather than to the actual exemplary hospitalists. The exemplary hospitalists were described as role modeling personal qualities more frequently, whereas the ideal hospitalists were described more often as role modeling patient care skills.
Summary of findings
Table 2 summarizes our results. Although some of the qualities and skills that emerged as themes in our study have been previously identified in studies of hospitalists and other clinical teachers, we identified some for the first time. Within each domain, we identified or gave new emphasis to one or two themes. Among teaching skills, patient/work-centered teaching and stimulation of learning and problem-solving were the most notable. The personal quality of self-reflection/insight and the patient care skill of knowledge acquisition were added to the skills that exemplary hospitalist educators should demonstrate. Role modeling, for the first time, was recognized as its own domain of skills.
To the best of our knowledge, this study is the first to prospectively and qualitatively identify the qualities and skills of exemplary pediatric hospitalist educators as perceived and identified by residents, students, and exemplary pediatric hospitalists. The qualities and skills we identified conform to a tripartite model that includes the domains of teaching skills, personal qualities, and patient care skills, as previously formulated in the clinical teaching literature.3,4 However, this study expands that model by adding role modeling as a unique domain.
Among these skills and qualities are some that have not been identified or emphasized previously (see Table 2). For example, patient/work-centered teaching emerged as an important teaching skill. In the era of the 80-hour workweek, a balance between patient care work and learning must be maintained. Knowledge acquisition, the ability to keep up-to-date with current evidence and literature, emerged as an important patient care skill. Prior studies2,3,9,14 have emphasized the importance of medical knowledge in clinical teaching, but those studies defined that knowledge as static and based on experience. This study demonstrates that learners and hospitalists value the ongoing, dynamic process of knowledge acquisition more than existing medical knowledge. Interestingly, Freed et al16 found that 70% of program directors thought that hospitalists have improved the use of evidence-based medicine; we believe this improvement most likely occurs through their role modeling of knowledge acquisition.
Self-reflection/insight, which emerged as an important personal quality, was identified by Irby18 as a quality of exemplary clinical teachers and by Sutkin et al3 as both a teaching and a human characteristic, but it has been commented on infrequently by other investigators. Medical schools and training programs have begun to recognize the importance of reflection in education, and the emphasis placed on self-reflection/insight in this study may reflect that view. Reflection can be used as a self-directed learning tool, so it is not surprising that physicians who reflect on their own performance can use that reflection to improve their teaching and, thus, may be perceived as better educators.
This study also emphasizes the importance and impact of role modeling by recognizing it as a separate and unique domain to underscore its relationship with the other domains. While only one prior study14 has identified role modeling qualities and skills as important in clinical educators or hospitalists, learners in our study identified the hospitalist's role modeling as an important quality in the other three domains. Role modeling can be enhanced when educators have an active awareness of themselves as role models and are mindful of making explicit their processes of thinking and acting in all three primary domains. The exemplary hospitalists in our study had this awareness regarding both their personal qualities and patient care skills.
In addition, we identified areas in which the two groups of learners (residents and medical students) differed in their responses regarding the skills and qualities they viewed as important. Residents stressed teaching skills such as patient/work-centered teaching, stimulation of learning and problem-solving, and encouraging autonomy, whereas students emphasized personal qualities such as learner climate, giving feedback, empathy/caring, and supportiveness. These differences may reflect their different objectives. Residents are challenged to learn in the context of a busy 80-hour workweek, and a hospitalist who can teach efficiently in the work setting aids them with this goal. By contrast, medical students are primarily learners, and they focus on qualities that make learning more comfortable. This is an important distinction for hospitalists to recognize as they work with both sets of learners on a daily basis.
Residents were the only group that made more comments in the domain of patient care skills than in the domain of personal qualities, which may reflect their preference to focus on practical skills that are useful to patient care. They may need more explicitly articulated role modeling to relate personal qualities to patient care processes and outcomes.
Prior studies have rarely identified time management as a skill of effective clinical teachers and have identified encouraging autonomy only in the context of concerns regarding residents' potential loss of autonomy when they are supervised by hospitalists.8,10 Indeed, Freed and colleagues16 indicated that 33% of the residency programs they surveyed reported decreased senior resident autonomy in the presence of hospitalists. Our results suggest that learners value time management and encouraging autonomy among teaching skills but that their hospitalist educators did not demonstrate or recognize the need for these, indicating a gap in expectations and an area for hospitalist development. Conversely, the exemplary hospitalists mentioned self-reflection/insight, but learners seldom did. This skill tends to be invisible to learners, unless hospitalists role model the skill by making their reflections accessible to learners via metacognitive comments so that learners can emulate them.
