Kevin J. O'Leary, MD, MS

The patient‐physician relationship is fundamental to safe and effective care. Hospital settings present unique challenges to this partnership, including the lack of a prior relationship for hospital‐based physicians, rapid pace of clinical care, and dynamic nature of inpatient medical teams. Prior studies document that a majority of hospitalized patients are unable to correctly identify their physicians or nurses, and patients in teaching hospitals have difficulty understanding their physicians' level of training.[1, 2, 3, 4] Acknowledging these deficits, professional societies and the Accreditation Council for Graduate Medical Education (ACMGE) have issued policies stating that patients and caregivers need to know who is responsible at every point during patient care.[5, 6] These policies do not, however, make recommendations on methods to achieve better understanding.

Simple interventions improve patients' ability to correctly identify the names and roles of their hospital physicians. Maniaci and colleagues found that patients were better able to identify attending physicians when their names were written on the dry‐erase board in the room.[7] Arora and colleagues asked hospital physicians to give facecards, which included their picture and a description of their role, to patients.[8] Patients were more likely to correctly identify 1 physicians, but, surprisingly, less likely to understand physicians' roles. In a similar study, Francis and colleagues placed photographs with names of the attending and resident physicians on the wall in patient rooms.[9] Patients who had photographs of their physicians on the wall were more likely to correctly identify physicians on their team compared with patients who had no photographs. Additionally, patients who were able to identify more physicians rated satisfaction with physicians higher in 2 of 6 survey questions used. However, the study was limited by the use of a nonvalidated instrument to assess patient satisfaction and the use of an intermediate outcome (ie, ability to identify physicians) as the independent variable rather than the intervention itself (ie, physician photographs).

Beyond satisfaction, lack of familiarity may negatively impact patients' trust and agreement with hospital physicians. Trust and agreement are important predictors of adherence to recommended treatment in outpatient settings[10, 11, 12, 13, 14, 15, 16, 17, 18] but have not been adequately evaluated in hospital settings. Therefore, we sought to pilot the use of physician facecards and assess their potential impact on patients' knowledge of physicians' names and roles as well as patient satisfaction, trust, and agreement with physicians.

METHODS

Physician Facecard

We created draft physician facecards featuring pictures of physicians and descriptions of their roles. We used Lexile analysis, a widely used measure of reading difficulty, to improve readability in an iterative fashion.[20, 21] We then sought feedback at hospitalist and resident meetings. Specifically, we asked for suggested revisions to content and recommendations on reliable methods to deliver facecards to patients. Teaching physicians felt strongly that each team member should be listed and shown on 1 card, which would fit easily into a lab‐coat pocket. We similarly engaged the NMH Patient and Family Advisory Council to seek recommended revisions to content and delivery of the facecards. The Council consists of 18 patient and caregiver members who meet regularly to provide input on hospital programs and proposals. Council members felt strongly that physicians should deliver the cards themselves during their initial introduction, rather than having patients receive cards by other means (eg, as part of unit orientation materials delivered by nonphysician staff members). We incorporated feedback from these stakeholder groups into a final version of the physician facecard and method for delivery (Figure 1).

Figure 1

We implemented the use of facecards from May to June 2012. Physicians on intervention units were informed of the study via email, and one of the co‐investigators (T.C.) distributed a supply of facecards to these physicians at the start of each rotation. This distribution was performed in person, and physicians were instructed to provide a facecard to each new patient during their first encounter. We also placed facecards in easily visible cardholders at the nurses' station on intervention units. Reminder emails were sent once each week to reinforce physician delivery of facecards.

RESULTS

DISCUSSION

We found that receipt of physician facecards significantly improved patients' knowledge of the names and roles of hospital physicians but had little to no impact on satisfaction, trust, or agreement with physicians. Our finding of improved knowledge of the names and roles of physician providers is consistent with prior studies using similar interventions.[7, 8, 9] Facecards may have prompted more effective introductions on the part of physicians and may have served as memory aids for patients to better retain information about their newly introduced hospital physicians.

Patient receipt of the facecard on intervention units was incomplete in our study. Despite engagement of physicians in designing cards that could easily fit into lab coats and a robust strategy to inform and motivate physician delivery of facecards, only 68% of intended patients received them. Although not explicitly reported, prior studies appear to have similarly struggled to deliver interventions consistently. Arora and colleagues reported that facecards were visible in only 59% of patients' rooms among those able to correctly identify 1 of their physicians.[8] A post hoc survey of physicians involved in our study revealed the biggest impediment to delivering facecards was simply forgetting to do so (data not shown). Technologic innovations may help by automating the identification of providers. For example, the University of Pittsburgh Medical Center has piloted smart rooms that use sensor technology to announce the name and role of providers as they enter patients' rooms.[26]

We hypothesized that facecards might improve other important aspects of the patient‐physicians relationship. Although levels of patient satisfaction were slightly higher in patients who had received facecards, the results were not statistically significant. Levels of trust and agreement were minimally higher in patients who received facecards, and the results were not statistically significant. Notably, baseline levels of trust and agreement were higher than we had expected. In fact, levels of trust were nearly identical to those seen in a prior study of outpatients who had been with the same physician for a median of 4 years.[22] Patients in our study may have had high levels of trust in the hospital and transferred this trust to their assigned physicians as representatives of the organization. The high level of agreement may relate to patients' tendency to prefer a more passive role as they encounter serious illness.[27, 28] Paradoxically, these findings may impede optimal patient care. The high levels of trust and agreement in the current study suggest that patients may not question their physicians to clarify plans and the rationale behind them. Prior research has shown that deficits in patients' comprehension of the care plan are often not apparent to patients or their physicians.[4, 29, 30]

Our study has several limitations. First, we assessed an intervention involving 4 units in a single hospital. Generalizability may be limited, as physician‐staffing models, hospitals, and the patients they serve vary. Second, as previously mentioned, patients in the intervention group did not receive physician facecards as consistently as intended. We conducted analyses based on treatment received in an effort to evaluate the impact of facecards if optimally delivered. Third, questions assessing satisfaction, trust, and agreement did not specifically ask patients to reflect on care provided by the primary physician team. It is possible that interactions with other physicians (ie, consultants) may have influenced these results. Fourth, we were underpowered to detect statistically significant improvements in satisfaction, trust, or agreement resulting from our intervention. Assuming the intervention might truly improve satisfaction with physicians from 54.2% to 63.6%, we would have needed 900 patients (ie, 450 each for the intervention and control groups) to have 80% power to detect a statistically significant difference. However, our results show that patients have high levels of trust and agreement with hospital physicians despite the relative lack of familiarity. Therefore, any existing deficits in hospitalized patients' comprehension of the care plan do not appear to be exacerbated by a lack of trust and/or agreement with treating physicians.

CONCLUSION

In summary, we found that physician facecards significantly improved patients' knowledge of the names and roles of hospital physicians but had little to no impact on satisfaction, trust, or agreement with physicians. Baseline levels of satisfaction, trust, and agreement were high, suggesting lack of familiarity with hospital physicians does not impede these important aspects of the patient‐physician relationship. Larger studies are needed to definitively assess the impact of facecards on satisfaction, trust, and agreement with physicians.

The patient‐physician relationship is fundamental to safe and effective care. Hospital settings present unique challenges to this partnership, including the lack of a prior relationship for hospital‐based physicians, rapid pace of clinical care, and dynamic nature of inpatient medical teams. Prior studies document that a majority of hospitalized patients are unable to correctly identify their physicians or nurses, and patients in teaching hospitals have difficulty understanding their physicians' level of training.[1, 2, 3, 4] Acknowledging these deficits, professional societies and the Accreditation Council for Graduate Medical Education (ACMGE) have issued policies stating that patients and caregivers need to know who is responsible at every point during patient care.[5, 6] These policies do not, however, make recommendations on methods to achieve better understanding.

Simple interventions improve patients' ability to correctly identify the names and roles of their hospital physicians. Maniaci and colleagues found that patients were better able to identify attending physicians when their names were written on the dry‐erase board in the room.[7] Arora and colleagues asked hospital physicians to give facecards, which included their picture and a description of their role, to patients.[8] Patients were more likely to correctly identify 1 physicians, but, surprisingly, less likely to understand physicians' roles. In a similar study, Francis and colleagues placed photographs with names of the attending and resident physicians on the wall in patient rooms.[9] Patients who had photographs of their physicians on the wall were more likely to correctly identify physicians on their team compared with patients who had no photographs. Additionally, patients who were able to identify more physicians rated satisfaction with physicians higher in 2 of 6 survey questions used. However, the study was limited by the use of a nonvalidated instrument to assess patient satisfaction and the use of an intermediate outcome (ie, ability to identify physicians) as the independent variable rather than the intervention itself (ie, physician photographs).

Beyond satisfaction, lack of familiarity may negatively impact patients' trust and agreement with hospital physicians. Trust and agreement are important predictors of adherence to recommended treatment in outpatient settings[10, 11, 12, 13, 14, 15, 16, 17, 18] but have not been adequately evaluated in hospital settings. Therefore, we sought to pilot the use of physician facecards and assess their potential impact on patients' knowledge of physicians' names and roles as well as patient satisfaction, trust, and agreement with physicians.

METHODS

Physician Facecard

We created draft physician facecards featuring pictures of physicians and descriptions of their roles. We used Lexile analysis, a widely used measure of reading difficulty, to improve readability in an iterative fashion.[20, 21] We then sought feedback at hospitalist and resident meetings. Specifically, we asked for suggested revisions to content and recommendations on reliable methods to deliver facecards to patients. Teaching physicians felt strongly that each team member should be listed and shown on 1 card, which would fit easily into a lab‐coat pocket. We similarly engaged the NMH Patient and Family Advisory Council to seek recommended revisions to content and delivery of the facecards. The Council consists of 18 patient and caregiver members who meet regularly to provide input on hospital programs and proposals. Council members felt strongly that physicians should deliver the cards themselves during their initial introduction, rather than having patients receive cards by other means (eg, as part of unit orientation materials delivered by nonphysician staff members). We incorporated feedback from these stakeholder groups into a final version of the physician facecard and method for delivery (Figure 1).

