ACADEMIC EMERGENCY MEDICINE • November 2000, Volume 7, Number 11
1189
System Contributions to Error JAMES G. ADAMS, MD, J. STEPHEN BOHAN, MD
Abstract. An unacceptably high rate of medical error occurs in the emergency department (ED). Professional accountability requires that EDs be managed to systematically eliminate error. This requires advocacy and leadership at every level of the specialty and at each institution in order to be effective and sustainable. At the same time, the significant operational challenges that face the ED, such as excessive patient care requirements, should be recognized if error reduction efforts are to remain credible. Proper staffing levels, for example, are an important prerequisite if medical error is to be minimized. Even at times of low volume, however, medical error is
A
NY SYSTEM or any human performance is prone to error. Proper system design should always include error prevention mechanisms, including those directed at individual behavior. There has been scant attention to this within emergency medicine (EM). There are no systematic identification and analysis of errors and little substantive research regarding medical error in the emergency setting. In order to begin this process, we must ensure insight into the current environment of emergency care by confronting essential operational, environmental, and professional factors that influence medical error. These views were expressed by emergency department (ED) leaders, academicians, opinion leaders, and practicing emergency physicians. Some views are, admittedly, editorial and not assured to be unanimous. All are thought-provoking, insightful, and important to consider as we embark on efforts to reduce error in EM.
THE CURRENT LANDSCAPE OF EM Many EDs, especially the busiest centers, are operating under conditions of extraordinary stress. Facing record numbers of patients seeking care, highly complex patients in need of detailed evaluFrom the Department of Emergency Medicine, Northwestern Memorial Hospital, and the Division of Emergency Medicine, Northwestern University Medical School, Chicago, IL (JGA); and the Department of Emergency Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA (JSB). Received June 28, 2000; accepted July 12, 2000. Address for correspondence and reprints: James G. Adams, MD, 216 East Superior Street, Suite 100, Chicago, IL 60611. Fax: 312-908-6004; e-mail:
[email protected]
probably common. Engineering human factors and operational procedures, promoting team coordination, and standardizing care processes can decrease error and are strongly promoted. Such efforts should be coupled to systematic analysis of errors that occur. Reliable reporting is likely only if the system is based within the specialty to help ensure proper analysis and decrease threat. Ultimate success will require dedicated effort, continued advocacy, and promotion of research. Key words: errors; emergency department; public health. ACADEMIC EMERGENCY MEDICINE 2000; 7:1189–1193
ation, increased expectations for extensive diagnostic testing, an aging population, demands to control costs, increasing service expectations, and financial operating deficits, many ED leaders have declared their environment catastrophic. Any dialogue on medical error must begin with an understanding of the pressures that are placed on motivated and talented professionals. On a daily basis, there are often insufficient space, minimal staffing, barely sufficient support services, and lack of timely access to inpatient beds. Many of these problems are most acute in departments without the financial resources to correct deficiencies. It would be deceptive and unfair to declare efforts to reduce medical error without recognizing the immense daily challenges presented. Even in this setting, or perhaps because of these circumstances, a focus on error reduction is needed. A concurrent dialogue is required in order to address these serious operational concerns if calls for error reduction are to remain credible. In addition, a unifying philosophy must be accepted. All stakeholders, including physicians, nurses, and hospital administrators, must have genuine commitment to serve all patients well. Patients, therefore, must not be blamed for seeking care. Emergency caregivers, primary care physicians, insurers, and others, continue to believe that much ED care is ‘‘unnecessary.’’ Patients who are willing to put up with the distress of an ED visit obviously have self-identified a need. Debating the credibility of the needs is less productive than identifying efficient ways to meet it. The past generation of experience and current cultural observations tell us that attempts to limit demand for unscheduled care will not succeed.
