Trauma Leadership: Does Perception Drive Reality?

ORIGINAL REPORTS

Trauma Leadership: Does Perception Drive Reality? Joseph V. Sakran, MD,* Bo Finneman, BS,† Chris Maxwell, PhD,† Seema S. Sonnad, PhD,* Babak Sarani, MD,* Jose Pascual, PhD,* Patrick Kim, MD,* C. William Schwab, MD,* and Carrie Sims, MD* *Trauma Center at Penn and †Wharton Undergraduate Leadership Program, University of Pennsylvania, Philadelphia, Pennsylvania INTRODUCTION: Leadership plays a key role in trauma team

management and might affect the efficiency of patient care. Our hypothesis was that a positive relationship exists between the trauma team members’ perception of leadership and the efficiency of the injured patient’s initial evaluation. METHODS: We conducted a prospective observational study

evaluating trauma attending leadership (TAL) over 5 months at a level 1 trauma center. After the completion of patient care, trauma team members evaluated the TAL’s ability using a modified Campbell Leadership Descriptor Survey tool. Scores ranged from 18 (ineffective leader) to 72 (perfect score). Clinical efficiency was measured prospectively by recording the time needed to complete an advanced trauma life support (ATLS)directed resuscitation. Assessment times across Leadership score groups were compared using Kruskal-Wallis and Mann-Whitney tests (p ⬍ 0.05, statistically significant). RESULTS: Seven attending physicians were included with a postfellowship experience ranging from ⱕ1 to 11 years. The average leadership score was 59.8 (range, 27–72). Leadership scores were divided into 3 groups post facto: low (18 – 45), medium (46 – 67), and high (68 –72). The teams directed by surgeons with low scores took significantly longer than teams directed by surgeons with high scores to complete the secondary survey (14 ⫾ 4 minutes in contrast to 11 ⫾ 2 minutes, p ⬍ 0.009) and to transport the patient for CT evaluation (19 ⫾ 5 minutes in contrast to 14 ⫾ 4 minutes; p ⬍ 0.001). Attending surgeon experience also affected clinical efficiency with teams directed by less experienced surgeons taking significantly longer to complete the primary survey (p ⬍ 0.05). CONCLUSION: The trauma team’s perception of leadership

is associated positively with clinical efficiency. As such, more formal leadership training could potentially improve patient

Correspondence: Inquiries to Joseph V. Sakran, MD, Division of Traumatology, Surgical Critical Care, and Emergency Surgery, The Trauma Center at Penn, 3400 Spruce Street, 5 Maloney, Philadelphia, PA 19104; fax: (215) 349-5917; e-mail: Joseph. [email protected]

236

care and should be included in surgical education. (J Surg 69: 236-240. © 2012 Association of Program Directors in Surgery. Published by Elsevier Inc. All rights reserved.) KEY WORDS: trauma, leadership, efficiency patient care COMPETENCIES: Patient Care, Medical Knowledge, Professionalism, Interpersonal and Communication Skills, Practice Based Learning and Improvement, Systems Based Practice

INTRODUCTION Trauma centers were introduced first in the late 1960s and early 1970s to provide better care to patients with serious and potentially life-threatening injuries. From these initial models, trauma care in the United States has evolved substantially with regionalized, designated trauma centers providing care to approximately 700,000 injured patients annually. As time passed, these centers and teams showed improved outcomes and notably decreased mortality and less morbidity from organ failure. Subsequently, several studies showed even better outcomes as patient volume increased, and ultimately, the concept of regionalization of severe injuries to designated trauma centers was born.1,2 Nathens et al.1 demonstrated recently an improvement in mortality when volume at trauma centers exceeds 650 cases. Evidence suggests that outcomes in critically injured patients correlate directly with the time between injury and definitive care. Reduction in time clearly impacts patient morbidity and mortality and, therefore, stresses the need for clinical efficiency at regional trauma centers.3,4 Frequently, trauma care is rendered by a multiperson interdisciplinary team of individuals that work collectively in a coordinated fashion to perform the simultaneous resuscitation, diagnosis, and management of life-threatening injuries according to Advanced Trauma Life Support (ATLS) Guidelines. Trauma teams can be small, working sequentially through the established protocols of “vertical” resuscitation, or they can be comprised of a larger number of providers where many tasks are simultaneously performed and completed in a “horizontal”

Journal of Surgical Education • © 2012 Association of Program Directors in Surgery Published by Elsevier Inc. All rights reserved.

