Brief Reports

Patient complexity,1 productivity, and documentation pressures have increased substantially over the past 2 decades. Within this environment, time for teaching is often limited. The same pressures which limit faculty members' teaching time also limit their availability to learn how to teach; faculty development efforts need to be both effective and efficient.

In a seminal study of exemplary clinical teachers, Irby discovered that expert teachers often developed and utilized teaching scripts for commonly encountered teachable moments.2 Teaching scripts consist of a trigger, key teaching points, and teaching strategies.2 A trigger may be a specific clinical situation or a learner knowledge gap identified by the teacher. The trigger prompts the teacher to select key teaching points about the topic (the content), and utilize strategies for making these teaching points comprehensible (the process).2 Through a reflective process, these expert teachers evaluated the effectiveness of each teaching session and honed their scripts over time.2 While additional reports have described the use of teaching scripts,35 we found no studies evaluating the impact of collaboratively developing teaching scripts. In the present study, we sought to understand faculty members' early experiences with a program of collaboratively developing teaching scripts and the impact on their self‐efficacy with teaching about commonly encountered clinical conditions on attending rounds.

METHODS

Participants were the 22 internal medicine, or combined internal medicine and pediatrics (med‐peds), hospitalists in a 750‐bed university teaching hospital in upstate New York. Nine hospitalists worked for only 1 year (eg, chief residents and recent graduates awaiting fellowship training), and were present for half of the program year. All hospitalists conducted daily bedside attending rounds, lasting 1.52 hours, with a dual purpose of teaching the residents and students, and making management decisions for their shared patients.

Hospitalists were surveyed to identify 10 commonly encountered diagnoses about which they wanted to learn how to teach. The faculty development director (V.J.L.) conducted a 1‐hour workshop to introduce the concept of teaching scripts, and role‐play a teaching script. Nine hospitalists volunteered to write scripts for the remaining target diagnoses. They were provided with a template; example teaching script (see Supporting Information, Supplemental Content 1, in the online version of this article); and guidelines on writing scripts which highlighted effective clinical teaching principles for hospitalists, including: managing time with short scripts and high‐yield teaching points, knowledge acquisition with evidence‐based resources, self‐reflection/emnsight, patient‐centered teaching (identifying triggers among commonly encountered situations), and learner‐centered teaching (identifying common misconceptions and strategies for engaging all levels of learners) (Figure 1).2, 6 Faculty were encouraged to practice their scripts on attending rounds, using lessons learned to refine and write the script for presentation. Each script was presented verbally and on paper at a monthly 1‐hour interactive workshop where lunch was provided. Authors received feedback and incorporated suggestions for teaching strategies from the other hospitalists. Revised scripts were distributed electronically.

Figure 1

Baseline surveys measured prior teaching and faculty development experience, and self‐efficacy with teaching about the 10 target diagnoses, ranging from Not confident at all to Very confident on a 4‐point Likert scale. Using open‐ended surveys, we asked all of the hospitalists about their experiences with presenting scripts and participating in peer feedback, and the impact of the program on their teaching skills and patient care.

Because the learning objectives for each teaching script were determined by each script's author and were not known prior to the program, we were unable to assess changes in residents' and students' knowledge directly. As a surrogate measure, we surveyed students, residents, and faculty regarding how often the hospitalist taught about the 10 target diagnoses and whether teaching points were applicable to current or future patients. We administered the surveys online weekly for 8 weeks before and after the program. Residents and students were notified that participation had no impact on their evaluations. They received a $2.50 coffee gift card for each survey. The study received an exemption from the university's Institutional Review Board.

The number of teaching episodes per week related to the target diagnoses was averaged across survey weeks. Student t tests were used to compare results before versus after the intervention, and 95% confidence interval (CI) calculated. We considered P 0.05 to be statistically significant. Data were analyzed using SAS version 9.2 (Cary, NC).

Qualitative data were analyzed by coding each statement, then developing themes using an iterative process. Three investigators independently developed themes, and met twice to review the categorization of each statement until consensus was achieved. Two of the investigators were involved in the program (V.J.L. and A.B.) and one did not participate in the workshops (C.G.).

RESULTS

The 22 faculty had an average of 5 years' experience as hospitalists (range 0.824 years). Previous experience formally learning how to teach ranged from 0 to 150 hours (average 33.1 hours; median 15 hours). A mean of 9.4 hospitalists attended each of the 10 1‐hour workshops. Script writers estimated that scripts required a mean of 4.3 hours to prepare. A total of 105 (59%) resident/student and 22 (55%) faculty surveys were returned preintervention, and 83 (47%) resident/student and 19 (48%) faculty surveys were returned postintervention. There were no significant differences in the number or applicability of teaching events from before to after the program. Faculty self‐efficacy with teaching was available for 7 of the 10 diagnoses, and increased from a mean of 3.26 (n = 77) preintervention to 3.72 (n = 52) postintervention (95% CI for the difference in means 0.350.51; P 0.0001).

A total of 8 (80%) script‐writers and 5 (42%) non‐writers responded to the qualitative survey, and 77 comments were coded. Three major themes and 8 subthemes were identified (for representative comments, see Supporting Information, Supplemental Content 2, in the online version of this article). The major theme of individual professional development related especially to the personal satisfaction of researching a topic and becoming a local expert. While most comments were positive, 2 described apprehension about presenting to peers. Fifteen comments specifically addressed the development of teaching skills, 13 positive and 2 neutral. Some focused on strategies consistent with the teaching script framework, including recognizing teachable moments and the importance of preparation for teaching. Others focused on changes in teaching style, shifting to a more interactive method and involving multiple levels of learners. Others revealed that participants adjusted the content of their teaching, adding new material and changing the focus to important clinical pearls. Another subtheme was the impact on clinical care and medical knowledge base. Of the 11 comments, 7 were positive and emphasized the development of a framework for making decisions, based on an understanding of the evidence behind those decisions. Four were neutral, noting that care of patients had not changed. Two comments remarked on the time invested in developing teaching scripts. A second major theme was the development of a shared mental model of professional responsibility. This was demonstrated by comments relating to participants' motivation for learning, and development or strengthening of responsibility for teaching. The third major theme described interpersonal relationships among colleagues. Four commented on how the opportunity to see how others teach led them to appreciate the diversity of approaches, while 14 focused on collegiality among the faculty. Thirteen of these identified an increased sense of community and camaraderie, while one was neutral.

CONCLUSIONS

We had successful early experience with a faculty development intervention that involved hospitalists in creating and implementing teaching scripts related to commonly encountered diagnoses. The intervention was time‐ and resource‐efficient. Following the intervention, we found increased faculty self‐efficacy and beneficial effects in several domains related to professional development and satisfaction. We found no significant difference in the frequency or applicability of teaching about the targeted diagnoses.

In addition to the formal program evaluation results, we learned several additional lessons informally. Faculty who developed scripts had varying levels of familiarity with evidence‐based approaches to teaching. Some faculty requested to have their scripts reviewed by the program leader before presentation, and small revisions were made, emphasizing use of the tips included in Figure 1. Using volunteers, rather than assigning the responsibility for script development, ensured that we had a group of enthusiastic participants. In fact, several hospitalists volunteered to write additional scripts the following year.

This program used a conceptual framework of best practices, namely evidence‐based principles of effective faculty development for teaching in medical education.7 Different instructional methods were utilized: experiential learning was simulated by demonstrating scripts; the reasoning underlying scripts was provided; feedback was provided; and scripts were provided in written, electronic, and verbal formats. Allowing hospitalists to choose which script to develop gave them a chance to showcase an area of strength or explore an area of weakness, a feature of self‐directed learning. Focusing scripts on common diagnoses and easily identifiable triggers enhanced the functional value of the workshops. By having each hospitalist develop a script with input from each other, the unit built a body of knowledge and skill, enhancing collegiality and building a community of learners. Studies of other longitudinal faculty development programs have found that they create a supportive, learner‐centered environment that fosters a sense of commonality and interdisciplinary collegiality.8, 9

Other faculty development initiatives specific to hospitalists have been described, several focusing on the care of geriatric patients,1012 and one focusing on general academic development.13 While effective, these programs depended on a few individuals to develop the materials, and one required extensive time away from clinical duties for attendance.12 By sharing responsibility for developing teaching scripts, our program was efficient to conduct and capitalized on unique contributions from each faculty member.

This study has several limitations. While we attempted to quantify the amount and applicability of teaching, we were not able to account for the number of inpatients on the teams who had the diagnoses for which teaching scripts had been developed. It was impossible to determine whether these diagnoses were the most important topics to discuss on rounds. Because learning objectives were developed as each script was written, we were unable to assess changes in resident and student knowledge or patient outcomes. The study was conducted at a single center with interested faculty.

Future studies are needed to compare the effectiveness of collaborative teaching script development programs with other faculty development initiatives, and assess the impact on downstream outcomes, such as learners' decision‐making, patient outcomes, and faculty retention.

Patient complexity,1 productivity, and documentation pressures have increased substantially over the past 2 decades. Within this environment, time for teaching is often limited. The same pressures which limit faculty members' teaching time also limit their availability to learn how to teach; faculty development efforts need to be both effective and efficient.

In a seminal study of exemplary clinical teachers, Irby discovered that expert teachers often developed and utilized teaching scripts for commonly encountered teachable moments.2 Teaching scripts consist of a trigger, key teaching points, and teaching strategies.2 A trigger may be a specific clinical situation or a learner knowledge gap identified by the teacher. The trigger prompts the teacher to select key teaching points about the topic (the content), and utilize strategies for making these teaching points comprehensible (the process).2 Through a reflective process, these expert teachers evaluated the effectiveness of each teaching session and honed their scripts over time.2 While additional reports have described the use of teaching scripts,35 we found no studies evaluating the impact of collaboratively developing teaching scripts. In the present study, we sought to understand faculty members' early experiences with a program of collaboratively developing teaching scripts and the impact on their self‐efficacy with teaching about commonly encountered clinical conditions on attending rounds.

METHODS

Participants were the 22 internal medicine, or combined internal medicine and pediatrics (med‐peds), hospitalists in a 750‐bed university teaching hospital in upstate New York. Nine hospitalists worked for only 1 year (eg, chief residents and recent graduates awaiting fellowship training), and were present for half of the program year. All hospitalists conducted daily bedside attending rounds, lasting 1.52 hours, with a dual purpose of teaching the residents and students, and making management decisions for their shared patients.

Hospitalists were surveyed to identify 10 commonly encountered diagnoses about which they wanted to learn how to teach. The faculty development director (V.J.L.) conducted a 1‐hour workshop to introduce the concept of teaching scripts, and role‐play a teaching script. Nine hospitalists volunteered to write scripts for the remaining target diagnoses. They were provided with a template; example teaching script (see Supporting Information, Supplemental Content 1, in the online version of this article); and guidelines on writing scripts which highlighted effective clinical teaching principles for hospitalists, including: managing time with short scripts and high‐yield teaching points, knowledge acquisition with evidence‐based resources, self‐reflection/emnsight, patient‐centered teaching (identifying triggers among commonly encountered situations), and learner‐centered teaching (identifying common misconceptions and strategies for engaging all levels of learners) (Figure 1).2, 6 Faculty were encouraged to practice their scripts on attending rounds, using lessons learned to refine and write the script for presentation. Each script was presented verbally and on paper at a monthly 1‐hour interactive workshop where lunch was provided. Authors received feedback and incorporated suggestions for teaching strategies from the other hospitalists. Revised scripts were distributed electronically.

Figure 1

Baseline surveys measured prior teaching and faculty development experience, and self‐efficacy with teaching about the 10 target diagnoses, ranging from Not confident at all to Very confident on a 4‐point Likert scale. Using open‐ended surveys, we asked all of the hospitalists about their experiences with presenting scripts and participating in peer feedback, and the impact of the program on their teaching skills and patient care.

Because the learning objectives for each teaching script were determined by each script's author and were not known prior to the program, we were unable to assess changes in residents' and students' knowledge directly. As a surrogate measure, we surveyed students, residents, and faculty regarding how often the hospitalist taught about the 10 target diagnoses and whether teaching points were applicable to current or future patients. We administered the surveys online weekly for 8 weeks before and after the program. Residents and students were notified that participation had no impact on their evaluations. They received a $2.50 coffee gift card for each survey. The study received an exemption from the university's Institutional Review Board.

The number of teaching episodes per week related to the target diagnoses was averaged across survey weeks. Student t tests were used to compare results before versus after the intervention, and 95% confidence interval (CI) calculated. We considered P 0.05 to be statistically significant. Data were analyzed using SAS version 9.2 (Cary, NC).

Qualitative data were analyzed by coding each statement, then developing themes using an iterative process. Three investigators independently developed themes, and met twice to review the categorization of each statement until consensus was achieved. Two of the investigators were involved in the program (V.J.L. and A.B.) and one did not participate in the workshops (C.G.).

RESULTS

The 22 faculty had an average of 5 years' experience as hospitalists (range 0.824 years). Previous experience formally learning how to teach ranged from 0 to 150 hours (average 33.1 hours; median 15 hours). A mean of 9.4 hospitalists attended each of the 10 1‐hour workshops. Script writers estimated that scripts required a mean of 4.3 hours to prepare. A total of 105 (59%) resident/student and 22 (55%) faculty surveys were returned preintervention, and 83 (47%) resident/student and 19 (48%) faculty surveys were returned postintervention. There were no significant differences in the number or applicability of teaching events from before to after the program. Faculty self‐efficacy with teaching was available for 7 of the 10 diagnoses, and increased from a mean of 3.26 (n = 77) preintervention to 3.72 (n = 52) postintervention (95% CI for the difference in means 0.350.51; P 0.0001).

A total of 8 (80%) script‐writers and 5 (42%) non‐writers responded to the qualitative survey, and 77 comments were coded. Three major themes and 8 subthemes were identified (for representative comments, see Supporting Information, Supplemental Content 2, in the online version of this article). The major theme of individual professional development related especially to the personal satisfaction of researching a topic and becoming a local expert. While most comments were positive, 2 described apprehension about presenting to peers. Fifteen comments specifically addressed the development of teaching skills, 13 positive and 2 neutral. Some focused on strategies consistent with the teaching script framework, including recognizing teachable moments and the importance of preparation for teaching. Others focused on changes in teaching style, shifting to a more interactive method and involving multiple levels of learners. Others revealed that participants adjusted the content of their teaching, adding new material and changing the focus to important clinical pearls. Another subtheme was the impact on clinical care and medical knowledge base. Of the 11 comments, 7 were positive and emphasized the development of a framework for making decisions, based on an understanding of the evidence behind those decisions. Four were neutral, noting that care of patients had not changed. Two comments remarked on the time invested in developing teaching scripts. A second major theme was the development of a shared mental model of professional responsibility. This was demonstrated by comments relating to participants' motivation for learning, and development or strengthening of responsibility for teaching. The third major theme described interpersonal relationships among colleagues. Four commented on how the opportunity to see how others teach led them to appreciate the diversity of approaches, while 14 focused on collegiality among the faculty. Thirteen of these identified an increased sense of community and camaraderie, while one was neutral.

