Karen DeSalvo, MD

Consensus Conference Steering Committee

The consensus conference was developed first by the sponsoring professional societies (SGIM, SHM, and ACGIM) being asked to nominate 2 people to be part of the consensus conference steering committee. The steering committee's main functions were to identify key tasks for the consensus conference, invite consensus conference attendees, ensure adequate representation from all participating organizations, synthesize the results of the consensus conference, and work with the individual professional societies so that results from the consensus conference were acted upon in a coordinated and effective manner.

Consensus Conference Prework

The consensus conference co‐chairs convened a series of conference calls in the spring of 2007, during which the steering committee developed a series of key areas to be explored during the conference. Topic areas were selected on the basis of the group's expertise and referred to past work by AHM taskforces convened by both SGIM/ACGIM and SHM.

The steering committee then invited stakeholders from each invited society so that each professional organization would provide at least 1 representative with expertise appropriate to 1 of the key domains identified:

• 1

  Clinical and financial issues (within which topics such as optimal job descriptions and salary structures would be explored).

• 2

  Teaching and education mission (within which topics such as mentorship for AHM junior faculty might be discussed).

• 3

  Research and promotable activities (within which issues related to the development of promotable activities for AHM would be discussed).

Invitees to the consensus conference were assigned to one working group, given a general description of the potential areas within their domain, and instructed to consider a number of broad questions relevant to the topic area. These questions were as follows:

• 1

  What are the key barriers to AHM in each key domain?

• 2

  What viewpoints or barriers are the most pressing and/or actionable?

• 3

  What solutions could be implemented or initiated in the next 1 to 3 years?

In order to facilitate discussion, consensus conference invitees were provided copies of findings from the SGIM/ACGIM and SHM Academic Hospitalist Task Forces, preliminary results from a survey of AHM leaders, and key literature related to the field.

Consensus Conference Format

The AHM consensus conference followed a modified Delphi consensus‐building format, in which the members of each working group developed findings relevant to their area, presented these viewpoints back to the overall group for feedback, and returned to their working group to refine their initial recommendations or move on to subsequent areas.

We used Day 1 of the meeting to confirm and refine lists of key barriers and opportunities to AHM. On Day 2, we developed actionable solutions and identified barriers with no ready solution but which were felt to be worth highlighting.

Each cycle of feedback (1 on Day 1 and 2 on Day 2) was used to identify additional barriers or opportunities prespecified by the steering committee, prioritize issues/opportunities, clarify uncertainties or point them out when they existed, and identify new areas requiring consensus. Between each cycle, workflow and interim results were summarized by the co‐chairs and a professional meeting coordinator to ensure that the group felt consensus had been achieved and to identify where additional work was required.

Writing Group/Peer Review

After the consensus conference adjourned, minutes were circulated to the group and approved, whereupon a summary of the meeting was reformatted into manuscript form. The manuscript was circulated to the steering committee, consensus conference attendees, and 2 selected peer reviewers as an additional check on the external validity of the study's results.

Clinical and Financial Issues in AHM

The consensus group identified misalignment of the mission of hospitals (which often provide substantial financial support for hospital medicine programs) and the mission of departments of internal medicine (or divisions of general internal medicine) in which adult hospitalists reside as a fundamental barrier in AHM. Misalignment of missions produces challenges to the development of hospitalist groups in that their primary funder, the hospital, focuses on clinical care delivery, productivity, efficiency, and, in some cases, participation in patient safety and quality improvement efforts, whereas academic departments place considerable value on education, research, grants received, dissemination of scholarly work, and the national reputation of its faculty. Further exacerbating this tension is the fact that hospitalists do not always reside within traditional academic divisions (such as divisions of general medicine) and are therefore viewed by the hospital and their peers as hospital employees more than academic faculty.

As yet, few hospital medicine programs have successfully integrated academic and clinical needs. In many AHM programs, clinical demands have trumped academic pursuits and, as a result, produced jobs that have frequent turnover. This occurs most often when hospitalists are hired by academic medical centers primarily to staff nonresident services. Hospitalists who join these academic programs expecting ample opportunity to teach and pursue scholarly work often leave when they realize these jobs differ little from those in community settings (with the exception of less pay and, in most cases, a less efficient clinical delivery system). This turnover contributes to the perception of hospitalists as transient nonacademic faculty. The participants felt that we needed to define the ideal academic hospitalist job description.

Teaching and Education Mission in AHM

Traditionally, faculty in academic medical centers have had prominent roles in resident teaching services, supervising medical residents, interns, and students. Hospitalists fill these roles at some institutions and in many cases have replaced senior faculty who are no longer able (because of competing demands from clinics or labs) or willing (because of an increased need for oversight and availability) to staff the teaching service. The teaching hospitalists start at these positions straight out of residency with little experience, training, or mentoring in how to succeed as a clinician educator. The creation of nonresident hospitalist services to address residency work hour requirements has removed many hospitalists from teaching opportunities as these services often have few if any teaching opportunities. The consensus group identified the lack of teaching opportunities and a lack of any formal preparation for those who do teach as the key challenges for new hospitalist clinician educators.

