The meteoric rise in the stature of academic surgery departments (ASDs) in the last 50 years was fueled by changes in reimbursement, changes in the science of surgery, and the refinement of surgical techniques.1–3 ASDs were able to excel in the three mission-critical areas of clinical care, research, and teaching.
However, reform in health care financing resulted in reductions in reimbursement for procedure- based specialties.1–3 ASDs found their revenues declining as expenses remained stable or increased.4–11 By necessity, these economic pressures resulted in the surgeon scientist being redefined as the clinician surgeon to meet the ever-increasing economic demands of modern health care. Today, many ASDs continue to find themselves not only struggling to survive financially, but also experiencing decreases in their academic productivity. A new model is needed for the design and development of an ASD in today’s health care environment that allows success in all areas of its tripartite mission.1–11
With the above issues in mind, I conducted a project to collect data from key stakeholders in ASDs at 12 institutions in an attempt to gather information on
- ▪ the current issues of significance to an ASD,
- ▪ the alignment among the stakeholders involved in managing and operating these ASDs, and
- ▪ the effect of alignment on the mission critical outcomes of these ASDs.
The project had two purposes:
- ▪ Purpose 1: To better understand the current issues important to ASDs and to query participants about possible solutions to these issues.
- ▪ Purpose 2: To test the following hypothesis: that alignment among stakeholders involved in the management and operation of today’s ASD is directly correlated with an improvement in mission-critical outcomes.
It is my hope that the data collected during this two-year project will help others in establishing guidelines for the design and development of successful ASDs in today’s health care environment.
In 2003, I conducted a pilot survey consisting of open-ended questions using the Delphi Technique12 and the process of dialectic inquiry.13,14 Members of four stakeholder groups from each of 12 randomly chosen ASDs were selected as participants: a medical school dean, a hospital/health system practice plan chief executive officers (CEO) or chief financial officers (CFO), an academic surgery department chair, and a surgery faculty member were identified at each site. These stakeholders were queried to obtain information used to design the questions contained in the survey instrument. The final survey was formulated using stakeholders’ responses that were deemed helpful in understanding the current issues important to ASDs and suggested questions for survey participants to elicit possible solutions to these issues (i.e., the goal of Purpose 1). After compilation of an initial list of 22 questions, the above-mentioned stakeholders were surveyed again using all 22 questions. This enabled me to reduce the number of survey questions and to customize the questions for each category of stakeholder.
Site selection and survey administration.
Taking geographic diversity into account, sites for administration of the final survey were selected from 12 well-known academic health centers with ASDs.* A total of 48 stakeholders at these schools received a cover letter and survey by mail at least one month before the scheduled interview. I administered all surveys by interviewing the participants, either by telephone or in person. All questions were asked in an open-ended fashion so as not to introduce interviewer bias into the responses.
For Purpose 1, the data from respondents were grouped and analyzed with either a Student t test or chi-square test, as appropriate. For Purpose 2, data were analyzed using nonparametric statistical tests (chi-square) where applicable to test the hypothesis. Correlation was calculated using Spearman’s correlation coefficient.
A total of 42 individuals from ASDs in all 12 institutions were interviewed: 11 medical school deans, 9 hospital CEO/CFOs, 12 department chairs, and 10 department faculty members (an overall response rate of 88%). Complete information was obtained from all four stakeholder groups at ASDs in nine of those institutions. Incomplete information was obtained at some institutions for one or more of the following reasons:
- ▪ Not all stakeholders agreed to participate at all institutions.
- ▪ A stakeholder held more than one role at an institution.
- ▪ Despite multiple attempts, and encouragement by the interviewer, some stakeholders declined to answer every question.
Some stakeholder groups declined to respond to particular issues by stating that they either felt inadequately informed to respond to an issue or that the issue was not within their purview.
In situations where multiple answers were offered to a question, the interviewer deferred to the answer that seemed to be most important to the stakeholder; that choice was included in the statistical analysis.
Regarding Purpose 1
As stated earlier, Purpose 1 was to better understand the current issues important to ASDs and to query participants about possible solutions to these issues. Based on data collected during the pilot focus groups as well as from the final survey, it became apparent that 12 issues (identified by italics below and listed in Table 1) were important to the stakeholders of ASDs.