1931-7204/$30.00 doi:10.1016/j.jsurg.2011.09.004

fashion. In either scenario, a team leader is vital to the flow of the resuscitation. The physician with the most experience in trauma centers, typically a surgeon, is the trauma attending leadership (TAL) and designated to oversee, direct, and coordinate all tasks. After analyzing video recordings of 425 trauma resuscitations, Hoff et al.5 concluded that presence of a command physician facilitated trauma care and provided essential personnel management. Other studies investigating trauma team structure have concluded also that the presence of organized leadership improves team performance.6-9 Ultimately, the goal of the trauma team leader is to assure efficiency, proficiency, and rapid delivery of definitive therapy. Thus, the team leader’s ability “to lead” the team becomes a key factor in the timely implementation of care.9-11 Despite the importance of leadership, however, limited data are available evaluating how the trauma team perceives leadership and whether or not these perceptions impact patient care. The purpose of our study was to evaluate the relationship between the perception of leadership ability and efficiency of trauma patient care.

METHODS The study design was reviewed and approved by the Institutional Review Board of the University of Pennsylvania. Data were obtained over a 5-month period (November 2009 to March 2010) from the Trauma Center at Penn, a state-accredited regional resource (level 1) trauma center. The evaluation of trauma patients at the University of Pennsylvania is based on the regulations set forth by the ATLS Guidelines. Trauma teams are composed of an interdisciplinary team consisting of an attending trauma surgeon, trauma fellow, trauma resident, emergency medicine resident, airway team, nursing staff, pharmacist, respiratory therapist, emergency medical technician, radiology technician, and pastoral care. Although not physically in the trauma bay, open lines of communication with the operating room and blood bank are also established for each trauma response. The resuscitations are run in a horizontal fashion, as described. The attending surgeon is in a predetermined command area, at the foot of the resuscitation stretcher, where he or she can have a complete view of the resuscitation while the other trauma team members perform preassigned resuscitation tasks. All trauma attendings at the University of Pennsylvania are fellowship trained in trauma and surgical critical care. Specific surgeon demographics were identified through the administrative trauma office and were blinded to ensure anonymity. Demographic variables included age, sex, and years in practice. As required by the Pennsylvania Trauma Systems Foundation (PTSF), all trauma cases were audited for the presence of PTSF-defined performance improvement occurrences as well as institution-specific occurrences. Positive audits were reviewed by the performance improvement coordinator and medical director, and the results were referred to the full trauma performance improvement committee for adjudication and assignment of responsibility to specific attending or the team as warranted. For the purposes of this study, data regarding each

attendings’ performance improvement occurrences over the last 5 years were collected. Using a convenience sample, a dedicated researcher independent of the trauma program observed trauma patient resuscitations. All resuscitations in this study were lead by the attending surgeon with a full complement of trauma team members. Only the resuscitations of hemodynamically stable patients who did not appear to require surgery or life-saving procedures were included in this study. Multiple-patient scenarios also were excluded. Efficiency was predefined as the cumulative time to (1) completion of the primary survey, (2) completion of the secondary survey, and (3) the time to definitive evaluation using computed tomography (CT) scan. Members of the trauma team were not aware that the researcher was timing the resuscitation events. At the completion of each resuscitation, individual team members (eg, medics, nurses, residents, and trauma fellows) were asked to evaluate the attending trauma surgeon anonymously using a modified Campbell Leadership Descriptor Survey. Team members were instructed to complete the survey based solely on the trauma resuscitation that just had taken place. Surveys were completed immediately postresuscitation and then collected by the researcher. The modified Campbell Leadership Descriptor is a validated survey tool that provides a quantitative measure of leadership ability.12 The survey consists of 9 sections describing important aspect of leadership, including: vision, management, empowerment, diplomacy, feedback, innovation, creativity, style, energy and overall leadership capacity. Respondents were asked to evaluate 18 leadership statements and rate their perception of the attending surgeon using a Likert scale ranging from 1 to 4 (Appendix A). A rating of 1 signified that the statement definitely did not describe the actions of the attending surgeon, whereas a rating of 4 meant that the statement definitely described the attending surgeon. Given that each statement was positive with regard to leadership ability, higher scores indicated perceptions of greater leadership ability. A perfect leader would receive a maximum score of 72, and a leader perceived as ineffective could receive a score as low as 18. Based on leadership scores attending surgeons were divided post facto into 3 leadership ability groups (low, moderate, and high). Clinical efficiency measured by the time to each intervention point was compared across leadership groups using the Kruskal-Wallis test followed by the Mann-Whitney test to determine significant pairwise differences. Statistical analyses were performed using SPSS in contrast to 17.0 (IBM Corporation, Somers, New York) and a p value ⬍ 0.05 was considered significant.

