The Potential for Improved Teamwork to Reduce Medical Errors in the Emergency Department

The Potential for Improved Teamwork to Reduce Medical Errors in the Emergency Department

CONCEPTS The Potential for Improved Teamwork to Reduce Medical Errors in the Emergency Department From Dynamics Research Corporation,* Andover, MA; D...

99KB Sizes 0 Downloads 21 Views

CONCEPTS

The Potential for Improved Teamwork to Reduce Medical Errors in the Emergency Department From Dynamics Research Corporation,* Andover, MA; Departments of MedicineII and Pediatrics,‡ Brown University School of Medicine, Providence, RI; Section of Emergency Medicine, Madigan Army Medical Center, Department of Emergency Medicine,§ Fort Lewis, WA.

Daniel T Risser, PhD* Matthew M Rice, MD, JD§ Mary L Salisbury, RN, MSNII Robert Simon, EdD* Gregory D Jay, MD, PhDII Scott D Berns, MD, MPH‡ The MedTeams Research Consortium

Received for publication June 23, 1998. Revision received March 5, 1999. Accepted for publication April 29, 1999. Presented at the annual meeting of the Society for Academic Emergency Medicine, Chicago, IL, May 1998, and the American College of Emergency Physicians Management Academy, New Orleans, LA, May 1998. Supported by Army Research Laboratory Contract #DAAL01-96C-0091. Address for reprints: Daniel T Risser, PhD, Crew Performance Group, Dynamics Research Corp, 60 Frontage Road, Andover, MA 01810; 978-475-9090, fax 978-474-9059; E-mail [email protected]. Copyright © 1999 by the American College of Emergency Physicians. 0196-0644/99/$8.00 + 0 47/1/99701

SEPTEMBER 1999

34:3

See related article, p. 370. This article describes emergency department care work teams designed to improve team communication and coordination and reduce error. The core of this teamwork system is the teaching of teamwork behaviors and skills, development of teamwork habits, and creation of small work teams, all of which are key teamwork concepts largely drawn from successful aviation programs. Arguments for enculturating teamwork into ED practice are drawn from a retrospective study of ED malpractice incidents. Fifty-four incidents (1985-1996), a sample of convenience drawn from 8 hospitals, were identified and judged mitigable or preventable by better teamwork. An average of 8.8 teamwork failures occurred per case. More than half of the deaths and permanent disabilities that occurred were judged avoidable. Better teamwork could save nearly $3.50 per ED patient visit. Caregivers must improve teamwork skills to reduce errors, improve care quality, and reduce litigation risks. [Risser DT, Rice MM, Salisbury ML, Simon R, Jay GD, Berns SD, The MedTeams Research Consortium: The potential for improved teamwork to reduce medical errors in the emergency department. Ann Emerg Med September 1999;34:373-383.] INTRODUCTION

Over a decade of aviation research has shown that effective teamwork is essential to flight safety.1-4 Both military and commercial aviation organizations have standardized teamwork training systems in place because experience has shown that effective teamwork does not arise spontaneously but rather requires specific skill development and practice. The objective of the training is to reduce the risk that crews will make a fatal error or permit a fatal chain of errors to unfold because they failed to foster teamwork, solve problems, communicate, and man-

ANNALS OF EMERGENCY MEDICINE

3 7 3

TEAMWORK AND ERROR REDUCTION Risser et al

age workload. Validated training in aviation has significantly improved teamwork, resulting in saved aircraft and lives.2,4,5 Emergency medicine, with its needs for quick decisionmaking using incomplete information and demand for effective coordination of groups of caregivers for rapid critical care delivery, has much in common with aviation. 6 Good patient outcomes can be undermined when emergency caregivers improperly coordinate care or fail to help each other avoid clinical errors. In their most extreme forms, such clinical errors trigger patient morbidity and mortality.7-9 In less extreme forms these clinical errors do no permanent harm but do cause frustration and waste time and resources. To protect against errors and their effects, it is essential that emergency care systems begin to recognize the reality that all people are prone to err and will frequently make errors.10,11 Moreover, given complex tasks, long work hours, and difficult vigilance duties, even the brightest, most diligent, and conscientious clinician will make errors frequently.6,12,13 A willingness to squarely face this problem of inherent human error and seek systematic solutions has been missing in the health care community.6-9,13 The education systems in nursing and medicine have historically done an excellent job of teaching clinical skills to individuals, but superb individual clinical skills do not guarantee effective team performance in care delivery.14 Teaching of teamwork skills and team concepts as integral to the delivery of emergency department clinical care is virtually nonexistent. To help emergency caregivers develop more effective teamwork skills, the teamwork lessons learned in aviation have been organized into a behavior-based teamwork course focused on specific learnable skills that support the emergency medicine environment.15,16 The ultimate goal is to provide training that will enable ED staff to form effective teams17 and improve department performance and patient care. The goals of this article are as follows: (1) to introduce the ED community to important constructs and behaviors associated with teamwork systems, and (2) to describe the teamwork failures that commonly occur during ED clinical care and the consequences of failures to use teamwork to break clinical error chains. The following summary of an actual closed malpractice case provides a sense of how clinical and nonclinical actions are intertwined in the delivery of quality clinical care.18 This case reveals several common teamwork failures and demonstrates the potentially dramatic consequences of such failure. A 39-year-old woman with a history of documented coronary artery disease came to the ED