The first issue identified was the design of an ASD. Two designs were proposed by stakeholders: a traditional, centrally controlled department versus a disease-specific model that crossed specialty lines. Medical school deans favored the traditional, centrally controlled department (9 of 11; 82%). CEO/CFOs and department chairs had no favored departmental structure, whereas department faculty members favored a disease-specific model (7 of 10; 70%).
The stakeholder groups differed in their preference for the second issue, an appropriate clinical performance metric. Medical school deans (8 of 10; 80%), CEO/CFOs (7 of 9; 78%), and department faculty members (10; 100%) strongly favored the work relative value unit (RVU), whereas department chairs favored both the work RVU (6 of 11; 55%) and collections (5 of 11; 45%) (P = .167 department chair versus CEO or medical school dean; P = .035 department chair versus department faculty member).
The third issue, the appropriate research performance metric, revealed more differences between stakeholder groups. Medical school deans and department chairs (10 of 10 and 11 of 11; 100%) favored extramural funding, whereas department faculty members, though they felt that extramural funding was important (6 of 10; 60%), also clearly suggested that other measures such as publications or presentations (4 of 10; 40%) should also be considered as appropriate research performance metrics.
Responses for an appropriate salary benchmark for surgery department faculty, the fourth issue, demonstrated that all stakeholder groups recognized a need to use multiple benchmarks. Suggested benchmarks were standards proposed by the Medical Group Management Association (MGMA) or by the Association of American Medical Colleges (AAMC), and salaries in local markets. A total of 7 of the 10 department faculty members (70%) favored the AAMC standards as the benchmark of choice.
The relative position of the department chair on the organizational chart was identified as important because all stakeholder groups felt that today’s department chair should be accountable to multiple individuals within the health care organization. Stakeholders noted that department chairs should have accountability in academics to the medical school dean, finances to the CFO practice plan, and operations, to the CEO of the hospital.
The sixth issue, promotion and tenure, was identified by all stakeholder groups as important. However, no readily identifiable solutions or responses were reported by the stakeholder groups. The most constant response from all groups was that tenure, in its current form, needs to change.
The appropriate teaching performance metric, issue seven, was identified as a very important issue, but again, there was no prevailing solution offered. Statistical analysis for this issue was hampered by the large number of I don’t know responses when stakeholders were queried for a possible teaching performance metric.
The need for and use of a dean’s tax was identified as necessary by all stakeholder groups, who suggested its use to be either for academic needs or for program development.
Stakeholders demonstrated differences on issue 9, the importance of medical student education to an ASD. Whereas most medical school deans (9 of 10; 90%) and department chairs (8 of 11; 73%) felt it was a “critical” or “very important” issue, the majority (6 of 10; 60%) of the department faculty members felt it was only “moderately important” or “unimportant.”
Issue 10, the design of an individual department faculty member job, was identified as an important issue in today’s ASD. Solutions and responses favored either a well-defined job or a goal-oriented job, where department faculty members are given goals to accomplish for which they define their own time and effort. Stakeholders were relatively evenly divided between these choices. However, a caveat that was added by most stakeholders suggested that a well-defined job structure for a young faculty member could evolve to a more goal-oriented structure as the individual matured.
Lack of money and other resources was cited by all stakeholders as the most important impediment to the development of an ideal ASD, issue 11. Though stakeholders listed a variety of impediments, the most consistently reported and seemingly important resource identified was money. Stakeholders clearly articulated that with adequate funding, development and success of an ASD would be assured.
Finally, the concepts of mission, vision, and values were identified as “critical” or “very important” by most CEO/CFOs (7 of 9; 78%) and medical school deans (6 of 10; 60%) and reported as “moderately important” or “unimportant” by department chairs (4 of 12; 33%, and 2 of 11; 54%, respectively) and department faculty members (8 of 10; 80%).
Regarding Purpose 2
As stated earlier, Purpose 2 was to test the hypothesis that alignment between the stakeholders involved in the management and operation of today’s ASDs is directly correlated with improvement in mission critical outcomes. The alignment analysis was incomplete because objective outcome data are sparse and a consensus on appropriate outcomes to measure when reporting performance of an ASD has not been achieved. National Institutes of Health (NIH) funding was the only objective measure uniformly reported, and it was used as representative of an ASD’s research performance metric.