RESULTS Seven attending trauma surgeons were evaluated during the 5-month study period. All attendings were male with ages ranging from 38 to 56 years. Postfellowship experience ranged from 1 to 11 years (Table 1). Survey data included 81 leadership surveys from 22 separate patient encounters. The response rate was 100%. Leadership

Journal of Surgical Education • Volume 69/Number 2 • March/April 2012

237

TABLE 1. Demographics of Attending Surgeons Attending Surgeon

Sex

Age (years)

Postfellowship Experience (years)

Leadership Score (Average)

Performance Improvement (Occurrences over 5 years)

1 2 3 4 5 6 7

M M M M M M M

36 38 38 39 40 46 49

1 3 4 1 5 11 1

70 68 68 69 53 66 54

0 15 16 0 2 11 2

scores ranged from 27 to 72, with a mean of 59 (standard deviation SD 12). The 3 post facto attending surgeon groups were compared [low (18-45), medium (46-67), and high (6872)]. The time to primary survey completion did not differ significantly across leadership groups; however, significant differences were found in time to completion of the secondary survey and time to CT scan (Table 2). Specifically, teams directed by surgeons receiving low leadership scores took significantly longer to complete these tasks when compared with teams that were directed by surgeons perceived as having high leadership ability (Table 3). When individual surgeon experience was examined, time in practice was associated with clinical efficiency. Teams led by surgeons with less than 5 years in practice took significantly longer to complete the primary survey when compared with teams led by more experienced surgeons (p ⬍ 0.05). No difference based on experience was observed for the secondary survey or time to CT scan interventions. Leadership scores were not found to correlate with performance improvement occurrences (Table 1).

DISCUSSION Leadership plays a key role in the coordination and function of any team. Strong leadership skills may be important particularly when teams are performing complex activities in high-stress situations, as is the case with trauma care. Although several publications delineated the importance of leadership in crisis management, there are limited data on how leadership is perceived by interdisciplinary trauma team members and whether or not this perception impacts clinical care. In this study, we observed seven fellowship-trained trauma surgeons during 22 trauma resuscitations and demonstrated a positive relationship

between clinical efficiency and the perception of leadership. In particular, teams directed by surgeons perceived as having low leadership ability took significantly longer to complete the key steps in initial trauma patient evaluation. Although leadership ability did not seems to be related to surgeon experience, time in practice seemed to influence positively the efficiency of completing the primary survey. Although leadership is clearly important, the implementation of effective leadership in trauma may be more of a systems process rather than an individual person’s assignment. The importance of the trauma surgeon’s leadership role has not been clearly validated. In an extensive study investigating trauma leadership, Klein et al.8 interviewed over 30 members of the trauma center team, and observed over 150 hours of trauma resuscitations. Effective trauma leaders performed 4 key functions. They (1) provided strategic direction, (2) monitored the course of clinical care, (3) provided hands-on treatment, and (4) and taught other team members. Although having an identifiable leader contributes to the coordination, learning, and reliability of the team process, Klein et al.8 suggested that leadership in the trauma bay is a dynamic phenomenon that allows for the interchangeable assignment of “leader” from the attending surgeon, to the surgical fellow to the trauma resident.8 At our institute, trauma teams function in a highly organized manner with predetermined, defined roles for each member of the team. This mature type of trauma system may have had an impact on our findings. In a study by Driscoll et al. evaluating three different trauma centers, the total resuscitation time was reduced by 54% when there was precise task allocation, larger trauma teams, and the adoption of horizontal trauma resuscitation. Similarly, our center functions with a high level of organizational skill and a robust team infrastructure. This may have contributed to the short resuscitation times observed. Although

TABLE 2. Comparison of Leadership Ability Using the Resuscitation Endpoints of Time to Primary Survey, Secondary Survey, and CT Scan Time (min) to Primary Survey