Research and Promotable Activities in AHM

Numerous challenges to promotion and success in hospital medicine research were identified. Most conference attendees felt that chairs of departments of medicine do not fully understand what the roles of academic hospitalists are, how they fit into the department's mission, or what is needed to better integrate hospitalists into the research and academic activities of the department. In addition, there are few hospital medicine fellowship programs, and those that have been created focus primarily on improving teaching skills or quality improvement rather than on research or the development of academic products. Aspiring academic hospitalists could pursue research fellowship training in existing programs (ie, the Robert Wood Johnson Foundation), but few graduates currently pursue these opportunities, and federally funded fellowships (eg, the National Research Service Awards and Health Resources and Services Administration T32 awards) explicitly exclude physicians who are not focused on primary care research. The group noted that a number of Veterans Administration fellowships (such as the Quality Scholars programs) may provide avenues for the training of hospital medicinefocused researchers, but they have been underused.

For researchers who focus on hospital medicine, federal funding sources are limited for both career development awards (K‐series) and later (R‐series) grants, particularly those funding the quality and safety research that hospitalists often pursue. Agencies of the National Institutes of Health currently do not provide many opportunities for hospital‐based general internal medicine research, and thus academic hospitalist research is undervalued by many promotion committees.

Cross‐Cutting Issues

Challenges in leadership and mentorship were identified as cross‐cutting. Many AHM programs are young, and so are their leaders. As a result, hospital medicine leaders often lack the experience and skills necessary to successfully negotiate for the support that is critical for the ideal program's success. As a young field, hospital medicine lacks faculty who have succeeded in careers as hospitalists, have been promoted in tenure tracks, and can mentor and guide young faculty through the complexities of academic medicine. Absent leadership and mentoring, few hospital medicine programs will succeed in traditional academic pursuits.

Outreach to and Education of Stakeholders in Academic Medicine

The focus of the educational and outreach efforts suggested by the consensus group is to help leaders in academic medicine (not just AHM) and academic medical centers understand the challenges facing AHM. More importantly, efforts should reinforce the value of academic hospitalists to their hospital, department, and division. Efforts to engage these critical stakeholders to discuss and potentially address a number of the conference's proposed solutions are needed. Leaders include deans of medical schools, chairs of departments of medicine, division chiefs, and hospital administrative leadership.

Suggested outreach and educational activities included the publication of articles in key journals with the goal of increasing the visibility of AHM in professional societies as well as meetings and workshops focusing on teaching hospitalists and academic leaders methods to overcome challenges. Professional societies with a stake in AHM should better understand the challenges and position themselves to address these issues. The AHM task forces of SHM and SGIM can help give academic hospitalists a voice in having their needs addressed.

Defining a Sustainable Job Description for Academic Hospitalists

The group strongly endorsed the need for transparent and readily available data aimed at developing sustainable academic hospitalist positions. For example, required information would include how academic jobs are constructed (in terms of months on service per year and the number of nights or weekends of coverage) and what successful programs and their hospitalists have found to be acceptable. Over the longer term, empiric comparisons based on key metrics are needed to not only help guide career development and retention but also facilitate negotiations for programmatic support.

The group pointed out that embedded in delineating an optimal academic hospitalist job description is the longstanding work of general medicine societies in supporting and fostering the development of clinician educators. In many ways, the pressures of academic physicians to be mostly clinician and less educator versus someone who focuses heavily on educational work is similar for hospitalists and outpatient generalists. Academic general internal medicine divisions hired many general internists in the early 1990s to expand the reach of academic medical centers and increase the outpatient base.6 Many university hospitals are now hiring hospitalists to provide the inpatient care for these patients, but residency work hour reductions have added a layer of complexity, creating the need for entirely new roles for academic generalists (such as surgical comanagement of medically complex patients).7, 8 Past experiences in refining and reinforcing education as a key function (http://sgim.org/clinicianteach.cfm) would provide a template for future activities. Again, in the context of a rapidly growing number of nonteaching services, these descriptions have great importance.

Development of a Quality‐Improvement Portfolio Akin to an Educator's Portfolio

Many hospitalists actively participate in administrative work related to quality improvement activities, and we should develop this additional pathway for promotable academic activities (eg, clinician administrator); however, such a pathway may not be recognized by all promotion committees. The group observed that many aspects of quality improvement are similar to those of education (eg, developing a curriculum, leading a team, evaluating a process, defining generalizability, and disseminating locally proven interventions) and as such would be amenable to the development of a quality improvement portfolio, which candidates could submit to promotion committees. Again, past work in developing the importance and value of the educator's portfolio would facilitate the development of a quality improvement portfolio, which would require endorsement from key stakeholders (eg, the Association of Professors of Medicine, SGIM, and SHM).9 Importantly, this work may also benefit many outpatient‐based generalists who are increasingly focusing their careers on quality and safety improvement.