Total organizational alignment was calculated from the responses noted in the survey regarding the issues important to ASDs. First an average alignment value was calculated per organization per issue. If all stakeholders at an ASD posited the same solution to an issue, the alignment value was 100, representing 100% agreement, whereas if all solutions were different the alignment value was 33 or 25 (dependent on the total number, either 3 or 4, of stakeholders who responded to an issue). Finally, the average alignment value for all issues was calculated to give the average total organizational alignment. Average total organizational alignment varied from a low score of 60 to a high of 92. There was no correlation noted between average organizational alignment and NIH funding.
When the intraorganizational alignment on each individual issue was analyzed, a significant (P = .01) negative correlation (−0.798) was found between alignment on promotion and tenure and NIH funding.
In this study, I attempted to define current issues of importance to stakeholders in ASDs and was able to identify 12 issues with potential solutions and responses. While I was conducting the interviews, it became evident that the issues identified by stakeholders were similar across ASDs. Though most survey questions were asked in an open-ended manner, it is interesting to note that the responses fell into very few, finite categories, thus suggesting that despite geographic and marketplace separation, ASDs face similar problems. The paucity and similarity of issues coupled with the uniformity of posited solutions and responses suggest a limited outlook for success. Overall, this constrained view is not surprising, given the fairly similar education, training, and work environments of academic surgeons.1–11 The uniformity of issues signals a need to change the paradigm and think “outside the box.”
Regarding Purpose 1
The first issue identified was the design of an ideal ASD. Stakeholders’ responses were limited, revealing a preference for a traditional, vertically oriented, centrally controlled department or for a patient-centered department organized along disease lines. Though significance was not attained, the P value (P = .066) suggested a trend toward significant differences between the groups. The groups most different in a 2 × 2 comparison were medical school dean versus department chair and department chair versus department faculty members. Department faculty members favored a patient-centered department more than department chairs did, who, in turn, favored a patient-centered department more than medical school deans did. These findings suggest that those most closely involved in clinical care in ASDs understand the need for patient-centered care. A new department structure to meet both the needs of the academic health center and the patient should be developed.
An appropriate clinical performance metric was identified as an important and controversial issue by nearly all stakeholders. Stakeholders identified three potential choices, but the work RVU was the most consistent choice for all stakeholder groups. The P value of .096 approached statistical significance, suggesting a trend toward differences between stakeholder groups, and in the 2 × 2 direct comparison, the largest difference was between department chairs and department faculty members. This discrepancy reflects the data that demonstrated that 45% of department chairs (5 of 11) reported that the appropriate clinical metric was collections or billings, whereas 100% of department faculty members favored the work RVU. The economic pressures faced by department chairs to meet or exceed budget may account for this discrepancy. Department faculty members favored the work RVU because it was the metric over which they had the greatest control.
Stakeholders considered the appropriate research performance metric for an ASD to be either extramural funding (usually identified as the “gold standard” NIH funding) or presentations/publications. A p value of .008 suggested that the stakeholder groups’ views were different. The data reveal that although all three stakeholder groups who responded to this issue mentioned both of the possible choices as important, medical school deans (100%) and department chairs (100%) indicated that extramural funding was the most important metric, whereas 40% of department faculty members (4 of 10) favored publications/presentations as the metric of choice. Findings for this particular issue revealed the need for stakeholder groups to openly discuss and identify a metric or metrics that will be used to measure research performance and establish the relative value of the metrics.
The appropriate salary benchmark to use for reimbursement for an ASD was identified as an important issue. Possible benchmark choices were (1) standards from a recognized nonacademic source, the MGMA, (2) the recognized academic standard recommended by the AAMC, or (3) salary levels found in the local market The P value of .02 suggested that the stakeholder groups were different in their views, and in the 2 × 2 direct comparison it was noted that the CEO/ CFOs favored the nonacademic source (7 of 9; 78% included either the MGMA and/or the local market) compared with either the medical school dean or department faculty members. Department faculty members favored the traditional academic salary benchmark AAMC (100%). Most stakeholders recognized that salary expectations needed to be based on multiple benchmarks. These expectations should be considered when devising an ASD compensation schema.