3 7 4

complaining of increased frequency of anginal chest pains over the preceding 2 weeks. She denied any current chest pain at triage but did report mild shortness of breath. She was triaged as “urgent,” the second highest triage category in a 4-tier triage system, even though she had abnormal vital signs and a history that should have placed her in the highest category at this institution. The ED was extremely busy, and almost 1 hour elapsed before she was evaluated by a medical student. At that time she complained of (3/10) chest pain and was found to have weak-to-absent peripheral pulses. Ninety minutes after presentation to the ED, repeat vital signs showed a blood pressure of 61/32 mm Hg, but this was not communicated to the medical student or the physician. To relieve the patient’s chest pain, the physician ordered sublingual nitroglycerin. After the incident, the nurse reported on a written statement that she was uncomfortable giving nitroglycerin, a drug she knew could lower blood pressure, to a hypotensive patient but assumed that the physician “knew what he was doing.” The patient continued to complain of chest pain and shortness of breath, and morphine sulfate was given and a nitroglycerin drip was started. Almost one-half hour after the initial hypotensive episode, her low blood pressure was acknowledged by the physician, and the nitroglycerin infusion was discontinued. At this time a consult with an internal medicine physician was called. The internal medicine resident arrived in the ED one-half hour later. The patient remained hypotensive and dyspneic and continued to have chest pain. Finally she became extremely bradycardic, lost her blood pressure, and a “code” was called. Resuscitation per advanced cardiac life support protocols, including epinephrine, atropine, defibrillation, external pacing, and pericardiocentesis was unsuccessful, and the patient was pronounced dead 3 hours and 10 minutes after entering the ED. This case demonstrates a chain of errors in which poor organizational climate, lack of team structure, poor task prioritization, poor communication, lack of cross-monitoring (team members checking each other’s actions), and lack of assertiveness within the ED contributed to a catastrophic patient outcome. The consequences of this team failure were dramatic: the patient died, the family was devastated, the staff was distressed and demoralized, the hospital’s reputation was harmed,

ANNALS OF EMERGENCY MEDICINE

34:3 SEPTEMBER 1999

TEAMWORK AND ERROR REDUCTION Risser et al

and more than $2 million was paid in settlement. 18 Teamwork failures like these, although not usually resulting in such a disaster, are more common than most clinicians or patients want to believe. THE TEAMWORK SYSTEM

The teamwork model presented here is an overview of the system created by the MedTeams Project,* a large applied translational research project that has developed an emergency care teamwork system based on a successful aviation model. The system is currently undergoing field evaluation in 10 US EDs. The teamwork system is designed to improve care delivery performance and reduce the number of clinical errors that occur. It encourages team members to actively coordinate and support each other in the course of clinical task execution by using the structure of work teams. Teams and teamwork behaviors do not replace clinical skills. Teamwork actions ensure that clinical activities are properly integrated and executed to deliver effective emergency care. Teamwork gives caregivers increased control over their constantly changing environment and a safety net to help protect patients and caregivers from inevitable system and human failings and their consequences. ED caregivers typically have a general concept of a “team” that lacks the precision necessary to create practical, manageable work teams. Often included within the “team boundary” are not only the many members of the ED staff but also members of other departments, such as laboratory and housekeeping. This is “team” in the large loose sense of the word and not “team” in the desired sense of tightly coordinated, mission-focused, technically skilled small groups (eg, athletic teams or cockpit crews). In the MedTeams system a work team, 17,19 referred to as an ED core team, is a set of 3 to as many as 10 clinically skilled caregivers (average team size=6) who work together during a shift and have been trained to use specific teamwork behaviors to tightly coordinate and manage their clinical actions. Their goal is to deliver high-quality ED clinical care to the set of ED patients assigned to them. Each core team contains at least 1 physician and 1 nurse. The most experienced physician on the team is the designated team leader. The team leader only serves as leader to 1 team during a work shift. Team mem*For more details on the MedTeams Project, go to the project Web site at http://teams.drc.com.

