Patient Safety in the Cardiac Operating Room: Human Factors and Teamwork

HomeCirculationVol. 128, No. 10Patient Safety in the Cardiac Operating Room: Human Factors and Teamwork Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissionsDownload Articles + Supplements ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toSupplemental MaterialFree AccessResearch ArticlePDF/EPUBPatient Safety in the Cardiac Operating Room: Human Factors and TeamworkA Scientific Statement From the American Heart Association Joyce A. Wahr, MD, FAHA, Co-Chair, Richard L. Prager, MD, FAHA, J.H. AbernathyIII, MD, Elizabeth A. Martinez, MD, Eduardo Salas, PhD, Patricia C. Seifert, MSN, Robert C. Groom, CCP, Bruce D. Spiess, MD, FAHA, Bruce E. Searles, MS, CCP, Thoralf M. SundtIII, MD, Juan A. Sanchez, MD, Scott A. Shappell, PhD, Michael H. Culig, MD, Elizabeth H. Lazzara, PhD, David C. Fitzgerald, CCP, FAHA, Vinod H. Thourani, MD, Pirooz Eghtesady, MD, PhD, FAHA, John S. Ikonomidis, MD, PhD, FAHA, Michael R. England, MD, Frank W. Sellke, MD, FAHA and Nancy A. Nussmeier, MD, FAHA, Co-Chairon behalf of the American Heart Association Council on Cardiovascular Surgery and Anesthesia, Council on Cardiovascular and Stroke Nursing, and Council on Quality of Care and Outcomes Research Joyce A. WahrJoyce A. Wahr Search for more papers by this author , Richard L. PragerRichard L. Prager Search for more papers by this author , J.H. AbernathyIIIJ.H. AbernathyIII Search for more papers by this author , Elizabeth A. MartinezElizabeth A. Martinez Search for more papers by this author , Eduardo SalasEduardo Salas Search for more papers by this author , Patricia C. SeifertPatricia C. Seifert Search for more papers by this author , Robert C. GroomRobert C. Groom Search for more papers by this author , Bruce D. SpiessBruce D. Spiess Search for more papers by this author , Bruce E. SearlesBruce E. Searles Search for more papers by this author , Thoralf M. SundtIIIThoralf M. SundtIII Search for more papers by this author , Juan A. SanchezJuan A. Sanchez Search for more papers by this author , Scott A. ShappellScott A. Shappell Search for more papers by this author , Michael H. CuligMichael H. Culig Search for more papers by this author , Elizabeth H. LazzaraElizabeth H. Lazzara Search for more papers by this author , David C. FitzgeraldDavid C. Fitzgerald Search for more papers by this author , Vinod H. ThouraniVinod H. Thourani Search for more papers by this author , Pirooz EghtesadyPirooz Eghtesady Search for more papers by this author , John S. IkonomidisJohn S. Ikonomidis Search for more papers by this author , Michael R. EnglandMichael R. England Search for more papers by this author , Frank W. SellkeFrank W. Sellke Search for more papers by this author and Nancy A. NussmeierNancy A. Nussmeier Search for more papers by this author and on behalf of the American Heart Association Council on Cardiovascular Surgery and Anesthesia, Council on Cardiovascular and Stroke Nursing, and Council on Quality of Care and Outcomes Research Originally published5 Aug 2013https://doi.org/10.1161/CIR.0b013e3182a38efaCirculation. 2013;128:1139–1169Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2013: Previous Version 1 IntroductionThe cardiac surgical operating room (OR) is a complex environment in which highly trained subspecialists interact with each other using sophisticated equipment to care for patients with severe cardiac disease and significant comorbidities. Thousands of patient lives have been saved or significantly improved with the advent of modern cardiac surgery. Indeed, both mortality and morbidity for coronary artery bypass surgery have decreased during the past decade (Figure 1).1 Nonetheless, the highly skilled and dedicated personnel in cardiac ORs are human and will make errors. In 1991, Leape and colleagues2,3 estimated that among the 2 million patients hospitalized in New York in 1984, there were 27 179 adverse events that involved negligence; other evidence suggests that up to 16% of hospital inpatients are harmed.4 Gawande and associates5 found that the incidence of surgical adverse events was 12% among cardiac surgery patients versus 3% in other surgical patients; 54% of the adverse events were considered preventable. Of the roughly 350 000 to 500 000 patients who undergo cardiac surgery each year, 28 000 will have an adverse event, and one third of deaths associated with coronary artery bypass graft (CABG) operations may be preventable.6Download figureDownload PowerPointFigure 1. Change in mortality and stroke rates in patients undergoing isolated coronary artery bypass graft (CABG) surgery, 2000 to 2009. There was a 24.4% and 26.4% reduction in the unadjusted observed operative mortality (2.4% vs 1.9%) and stroke rates (1.6% vs 1.2%), respectively, during the course of the study period. Reprinted from ElBardissi et al1 with permission from Elsevier. Copyright © 2012, The American Association for Thoracic Surgery.Refined techniques, advanced