CONCLUSIONS

We had successful early experience with a faculty development intervention that involved hospitalists in creating and implementing teaching scripts related to commonly encountered diagnoses. The intervention was time‐ and resource‐efficient. Following the intervention, we found increased faculty self‐efficacy and beneficial effects in several domains related to professional development and satisfaction. We found no significant difference in the frequency or applicability of teaching about the targeted diagnoses.

In addition to the formal program evaluation results, we learned several additional lessons informally. Faculty who developed scripts had varying levels of familiarity with evidence‐based approaches to teaching. Some faculty requested to have their scripts reviewed by the program leader before presentation, and small revisions were made, emphasizing use of the tips included in Figure 1. Using volunteers, rather than assigning the responsibility for script development, ensured that we had a group of enthusiastic participants. In fact, several hospitalists volunteered to write additional scripts the following year.

This program used a conceptual framework of best practices, namely evidence‐based principles of effective faculty development for teaching in medical education.7 Different instructional methods were utilized: experiential learning was simulated by demonstrating scripts; the reasoning underlying scripts was provided; feedback was provided; and scripts were provided in written, electronic, and verbal formats. Allowing hospitalists to choose which script to develop gave them a chance to showcase an area of strength or explore an area of weakness, a feature of self‐directed learning. Focusing scripts on common diagnoses and easily identifiable triggers enhanced the functional value of the workshops. By having each hospitalist develop a script with input from each other, the unit built a body of knowledge and skill, enhancing collegiality and building a community of learners. Studies of other longitudinal faculty development programs have found that they create a supportive, learner‐centered environment that fosters a sense of commonality and interdisciplinary collegiality.8, 9

Other faculty development initiatives specific to hospitalists have been described, several focusing on the care of geriatric patients,1012 and one focusing on general academic development.13 While effective, these programs depended on a few individuals to develop the materials, and one required extensive time away from clinical duties for attendance.12 By sharing responsibility for developing teaching scripts, our program was efficient to conduct and capitalized on unique contributions from each faculty member.

This study has several limitations. While we attempted to quantify the amount and applicability of teaching, we were not able to account for the number of inpatients on the teams who had the diagnoses for which teaching scripts had been developed. It was impossible to determine whether these diagnoses were the most important topics to discuss on rounds. Because learning objectives were developed as each script was written, we were unable to assess changes in resident and student knowledge or patient outcomes. The study was conducted at a single center with interested faculty.

Future studies are needed to compare the effectiveness of collaborative teaching script development programs with other faculty development initiatives, and assess the impact on downstream outcomes, such as learners' decision‐making, patient outcomes, and faculty retention.

Patient complexity,1 productivity, and documentation pressures have increased substantially over the past 2 decades. Within this environment, time for teaching is often limited. The same pressures which limit faculty members' teaching time also limit their availability to learn how to teach; faculty development efforts need to be both effective and efficient.

In a seminal study of exemplary clinical teachers, Irby discovered that expert teachers often developed and utilized teaching scripts for commonly encountered teachable moments.2 Teaching scripts consist of a trigger, key teaching points, and teaching strategies.2 A trigger may be a specific clinical situation or a learner knowledge gap identified by the teacher. The trigger prompts the teacher to select key teaching points about the topic (the content), and utilize strategies for making these teaching points comprehensible (the process).2 Through a reflective process, these expert teachers evaluated the effectiveness of each teaching session and honed their scripts over time.2 While additional reports have described the use of teaching scripts,35 we found no studies evaluating the impact of collaboratively developing teaching scripts. In the present study, we sought to understand faculty members' early experiences with a program of collaboratively developing teaching scripts and the impact on their self‐efficacy with teaching about commonly encountered clinical conditions on attending rounds.

METHODS

Participants were the 22 internal medicine, or combined internal medicine and pediatrics (med‐peds), hospitalists in a 750‐bed university teaching hospital in upstate New York. Nine hospitalists worked for only 1 year (eg, chief residents and recent graduates awaiting fellowship training), and were present for half of the program year. All hospitalists conducted daily bedside attending rounds, lasting 1.52 hours, with a dual purpose of teaching the residents and students, and making management decisions for their shared patients.

Hospitalists were surveyed to identify 10 commonly encountered diagnoses about which they wanted to learn how to teach. The faculty development director (V.J.L.) conducted a 1‐hour workshop to introduce the concept of teaching scripts, and role‐play a teaching script. Nine hospitalists volunteered to write scripts for the remaining target diagnoses. They were provided with a template; example teaching script (see Supporting Information, Supplemental Content 1, in the online version of this article); and guidelines on writing scripts which highlighted effective clinical teaching principles for hospitalists, including: managing time with short scripts and high‐yield teaching points, knowledge acquisition with evidence‐based resources, self‐reflection/emnsight, patient‐centered teaching (identifying triggers among commonly encountered situations), and learner‐centered teaching (identifying common misconceptions and strategies for engaging all levels of learners) (Figure 1).2, 6 Faculty were encouraged to practice their scripts on attending rounds, using lessons learned to refine and write the script for presentation. Each script was presented verbally and on paper at a monthly 1‐hour interactive workshop where lunch was provided. Authors received feedback and incorporated suggestions for teaching strategies from the other hospitalists. Revised scripts were distributed electronically.

Figure 1

Baseline surveys measured prior teaching and faculty development experience, and self‐efficacy with teaching about the 10 target diagnoses, ranging from Not confident at all to Very confident on a 4‐point Likert scale. Using open‐ended surveys, we asked all of the hospitalists about their experiences with presenting scripts and participating in peer feedback, and the impact of the program on their teaching skills and patient care.

Because the learning objectives for each teaching script were determined by each script's author and were not known prior to the program, we were unable to assess changes in residents' and students' knowledge directly. As a surrogate measure, we surveyed students, residents, and faculty regarding how often the hospitalist taught about the 10 target diagnoses and whether teaching points were applicable to current or future patients. We administered the surveys online weekly for 8 weeks before and after the program. Residents and students were notified that participation had no impact on their evaluations. They received a $2.50 coffee gift card for each survey. The study received an exemption from the university's Institutional Review Board.

The number of teaching episodes per week related to the target diagnoses was averaged across survey weeks. Student t tests were used to compare results before versus after the intervention, and 95% confidence interval (CI) calculated. We considered P 0.05 to be statistically significant. Data were analyzed using SAS version 9.2 (Cary, NC).

Qualitative data were analyzed by coding each statement, then developing themes using an iterative process. Three investigators independently developed themes, and met twice to review the categorization of each statement until consensus was achieved. Two of the investigators were involved in the program (V.J.L. and A.B.) and one did not participate in the workshops (C.G.).

RESULTS

The 22 faculty had an average of 5 years' experience as hospitalists (range 0.824 years). Previous experience formally learning how to teach ranged from 0 to 150 hours (average 33.1 hours; median 15 hours). A mean of 9.4 hospitalists attended each of the 10 1‐hour workshops. Script writers estimated that scripts required a mean of 4.3 hours to prepare. A total of 105 (59%) resident/student and 22 (55%) faculty surveys were returned preintervention, and 83 (47%) resident/student and 19 (48%) faculty surveys were returned postintervention. There were no significant differences in the number or applicability of teaching events from before to after the program. Faculty self‐efficacy with teaching was available for 7 of the 10 diagnoses, and increased from a mean of 3.26 (n = 77) preintervention to 3.72 (n = 52) postintervention (95% CI for the difference in means 0.350.51; P 0.0001).

A total of 8 (80%) script‐writers and 5 (42%) non‐writers responded to the qualitative survey, and 77 comments were coded. Three major themes and 8 subthemes were identified (for representative comments, see Supporting Information, Supplemental Content 2, in the online version of this article). The major theme of individual professional development related especially to the personal satisfaction of researching a topic and becoming a local expert. While most comments were positive, 2 described apprehension about presenting to peers. Fifteen comments specifically addressed the development of teaching skills, 13 positive and 2 neutral. Some focused on strategies consistent with the teaching script framework, including recognizing teachable moments and the importance of preparation for teaching. Others focused on changes in teaching style, shifting to a more interactive method and involving multiple levels of learners. Others revealed that participants adjusted the content of their teaching, adding new material and changing the focus to important clinical pearls. Another subtheme was the impact on clinical care and medical knowledge base. Of the 11 comments, 7 were positive and emphasized the development of a framework for making decisions, based on an understanding of the evidence behind those decisions. Four were neutral, noting that care of patients had not changed. Two comments remarked on the time invested in developing teaching scripts. A second major theme was the development of a shared mental model of professional responsibility. This was demonstrated by comments relating to participants' motivation for learning, and development or strengthening of responsibility for teaching. The third major theme described interpersonal relationships among colleagues. Four commented on how the opportunity to see how others teach led them to appreciate the diversity of approaches, while 14 focused on collegiality among the faculty. Thirteen of these identified an increased sense of community and camaraderie, while one was neutral.

CONCLUSIONS

We had successful early experience with a faculty development intervention that involved hospitalists in creating and implementing teaching scripts related to commonly encountered diagnoses. The intervention was time‐ and resource‐efficient. Following the intervention, we found increased faculty self‐efficacy and beneficial effects in several domains related to professional development and satisfaction. We found no significant difference in the frequency or applicability of teaching about the targeted diagnoses.

In addition to the formal program evaluation results, we learned several additional lessons informally. Faculty who developed scripts had varying levels of familiarity with evidence‐based approaches to teaching. Some faculty requested to have their scripts reviewed by the program leader before presentation, and small revisions were made, emphasizing use of the tips included in Figure 1. Using volunteers, rather than assigning the responsibility for script development, ensured that we had a group of enthusiastic participants. In fact, several hospitalists volunteered to write additional scripts the following year.

This program used a conceptual framework of best practices, namely evidence‐based principles of effective faculty development for teaching in medical education.7 Different instructional methods were utilized: experiential learning was simulated by demonstrating scripts; the reasoning underlying scripts was provided; feedback was provided; and scripts were provided in written, electronic, and verbal formats. Allowing hospitalists to choose which script to develop gave them a chance to showcase an area of strength or explore an area of weakness, a feature of self‐directed learning. Focusing scripts on common diagnoses and easily identifiable triggers enhanced the functional value of the workshops. By having each hospitalist develop a script with input from each other, the unit built a body of knowledge and skill, enhancing collegiality and building a community of learners. Studies of other longitudinal faculty development programs have found that they create a supportive, learner‐centered environment that fosters a sense of commonality and interdisciplinary collegiality.8, 9

Other faculty development initiatives specific to hospitalists have been described, several focusing on the care of geriatric patients,1012 and one focusing on general academic development.13 While effective, these programs depended on a few individuals to develop the materials, and one required extensive time away from clinical duties for attendance.12 By sharing responsibility for developing teaching scripts, our program was efficient to conduct and capitalized on unique contributions from each faculty member.

This study has several limitations. While we attempted to quantify the amount and applicability of teaching, we were not able to account for the number of inpatients on the teams who had the diagnoses for which teaching scripts had been developed. It was impossible to determine whether these diagnoses were the most important topics to discuss on rounds. Because learning objectives were developed as each script was written, we were unable to assess changes in resident and student knowledge or patient outcomes. The study was conducted at a single center with interested faculty.

Future studies are needed to compare the effectiveness of collaborative teaching script development programs with other faculty development initiatives, and assess the impact on downstream outcomes, such as learners' decision‐making, patient outcomes, and faculty retention.

Perioperative myocardial infarction (PMI) often remains unrecognized with higher mortality in the aged.13 Perioperative ischemic symptoms are often masked by analgesia, sedation, and transient and subtle electrocardiographic (ECG) changes. Postoperative troponin measurement is not routinely done for PMI diagnosis. Hip fracture surgery is the most common non‐cardiac surgical procedure in the elderly, with limited data on clinical presentation of PMI.46 Moreover, the elderly are significantly underrepresented in clinical studies.7 We therefore examined the clinical presentation of PMI and its outcomes among elderly patients admitted for hip fracture repair.

METHODS

RESULTS

In the cohort of 1212 with hip fracture surgeries, 167 (13.8%) cases of PMI occurred in the first 7 days, of which 153 (92%) occurred within the first 48 hours. A total of 334 controls were matched with 167 cases of PMI. Table 1 summarizes the demographic characteristics of the study participants. Of the patients with PMI, 25.2% experienced symptoms of ischemia; 7% reported chest pain, and 12% reported dyspnea. Only 22.8% of patients with PMI had ECG changes consistent with ischemia. ST elevation MI was present in 7.2% patients. PMI patients had a lower mean hemoglobin compared to the patients without PMI (8.9 mg/dL vs 9.4 mg/dL, P 0.001). Median length of stay (LOS) in the hospital was higher among patients who experienced PMI (11.6 vs 7.4 days, P 0.001). Overall in‐hospital mortality was 5.6%. There were 24 deaths (14.4%) in the PMI group compared to 4 (1.2%) in‐hospital deaths in patients without PMI (P 0.001). A total of 473 (94%) patients survived to discharge. At 30‐day follow‐up, there were 29 (17.4%) deaths in the PMI group and 14 (4.2%) deaths in non‐PMI group. During the follow‐up for 1 year, there were 143 (29%) deaths: PMI 66 (39.5%) and 77 (23%) non‐PMI group (P 0.01).

Table 2 describes the risk factors associated with PMI in‐hospital, 30‐day, and 1‐year mortality. Risk factors for PMI were coronary artery disease (CAD) (odds ratio [OR], 3.5; confidence interval [CI], 2.25.6), and serum creatinine >2 mg/dL (OR, 2.4; CI, 1.34.4). Risk factors for in‐hospital mortality were age 8589 (OR, 5.3; CI, 1.617.7), age 90 (OR, 8.9; CI, 2.630.8), PMI (OR 15.1; CI, 4.648.8), male gender (OR 5.8; CI, 2.215.2), dyspnea (OR 5.4; CI, 1.816.9), and hemoglobin 8.0 gm/dL (OR, 3.5; CI, 1.29.9). PMI was a strong predictor for 30‐day mortality (hazard ratio [HR], 4.3; CI, 2.18.9). Risk factors for 1‐year mortality were: age 90 (HR, 2.0; CI, 1.43.1), male gender (HR, 2.1; CI, 1.53.0), and PMI (HR, 1.9; CI, 1.42.7). Figures 1 and 2 describe the Kaplan‐Meier survival curves for patients with and without PMI.