Developing Mentoring and Training Opportunities for Newly Hired and Junior Hospitalists

We reached a strong consensus about the need to develop a retreat‐format training opportunity by which junior academic hospitalists would be able to gain training in tasks critical to early‐career success. These were envisioned as an initial 2‐ to 3‐day meeting followed by mentorship at a distance and continued collaboration within the class of attendees. Topics would include key functions in AHM, such as becoming an effective attending physician and teacher, leadership, quality improvement, the business of medicine, effective billing, and maintaining a curriculum vitae. A number of professional societies have developed leadership or mentoring retreats, and at the time of this article's preparation, both regional and national efforts were underway to develop these products.

Developing Training and Mentorship Pathways for Hospitalist Researchers

There are few funded hospitalist researchers in the midcareer phase and a small but growing number of academic hospitalists entering the field with a focus on research. Enhancing a pipeline of researchers is a critical need for the field, as cementing AHM as an equal member of the academic medical community will be predicated on the successful development of hospitalist investigators. To this end, academic hospitalist groups should be encouraged to partner with other established research units (particularly general internal medicine) to create mentoring relationships and increase collaborative activities. The emergence of the Clinical and Translational Science Awards consortium sites, with a focus on implementation and effectiveness research, may also provide local opportunities for hospitalists to partner in research important for early‐career grant submission. Furthermore, building the pipeline of academic hospitalist researchers will require a strong focus on identifying students and residents through outreach at individual sites as well as presentations at national meetings (eg, the American College of Physicians).

Two other issues were also thought to be important. First, professional societies should work to encourage funders of primary carefocused general medicine training programs (the National Research Service Awards and the Health Resources and Services Administration) to allow hospitalists to qualify for such critical research training. Second, continuing to advocate for increasing funding for implementation and effectiveness research, via either the Agency for Healthcare Research and Quality or individual agencies of the National Institutes of Health, will be key; the emergence of a medical effectiveness institute would also be a potential boon.

Improving Relationships Between the Professional Homes of Academic Generalists

Relationships between outpatient‐based general medicine and hospital medicine were rocky as the field of AHM first took shape, and some residua of initial tensions persist a decade later. These tensions persist in part because hospitalists remain underdeveloped members of the academic community, and this perhaps gives some license to aver that hospitalists are merely transient faculty in a stage between residency and fellowship hired to improve throughput.

Overcoming this perception will require more engagement between academic generalists of all types, not less. The consensus group felt strongly that there need not be a single professional home for academic hospitalists and that generalists should be willing and even encouraged to self‐identify as hospital‐ or clinic‐focused, much as they might be geriatrics‐focused, informatics‐focused, or women's healthfocused. In fact, in some academic centers, a few generalists have successfully integrated themselves into both clinic‐based and hospitalist roles. In this way, the emergence and growth of AHM should be viewed as a boon to the practice of general medicine, not a challenge.

Resources

Much of what is proposed to enhance AHM will require resources. Academic hospitals have a vested interest in supporting AHM as a way to reduce turnover in a group that is increasingly critical for hospital operations, not to mention key leadership roles. Negotiating for these resources should emphasize that hospitals benefit directly from the revenue and margin that comes from incremental hospital admissions, collect most of the federal graduate medical education dollars, and benefit from improved care processes that are a result of hospitalist quality improvement efforts.

Deans and Departments, a key audience for the conference findings, also have a clear stake in fostering a less transient, more professionally satisfied and academically successful work force, particularly when hospitalists are increasingly the key educators of medical residents. Moreover, schools have a vested interest in the academic accomplishments and national reputation of their hospitalists. The financial arrangements will be unique to each setting and institution, and it is clear that the sources to be tapped will vary from site to site, but these resources are clearly necessary for the field.

Consensus Conference Steering Committee

The consensus conference was developed first by the sponsoring professional societies (SGIM, SHM, and ACGIM) being asked to nominate 2 people to be part of the consensus conference steering committee. The steering committee's main functions were to identify key tasks for the consensus conference, invite consensus conference attendees, ensure adequate representation from all participating organizations, synthesize the results of the consensus conference, and work with the individual professional societies so that results from the consensus conference were acted upon in a coordinated and effective manner.

Consensus Conference Prework

The consensus conference co‐chairs convened a series of conference calls in the spring of 2007, during which the steering committee developed a series of key areas to be explored during the conference. Topic areas were selected on the basis of the group's expertise and referred to past work by AHM taskforces convened by both SGIM/ACGIM and SHM.

The steering committee then invited stakeholders from each invited society so that each professional organization would provide at least 1 representative with expertise appropriate to 1 of the key domains identified:

• 1

  Clinical and financial issues (within which topics such as optimal job descriptions and salary structures would be explored).

• 2

  Teaching and education mission (within which topics such as mentorship for AHM junior faculty might be discussed).

• 3

  Research and promotable activities (within which issues related to the development of promotable activities for AHM would be discussed).

Invitees to the consensus conference were assigned to one working group, given a general description of the potential areas within their domain, and instructed to consider a number of broad questions relevant to the topic area. These questions were as follows:

• 1

  What are the key barriers to AHM in each key domain?

• 2

  What viewpoints or barriers are the most pressing and/or actionable?

• 3

  What solutions could be implemented or initiated in the next 1 to 3 years?

In order to facilitate discussion, consensus conference invitees were provided copies of findings from the SGIM/ACGIM and SHM Academic Hospitalist Task Forces, preliminary results from a survey of AHM leaders, and key literature related to the field.