1190
PROCEEDINGS
While attempting to ensure adequacy of the current health care structure and meeting current demand, preparations should be made for future patient volume and complexity. While many EDs are now faced with unprecedented numbers of people seeking emergency care, it is likely that these numbers will still increase. It is probable that the current resource shifts in medical care delivery, resulting in fewer inpatient beds and increased delivery of health care outside of the hospital, are contributing to an increased reliance on the ED for diagnosis and treatment. Over the next 30 years there will be a near doubling of the population over the age of 65 years. With the aging of the population just beginning to affect the health care system, it is apparent that increasing numbers of people over 65 will require emergency care. In addition, there is an increased prevalence of chronic illness, as improved medical strategies allow increased survival. Improvements in the treatment of heart disease, pulmonary disease, kidney disease, neurological disease, liver disease, and others have increased survival of patients with serious chronic illness. At the other end of the age spectrum, success in neonatal and pediatric therapeutics has led to increased numbers of young patients with chronic illness. At every age, there are increased numbers of complex patients. These trends mean additional, continuing challenges for the health care system in general, and increased demands for the emergency care system in particular. While these challenges must be addressed and while social and political debate regarding health care financing must continue, the core of the emergency care process must still be critically analyzed. Even at times of low volume, the ED is prone to error. There are little automation, little consistency, little task standardization, few forced functions, questionable team coordination, and high variability. The system functions by virtue of human effort and human intention, but possesses few safety checks, backup systems, and redundancies to ensure safety. The system is primarily dependent on the expertise and judgment of individual physicians and nurses rather than being guided by standard procedures and automated systems that ensure consistency and minimize error. While it is clear that there is a need for standardization, automation, and safety functions, we must not lose sight of the primary determinant of health care quality. The centerpiece of a high-quality system remains the expert physician, with high levels of technical ability, knowledge, compassion, and effective communication skills, who has sufficient time to spend with a patient. In the context of a successful patient relationship and effective communication there is greater likelihood of deci-
Adams, Bohan • SYSTEM CONTRIBUTIONS TO ERROR
sional accuracy, medicolegal safety, patient satisfaction, and physician satisfaction. Error reduction efforts must preserve or enhance the time that the physician spends with the patient. Supporting this relationship would be good communication, reliable support systems, and teamwork among the ED staff. This teamwork would extend to hospital units and to consultant physicians. The primary care physicians, when available, should be accessed as team members given their potential ability to offer information, context, and ongoing care. Such elements might seem basic, but multiple competing demands can draw the physician away or fail to sufficiently enable these essential elements. With these fundamentals in mind, an administrative plan for error reduction can be begin.
SUPPLY –DEMAND MATCH At the outset of an error reduction plan, a sound understanding of system capacity is required. Emergency departments, because of a failure to objectively identify capacity, are assumed to be infinitely expansible. Excessive demand is currently managed by increasing patient waiting time and by caring for patients in the hallways. Long waiting times have been accepted as normative, even if unsettling to patients and staff. Caregivers and administrators must, defensively, develop diminished sensitivity to the discomfort of waiting patients since no effective alternative response has been traditionally available. This contributes to an unsatisfying and stressful environment that is likely error-prone. The specialty of EM must not accept these defective processes. Leaders of hospitals, in order to reduce medical error, must help set acceptable limits, critically analyze operational challenges, and attempt solutions. Much progress can be made in reducing medical error even in a chaotic, disorganized, overburdened, and inefficient ED, but the goal must be to restore order, organize, unburden, and make the unit operate with consistency and efficiency. If a hospital’s 15-bed intensive care unit had 20 patients, but no additional space or staffing, a crisis would be declared and the unit would be closed. If a 20-room operating suite carried out 30 simultaneous surgeries, a problem would be noted. If an airport increased take-offs and landings from 150/ day to 200/day with neither added resources nor deliberate attention to handling the increased volume, near-misses and dangerous errors would result. Yet if a 24 bed ED has 48 simultaneous patients, or if demand suddenly surges, this is dismissed as normal operations. This flawed logic is a dangerous contributor to error. Clearly, an ED does have safe limits of expansibility, but these
ACADEMIC EMERGENCY MEDICINE • November 2000, Volume 7, Number 11
must be defined. When excessive demand risks inability to meet acceptable standards, some operational response is required. A sufficient infrastructure is not a luxury, but a necessity. Excessive waiting times and overcrowding clearly lead to increased morbidity, although this morbidity is generally not attributable to conventional notions of medical error. There is growing evidence that time to treatment affects outcome (time to open coronary artery; time to open cerebral artery; time to antibiotics in pneumonia; time to pain relief ). We currently characterize delays in care as a ‘‘deficiency’’ but it is clear that a failure to perform an action that leads to an adverse outcome fits under the definition of error. Operating over capacity leads to error through excessive distractions, interruptions, and multiple competing demands. The pressure to move quickly risks a lack of thoroughness. While efficiency and situational awareness must be emphasized, there should be diminished reliance on crude notions of speed. Decreasing time spent with patients, cutting corners, and prematurely making decisions because of time pressure all potentially contribute to error. Efficiency emphasizes the systematic elimination of delays, barriers, obstructions, and unnecessary steps. This notion moves beyond a goal of ‘‘2.5 patients per hour seen per doctor’’ and begins to look at the relationship of the care providers to each other as well as to the environment and the system. An optimal system would also help quantify incoming demand. At threshold levels, additional services should be added. In order for this to occur, capacity must be defined and thresholds need to be related to capacity. For example, at 120% of capacity, a defined response might be triggered. This requires sufficient space, organization, and cooperation among units, because the response often must include other hospital departments and inpatient units. The ED might add physicians and nurses, for example, but this would trigger increased demand for lab, radiology, secretarial, technician, inpatient, and consultant services. To be effective, the entire system must coordinate a response. Such plans exist in every hospital for disasters. Yet some EDs operate frequently at volumes that overwhelm, meeting ‘‘disaster’’ criteria. The disaster plan is not triggered only because the collective group did not become injured at a single event. Inadequate or absent response to extreme demands contribute to error.