Time (min) to Secondary Survey

Time (min) to CT Scan

Leadership Ability

Leadership Survey, N (%)

Mean (SD)

p Value

Mean (SD)

p Value

Mean (SD)

p Value

Low Moderate High

24 (30) 26 (32) 31 (38)

1.99 (0.51) 1.79 (0.62) 1.71 (0.70)

0.16

13.77 (4.1) 12.07 (3.49) 11.09 (2.15)

0.009

18.97 (4.67) 16.01 (4.85) 14.40 (3.73)

⬍0.001

Statistical analysis: Kruskall-Wallis test. 238

Journal of Surgical Education • Volume 69/Number 2 • March/April 2012

TABLE 3. Comparison of Leadership Abilities Using the Statistically Significant Resuscitation Endpoints of Time to Secondary Survey and Time to CT Scan Objective Endpoints Secondary survey CT scan

Leadership Ability

p Value

Low in contrast to moderate Low in contrast to high Moderate in contrast to high Low in contrast to moderate Low in contrast to high Moderate in contrast to high

0.07 0.002 0.30 0.01 ⬍0.001 0.14

Statistical analysis: Mann–Whitney test.

we did not observe any differences in time to complete the primary survey, a positive relationship between perception of leadership ability and efficiency of patient care after the primary survey emerged. The lack of difference in completion of the primary survey might be explained by 2 factors: (1) in general the primary survey is short in duration taking less than a minute to complete and (2) minimal input or direction is needed from the attending surgeon to complete this part of the resuscitation. These factors are not accurate when examining time to completion of secondary survey and time to CT scan. The duration of time to completion can be long, and TAL input is critical in the guidance and management of the trauma team during this part of the resuscitation. Our research suggests that perception of team leadership ability can impact patient care and, therefore, suggests that providing enhanced leadership skill training, as part of the surgical and trauma curriculum might be beneficial. Although formal leadership training has not been traditionally offered in most academic institutions, leadership development has been shown be helpful in other high stress arenas.13 Leadership and team building could be enhanced via debriefing sessions after resuscitations take place allowing team members to discuss the responsibilities they were given and focus on areas where processes were not used as efficiently as possible. Finally, a positive work environment has been associated with improved patient outcomes and organizational performance.14,15 We recognize the inherent limitations of this type of study. The small sample size, short time frame of the project, few attending surgeons, lack of gender differences, and the limitation to a single trauma center may introduce significant bias. Additionally, our study included only single, stable patients to compare time points directly. One may argue that “true leadership” and/or the perception of good leadership may have a bigger impact during multiple patient scenarios or during the resuscitation of physiologically unstable patients. Furthermore, even though the survey instrument has been validated for leadership in business, this tool might not be applicable in clinical situations. Finally, even though differences between low and high leadership groups were observed, they are small time intervals. Therefore, the clinical significance on outcome remains to be determined. Broadening this pilot study to include other trauma centers might provide a more representative distribution of leadership

ability. Inclusion of critically injured patients might be able to provide more comprehensive data as to whether leadership remains a significant factor in efficiency given these altered circumstances. In fact, some might argue that these types of situations when efficiency of clinical care matter the most. Our study evaluated leadership ability through the perception of those being led. However, there is no current standard on leadership evaluation, and ultimately it is the perception of those being led that matters the most when a team is confronted by a critical clinical dilemma. Despite its limitations, this study is unique in that it moves beyond the traditional standard of quantitative variables measured for trauma team leaders (eg, number of resuscitations, admissions, or patients managed with Injury Severity Score ⬎ 16). It attempts to look at the quality of leadership as determined by trauma team members, helping to elucidate the importance of leadership in resuscitation of the trauma patient. Current training programs seem to produce excellent clinicians and academicians. We hope that this pilot study will result in more robust studies. This would allow us to better answer whether enhancing the training required of surgeons should include leadership development, with the ultimate goal of optimizing trauma care.