Consensus Conference Format

The AHM consensus conference followed a modified Delphi consensus‐building format, in which the members of each working group developed findings relevant to their area, presented these viewpoints back to the overall group for feedback, and returned to their working group to refine their initial recommendations or move on to subsequent areas.

We used Day 1 of the meeting to confirm and refine lists of key barriers and opportunities to AHM. On Day 2, we developed actionable solutions and identified barriers with no ready solution but which were felt to be worth highlighting.

Each cycle of feedback (1 on Day 1 and 2 on Day 2) was used to identify additional barriers or opportunities prespecified by the steering committee, prioritize issues/opportunities, clarify uncertainties or point them out when they existed, and identify new areas requiring consensus. Between each cycle, workflow and interim results were summarized by the co‐chairs and a professional meeting coordinator to ensure that the group felt consensus had been achieved and to identify where additional work was required.

Writing Group/Peer Review

After the consensus conference adjourned, minutes were circulated to the group and approved, whereupon a summary of the meeting was reformatted into manuscript form. The manuscript was circulated to the steering committee, consensus conference attendees, and 2 selected peer reviewers as an additional check on the external validity of the study's results.

Clinical and Financial Issues in AHM

The consensus group identified misalignment of the mission of hospitals (which often provide substantial financial support for hospital medicine programs) and the mission of departments of internal medicine (or divisions of general internal medicine) in which adult hospitalists reside as a fundamental barrier in AHM. Misalignment of missions produces challenges to the development of hospitalist groups in that their primary funder, the hospital, focuses on clinical care delivery, productivity, efficiency, and, in some cases, participation in patient safety and quality improvement efforts, whereas academic departments place considerable value on education, research, grants received, dissemination of scholarly work, and the national reputation of its faculty. Further exacerbating this tension is the fact that hospitalists do not always reside within traditional academic divisions (such as divisions of general medicine) and are therefore viewed by the hospital and their peers as hospital employees more than academic faculty.

As yet, few hospital medicine programs have successfully integrated academic and clinical needs. In many AHM programs, clinical demands have trumped academic pursuits and, as a result, produced jobs that have frequent turnover. This occurs most often when hospitalists are hired by academic medical centers primarily to staff nonresident services. Hospitalists who join these academic programs expecting ample opportunity to teach and pursue scholarly work often leave when they realize these jobs differ little from those in community settings (with the exception of less pay and, in most cases, a less efficient clinical delivery system). This turnover contributes to the perception of hospitalists as transient nonacademic faculty. The participants felt that we needed to define the ideal academic hospitalist job description.

Teaching and Education Mission in AHM

Traditionally, faculty in academic medical centers have had prominent roles in resident teaching services, supervising medical residents, interns, and students. Hospitalists fill these roles at some institutions and in many cases have replaced senior faculty who are no longer able (because of competing demands from clinics or labs) or willing (because of an increased need for oversight and availability) to staff the teaching service. The teaching hospitalists start at these positions straight out of residency with little experience, training, or mentoring in how to succeed as a clinician educator. The creation of nonresident hospitalist services to address residency work hour requirements has removed many hospitalists from teaching opportunities as these services often have few if any teaching opportunities. The consensus group identified the lack of teaching opportunities and a lack of any formal preparation for those who do teach as the key challenges for new hospitalist clinician educators.

Research and Promotable Activities in AHM

Numerous challenges to promotion and success in hospital medicine research were identified. Most conference attendees felt that chairs of departments of medicine do not fully understand what the roles of academic hospitalists are, how they fit into the department's mission, or what is needed to better integrate hospitalists into the research and academic activities of the department. In addition, there are few hospital medicine fellowship programs, and those that have been created focus primarily on improving teaching skills or quality improvement rather than on research or the development of academic products. Aspiring academic hospitalists could pursue research fellowship training in existing programs (ie, the Robert Wood Johnson Foundation), but few graduates currently pursue these opportunities, and federally funded fellowships (eg, the National Research Service Awards and Health Resources and Services Administration T32 awards) explicitly exclude physicians who are not focused on primary care research. The group noted that a number of Veterans Administration fellowships (such as the Quality Scholars programs) may provide avenues for the training of hospital medicinefocused researchers, but they have been underused.

For researchers who focus on hospital medicine, federal funding sources are limited for both career development awards (K‐series) and later (R‐series) grants, particularly those funding the quality and safety research that hospitalists often pursue. Agencies of the National Institutes of Health currently do not provide many opportunities for hospital‐based general internal medicine research, and thus academic hospitalist research is undervalued by many promotion committees.

Cross‐Cutting Issues

Challenges in leadership and mentorship were identified as cross‐cutting. Many AHM programs are young, and so are their leaders. As a result, hospital medicine leaders often lack the experience and skills necessary to successfully negotiate for the support that is critical for the ideal program's success. As a young field, hospital medicine lacks faculty who have succeeded in careers as hospitalists, have been promoted in tenure tracks, and can mentor and guide young faculty through the complexities of academic medicine. Absent leadership and mentoring, few hospital medicine programs will succeed in traditional academic pursuits.

