PDFlib PLOP: PDF Linearization, Optimization, Protection Page inserted by evaluation version www.pdflib.com –
[email protected]
ACAD EMERG MED
d
December 2004, Vol. 11, No. 12
d
1341
www.aemj.org
Profiles in Patient Safety: Authority Gradients in Medical Error Karen S. Cosby, MD, Pat Croskerry, MD, PhD Abstract The term ‘‘authority gradient’’ was first defined in aviation when it was noted that pilots and copilots may not communicate effectively in stressful situations if there is a significant difference in their experience, perceived expertise, or authority. A number of unintentional aviation, aerospace, and industrial incidents have been attributed, in part, to authority gradients. The concept of authority gradient was introduced to medicine in the Institute of Medicine report To Err Is Human, yet little has been written or acknowledged in the medical literature regarding its role in medical error. The practice of medicine and medical
training programs are highly organized, hierarchical structures that depend on supervision by authority figures. The concept that authority gradients might contribute to medical error is largely unrecognized. This article presents one case and a series of examples to detail how authority gradients can contribute to medical error, and describes methods used in other disciplines to avoid their potentially negative impact. Key words: authority gradients; medical error; quality assurance. ACADEMIC EMERGENCY MEDICINE 2004; 11:1341–1345.
A resident encountered a pediatric patient with a varicellalike illness. However, the child appeared more ill than expected. On close inspection, several lesions had surrounding erythema and tenderness; one had a necrotic edge suggesting a potentially serious secondary deep soft-tissue infection. The resident had just finished reading a review of similar cases and was concerned that this child might have a life-threatening condition. He discussed the case with the attending physician in the emergency department. He was surprised to find that the attending physician considered the case a routine, uncomplicated presentation and instructed him to discharge the patient. The young physician reiterated his concern and requested that the supervising physician reassess the child. The attending physician once again insisted that the illness was minor and advised the resident to send the child home without further delay. The resident, convinced that his impression had not been taken seriously, chose to ‘‘draw some labs’’ and stall, hoping to have the opportunity to discuss the case with a different
attending physician at the change of shift. Eventually the resident had to leave for other duties and left the child in the care of another resident awaiting a second review. He returned the following day to learn that the child had been discharged and had died later that night at another hospital. The patient died of varicella complicated by a secondary streptococcal fasciitis, a relatively rare complication of a common, typically benign, childhood infection.1
From the Department of Emergency Medicine, The John H. Stroger, Jr., Hospital of Cook County/Rush Medical College (KSC), Chicago, IL; and the Department of Emergency Medicine, Dalhousie University (PC), Halifax, Nova Scotia, Canada. Received November 14, 2003; revisions received March 18, 2004, and May 12, 2004; accepted June 11, 2004. Series editors: Pat Croskerry, MD, PhD, Dartmouth General Hospital Site, Dalhousie University, Halifax, Nova Scotia, Canada; and Marc J. Shapiro, MD, Rhode Island Hospital, Brown University School of Medicine, Providence, RI. Address for correspondence and reprints: Karen S. Cosby, MD, Department of Emergency Medicine, The John H. Stroger, Jr., Hospital of Cook County/Rush Medical College, 1900 West Polk Street, Chicago, IL 60612. Fax: 312-864-9656; e-mail: kcosby@ ccbh.org. doi:10.1197/j.aem.2004.07.005
DISCUSSION Ideally, in complex organizations, information should flow seamlessly between team members. A variety of conflicts may impede the free exchange of information and the objective assessment of risk. Communication may be affected when individuals on teams have differing levels of professional stature and seniority, expertise, or experience. This authority gradient may be especially intrusive if senior team leaders wield some influence in the career advancement of those they supervise. The concept of authority gradient was introduced to medicine in the Institute of Medicine report To Err Is Human,2 but little has been written or acknowledged in the medical literature regarding the impact of authority gradients in medical care. Medicine is taught and practiced in a largely hierarchical system, one that utilizes the benefits of authority (and clinical expertise) to guide trainees and structure teams. In fact, the role of experience and authority is highly valued; many would argue that when properly used, authority gradients ought to minimize medical error. The concept that authority gradients might contribute to medical errors is provocative.