EXPERT PERSONNEL, PROFESSIONALISM, AND ACCOUNTABILITY The emergence of specialty training for emergency physicians over the past three decades has led to
1191
enormous improvements in the quality of diagnostic and therapeutic decision making. A related landmark shift occurred in the late 1980s, when attending oversight, 24 hours/day, 7 days/week, was required in academic centers. The mandate that attending physicians must supervise trainees 24 hours/day marked a radical shift in educational philosophy. No longer were residents charged to learn by making mistakes. Now the residents were charged to provide expert care, avoiding mistakes, while learning from experienced mentors. These ideals were a matter of debate in the 1980s because they went against the traditional notions of educational philosophy. In the end, the interests of the patient prevailed. Still, in many areas of academic medicine, relatively inexperienced residents continue to be given high levels of authority, directly impacting the quality and efficiency of care for ED patients. Such traditional notions must be scrutinized as well. Traditional reliance on resident autonomy has been coupled with strong belief in individual accountability. Residency training has traditionally been a hierarchical process of questioning, probing, ridiculing, ‘‘pimping,’’ and second-guessing, providing patient safety through multiple levels of checks by more senior physicians. The junior physician is made to feel personally and professionally accountable. No excuses are accepted; no blaming of the ‘‘system’’ is allowed. Residents present mistakes at morbidity and mortality conferences for additional professional, open scrutiny. This belief in individual accountability is good, if it does not extend to a pathological extreme, but an additional component is required. Genuine accountability requires a systematic analysis of error and administrative effort to change the system to eliminate errors. In addition to individual accountability, there must be collective professional accountability, demonstrated through organizational and operational change. A similar notion of accountability must be shared by all who have power in the organization. This includes administrative leaders and nursing leaders. These groups are similarly bound by tradition. All groups must share a notion of accountability, duty, and professionalism if the difficult changes necessary to reduce medical error are to be successful. If efforts to reduce medical error are to be more than a passing fad, then a deep, ongoing, abiding commitment must be made. This commitment must be borne of individual and collective values, shared by key leaders. Although intentions are almost always good and talk can be persuasive, it is not certain that the necessary depth of commitment and passion is common in most health care organizations. Attention is necessarily directed at maintaining financial solvency. Because
1192
PROCEEDINGS
multiple leaders, physicians, nurses, and administrators, share influence, the commitment of all is required. Since there is generally no single organizational leader, they must band together with a common vision.
TEAM EFFORT Local Emergency Department Team. A supportive physician–nurse–secretary–technician team is important to a well-functioning unit and can play an important role in error reduction. In this spirit, genuine unity and respect must be developed, although it is recognized that respect is earned through technical competence, positive interpersonal dynamics, and a strong work ethic. The current shortage of highly qualified emergency nurses threatens team unity and, perhaps, achievement of ideal levels of unit cohesion. Technical expertise of all team members, supported by ongoing training, is an important requirement for team development. Additional explicit training in teamwork may also substantially improve departmental function, while reducing oversights, errors, and miscommunications. Currently, there is high shift to shift variation, due to individual preferences, unique practice styles, and individual opinions regarding diagnostic, therapeutic, and operational standards. An error-free unit cannot run in this fashion. Some common standards and consistency are required for efficiency and safety. This is not to assert that individual judgment and decision making are to be superseded by rules. It does mean that thoughtful guidelines and accepted standard operating procedures will lead to less error because nurses, technicians, and even secretaries can then learn the procedures and help to make the right thing happen. Part of team coordination is understanding the role and responsibilities of other team members, supporting each other, communicating clearly, receiving communication positively, and having mechanisms to resolve conflict. The team should have feedback regarding the quality of functioning. Hospital-wide Team. The ED system of care is a complex orchestration of multiple services. The delivery of medical care relies on segregated services that have developed along specialty lines, often leading to competition instead of cooperation. Further, the nursing and administrative services are usually managed independently from physicians. There is generally no single leader. Issues and goals are negotiated among the emergency physician and nurse as well as with in-hospital units, laboratory and radiology, consulting services, and other hospital departments. All contribute to the
Adams, Bohan • SYSTEM CONTRIBUTIONS TO ERROR
success or failure of the ED. It is incumbent upon all leaders to attempt to achieve supportive relationships and consistent service. Blame is frequently cast to ED when the system fails, but the contributions of many others are sometimes less appreciated. Despite claims of sophisticated decision making, managers are often prone to error in the analysis of system problems. Notably, proximity bias leads those outside of the ED to attribute problems that are based close to the ED to emergency caregivers. Proper analysis is required for productive discussion and problem solving to occur. Those who work directly with the ED must operate as partners in the system, ensuring effective communication and good mutual understandings, while possessing insight into ED operational norms and policies. High levels of responsibility are given to consultants. If consultants in the ED view themselves as autonomous rather than partners with the ED, there is increased risk of miscommunication, frustration, and error. Consultants are sometimes unfamiliar with ED standards and the ED systems of care, know few of the personnel, and are unfamiliar with the environment, the equipment, and the processes. They are not oriented to the unit. It is predictable that frustration could result in disagreement and conflict. Unfamiliarity and conflict might enhance the likelihood of error.