Outreach to and Education of Stakeholders in Academic Medicine

The focus of the educational and outreach efforts suggested by the consensus group is to help leaders in academic medicine (not just AHM) and academic medical centers understand the challenges facing AHM. More importantly, efforts should reinforce the value of academic hospitalists to their hospital, department, and division. Efforts to engage these critical stakeholders to discuss and potentially address a number of the conference's proposed solutions are needed. Leaders include deans of medical schools, chairs of departments of medicine, division chiefs, and hospital administrative leadership.

Suggested outreach and educational activities included the publication of articles in key journals with the goal of increasing the visibility of AHM in professional societies as well as meetings and workshops focusing on teaching hospitalists and academic leaders methods to overcome challenges. Professional societies with a stake in AHM should better understand the challenges and position themselves to address these issues. The AHM task forces of SHM and SGIM can help give academic hospitalists a voice in having their needs addressed.

Defining a Sustainable Job Description for Academic Hospitalists

The group strongly endorsed the need for transparent and readily available data aimed at developing sustainable academic hospitalist positions. For example, required information would include how academic jobs are constructed (in terms of months on service per year and the number of nights or weekends of coverage) and what successful programs and their hospitalists have found to be acceptable. Over the longer term, empiric comparisons based on key metrics are needed to not only help guide career development and retention but also facilitate negotiations for programmatic support.

The group pointed out that embedded in delineating an optimal academic hospitalist job description is the longstanding work of general medicine societies in supporting and fostering the development of clinician educators. In many ways, the pressures of academic physicians to be mostly clinician and less educator versus someone who focuses heavily on educational work is similar for hospitalists and outpatient generalists. Academic general internal medicine divisions hired many general internists in the early 1990s to expand the reach of academic medical centers and increase the outpatient base.6 Many university hospitals are now hiring hospitalists to provide the inpatient care for these patients, but residency work hour reductions have added a layer of complexity, creating the need for entirely new roles for academic generalists (such as surgical comanagement of medically complex patients).7, 8 Past experiences in refining and reinforcing education as a key function (http://sgim.org/clinicianteach.cfm) would provide a template for future activities. Again, in the context of a rapidly growing number of nonteaching services, these descriptions have great importance.

Development of a Quality‐Improvement Portfolio Akin to an Educator's Portfolio

Many hospitalists actively participate in administrative work related to quality improvement activities, and we should develop this additional pathway for promotable academic activities (eg, clinician administrator); however, such a pathway may not be recognized by all promotion committees. The group observed that many aspects of quality improvement are similar to those of education (eg, developing a curriculum, leading a team, evaluating a process, defining generalizability, and disseminating locally proven interventions) and as such would be amenable to the development of a quality improvement portfolio, which candidates could submit to promotion committees. Again, past work in developing the importance and value of the educator's portfolio would facilitate the development of a quality improvement portfolio, which would require endorsement from key stakeholders (eg, the Association of Professors of Medicine, SGIM, and SHM).9 Importantly, this work may also benefit many outpatient‐based generalists who are increasingly focusing their careers on quality and safety improvement.

Developing Mentoring and Training Opportunities for Newly Hired and Junior Hospitalists

We reached a strong consensus about the need to develop a retreat‐format training opportunity by which junior academic hospitalists would be able to gain training in tasks critical to early‐career success. These were envisioned as an initial 2‐ to 3‐day meeting followed by mentorship at a distance and continued collaboration within the class of attendees. Topics would include key functions in AHM, such as becoming an effective attending physician and teacher, leadership, quality improvement, the business of medicine, effective billing, and maintaining a curriculum vitae. A number of professional societies have developed leadership or mentoring retreats, and at the time of this article's preparation, both regional and national efforts were underway to develop these products.

Developing Training and Mentorship Pathways for Hospitalist Researchers

There are few funded hospitalist researchers in the midcareer phase and a small but growing number of academic hospitalists entering the field with a focus on research. Enhancing a pipeline of researchers is a critical need for the field, as cementing AHM as an equal member of the academic medical community will be predicated on the successful development of hospitalist investigators. To this end, academic hospitalist groups should be encouraged to partner with other established research units (particularly general internal medicine) to create mentoring relationships and increase collaborative activities. The emergence of the Clinical and Translational Science Awards consortium sites, with a focus on implementation and effectiveness research, may also provide local opportunities for hospitalists to partner in research important for early‐career grant submission. Furthermore, building the pipeline of academic hospitalist researchers will require a strong focus on identifying students and residents through outreach at individual sites as well as presentations at national meetings (eg, the American College of Physicians).

Two other issues were also thought to be important. First, professional societies should work to encourage funders of primary carefocused general medicine training programs (the National Research Service Awards and the Health Resources and Services Administration) to allow hospitalists to qualify for such critical research training. Second, continuing to advocate for increasing funding for implementation and effectiveness research, via either the Agency for Healthcare Research and Quality or individual agencies of the National Institutes of Health, will be key; the emergence of a medical effectiveness institute would also be a potential boon.