1342
Cosby and Croskerry
The Historical Context of Authority Gradients. History has given us many examples in which authority gradients contributed to harm. One of the world’s greatest maritime disasters happened in 1707 when Sir Clowdisley Shovell led 18 ships into disaster in the Isles of Scilly where they crashed on heavy rocks, killing 2,000 men.3 Some accounts claim that a seaman had warned him the day before that the projected route was ill-advised; the lowly seaman was hanged on the spot for mutiny for questioning his admiral’s decision. The aviation and aerospace industries have long recognized the role of authority gradients in collisions. More than 25 years ago, Edwards noted that differences among crew members affect the ability to give and receive information at critical moments and contribute to failure to detect and abort dangerous actions.4,5 Since then, a growing body of evidence has shown the significance of authority gradients in aviation incidents. When officers of different ranks occupy a helicopter cockpit together, the rate of aircrew mishaps increases.6 As the transcockpit authority gradient increases, so does the number of incidents. As many as 40% of juniorranking copilots reported a failure to relay significant doubts about safety to their pilots. When junior officers do offer their concerns, their senior-ranking officers may dismiss them without careful consideration. The failures of subordinate first officers to challenge their captains despite dangerous actions and violations of safety rules, and the failures of captains to heed warnings from crew members, are cited as factors in a number of airplane crashes, including the runway collision of two airliners at Tenerife in 1977 in which 583 people died.7 In another highly visible incident, critical information known by National Aeronautics and Space Administration (NASA) engineers failed to reach authorities responsible for making the final decision to launch the space shuttle Challenger, leading to disaster and loss of life.8 Authority Gradients in Medical Error. There have been recent high-profile cases in which authority gradients played a role in medical errors. A junior resident knowingly administered intrathecal vincristine to a patient, against his own judgment, pressured by a more senior physician. After repeatedly questioning his superior, he finally accepted the reassurances of his supervising physician. The patient died.9–11 In another recent event, a resident physician prepared to perform a left hemivulvectomy for microinvasive vulvar carcinoma when the attending surgeon redirected him to make the incision on the right. The resident complied despite formal reports that documented disease on the left. This wrong-site surgery made an already devastating procedure all the more agonizing for the patient.12 Anecdotal accounts of authority gradients in medical errors abound, yet there is a relative dearth of reports in the literature
d
AUTHORITY GRADIENTS IN MEDICAL ERROR
that acknowledge the role authority gradients contribute to poor decisions. This fact itself may represent an authority gradient effect. Authority Structure in Medical Education. Medicine is an organization with a rigid and steep hierarchical structure. It may have evolved that way in part because much of medicine involves pattern recognition, and senior physicians are likely to have the most experience to draw on to recognize variations in the presentation of disease. There may be the implicit assumption that the most senior team member will also have the best judgment. In a study of the professional socialization process in medical student training, Sinclair13 found that physician trainees were inclined to value their seniors’ experience and responsibility over their own knowledge, and were willing to give deference to personal authority over scientific merit in clinical decisions. Although there is acknowledgment that teamwork is important in medical systems, there are still those in medicine who believe that junior team members should not question the decisions made by more senior team members.14 Conflict in Medicine and Medical Training. Authority gradients and conflicting roles may affect the relationship between attending and resident physicians.15 The resident has a strong sense of personal accountability to individual patients. Emergency medicine residents are taught to focus on recognizing serious disease; their role is to detect and warn of any potential life-threatening condition. In contrast, attending physicians assume a greater responsibility to the overall system and are held accountable for flow and efficiency. Their role, unlike the resident, is often to assuage fear and create a sense of calm and control. In order to oversee multiple patients simultaneously, attending physicians must prioritize their time to focus on cases they perceive to be difficult or unstable. The perception that an illness is common and typically benign might lead to a premature judgment that results in an incomplete assessment of the patient. Both physicians in the case presented here were acting appropriately for their respective roles. Attending physicians and residents may find conflict as they struggle to find a balance that optimizes the needs of individual patients and the demands of the organization as a whole. This delicate balance can be lost if either physician fails to address the concerns of the other. Differences of opinion in patient care are not uncommon. However, residents are often reluctant to voice a dissenting view and attending physicians may be unaware of their trainees’ concerns.16–18 Conflict is common in health care teams and, left unaddressed, can impair individual and team performance.19 Medical school does little to prepare
ACAD EMERG MED
d
December 2004, Vol. 11, No. 12
d
www.aemj.org