CONTINUOUS IMPROVEMENT AND ERROR ANALYSIS An effort to reduce medical error requires systematic identification and analysis of existing errors. Such effort has not traditionally been part of the practice or administration of EM. This process, although intended for the patient’s benefit, is not essentially a patient-centered process. It is a technical, administrative, and medical process. Public relations and marketing should not be central, so analysis and deliberations regarding error and potential error should be done in private. If done properly, nonmedical personnel could ultimately be incorporated as a partner so effective communication and enlistment of patients could be beneficial. Provisions must be made for confidentiality of individual reports, providing a safe mechanism for reporting, driving out fear. This must be coupled with a meaningful, effective process to analyze errors and intervene to prevent future errors. High threat would predictably interfere with reporting. Such threat could come from many sources: supervisor, hospital, licensing board, public health agencies, press, malpractice attorneys, and peers. Error analysis must be performed by practicing physicians and nurses to ensure full insight and avoid
1193
ACADEMIC EMERGENCY MEDICINE • November 2000, Volume 7, Number 11
misattribution. Error analysis must also include those who can provide insight into factors outside of the ED that impact the operation. The error reduction team also must be able to implement change, so it must include leaders with operational and financial control. Potential negative unintended consequences of quality improvement efforts must be recognized. Solutions that seem clear in a meeting can introduce subtle problems when implemented. For example, automation, computerization, and forced functions that address one specific operational need can add work and distract the physician from other core tasks. Physicians, notoriously uncooperative with change efforts, are still adaptable and could be caused to conform to detrimental solutions. Even if the physicians adapt to distracting steps, time might be taken away from the patient evaluation, communication with family, time spent listening to the nurse, or other important, but unmeasured, tasks. When introducing change, careful analysis of potential consequences must be undertaken, even while holding the possibility of unpredictable, subtle, but negative shifts. Even important change that has minor negative potential can be difficult to introduce and requires strong leadership. For example, it is probable that shift change is a high-risk time for error. In order to decrease hand-offs of patients, overlap of shifts might be instituted. Implementing such change, even if the change is minimally controversial, takes effective leadership in order to generate understanding, consensus, and acceptance. For changes that are not easily accepted, the leaders must be able to communicate the importance. Any change requires effective leadership. Error reduction efforts may lead to hotly contestable suggestions, such as restrictive rules and regulations. Mandatory rest periods, consecutive hour limits, and prohibitions against substances that could impair psychomotor functioning would be logical, but possibly controversial, targets. Some solutions are logical and are mandated by prudence. Some regulations might appear logical, but would have little effect in decreasing error. Some regulations would require evidence before physi-
cians could be convinced of the necessity. Since the causes and incidence of error in EM are unknown, it is difficult to defend specific solutions. Others might rightly note that more would be gained by directing initial efforts toward education, training, and team coordination. Only after the landscape of error in EM is understood can any but the most common-sense solutions be rationally supported. Surveillance networks, based within the specialty of EM, should be established to identify sources of error. Other reporting methods, such as filings to public agencies, would stimulate defensive and potentially adversarial posturing. Reporting to any source unable to provide helpful insight or meaningfully assist with the reduction of error is unlikely to be effective. The role of public agencies is not to operate the monitoring system, but to ensure that a sound system for identifying, monitoring, and correcting error is in place. Ultimately, accountability is to the public, so statistics and data could be made available. The goal is not a unit where error is not possible, but a unit that can rapidly identify potential error and self-correct long before there is any negative impact on a patient. Multiple levels of safeguards would ensure that no error results in harm. In order for this to become reality, sustained advocacy and coordination among all who have influence are required.
CONCLUSIONS Emergency departments have not been managed to systematically eliminate error. The effort of the past generation has been directed toward achieving technical competence. This effort must not be taken for granted, and much work remains. The technical expertise of nursing and technicians must also be promoted. At the same time, human factors, operational procedures, team coordination, and care delivery systems should be engineered to decrease error. These efforts can be undermined by patient care demands that overwhelm the emergency care system. Ultimate success will require dedicated effort, continued advocacy, and ongoing analysis of ED errors.