Figure 1

Figure 2

DISCUSSION

We report the high incidence of PMI (13.8%) in the cohort of 1212 elderly patients (mean age 85 years) undergoing hip fracture surgery. Most PMI events (92%) occurred within the first 48 hours of surgery. Most of the events (75%) were asymptomatic. Elderly patients with PMI had an increased hospital LOS by 4.2 days, with high in‐hospital mortality (13.8%), 30‐day mortality (17.4%), and 1‐year mortality (39.5%).

Most of the PMI patients were identified with cardiac biomarkers on the basis of universal definition of MI within the first 48 hours. Although universal definition of MI does not define PMI as a separate type, PMI shares common pathophysiological pathways of Type 1 MI (primary coronary event) and Type 2 MI (myocardial oxygen supplydemand imbalance). Postoperative tachycardia, hemodynamic instability, anemia, and hypoxemia may initiate pathways causing more Type 2 MI. Our study highlights the continued need for active surveillance of clinical symptoms, postoperative ECG monitoring for STT changes, and utilizing cardiac troponin in older postoperative patients to improve diagnostic accuracy of PMI.

The current study has higher asymptomatic PMI events when compared to a study of Devereaux et al.11 The current study had an older population undergoing urgent hip fracture surgery, with a higher burden of CAD (60%) and renal failure (20%) with serum creatinine >2 gm/dL (see Supporting Information, Appendix 1, in the online version of this article). Older age and a higher burden of these risk factors may explain the higher incidence of PMI in the current study. Perioperative liberal use of analgesics in hip fracture surgery may explain more asymptomatic patients.

In light of the recently published FOCUS12 trial, an important finding from our study is that postoperative anemia among elderly (8.0 gm/dL) is associated with a 3.5‐fold increased in‐hospital mortality. It is critical to maintain perioperative hemoglobin above 8.0 gm/dL in very elderly patients, due to asymptomatic presentation of PMI.

In the current study, PMI is associated with a 15‐fold increased risk of in‐hospital death and a 4.3‐fold increased risk of 30‐day mortality in the elderly. Advanced age (85 years) is a well known strong predictor of initial hospital admission and death in elderly patients after outpatient surgery.13 Furthermore, the odds for an in‐hospital death increase by 70% for each 10‐year increase in age.14 Therefore, early detection of silent PMI among at‐risk elderly patients by cardiac biomarkers may help in optimization of cardiac pharmacotherapy known to decrease short‐ and long‐term mortality.

There are limitations inherent to the retrospective design and methodology. Data collection was done through the year 2002. CK was used for the period that spans from 1988 to mid‐2000. Troponin was used from 2000 to 2002. Statin use was not analyzed for lack of significant data. Limited use of beta‐blockers (15%) and angiotensin‐converting‐enzyme (ACE) inhibitors (25%) may also contribute to higher events (see Supporting Information, Appendix 1, in the online version of this article).

CONCLUSIONS

Elderly patients have a higher incidence of PMI and mortality after hip fracture surgery than what guidelines indicate. The majority of the elderly patients with PMI did not experience ischemic symptoms and required cardiac biomarkers for diagnosis. The results of our study support the measurement of troponin in postoperative elderly patients for the diagnosis of PMI to implement in‐hospital preventive strategies to reduce PMI‐associated mortality.

Perioperative myocardial infarction (PMI) often remains unrecognized with higher mortality in the aged.13 Perioperative ischemic symptoms are often masked by analgesia, sedation, and transient and subtle electrocardiographic (ECG) changes. Postoperative troponin measurement is not routinely done for PMI diagnosis. Hip fracture surgery is the most common non‐cardiac surgical procedure in the elderly, with limited data on clinical presentation of PMI.46 Moreover, the elderly are significantly underrepresented in clinical studies.7 We therefore examined the clinical presentation of PMI and its outcomes among elderly patients admitted for hip fracture repair.

METHODS

RESULTS

In the cohort of 1212 with hip fracture surgeries, 167 (13.8%) cases of PMI occurred in the first 7 days, of which 153 (92%) occurred within the first 48 hours. A total of 334 controls were matched with 167 cases of PMI. Table 1 summarizes the demographic characteristics of the study participants. Of the patients with PMI, 25.2% experienced symptoms of ischemia; 7% reported chest pain, and 12% reported dyspnea. Only 22.8% of patients with PMI had ECG changes consistent with ischemia. ST elevation MI was present in 7.2% patients. PMI patients had a lower mean hemoglobin compared to the patients without PMI (8.9 mg/dL vs 9.4 mg/dL, P 0.001). Median length of stay (LOS) in the hospital was higher among patients who experienced PMI (11.6 vs 7.4 days, P 0.001). Overall in‐hospital mortality was 5.6%. There were 24 deaths (14.4%) in the PMI group compared to 4 (1.2%) in‐hospital deaths in patients without PMI (P 0.001). A total of 473 (94%) patients survived to discharge. At 30‐day follow‐up, there were 29 (17.4%) deaths in the PMI group and 14 (4.2%) deaths in non‐PMI group. During the follow‐up for 1 year, there were 143 (29%) deaths: PMI 66 (39.5%) and 77 (23%) non‐PMI group (P 0.01).

Table 2 describes the risk factors associated with PMI in‐hospital, 30‐day, and 1‐year mortality. Risk factors for PMI were coronary artery disease (CAD) (odds ratio [OR], 3.5; confidence interval [CI], 2.25.6), and serum creatinine >2 mg/dL (OR, 2.4; CI, 1.34.4). Risk factors for in‐hospital mortality were age 8589 (OR, 5.3; CI, 1.617.7), age 90 (OR, 8.9; CI, 2.630.8), PMI (OR 15.1; CI, 4.648.8), male gender (OR 5.8; CI, 2.215.2), dyspnea (OR 5.4; CI, 1.816.9), and hemoglobin 8.0 gm/dL (OR, 3.5; CI, 1.29.9). PMI was a strong predictor for 30‐day mortality (hazard ratio [HR], 4.3; CI, 2.18.9). Risk factors for 1‐year mortality were: age 90 (HR, 2.0; CI, 1.43.1), male gender (HR, 2.1; CI, 1.53.0), and PMI (HR, 1.9; CI, 1.42.7). Figures 1 and 2 describe the Kaplan‐Meier survival curves for patients with and without PMI.

Figure 1

Figure 2

DISCUSSION

We report the high incidence of PMI (13.8%) in the cohort of 1212 elderly patients (mean age 85 years) undergoing hip fracture surgery. Most PMI events (92%) occurred within the first 48 hours of surgery. Most of the events (75%) were asymptomatic. Elderly patients with PMI had an increased hospital LOS by 4.2 days, with high in‐hospital mortality (13.8%), 30‐day mortality (17.4%), and 1‐year mortality (39.5%).

Most of the PMI patients were identified with cardiac biomarkers on the basis of universal definition of MI within the first 48 hours. Although universal definition of MI does not define PMI as a separate type, PMI shares common pathophysiological pathways of Type 1 MI (primary coronary event) and Type 2 MI (myocardial oxygen supplydemand imbalance). Postoperative tachycardia, hemodynamic instability, anemia, and hypoxemia may initiate pathways causing more Type 2 MI. Our study highlights the continued need for active surveillance of clinical symptoms, postoperative ECG monitoring for STT changes, and utilizing cardiac troponin in older postoperative patients to improve diagnostic accuracy of PMI.

The current study has higher asymptomatic PMI events when compared to a study of Devereaux et al.11 The current study had an older population undergoing urgent hip fracture surgery, with a higher burden of CAD (60%) and renal failure (20%) with serum creatinine >2 gm/dL (see Supporting Information, Appendix 1, in the online version of this article). Older age and a higher burden of these risk factors may explain the higher incidence of PMI in the current study. Perioperative liberal use of analgesics in hip fracture surgery may explain more asymptomatic patients.

In light of the recently published FOCUS12 trial, an important finding from our study is that postoperative anemia among elderly (8.0 gm/dL) is associated with a 3.5‐fold increased in‐hospital mortality. It is critical to maintain perioperative hemoglobin above 8.0 gm/dL in very elderly patients, due to asymptomatic presentation of PMI.

In the current study, PMI is associated with a 15‐fold increased risk of in‐hospital death and a 4.3‐fold increased risk of 30‐day mortality in the elderly. Advanced age (85 years) is a well known strong predictor of initial hospital admission and death in elderly patients after outpatient surgery.13 Furthermore, the odds for an in‐hospital death increase by 70% for each 10‐year increase in age.14 Therefore, early detection of silent PMI among at‐risk elderly patients by cardiac biomarkers may help in optimization of cardiac pharmacotherapy known to decrease short‐ and long‐term mortality.

There are limitations inherent to the retrospective design and methodology. Data collection was done through the year 2002. CK was used for the period that spans from 1988 to mid‐2000. Troponin was used from 2000 to 2002. Statin use was not analyzed for lack of significant data. Limited use of beta‐blockers (15%) and angiotensin‐converting‐enzyme (ACE) inhibitors (25%) may also contribute to higher events (see Supporting Information, Appendix 1, in the online version of this article).

CONCLUSIONS

Elderly patients have a higher incidence of PMI and mortality after hip fracture surgery than what guidelines indicate. The majority of the elderly patients with PMI did not experience ischemic symptoms and required cardiac biomarkers for diagnosis. The results of our study support the measurement of troponin in postoperative elderly patients for the diagnosis of PMI to implement in‐hospital preventive strategies to reduce PMI‐associated mortality.

Perioperative myocardial infarction (PMI) often remains unrecognized with higher mortality in the aged.13 Perioperative ischemic symptoms are often masked by analgesia, sedation, and transient and subtle electrocardiographic (ECG) changes. Postoperative troponin measurement is not routinely done for PMI diagnosis. Hip fracture surgery is the most common non‐cardiac surgical procedure in the elderly, with limited data on clinical presentation of PMI.46 Moreover, the elderly are significantly underrepresented in clinical studies.7 We therefore examined the clinical presentation of PMI and its outcomes among elderly patients admitted for hip fracture repair.

METHODS

RESULTS

In the cohort of 1212 with hip fracture surgeries, 167 (13.8%) cases of PMI occurred in the first 7 days, of which 153 (92%) occurred within the first 48 hours. A total of 334 controls were matched with 167 cases of PMI. Table 1 summarizes the demographic characteristics of the study participants. Of the patients with PMI, 25.2% experienced symptoms of ischemia; 7% reported chest pain, and 12% reported dyspnea. Only 22.8% of patients with PMI had ECG changes consistent with ischemia. ST elevation MI was present in 7.2% patients. PMI patients had a lower mean hemoglobin compared to the patients without PMI (8.9 mg/dL vs 9.4 mg/dL, P 0.001). Median length of stay (LOS) in the hospital was higher among patients who experienced PMI (11.6 vs 7.4 days, P 0.001). Overall in‐hospital mortality was 5.6%. There were 24 deaths (14.4%) in the PMI group compared to 4 (1.2%) in‐hospital deaths in patients without PMI (P 0.001). A total of 473 (94%) patients survived to discharge. At 30‐day follow‐up, there were 29 (17.4%) deaths in the PMI group and 14 (4.2%) deaths in non‐PMI group. During the follow‐up for 1 year, there were 143 (29%) deaths: PMI 66 (39.5%) and 77 (23%) non‐PMI group (P 0.01).

Table 2 describes the risk factors associated with PMI in‐hospital, 30‐day, and 1‐year mortality. Risk factors for PMI were coronary artery disease (CAD) (odds ratio [OR], 3.5; confidence interval [CI], 2.25.6), and serum creatinine >2 mg/dL (OR, 2.4; CI, 1.34.4). Risk factors for in‐hospital mortality were age 8589 (OR, 5.3; CI, 1.617.7), age 90 (OR, 8.9; CI, 2.630.8), PMI (OR 15.1; CI, 4.648.8), male gender (OR 5.8; CI, 2.215.2), dyspnea (OR 5.4; CI, 1.816.9), and hemoglobin 8.0 gm/dL (OR, 3.5; CI, 1.29.9). PMI was a strong predictor for 30‐day mortality (hazard ratio [HR], 4.3; CI, 2.18.9). Risk factors for 1‐year mortality were: age 90 (HR, 2.0; CI, 1.43.1), male gender (HR, 2.1; CI, 1.53.0), and PMI (HR, 1.9; CI, 1.42.7). Figures 1 and 2 describe the Kaplan‐Meier survival curves for patients with and without PMI.

Figure 1

Figure 2

DISCUSSION

We report the high incidence of PMI (13.8%) in the cohort of 1212 elderly patients (mean age 85 years) undergoing hip fracture surgery. Most PMI events (92%) occurred within the first 48 hours of surgery. Most of the events (75%) were asymptomatic. Elderly patients with PMI had an increased hospital LOS by 4.2 days, with high in‐hospital mortality (13.8%), 30‐day mortality (17.4%), and 1‐year mortality (39.5%).

Most of the PMI patients were identified with cardiac biomarkers on the basis of universal definition of MI within the first 48 hours. Although universal definition of MI does not define PMI as a separate type, PMI shares common pathophysiological pathways of Type 1 MI (primary coronary event) and Type 2 MI (myocardial oxygen supplydemand imbalance). Postoperative tachycardia, hemodynamic instability, anemia, and hypoxemia may initiate pathways causing more Type 2 MI. Our study highlights the continued need for active surveillance of clinical symptoms, postoperative ECG monitoring for STT changes, and utilizing cardiac troponin in older postoperative patients to improve diagnostic accuracy of PMI.

The current study has higher asymptomatic PMI events when compared to a study of Devereaux et al.11 The current study had an older population undergoing urgent hip fracture surgery, with a higher burden of CAD (60%) and renal failure (20%) with serum creatinine >2 gm/dL (see Supporting Information, Appendix 1, in the online version of this article). Older age and a higher burden of these risk factors may explain the higher incidence of PMI in the current study. Perioperative liberal use of analgesics in hip fracture surgery may explain more asymptomatic patients.

In light of the recently published FOCUS12 trial, an important finding from our study is that postoperative anemia among elderly (8.0 gm/dL) is associated with a 3.5‐fold increased in‐hospital mortality. It is critical to maintain perioperative hemoglobin above 8.0 gm/dL in very elderly patients, due to asymptomatic presentation of PMI.

In the current study, PMI is associated with a 15‐fold increased risk of in‐hospital death and a 4.3‐fold increased risk of 30‐day mortality in the elderly. Advanced age (85 years) is a well known strong predictor of initial hospital admission and death in elderly patients after outpatient surgery.13 Furthermore, the odds for an in‐hospital death increase by 70% for each 10‐year increase in age.14 Therefore, early detection of silent PMI among at‐risk elderly patients by cardiac biomarkers may help in optimization of cardiac pharmacotherapy known to decrease short‐ and long‐term mortality.