Figure 1

We implemented the use of facecards from May to June 2012. Physicians on intervention units were informed of the study via email, and one of the co‐investigators (T.C.) distributed a supply of facecards to these physicians at the start of each rotation. This distribution was performed in person, and physicians were instructed to provide a facecard to each new patient during their first encounter. We also placed facecards in easily visible cardholders at the nurses' station on intervention units. Reminder emails were sent once each week to reinforce physician delivery of facecards.

RESULTS

DISCUSSION

We found that receipt of physician facecards significantly improved patients' knowledge of the names and roles of hospital physicians but had little to no impact on satisfaction, trust, or agreement with physicians. Our finding of improved knowledge of the names and roles of physician providers is consistent with prior studies using similar interventions.[7, 8, 9] Facecards may have prompted more effective introductions on the part of physicians and may have served as memory aids for patients to better retain information about their newly introduced hospital physicians.

Patient receipt of the facecard on intervention units was incomplete in our study. Despite engagement of physicians in designing cards that could easily fit into lab coats and a robust strategy to inform and motivate physician delivery of facecards, only 68% of intended patients received them. Although not explicitly reported, prior studies appear to have similarly struggled to deliver interventions consistently. Arora and colleagues reported that facecards were visible in only 59% of patients' rooms among those able to correctly identify 1 of their physicians.[8] A post hoc survey of physicians involved in our study revealed the biggest impediment to delivering facecards was simply forgetting to do so (data not shown). Technologic innovations may help by automating the identification of providers. For example, the University of Pittsburgh Medical Center has piloted smart rooms that use sensor technology to announce the name and role of providers as they enter patients' rooms.[26]

We hypothesized that facecards might improve other important aspects of the patient‐physicians relationship. Although levels of patient satisfaction were slightly higher in patients who had received facecards, the results were not statistically significant. Levels of trust and agreement were minimally higher in patients who received facecards, and the results were not statistically significant. Notably, baseline levels of trust and agreement were higher than we had expected. In fact, levels of trust were nearly identical to those seen in a prior study of outpatients who had been with the same physician for a median of 4 years.[22] Patients in our study may have had high levels of trust in the hospital and transferred this trust to their assigned physicians as representatives of the organization. The high level of agreement may relate to patients' tendency to prefer a more passive role as they encounter serious illness.[27, 28] Paradoxically, these findings may impede optimal patient care. The high levels of trust and agreement in the current study suggest that patients may not question their physicians to clarify plans and the rationale behind them. Prior research has shown that deficits in patients' comprehension of the care plan are often not apparent to patients or their physicians.[4, 29, 30]

Our study has several limitations. First, we assessed an intervention involving 4 units in a single hospital. Generalizability may be limited, as physician‐staffing models, hospitals, and the patients they serve vary. Second, as previously mentioned, patients in the intervention group did not receive physician facecards as consistently as intended. We conducted analyses based on treatment received in an effort to evaluate the impact of facecards if optimally delivered. Third, questions assessing satisfaction, trust, and agreement did not specifically ask patients to reflect on care provided by the primary physician team. It is possible that interactions with other physicians (ie, consultants) may have influenced these results. Fourth, we were underpowered to detect statistically significant improvements in satisfaction, trust, or agreement resulting from our intervention. Assuming the intervention might truly improve satisfaction with physicians from 54.2% to 63.6%, we would have needed 900 patients (ie, 450 each for the intervention and control groups) to have 80% power to detect a statistically significant difference. However, our results show that patients have high levels of trust and agreement with hospital physicians despite the relative lack of familiarity. Therefore, any existing deficits in hospitalized patients' comprehension of the care plan do not appear to be exacerbated by a lack of trust and/or agreement with treating physicians.

CONCLUSION

In summary, we found that physician facecards significantly improved patients' knowledge of the names and roles of hospital physicians but had little to no impact on satisfaction, trust, or agreement with physicians. Baseline levels of satisfaction, trust, and agreement were high, suggesting lack of familiarity with hospital physicians does not impede these important aspects of the patient‐physician relationship. Larger studies are needed to definitively assess the impact of facecards on satisfaction, trust, and agreement with physicians.

The patient‐physician relationship is fundamental to safe and effective care. Hospital settings present unique challenges to this partnership, including the lack of a prior relationship for hospital‐based physicians, rapid pace of clinical care, and dynamic nature of inpatient medical teams. Prior studies document that a majority of hospitalized patients are unable to correctly identify their physicians or nurses, and patients in teaching hospitals have difficulty understanding their physicians' level of training.[1, 2, 3, 4] Acknowledging these deficits, professional societies and the Accreditation Council for Graduate Medical Education (ACMGE) have issued policies stating that patients and caregivers need to know who is responsible at every point during patient care.[5, 6] These policies do not, however, make recommendations on methods to achieve better understanding.

Simple interventions improve patients' ability to correctly identify the names and roles of their hospital physicians. Maniaci and colleagues found that patients were better able to identify attending physicians when their names were written on the dry‐erase board in the room.[7] Arora and colleagues asked hospital physicians to give facecards, which included their picture and a description of their role, to patients.[8] Patients were more likely to correctly identify 1 physicians, but, surprisingly, less likely to understand physicians' roles. In a similar study, Francis and colleagues placed photographs with names of the attending and resident physicians on the wall in patient rooms.[9] Patients who had photographs of their physicians on the wall were more likely to correctly identify physicians on their team compared with patients who had no photographs. Additionally, patients who were able to identify more physicians rated satisfaction with physicians higher in 2 of 6 survey questions used. However, the study was limited by the use of a nonvalidated instrument to assess patient satisfaction and the use of an intermediate outcome (ie, ability to identify physicians) as the independent variable rather than the intervention itself (ie, physician photographs).

Beyond satisfaction, lack of familiarity may negatively impact patients' trust and agreement with hospital physicians. Trust and agreement are important predictors of adherence to recommended treatment in outpatient settings[10, 11, 12, 13, 14, 15, 16, 17, 18] but have not been adequately evaluated in hospital settings. Therefore, we sought to pilot the use of physician facecards and assess their potential impact on patients' knowledge of physicians' names and roles as well as patient satisfaction, trust, and agreement with physicians.

METHODS

Physician Facecard

We created draft physician facecards featuring pictures of physicians and descriptions of their roles. We used Lexile analysis, a widely used measure of reading difficulty, to improve readability in an iterative fashion.[20, 21] We then sought feedback at hospitalist and resident meetings. Specifically, we asked for suggested revisions to content and recommendations on reliable methods to deliver facecards to patients. Teaching physicians felt strongly that each team member should be listed and shown on 1 card, which would fit easily into a lab‐coat pocket. We similarly engaged the NMH Patient and Family Advisory Council to seek recommended revisions to content and delivery of the facecards. The Council consists of 18 patient and caregiver members who meet regularly to provide input on hospital programs and proposals. Council members felt strongly that physicians should deliver the cards themselves during their initial introduction, rather than having patients receive cards by other means (eg, as part of unit orientation materials delivered by nonphysician staff members). We incorporated feedback from these stakeholder groups into a final version of the physician facecard and method for delivery (Figure 1).

Figure 1

We implemented the use of facecards from May to June 2012. Physicians on intervention units were informed of the study via email, and one of the co‐investigators (T.C.) distributed a supply of facecards to these physicians at the start of each rotation. This distribution was performed in person, and physicians were instructed to provide a facecard to each new patient during their first encounter. We also placed facecards in easily visible cardholders at the nurses' station on intervention units. Reminder emails were sent once each week to reinforce physician delivery of facecards.

RESULTS

DISCUSSION

We found that receipt of physician facecards significantly improved patients' knowledge of the names and roles of hospital physicians but had little to no impact on satisfaction, trust, or agreement with physicians. Our finding of improved knowledge of the names and roles of physician providers is consistent with prior studies using similar interventions.[7, 8, 9] Facecards may have prompted more effective introductions on the part of physicians and may have served as memory aids for patients to better retain information about their newly introduced hospital physicians.

Patient receipt of the facecard on intervention units was incomplete in our study. Despite engagement of physicians in designing cards that could easily fit into lab coats and a robust strategy to inform and motivate physician delivery of facecards, only 68% of intended patients received them. Although not explicitly reported, prior studies appear to have similarly struggled to deliver interventions consistently. Arora and colleagues reported that facecards were visible in only 59% of patients' rooms among those able to correctly identify 1 of their physicians.[8] A post hoc survey of physicians involved in our study revealed the biggest impediment to delivering facecards was simply forgetting to do so (data not shown). Technologic innovations may help by automating the identification of providers. For example, the University of Pittsburgh Medical Center has piloted smart rooms that use sensor technology to announce the name and role of providers as they enter patients' rooms.[26]

We hypothesized that facecards might improve other important aspects of the patient‐physicians relationship. Although levels of patient satisfaction were slightly higher in patients who had received facecards, the results were not statistically significant. Levels of trust and agreement were minimally higher in patients who received facecards, and the results were not statistically significant. Notably, baseline levels of trust and agreement were higher than we had expected. In fact, levels of trust were nearly identical to those seen in a prior study of outpatients who had been with the same physician for a median of 4 years.[22] Patients in our study may have had high levels of trust in the hospital and transferred this trust to their assigned physicians as representatives of the organization. The high level of agreement may relate to patients' tendency to prefer a more passive role as they encounter serious illness.[27, 28] Paradoxically, these findings may impede optimal patient care. The high levels of trust and agreement in the current study suggest that patients may not question their physicians to clarify plans and the rationale behind them. Prior research has shown that deficits in patients' comprehension of the care plan are often not apparent to patients or their physicians.[4, 29, 30]

Our study has several limitations. First, we assessed an intervention involving 4 units in a single hospital. Generalizability may be limited, as physician‐staffing models, hospitals, and the patients they serve vary. Second, as previously mentioned, patients in the intervention group did not receive physician facecards as consistently as intended. We conducted analyses based on treatment received in an effort to evaluate the impact of facecards if optimally delivered. Third, questions assessing satisfaction, trust, and agreement did not specifically ask patients to reflect on care provided by the primary physician team. It is possible that interactions with other physicians (ie, consultants) may have influenced these results. Fourth, we were underpowered to detect statistically significant improvements in satisfaction, trust, or agreement resulting from our intervention. Assuming the intervention might truly improve satisfaction with physicians from 54.2% to 63.6%, we would have needed 900 patients (ie, 450 each for the intervention and control groups) to have 80% power to detect a statistically significant difference. However, our results show that patients have high levels of trust and agreement with hospital physicians despite the relative lack of familiarity. Therefore, any existing deficits in hospitalized patients' comprehension of the care plan do not appear to be exacerbated by a lack of trust and/or agreement with treating physicians.