As academic medical centers have responded to the financial pressures of today’s health care environment, the appropriate position of the department chair within the organizational chart was identified as an important issue facing ASDs today. Traditionally, department chairs were noted to be fiercely independent; they reported only to the medical school dean on matters pertaining to academics. The stakeholders interviewed reported that today’s department chair should be accountable in multiple areas: to the medical school dean regarding academic issues; to the CEO/CFO of the practice, who is responsible for the financial health of the organization; and to the CEO/CFO of the hospital regarding operational matters. This finding has important implications for the once relatively autonomous department chair.
Promotion and tenure has become a contentious topic in most ASDs, and all stakeholders identified it as such in their responses. The data reveal that stakeholders reached nearly uniform consensus that the current system should be discontinued. Only 1 of the 31 who responded (3%) answered that the current system is adequate. This uniquely academic concept needs to be carefully reconsidered in the design and development of an ASD.
All stakeholders identified an appropriate teaching metric to be an area of concern and uncertainty. During the survey construction and interview process, all stakeholders interviewed uniformly acknowledged the need and desire for an appropriate teaching metric. However, they also stated that an objective, appropriate performance metric did not exist. Many potential choices were identified, but all were noted to be inadequate. Perhaps the department faculty members most clearly articulated the issue with the response, “I don’t know, but there should be one.” Despite the consensus for its importance, no statistical analysis was performed due to the overall uncertainty in choosing an appropriate teaching metric. All ASDs must consider the importance of teaching and establish appropriate metrics.
A dean’s tax was noted as an issue facing today’s ASDs, yet stakeholders in all categories recognized it as an appropriate part of an ASD. Stakeholders recognized the need for a contribution of resources for the common good, but had strong opinions concerning the use of a dean’s tax that included program development and academic needs (teaching and research support).
Stakeholder groups noted medical student education as a mission-critical area for any academic health center and deemed it necessary for the continued vitality and viability of an ASD. However, the groups differed in their ideas concerning its relative importance. Though there was no statistical difference between the responses of the stakeholder groups, a disturbing trend was apparent in the responses. Ninety percent of medical school deans (9 of 10) indicated that the issue was either “critical” or “very important” as did 73% of department chairs (8 of 11), whereas only 40% of department faculty members (4 of 11) regarded it as “very important.” Sixty percent of department faculty members (6 of 10) responded that medical student education was “moderately important” or “unimportant,” although they have the greatest level of direct interaction with medical students.
Most department faculty members interviewed reported that the recent focus on revenue generating activities was the primary cause for their lack of interest in such non-revenue-generating activities as medical student education. ASDs will need to address and correct this disturbing finding.
Stakeholders identified two possible methods to design the job of the individual department faculty member in an ASD; either a well-defined job that exactly outlines the percentage of effort for each faculty member, or a goal-oriented job where department faculty members are given goals to accomplish for which they define their own time and effort. There was no significant difference between stakeholder groups on this issue. However, nearly all stakeholders added a caveat that young faculty would benefit from a three- to five-year period of a well-defined job, and this recommendation should be considered when young department faculty members are hired.
Another issue identified concerned impediments for the development of an ideal ASD. The possible impediments were narrowed to four by the stakeholders (egos, leadership, culture, and money), but the most consistent response indicated was money. There were no differences noted in this category, suggesting that all stakeholders held the same view, perhaps illusory, that money and other resources would ensure the success of an ASD.
The concepts of mission, vision, and values, though part of the culture of nearly all business organizations, were not found to be as universally accepted in ASDs. Though the results did not reach statistical significance, there was a clear trend among the stakeholder groups. Sixty percent of medical school deans and 78% of CEO/CFOs felt mission, vision, and values were “critical” or “very important” to the organization. Only 46% of department chairs (5 of 11) identified these concepts as “very important”; none chose “critical.” Surprisingly, 80% of department faculty members (8 of 10) felt that mission, vision, and values were “moderately important” or “unimportant” to the functioning or success of an ASD. During the interview process, most department faculty members and many department chairs stated they did not know the mission statement of their organization. Most stakeholders who felt that mission, vision, and values were “critical’ or “very important” added that either (1) they did not know how to disseminate the concepts down through the organization, or (2) mission, vision, and values were only important to leadership.