There are limitations inherent to the retrospective design and methodology. Data collection was done through the year 2002. CK was used for the period that spans from 1988 to mid‐2000. Troponin was used from 2000 to 2002. Statin use was not analyzed for lack of significant data. Limited use of beta‐blockers (15%) and angiotensin‐converting‐enzyme (ACE) inhibitors (25%) may also contribute to higher events (see Supporting Information, Appendix 1, in the online version of this article).

CONCLUSIONS

Elderly patients have a higher incidence of PMI and mortality after hip fracture surgery than what guidelines indicate. The majority of the elderly patients with PMI did not experience ischemic symptoms and required cardiac biomarkers for diagnosis. The results of our study support the measurement of troponin in postoperative elderly patients for the diagnosis of PMI to implement in‐hospital preventive strategies to reduce PMI‐associated mortality.

In 2003, the Accreditation Council for Graduate Medical Education (ACGME) announced the first in a series of guidelines related to the regulation and oversight of residency training.1 The initial iteration specifically focused on the total and consecutive numbers of duty hours worked by trainees. These limitations began a new era of shift work in internal medicine residency training. With decreases in housestaff admitting capacity, clinical work has frequently been offloaded to non‐teaching or attending‐only services, increasing the demand for hospitalists to fill the void in physician‐staffed care in the hospital.2, 3 Since the implementation of the 2003 ACGME guidelines and a growing focus on patient safety, there has been increased study of, and call for, oversight of trainees in medicine; among these was the 2008 Institute of Medicine report,4 calling for 24/7 attending‐level supervision. The updated ACGME requirements,5 effective July 1, 2011, mandate enhanced on‐site supervision of trainee physicians. These new regulations not only define varying levels of supervision for trainees, including direct supervision with the physical presence of a supervisor and the degree of availability of said supervisor, they also describe ensuring the quality of supervision provided.5 While continuous attending‐level supervision is not yet mandated, many residency programs look to their academic hospitalists to fill the supervisory void, particularly at night. However, what specific roles hospitalists play in the nighttime supervision of trainees or the impact of this supervision remains unclear. To date, no study has examined a broad sample of hospitalist programs in teaching hospitals and the types of resident oversight they provide. We aimed to describe the current state of academic hospitalists in the clinical supervision of housestaff, specifically during the overnight period, and hospitalist perceptions of how the new ACGME requirements would impact traineehospitalist interactions.

METHODS

The Housestaff Oversight Subcommittee, a working group of the Society of General Internal Medicine (SGIM) Academic Hospitalist Task Force, surveyed a sample of academic hospitalist program leaders to assess the current status of trainee supervision performed by hospitalists. Programs were considered academic if they were located in the primary hospital of a residency that participates in the National Resident Matching Program for Internal Medicine. To obtain a broad geographic spectrum of academic hospitalist programs, all programs, both university and community‐based, in 4 states and 2 metropolitan regions were sampled: Washington, Oregon, Texas, Maryland, and the Philadelphia and Chicago metropolitan areas. Hospitalist program leaders were identified by members of the Taskforce using individual program websites and by querying departmental leadership at eligible teaching hospitals. Respondents were contacted by e‐mail for participation. None of the authors of the manuscript were participants in the survey.

The survey was developed by consensus of the working group after reviewing the salient literature and included additional questions queried to internal medicine program directors.6 The 19‐item SurveyMonkey instrument included questions about hospitalists' role in trainees' education and evaluation. A Likert‐type scale was used to assess perceptions regarding the impact of on‐site hospitalist supervision on trainee autonomy and hospitalist workload (1 = strongly disagree to 5 = strongly agree). Descriptive statistics were performed and, where appropriate, t test and Fisher's exact test were performed to identify associations between program characteristics and perceptions. Stata SE was used (STATA Corp, College Station, TX) for all statistical analysis.

RESULTS

The survey was sent to 47 individuals identified as likely hospitalist program leaders and completed by 41 individuals (87%). However, 7 respondents turned out not to be program leaders and were therefore excluded, resulting in a 72% (34/47) survey response rate.

The programs for which we did not obtain responses were similar to respondent programs, and did not include a larger proportion of community‐based programs or overrepresent a specific geographic region. Twenty‐five (73%) of the 34 hospitalist program leaders were male, with an average age of 44.3 years, and an average of 12 years post‐residency training (range, 530 years). They reported leading groups with an average of 18 full‐time equivalent (FTE) faculty (range, 350 persons).

DISCUSSION

This study provides important information about how academic hospitalists currently contribute to the supervision of internal medicine residents. While academic hospitalist groups frequently have faculty providing clinical care on‐site at night, and often hospitalists provide overnight supervision of internal medicine trainees, formal supervision of trainees is not uniform, and few hospitalists groups have a mechanism to provide training or faculty development on how to effectively supervise resident trainees. Hospitalist leaders expressed concerns that creating additional formal overnight supervisory responsibilities may add to an already burdened overnight hospitalist. Formalizing this supervisory role, including explicit role definitions and faculty training for trainee supervision, is necessary.

Though our sample size is small, we captured a diverse geographic range of both university and community‐based academic hospitalist programs by surveying group leaders in several distinct regions. We are unable to comment on differences between responding and non‐responding hospitalist programs, but there does not appear to be a systematic difference between these groups.

Our findings are consistent with work describing a lack of structured conceptual frameworks in effectively supervising trainees,7, 8 and also, at times, nebulous expectations for hospitalist faculty. We found that the existence of a formal supervisory policy within the associated residency program, as well as defined roles for hospitalists, increases the likelihood of positive perceptions of the new ACGME supervisory recommendations. However, the existence of these requirements does not mean that all programs are capable of following them. While additional discussion is required to best delineate a formal overnight hospitalist role in trainee supervision, clearly defining expectations for both faculty and trainees, and their interactions, may alleviate the struggles that exist in programs with ill‐defined roles for hospitalist faculty supervision. While faculty duty hours standards do not exist, additional duties of nighttime coverage for hospitalists suggests that close attention should be paid to burn‐out.9 Faculty development on nighttime supervision and teaching may help maximize both learning and patient care efficiency, and provide a framework for this often unstructured educational time.

In 2003, the Accreditation Council for Graduate Medical Education (ACGME) announced the first in a series of guidelines related to the regulation and oversight of residency training.1 The initial iteration specifically focused on the total and consecutive numbers of duty hours worked by trainees. These limitations began a new era of shift work in internal medicine residency training. With decreases in housestaff admitting capacity, clinical work has frequently been offloaded to non‐teaching or attending‐only services, increasing the demand for hospitalists to fill the void in physician‐staffed care in the hospital.2, 3 Since the implementation of the 2003 ACGME guidelines and a growing focus on patient safety, there has been increased study of, and call for, oversight of trainees in medicine; among these was the 2008 Institute of Medicine report,4 calling for 24/7 attending‐level supervision. The updated ACGME requirements,5 effective July 1, 2011, mandate enhanced on‐site supervision of trainee physicians. These new regulations not only define varying levels of supervision for trainees, including direct supervision with the physical presence of a supervisor and the degree of availability of said supervisor, they also describe ensuring the quality of supervision provided.5 While continuous attending‐level supervision is not yet mandated, many residency programs look to their academic hospitalists to fill the supervisory void, particularly at night. However, what specific roles hospitalists play in the nighttime supervision of trainees or the impact of this supervision remains unclear. To date, no study has examined a broad sample of hospitalist programs in teaching hospitals and the types of resident oversight they provide. We aimed to describe the current state of academic hospitalists in the clinical supervision of housestaff, specifically during the overnight period, and hospitalist perceptions of how the new ACGME requirements would impact traineehospitalist interactions.

METHODS

The Housestaff Oversight Subcommittee, a working group of the Society of General Internal Medicine (SGIM) Academic Hospitalist Task Force, surveyed a sample of academic hospitalist program leaders to assess the current status of trainee supervision performed by hospitalists. Programs were considered academic if they were located in the primary hospital of a residency that participates in the National Resident Matching Program for Internal Medicine. To obtain a broad geographic spectrum of academic hospitalist programs, all programs, both university and community‐based, in 4 states and 2 metropolitan regions were sampled: Washington, Oregon, Texas, Maryland, and the Philadelphia and Chicago metropolitan areas. Hospitalist program leaders were identified by members of the Taskforce using individual program websites and by querying departmental leadership at eligible teaching hospitals. Respondents were contacted by e‐mail for participation. None of the authors of the manuscript were participants in the survey.

The survey was developed by consensus of the working group after reviewing the salient literature and included additional questions queried to internal medicine program directors.6 The 19‐item SurveyMonkey instrument included questions about hospitalists' role in trainees' education and evaluation. A Likert‐type scale was used to assess perceptions regarding the impact of on‐site hospitalist supervision on trainee autonomy and hospitalist workload (1 = strongly disagree to 5 = strongly agree). Descriptive statistics were performed and, where appropriate, t test and Fisher's exact test were performed to identify associations between program characteristics and perceptions. Stata SE was used (STATA Corp, College Station, TX) for all statistical analysis.

RESULTS

The survey was sent to 47 individuals identified as likely hospitalist program leaders and completed by 41 individuals (87%). However, 7 respondents turned out not to be program leaders and were therefore excluded, resulting in a 72% (34/47) survey response rate.

The programs for which we did not obtain responses were similar to respondent programs, and did not include a larger proportion of community‐based programs or overrepresent a specific geographic region. Twenty‐five (73%) of the 34 hospitalist program leaders were male, with an average age of 44.3 years, and an average of 12 years post‐residency training (range, 530 years). They reported leading groups with an average of 18 full‐time equivalent (FTE) faculty (range, 350 persons).

DISCUSSION

This study provides important information about how academic hospitalists currently contribute to the supervision of internal medicine residents. While academic hospitalist groups frequently have faculty providing clinical care on‐site at night, and often hospitalists provide overnight supervision of internal medicine trainees, formal supervision of trainees is not uniform, and few hospitalists groups have a mechanism to provide training or faculty development on how to effectively supervise resident trainees. Hospitalist leaders expressed concerns that creating additional formal overnight supervisory responsibilities may add to an already burdened overnight hospitalist. Formalizing this supervisory role, including explicit role definitions and faculty training for trainee supervision, is necessary.

Though our sample size is small, we captured a diverse geographic range of both university and community‐based academic hospitalist programs by surveying group leaders in several distinct regions. We are unable to comment on differences between responding and non‐responding hospitalist programs, but there does not appear to be a systematic difference between these groups.

Our findings are consistent with work describing a lack of structured conceptual frameworks in effectively supervising trainees,7, 8 and also, at times, nebulous expectations for hospitalist faculty. We found that the existence of a formal supervisory policy within the associated residency program, as well as defined roles for hospitalists, increases the likelihood of positive perceptions of the new ACGME supervisory recommendations. However, the existence of these requirements does not mean that all programs are capable of following them. While additional discussion is required to best delineate a formal overnight hospitalist role in trainee supervision, clearly defining expectations for both faculty and trainees, and their interactions, may alleviate the struggles that exist in programs with ill‐defined roles for hospitalist faculty supervision. While faculty duty hours standards do not exist, additional duties of nighttime coverage for hospitalists suggests that close attention should be paid to burn‐out.9 Faculty development on nighttime supervision and teaching may help maximize both learning and patient care efficiency, and provide a framework for this often unstructured educational time.

In 2003, the Accreditation Council for Graduate Medical Education (ACGME) announced the first in a series of guidelines related to the regulation and oversight of residency training.1 The initial iteration specifically focused on the total and consecutive numbers of duty hours worked by trainees. These limitations began a new era of shift work in internal medicine residency training. With decreases in housestaff admitting capacity, clinical work has frequently been offloaded to non‐teaching or attending‐only services, increasing the demand for hospitalists to fill the void in physician‐staffed care in the hospital.2, 3 Since the implementation of the 2003 ACGME guidelines and a growing focus on patient safety, there has been increased study of, and call for, oversight of trainees in medicine; among these was the 2008 Institute of Medicine report,4 calling for 24/7 attending‐level supervision. The updated ACGME requirements,5 effective July 1, 2011, mandate enhanced on‐site supervision of trainee physicians. These new regulations not only define varying levels of supervision for trainees, including direct supervision with the physical presence of a supervisor and the degree of availability of said supervisor, they also describe ensuring the quality of supervision provided.5 While continuous attending‐level supervision is not yet mandated, many residency programs look to their academic hospitalists to fill the supervisory void, particularly at night. However, what specific roles hospitalists play in the nighttime supervision of trainees or the impact of this supervision remains unclear. To date, no study has examined a broad sample of hospitalist programs in teaching hospitals and the types of resident oversight they provide. We aimed to describe the current state of academic hospitalists in the clinical supervision of housestaff, specifically during the overnight period, and hospitalist perceptions of how the new ACGME requirements would impact traineehospitalist interactions.

METHODS

The Housestaff Oversight Subcommittee, a working group of the Society of General Internal Medicine (SGIM) Academic Hospitalist Task Force, surveyed a sample of academic hospitalist program leaders to assess the current status of trainee supervision performed by hospitalists. Programs were considered academic if they were located in the primary hospital of a residency that participates in the National Resident Matching Program for Internal Medicine. To obtain a broad geographic spectrum of academic hospitalist programs, all programs, both university and community‐based, in 4 states and 2 metropolitan regions were sampled: Washington, Oregon, Texas, Maryland, and the Philadelphia and Chicago metropolitan areas. Hospitalist program leaders were identified by members of the Taskforce using individual program websites and by querying departmental leadership at eligible teaching hospitals. Respondents were contacted by e‐mail for participation. None of the authors of the manuscript were participants in the survey.

The survey was developed by consensus of the working group after reviewing the salient literature and included additional questions queried to internal medicine program directors.6 The 19‐item SurveyMonkey instrument included questions about hospitalists' role in trainees' education and evaluation. A Likert‐type scale was used to assess perceptions regarding the impact of on‐site hospitalist supervision on trainee autonomy and hospitalist workload (1 = strongly disagree to 5 = strongly agree). Descriptive statistics were performed and, where appropriate, t test and Fisher's exact test were performed to identify associations between program characteristics and perceptions. Stata SE was used (STATA Corp, College Station, TX) for all statistical analysis.

RESULTS

The survey was sent to 47 individuals identified as likely hospitalist program leaders and completed by 41 individuals (87%). However, 7 respondents turned out not to be program leaders and were therefore excluded, resulting in a 72% (34/47) survey response rate.

The programs for which we did not obtain responses were similar to respondent programs, and did not include a larger proportion of community‐based programs or overrepresent a specific geographic region. Twenty‐five (73%) of the 34 hospitalist program leaders were male, with an average age of 44.3 years, and an average of 12 years post‐residency training (range, 530 years). They reported leading groups with an average of 18 full‐time equivalent (FTE) faculty (range, 350 persons).