students for the role they will carry in health care teams, and its socialization climate may inadvertently both beget and perpetuate preventable error.20 A majority of fourth-year medical students encounter significant interpersonal conflicts early in their clinical training, primarily in their interactions with authority figures.21 As students and residents struggle to obtain clinical expertise and find their role in the team, they experience daily conflict that plays out at the bedside. How well their concerns are heard may determine how effective the team will be in delivering quality care. Although this case deals with communication barriers between attending physician and resident, similar barriers are well known between doctor and nurse and probably exist between other professional roles (e.g., specialist and generalist, surgeon and internist, consultant and primary care physician). Nurses have long argued that their voice is not listened to when doctors make rounds. They have fought to gain sufficient recognition in the culture of medicine to be heard and to influence critical patient care decisions.22,23 Authority, professional stature, and perceived expertise probably all contribute to some extent to communication barriers. Finding Potential Solutions Outside of Medicine. The military has a deeply entrenched hierarchical structure, yet it recognizes that at certain times safety takes precedence over rank. The ‘‘foreign object damage’’ walk across aircraft carriers has special rules; officers and enlisted crew members walk the decks of aircraft carriers together, equally sharing the responsibility to detect foreign objects that can threaten aircraft safety.24,25 Aviation has likewise adopted policies that empower crew members to take command away from the pilot if safety is compromised.26 Specific training can equip less-experienced and lower-rank team members to effectively challenge their superiors. Likewise, superiors can be trained to recognize prompts and cues to alert them to risk and danger. Business and management schools recognize that academic success alone does not necessarily predict success in business. Some of the softer human skills such as empathy and effective communication skills, referred to as emotional intelligence, are ultimately as important to individual and organizational success.27 Business and management schools include teamwork training, leadership, and communication skills in their curricula. In contrast to other professional schools, medical education has focused almost entirely on the acquisition of medical knowledge. Students are rewarded for academic success largely on their individual merit, neglecting the importance of interpersonal skills. The ability to interact with others or within a team structure is neither taught nor recognized. It is in-
1343 teresting to note that many disciplines outside of medicine provide promising ideas for improving communication that may benefit physicians. Potential Solutions: Teamwork Training. The recent Institute of Medicine report on medical error emphasizes the importance of teams and the need to improve communication among caregivers.2,28 Risser et al.29 propose a structured teamwork approach emphasizing communication and openness as one approach to minimizing error. This involves assigning responsibility for communicating a common understanding of the patient condition and ‘‘situational awareness’’ to each team member.30 The model of Crew Resource Management (CRM), originating from the aviation industry, emphasizes an open communication environment where decision making is shared equally by all team members.31 It is regarded as a potential model for improving patient care. Emergency response teams in nuclear power plants have shown that effective teamwork can improve the flow of information and increase team efficiency. It is likely that well-organized teamwork could similarly improve efficiency and accuracy in the medical domain.32 Throughout medical training, students and residents are encouraged to ask questions and seek explanations. However, when curiosity turns to concern and there is a perception of imminent risk, both the resident and the supervisor need a language to communicate their degree of concern. Medical education may do well to acknowledge the importance of formal communication skills, similar to those taught by CRM and employed by aviation. CRM teaches skills to communicate escalating threats, beginning with an objective statement or question (‘‘How might I recognize this complication?’’), progressing to a more active statement of concern (‘‘I’m worried’’), and then, in extreme cases, culminating in a more direct challenge (‘‘Something is wrong! You need to see this patient now’’).33 This technique helps both the trainee and the supervisor distinguish between simple curiosity and real threat. Disagreements and conflict are common and can be anticipated. Without a formal method to settle disputes, residents and students may be less likely to speak up when they believe a patient is at risk of a medical error, and attending physicians may be ill prepared to handle conflicts when they arise. One proposed solution is to obtain a mandatory second opinion by an independent, senior physician whenever significant disputes occur. Even if such a review cannot be performed immediately, the request for a review may be sufficient to prompt attending physicians to reconsider their conclusions and defend their decisions. Such a policy can provide a critical opportunity for both the residents and their attending physicians to address issues in patient safety.
1344
Cosby and Croskerry
d
AUTHORITY GRADIENTS IN MEDICAL ERROR
TABLE 1. Proposed System Changes and Educational Recommendations 1. Trainees should be empowered to present their opinions and information they consider relevant. 2. Authority figures need to recognize that valuable information comes from many sources and to carefully consider the contributions others provide. 3. Experienced clinicians should use clinical narratives to illustrate their own vulnerability to error. 4. Teamwork factors and patient safety should take precedence over authority roles. 5. Changes in medical education should begin early and continue throughout the professional lives of physicians to help them adopt healthy communication styles. 6. Systems should be designed to facilitate effective teamwork and provide support for clinicians.