SEPTEMBER 1999

34:3

ANNALS OF EMERGENCY MEDICINE

bers always know who is on their team and who is the team leader. Different core teams in a department are distinguished by having members of each team wear a readily visible team identifier (eg, colored patches, armbands, badges, or colored scrubs) that denotes an individual as belonging to a particular team. Visible identifiers make caregiver coordination easier and also make it easier for patients to know who is responsible for their care and who they can expect to be immediately willing to help them. Note that the team system discussed here does not establish a fixed set of caregivers as a permanent core team. Rather it expects that all ED staff know how to work in core teams and are skilled in the use of standard teamwork behaviors. Specific core team composition will vary in most EDs from one day to the next. TEAM DIMENSIONS AND TEAMWORK BEHAVIORS

Studies on aircrew coordination2,3,5 have led to an integrating framework composed of 5 behavioral team dimensions. The emergency care–specific teamwork actions described here resulted from tailoring aviation teamwork behaviors to the emergency care setting through systematic observations by behavioral scientists, data collection and analysis, and a series of expert panel meetings of emergency physicians, emergency nurses, and behavioral scientists over a 2-year period (Table). Specific teamwork behaviors were identified that support each of the particular dimensions. These teamwork behaviors, the practical daily actions actually taken by core team members, are shown in column 4 of the Table (“teamwork actions by core team”). The primary descriptors (Table, column 3) are the major subheadings under the team dimensions that are used to organize the training course. The interrelationships of the 5 team dimensions are shown in Figure 1. The first team dimension, “maintain team structure and climate,” establishes and maintains appropriate team structures and an organizational climate conducive to teamwork. The teamwork behavior focuses on the daily formation and preparation of the core team for the work shift and expectations regarding professional interactions. Success in meeting the other 4 objectives presupposes success in meeting this first objective, which provides the foundation. Team dimensions 2, 3, and 4 address daily operational teamwork objectives within the ED. These objectives are not applied in any specific sequence but rather come into

3 7 5

TEAMWORK AND ERROR REDUCTION Risser et al

focus many times in the course of a clinical shift and are often intertwined with each other. Moreover, teams and team members rapidly and frequently change their focus

between clinical task execution and team coordination issues. A team member’s focus at any given point in time is driven by the need to balance 3 team responsibilities: (1)

Table.

MedTeams teamwork behavior matrix. Team Dimension No. 1

Team Dimension Maintain team structure and climate

Primary Descriptors *Establish leadership *Organize team *Cultivate team climate *Resolve conflicts constructively

2

Apply problem-solving strategies

*Conduct situational planning *Apply decisionmaking methods *Engage in error-correction actions

3

Communicate with the team

*Use standards of effective communication *Supporting information offered

*Supporting information requested 4

Execute plans and manage workload

*Implement plan *Primary and secondary triage *Prioritize tasks *Manage team resources and workload

*Cross-monitor team member actions *Maintain situation awareness 5

Improve team skill

*Engage in informal team improvement strategies *Engage in formal team improvement strategies

Teamwork Actions by the Core Team (% of cases in retrospective study in which teamwork failure contributed to clinical error/primary contributor to clinical error) (a) Establish the leader (13/7) (b) Form the team (19/9) (c) Set team goals (15/4) (d) Assign roles and responsibilities to team members (24/9) (e) Acknowledge team member contribution to performance (4/2) (f) Demonstrate respect in verbal communication (11/4) (g) Hold team members accountable (30/19) (h) Address professional concerns directly with constructive exchanges (17/9) (I) Explore alternatives when time permits (19/9) (j) Achieve acceptable resolution with follow-up discussions if needed (13/7) (k) Defer to leader’s decision when time is critical (4/2) (a) Engage team members in planning process (19/4) (b) Identify established protocol to be used or develop a plan (35/20) (c) Engage team members in decision-making process (20/7) (d) Alert team to potential biases and errors (30/4) (e) Report slips, lapses, and mistakes to team (19/15) (f) Advocate and assert a position or corrective action (31/28) (g) Apply the 2-challenge rule (22/17) (a) Use common ED terminology in communications (2/0) (b) Call out request for information input (7/4) (c) Use check back process to verify communication (22/11) (d) Systematically hand off responsibilities during team transitions (19/9) (e) Offer information to support decision making (20/9) (f) Communicate decision made to team members (19/7) (g) Offer information to support planning (19/6) (h) Communicate plans to team members (22/9) (i) Seek information for decision making (28/17) (j) Seek information for planning (9/4) (a) Execute protocol or team-established plan (30/19) (b) Resolve deviations from protocol or team-established plan (20/13) (c) Integrate individual assessments of patient needs (31/17) (d) Re-plan patient care in response to overall caseload of team (17/11) (e) Prioritize tasks for a patient (31/22) (f) Prioritize all tasks for all patients belonging to the team (9/4) (g) Balance workload within the team (7/4) (h) Request help with task overload (9/4) (i) Offer help for task overload (7/4) (j) Constructively use periods of low workload (0/0) (k) Cross-monitor actions of team members (52/35) (l) Redirect team focus back to immediate team and clinical tasks as necessary (6/4) (m) Request situation awareness updates from team members (9/4) (n) Provide situation awareness updates to team members (9/6) (o) Monitor execution of protocol or team-established plan (20/13) (a) Engage in situational learning or coaching (0/0) (b) Conduct event or shift reviews of teamwork (13/0) (c) Explain actions previously taken without explanation because of original incident urgency (4/0) (d) Include teamwork considerations in educational forums (9/0) (e) Review teamwork in clinical case reviews (9/0)