DISCUSSION

This study provides important information about how academic hospitalists currently contribute to the supervision of internal medicine residents. While academic hospitalist groups frequently have faculty providing clinical care on‐site at night, and often hospitalists provide overnight supervision of internal medicine trainees, formal supervision of trainees is not uniform, and few hospitalists groups have a mechanism to provide training or faculty development on how to effectively supervise resident trainees. Hospitalist leaders expressed concerns that creating additional formal overnight supervisory responsibilities may add to an already burdened overnight hospitalist. Formalizing this supervisory role, including explicit role definitions and faculty training for trainee supervision, is necessary.

Though our sample size is small, we captured a diverse geographic range of both university and community‐based academic hospitalist programs by surveying group leaders in several distinct regions. We are unable to comment on differences between responding and non‐responding hospitalist programs, but there does not appear to be a systematic difference between these groups.

Our findings are consistent with work describing a lack of structured conceptual frameworks in effectively supervising trainees,7, 8 and also, at times, nebulous expectations for hospitalist faculty. We found that the existence of a formal supervisory policy within the associated residency program, as well as defined roles for hospitalists, increases the likelihood of positive perceptions of the new ACGME supervisory recommendations. However, the existence of these requirements does not mean that all programs are capable of following them. While additional discussion is required to best delineate a formal overnight hospitalist role in trainee supervision, clearly defining expectations for both faculty and trainees, and their interactions, may alleviate the struggles that exist in programs with ill‐defined roles for hospitalist faculty supervision. While faculty duty hours standards do not exist, additional duties of nighttime coverage for hospitalists suggests that close attention should be paid to burn‐out.9 Faculty development on nighttime supervision and teaching may help maximize both learning and patient care efficiency, and provide a framework for this often unstructured educational time.

Over 14% of all patients hospitalized in the United States are readmitted within 30 days of discharge.1 Numerous studies have used administrative data in order to identify clinical and operational predictors of readmission. However, few studies have explored patients' perspectives on readmission.27 As a result, we know little about potentially modifiable challenges which patients face during the transition from hospital to home. Lack of understanding of the patient perspective has hampered the ability of hospitals to create interventions which address these underlying causes of readmissions.

Patients with low socioeconomic status (SES) are up to 43% more likely to require readmission than their higher‐SES counterparts,8, 9 and qualitative data has described unique challenges faced by low‐SES patients during transition.2 Our objectives were to understand the transition experiences of readmitted patients and to compare these experiences across SES and diagnostic categories.

METHODS

RESULTS

DISCUSSION

Several findings from this study are of interest to practicing hospitalists or hospital administrators. First, of the issues to which patients most commonly attributed readmission, lack of discharge preparedness is the only one which occurs during index hospitalization; in order to address most transition challenges, hospitals must think beyond their walls. By penalizing hospitals for excess rates of readmission, The Hospital Readmission Reduction Program (HRRP) will effectively hold hospitals accountable for addressing issues which occur in patients' homes and communities.17 Hospitals that have robust partnerships with community pharmacies, social service agencies, and PCPs may have the most influence on these issues and the most success in reducing readmissions. Second, consistent with other literature describing increased rates of readmission with enhanced PCP follow‐up,18 our findings demonstrate that PCPs often refer their patients to the emergency room for readmission. This suggests that PCP follow‐up, while perhaps essential for patient care, may not necessarily reduce readmissions and may actually facilitate readmission. Third, this study describes underlying reasons for patient nonadherence with discharge medications: side effects, cost, and transportation. Targeted interventions to improve adherence may include floor‐based pharmacists who counsel on side effects, determine co‐pays prior to discharge, and encourage patients to fill prescriptions from the hospital pharmacy to avoid transportation barriers.

Finally, and perhaps most importantly, these data suggest that one transition experience does not fit all. Patients with low SES appear to have a distinct and challenging transition experience. Currently, there is an emphasis on tailoring transition interventions to specific disease populations, such as patients with congestive heart failure. Our study suggests that it may be more effective to tailor interventions for low‐SES patients across diagnostic category, helping these patients gain access to outpatient medical resources and address competing issues, such as food insecurity or substance abuse.

Our study has several limitations. First, the low survey response rate makes it susceptible to nonresponse bias. Second, survey administration by a member of the care team may have increased social desirability bias. Third, because it was important to the study team to incorporate our survey into hospital workflow, survey responses were recorded directly into the EMR which limited administrators to recording a yes response for each answer choice which the participant endorsed. Therefore, in our dataset, we are unable to distinguish a definite no from a missing response; however, the survey was short, making it unlikely that questions were skipped. Fourth, closed‐ended questions may have failed to capture the range of participant responses, although the inclusion of an open‐ended answer choice ameliorates this issue. Finally, we are unable to draw conclusions regarding association of survey responses with the risk of readmission, because this study was administered only to readmitted patients.

CONCLUSIONS

This report of patients' perspectives challenges many commonly held assumptions regarding readmission. Readmission reflects not only the quality of hospital care, but a variety of factors in patients' homes and communities. PCP follow‐up, while perhaps critical for patient care, may not be a panacea for reducing hospital readmissions. Targeted medication counseling focused on side effects, co‐pay, and medication delivery may address patients' underlying reasons for nonadherence. And most importantly, one transition experience does not fit all. Hospitalists and administrators must tailor interventions to address challenges reported by their patients, particularly those of low SES.

Over 14% of all patients hospitalized in the United States are readmitted within 30 days of discharge.1 Numerous studies have used administrative data in order to identify clinical and operational predictors of readmission. However, few studies have explored patients' perspectives on readmission.27 As a result, we know little about potentially modifiable challenges which patients face during the transition from hospital to home. Lack of understanding of the patient perspective has hampered the ability of hospitals to create interventions which address these underlying causes of readmissions.

Patients with low socioeconomic status (SES) are up to 43% more likely to require readmission than their higher‐SES counterparts,8, 9 and qualitative data has described unique challenges faced by low‐SES patients during transition.2 Our objectives were to understand the transition experiences of readmitted patients and to compare these experiences across SES and diagnostic categories.

METHODS

RESULTS

DISCUSSION

Several findings from this study are of interest to practicing hospitalists or hospital administrators. First, of the issues to which patients most commonly attributed readmission, lack of discharge preparedness is the only one which occurs during index hospitalization; in order to address most transition challenges, hospitals must think beyond their walls. By penalizing hospitals for excess rates of readmission, The Hospital Readmission Reduction Program (HRRP) will effectively hold hospitals accountable for addressing issues which occur in patients' homes and communities.17 Hospitals that have robust partnerships with community pharmacies, social service agencies, and PCPs may have the most influence on these issues and the most success in reducing readmissions. Second, consistent with other literature describing increased rates of readmission with enhanced PCP follow‐up,18 our findings demonstrate that PCPs often refer their patients to the emergency room for readmission. This suggests that PCP follow‐up, while perhaps essential for patient care, may not necessarily reduce readmissions and may actually facilitate readmission. Third, this study describes underlying reasons for patient nonadherence with discharge medications: side effects, cost, and transportation. Targeted interventions to improve adherence may include floor‐based pharmacists who counsel on side effects, determine co‐pays prior to discharge, and encourage patients to fill prescriptions from the hospital pharmacy to avoid transportation barriers.

Finally, and perhaps most importantly, these data suggest that one transition experience does not fit all. Patients with low SES appear to have a distinct and challenging transition experience. Currently, there is an emphasis on tailoring transition interventions to specific disease populations, such as patients with congestive heart failure. Our study suggests that it may be more effective to tailor interventions for low‐SES patients across diagnostic category, helping these patients gain access to outpatient medical resources and address competing issues, such as food insecurity or substance abuse.

Our study has several limitations. First, the low survey response rate makes it susceptible to nonresponse bias. Second, survey administration by a member of the care team may have increased social desirability bias. Third, because it was important to the study team to incorporate our survey into hospital workflow, survey responses were recorded directly into the EMR which limited administrators to recording a yes response for each answer choice which the participant endorsed. Therefore, in our dataset, we are unable to distinguish a definite no from a missing response; however, the survey was short, making it unlikely that questions were skipped. Fourth, closed‐ended questions may have failed to capture the range of participant responses, although the inclusion of an open‐ended answer choice ameliorates this issue. Finally, we are unable to draw conclusions regarding association of survey responses with the risk of readmission, because this study was administered only to readmitted patients.

CONCLUSIONS

This report of patients' perspectives challenges many commonly held assumptions regarding readmission. Readmission reflects not only the quality of hospital care, but a variety of factors in patients' homes and communities. PCP follow‐up, while perhaps critical for patient care, may not be a panacea for reducing hospital readmissions. Targeted medication counseling focused on side effects, co‐pay, and medication delivery may address patients' underlying reasons for nonadherence. And most importantly, one transition experience does not fit all. Hospitalists and administrators must tailor interventions to address challenges reported by their patients, particularly those of low SES.

Over 14% of all patients hospitalized in the United States are readmitted within 30 days of discharge.1 Numerous studies have used administrative data in order to identify clinical and operational predictors of readmission. However, few studies have explored patients' perspectives on readmission.27 As a result, we know little about potentially modifiable challenges which patients face during the transition from hospital to home. Lack of understanding of the patient perspective has hampered the ability of hospitals to create interventions which address these underlying causes of readmissions.

Patients with low socioeconomic status (SES) are up to 43% more likely to require readmission than their higher‐SES counterparts,8, 9 and qualitative data has described unique challenges faced by low‐SES patients during transition.2 Our objectives were to understand the transition experiences of readmitted patients and to compare these experiences across SES and diagnostic categories.

METHODS

RESULTS

DISCUSSION

Several findings from this study are of interest to practicing hospitalists or hospital administrators. First, of the issues to which patients most commonly attributed readmission, lack of discharge preparedness is the only one which occurs during index hospitalization; in order to address most transition challenges, hospitals must think beyond their walls. By penalizing hospitals for excess rates of readmission, The Hospital Readmission Reduction Program (HRRP) will effectively hold hospitals accountable for addressing issues which occur in patients' homes and communities.17 Hospitals that have robust partnerships with community pharmacies, social service agencies, and PCPs may have the most influence on these issues and the most success in reducing readmissions. Second, consistent with other literature describing increased rates of readmission with enhanced PCP follow‐up,18 our findings demonstrate that PCPs often refer their patients to the emergency room for readmission. This suggests that PCP follow‐up, while perhaps essential for patient care, may not necessarily reduce readmissions and may actually facilitate readmission. Third, this study describes underlying reasons for patient nonadherence with discharge medications: side effects, cost, and transportation. Targeted interventions to improve adherence may include floor‐based pharmacists who counsel on side effects, determine co‐pays prior to discharge, and encourage patients to fill prescriptions from the hospital pharmacy to avoid transportation barriers.

Finally, and perhaps most importantly, these data suggest that one transition experience does not fit all. Patients with low SES appear to have a distinct and challenging transition experience. Currently, there is an emphasis on tailoring transition interventions to specific disease populations, such as patients with congestive heart failure. Our study suggests that it may be more effective to tailor interventions for low‐SES patients across diagnostic category, helping these patients gain access to outpatient medical resources and address competing issues, such as food insecurity or substance abuse.

Our study has several limitations. First, the low survey response rate makes it susceptible to nonresponse bias. Second, survey administration by a member of the care team may have increased social desirability bias. Third, because it was important to the study team to incorporate our survey into hospital workflow, survey responses were recorded directly into the EMR which limited administrators to recording a yes response for each answer choice which the participant endorsed. Therefore, in our dataset, we are unable to distinguish a definite no from a missing response; however, the survey was short, making it unlikely that questions were skipped. Fourth, closed‐ended questions may have failed to capture the range of participant responses, although the inclusion of an open‐ended answer choice ameliorates this issue. Finally, we are unable to draw conclusions regarding association of survey responses with the risk of readmission, because this study was administered only to readmitted patients.

CONCLUSIONS

This report of patients' perspectives challenges many commonly held assumptions regarding readmission. Readmission reflects not only the quality of hospital care, but a variety of factors in patients' homes and communities. PCP follow‐up, while perhaps critical for patient care, may not be a panacea for reducing hospital readmissions. Targeted medication counseling focused on side effects, co‐pay, and medication delivery may address patients' underlying reasons for nonadherence. And most importantly, one transition experience does not fit all. Hospitalists and administrators must tailor interventions to address challenges reported by their patients, particularly those of low SES.

Deficiencies in physical examination skills among medical students, housestaff, and even faculty have been reported for decades. For example, cardiac examination skills have been shown to improve during the early years of medical school and then plateau by the third year, with no measurable improvement through residency or beyond without fellowship training.13

Despite efforts by residency programs to promote bedside teaching, many general medicine faculty shy away due to lack of confidence and comfort in teaching physical diagnosis.4 The presence of hospitalists increases satisfaction among trainees with inpatient teaching,5 and medical faculty's ability has been shown to positively impact medical students' test scores.6 This lack of bedside teaching is a missed opportunity, as general medicine faculty are assuming more teaching responsibility.

The Merrin Bedside Teaching Program was founded in 2004 at New York University (NYU) School of Medicine to improve the quality of bedside teaching in the Department of Medicine/Division of General Internal Medicine. In this article, we describe the development process and early outcomes of this program.

METHODS

The principal teaching institution of NYU is Bellevue Hospital, an 800‐bed tertiary care safety‐net facility located in New York City. Approximately 30 general medicine faculty members at Bellevue Hospital attend on the inpatient teaching service 12 weeks per year each, with the remaining time spent in the hospital‐based clinics and other inpatient services.

Prior to the initiation of this program, our faculty did not receive formal instruction in bedside teaching, in general, or in teaching physical diagnosis, in particular. Housestaff and students were reporting that bedside rounds were not being conducted consistently and were of variable educational quality. Among a number of our residency and fellowship programs, there were concerns about meeting Accreditation Counsel for Graduate Medical Education (ACGME) requirements for quantity and quality of bedside rounds and faculty development.

To inform program development, we conducted a literature review, a series of focus groups with residents and faculty, and a survey of 39 general medicine hospital and ambulatory‐based clinician teachers to determine the perceived bedside teaching faculty development needs of our department. Then, over the next 2 years, we recruited 32 hospitalists and fellowship faculty to participate in a videotaped bedside teaching simulation. Each attending reviewed the videotape one‐on‐one with an experienced facilitator (A.K.), who elicited the attending's goals and instructional models and methods used for clinical teaching. To address the wide variety in teaching approaches identified in the assessment, 4 to 6 attendings met weekly for 1 month to observe each other conducting rounds with their teams. The facilitator of these groups used materials derived from relevant medical education theory79 and related conceptual models to frame the debriefing. Participants enthusiastically supported the educational value of this learner‐centered, experiential teaching approach.