Improving Relationships Between the Professional Homes of Academic Generalists

Relationships between outpatient‐based general medicine and hospital medicine were rocky as the field of AHM first took shape, and some residua of initial tensions persist a decade later. These tensions persist in part because hospitalists remain underdeveloped members of the academic community, and this perhaps gives some license to aver that hospitalists are merely transient faculty in a stage between residency and fellowship hired to improve throughput.

Overcoming this perception will require more engagement between academic generalists of all types, not less. The consensus group felt strongly that there need not be a single professional home for academic hospitalists and that generalists should be willing and even encouraged to self‐identify as hospital‐ or clinic‐focused, much as they might be geriatrics‐focused, informatics‐focused, or women's healthfocused. In fact, in some academic centers, a few generalists have successfully integrated themselves into both clinic‐based and hospitalist roles. In this way, the emergence and growth of AHM should be viewed as a boon to the practice of general medicine, not a challenge.

Resources

Much of what is proposed to enhance AHM will require resources. Academic hospitals have a vested interest in supporting AHM as a way to reduce turnover in a group that is increasingly critical for hospital operations, not to mention key leadership roles. Negotiating for these resources should emphasize that hospitals benefit directly from the revenue and margin that comes from incremental hospital admissions, collect most of the federal graduate medical education dollars, and benefit from improved care processes that are a result of hospitalist quality improvement efforts.

Deans and Departments, a key audience for the conference findings, also have a clear stake in fostering a less transient, more professionally satisfied and academically successful work force, particularly when hospitalists are increasingly the key educators of medical residents. Moreover, schools have a vested interest in the academic accomplishments and national reputation of their hospitalists. The financial arrangements will be unique to each setting and institution, and it is clear that the sources to be tapped will vary from site to site, but these resources are clearly necessary for the field.

Consensus Conference Steering Committee

The consensus conference was developed first by the sponsoring professional societies (SGIM, SHM, and ACGIM) being asked to nominate 2 people to be part of the consensus conference steering committee. The steering committee's main functions were to identify key tasks for the consensus conference, invite consensus conference attendees, ensure adequate representation from all participating organizations, synthesize the results of the consensus conference, and work with the individual professional societies so that results from the consensus conference were acted upon in a coordinated and effective manner.

Consensus Conference Prework

The consensus conference co‐chairs convened a series of conference calls in the spring of 2007, during which the steering committee developed a series of key areas to be explored during the conference. Topic areas were selected on the basis of the group's expertise and referred to past work by AHM taskforces convened by both SGIM/ACGIM and SHM.

The steering committee then invited stakeholders from each invited society so that each professional organization would provide at least 1 representative with expertise appropriate to 1 of the key domains identified:

• 1

  Clinical and financial issues (within which topics such as optimal job descriptions and salary structures would be explored).

• 2

  Teaching and education mission (within which topics such as mentorship for AHM junior faculty might be discussed).

• 3

  Research and promotable activities (within which issues related to the development of promotable activities for AHM would be discussed).

Invitees to the consensus conference were assigned to one working group, given a general description of the potential areas within their domain, and instructed to consider a number of broad questions relevant to the topic area. These questions were as follows:

• 1

  What are the key barriers to AHM in each key domain?

• 2

  What viewpoints or barriers are the most pressing and/or actionable?

• 3

  What solutions could be implemented or initiated in the next 1 to 3 years?

In order to facilitate discussion, consensus conference invitees were provided copies of findings from the SGIM/ACGIM and SHM Academic Hospitalist Task Forces, preliminary results from a survey of AHM leaders, and key literature related to the field.

Consensus Conference Format

The AHM consensus conference followed a modified Delphi consensus‐building format, in which the members of each working group developed findings relevant to their area, presented these viewpoints back to the overall group for feedback, and returned to their working group to refine their initial recommendations or move on to subsequent areas.

We used Day 1 of the meeting to confirm and refine lists of key barriers and opportunities to AHM. On Day 2, we developed actionable solutions and identified barriers with no ready solution but which were felt to be worth highlighting.

Each cycle of feedback (1 on Day 1 and 2 on Day 2) was used to identify additional barriers or opportunities prespecified by the steering committee, prioritize issues/opportunities, clarify uncertainties or point them out when they existed, and identify new areas requiring consensus. Between each cycle, workflow and interim results were summarized by the co‐chairs and a professional meeting coordinator to ensure that the group felt consensus had been achieved and to identify where additional work was required.

Writing Group/Peer Review

After the consensus conference adjourned, minutes were circulated to the group and approved, whereupon a summary of the meeting was reformatted into manuscript form. The manuscript was circulated to the steering committee, consensus conference attendees, and 2 selected peer reviewers as an additional check on the external validity of the study's results.

Clinical and Financial Issues in AHM

The consensus group identified misalignment of the mission of hospitals (which often provide substantial financial support for hospital medicine programs) and the mission of departments of internal medicine (or divisions of general internal medicine) in which adult hospitalists reside as a fundamental barrier in AHM. Misalignment of missions produces challenges to the development of hospitalist groups in that their primary funder, the hospital, focuses on clinical care delivery, productivity, efficiency, and, in some cases, participation in patient safety and quality improvement efforts, whereas academic departments place considerable value on education, research, grants received, dissemination of scholarly work, and the national reputation of its faculty. Further exacerbating this tension is the fact that hospitalists do not always reside within traditional academic divisions (such as divisions of general medicine) and are therefore viewed by the hospital and their peers as hospital employees more than academic faculty.