Teamwork actions in bold type were judged to be a contributing factor to error in 20% or more of cases reviewed. Teamwork actions implicated as a primary contributor to error in 20% or more of these cases are in bold type and italics.

3 7 6

ANNALS OF EMERGENCY MEDICINE

34:3 SEPTEMBER 1999

TEAMWORK AND ERROR REDUCTION Risser et al

the real-time demands of immediate patient care responsibilities, (2) responsibilities to monitor other caregivers’ actions in real time to help ensure proper care delivery, and (3) a responsibility to maintain accurate situation awareness to help ensure sound care delivery in the near future. Situation awareness20 refers to an individual’s level of awareness of important care-related information and events that are present in the immediate ED working environment. Situation data are tracked because they

provide interim measures related to current system performance that can be used to make future care decisions that influence ultimate patient outcome. In emergency care teams, this means an awareness of patient status, care plans, and plan status for patients assigned to the team and awareness of the workload of fellow team members. Team members must realize that 1 of the consequences of this 3-pronged balancing act is that as a caregiver becomes overloaded with demands from

Figure 1.

Interrelationships of the 5 team dimensions.

SEPTEMBER 1999

34:3

ANNALS OF EMERGENCY MEDICINE

3 7 7

TEAMWORK AND ERROR REDUCTION Risser et al

direct patient care responsibilities (1 above), ability to meet core team responsibilities of cross-monitoring (2) and maintaining situation awareness (3) is undermined. An overloaded team member is more likely to err and increases the likelihood that other team member’s errors will go unrecognized, unchallenged, and uncorrected. Team dimension 2, “apply problem-solving strategies,” emphasizes the importance of engaging core team members in planning and decisionmaking, in particular ensuring that all relevant patient information has been provided to the decisionmaker. It also emphasizes the importance of clear identification of the protocol or care plan to be used. A common understanding of the protocol or plan by all team members improves the likelihood of effective, error-free care. Caregivers on a core team are trained to ask questions and be respectfully assertive on behalf of a team’s patient whenever a patient appears to be at risk. This dimension also introduces 3 basic types of errors that all caregivers are prone to make and teaches them to take corrective actions when they recognize them11: 1. Slips—failures to properly adjust well-practiced tasks that require little conscious attention to the characteristics of a new situation (eg, without thinking, ordering the adult dose of a medication for a pediatric patient). The slip is the common failure of the seasoned expert. 2. Lapses—failures of memory that cause tasks not to be done (eg, forgetting which ankle is to be radiographed after leaving the bedside). The lapse is a common error for caregivers facing task overload or distraction. 3. Mistakes—the selection of incorrect actions caused by misclassifying a situation or failing to take into account all relevant factors in a decision (eg, evaluating a patient for nausea, vomiting, and dehydration but failing to recognize new-onset diabetes as a likely cause). A mistake can occur whenever any caregiver’s analytic processing activities are disturbed, disrupted, or missing key information; mistakes can be triggered by many factors, including personal stress, fatigue, task overload, environmental distractions, as well as a lack of clinical knowledge. Perfect execution of poor or inaccurate care plans distinguish mistakes from slips and lapses. Team dimension 3, “communicate with the team,” focuses on communication activities, a recurring problem area in clinical care,21,22 that help team members establish and maintain a common understanding of patient and operational issues affecting team performance. The objectives are to ensure timely and accurate information transfer and to maintain a common situation awareness so team members can effectively coordinate actions and