We integrated this needs assessment with an individualized approach, incorporating learner goal‐setting with interactive and highly experiential teaching strategies,7,9,10 to create the Merrin Bedside Teaching Program in 2004. This program recruits faculty, with reputations as excellent teachers, to design a program to develop their own bedside teaching skills and disseminate what they learn to their peers. Faculty fellows are recruited through an open call for applications, which includes a letter of support from a supervisor and a detailed independent learning plan, including an identified mentor. Fellows are selected by the program's executive committee based on the likelihood of the success of their proposed program. A stipend equivalent of between 5% and 10% of base salary is provided to each fellow for a period of 2 years. Selected faculty fellows are encouraged to focus on an aspect of the physical examination, work in groups of 2 or 3, and to identify and recruit a mentor who is considered a master clinician in the target specialty. Master clinicians are given an honorarium to acknowledge their selection and incentivize them to spend time with the faculty fellows. This is funded with philanthropic support from the Merrin Family Foundation.

Fellows are guided by program leadership in their independent study, development of clinical teaching skills, and curriculum development using the same theory‐driven, systematic approach that framed program inception.911 Bedside rounds are the core instructional method used by each group of fellows and are supplemented by lectures, interactive small‐group seminars, and Web‐based modules in certain cases. Bedside sessions are run by the master‐clinician mentor until the faculty fellows are deemed competent by the mentor and feel confident enough to lead independently.

Since 2004, there have been 14 fellows who have developed programs focused on the examination of the heart, skin, knee, and shoulder. Program development is underway in motivational interviewing, the pulmonary examination, and the examination of the critically ill patient. We describe the work of the first 4 fellows as an example of how this fellowship creates value for the individual fellows, our departmental teaching programs, and the medical school.

Our first cohort of fellows chose, out of personal interest, to concentrate on the cardiac examination. They spent the first year working with highly respected cardiologists to hone their own clinical skills, reviewing the literature on the evidence‐based cardiac physical examination and effective teaching methods,12,13 researching the use of electronic stethoscopes and related technology for teaching at the bedside, and piloting a variety of approaches to teaching their busy colleagues these skills.

Bedside rounds focus on pertinent physical findings with an emphasis on an evidence‐based approach. We find we are most effective when the patient's diagnosis is unknown by the group leader to avoid bias when formulating the differential diagnosis. Sessions include a discussion of how the exam correlates with the diagnosis, relevant pathophysiology, imaging, and treatment options.

Two, 1‐hour‐long lectures in cardiac examination are delivered: the first reviews basics of heart sounds, both normal and abnormal, and the second reviews the most common systolic and diastolic murmurs. These lectures, scheduled into routine faculty conference time, utilize a PowerPoint format, with an overview of basic physiology and pathophysiology, aided by phonocardiograms, frequency spectrographs, and audio recordings delivered via a loudspeaker. Interactive cases offered by Blaufuss Multimedia (Rolling Hills Estates, CA) are an excellent teaching tool that incorporate case presentations, videos of key physical findings, auscultatory recordings, and relevant pathophysiology. We initially used this resource because of its high quality and ease of use; we now use our own interactive case presentations, which allow for flexibility with content and style, and which reinforce the prevalence of interesting cases at our institution to the audience members.

Technology has proven to be an invaluable tool in teaching cardiac physical diagnosis, both at the bedside and in the classroom. Electronic stethoscopes provide the ability to record heart sounds for use in teaching venues on short notice, such as morning report, and for use in creating the interactive case presentations described above. The electronic stethoscopes we use can be wired to peripheral devices, such as camcorders, iPods, and speaker pads. Speaker pads are devices, approximately the size of a stethoscope head, that can be connected by wires in series, each attached to a stethoscope, allowing a group of people to listen to the same sounds simultaneously with excellent sound reproduction. This technology allows each person standing around the bedside to listen to a patient while the group leader auscultates and explains the findings in real time. There are distinct advantages of simultaneous auscultation both for describing auditory findings and minimizing discomfort to the patient.

Applications are available for the iPod (Stethoscope App, Thinklabs Technology, LLC, Centennial, CO) which can record and display real‐time phonocardiography when attached to an electronic stethoscope, even at the terminus of a speaker pad chain. This application also allows recorded sounds to be played directly through a speaker, or transferred to a computer with the corresponding phonocardiographic and spectrographic images, that can all be incorporated into an interactive case presentation. Frequency spectrographs allow visualization of differences between low‐ and high‐frequency sounds, which, in conjunction with the timing and amplitude displayed by phonocardiography, can aid in teaching subtle findings, such as shapes of murmurs, patterns of splitting, gallups, etc.14 Playback of heart sounds in a conference room setting can be challenging, given the often subtle and low‐frequency findings typical of cardiac pathology, and is effectively achieved by using a musician‐quality loudspeaker. We have found that speaker pads offer the best sound quality at the bedside, although they are inconvenient for larger groups.

DISCUSSION

A new framework has been proposed for considering faculty development programs that focuses on the participants, program, content, facilitator, and the context in which the program occurs.15 We have effectively addressed and synthesized these components in a rich, high impact, learner‐centered faculty development program that also responds to challenges raised by changes in the health delivery system, concerns about accreditation requirements, and targeted local needs assessment.

We have been fortunate to recruit specialty faculty who are outstanding teachers, have welcomed the fellows into their clinics, and have dedicated countless hours to supervision and education. An unintended, but important, outcome of the program is that we are able to highlight the exceptional skills of our senior, experienced clinicians. These are colleagues who all too often do not receive adequate recognition in the modern‐day academic medical center environment, but who are undoubtedly invaluable to the education mission of these centers.

The existence of the program has resulted in our general medicine faculty showing great enthusiasm, both to develop an area of expertise and to participate as learners in the programs developed by peers. The faculty fellows in each specialty have become a valuable resource to peer faculty, residents, and medical students alike, who are now less dependent on consultants to identify and explain physical findings. The faculty teaching the knee and shoulder exams started a Sports Medicine Clinic within primary care, and assist with joint injections throughout the clinic. In addition to providing clinical support, their educational curriculum is included in both the attending and housestaff conference schedules. The cardiac lectures, both didactic and interactive case presentations, are included in the attending conference schedule, intern and resident core curricula, and the third‐year medicine clerkship lecture series. The dermatology group has created a series of comprehensive online modules that provide content tailored to general medicine. All this durable material is available broadly to trainees of all professions in our medical center.

Given the ever‐growing burden of patient care and extra‐clinical responsibilities, the principal factor limiting the effectiveness of bedside rounds is faculty availability. Despite this, all of our hospitalists have attended at least 1 bedside cardiac session, and the majority have attended multiple times. Varying the time and day of the sessions, offering to join attending rounds, and being available for impromptu diagnostic consultations have maximized the fellows' contact with faculty, residents, and students.

Although funding for evaluation of the program has been limited, a research agenda is emerging. Both the pulmonary physical exam and critical care groups are in the process of evaluating the effectiveness of their programs on the quality of bedside rounds, student and resident learning, and, to the extent possible, on patient care.

CONCLUSION

We believe wholeheartedly that bedside instruction both in physical diagnosis and interview skills must not become a lost art. General medicine faculty are ideally situated to take on this challenge. An educational program targeting hospitalists and general medicine faculty energizes faculty and leverages local resources to fill in gaps in skills for faculty and then for trainees. Generalist faculty relish the opportunity to champion a particular element of the doctorpatient encounter, which has contributed to our ultimate goal of strengthening the core diagnostic skills of our faculty who are at the forefront of clinical care and medical education.

Deficiencies in physical examination skills among medical students, housestaff, and even faculty have been reported for decades. For example, cardiac examination skills have been shown to improve during the early years of medical school and then plateau by the third year, with no measurable improvement through residency or beyond without fellowship training.13

Despite efforts by residency programs to promote bedside teaching, many general medicine faculty shy away due to lack of confidence and comfort in teaching physical diagnosis.4 The presence of hospitalists increases satisfaction among trainees with inpatient teaching,5 and medical faculty's ability has been shown to positively impact medical students' test scores.6 This lack of bedside teaching is a missed opportunity, as general medicine faculty are assuming more teaching responsibility.

The Merrin Bedside Teaching Program was founded in 2004 at New York University (NYU) School of Medicine to improve the quality of bedside teaching in the Department of Medicine/Division of General Internal Medicine. In this article, we describe the development process and early outcomes of this program.

METHODS

The principal teaching institution of NYU is Bellevue Hospital, an 800‐bed tertiary care safety‐net facility located in New York City. Approximately 30 general medicine faculty members at Bellevue Hospital attend on the inpatient teaching service 12 weeks per year each, with the remaining time spent in the hospital‐based clinics and other inpatient services.

Prior to the initiation of this program, our faculty did not receive formal instruction in bedside teaching, in general, or in teaching physical diagnosis, in particular. Housestaff and students were reporting that bedside rounds were not being conducted consistently and were of variable educational quality. Among a number of our residency and fellowship programs, there were concerns about meeting Accreditation Counsel for Graduate Medical Education (ACGME) requirements for quantity and quality of bedside rounds and faculty development.

To inform program development, we conducted a literature review, a series of focus groups with residents and faculty, and a survey of 39 general medicine hospital and ambulatory‐based clinician teachers to determine the perceived bedside teaching faculty development needs of our department. Then, over the next 2 years, we recruited 32 hospitalists and fellowship faculty to participate in a videotaped bedside teaching simulation. Each attending reviewed the videotape one‐on‐one with an experienced facilitator (A.K.), who elicited the attending's goals and instructional models and methods used for clinical teaching. To address the wide variety in teaching approaches identified in the assessment, 4 to 6 attendings met weekly for 1 month to observe each other conducting rounds with their teams. The facilitator of these groups used materials derived from relevant medical education theory79 and related conceptual models to frame the debriefing. Participants enthusiastically supported the educational value of this learner‐centered, experiential teaching approach.

We integrated this needs assessment with an individualized approach, incorporating learner goal‐setting with interactive and highly experiential teaching strategies,7,9,10 to create the Merrin Bedside Teaching Program in 2004. This program recruits faculty, with reputations as excellent teachers, to design a program to develop their own bedside teaching skills and disseminate what they learn to their peers. Faculty fellows are recruited through an open call for applications, which includes a letter of support from a supervisor and a detailed independent learning plan, including an identified mentor. Fellows are selected by the program's executive committee based on the likelihood of the success of their proposed program. A stipend equivalent of between 5% and 10% of base salary is provided to each fellow for a period of 2 years. Selected faculty fellows are encouraged to focus on an aspect of the physical examination, work in groups of 2 or 3, and to identify and recruit a mentor who is considered a master clinician in the target specialty. Master clinicians are given an honorarium to acknowledge their selection and incentivize them to spend time with the faculty fellows. This is funded with philanthropic support from the Merrin Family Foundation.

Fellows are guided by program leadership in their independent study, development of clinical teaching skills, and curriculum development using the same theory‐driven, systematic approach that framed program inception.911 Bedside rounds are the core instructional method used by each group of fellows and are supplemented by lectures, interactive small‐group seminars, and Web‐based modules in certain cases. Bedside sessions are run by the master‐clinician mentor until the faculty fellows are deemed competent by the mentor and feel confident enough to lead independently.

Since 2004, there have been 14 fellows who have developed programs focused on the examination of the heart, skin, knee, and shoulder. Program development is underway in motivational interviewing, the pulmonary examination, and the examination of the critically ill patient. We describe the work of the first 4 fellows as an example of how this fellowship creates value for the individual fellows, our departmental teaching programs, and the medical school.

Our first cohort of fellows chose, out of personal interest, to concentrate on the cardiac examination. They spent the first year working with highly respected cardiologists to hone their own clinical skills, reviewing the literature on the evidence‐based cardiac physical examination and effective teaching methods,12,13 researching the use of electronic stethoscopes and related technology for teaching at the bedside, and piloting a variety of approaches to teaching their busy colleagues these skills.

Bedside rounds focus on pertinent physical findings with an emphasis on an evidence‐based approach. We find we are most effective when the patient's diagnosis is unknown by the group leader to avoid bias when formulating the differential diagnosis. Sessions include a discussion of how the exam correlates with the diagnosis, relevant pathophysiology, imaging, and treatment options.

Two, 1‐hour‐long lectures in cardiac examination are delivered: the first reviews basics of heart sounds, both normal and abnormal, and the second reviews the most common systolic and diastolic murmurs. These lectures, scheduled into routine faculty conference time, utilize a PowerPoint format, with an overview of basic physiology and pathophysiology, aided by phonocardiograms, frequency spectrographs, and audio recordings delivered via a loudspeaker. Interactive cases offered by Blaufuss Multimedia (Rolling Hills Estates, CA) are an excellent teaching tool that incorporate case presentations, videos of key physical findings, auscultatory recordings, and relevant pathophysiology. We initially used this resource because of its high quality and ease of use; we now use our own interactive case presentations, which allow for flexibility with content and style, and which reinforce the prevalence of interesting cases at our institution to the audience members.

Technology has proven to be an invaluable tool in teaching cardiac physical diagnosis, both at the bedside and in the classroom. Electronic stethoscopes provide the ability to record heart sounds for use in teaching venues on short notice, such as morning report, and for use in creating the interactive case presentations described above. The electronic stethoscopes we use can be wired to peripheral devices, such as camcorders, iPods, and speaker pads. Speaker pads are devices, approximately the size of a stethoscope head, that can be connected by wires in series, each attached to a stethoscope, allowing a group of people to listen to the same sounds simultaneously with excellent sound reproduction. This technology allows each person standing around the bedside to listen to a patient while the group leader auscultates and explains the findings in real time. There are distinct advantages of simultaneous auscultation both for describing auditory findings and minimizing discomfort to the patient.

Applications are available for the iPod (Stethoscope App, Thinklabs Technology, LLC, Centennial, CO) which can record and display real‐time phonocardiography when attached to an electronic stethoscope, even at the terminus of a speaker pad chain. This application also allows recorded sounds to be played directly through a speaker, or transferred to a computer with the corresponding phonocardiographic and spectrographic images, that can all be incorporated into an interactive case presentation. Frequency spectrographs allow visualization of differences between low‐ and high‐frequency sounds, which, in conjunction with the timing and amplitude displayed by phonocardiography, can aid in teaching subtle findings, such as shapes of murmurs, patterns of splitting, gallups, etc.14 Playback of heart sounds in a conference room setting can be challenging, given the often subtle and low‐frequency findings typical of cardiac pathology, and is effectively achieved by using a musician‐quality loudspeaker. We have found that speaker pads offer the best sound quality at the bedside, although they are inconvenient for larger groups.

DISCUSSION

A new framework has been proposed for considering faculty development programs that focuses on the participants, program, content, facilitator, and the context in which the program occurs.15 We have effectively addressed and synthesized these components in a rich, high impact, learner‐centered faculty development program that also responds to challenges raised by changes in the health delivery system, concerns about accreditation requirements, and targeted local needs assessment.