CONCLUSION

In summary, we found that physician facecards significantly improved patients' knowledge of the names and roles of hospital physicians but had little to no impact on satisfaction, trust, or agreement with physicians. Baseline levels of satisfaction, trust, and agreement were high, suggesting lack of familiarity with hospital physicians does not impede these important aspects of the patient‐physician relationship. Larger studies are needed to definitively assess the impact of facecards on satisfaction, trust, and agreement with physicians.

In this issue of the Journal of Hospital Medicine, Hansen and colleagues provide a first, early look at the effectiveness of the BOOST intervention to reduce 30‐day readmissions among hospitalized patients.[1] BOOST[2] is 1 of a number of care transition improvement methodologies that have been applied to the problem of readmissions, each of which has evidence to support its effectiveness in its initial settings[3, 4] but has proven to be difficult to translate to other sites.[5, 6, 7]

BOOST stands in contrast with other, largely research protocol‐derived, programs in that it allows sites to tailor adoption of recommendations to local contexts and is therefore potentially more feasible to implement. Feasibility and practicality has led BOOST to be adopted in large national settings, even if it has had little evidence to support its effectiveness to date.

Given the nonstandardized and ad hoc nature of most multicenter collaboratives generally, and the flexibility of the BOOST model specifically, the BOOST authors are to be commended for undertaking any evaluation at all. Perhaps, not surprisingly, they encountered many of the problems associated with a multicenter studydropout of sites, problematic data, and limited evidence for adoption of the intervention at participating hospitals. Although these represent real‐world experiences of a quality‐improvement program, as a group they pose a number of problems that limit the study's robustness, and generate important caveats that readers should use to temper their interpretation of the authors' findings.

The first caveat relates to the substantial number of sites that either dropped out of BOOST or failed to submit data after enlisting in the collaborative. Although this may be common in quality improvement collaboratives, similar problems would not be permissible in a trial of a new drug or device. Dropout and selected ability to contribute data suggest that the ability to fully adopt BOOST may not be universal, and raises the possibility of bias, because the least successful sites may have had less interest in remaining engaged and submitting data.

The second caveat relates to how readmission rates were assessed. Because sites provided rates of readmissions at the unit level rather than the actual counts of admissions or readmissions, the authors were unable to conduct statistical analyses typically performed for these interventions, such as time series or difference‐in‐difference analyses. More importantly, one cannot discern whether their results are driven by a small absolute but large relative change in the number of readmissions at small sites. That is, large percentage changes of low statistical significance could have misleadingly affected the overall results. Conversely, we cannot identify large sites where a similar relative reduction could be statistically significant and more broadly interpreted as representing the real effectiveness of BOOST efforts.

The third caveat is in regard to the data describing the sites' performance. The effectiveness of BOOST in this analysis varied greatly among sites, with only 1 site showing a strong reduction in readmission rate, and nearly all others showing no statistical improvements. In fact, it appears that their overall results were almost entirely driven by the improvements at that 1 site.

Variable effectiveness of an intervention can be related to variable adoption or contextual factors (such as availability of personnel to implement the program). Although these authors have data on BOOST programmatic adoption, they do not have qualitative data on local barriers and facilitators to BOOST implementation, which at this stage of evaluation would be particularly valuable in understanding the results. Analyzing site‐level effectiveness is of growing relevance to multicenter quality improvement collaboratives,[8, 9] but this evaluation provides little insight into reasons for variable success across institutions.

Finally, their study design does not allow us to understand a number of key questions. How many patients were involved in the intervention? How many patients received all BOOST‐recommended interventions? Which of these interventions seemed most effective in which patients? To what degree did patient severity of illness, cognitive status, social supports, or access to primary care influence readmission risk? Such information would help frame cost‐effective deployment of BOOST or related tools.

In the end, it seems unlikely that this iteration of the BOOST program produced broad reductions in readmission rates. Having said this, the authors provide the necessary start down the road toward a fuller understanding of real‐world efforts to reduce readmissions. Stated alternately, the nuances and flaws of this study provide ample fodder for others working in the field. BOOST is in good stead with other care transition models that have not translated well from their initial research environment to real‐world practices. The question now is: Do any of these interventions actually work in clinical practice settings, and will we ever know? Even more fundamentally, how important and meaningful are these hospital‐based care transition interventions? Where is the engagement with primary care? Where are the primary care outcomes? Does BOOST truly impact outcomes other than readmission?[10]

Doing high‐quality research in the context of a rapidly evolving quality improvement program is hard. Doing it at more than 1 site is harder. BOOST's flexibility is both a great source of strength and a clear challenge to rigorous evaluation. However, when the costs of care transition programs are so high, and the potential consequences of high readmission rates are so great for patients and for hospitals, the need to address these issues with real data and better evidence is paramount. We look forward to the next phase of BOOST and to the growth and refinement of the evidence base for how to improve care coordination and transitions effectively.

In this issue of the Journal of Hospital Medicine, Hansen and colleagues provide a first, early look at the effectiveness of the BOOST intervention to reduce 30‐day readmissions among hospitalized patients.[1] BOOST[2] is 1 of a number of care transition improvement methodologies that have been applied to the problem of readmissions, each of which has evidence to support its effectiveness in its initial settings[3, 4] but has proven to be difficult to translate to other sites.[5, 6, 7]

BOOST stands in contrast with other, largely research protocol‐derived, programs in that it allows sites to tailor adoption of recommendations to local contexts and is therefore potentially more feasible to implement. Feasibility and practicality has led BOOST to be adopted in large national settings, even if it has had little evidence to support its effectiveness to date.

Given the nonstandardized and ad hoc nature of most multicenter collaboratives generally, and the flexibility of the BOOST model specifically, the BOOST authors are to be commended for undertaking any evaluation at all. Perhaps, not surprisingly, they encountered many of the problems associated with a multicenter studydropout of sites, problematic data, and limited evidence for adoption of the intervention at participating hospitals. Although these represent real‐world experiences of a quality‐improvement program, as a group they pose a number of problems that limit the study's robustness, and generate important caveats that readers should use to temper their interpretation of the authors' findings.

The first caveat relates to the substantial number of sites that either dropped out of BOOST or failed to submit data after enlisting in the collaborative. Although this may be common in quality improvement collaboratives, similar problems would not be permissible in a trial of a new drug or device. Dropout and selected ability to contribute data suggest that the ability to fully adopt BOOST may not be universal, and raises the possibility of bias, because the least successful sites may have had less interest in remaining engaged and submitting data.

The second caveat relates to how readmission rates were assessed. Because sites provided rates of readmissions at the unit level rather than the actual counts of admissions or readmissions, the authors were unable to conduct statistical analyses typically performed for these interventions, such as time series or difference‐in‐difference analyses. More importantly, one cannot discern whether their results are driven by a small absolute but large relative change in the number of readmissions at small sites. That is, large percentage changes of low statistical significance could have misleadingly affected the overall results. Conversely, we cannot identify large sites where a similar relative reduction could be statistically significant and more broadly interpreted as representing the real effectiveness of BOOST efforts.

The third caveat is in regard to the data describing the sites' performance. The effectiveness of BOOST in this analysis varied greatly among sites, with only 1 site showing a strong reduction in readmission rate, and nearly all others showing no statistical improvements. In fact, it appears that their overall results were almost entirely driven by the improvements at that 1 site.

Variable effectiveness of an intervention can be related to variable adoption or contextual factors (such as availability of personnel to implement the program). Although these authors have data on BOOST programmatic adoption, they do not have qualitative data on local barriers and facilitators to BOOST implementation, which at this stage of evaluation would be particularly valuable in understanding the results. Analyzing site‐level effectiveness is of growing relevance to multicenter quality improvement collaboratives,[8, 9] but this evaluation provides little insight into reasons for variable success across institutions.

Finally, their study design does not allow us to understand a number of key questions. How many patients were involved in the intervention? How many patients received all BOOST‐recommended interventions? Which of these interventions seemed most effective in which patients? To what degree did patient severity of illness, cognitive status, social supports, or access to primary care influence readmission risk? Such information would help frame cost‐effective deployment of BOOST or related tools.

In the end, it seems unlikely that this iteration of the BOOST program produced broad reductions in readmission rates. Having said this, the authors provide the necessary start down the road toward a fuller understanding of real‐world efforts to reduce readmissions. Stated alternately, the nuances and flaws of this study provide ample fodder for others working in the field. BOOST is in good stead with other care transition models that have not translated well from their initial research environment to real‐world practices. The question now is: Do any of these interventions actually work in clinical practice settings, and will we ever know? Even more fundamentally, how important and meaningful are these hospital‐based care transition interventions? Where is the engagement with primary care? Where are the primary care outcomes? Does BOOST truly impact outcomes other than readmission?[10]

Doing high‐quality research in the context of a rapidly evolving quality improvement program is hard. Doing it at more than 1 site is harder. BOOST's flexibility is both a great source of strength and a clear challenge to rigorous evaluation. However, when the costs of care transition programs are so high, and the potential consequences of high readmission rates are so great for patients and for hospitals, the need to address these issues with real data and better evidence is paramount. We look forward to the next phase of BOOST and to the growth and refinement of the evidence base for how to improve care coordination and transitions effectively.