Regarding Purpose 2
Purpose 2 sought to determine the relationship between organizational alignment and performance. The data initially revealed that average intraorganizational alignment varied from a high of 92% to a low of 60% among the nine organizations whose stakeholders’ responses were complete enough to be analyzed. The relationship between alignment and performance was difficult to examine because of the paucity of nationally recognized, accepted, and reported outcome measures available to assess an ASD’s success. After intense investigation, the only outcome measure felt to be of value was that of total NIH funding dollars per ASD. Though this metric is certainly important to stakeholders in ASDs, it measures only one mission-critical area. Objective, acceptable metrics representative of the other two mission-critical areas, clinical care and teaching, are needed not only to more adequately examine the effect of alignment on performance, but, more important, to allow ASDs to accurately gauge their performance in all mission-critical areas.
The alignment analysis found no correlation between the average overall organizational alignment and NIH funding. A single, significant linear relationship between intraorganizational, single issue alignment and NIH funding was identified. A significant linear relationship (P = .01) was found between intraorganizational alignment on the issue of promotion and tenure and NIH funding. However, this was a negative correlation (correlation coefficient = −0.798), suggesting that the more aligned an organization’s stakeholders were in the area of promotion and tenure, the less NIH funding it would secure. Given that promotion and tenure represents a traditional academic concept and NIH funding remains the “gold standard” of success in academic research, the finding seems counterintuitive. However, a look at survey results pertaining to promotion and tenure reveals that most stakeholders (98%) felt that promotion and tenure needed to change. Forty percent (4 of 10) of department faculty members (those most responsible for securing NIH funding) responded that promotion and tenure should be eliminated. The survey results suggested a very dissatisfied view of promotion and tenure, and in this climate, the negative correlation between dissatisfaction with the gold standard of academic success (promotion and tenure) and the traditional measure of research success (NIH funding) is understandable. ASDs will need to consider ways to reverse the negative correlation between promotion and tenure and NIH funding.
A limited number of issues felt to be important to the stakeholders in today’s ASD were revealed. The issues represented the broad concepts of organizational structure, individual job structure, performance metrics, salary, and organizational behavior. When viewed in the aggregate, it seemed that many of the issues facing ASDs involved discrepancies between expectations of leaders and followers. Also, there seemed to be confusion among all groups as to their respective roles, responsibilities, and performance metrics. Finally, there was the disturbing finding from the department faculty members concerning the relative unimportance of medical student education.
Though ASDs have missions unique to the profession (specifically, clinical care, research, and teaching), many issues identified in this project are the same as those that nonmedical organizations struggle to properly balance. The principles of management and strategic planning found in nonmedical organizations are valuable tools that could be applied to help complex medical organizations function more efficiently and effectively. ASDs must invest the time and resources necessary for formal strategic planning. Understanding an ASD’s position within local, regional, national, and international markets will allow a more accurate plan for future growth and development. ASDs must also identify their strengths and weaknesses to more efficiently allocate resources; targeting areas and services for growth as well as areas that should have support gradually withdrawn. It is recognized that clinical excellence and the achievement of the clinical endeavor will be the engine that drives success in all the mission-critical areas of an ASD. By maintaining a clear focus on strategic planning and examining all decisions in light of the strategic plan, today’s ASD can become an organization that is run by the process of strategic management—a management style that is likely to be successful.
Recommendations for the future
During the conduct of this project, it became apparent to me that there were certain actions that ASDs could take to improve their performance. The following recommendations are based on survey responses, intuition, and subjective information that I collected from a variety of ASD stakeholders during the 2003–2005 study period. Although not tested or proven to be true, nor likely to apply in their entirety to any one ASD, these recommendations may be helpful to individual organizations. I present these recommendations as a vision for the future; their validity remains, given the measured view of change found within many ASDs.
Clearly delineate and disseminate the mission, vision, and values of an ASD. A series of focus meetings using input from all ASD members could be held to articulate a mission statement that would encompass the department’s tripartite mission as well as define the department’s vision and values. The mission statement created by the ASD would require alignment with the core mission, vision, and values of its medical school, the academic health center, and the affiliated university. A key job of leadership would be to ensure that the ASD’s mission statement is disseminated and adopted throughout every level of the organization. All ASD decisions would be mission based. The simplified decision making resulting from clearly delineated mission, vision, and values will reverse the current trend of “margin drives mission” to “mission drives margin.”