Conflict in the Face of Uncertainty and Risk. The solutions may be more complex in medicine where real-time decisions often take place in the face of incomplete information, uncertainty, and risk. Unlike foreign object damage detection, most clinical situations do not have an obvious right or wrong action and may generate conflict. Often, diagnoses and actions are proven only after a test of time, when the patient responds (or fails to respond) to therapy. Uncertainty is a part of medical decision making, and teaching to help the individual and team manage uncertainty and risk is necessary. The trainee may not know when his or her concern is justified; the trainee’s ability to persuade may be affected by his or her own indecision. One lesson that can be gleaned for medicine is the need for team members to listen respectfully to each other’s concerns.34 Should an attending physician overrule a resident when a serious life threat is under consideration, at the very least, the attending physician ought to offer a cogent argument for his or her point of view. The trainee has a duty to the patient and should persist when he or she perceives real risk.
Implementing Change. The development of guidelines to effectively challenge the experienced clinician will be difficult. A certain degree of compliance and obedience is necessary in those in training, especially within complex organizations; without it, chaos would ensue and safety violations would increase. At the same time, however, obedience should never become blind. An overriding principle throughout medical training is that every individual is ultimately responsible for his or her actions in all situations. This is a guiding principle in the development of professional autonomy. Authority figures should be sensitive to those they supervise and aware of the potential for abuse of power. They should recognize, too, the need for different roles in different situations. Structured teaching sessions clearly call for a different approach from the clinical situation where real morbidity and mortality may result. Insight into error theory and teamwork dynamics should remove or at least attenuate any threat from being challenged by others. Authority is an inherent and necessary feature of any hierarchical system. The role of the expert carries
with it accountability and responsibility, traits that we should not undermine or lose. Authorities should not lose sight of their own fallibility and their interdependence with other team members. Consideration should be given to developing teaching techniques that minimize the negative aspects of authority gradients. A powerful example is when senior clinicians use clinical narratives from their own experience to illustrate errors and judgment failures. This serves the important purpose of bridging the gradient by permitting junior staff, and other members of the health care team, to witness secondhand the vulnerability of their mentors. Given that mimicry is an important feature of learning, it also models desirable behavior for those in training. As with many proposed ideas in patient safety, changes in professional roles and teamwork concepts will require cultural change in medicine and innovations in medical education, beginning at the undergraduate level and continuing throughout the professional careers of physicians. Recognizing authority gradients, improving teamwork, and finding strategies to optimize communication can facilitate the flow of information and optimize patient care decisions. A safety culture that acknowledges safety as everyone’s responsibility, promotes shared knowledge, and emphasizes teamwork can minimize events that endanger patients.
Complexity of Medical Error. This case is presented to define and highlight the concept of authority gradient, but it is important to acknowledge that most medical errors occur as a result of the complex interplay of many factors, both human and system. Cases of medical error deserve a thorough incident investigation to uncover all the factors that contribute to risk and potential for harm. Authority gradients are more likely to exert a negative influence when medical information is incomplete, decisions are rushed, and actions are irreversible—all characteristics typical of emergency department practice. In the case described in this article, a peak season of viral illness led to high volume and acuity. Inadequate staffing, disorganized teamwork, and limited support systems all contributed to the tendency to prematurely discharge patients. Awareness of authority gradients may help clinicians avoid their negative impact, but
ACAD EMERG MED
d
December 2004, Vol. 11, No. 12
d
www.aemj.org
more importantly, systems can be designed to minimize the overall risk of a single bad decision. System improvements can help by decreasing cognitive load, improving support, and providing checkpoints for rescue and recovery. A healthy and productive response (for example, the proposed changes listed in Table 1) to an event such as this should acknowledge the role of authority gradients, but also address hospital-wide, system-wide improvements to optimize the working environment of the front-line clinician. References 1. Brogan TV, Nizet V, Waldhausen JH, Rubens CE, Clarke WR. Group A streptococcal necrotizing fasciitis complicating primary varicella: a series of fourteen patients. Pediatr Infect Dis J. 1994; 14:588–94. 2. Kohn LT, Corrigan JM, Donaldson MS (eds). To Err Is Human: Building a Safer Health System. Institute of Medicine Report. Washington, DC: National Academy Press, 2000, pp. 180–1. 3. Sobel D. Longitude. The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time. New York: Penguin, 1995. 4. Edwards D. Stress and the airline pilot. Presented at the BALPA Technical Symposium: Aviation Medicine and the Airline Pilot. Department of Human Sciences, University of Technology, Loughborough, October 1975. 5. Hale AR, Glendon AI. Individual Behaviour in the Control of Danger, Industrial Safety Series 2. Amsterdam: Elsevier; 1987. 6. Alkov RA, Borowsky MS, Williamson DW, Yacavone DW. The effect of trans-cockpit authority gradient on navy/marine helicopter mishaps. Aviat Space Environ Med. 1992; 63:659–61. 7. ‘‘CRM: The missing link.’’ Available at: http:// www.airlinesafety.com/editorials/editorial3.htm. Accessed Oct 17, 2003. 8. Vaughan D. The Challenger Launch Decision. Risky Technology, Culture, and Deviance at NASA. Chicago: University of Chicago Press, 1996. 9. Cancer patient, 18, critical after drug injection blunder. Daily Mail. 2001;Jan 24:11. 10. Berwick DM. Not again! [letter] BMJ. 2001; 322:247–8. 11. Sanderson I. Can an organization have a memory? Available at: http://www.arch.net.au/content/file/download.phtml/ type/ARCHIPresentation/id/1657/field/attachment/name/ Sanderson_PPTslides.pdf. Accessed Oct 17, 2003. 12. Vincent C. Hemivulvectomy: the wrong side removed. Available at: http://www.webmm.ahrq.gov/ spotlightcases.aspx?icd=33. Accessed Oct 23, 2003. 13. Sinclair S. Making Doctors: An Institutional Apprenticeship. Oxford: Berg; 1997. 14. Sexton JB, Thomas EJ, Helmreich RL. Error, stress, and teamwork in medicine and aviation: cross sectional surveys. BMJ. 2000; 320:745–9. 15. McCue JD, Beach KJ. Communication barriers between attending physicians and residents. J Gen Intern Med. 1994; 9:158–61.