As yet, few hospital medicine programs have successfully integrated academic and clinical needs. In many AHM programs, clinical demands have trumped academic pursuits and, as a result, produced jobs that have frequent turnover. This occurs most often when hospitalists are hired by academic medical centers primarily to staff nonresident services. Hospitalists who join these academic programs expecting ample opportunity to teach and pursue scholarly work often leave when they realize these jobs differ little from those in community settings (with the exception of less pay and, in most cases, a less efficient clinical delivery system). This turnover contributes to the perception of hospitalists as transient nonacademic faculty. The participants felt that we needed to define the ideal academic hospitalist job description.

Teaching and Education Mission in AHM

Traditionally, faculty in academic medical centers have had prominent roles in resident teaching services, supervising medical residents, interns, and students. Hospitalists fill these roles at some institutions and in many cases have replaced senior faculty who are no longer able (because of competing demands from clinics or labs) or willing (because of an increased need for oversight and availability) to staff the teaching service. The teaching hospitalists start at these positions straight out of residency with little experience, training, or mentoring in how to succeed as a clinician educator. The creation of nonresident hospitalist services to address residency work hour requirements has removed many hospitalists from teaching opportunities as these services often have few if any teaching opportunities. The consensus group identified the lack of teaching opportunities and a lack of any formal preparation for those who do teach as the key challenges for new hospitalist clinician educators.

Research and Promotable Activities in AHM

Numerous challenges to promotion and success in hospital medicine research were identified. Most conference attendees felt that chairs of departments of medicine do not fully understand what the roles of academic hospitalists are, how they fit into the department's mission, or what is needed to better integrate hospitalists into the research and academic activities of the department. In addition, there are few hospital medicine fellowship programs, and those that have been created focus primarily on improving teaching skills or quality improvement rather than on research or the development of academic products. Aspiring academic hospitalists could pursue research fellowship training in existing programs (ie, the Robert Wood Johnson Foundation), but few graduates currently pursue these opportunities, and federally funded fellowships (eg, the National Research Service Awards and Health Resources and Services Administration T32 awards) explicitly exclude physicians who are not focused on primary care research. The group noted that a number of Veterans Administration fellowships (such as the Quality Scholars programs) may provide avenues for the training of hospital medicinefocused researchers, but they have been underused.

For researchers who focus on hospital medicine, federal funding sources are limited for both career development awards (K‐series) and later (R‐series) grants, particularly those funding the quality and safety research that hospitalists often pursue. Agencies of the National Institutes of Health currently do not provide many opportunities for hospital‐based general internal medicine research, and thus academic hospitalist research is undervalued by many promotion committees.

Cross‐Cutting Issues

Challenges in leadership and mentorship were identified as cross‐cutting. Many AHM programs are young, and so are their leaders. As a result, hospital medicine leaders often lack the experience and skills necessary to successfully negotiate for the support that is critical for the ideal program's success. As a young field, hospital medicine lacks faculty who have succeeded in careers as hospitalists, have been promoted in tenure tracks, and can mentor and guide young faculty through the complexities of academic medicine. Absent leadership and mentoring, few hospital medicine programs will succeed in traditional academic pursuits.

Outreach to and Education of Stakeholders in Academic Medicine

The focus of the educational and outreach efforts suggested by the consensus group is to help leaders in academic medicine (not just AHM) and academic medical centers understand the challenges facing AHM. More importantly, efforts should reinforce the value of academic hospitalists to their hospital, department, and division. Efforts to engage these critical stakeholders to discuss and potentially address a number of the conference's proposed solutions are needed. Leaders include deans of medical schools, chairs of departments of medicine, division chiefs, and hospital administrative leadership.

Suggested outreach and educational activities included the publication of articles in key journals with the goal of increasing the visibility of AHM in professional societies as well as meetings and workshops focusing on teaching hospitalists and academic leaders methods to overcome challenges. Professional societies with a stake in AHM should better understand the challenges and position themselves to address these issues. The AHM task forces of SHM and SGIM can help give academic hospitalists a voice in having their needs addressed.

Defining a Sustainable Job Description for Academic Hospitalists

The group strongly endorsed the need for transparent and readily available data aimed at developing sustainable academic hospitalist positions. For example, required information would include how academic jobs are constructed (in terms of months on service per year and the number of nights or weekends of coverage) and what successful programs and their hospitalists have found to be acceptable. Over the longer term, empiric comparisons based on key metrics are needed to not only help guide career development and retention but also facilitate negotiations for programmatic support.