There are several limitations to this study. We elicited perspectives via interviews and focus groups led by three different investigators. Although we strove for consistency in interviewing styles, there may have been differences that influenced participants' comments. Our study focused only on pediatric hospitalists, so generalizability to the hospitalist profession as a whole may be limited. However, the qualities and skills we identified confirm results of prior studies of clinical educators, and the new qualities and skills we identified are not specific to hospitalists. Further, none of the respondents focused on the unique aspects of pediatric hospitalist teaching. These points support the applicability of hospitalist studies across specialties.
An observational study may offer further insight into which teaching behaviors hospitalist educators actually demonstrate. Future studies should also address the skills and qualities that make hospitalists successful as clinicians and administrators.
A review of the Society of Hospital Medicine (SHM) Web site reveals that, beyond a 2009 national meeting precourse19 and two or three workshops each year at SHM and American Academy of Pediatrics' pediatric hospital medicine conferences,20,21 no significant courses have been offered on teaching skills for hospitalists in the past several years. The SHM core competencies, developed in 2006 to provide guidelines for hospitalist training, do not address hospitalists as educators.22 Our study identifies and describes important teaching qualities and skills for residency and fellowship programs to address in their curricula and draws attention to the importance of role modeling and recognizing the varying needs of different learner groups.
The specialty of hospital medicine is now in its second decade, and hospitalists are being called on to act as the primary inpatient educators of medical students and residents. Being an effective clinical teacher takes preparation, training, and attention to the process. It is essential that hospitalists focus on their development as educators and appreciate their unique opportunities to enrich the learning of residents and students.
This study received IRB approval from the three medical schools at which interviews and focus groups took place—Baylor College of Medicine, University of Chicago Pritzker School of Medicine, and University of Connecticut School of Medicine—as well as the University of Illinois at Chicago College of Medicine.
Appendix Domains and...Image Tools
Appendix, continued...Image Tools
Appendix, continued...Image Tools
1 Irby DM. Clinical teacher effectiveness in medicine. J Med Educ. 1978;53:808–815.
2 Irby DM. What clinical teachers in medicine need to know. Acad Med. 1994;69:333–342.
3 Sutkin G, Wagner E, Harris I, Schiffer R. What makes a good clinical teacher in medicine? A review of the literature. Acad Med. 2008;83:452–466.
4 Ullian JA, Bland CJ, Simpson DE. An alternative approach to defining the role of the clinical teacher. Acad Med. 1994;69:832–838.
7 Shea JA, Wasfi YS, Kovath KJ, Asch DA, Bellini LM. The presence of hospitalists in medical education. Acad Med. 2000;75(10 suppl):S34–S36.
8 Chung P, Morrison J, Jin L, Levinson W, Humphrey H, Meltzer D. Resident satisfaction on an academic hospitalist service: Time to teach. Am J Med. 2002;112:597–601.
9 Hauer KE, Wachter RM, McCulloch CE, Woo GA, Auerbach AD. Effects of hospitalist attending physicians on trainee satisfaction with teaching and with internal medicine rotations. Arch Intern Med. 2004;164:1866–1871.
10 Kulaga ME, Charney P, O'Mahony SP, et al. The positive impact of initiation of hospitalist clinician educators. J Gen Intern Med. 2004;19:293–301.
11 Kripalani S, Pope AC, Rask K, et al. Hospitalists as teachers. J Gen Intern Med. 2004;19:8–15.
12 Navaneethan SD, Nautiyal A, Abdel-Gadir K, et al. Hospitalist physicians as educators in a community hospital: The trainee's view. South Med J. 2006;99:550–551.
13 Freed GL, Uren RL. Hospitalists in children's hospitals: What we know now and what we need to know. J Pediatr. 2006;148:296–299.
14 Landrigan CP, Muret-Wagstaff S, Chiang VW, Nigrin DJ, Goldmann DA, Finkelstein JA. Effect of a pediatric hospitalist system on housestaff education and experience. Arch Pediatr Adolesc Med. 2002;156:877–883.
15 Geskey JM, Kees-Folts D. Third-year medical students' evaluation of hospitalist and nonhospitalist faculty during the inpatient portion of their pediatrics clerkships. J Hosp Med. 2007;2:17–22.
16 Freed GL, Dunham KM, Lamarand KE. Hospitalists' involvement in pediatrics training: Perspectives from pediatric residency program and clerkship directors. Acad Med. 2009;84:1617–1621.
17 Corbin J, Strauss AC. Basics of Qualitative Research: Techniques and Procedures for Development of Grounded Theory. Thousand Oaks, Calif: Sage; 2008.
18 Irby DM. How attending physicians make instructional decisions when conducting teaching rounds. Acad Med. 1992;67:630–638.