In this issue of the Journal of Hospital Medicine, Hansen and colleagues provide a first, early look at the effectiveness of the BOOST intervention to reduce 30‐day readmissions among hospitalized patients.[1] BOOST[2] is 1 of a number of care transition improvement methodologies that have been applied to the problem of readmissions, each of which has evidence to support its effectiveness in its initial settings[3, 4] but has proven to be difficult to translate to other sites.[5, 6, 7]

BOOST stands in contrast with other, largely research protocol‐derived, programs in that it allows sites to tailor adoption of recommendations to local contexts and is therefore potentially more feasible to implement. Feasibility and practicality has led BOOST to be adopted in large national settings, even if it has had little evidence to support its effectiveness to date.

Given the nonstandardized and ad hoc nature of most multicenter collaboratives generally, and the flexibility of the BOOST model specifically, the BOOST authors are to be commended for undertaking any evaluation at all. Perhaps, not surprisingly, they encountered many of the problems associated with a multicenter studydropout of sites, problematic data, and limited evidence for adoption of the intervention at participating hospitals. Although these represent real‐world experiences of a quality‐improvement program, as a group they pose a number of problems that limit the study's robustness, and generate important caveats that readers should use to temper their interpretation of the authors' findings.

The first caveat relates to the substantial number of sites that either dropped out of BOOST or failed to submit data after enlisting in the collaborative. Although this may be common in quality improvement collaboratives, similar problems would not be permissible in a trial of a new drug or device. Dropout and selected ability to contribute data suggest that the ability to fully adopt BOOST may not be universal, and raises the possibility of bias, because the least successful sites may have had less interest in remaining engaged and submitting data.

The second caveat relates to how readmission rates were assessed. Because sites provided rates of readmissions at the unit level rather than the actual counts of admissions or readmissions, the authors were unable to conduct statistical analyses typically performed for these interventions, such as time series or difference‐in‐difference analyses. More importantly, one cannot discern whether their results are driven by a small absolute but large relative change in the number of readmissions at small sites. That is, large percentage changes of low statistical significance could have misleadingly affected the overall results. Conversely, we cannot identify large sites where a similar relative reduction could be statistically significant and more broadly interpreted as representing the real effectiveness of BOOST efforts.

The third caveat is in regard to the data describing the sites' performance. The effectiveness of BOOST in this analysis varied greatly among sites, with only 1 site showing a strong reduction in readmission rate, and nearly all others showing no statistical improvements. In fact, it appears that their overall results were almost entirely driven by the improvements at that 1 site.

Variable effectiveness of an intervention can be related to variable adoption or contextual factors (such as availability of personnel to implement the program). Although these authors have data on BOOST programmatic adoption, they do not have qualitative data on local barriers and facilitators to BOOST implementation, which at this stage of evaluation would be particularly valuable in understanding the results. Analyzing site‐level effectiveness is of growing relevance to multicenter quality improvement collaboratives,[8, 9] but this evaluation provides little insight into reasons for variable success across institutions.

Finally, their study design does not allow us to understand a number of key questions. How many patients were involved in the intervention? How many patients received all BOOST‐recommended interventions? Which of these interventions seemed most effective in which patients? To what degree did patient severity of illness, cognitive status, social supports, or access to primary care influence readmission risk? Such information would help frame cost‐effective deployment of BOOST or related tools.

In the end, it seems unlikely that this iteration of the BOOST program produced broad reductions in readmission rates. Having said this, the authors provide the necessary start down the road toward a fuller understanding of real‐world efforts to reduce readmissions. Stated alternately, the nuances and flaws of this study provide ample fodder for others working in the field. BOOST is in good stead with other care transition models that have not translated well from their initial research environment to real‐world practices. The question now is: Do any of these interventions actually work in clinical practice settings, and will we ever know? Even more fundamentally, how important and meaningful are these hospital‐based care transition interventions? Where is the engagement with primary care? Where are the primary care outcomes? Does BOOST truly impact outcomes other than readmission?[10]

Doing high‐quality research in the context of a rapidly evolving quality improvement program is hard. Doing it at more than 1 site is harder. BOOST's flexibility is both a great source of strength and a clear challenge to rigorous evaluation. However, when the costs of care transition programs are so high, and the potential consequences of high readmission rates are so great for patients and for hospitals, the need to address these issues with real data and better evidence is paramount. We look forward to the next phase of BOOST and to the growth and refinement of the evidence base for how to improve care coordination and transitions effectively.

Hospital settings present unique challenges to patient‐clinician communication and collaboration. Patients frequently have multiple, active conditions. Interprofessional teams are large and care for multiple patients at the same time, and team membership is dynamic and dispersed. Moreover, physicians spend relatively little time with patients[1, 2] and seldom receive training in communication skills after medical school.

The Agency for Healthcare Research and Quality (AHRQ) has developed the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey to assess hospitalized patients' experiences with care.[3, 4, 5] Results are publicly reported on the US Department of Health and Human Services Hospital Compare Web site[6] and now affect hospital payment through the Center for Medicare and Medicaid Services Hospital Value‐Based Purchasing Program.[7]

Despite this increased transparency and accountability for performance related to the patient experience, little research has been conducted on how hospitals or clinicians might improve performance. Although interventions to enhance physician communication skills have shown improvements in observed behaviors, few studies have assessed benefit from the patient's perspective and few interventions have been integrated into practice.[8] We sought to assess the impact of a communication‐skills training program, based on a common framework used by hospitals, on patient satisfaction with doctor communication and overall hospital care.

METHODS

RESULTS

Overall, 61 (97%) of 63 hospitalists completed the first session, 44 (70%) completed the second session, and 25 (40%) completed the third session of program. Patient‐satisfaction data were available for 278 patients during the preintervention period and 186 patients during the postintervention period. Patient demographic characteristics were similar for the 2 periods (Table 2).

DISCUSSION

We found no significant improvement in patient satisfaction with doctor communication or overall rating of hospital care after implementation of a communication‐skills training program for hospitalists. There are several potential explanations for our results. First, though we used sound educational methods and attempted to replicate common clinical scenarios during simulation exercises, our program may not have resulted in improved communication behaviors during actual clinical care. We attempted to balance instructional methods that would result in behavioral change with a feasible investment of time and effort on the part of our learners (ie, practicing hospitalists). It is possible that additional time, feedback, and practice of communication skills would be necessary to change behaviors in the clinical setting. However, prior communication‐skills interventions have similarly struggled to show an impact on patient satisfaction.[13, 14] Second, we had incomplete participation in the program, with only 40% of hospitalists completing all 3 planned sessions. We encouraged all hospitalists, regardless of job type, to participate in the program. Participation rates were lower for 1‐year hospitalists compared with career hospitalists. The results of our analyses based on level of hospitalist participation in the training program, although not achieving statistical significance, suggest a greater effect of the program with higher degrees of participation.

Most important, the study was likely underpowered to detect a statistically significant difference in satisfaction results. Leaders were committed to providing communication‐skills training throughout our organization. We did not know the magnitude of potential improvement in satisfaction scores that might arise from our efforts, and therefore we did not conduct power calculations before designing and implementing the training program. Our HCAHPS composite doctor‐communication domain performance was 76% during the preintervention period and 79% during the postintervention period. Assuming an absolute 3% improvement is indeed possible, we would have needed >3000 patients in each period to have 80% power to detect a significant difference. Similarly, we would have needed >2000 patients during each period to have 80% power to detect an absolute 4% improvement in global rating of hospital care.

In an attempt to discern whether our favorable results were due to secular trends, we conducted post hoc analyses of HCAHPS nurse‐communication and hospital‐environment domains for the preintervention vs postintervention periods. Two of the 3 nurse‐communication items were rated lower during the postintervention period, but no result was statistically significant. Both hospital‐environment domain items were rated lower during the postintervention period, and 1 result was statistically significant (quiet at night). This post hoc evaluation lends additional support to the potential benefit of the communication‐skills training program.

The findings from this study represent an important issue for leaders attempting to improve quality performance within their organizations. What level of proof is needed before investing time and effort in implementing an intervention? With mounting pressure to improve performance, leaders are often left to make informed decisions based on data that fall short of scientifically rigorous evidence. Importantly, an increase in composite doctor‐communication ratings from 76% to 79% would translate into an improvement from the 25th percentile to 50th‐percentile performance in the fiscal‐year 2011 Press Ganey University Healthcare Consortium benchmark comparison (based on surveys received from September 1, 2010, to August 31, 2011).[15]

Our study has several limitations. First, we assessed an intervention on a single service in a single hospital. Generalizability may be limited, as hospital medicine groups, hospitals, and the patients they serve vary. Second, our intervention was based on a framework (ie, AIDET) that has face validity but has not undergone extensive study to confirm that the underlying constructs, and the behaviors related to them, are tightly linked to patient satisfaction. Third, as previously mentioned, we were likely underpowered to detect a significant improvement in satisfaction resulting from our intervention. Incomplete participation in the training program may have also limited the effect of our intervention. Finally, our comparisons by hospitalist level of participation were based on the discharging physician. Attribution of a patient response to a single physician is problematic because many patients encounter more than 1 hospitalist and 1 or more specialist physicians during their stay.

CONCLUSION

In summary, we found improvements in patient satisfaction with doctor communication, which were not statistically significant, after implementation of a communication‐skills training program for hospitalists. Larger studies are needed to assess whether a communication‐skills training program can truly improve patient satisfaction with doctor communication and overall hospital care.

Hospital settings present unique challenges to patient‐clinician communication and collaboration. Patients frequently have multiple, active conditions. Interprofessional teams are large and care for multiple patients at the same time, and team membership is dynamic and dispersed. Moreover, physicians spend relatively little time with patients[1, 2] and seldom receive training in communication skills after medical school.

The Agency for Healthcare Research and Quality (AHRQ) has developed the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey to assess hospitalized patients' experiences with care.[3, 4, 5] Results are publicly reported on the US Department of Health and Human Services Hospital Compare Web site[6] and now affect hospital payment through the Center for Medicare and Medicaid Services Hospital Value‐Based Purchasing Program.[7]

Despite this increased transparency and accountability for performance related to the patient experience, little research has been conducted on how hospitals or clinicians might improve performance. Although interventions to enhance physician communication skills have shown improvements in observed behaviors, few studies have assessed benefit from the patient's perspective and few interventions have been integrated into practice.[8] We sought to assess the impact of a communication‐skills training program, based on a common framework used by hospitals, on patient satisfaction with doctor communication and overall hospital care.