We have been fortunate to recruit specialty faculty who are outstanding teachers, have welcomed the fellows into their clinics, and have dedicated countless hours to supervision and education. An unintended, but important, outcome of the program is that we are able to highlight the exceptional skills of our senior, experienced clinicians. These are colleagues who all too often do not receive adequate recognition in the modern‐day academic medical center environment, but who are undoubtedly invaluable to the education mission of these centers.

The existence of the program has resulted in our general medicine faculty showing great enthusiasm, both to develop an area of expertise and to participate as learners in the programs developed by peers. The faculty fellows in each specialty have become a valuable resource to peer faculty, residents, and medical students alike, who are now less dependent on consultants to identify and explain physical findings. The faculty teaching the knee and shoulder exams started a Sports Medicine Clinic within primary care, and assist with joint injections throughout the clinic. In addition to providing clinical support, their educational curriculum is included in both the attending and housestaff conference schedules. The cardiac lectures, both didactic and interactive case presentations, are included in the attending conference schedule, intern and resident core curricula, and the third‐year medicine clerkship lecture series. The dermatology group has created a series of comprehensive online modules that provide content tailored to general medicine. All this durable material is available broadly to trainees of all professions in our medical center.

Given the ever‐growing burden of patient care and extra‐clinical responsibilities, the principal factor limiting the effectiveness of bedside rounds is faculty availability. Despite this, all of our hospitalists have attended at least 1 bedside cardiac session, and the majority have attended multiple times. Varying the time and day of the sessions, offering to join attending rounds, and being available for impromptu diagnostic consultations have maximized the fellows' contact with faculty, residents, and students.

Although funding for evaluation of the program has been limited, a research agenda is emerging. Both the pulmonary physical exam and critical care groups are in the process of evaluating the effectiveness of their programs on the quality of bedside rounds, student and resident learning, and, to the extent possible, on patient care.

CONCLUSION

We believe wholeheartedly that bedside instruction both in physical diagnosis and interview skills must not become a lost art. General medicine faculty are ideally situated to take on this challenge. An educational program targeting hospitalists and general medicine faculty energizes faculty and leverages local resources to fill in gaps in skills for faculty and then for trainees. Generalist faculty relish the opportunity to champion a particular element of the doctorpatient encounter, which has contributed to our ultimate goal of strengthening the core diagnostic skills of our faculty who are at the forefront of clinical care and medical education.

Deficiencies in physical examination skills among medical students, housestaff, and even faculty have been reported for decades. For example, cardiac examination skills have been shown to improve during the early years of medical school and then plateau by the third year, with no measurable improvement through residency or beyond without fellowship training.13

Despite efforts by residency programs to promote bedside teaching, many general medicine faculty shy away due to lack of confidence and comfort in teaching physical diagnosis.4 The presence of hospitalists increases satisfaction among trainees with inpatient teaching,5 and medical faculty's ability has been shown to positively impact medical students' test scores.6 This lack of bedside teaching is a missed opportunity, as general medicine faculty are assuming more teaching responsibility.

The Merrin Bedside Teaching Program was founded in 2004 at New York University (NYU) School of Medicine to improve the quality of bedside teaching in the Department of Medicine/Division of General Internal Medicine. In this article, we describe the development process and early outcomes of this program.

METHODS

The principal teaching institution of NYU is Bellevue Hospital, an 800‐bed tertiary care safety‐net facility located in New York City. Approximately 30 general medicine faculty members at Bellevue Hospital attend on the inpatient teaching service 12 weeks per year each, with the remaining time spent in the hospital‐based clinics and other inpatient services.

Prior to the initiation of this program, our faculty did not receive formal instruction in bedside teaching, in general, or in teaching physical diagnosis, in particular. Housestaff and students were reporting that bedside rounds were not being conducted consistently and were of variable educational quality. Among a number of our residency and fellowship programs, there were concerns about meeting Accreditation Counsel for Graduate Medical Education (ACGME) requirements for quantity and quality of bedside rounds and faculty development.

To inform program development, we conducted a literature review, a series of focus groups with residents and faculty, and a survey of 39 general medicine hospital and ambulatory‐based clinician teachers to determine the perceived bedside teaching faculty development needs of our department. Then, over the next 2 years, we recruited 32 hospitalists and fellowship faculty to participate in a videotaped bedside teaching simulation. Each attending reviewed the videotape one‐on‐one with an experienced facilitator (A.K.), who elicited the attending's goals and instructional models and methods used for clinical teaching. To address the wide variety in teaching approaches identified in the assessment, 4 to 6 attendings met weekly for 1 month to observe each other conducting rounds with their teams. The facilitator of these groups used materials derived from relevant medical education theory79 and related conceptual models to frame the debriefing. Participants enthusiastically supported the educational value of this learner‐centered, experiential teaching approach.

We integrated this needs assessment with an individualized approach, incorporating learner goal‐setting with interactive and highly experiential teaching strategies,7,9,10 to create the Merrin Bedside Teaching Program in 2004. This program recruits faculty, with reputations as excellent teachers, to design a program to develop their own bedside teaching skills and disseminate what they learn to their peers. Faculty fellows are recruited through an open call for applications, which includes a letter of support from a supervisor and a detailed independent learning plan, including an identified mentor. Fellows are selected by the program's executive committee based on the likelihood of the success of their proposed program. A stipend equivalent of between 5% and 10% of base salary is provided to each fellow for a period of 2 years. Selected faculty fellows are encouraged to focus on an aspect of the physical examination, work in groups of 2 or 3, and to identify and recruit a mentor who is considered a master clinician in the target specialty. Master clinicians are given an honorarium to acknowledge their selection and incentivize them to spend time with the faculty fellows. This is funded with philanthropic support from the Merrin Family Foundation.

Fellows are guided by program leadership in their independent study, development of clinical teaching skills, and curriculum development using the same theory‐driven, systematic approach that framed program inception.911 Bedside rounds are the core instructional method used by each group of fellows and are supplemented by lectures, interactive small‐group seminars, and Web‐based modules in certain cases. Bedside sessions are run by the master‐clinician mentor until the faculty fellows are deemed competent by the mentor and feel confident enough to lead independently.

Since 2004, there have been 14 fellows who have developed programs focused on the examination of the heart, skin, knee, and shoulder. Program development is underway in motivational interviewing, the pulmonary examination, and the examination of the critically ill patient. We describe the work of the first 4 fellows as an example of how this fellowship creates value for the individual fellows, our departmental teaching programs, and the medical school.

Our first cohort of fellows chose, out of personal interest, to concentrate on the cardiac examination. They spent the first year working with highly respected cardiologists to hone their own clinical skills, reviewing the literature on the evidence‐based cardiac physical examination and effective teaching methods,12,13 researching the use of electronic stethoscopes and related technology for teaching at the bedside, and piloting a variety of approaches to teaching their busy colleagues these skills.

Bedside rounds focus on pertinent physical findings with an emphasis on an evidence‐based approach. We find we are most effective when the patient's diagnosis is unknown by the group leader to avoid bias when formulating the differential diagnosis. Sessions include a discussion of how the exam correlates with the diagnosis, relevant pathophysiology, imaging, and treatment options.

Two, 1‐hour‐long lectures in cardiac examination are delivered: the first reviews basics of heart sounds, both normal and abnormal, and the second reviews the most common systolic and diastolic murmurs. These lectures, scheduled into routine faculty conference time, utilize a PowerPoint format, with an overview of basic physiology and pathophysiology, aided by phonocardiograms, frequency spectrographs, and audio recordings delivered via a loudspeaker. Interactive cases offered by Blaufuss Multimedia (Rolling Hills Estates, CA) are an excellent teaching tool that incorporate case presentations, videos of key physical findings, auscultatory recordings, and relevant pathophysiology. We initially used this resource because of its high quality and ease of use; we now use our own interactive case presentations, which allow for flexibility with content and style, and which reinforce the prevalence of interesting cases at our institution to the audience members.

Technology has proven to be an invaluable tool in teaching cardiac physical diagnosis, both at the bedside and in the classroom. Electronic stethoscopes provide the ability to record heart sounds for use in teaching venues on short notice, such as morning report, and for use in creating the interactive case presentations described above. The electronic stethoscopes we use can be wired to peripheral devices, such as camcorders, iPods, and speaker pads. Speaker pads are devices, approximately the size of a stethoscope head, that can be connected by wires in series, each attached to a stethoscope, allowing a group of people to listen to the same sounds simultaneously with excellent sound reproduction. This technology allows each person standing around the bedside to listen to a patient while the group leader auscultates and explains the findings in real time. There are distinct advantages of simultaneous auscultation both for describing auditory findings and minimizing discomfort to the patient.

Applications are available for the iPod (Stethoscope App, Thinklabs Technology, LLC, Centennial, CO) which can record and display real‐time phonocardiography when attached to an electronic stethoscope, even at the terminus of a speaker pad chain. This application also allows recorded sounds to be played directly through a speaker, or transferred to a computer with the corresponding phonocardiographic and spectrographic images, that can all be incorporated into an interactive case presentation. Frequency spectrographs allow visualization of differences between low‐ and high‐frequency sounds, which, in conjunction with the timing and amplitude displayed by phonocardiography, can aid in teaching subtle findings, such as shapes of murmurs, patterns of splitting, gallups, etc.14 Playback of heart sounds in a conference room setting can be challenging, given the often subtle and low‐frequency findings typical of cardiac pathology, and is effectively achieved by using a musician‐quality loudspeaker. We have found that speaker pads offer the best sound quality at the bedside, although they are inconvenient for larger groups.

DISCUSSION

A new framework has been proposed for considering faculty development programs that focuses on the participants, program, content, facilitator, and the context in which the program occurs.15 We have effectively addressed and synthesized these components in a rich, high impact, learner‐centered faculty development program that also responds to challenges raised by changes in the health delivery system, concerns about accreditation requirements, and targeted local needs assessment.

We have been fortunate to recruit specialty faculty who are outstanding teachers, have welcomed the fellows into their clinics, and have dedicated countless hours to supervision and education. An unintended, but important, outcome of the program is that we are able to highlight the exceptional skills of our senior, experienced clinicians. These are colleagues who all too often do not receive adequate recognition in the modern‐day academic medical center environment, but who are undoubtedly invaluable to the education mission of these centers.

The existence of the program has resulted in our general medicine faculty showing great enthusiasm, both to develop an area of expertise and to participate as learners in the programs developed by peers. The faculty fellows in each specialty have become a valuable resource to peer faculty, residents, and medical students alike, who are now less dependent on consultants to identify and explain physical findings. The faculty teaching the knee and shoulder exams started a Sports Medicine Clinic within primary care, and assist with joint injections throughout the clinic. In addition to providing clinical support, their educational curriculum is included in both the attending and housestaff conference schedules. The cardiac lectures, both didactic and interactive case presentations, are included in the attending conference schedule, intern and resident core curricula, and the third‐year medicine clerkship lecture series. The dermatology group has created a series of comprehensive online modules that provide content tailored to general medicine. All this durable material is available broadly to trainees of all professions in our medical center.

Given the ever‐growing burden of patient care and extra‐clinical responsibilities, the principal factor limiting the effectiveness of bedside rounds is faculty availability. Despite this, all of our hospitalists have attended at least 1 bedside cardiac session, and the majority have attended multiple times. Varying the time and day of the sessions, offering to join attending rounds, and being available for impromptu diagnostic consultations have maximized the fellows' contact with faculty, residents, and students.

Although funding for evaluation of the program has been limited, a research agenda is emerging. Both the pulmonary physical exam and critical care groups are in the process of evaluating the effectiveness of their programs on the quality of bedside rounds, student and resident learning, and, to the extent possible, on patient care.

CONCLUSION

We believe wholeheartedly that bedside instruction both in physical diagnosis and interview skills must not become a lost art. General medicine faculty are ideally situated to take on this challenge. An educational program targeting hospitalists and general medicine faculty energizes faculty and leverages local resources to fill in gaps in skills for faculty and then for trainees. Generalist faculty relish the opportunity to champion a particular element of the doctorpatient encounter, which has contributed to our ultimate goal of strengthening the core diagnostic skills of our faculty who are at the forefront of clinical care and medical education.

Non‐polio enteroviruses are the most common cause of aseptic meningitis in children.1 While bacterial meningitis requires parenteral antibiotics, aseptic meningitis requires only supportive care.1 Enteroviral reverse transcription polymerase chain reaction (EVPCR) testing of the cerebrospinal fluid (CSF) allows the virus to be detected with high sensitivity and specificity.2 Because children with a positive EVPCR test are at low risk of bacterial meningitis,3 access to rapid EVPCR results has the potential to impact the clinical management of children with meningitis.4, 5 We studied the impact of implementing an in‐hospital EVPCR testing protocol on the clinical management of children with meningitis in a single‐center retrospective cohort.

MATERIALS AND METHODS

RESULTS

Of the 593 children with meningitis, 152 (26%) were excluded for the reasons noted above. The 441 patients included in our analyses had the following final diagnoses: bacterial meningitis (1 patient with Streptococcus pneumoniae, 0.2%), pretreated culture‐negative meningitis (42 patients, 10%), and aseptic meningitis (398 patients, 90%).

We compared patient populations and EVPCR testing characteristics between the pre‐ and post‐study periods (Table 1). While CSF glucose differed between study periods, the difference was not felt to be clinically significant. However, during the post‐period, more children presented during enteroviral season. Clinicians were more likely to order an EVPCR test for children with aseptic, than bacterial, meningitis (213/370 [58%] vs 0/1 [0%]).

We evaluated the impact of the in‐hospital EVPCR test on the length of stay and duration of parenteral antibiotics for the 6 predefined patient groups (Table 2). Length of stay could be determined for 432 (98%) of study patients, and duration of parenteral antibiotics for 365 (83%). We found a clinically important decrease in both length of stay and duration of parenteral antibiotics for children with a positive EVPCR test in the post‐period. For every hour earlier the EVPCR results returned, length of stay was reduced by 0.3 hours ( = 0.3, 95% confidence interval [CI] 0.10.5), and parenteral antibiotics were reduced by 0.3 hours ( = 0.3, 95% CI 0.10.5). However, even in the post‐period, the median length of time from a positive EVPCR test result to hospital discharge was 14 hours (interquartile range, 533 hours).

We observed no trend in length of stay in either testing period ( = 0.17, 95% CI 3.9 to 3.6 pre vs = 1.64, 95% CI 6.3 to 3.0 post), with no change following the introduction of the faster EVPCR protocol (P = 0.52). We observed an increase in duration of parenteral antibiotics in the pre‐period ( = 5.4, 95% CI 0.3 to 10.6), with no trend in the post‐period ( = 1.7, 95% CI 5.2 to 1.8), but the difference was not significant (P = 0.08).