REFERENCES 1. Nathens AB, Jurkovich GJ, Maier RV, et al. Relationship

between trauma center volume and outcomes. JAMA. 2001;285:1164-1171. 2. Bazzoli GJ, Madura KJ, Cooper GF, MacKenzie EJ,

Maier RV. Progress in the development of trauma systems in the United States: results of a national survey. JAMA. 1995;273:395-401. 3. Foster JT. Helicopters make sense in medical care. Mod

Hosp. 1969;112:78-82. 4. Sampalis JS, Denis R, Lavoie A, et al. Trauma care region-

alization: a process-outcome evaluation. J Trauma. 1999; 46:565-579. 5. Hoff WS, Reilly PM, Rotondo MF, DiGiacomo JC,

Schwab CW. The Importance of the Command-Physician in Trauma Resuscitation. J Trauma. 1997;43: 772-777. 6. Driscoll PA, Vincent CA. Organizing an efficient trauma

team. Injury. 1992;23:107-110. 7. Xiao Y, Seagull FJ, Mackenzie C, Klein K. Adaptive lead-

ership in trauma resuscitation teams: a grounded theory approach to video analysis. Cogn Tech Works. 2004;6:158164. 8. Klein K, Zegert J, Knight A, Xiao Y. Dynamic delegation:

shared, hierarchial, and deindividualized leadership in extreme action teams. Admin Sci Q. 2006;51:590-621.

Journal of Surgical Education • Volume 69/Number 2 • March/April 2012

239

9. Driscoll PA, Vincent CA. Variation in trauma resuscita-

tion and its effect on patient outcome. Injury. 1992;23: 111-115. 10. Hallam J, Stammers RB. Task allocation and the balanc-

ing of task demands in the multiman-machine system: some case studies. Appl Egron. 1985;16:251. 11. Lanzetta JT, Roby TB. Group performance as a function

of work distribution patterns and task load. Sociometry. 1956;19:95. 12. Campbell DC. Leadership Descriptor: Facilitator’s

Guide. San Francisco, Calif: Jossey-Bass/Pfeiffer; 2002. 13. Duda RB. Physician and scientist leadership in academic

medicine: strategic planning for a successful academic leadership career. Curr Surg. 2004;61:175-177. 14. Aiken LH, Clarke SP, Sloane DM, Lake ET, Cheney T.

Effects of hospital care environment on patient mortality and nurse outcomes. J Nurs Adm. 2008;38:223-229. 15. Aiken L, Clarke S, Sloane D, Lake E, Cheney T. Effects of

hospital care environment on patient mortality and nurse outcomes. J Nurs Adm. 2009;39:45-51.

Appendix A. Leadership Survey Please circle your role: Resident RN FellowMedical StudentTech The following survey is derived from the Campbell Leadership Descriptor* questions making statements relevant to the healthcare industry and trauma field. Using the following scale, please rate Dr ________________ in the following leadership areas. *The Campbell Leadership Descriptor is a proven method used to ascertain leadership quality. This survey has been assessed and approved by a professor in the Wharton Leadership Program as accurate and reliable for the means of this research proposal. 1 ⴝ Definitely Not Descriptive; 2 ⴝ Not Descriptive; 3 ⴝ Descriptive; 4 ⴝ Definitely Descriptive Area A: Vision

240

1 2 3 4 Sees the big picture in developing a vision for each patient’s plan of care. 1 2 3 4 Likes to take on new projects and programs Area B: Management 1 2 3 4 Effectively assigns responsibility and the necessary authority to others 1 2 3 4 Sets clear work priorities for others and themselves 1 2 3 4 Develops systems and procedures for efficiently organizing people andresources Area C: Empowerment 1 2 3 4 Helps others to achieve more than they thought they were capable of achieving 1 2 3 4 Helps others deal with difficult situations 1 2 3 4 Provides challenging assignments to others and coaches them Area D: Diplomacy 1 2 3 4 Knows a wide range of people to help get things done 1 2 3 4 Develops teamwork among individuals of differing backgrounds Area E: Feedback 1 2 3 4 Gives constructive feedback in a way that benefits individuals 1 2 3 4 Open and responsive when receiving ideas from others Area F: Innovative/Creative 1 2 3 4 Capable of thinking independently and coming up with novel ideas 1 2 3 4 Persists in the face of criticism or failure; hard to discourage Area G: Style 1 2 3 4 Sees many positive possibilities; upbeat 1 2 3 4 Believable, ethical, trustworthy Area H: Energy 1 2 3 4 Actively brings energy to a team 1 2 3 4 Adapts well to conflicting work and personal demands Area I: Leadership 1 2 3 4 Is a great leader Total Score:/72

Journal of Surgical Education • Volume 69/Number 2 • March/April 2012