METHODS

RESULTS

Overall, 61 (97%) of 63 hospitalists completed the first session, 44 (70%) completed the second session, and 25 (40%) completed the third session of program. Patient‐satisfaction data were available for 278 patients during the preintervention period and 186 patients during the postintervention period. Patient demographic characteristics were similar for the 2 periods (Table 2).

DISCUSSION

We found no significant improvement in patient satisfaction with doctor communication or overall rating of hospital care after implementation of a communication‐skills training program for hospitalists. There are several potential explanations for our results. First, though we used sound educational methods and attempted to replicate common clinical scenarios during simulation exercises, our program may not have resulted in improved communication behaviors during actual clinical care. We attempted to balance instructional methods that would result in behavioral change with a feasible investment of time and effort on the part of our learners (ie, practicing hospitalists). It is possible that additional time, feedback, and practice of communication skills would be necessary to change behaviors in the clinical setting. However, prior communication‐skills interventions have similarly struggled to show an impact on patient satisfaction.[13, 14] Second, we had incomplete participation in the program, with only 40% of hospitalists completing all 3 planned sessions. We encouraged all hospitalists, regardless of job type, to participate in the program. Participation rates were lower for 1‐year hospitalists compared with career hospitalists. The results of our analyses based on level of hospitalist participation in the training program, although not achieving statistical significance, suggest a greater effect of the program with higher degrees of participation.

Most important, the study was likely underpowered to detect a statistically significant difference in satisfaction results. Leaders were committed to providing communication‐skills training throughout our organization. We did not know the magnitude of potential improvement in satisfaction scores that might arise from our efforts, and therefore we did not conduct power calculations before designing and implementing the training program. Our HCAHPS composite doctor‐communication domain performance was 76% during the preintervention period and 79% during the postintervention period. Assuming an absolute 3% improvement is indeed possible, we would have needed >3000 patients in each period to have 80% power to detect a significant difference. Similarly, we would have needed >2000 patients during each period to have 80% power to detect an absolute 4% improvement in global rating of hospital care.

In an attempt to discern whether our favorable results were due to secular trends, we conducted post hoc analyses of HCAHPS nurse‐communication and hospital‐environment domains for the preintervention vs postintervention periods. Two of the 3 nurse‐communication items were rated lower during the postintervention period, but no result was statistically significant. Both hospital‐environment domain items were rated lower during the postintervention period, and 1 result was statistically significant (quiet at night). This post hoc evaluation lends additional support to the potential benefit of the communication‐skills training program.

The findings from this study represent an important issue for leaders attempting to improve quality performance within their organizations. What level of proof is needed before investing time and effort in implementing an intervention? With mounting pressure to improve performance, leaders are often left to make informed decisions based on data that fall short of scientifically rigorous evidence. Importantly, an increase in composite doctor‐communication ratings from 76% to 79% would translate into an improvement from the 25th percentile to 50th‐percentile performance in the fiscal‐year 2011 Press Ganey University Healthcare Consortium benchmark comparison (based on surveys received from September 1, 2010, to August 31, 2011).[15]

Our study has several limitations. First, we assessed an intervention on a single service in a single hospital. Generalizability may be limited, as hospital medicine groups, hospitals, and the patients they serve vary. Second, our intervention was based on a framework (ie, AIDET) that has face validity but has not undergone extensive study to confirm that the underlying constructs, and the behaviors related to them, are tightly linked to patient satisfaction. Third, as previously mentioned, we were likely underpowered to detect a significant improvement in satisfaction resulting from our intervention. Incomplete participation in the training program may have also limited the effect of our intervention. Finally, our comparisons by hospitalist level of participation were based on the discharging physician. Attribution of a patient response to a single physician is problematic because many patients encounter more than 1 hospitalist and 1 or more specialist physicians during their stay.

CONCLUSION

In summary, we found improvements in patient satisfaction with doctor communication, which were not statistically significant, after implementation of a communication‐skills training program for hospitalists. Larger studies are needed to assess whether a communication‐skills training program can truly improve patient satisfaction with doctor communication and overall hospital care.

Hospital settings present unique challenges to patient‐clinician communication and collaboration. Patients frequently have multiple, active conditions. Interprofessional teams are large and care for multiple patients at the same time, and team membership is dynamic and dispersed. Moreover, physicians spend relatively little time with patients[1, 2] and seldom receive training in communication skills after medical school.

The Agency for Healthcare Research and Quality (AHRQ) has developed the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey to assess hospitalized patients' experiences with care.[3, 4, 5] Results are publicly reported on the US Department of Health and Human Services Hospital Compare Web site[6] and now affect hospital payment through the Center for Medicare and Medicaid Services Hospital Value‐Based Purchasing Program.[7]

Despite this increased transparency and accountability for performance related to the patient experience, little research has been conducted on how hospitals or clinicians might improve performance. Although interventions to enhance physician communication skills have shown improvements in observed behaviors, few studies have assessed benefit from the patient's perspective and few interventions have been integrated into practice.[8] We sought to assess the impact of a communication‐skills training program, based on a common framework used by hospitals, on patient satisfaction with doctor communication and overall hospital care.

METHODS

RESULTS

Overall, 61 (97%) of 63 hospitalists completed the first session, 44 (70%) completed the second session, and 25 (40%) completed the third session of program. Patient‐satisfaction data were available for 278 patients during the preintervention period and 186 patients during the postintervention period. Patient demographic characteristics were similar for the 2 periods (Table 2).

DISCUSSION

We found no significant improvement in patient satisfaction with doctor communication or overall rating of hospital care after implementation of a communication‐skills training program for hospitalists. There are several potential explanations for our results. First, though we used sound educational methods and attempted to replicate common clinical scenarios during simulation exercises, our program may not have resulted in improved communication behaviors during actual clinical care. We attempted to balance instructional methods that would result in behavioral change with a feasible investment of time and effort on the part of our learners (ie, practicing hospitalists). It is possible that additional time, feedback, and practice of communication skills would be necessary to change behaviors in the clinical setting. However, prior communication‐skills interventions have similarly struggled to show an impact on patient satisfaction.[13, 14] Second, we had incomplete participation in the program, with only 40% of hospitalists completing all 3 planned sessions. We encouraged all hospitalists, regardless of job type, to participate in the program. Participation rates were lower for 1‐year hospitalists compared with career hospitalists. The results of our analyses based on level of hospitalist participation in the training program, although not achieving statistical significance, suggest a greater effect of the program with higher degrees of participation.

Most important, the study was likely underpowered to detect a statistically significant difference in satisfaction results. Leaders were committed to providing communication‐skills training throughout our organization. We did not know the magnitude of potential improvement in satisfaction scores that might arise from our efforts, and therefore we did not conduct power calculations before designing and implementing the training program. Our HCAHPS composite doctor‐communication domain performance was 76% during the preintervention period and 79% during the postintervention period. Assuming an absolute 3% improvement is indeed possible, we would have needed >3000 patients in each period to have 80% power to detect a significant difference. Similarly, we would have needed >2000 patients during each period to have 80% power to detect an absolute 4% improvement in global rating of hospital care.

In an attempt to discern whether our favorable results were due to secular trends, we conducted post hoc analyses of HCAHPS nurse‐communication and hospital‐environment domains for the preintervention vs postintervention periods. Two of the 3 nurse‐communication items were rated lower during the postintervention period, but no result was statistically significant. Both hospital‐environment domain items were rated lower during the postintervention period, and 1 result was statistically significant (quiet at night). This post hoc evaluation lends additional support to the potential benefit of the communication‐skills training program.

The findings from this study represent an important issue for leaders attempting to improve quality performance within their organizations. What level of proof is needed before investing time and effort in implementing an intervention? With mounting pressure to improve performance, leaders are often left to make informed decisions based on data that fall short of scientifically rigorous evidence. Importantly, an increase in composite doctor‐communication ratings from 76% to 79% would translate into an improvement from the 25th percentile to 50th‐percentile performance in the fiscal‐year 2011 Press Ganey University Healthcare Consortium benchmark comparison (based on surveys received from September 1, 2010, to August 31, 2011).[15]

Our study has several limitations. First, we assessed an intervention on a single service in a single hospital. Generalizability may be limited, as hospital medicine groups, hospitals, and the patients they serve vary. Second, our intervention was based on a framework (ie, AIDET) that has face validity but has not undergone extensive study to confirm that the underlying constructs, and the behaviors related to them, are tightly linked to patient satisfaction. Third, as previously mentioned, we were likely underpowered to detect a significant improvement in satisfaction resulting from our intervention. Incomplete participation in the training program may have also limited the effect of our intervention. Finally, our comparisons by hospitalist level of participation were based on the discharging physician. Attribution of a patient response to a single physician is problematic because many patients encounter more than 1 hospitalist and 1 or more specialist physicians during their stay.

CONCLUSION

In summary, we found improvements in patient satisfaction with doctor communication, which were not statistically significant, after implementation of a communication‐skills training program for hospitalists. Larger studies are needed to assess whether a communication‐skills training program can truly improve patient satisfaction with doctor communication and overall hospital care.

Teamwork is essential to delivering safe and effective hospital care,15 yet the fluidity and geographic dispersion of team members in the hospital setting presents a significant barrier to teamwork.6 Physicians, nurses, and other hospital professionals frequently lack convenient and reliable opportunities to interact, and may struggle in efforts to discuss the care of their patients in person. Research studies show that nurses and physicians on patient care units do not communicate consistently and frequently do not agree on key aspects of their patients' plans of care.7, 8

Interdisciplinary rounds (IDR), also known as multidisciplinary rounds, provide a means to assemble hospital care team members and improve collaboration.913 Prior research on the use of IDR has demonstrated improved ratings of collaboration,11, 12 but inconsistent effects on length of stay and cost.10, 12, 13 Notably, the format, frequency, and duration of IDR in prior studies has been variable and no studies, to our knowledge, have evaluated teamwork performance during IDR. Lamb and colleagues conducted observations of cancer teams during multidisciplinary meetings.14 Trained observers used a validated observation tool to rate teamwork and found significant variation in performance by subteams. However, the study focused mainly on discussion among physician team members during meetings to plan longitudinal care for oncology patients.