3 7 8

recognize pending errors. The dimension establishes specific communication actions (Table) to accomplish these objectives. Team dimension 4, “execute plans and manage workload,” focuses on eliminating work overload and weak integration of care delivery. Individual team members learn to help each other with tasks. Task assistance is a risk avoidance activity that reduces the potential for clinical errors that stem from stress, fatigue, distractions, limited skills, or individual overload. The concept is simple: help others and ask for help. This dimension also ensures that various caregiver assessments for an individual patient are integrated into a single patient care plan and clinical tasks are properly prioritized. The objective is to avoid care delivery circumstances that increase the risk of error. The fifth team dimension, “improve team skills,” focuses on improving teamwork skills23 through team review meetings and situation-specific teaching conducted during real-time patient care activities. In team dimension 5, shift reviews, held near the end of the shift, examine the experiences of the team during the shift, discuss the teamwork and care implications, and provide feedback to improve teamwork and clinical skills. Establishing new habits requires regular practice and frequent feedback. Team dimension 5 also seeks to leverage educational forums, like morbidity and mortality conferences, and clinical case reviews to provide examples of teamwork failures to emergency caregivers. D A I LY T E A M M E M B E R A C T I V I T I E S

Team members coordinate directly and repeatedly with each other to ensure proper and timely clinical task execution and to assist overloaded teammates. Each team member works to maintain a clear and accurate understanding (a common situation awareness) of the care status and care plan for each patient assigned to the team and the workload status of each team member. The team oversees and directly manages the use of all other care resources needed by the patients assigned to the team. Figure 2 shows the relationship of the core team to these other commonly used resources. The team oversees all care actions involving the team’s patients. The team must always be provided with a clear explanation of any patient care action taken by a nonteam member because the team is ultimately responsible for the patient. Team members may not leave the team without first making appropriate hand-offs of their responsibilities to the remaining team members.

ANNALS OF EMERGENCY MEDICINE

34:3 SEPTEMBER 1999

TEAMWORK AND ERROR REDUCTION Risser et al

In most settings an additional structural piece, a coordinating team, is needed to manage the overall ED operations. This coordinating team is normally composed of the members shown in Figure 2. In large facilities, the coordinating team may oversee 2 or 3 core teams. The coordinating team decides which patients will be assigned to each team and when a team is in need of additional resources. The coordinating team’s goal is to distribute workload and resources so that no core team is overloaded. In contrast to this, the core team’s goal is to internally manage the team’s assigned responsibilities and resources so that (1) no individual team member is overloaded, and (2) all patients assigned to the team move efficiently through the system while receiving respectful, error-free, quality care. For small facilities there is usually only one core team, and the coordinating team and the core team are one and the same. There are 2 basic classes of teamwork activity: team meetings and individual teamwork actions. Core teams hold brief 2- to 5-minute team meetings to organize and learn. The issues addressed in team meetings include identifying team members and leaders, establishing/

reestablishing situation awareness, assigning/reassigning responsibilities and tasks, making team decisions, discussing problems, and reviewing lessons learned. Because it is often difficult to get everyone together for a meeting, most core teams hold 1 organizing meeting at the beginning of the shift and a review meeting near the end of the shift. When great chaos arises for a team, however, calling a 1- to 3-minute team meeting to quickly assess the situation and reorganize is extremely important. The other side of teamwork, individual teamwork action, is actually the most common teamwork activity. Individual teamwork is caregivers operating as single team members observing or briefly connecting with 1 other team member. Common individual teamwork actions include the following: 1. Caregiver A helps caregiver B with a clinical task or a nonclinical task. 2. Caregiver A observes caregiver B’s actions (crossmonitors) but notes nothing unusual, and therefore no direct interaction occurs. 3. Caregiver A observes caregiver B’s actions (crossmonitors), notes an unexpected care action by B, and acts

Figure 2.

Care resources managed by the core team. BLS, Basic life support; ALS, advanced life support.

Facility resources Medical equipment and supplies Drugs and IV fluids Information systems Out-of-hospital resources BLS ALS

Local and regional

Coordinating team Staff attending Charge nurse Triage nurse Clinical nurse specialist Clerk Patient resources Patient Chief complaint information Level of cooperation Patient history Significant others

SEPTEMBER 1999

34:3

ANNALS OF EMERGENCY MEDICINE

ED core team • Clinical skills • Teamwork skills

Technical support Radiology Laboratory Respiratory Phlebotomy Dietary

Departmental support Security Registration Transport Volunteers Clergy

Referral and consultations In-house staff On-call staff External medical center Inhospital nursing (eg, unit nurse)