The group pointed out that embedded in delineating an optimal academic hospitalist job description is the longstanding work of general medicine societies in supporting and fostering the development of clinician educators. In many ways, the pressures of academic physicians to be mostly clinician and less educator versus someone who focuses heavily on educational work is similar for hospitalists and outpatient generalists. Academic general internal medicine divisions hired many general internists in the early 1990s to expand the reach of academic medical centers and increase the outpatient base.6 Many university hospitals are now hiring hospitalists to provide the inpatient care for these patients, but residency work hour reductions have added a layer of complexity, creating the need for entirely new roles for academic generalists (such as surgical comanagement of medically complex patients).7, 8 Past experiences in refining and reinforcing education as a key function (http://sgim.org/clinicianteach.cfm) would provide a template for future activities. Again, in the context of a rapidly growing number of nonteaching services, these descriptions have great importance.

Development of a Quality‐Improvement Portfolio Akin to an Educator's Portfolio

Many hospitalists actively participate in administrative work related to quality improvement activities, and we should develop this additional pathway for promotable academic activities (eg, clinician administrator); however, such a pathway may not be recognized by all promotion committees. The group observed that many aspects of quality improvement are similar to those of education (eg, developing a curriculum, leading a team, evaluating a process, defining generalizability, and disseminating locally proven interventions) and as such would be amenable to the development of a quality improvement portfolio, which candidates could submit to promotion committees. Again, past work in developing the importance and value of the educator's portfolio would facilitate the development of a quality improvement portfolio, which would require endorsement from key stakeholders (eg, the Association of Professors of Medicine, SGIM, and SHM).9 Importantly, this work may also benefit many outpatient‐based generalists who are increasingly focusing their careers on quality and safety improvement.

Developing Mentoring and Training Opportunities for Newly Hired and Junior Hospitalists

We reached a strong consensus about the need to develop a retreat‐format training opportunity by which junior academic hospitalists would be able to gain training in tasks critical to early‐career success. These were envisioned as an initial 2‐ to 3‐day meeting followed by mentorship at a distance and continued collaboration within the class of attendees. Topics would include key functions in AHM, such as becoming an effective attending physician and teacher, leadership, quality improvement, the business of medicine, effective billing, and maintaining a curriculum vitae. A number of professional societies have developed leadership or mentoring retreats, and at the time of this article's preparation, both regional and national efforts were underway to develop these products.

Developing Training and Mentorship Pathways for Hospitalist Researchers

There are few funded hospitalist researchers in the midcareer phase and a small but growing number of academic hospitalists entering the field with a focus on research. Enhancing a pipeline of researchers is a critical need for the field, as cementing AHM as an equal member of the academic medical community will be predicated on the successful development of hospitalist investigators. To this end, academic hospitalist groups should be encouraged to partner with other established research units (particularly general internal medicine) to create mentoring relationships and increase collaborative activities. The emergence of the Clinical and Translational Science Awards consortium sites, with a focus on implementation and effectiveness research, may also provide local opportunities for hospitalists to partner in research important for early‐career grant submission. Furthermore, building the pipeline of academic hospitalist researchers will require a strong focus on identifying students and residents through outreach at individual sites as well as presentations at national meetings (eg, the American College of Physicians).

Two other issues were also thought to be important. First, professional societies should work to encourage funders of primary carefocused general medicine training programs (the National Research Service Awards and the Health Resources and Services Administration) to allow hospitalists to qualify for such critical research training. Second, continuing to advocate for increasing funding for implementation and effectiveness research, via either the Agency for Healthcare Research and Quality or individual agencies of the National Institutes of Health, will be key; the emergence of a medical effectiveness institute would also be a potential boon.

Improving Relationships Between the Professional Homes of Academic Generalists

Relationships between outpatient‐based general medicine and hospital medicine were rocky as the field of AHM first took shape, and some residua of initial tensions persist a decade later. These tensions persist in part because hospitalists remain underdeveloped members of the academic community, and this perhaps gives some license to aver that hospitalists are merely transient faculty in a stage between residency and fellowship hired to improve throughput.

Overcoming this perception will require more engagement between academic generalists of all types, not less. The consensus group felt strongly that there need not be a single professional home for academic hospitalists and that generalists should be willing and even encouraged to self‐identify as hospital‐ or clinic‐focused, much as they might be geriatrics‐focused, informatics‐focused, or women's healthfocused. In fact, in some academic centers, a few generalists have successfully integrated themselves into both clinic‐based and hospitalist roles. In this way, the emergence and growth of AHM should be viewed as a boon to the practice of general medicine, not a challenge.

Resources

Much of what is proposed to enhance AHM will require resources. Academic hospitals have a vested interest in supporting AHM as a way to reduce turnover in a group that is increasingly critical for hospital operations, not to mention key leadership roles. Negotiating for these resources should emphasize that hospitals benefit directly from the revenue and margin that comes from incremental hospital admissions, collect most of the federal graduate medical education dollars, and benefit from improved care processes that are a result of hospitalist quality improvement efforts.

Deans and Departments, a key audience for the conference findings, also have a clear stake in fostering a less transient, more professionally satisfied and academically successful work force, particularly when hospitalists are increasingly the key educators of medical residents. Moreover, schools have a vested interest in the academic accomplishments and national reputation of their hospitalists. The financial arrangements will be unique to each setting and institution, and it is clear that the sources to be tapped will vary from site to site, but these resources are clearly necessary for the field.