We recently reported on the use of structured interdisciplinary rounds (SIDR) on 2 medical units in our hospital.15, 16 SIDR combines a structured format for communication, similar to a goals‐of‐care form,17, 18 with a forum for daily interdisciplinary meetings. Though no effect was seen on length of stay or cost, SIDR was associated with significantly higher ratings of the quality of collaboration and teamwork climate, and a reduction in the rate of adverse events.19 In March 2010, we implemented SIDR across all medical units in our hospital. We subjectively noted variation in teamwork performance during SIDR after a modification of nurse manager roles. We sought to evaluate teamwork during SIDR and to determine whether variation in performance existed and, if present, to characterize it.

METHODS

RESULTS

DISCUSSION

We found that the adapted OTAS instrument demonstrated acceptable reliability for assessing teamwork during SIDR across units, domains, and most subteams. Although teamwork scores during SIDR were generally high, we found variation in performance across units, domains, and subteams. Variation in performance is notable in light of our efforts to implement a consistent format for SIDR across units. Specifically, all units have similar timing, duration, frequency, and location of SIDR, use a structured communication tool for new patients, expect the same professions to be represented, and use coleaders to facilitate discussion. We believe teamwork within IDR likely varies across units in other hospitals, and perhaps to a larger degree, given the emphasis on purposeful design and implementation of SIDR in our hospital.

Our findings are important for several reasons. First, though commonly used in hospital settings, the effectiveness of IDR is seldom assessed. Hospitalists and other professionals may not be able to identify or characterize deficiencies in teamwork during IDR without objective assessment. The adapted OTAS instrument provides a useful tool to evaluate team performance during IDR. Second, professionals may conclude that the mere implementation of an intervention such as SIDR will improve teamwork ratings and improve patient safety. Importantly, published studies evaluating the benefits of SIDR reflected a pilot study occurring on 2 units.15, 16, 19 The current study emphasizes the need to ensure that interventions proven to be effective on a small scale are implemented consistently when put into place on a larger scale.

Despite good reliability for assessing teamwork during SIDR across units, domains, and most subteams, we found poor inter‐rater reliability for the physician subteam. The explanation for this finding is not entirely clear. We reviewed the anchors for the physician subteam behaviors and were unable to identify ambiguity in anchor definitions. An analysis of domain scores within the physician subteam did not reveal any specific pattern to explain the poor correlation.

We found that the leadership domain and coleader subteam received particularly low scores. The explanation for this finding likely relates to changes in the nurse management structure shortly before our study, which reduced attendance by nurse managers and created a need for clinical coordinators to take on a leadership role during SIDR. Although we provided simulation‐based training to unit medical directors and nurse managers prior to implementing SIDR in March 2010, clinical coordinators were not part of the initial training. Our study suggests a need to provide additional training to coleaders, including clinical coordinators, to enhance their ability to facilitate discussion in SIDR.

We found no difference in overall teamwork scores when comparing teaching service units to nonteaching hospitalist service units. Duration of SIDR was significantly longer on teaching service units, but there was no association between duration of discussion and overall team score. The difference in duration of SIDR is likely explained by less succinct discussions on the part of housestaff physicians compared to more experienced hospitalists. Importantly, the quality of input, and its impact on teamwork during SIDR, does not appear to suffer when physician discussion is less efficient.

Our study has several limitations. First, we evaluated IDR in a single, urban, academic institution, which may limit generalizability. Our version of IDR (ie, SIDR) was designed to improve teamwork and incorporate a structured communication tool with regularly held interdisciplinary meetings. Features of IDR may differ in other hospitals. Second, the high teamwork scores seen in our study may not be generalizable to hospitals which have used a less rigorous, less standardized approach to IDR. Third, SIDR did not include patients or caregivers. Research is needed to test strategies to include patients and caregivers as active team members and participants in clinical decisions during hospitalization. Finally, we used the term interdisciplinary rounds to be consistent with prior published research. The term interprofessional may be more appropriate, as it specifically describes interactions among members of different professions (eg, physicians, nurses, social workers) rather than among different disciplines within a profession (eg, cardiologists, hospitalists, surgeons).

In summary, we found that teamwork during IDR could be reliably assessed using an adapted OTAS instrument. Although scores were generally high, we found variation in performance across units and domains suggesting a need to improve consistency of teamwork performance across units, domains, and subteams. Our study fills an important gap in the literature. Although IDR is commonly used in hospitals, and research shows improvements in ratings of collaboration,11, 12 little if any research has evaluated teamwork during IDR. Beyond the mere implementation of IDR, our study suggests the need to confirm that teamwork is optimal and consistent. Furthermore, hospital leaders should consider specific training for clinicians leading discussion during IDR.

Teamwork is essential to delivering safe and effective hospital care,15 yet the fluidity and geographic dispersion of team members in the hospital setting presents a significant barrier to teamwork.6 Physicians, nurses, and other hospital professionals frequently lack convenient and reliable opportunities to interact, and may struggle in efforts to discuss the care of their patients in person. Research studies show that nurses and physicians on patient care units do not communicate consistently and frequently do not agree on key aspects of their patients' plans of care.7, 8

Interdisciplinary rounds (IDR), also known as multidisciplinary rounds, provide a means to assemble hospital care team members and improve collaboration.913 Prior research on the use of IDR has demonstrated improved ratings of collaboration,11, 12 but inconsistent effects on length of stay and cost.10, 12, 13 Notably, the format, frequency, and duration of IDR in prior studies has been variable and no studies, to our knowledge, have evaluated teamwork performance during IDR. Lamb and colleagues conducted observations of cancer teams during multidisciplinary meetings.14 Trained observers used a validated observation tool to rate teamwork and found significant variation in performance by subteams. However, the study focused mainly on discussion among physician team members during meetings to plan longitudinal care for oncology patients.

We recently reported on the use of structured interdisciplinary rounds (SIDR) on 2 medical units in our hospital.15, 16 SIDR combines a structured format for communication, similar to a goals‐of‐care form,17, 18 with a forum for daily interdisciplinary meetings. Though no effect was seen on length of stay or cost, SIDR was associated with significantly higher ratings of the quality of collaboration and teamwork climate, and a reduction in the rate of adverse events.19 In March 2010, we implemented SIDR across all medical units in our hospital. We subjectively noted variation in teamwork performance during SIDR after a modification of nurse manager roles. We sought to evaluate teamwork during SIDR and to determine whether variation in performance existed and, if present, to characterize it.

METHODS

RESULTS

DISCUSSION

We found that the adapted OTAS instrument demonstrated acceptable reliability for assessing teamwork during SIDR across units, domains, and most subteams. Although teamwork scores during SIDR were generally high, we found variation in performance across units, domains, and subteams. Variation in performance is notable in light of our efforts to implement a consistent format for SIDR across units. Specifically, all units have similar timing, duration, frequency, and location of SIDR, use a structured communication tool for new patients, expect the same professions to be represented, and use coleaders to facilitate discussion. We believe teamwork within IDR likely varies across units in other hospitals, and perhaps to a larger degree, given the emphasis on purposeful design and implementation of SIDR in our hospital.

Our findings are important for several reasons. First, though commonly used in hospital settings, the effectiveness of IDR is seldom assessed. Hospitalists and other professionals may not be able to identify or characterize deficiencies in teamwork during IDR without objective assessment. The adapted OTAS instrument provides a useful tool to evaluate team performance during IDR. Second, professionals may conclude that the mere implementation of an intervention such as SIDR will improve teamwork ratings and improve patient safety. Importantly, published studies evaluating the benefits of SIDR reflected a pilot study occurring on 2 units.15, 16, 19 The current study emphasizes the need to ensure that interventions proven to be effective on a small scale are implemented consistently when put into place on a larger scale.

Despite good reliability for assessing teamwork during SIDR across units, domains, and most subteams, we found poor inter‐rater reliability for the physician subteam. The explanation for this finding is not entirely clear. We reviewed the anchors for the physician subteam behaviors and were unable to identify ambiguity in anchor definitions. An analysis of domain scores within the physician subteam did not reveal any specific pattern to explain the poor correlation.

We found that the leadership domain and coleader subteam received particularly low scores. The explanation for this finding likely relates to changes in the nurse management structure shortly before our study, which reduced attendance by nurse managers and created a need for clinical coordinators to take on a leadership role during SIDR. Although we provided simulation‐based training to unit medical directors and nurse managers prior to implementing SIDR in March 2010, clinical coordinators were not part of the initial training. Our study suggests a need to provide additional training to coleaders, including clinical coordinators, to enhance their ability to facilitate discussion in SIDR.

We found no difference in overall teamwork scores when comparing teaching service units to nonteaching hospitalist service units. Duration of SIDR was significantly longer on teaching service units, but there was no association between duration of discussion and overall team score. The difference in duration of SIDR is likely explained by less succinct discussions on the part of housestaff physicians compared to more experienced hospitalists. Importantly, the quality of input, and its impact on teamwork during SIDR, does not appear to suffer when physician discussion is less efficient.

Our study has several limitations. First, we evaluated IDR in a single, urban, academic institution, which may limit generalizability. Our version of IDR (ie, SIDR) was designed to improve teamwork and incorporate a structured communication tool with regularly held interdisciplinary meetings. Features of IDR may differ in other hospitals. Second, the high teamwork scores seen in our study may not be generalizable to hospitals which have used a less rigorous, less standardized approach to IDR. Third, SIDR did not include patients or caregivers. Research is needed to test strategies to include patients and caregivers as active team members and participants in clinical decisions during hospitalization. Finally, we used the term interdisciplinary rounds to be consistent with prior published research. The term interprofessional may be more appropriate, as it specifically describes interactions among members of different professions (eg, physicians, nurses, social workers) rather than among different disciplines within a profession (eg, cardiologists, hospitalists, surgeons).