3 7 9

TEAMWORK AND ERROR REDUCTION Risser et al

or communicates as necessary to aid or adjust B’s actions or understanding or to adjust A’s own understanding. Caregivers A and B could be any 2 members of the team, regardless of professional roles. Except when scope of practice issues limit clinical actions, teamwork behaviors are pointedly not constrained by traditional status and hierarchical boundaries; in fact, many teamwork behaviors are designed to help break down traditional barriers (eg, between nurses and physicians) that inhibit communication and cooperation and undermine quality of care. These individual teamwork actions give teamwork a major part of its operational power. Teamwork at the operational level involves an integration of these actions into a recurring cycle24 of monitoring, intervening, and correcting errors or deviations in situational awareness (Figure 3). By independently executing this cyclical mon-

itoring process, each team member significantly contributes to the following: • The maintenance of his or her own situation awareness, • The maintenance of the situation awareness of teammates questioned, • The catching of simple errors (eg, slips, lapses, and mistakes) made by teammates monitored, and • On rare occasions, identifying true best practice conflicts that require significant clinical discussion for resolution. TEAMWORK FAILURES AND THEIR CONSEQUENCES

To better understand the nature of teamwork failures in the ED, a retrospective review of claim files was con-

Figure 3.

The teamwork check cycle.

Monitor

Intervene

Correct

Start 1 Check own situation awareness (SA)

2 Cross-monitor actions of teammate(s)

3 Error/ potential error observed?

NO—continue monitoring

YES—ask question

4 Is it a simple error?

NO

3 8 0

YES

5 Erring teammate quickly recognizes and corrects error and continues

6 Teammates resolve complex error disagreement

ANNALS OF EMERGENCY MEDICINE

34:3 SEPTEMBER 1999

TEAMWORK AND ERROR REDUCTION Risser et al

ducted by the MedTeams project. 15,16,18 The cases examined were a convenience sample drawn from 8 participating hospitals. The hospitals included large teaching hospitals, as well as medium-sized and small community hospitals from both military and civilian environments. Fifty-four cases, 25 risk cases (open claims) from a total population of unknown size and 29 closed malpractice claims from a total population of 68 closed claims, were judged by teamwork-trained physician-nurse pairs to be either preventable or mitigable by better teamwork. Physician-nurse pairs used a 2page teamwork failure checklist constructed from the teamwork actions listed in the Table to assess specific teamwork failures in each of the 54 cases. For closed malpractice claims, indemnity payment data were also collected. Each malpractice incident was reviewed in detail to determine whether each of the 48 teamwork actions was present or absent in the case. Teamwork actions absent in the case were identified as teamwork failures (eg, failure to “assign roles and responsibilities to team members”). Each teamwork failure was judged for degree of contribution to the occurrence of the clinical error in the case. The reviewing pairs classified the impact of each teamwork failure into 1 of 3 categories: 1. Noncontributor—the teamwork failure did not contribute to the error(s). 2. Minor contributor—the teamwork failure contributed to the error(s) in a minor way. 3. Primary contributor—the teamwork failure was a powerful, primary contributor to the error(s). The final question on the checklist asked the physician-nurse pair to make a global judgment regarding whether proper teamwork would have “prevented,” “mitigated,” or “had no impact” on the clinical error that occurred in the case. A reliability analysis demonstrated good correspondence between the rater pairs for the count of teamwork failures per case (r=0.85, P<.001) and for the global judgment of the impact (prevent, mitigate, or no impact) that proper teamwork behavior would have had on clinical error avoidance (τb=0.61, P=.006). The percentage of cases in which the caregivers failed to produce the teamwork action and that failure contributed to the clinical error is shown immediately after each teamwork action in the Table. The first number is the percentage of cases in which that teamwork failure was judged to be either a minor or primary contributor to the clinical error(s); the second number is the percent of cases in which the teamwork failure was judged a primary contributor to the clinical error(s).