1345 16. Shreves JG, Moss AH. Residents’ ethical disagreements with attending physicians: an unrecognized problem. Acad Med. 1996; 71:1103–5. 17. Coats RD, Burd FS. Intraoperative communication of residents with faculty: perception versus reality. J Surg Res. 2002; 104: 40–5. 18. Farber NJ, Weiner JL, Boyer EG, Robinson EJ. How internal medicine residents resolve conflicts with attending physicians. Acad Med. 1990; 65:713–5. 19. Jellinek MS. Recognition and management of discord within house staff teams [commentary]. JAMA. 1986; 256:754–5. 20. Lester H, Tritter JQ. Medical error: a discussion of the medical construction of error and suggestions for reforms of medical education to decrease error. Med Educ. 2001; 35: 855–61. 21. Spiegel DA, Smolen RC, Jonas CK. Interpersonal conflicts involving students in clinical medical education. J Med Educ. 1985; 60:819–29. 22. Zwarenstein M, Reeve S. Working together but apart: barriers and routes to nurse–physician collaboration. Jt Comm J Qual Improv. 2002; 28:242–7. 23. Coombs M. Power and conflict in intensive care clinical decision making. Intensive Crit Care Nurs. 2003; 19:125–35. 24. Pfeiffer J. The secret of life at the limits: cogs become big wheels. Smithsonian. 1989; 20:38–48. 25. Tuthill W. FOD.deadly but preventable. Mech. The Naval Safety Center’s Aviation Maintenance Magazine. 1999; 2:10–3. 26. Air Force Materiel Command. AFI11-290_AFMCS1: Cockpit/Crew Resource Management Training Program. June 25, 2002. Available at: http://www.afmc.wpafb.af.mil/pdl/ afmc/sup/11series/11_290/11-290s.pdf. Accessed Nov 15, 2003. 27. Goleman D. Working with Emotional Intelligence. New York: Bantam, 1998. 28. Institute of Medicine: Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: Committee on Quality of Health Care in America; 2001. 29. Risser DT, Simon R, Rice MM, Salisbury ML. A structured teamwork system to reduce clinical errors. In: Spath PL (ed). Error Reduction in Health Care. A Systems Approach to Improving Patient Safety. San Francisco: AHA Press, 2000, pp 235–78. 30. Risser DT, Rice MM, Salisbury ML, Simon R, Jay GD, Berns SD. The potential for improved teamwork to reduce medical errors in the emergency department. MedTeams Research Consortium. Ann Emerg Med. 1999; 34:373–83. 31. Kosnik LK. The new paradigm of crew resource management: just what is needed to reengage the stalled collaborative movement? Jt Comm J Qual Improv. 2002; 28:235–41. 32. Klein G. The power of the team mind. In: Sources of Power. How People Make Decisions. Cambridge, MA: MIT Press, 1999, pp 233–57. 33. Prinea S. Communication and patient safety. Presented at the Third Halifax Symposium on Healthcare Safety: From Theory to Reality. Dalhousie University, Halifax, NS, Canada, Oct 18, 2003. 34. Norman GV. Interdisciplinary team issues. Available at: http://www.eduserv.hscer.washington.edu/bioethics/ topics/team.html. Accessed Oct 17, 2003.