DISCUSSION

The in‐hospital EVPCR testing protocol reduced test turnaround time and increased testing. Children with a positive test had a shorter length of stay and duration of parenteral antibiotics. Decreasing the test turnaround time for EVPCR improved the care of children with enteroviral meningitis by reducing the length of unnecessary hospitalizations and parenteral antibiotics, with the potential for reducing the costs associated with these admissions.

Accurate identification of children with enteroviral meningitis, an often self‐limited infection requiring supportive care, can reduce hospitalization and unnecessary antibiotics. Previously, a positive EVPCR test result has been associated with a reduction in length of stay and of parenteral antibiotics,4, 5, 1012 with a direct correlation between test turnaround time and length of stay.12, 13 Moreover, positive EVPCR test results that were available prior to hospital discharge resulted in shorter length of hospital stay and duration of parenteral antibiotics.10

Our study is the largest to investigate the impact of implementing an in‐hospital EVPCR testing protocol, with the goal of making results available in a clinically useful time frame for all patients. Older EVPCR tests were typically performed in batches, or at centralized laboratories.4, 5, 1013 The in‐hospital EVPCR test utilized is a simple testing platform, which can be run multiple times daily. While there were higher charges associated with increased testing in the post‐period, these were more than offset by a reduced length of stay. Using study institution patient charges, we estimate that overall patient charges decreased approximately $80,000 in the post‐period, compared to the pre‐period (an average reduction of $375 per patient).

Many children were not discharged when a positive EVPCR test result became available. Some children with enteroviral meningitis will have persistent symptoms that require inpatient management. In addition, results that returned in the evening or nighttime were less likely to result in immediate hospital discharge. However, children with a positive EVPCR test are at very low risk for bacterial meningitis.3 As clinicians' knowledge of, and comfort with, the EVPCR test increase, this technology has the potential to further decrease the costs of caring for children with enteroviral meningitis.14

Our study had several limitations. First, it was retrospective; however, primary outcomes were objective measures accurately recorded in the medical record for most patients. Second, our study was single‐center, and findings may not be generalizable to other settings. Third, the management of children with meningitis may have been changing over the study period, independent of the in‐hospital EVPCR test. However, among children with a negative test, we observed no change in either of our primary outcomes. Fourth, given the large number of physicians involved with testing and treatment decisions, we could not adjust for clustering at the physician level. Fifth, we corrected CSF WBC in the case of a traumatic lumbar puncture (LP). Although use of this correction might underestimate the true CSF WBC count,6 the percentage of children with traumatic lumbar punctures was the same in both testing periods. Lastly, we evaluated the impact of a diagnostic test immediately after introduction into the clinical setting. As new medical technologies often take time to be adopted into clinical practice,15 we would expect the impact to increase over time.

CONCLUSIONS

In‐hospital EVPCR testing can improve the care of children with meningitis by reducing the length of unnecessary hospitalizations and parenteral antibiotics. Clinicians caring for children with meningitis should have access to in‐hospital EVPCR testing.

Non‐polio enteroviruses are the most common cause of aseptic meningitis in children.1 While bacterial meningitis requires parenteral antibiotics, aseptic meningitis requires only supportive care.1 Enteroviral reverse transcription polymerase chain reaction (EVPCR) testing of the cerebrospinal fluid (CSF) allows the virus to be detected with high sensitivity and specificity.2 Because children with a positive EVPCR test are at low risk of bacterial meningitis,3 access to rapid EVPCR results has the potential to impact the clinical management of children with meningitis.4, 5 We studied the impact of implementing an in‐hospital EVPCR testing protocol on the clinical management of children with meningitis in a single‐center retrospective cohort.

MATERIALS AND METHODS

RESULTS

Of the 593 children with meningitis, 152 (26%) were excluded for the reasons noted above. The 441 patients included in our analyses had the following final diagnoses: bacterial meningitis (1 patient with Streptococcus pneumoniae, 0.2%), pretreated culture‐negative meningitis (42 patients, 10%), and aseptic meningitis (398 patients, 90%).

We compared patient populations and EVPCR testing characteristics between the pre‐ and post‐study periods (Table 1). While CSF glucose differed between study periods, the difference was not felt to be clinically significant. However, during the post‐period, more children presented during enteroviral season. Clinicians were more likely to order an EVPCR test for children with aseptic, than bacterial, meningitis (213/370 [58%] vs 0/1 [0%]).

We evaluated the impact of the in‐hospital EVPCR test on the length of stay and duration of parenteral antibiotics for the 6 predefined patient groups (Table 2). Length of stay could be determined for 432 (98%) of study patients, and duration of parenteral antibiotics for 365 (83%). We found a clinically important decrease in both length of stay and duration of parenteral antibiotics for children with a positive EVPCR test in the post‐period. For every hour earlier the EVPCR results returned, length of stay was reduced by 0.3 hours ( = 0.3, 95% confidence interval [CI] 0.10.5), and parenteral antibiotics were reduced by 0.3 hours ( = 0.3, 95% CI 0.10.5). However, even in the post‐period, the median length of time from a positive EVPCR test result to hospital discharge was 14 hours (interquartile range, 533 hours).

We observed no trend in length of stay in either testing period ( = 0.17, 95% CI 3.9 to 3.6 pre vs = 1.64, 95% CI 6.3 to 3.0 post), with no change following the introduction of the faster EVPCR protocol (P = 0.52). We observed an increase in duration of parenteral antibiotics in the pre‐period ( = 5.4, 95% CI 0.3 to 10.6), with no trend in the post‐period ( = 1.7, 95% CI 5.2 to 1.8), but the difference was not significant (P = 0.08).

DISCUSSION

The in‐hospital EVPCR testing protocol reduced test turnaround time and increased testing. Children with a positive test had a shorter length of stay and duration of parenteral antibiotics. Decreasing the test turnaround time for EVPCR improved the care of children with enteroviral meningitis by reducing the length of unnecessary hospitalizations and parenteral antibiotics, with the potential for reducing the costs associated with these admissions.

Accurate identification of children with enteroviral meningitis, an often self‐limited infection requiring supportive care, can reduce hospitalization and unnecessary antibiotics. Previously, a positive EVPCR test result has been associated with a reduction in length of stay and of parenteral antibiotics,4, 5, 1012 with a direct correlation between test turnaround time and length of stay.12, 13 Moreover, positive EVPCR test results that were available prior to hospital discharge resulted in shorter length of hospital stay and duration of parenteral antibiotics.10

Our study is the largest to investigate the impact of implementing an in‐hospital EVPCR testing protocol, with the goal of making results available in a clinically useful time frame for all patients. Older EVPCR tests were typically performed in batches, or at centralized laboratories.4, 5, 1013 The in‐hospital EVPCR test utilized is a simple testing platform, which can be run multiple times daily. While there were higher charges associated with increased testing in the post‐period, these were more than offset by a reduced length of stay. Using study institution patient charges, we estimate that overall patient charges decreased approximately $80,000 in the post‐period, compared to the pre‐period (an average reduction of $375 per patient).

Many children were not discharged when a positive EVPCR test result became available. Some children with enteroviral meningitis will have persistent symptoms that require inpatient management. In addition, results that returned in the evening or nighttime were less likely to result in immediate hospital discharge. However, children with a positive EVPCR test are at very low risk for bacterial meningitis.3 As clinicians' knowledge of, and comfort with, the EVPCR test increase, this technology has the potential to further decrease the costs of caring for children with enteroviral meningitis.14

Our study had several limitations. First, it was retrospective; however, primary outcomes were objective measures accurately recorded in the medical record for most patients. Second, our study was single‐center, and findings may not be generalizable to other settings. Third, the management of children with meningitis may have been changing over the study period, independent of the in‐hospital EVPCR test. However, among children with a negative test, we observed no change in either of our primary outcomes. Fourth, given the large number of physicians involved with testing and treatment decisions, we could not adjust for clustering at the physician level. Fifth, we corrected CSF WBC in the case of a traumatic lumbar puncture (LP). Although use of this correction might underestimate the true CSF WBC count,6 the percentage of children with traumatic lumbar punctures was the same in both testing periods. Lastly, we evaluated the impact of a diagnostic test immediately after introduction into the clinical setting. As new medical technologies often take time to be adopted into clinical practice,15 we would expect the impact to increase over time.

CONCLUSIONS

In‐hospital EVPCR testing can improve the care of children with meningitis by reducing the length of unnecessary hospitalizations and parenteral antibiotics. Clinicians caring for children with meningitis should have access to in‐hospital EVPCR testing.

Non‐polio enteroviruses are the most common cause of aseptic meningitis in children.1 While bacterial meningitis requires parenteral antibiotics, aseptic meningitis requires only supportive care.1 Enteroviral reverse transcription polymerase chain reaction (EVPCR) testing of the cerebrospinal fluid (CSF) allows the virus to be detected with high sensitivity and specificity.2 Because children with a positive EVPCR test are at low risk of bacterial meningitis,3 access to rapid EVPCR results has the potential to impact the clinical management of children with meningitis.4, 5 We studied the impact of implementing an in‐hospital EVPCR testing protocol on the clinical management of children with meningitis in a single‐center retrospective cohort.

MATERIALS AND METHODS

RESULTS

Of the 593 children with meningitis, 152 (26%) were excluded for the reasons noted above. The 441 patients included in our analyses had the following final diagnoses: bacterial meningitis (1 patient with Streptococcus pneumoniae, 0.2%), pretreated culture‐negative meningitis (42 patients, 10%), and aseptic meningitis (398 patients, 90%).

We compared patient populations and EVPCR testing characteristics between the pre‐ and post‐study periods (Table 1). While CSF glucose differed between study periods, the difference was not felt to be clinically significant. However, during the post‐period, more children presented during enteroviral season. Clinicians were more likely to order an EVPCR test for children with aseptic, than bacterial, meningitis (213/370 [58%] vs 0/1 [0%]).

We evaluated the impact of the in‐hospital EVPCR test on the length of stay and duration of parenteral antibiotics for the 6 predefined patient groups (Table 2). Length of stay could be determined for 432 (98%) of study patients, and duration of parenteral antibiotics for 365 (83%). We found a clinically important decrease in both length of stay and duration of parenteral antibiotics for children with a positive EVPCR test in the post‐period. For every hour earlier the EVPCR results returned, length of stay was reduced by 0.3 hours ( = 0.3, 95% confidence interval [CI] 0.10.5), and parenteral antibiotics were reduced by 0.3 hours ( = 0.3, 95% CI 0.10.5). However, even in the post‐period, the median length of time from a positive EVPCR test result to hospital discharge was 14 hours (interquartile range, 533 hours).

We observed no trend in length of stay in either testing period ( = 0.17, 95% CI 3.9 to 3.6 pre vs = 1.64, 95% CI 6.3 to 3.0 post), with no change following the introduction of the faster EVPCR protocol (P = 0.52). We observed an increase in duration of parenteral antibiotics in the pre‐period ( = 5.4, 95% CI 0.3 to 10.6), with no trend in the post‐period ( = 1.7, 95% CI 5.2 to 1.8), but the difference was not significant (P = 0.08).

DISCUSSION

The in‐hospital EVPCR testing protocol reduced test turnaround time and increased testing. Children with a positive test had a shorter length of stay and duration of parenteral antibiotics. Decreasing the test turnaround time for EVPCR improved the care of children with enteroviral meningitis by reducing the length of unnecessary hospitalizations and parenteral antibiotics, with the potential for reducing the costs associated with these admissions.

Accurate identification of children with enteroviral meningitis, an often self‐limited infection requiring supportive care, can reduce hospitalization and unnecessary antibiotics. Previously, a positive EVPCR test result has been associated with a reduction in length of stay and of parenteral antibiotics,4, 5, 1012 with a direct correlation between test turnaround time and length of stay.12, 13 Moreover, positive EVPCR test results that were available prior to hospital discharge resulted in shorter length of hospital stay and duration of parenteral antibiotics.10

Our study is the largest to investigate the impact of implementing an in‐hospital EVPCR testing protocol, with the goal of making results available in a clinically useful time frame for all patients. Older EVPCR tests were typically performed in batches, or at centralized laboratories.4, 5, 1013 The in‐hospital EVPCR test utilized is a simple testing platform, which can be run multiple times daily. While there were higher charges associated with increased testing in the post‐period, these were more than offset by a reduced length of stay. Using study institution patient charges, we estimate that overall patient charges decreased approximately $80,000 in the post‐period, compared to the pre‐period (an average reduction of $375 per patient).

Many children were not discharged when a positive EVPCR test result became available. Some children with enteroviral meningitis will have persistent symptoms that require inpatient management. In addition, results that returned in the evening or nighttime were less likely to result in immediate hospital discharge. However, children with a positive EVPCR test are at very low risk for bacterial meningitis.3 As clinicians' knowledge of, and comfort with, the EVPCR test increase, this technology has the potential to further decrease the costs of caring for children with enteroviral meningitis.14

Our study had several limitations. First, it was retrospective; however, primary outcomes were objective measures accurately recorded in the medical record for most patients. Second, our study was single‐center, and findings may not be generalizable to other settings. Third, the management of children with meningitis may have been changing over the study period, independent of the in‐hospital EVPCR test. However, among children with a negative test, we observed no change in either of our primary outcomes. Fourth, given the large number of physicians involved with testing and treatment decisions, we could not adjust for clustering at the physician level. Fifth, we corrected CSF WBC in the case of a traumatic lumbar puncture (LP). Although use of this correction might underestimate the true CSF WBC count,6 the percentage of children with traumatic lumbar punctures was the same in both testing periods. Lastly, we evaluated the impact of a diagnostic test immediately after introduction into the clinical setting. As new medical technologies often take time to be adopted into clinical practice,15 we would expect the impact to increase over time.

CONCLUSIONS

In‐hospital EVPCR testing can improve the care of children with meningitis by reducing the length of unnecessary hospitalizations and parenteral antibiotics. Clinicians caring for children with meningitis should have access to in‐hospital EVPCR testing.

Hospital discharge can be hazardous because discontinuity and fragmentation of care increase risks to the patient. Inadequate communication has been identified as a major etiology for errors and adverse events occurring shortly after discharge.1, 2 Another potential result of a failed hospital discharge is patient dissatisfaction. Increased patient involvement in care improves health outcomes, and may improve patient satisfaction.3 To engage patients in their care, healthcare providers must collaborate with patients to coordinate care across settings.

In this study, we sought to determine what patients and their caregivers view as essential elements of a safe and high‐quality discharge process. We developed a survey with a broad range of questions related to the hospital discharge process (see Supporting Information, Appendix A, in the online version of this article). The survey included several questions derived from Project BOOST (Better Outcomes for Older adults through Safe Transitions) discharge care plans.4

METHODS

RESULTS