SEPTEMBER 1999

34:3

ANNALS OF EMERGENCY MEDICINE

All teamwork actions except 2 were listed at least once as teamwork failures that contributed to clinical errors. The average number of teamwork failures per case was 8.8 (ranging from 1 to 32) of a possible maximum of 48. From the individual teamwork action perspective, slightly more than one third (17) of the teamwork actions in the matrix were implicated in a teamwork failure (minor or primary contributor) rate of 20% or more (see actions in bold in the Table). In 4 of the teamwork actions within this set of 17 (shown as bold and italics), the reviewing pairs judged the teamwork failure to be a primary contributor to the error in 20% or more of the cases. The implication is that these 4 individual teamwork behaviors, when properly used by the team, are powerful tools for avoiding serious errors and breaking error chains. • Identify established protocol to be used or develop a plan. It must be clear to everyone on the team what protocol or plan is being used. • Advocate and assert a position or corrective action. The caregiver must speak up when he or she believes the patient is at risk. Organizational leaders must create an organizational climate where this is possible. • Prioritize tasks for a patient. Caregivers must understand the care plan and prioritize tasks accordingly. • Cross-monitor actions of team members. Caregivers must watch each other’s behaviors for simple errors and act to correct these errors. Organizational leaders must create an organizational climate where cross-monitoring is an acceptable practice. Note that the single teamwork failure most frequently cited as a primary contributor to the occurrence of clinical error was cross-monitoring (35% of cases). Such failures occur all too frequently because in our culture it is generally unacceptable behavior for one caregiver to check another caregiver’s actions. Perhaps the most alarming finding was that 8 of the 12 deaths reviewed were judged to be preventable if appropriate teamwork action had been taken. Moreover, 5 of the 8 major permanent impairments (eg, significant heart damage, loss of a limb, or loss of ability to manage daily living activities) were judged to be preventable if appropriate teamwork actions had been taken.15,18 By using cost data from the closed claims only, it was estimated that legal costs of $16 million were avoidable. Improved ED teamwork, on average, would save $560,479 per closed case for the 43% of closed cases where teamwork can influence the outcome (ie, prevent or mitigate the error). Examined from another perspective, this is $345,460 in savings per 100,000 ED patient

3 8 1

TEAMWORK AND ERROR REDUCTION Risser et al

visits, or nearly $3.50 for every patient seen in the ED.15,16 Although all costs presented here are real costs paid in actual closed cases, the numbers provided are still a significant understatement of the total cost of error. First, risk management and malpractice cases are only a small fraction of the total number of iatrogenic injuries.7-9,25 In addition, the direct costs noted here do not reflect the wider staff and community heartaches, frustrations, and costs associated with these cases. Third, some risk management groups participating in the retrospective review were only willing to report indemnity payments that exceeded $1 million as $1 million. Also, in 2 cases with settlements of more than $1 million each, the risk managers were unwilling to permit inclusion of the cases in the study; that is, the number of cases preventable or mitigable should have been 56, not 54. Finally, most of the participating hospitals had only limited ability to identify legal defense costs. In summary, teamwork training for high-performance, high-stress teams has been shown to enhance mission effectiveness and improve the margin of safety in aviation. A similarly structured, behavior-based teamwork system can play an important role in learning to effectively manage system and caregiver imperfection in the ED environment. All caregivers, including physicians, need to openly acknowledge their human limitations and embrace the idea of help from others in managing system and human fallibility. Note, however, that effective teamwork behavior does not arise spontaneously; it must be learned and implemented through specific training and practice.3-5 The potential benefits of better teamwork are significant. More than half of the deaths that occurred in the malpractice cases were judged to be avoidable under conditions of better teamwork. Teamwork skills have significant potential to prevent and mitigate the effect of clinical errors in the ED setting. A number of teamwork behaviors emerged as potentially very valuable in breaking error chains and stopping errors. The retrospective claims review suggests that improvements in medical teamwork can significantly enhance the quality of emergency care and reduce future costs. Moreover, it seems likely that this type of teamwork system can positively influence other intense, high-stress units as well (ie, ICUs, labor and delivery, and operating rooms). The MedTeams Project is nearing the end of a multiyear, multisite validation experiment to assess the efficacy of aviation-like teamwork training in EDs. The initial results are promising. The full results will be reported in the next 12 to 18 months.

3 8 2

MedTeams Research Consortium: Gary Adamowicz, BS; Steven L Banks, DO; MAJ Tammie Chang, RN, MSN, CEN; MAJ James Cleveland, RN; CPT Robin Cody, RN, EMT-B, CEN; Teresa Czaplinski, RN, BSN; Amy Guilfoil-Dumont, RN; James Evangelista, RN, RPh,CEN; COL Daniel Fitzpatrick, DO, MPH; Nancy Gates; Marjorie Geist, RN, MSN, MHA, PhD; Lori A Hughes, RN, MS, CEN; Bruce Janiak, MD; Jorie Klein, RN; Vinette Langford, RN-CS, MSN, CEN; MAJ Constance Lavieri-Reynolds, MD; LTC Thomas Lenz, MD; Ann Locke, RN-CS, MSN; Sandra McDonald, RN, BSN; CDR Timothy McGuirk, DO, FACEP; John C Morey, PhD, CHFP; Todd Murray, MD; Dallas E Peak, MD, FACEP; Shawna J Perry, MD; LCDR James R Pierce, RN, MSN, CEN; MAJ Laura Rodgers, RN, MS; William D Rose, MD, FACEP; Harry Swiger, RN, BA; Carla G Tolbert, MS, RN, CCRN, CEN; LTC Clyde Turner, DO, MPH; Robert L Wears, MD, MS, FACEP; Kenneth A Williams, MD, FACEP; Charlotte S Yeh, MD, FACEP.