In summary, we found that teamwork during IDR could be reliably assessed using an adapted OTAS instrument. Although scores were generally high, we found variation in performance across units and domains suggesting a need to improve consistency of teamwork performance across units, domains, and subteams. Our study fills an important gap in the literature. Although IDR is commonly used in hospitals, and research shows improvements in ratings of collaboration,11, 12 little if any research has evaluated teamwork during IDR. Beyond the mere implementation of IDR, our study suggests the need to confirm that teamwork is optimal and consistent. Furthermore, hospital leaders should consider specific training for clinicians leading discussion during IDR.

Teamwork is essential to delivering safe and effective hospital care,15 yet the fluidity and geographic dispersion of team members in the hospital setting presents a significant barrier to teamwork.6 Physicians, nurses, and other hospital professionals frequently lack convenient and reliable opportunities to interact, and may struggle in efforts to discuss the care of their patients in person. Research studies show that nurses and physicians on patient care units do not communicate consistently and frequently do not agree on key aspects of their patients' plans of care.7, 8

Interdisciplinary rounds (IDR), also known as multidisciplinary rounds, provide a means to assemble hospital care team members and improve collaboration.913 Prior research on the use of IDR has demonstrated improved ratings of collaboration,11, 12 but inconsistent effects on length of stay and cost.10, 12, 13 Notably, the format, frequency, and duration of IDR in prior studies has been variable and no studies, to our knowledge, have evaluated teamwork performance during IDR. Lamb and colleagues conducted observations of cancer teams during multidisciplinary meetings.14 Trained observers used a validated observation tool to rate teamwork and found significant variation in performance by subteams. However, the study focused mainly on discussion among physician team members during meetings to plan longitudinal care for oncology patients.

We recently reported on the use of structured interdisciplinary rounds (SIDR) on 2 medical units in our hospital.15, 16 SIDR combines a structured format for communication, similar to a goals‐of‐care form,17, 18 with a forum for daily interdisciplinary meetings. Though no effect was seen on length of stay or cost, SIDR was associated with significantly higher ratings of the quality of collaboration and teamwork climate, and a reduction in the rate of adverse events.19 In March 2010, we implemented SIDR across all medical units in our hospital. We subjectively noted variation in teamwork performance during SIDR after a modification of nurse manager roles. We sought to evaluate teamwork during SIDR and to determine whether variation in performance existed and, if present, to characterize it.

METHODS

RESULTS

DISCUSSION

We found that the adapted OTAS instrument demonstrated acceptable reliability for assessing teamwork during SIDR across units, domains, and most subteams. Although teamwork scores during SIDR were generally high, we found variation in performance across units, domains, and subteams. Variation in performance is notable in light of our efforts to implement a consistent format for SIDR across units. Specifically, all units have similar timing, duration, frequency, and location of SIDR, use a structured communication tool for new patients, expect the same professions to be represented, and use coleaders to facilitate discussion. We believe teamwork within IDR likely varies across units in other hospitals, and perhaps to a larger degree, given the emphasis on purposeful design and implementation of SIDR in our hospital.

Our findings are important for several reasons. First, though commonly used in hospital settings, the effectiveness of IDR is seldom assessed. Hospitalists and other professionals may not be able to identify or characterize deficiencies in teamwork during IDR without objective assessment. The adapted OTAS instrument provides a useful tool to evaluate team performance during IDR. Second, professionals may conclude that the mere implementation of an intervention such as SIDR will improve teamwork ratings and improve patient safety. Importantly, published studies evaluating the benefits of SIDR reflected a pilot study occurring on 2 units.15, 16, 19 The current study emphasizes the need to ensure that interventions proven to be effective on a small scale are implemented consistently when put into place on a larger scale.

Despite good reliability for assessing teamwork during SIDR across units, domains, and most subteams, we found poor inter‐rater reliability for the physician subteam. The explanation for this finding is not entirely clear. We reviewed the anchors for the physician subteam behaviors and were unable to identify ambiguity in anchor definitions. An analysis of domain scores within the physician subteam did not reveal any specific pattern to explain the poor correlation.

We found that the leadership domain and coleader subteam received particularly low scores. The explanation for this finding likely relates to changes in the nurse management structure shortly before our study, which reduced attendance by nurse managers and created a need for clinical coordinators to take on a leadership role during SIDR. Although we provided simulation‐based training to unit medical directors and nurse managers prior to implementing SIDR in March 2010, clinical coordinators were not part of the initial training. Our study suggests a need to provide additional training to coleaders, including clinical coordinators, to enhance their ability to facilitate discussion in SIDR.

We found no difference in overall teamwork scores when comparing teaching service units to nonteaching hospitalist service units. Duration of SIDR was significantly longer on teaching service units, but there was no association between duration of discussion and overall team score. The difference in duration of SIDR is likely explained by less succinct discussions on the part of housestaff physicians compared to more experienced hospitalists. Importantly, the quality of input, and its impact on teamwork during SIDR, does not appear to suffer when physician discussion is less efficient.

Our study has several limitations. First, we evaluated IDR in a single, urban, academic institution, which may limit generalizability. Our version of IDR (ie, SIDR) was designed to improve teamwork and incorporate a structured communication tool with regularly held interdisciplinary meetings. Features of IDR may differ in other hospitals. Second, the high teamwork scores seen in our study may not be generalizable to hospitals which have used a less rigorous, less standardized approach to IDR. Third, SIDR did not include patients or caregivers. Research is needed to test strategies to include patients and caregivers as active team members and participants in clinical decisions during hospitalization. Finally, we used the term interdisciplinary rounds to be consistent with prior published research. The term interprofessional may be more appropriate, as it specifically describes interactions among members of different professions (eg, physicians, nurses, social workers) rather than among different disciplines within a profession (eg, cardiologists, hospitalists, surgeons).

In summary, we found that teamwork during IDR could be reliably assessed using an adapted OTAS instrument. Although scores were generally high, we found variation in performance across units and domains suggesting a need to improve consistency of teamwork performance across units, domains, and subteams. Our study fills an important gap in the literature. Although IDR is commonly used in hospitals, and research shows improvements in ratings of collaboration,11, 12 little if any research has evaluated teamwork during IDR. Beyond the mere implementation of IDR, our study suggests the need to confirm that teamwork is optimal and consistent. Furthermore, hospital leaders should consider specific training for clinicians leading discussion during IDR.

The discrepancy between what is taught about professionalism in formal medical education and what is witnessed in the hospital has received increasing attention.17 This latter aspect of medical education contributes to the hidden curriculum and impacts medical trainees' views on professionalism.8 The hidden curriculum refers to the lessons trainees learn through informal interactions within the multilayered educational learning environment.9 A growing body of work examines how the hidden curriculum and disruptive physicians impact the learning environment.9, 10 In response, regulatory agencies, such as the Liaison Committee on Medical Education (LCME) and Accreditation Council for Graduate Medical Education (ACGME), require training programs and medical schools to maintain standards of professionalism, and to regularly evaluate the learning environment and its impact on professionalism.11, 12 The ACGME in 2011 expanded its standards regarding professionalism by making certain that the program director and institution ensure a culture of professionalism that supports patient safety and personal responsibility.11 Given this increasing focus on professionalism in medical school and residency training programs, it is critical to examine faculty perceptions and actions that may perpetuate the discrepancy between the formal and hidden curriculum.13 This early exposure is especially significant because unprofessional behavior in medical school is strongly associated with later disciplinary action by a medical board.14, 15 Certain unprofessional behaviors can also compromise patient care and safety, and can detract from the hospital working environment.1618

In our previous work, we demonstrated that internal medicine interns reported increased participation in unprofessional behaviors regarding on‐call etiquette during internship.19, 20 Examples of these behaviors include refusing an admission (ie, blocking) and misrepresenting a test as urgent. Interestingly, students and residents have highlighted the powerful role of supervising faculty physicians in condoning or inhibiting such behavior. Given the increasing role of hospitalists as resident supervisors, it is important to consider the perceptions and actions of hospitalists with respect to perpetuating or hindering some unprofessional behaviors. Although hospital medicine is a relatively new specialty, many hospitalists are in frequent contact with medical trainees, perhaps because many residency programs and medical schools have a strong inpatient focus.2123 It is thus possible that hospitalists have a major influence on residents' behaviors and views of professionalism. In fact, the Society of Hospital Medicine's Core Competencies for Hospital Medicine explicitly state that hospitalists are expected to serve as a role model for professional and ethical conduct to house staff, medical students and other members of the interdisciplinary team.24

Therefore, the current study had 2 aims: first, to measure internal medicine hospitalists' perceptions of, and participation in, unprofessional behaviors using a previously validated survey; and second, to examine associations between job characteristics and participation in unprofessional behaviors.

METHODS

RESULTS

Seventy‐seven of the 101 hospitalists (76.2%) at 3 sites completed the survey. While response rates varied by site (site 1, 67%; site 2, 74%; site 3, 86%), the differences were not statistically significant (² = 2.9, P = 0.24). Most hospitalists (79.2%) completed residency after 2000. Over half (57.1%) of participants were male, and over half (61%) reported having worked with their current hospitalist group from 1 to 4 years. Almost 60% (59.7%) reported being unfamiliar with residents in the program. Over 40% of hospitalists did not do any night work. Hospitalists were largely clinical, one‐quarter of hospitalists reported working over 50% FTE, and the median was 80% FTE. While 78% of hospitalists reported some teaching time, median time on teaching service was low at 10% (Table 1).

Hospitalists perceived almost all behaviors as unprofessional (unprofessional or somewhat unprofessional on a 5‐point Likert Scale). The only behavior rated as professional with a mean of 4.25 (95% CI 4.014.49) was staying past shift limit to complete a patient‐care task that could have been signed out. This behavior also had the highest level of participation by hospitalists (81.7%). Hospitalists were most ambivalent when rating professionalism of attending an industry‐sponsored dinner or social event (mean 3.20, 95% CI 2.983.41) (Table 2).