technologies, and enhanced coordination of care have led to significant improvements in cardiac surgery outcomes. However, more than 10 years after the Institute of Medicine report,7 there is little evidence that much progress has been achieved in reducing or preventing errors.8 The tools to measure potential risks and interventions to improve patient safety are still in the early stages of development and testing,9 and funding for patient safety studies remains inadequate. Published studies provide only limited evidence of improved outcomes.8,9 Furthermore, much of the existing research is, by necessity, qualitative and descriptive and thus does not lend itself to traditional quantitative statistical analysis. Therefore, many clinicians are not conversant with such research.Preventable errors are often not related to failure of technical skill, training, or knowledge but represent cognitive, system, or teamwork failures (Figure 2).10–14 Nontechnical skills such as communication, cooperation, coordination, and leadership are critical components of teamwork, but limited interpersonal skills often underlie adverse events and errors.15–17 In a review of litigated surgical outcomes, communication failures accounted for 87% of the system failures that led to an indemnity payment.18 The communication failures occurred primarily between caregivers, rather than between caregiver and patient.Download figureDownload PowerPointFigure 2. Accident model. Active and latent failures in healthcare organizations, hospital management, and individual human error can all contribute to adverse events during high-risk procedures. Reprinted from Carthey et al13 with permission from Elsevier. Copyright © 2001, The Society of Thoracic Surgeons.Breakdowns in teamwork that lead to surgical flow or operative disruptions are exceedingly common, having been noted at a rate of 17.4 per hour in one cardiac surgery study19 and at 11 per case in another.20 Importantly, such disruptions add up, leading to technical errors and adverse patient outcomes.21–23 The majority of flow disruptions are related to teamwork failures, and these disruptions have been shown to be strongly predictive of surgical errors.20Even minor events in cardiac surgical procedures, that is, those not expected to affect outcome, reduce the team’s ability to recover from major events and appear significantly associated with both death and near misses.22 In one study, for every 3 minor problems above the mean of 9.9 per case, intraoperative performance was measurably reduced and operative duration increased.23 The accumulation of minor disruptions and events apparently reduced the ability of the cardiac team to compensate for major errors24; in short, “little things matter.”17,25Surgical team members vary in their awareness of their own and their colleagues’ teamwork skills. In multiple studies, self-assessment of communication and teamwork skills by surgeons and anesthesiologists is disturbingly discordant with the opinions of their associated nursing and perfusion staff.26,27 Surgeons rated the teamwork of other surgeons as high/very high 85% of the time, but nurses rated their collaboration with surgeons as high/very high only 48% of the time.28 Objective assessment of teamwork skill reveals differences between skill level of team members and can indicate opportunity for education and training.29The present scientific statement includes data regarding many teamwork skills but focuses on communication. Communication failures were the leading root cause of 65% of sentinel events reported by The Joint Commission between 2004 and 2012 and were a leading contributor to errors in medications, wrong-site procedures, and operative and postoperative events.30 In one cardiac surgery study, teamwork failures occurred frequently (5.4 per case with familiar teams and 15.4 per case with unfamiliar teams); communication issues were the primary cause of these teamwork failures (89%).21The American Heart Association commissioned this scientific statement to summarize the evidence regarding risks to patient safety and clarify interventions to reduce perioperative risks and human error in cardiac surgery. A comprehensive review of all potential risks to patient safety and tested interventions would be voluminous and could include wide-ranging topics such as surgical techniques (mammary arteries in CABG surgery), various cardiopulmonary bypass (CPB) strategies, or techniques to reduce infection or retained objects. We have chosen to focus primarily on those human, environmental, and cultural factors that affect teamwork, particularly how cardiac surgery teams communicate within the OR and with other unit teams. The statement is organized to describe current knowledge about communication within and between teams, the physical work environment and how it influences teamwork (space, equipment, and ergonomics), and the organizational culture (safety climate and quality improvement [QI]) of the cardiac OR.Our