REFERENCES 1. Helmreich R: Managing human error in aviation. Sci Am 1997;5:62-67. 2. Grubb G, Leedom D, Simon R, et al: Effects of Crew Coordination Training and Evaluation Methods on AH-64 Attack Helicopter Battalion Crew Performance. Fort Rucker, AL: US Army Research Institute, Aviation Research and Development Activity, 1993:14-17. 3. Leedom D, Simon R: US Army crew coordination training and evaluation, in Jensen RS (ed): Proceedings of the Seventh International Symposium on Aviation Psychology. Columbus, OH: The Ohio State University, 1993. 4. Gayman A, Gentner F, Crissy M, et al: Implications of crew resource management training for tank crews, in Proceedings of Interservice/Industry Training System and Education Conference. Orlando, FL: I/ITSEC, 1996:1-10. 5. Leedom D, Simon R: Improving team coordination: A case for behavior-based training. Mil Psych 1995;7:109-122. 6. Leape L. Error in medicine. JAMA 1994;272:1851-1857. 7. Leape LL: Keynote address for Proceedings of Examining Errors, in Health Care: Developing a Prevention, Education, and Research Agenda. Rancho Mirage, CA: Annenburg Center for Health Sciences, October 13-15, 1996. 8. Brennan TA, Leape LL, Laird NM, et al: The nature of adverse events in hospitalized patients: Results of the Harvard medical practice study I. N Engl J Med 1991;324:370-376. 9. Leape LL, Brennan TA, Laird NM, et al: The nature of adverse events in hospitalized patients: Results of the Harvard medical practice study II. N Engl J Med 1991;324:377-384. 10. Reason J, Mycielska K: Absent-minded? The Psychology of Mental Lapses and Everyday Errors? Englewood Cliffs, NJ: Prentice Hall, 1982. 11. Reason J: Human error. New York: Cambridge University Press, 1990. 12. Koss L: Cervical (Pap) smear: New directions. Cancer Suppl 1993;71:1406-1412. 13. Bogner S: Human Error in Medicine. Hillsdale, NJ: Lawrence Erlbaum Associates, 1994. 14. Joint Commission on Accreditation of Healthcare Organizations, in Sentinel Events: Evaluating Cause and Planning Improvement. Oakbrook, IL: Joint Commission on Accreditation of Healthcare Organizations, 1998. 15. Simon R, Morey J, Rice M, et al: Reducing errors in emergency medicine through team performance: The MedTeams Project, in Proceedings of Enhancing Patient Safety and Reducing Errors in Healthcare. Rancho Mirage, CA: Annenburg Center for Health Sciences, Nov 8-10, 1998. 16. Jay GD, Risser DT, Rice M, et al: The potential for improved teamwork to reduce medical error in the emergency department [abstract]. Acad Emerg Med 1998;5:526. 17. Phillips K: The Power of Health Care Teams: Strategies for Success. Oakbrook, IL: Joint Commission on Accreditation of Health Care Organizations, 1997. 18. Simon R, Morey J, Locke A, et al: Full Scale Development of Emergency Team Coordination Course (ETCC) and Evaluation Measures. Andover, MA: Dynamics Research Corporation, 1997. 19. Cannon-Bowers J, Oser R, Flanagan D: Work teams in industry: A selected review and proposed framework, in Swezey R, Salas E (eds): Teams: Their Training and Performance. Norwood, NJ: Ablex Publishing, 1992:355-377. 20. Endsley M: Toward a theory of situation awareness in dynamic systems. Hum Factors 1995;37:32-64.

ANNALS OF EMERGENCY MEDICINE

34:3 SEPTEMBER 1999

TEAMWORK AND ERROR REDUCTION Risser et al

21. Alt-White A, Charns M, Strayer R: Personal, organizational, and managerial factors related to nurse-physician collaboration. Nurs Admin Quart 1983;8:8-18. 22. Donchin Y, Gopher D, Olin M, et al: A look into the nature and causes of human errors in the intensive care unit. Crit Care Med 1995;23:294-300. 23. Berwick DM: Continuous improvement as an ideal in health care. N Engl J Med 1989;320:53-56. 24. Locke A, Langford V, Risser D, et al: Instructor Guide for Emergency Team Coordination Course (ETCC). Andover, MA: Dynamics Research Corporation, 1997. 25. Vincent C, Taylor-Adams S, Stanhope N: Framework for analyzing risk and safety in clinical medicine. BMJ 1998;316:1154-1157.

SEPTEMBER 1999

34:3

ANNALS OF EMERGENCY MEDICINE

3 8 3