Participation in egregious behaviors, such as falsifying patient records (6.49%) and performing medical or surgical procedures on a patient beyond self‐perceived level of skill (2.60%), was very low. The most common behaviors rated as unprofessional that hospitalists reported participating in were having nonmedical/personal conversations in patient corridors (67.1%), ordering a routine test as urgent to expedite care (62.3%), and making fun of other physicians to colleagues (40.3%). Forty percent of participants reported disparaging the emergency room (ER) team or primary care physician for findings later discovered, signing out over the phone when it could have been done in person, and texting or using smartphones during educational conferences. In particular, participation in unprofessional behaviors related to trainees was close to zero (eg, asking a student to discuss, with patients, medical or surgical information which is perceived to be beyond their level of knowledge). The least common behaviors that hospitalists reported participating in were discharging a patient before they are ready to go home in order to reduce one's census (2.56%) and reporting patient information as normal when uncertain of the true results (2.60%). Like previous studies of unprofessional behaviors, those that reported participation were less likely to report the behavior as unprofessional.8, 19

Observation of behaviors ranged from 4% to 80%. In all cases, observation of the behavior was reported at a higher level than participation. Correlation between observation and participation was also high, with the exception of a few behaviors that had zero or near zero participation rates (ie, reporting patient information as normal when unsure of true results.)

After performing factor analysis, 4 factors had eigenvalues greater than 1 and were therefore retained and extracted for further analysis. These 4 factors accounted for 76% of the variance in responses reported on the survey. By examining which items or groups of items most strongly loaded on each factor, the factors were named accordingly: factor 1 referred to behaviors related to making fun of others, factor 2 referred to workload management, factor 3 referred to behaviors related to the learning environment, and factor 4 referred to behaviors related to time pressure (Table 3).

Using site‐adjusted multivariate regression, certain hospitalist job characteristics were associated with certain patterns of participating in unprofessional behavior (Table 4). Those with less clinical time (<50% FTE) were more likely to participate in unprofessional behaviors related to making fun of others (factor 1, value = 0.94, 95% CI 0.32 to 1.56, P value <0.05). Hospitalists who had any administrative time ( value = 0.61, 95% CI 0.111.10, P value <0.05) were more likely to report participation in behaviors related to workload management. Hospitalists engaged in any night work were more likely to report participation in unprofessional behaviors related to time pressure ( value = 0.67, 95% CI 0.171.17, P value <0.05). Time devoted to teaching or research was not associated with greater participation in any of the domains of unprofessional behavior surveyed.

The only demographic characteristic that was significantly associated with unprofessional behavior was age. Specifically, those who were born after 1970 were more likely to participate in unprofessional behaviors related to workload management ( value = 0.87, 95% CI 0.071.67, P value <0.05). Site differences were also present. Specifically, one site was more likely to report participation in unprofessional behaviors related to workload management ( value site 1 = 1.01, 95% CI 0.15 to 1.86, P value <0.05), while another site was less likely to report participation in behaviors related to the learning environment ( value site 3 = 0.70, 95% CI 1.31 to 0.09, P value <0.05). Gender and familiarity with residents were not significant predictors of participation in unprofessional behaviors. Results remained robust in sensitivity analyses using different cutoffs of clinical time and teaching time.

DISCUSSION

This multisite study adds to what is known about the perceptions of, and participation in, unprofessional behaviors among internal medicine hospitalists. Hospitalists perceived almost all surveyed behaviors as unprofessional. Participation in egregious and trainee‐related unprofessional behaviors was very low. Four categories appeared to explain the variability in how hospitalists reported participation in unprofessional behaviors: making fun of others, workload management, learning environment, and time pressure. Participation in behaviors within these factors was associated with certain job characteristics, such as clinical time, administrative time, and night work, as well as age and site.

It is reassuring that participation in, and trainee‐related, unprofessional behaviors is very low, and it is noteworthy that attending an industry‐sponsored dinner is not considered unprofessional. This was surprising in the setting of increased external pressures to report and ban such interactions.28 Perception that attending such dinners is acceptable may reflect a lag between current practice and national recommendations.

It is important to explore why certain job characteristics are associated with participation in unprofessional behaviors. For example, those with less clinical time were more likely to participate in making fun of others. It may be the case that hospitalists with more clinical time may make a larger effort to develop and maintain positive relationships. Another possible explanation is that hospitalists with less clinical time are more easily influenced by those in the learning environment who make fun of others, such as residents who they are supervising for only a brief period.

For unprofessional behaviors related to workload management, those who were younger, and those with any administrative time, were more likely to participate in behaviors such as celebrating a blocked‐admission. Our prior work shows that behaviors related to workload management are more widespread in residency, and therefore younger hospitalists, who are often recent residency graduates, may be more prone to participating in these behaviors. While unproven, it is possible that those with more administrative time may have competing priorities with their administrative roles, which motivate them to more actively manage their workload, leading them to participate in workload management behaviors.

Hospitalists who did any night work were more likely to participate in unprofessional behaviors related to time pressure. This could reflect the high workloads that night hospitalists may face and the pressure they feel to wrap up work, resulting in a hasty handoff (ie, over the phone) or to defer work (ie, family questions). Site differences were also observed for participation in behaviors related to the learning environment, speaking to the importance of institutional culture.

It is worth mentioning that hospitalists who were teachers were not any less likely to report participating in certain behaviors. While 78% of hospitalists reported some level of teaching, the median reported percentage of teaching was 10% FTE. This level of teaching likely reflects the diverse nature of work in which hospitalists engage. While hospitalists spend some time working with trainees, services that are not staffed with residents (eg, uncovered services) are becoming increasingly common due to stricter resident duty hour restrictions. This may explain why 60% of hospitalists reported being unfamiliar with residents. We also used a high bar for familiarity, which we defined as knowing half of residents by name, and served as a proxy for those who may have trained at the institution where they currently work. In spite of hospitalists reporting a low fraction of their total clinical time devoted to resident services, a significant fraction of resident services were staffed by hospitalists at all sites, making them a natural target for interventions.

These results have implications for future work to assess and improve professionalism in the hospital learning environment. First, interventions to address unprofessional behaviors should focus on behaviors with the highest participation rates. Like our earlier studies of residents, participation is high in certain behaviors, such as misrepresenting a test as urgent, or disparaging the ER or primary care physician (PCP) for a missed finding.19, 20 While blocking an admission was common in our studies of residents, reported participation among hospitalists was low. Similar to a prior study of clinical year medical students at one of our sites, 1 in 5 hospitalists reported not correcting someone who mistakes a student for a physician, highlighting the role that hospitalists may have in perpetuating this behavior.8 Additionally, addressing the behaviors identified in this study, through novel curricular tools, may help to teach residents many of the interpersonal and communication skills called for in the 2011 ACGME Common Program Requirements.11 The ACGME requirements also include the expectation that faculty model how to manage their time before, during, and after clinical assignments, and recognize that transferring a patient to a rested provider is best. Given that most hospitalists believe staying past shift limit is professional, these requirements will be difficult to adopt without widespread culture change.

Moreover, interventions could be tailored to hospitalists with certain job characteristics. Interventions may be educational or systems based. An example of the former is stressing the impact of the learning and working environment on trainees, and an example of the latter is streamlining the process in which ordered tests are executed to result in a more timely completion of tests. This may result in fewer physicians misrepresenting a test as urgent in order to have the test done in a timely manner. Additionally, hospitalists with less clinical time could receive education on their impact as a role model for trainees. Hospitalists who are younger or with administrative commitments could be trained on the importance of avoiding behaviors related to workload management, such as blocking or turfing patients. Lastly, given the site differences, critical examination of institutional culture and policies is also important. With funding from the American Board of Internal Medicine (ABIM) Foundation, we are currently creating an educational intervention, targeting those behaviors that were most frequent among hospitalists and residents at our institutions to promote dialogue and critical reflection, with the hope of reducing the most prevalent behaviors encountered.

There are several limitations to this study. Despite the anonymity of the survey, participants may have inaccurately reported their participation in unprofessional behaviors due to socially desirable response. In addition, because we used factor analysis and multivariate regression models with a small sample size, item nonresponse limited the sample for regression analyses and raises the concern for response bias. However, all significant associations remained so after performing backwards stepwise elimination of covariates that were P > 0.10 in models that were larger (ranging from 65 to 69). Because we used self‐report and not direct observation of participation in unprofessional behaviors, it is not possible to validate the responses given. Future work could rely on the use of 360 degree evaluations or other methods to validate responses given by self‐report. It is also important to consider assessing whether these behaviors are associated with actual patient outcomes, such as length of stay or readmission. Some items may not always be unprofessional. For example, texting during an educational conference might be to advance care, which would not necessarily be unprofessional. The order in which the questions were asked could have led to bias. We asked about participation before perception to try to limit bias reporting in participation. Changing the order of these questions would potentially have resulted in under‐reporting participation in behaviors that one perceived to be unprofessional. This study was conducted at 3 institutions located in Chicago, limiting generalizability to institutions outside of this area. Only internal medicine hospitalists were surveyed, which also limits generalizability to other disciplines and specialties within internal medicine. Lastly, it is important to highlight that hospitalists are not the sole teachers on inpatient services, since residents encounter a variety of faculty who serve as teaching attendings. Future work should expand to other centers and other specialties.

In conclusion, in this multi‐institutional study of hospitalists, participation in egregious behaviors was low. Four factors or patterns underlie hospitalists' reports of participation in unprofessional behavior: making fun of others, learning environment, workload management, and time pressure. Job characteristics (clinical time, administrative time, night work), age, and site were all associated with different patterns of unprofessional behavior. Specifically, hospitalists with less clinical time were more likely to make fun of others. Hospitalists who were younger in age, as well as those who had any administrative work, were more likely to participate in behaviors related to workload management. Hospitalists who work nights were more likely to report behaviors related to time pressure. Interventions to promote professionalism should take institutional culture into account and should focus on behaviors with the highest participation rates. Efforts should also be made to address underlying reasons for participation in these behaviors.