Reestablish the culture of the ASD. Historically, the culture of ASDs was one of camaraderie that contributed to organizational success by motivating recruitment and retention that was nonmonetary in origin, resulting in a sense of duty to the organization, professionalism, and the recognition of a clear chain of command. Now, as ASDs enter the 21st century, their culture is a mix of the traditional mindset (surgeon scientist) along with the more egocentric, employee (clinician surgeon) mentality found in younger generations of surgeons. The result is the lack of a clear, unified organizational culture that is so widespread today. Development of a unified culture, a compelling vision, a clearly articulated mission, and accepted values will be essential to organizational success. ASDs are composed of highly motivated, well-educated, fiercely independent members who generally resist an overtly controlling leadership and management style. A unified culture will allow the diverse ASD members to understand “how things are done around here” and establish a sense of camaraderie. A compelling vision, in combination with a unified culture, will serve to not only get and keep “the right people on the bus,” but will also serve as an additional motivator to attract the best and brightest and retain them in the organization. A clearly articulated mission will guide ASD members in their daily decisions while accepted values serve as the guides for acceptable behaviors. In this way, an ASD will be able to function without stifling oversight, thereby allowing maximal organizational creativity and success.
Develop an organizational structure that supports the needs of a department chair who is responsible to multiple individuals in multiple areas by providing a structure that includes individuals to champion each of the mission-critical areas of clinical care, teaching, and research. Figure 1 illustrates one version of such an organizational structure.
In this structure, the vice chairs have responsibility for the three mission-critical areas of the ASD, and they report to the department chair. The department chair remains accountable to the dean as is traditional, but is also accountable to the CEO/CFO of the hospital/health system and/or the hospital practice plan as a member of a council of chairs composed of all clinical department chairs.
The findings of this study are, of course, contingent on the study design and methods. The following potential problems could affect the reported data and statistical analyses:
- ▪ Selection bias. Study sites represented ASDs across a wide geographic range, but I chose the sites randomly. The study was also dependent on the voluntary participation of individuals at each organization. These factors could introduce selection bias into the study and skew the results.
- ▪ Interviewer bias. As the study’s sole investigator, I was the only interviewer, thereby effectively eliminating inter-interviewer variability in data collection. A design of this type could introduce interviewer bias and influence the data and findings, despite attempts to the contrary.
- ▪ Small numbers. There were 12 sites selected for study and four stakeholders contacted for interviews at each site. Because of various factors, a total of 42 individuals were interviewed, and complete data were obtained for nine organizations. These small numbers could introduce statistical errors into the analyses.
- ▪ Lack of objective performance measures. In attempting to analyze alignment and its effect on performance, objective performance measures were required. Only one performance measure was readily available for all organizations—NIH funding. Lack of a definitive set of performance measures limited the applicability of the alignment analysis.
Summary and Conclusions
This project has outlined many issues facing ASDs today. It has also elicited various current solutions proposed by stakeholders in a number of ASDs. Interestingly, given the similarity in education, training, and work environment of most ASDs, these current solutions are well known and similar across the country. Effective solutions, however, remain elusive for most ASDs.
The findings of this project suggest that to succeed, an ASD will need to reestablish its culture and clearly articulate and disseminate its mission, vision, and values. It has been said that trying to lead a group of surgeons is like trying to herd cats. There is an inherent difficulty with the traditional, vertically oriented organization’s ability to motivate a group of highly educated, highly motivated, independent, and intelligent individuals in any direction in which they are not convinced they want to go.
The future success of the ASD will depend on a strong leader. Such a leader must not only be able to inspire a return to the valuable assets of the past, but must also be able to decentralize control, allowing students, residents, and faculty to be innovative; as well as stimulate original thought while making mission-based decisions. The decentralized control must be of the sort that sets mission-based goals within some value-driven guidelines, thus allowing department faculty members to achieve these goals unimpeded but with oversight. This type of organization will be most able to succeed if there is a coherent culture combined with fully adopted mission, vision, and values in the face of an academic health care environment that continues to change.
The author wishes to thank Deborah E. Powell, MD, dean, University of Minnesota Medical School, for her support, and Connie Lindberg, Department of Veterans Affairs Medical Center, for her help with survey implementation and editorial assistance.