process was to focus on studies in the cardiac surgical environment regarding teamwork, but we did draw on other literature as needed to present critical concepts that were specifically lacking in the cardiac surgical literature. Although many cardiac surgery studies identify communication as a significant source of error, discussion of the concepts that underlie effective or defective communication are found primarily in the cognitive psychology literature, and we have included these references in the “Communication and Teamwork” section. Similarly, although our focus is on cardiac surgery, we have included pertinent data from other surgical disciplines. We have attempted to identify the references specific to cardiac surgery, but the reader is encouraged to consult individual references for further information. Because of our focus, we excluded many dynamic areas of research that we hope will be summarized in other scientific statements or similar reviews. Finally, the present scientific statement aims to identify major knowledge gaps and potential areas for further research.The present statement was coauthored by a writing committee composed of members of the American Heart Association’s Council on Cardiovascular Surgery and Anesthesia, as well as collaborating members of the following nonprofit organizations: the Society of Cardiovascular Anesthesiologists and its FOCUS (Flawless Operative Cardiovascular Unified Systems) initiative (Society of Cardiovascular Anesthesiologists Foundation), the Society of Thoracic Surgeons, the Association of periOperative Registered Nurses, the Human Factors and Ergonomics Society, and the American Society of Extracorporeal Technology. We hope that these data and recommendations will motivate further research to address the challenges of reducing human error and improving patient safety in the cardiac OR. Such research should be widely applicable to all ORs, as well as to interventional cardiology and electrophysiology procedural settings. In particular, we hope that the present scientific statement will encourage similar reviews of patient safety in cardiology catheterization and electrophysiology laboratories, as well as in other interventional settings such as hybrid ORs designed for percutaneous management of valvular lesions, percutaneous assist devices, or stenting of aortic aneurysms.Assessing Patient SafetyTo understand how to improve patient safety, we must understand how researchers have assessed nontechnical skills and their impact. To begin with, we need a common vocabulary; terms for nontechnical skills must be defined to promote reliable comparison of studies and discussion. Second, the effect of specific nontechnical skills on the reduction of human error or on patient safety must be quantified. Third, interventions to improve individual and team nontechnical skills must be designed and tested for efficacy. Fourth, the effect of improved nontechnical skill(s) on error reduction and, hopefully, ultimately on patient outcomes must be studied to demonstrate progress.31Technical skills can be measured objectively (eg, knots tied per minute), but nontechnical skills assessment requires observational and often seemingly subjective assessment by experts. Observational research, although new to many clinicians, has already identified the number, type, and severity of adverse events that occur in the OR.13 Many team and individual behaviors that are precursors of adverse events, as well as the behaviors associated with surgical excellence, have been identified.12,32 Observational research, however, has limitations: Valid results require trained observers, and not all trainees will become expert.13,32,33 In one study, only 32% of all recorded events were captured by both observers, although events that were captured by both were rated equivalently.34Teaching nontechnical skills is particularly challenging given the difficulty in assessing performance and providing feedback. Appropriate attention is paid to assessing the quality of technical skills, but nontechnical skills also require assessment for competency and to identify opportunities for education. As noted, observational assessment of nontechnical skills requires trained and experienced observers; to date, use of trained observers has primarily been applied in research, not in training or certification of clinical competence. During surgical simulations, a strong correlation is found between the expert’s assessment and the resident surgeon’s self-assessment of technical skills, but the same is not true for nontechnical skills.35 Senior surgeons’ self-assessments of technical skills highly correlate with that of an observer, but both junior and senior physician surgical trainees (resident and fellows), as well as surgical faculty, all rated themselves higher on their nontechnical skill level than did the expert observers.36Objective observers are also necessary to accurately assess disruptions, errors, communication skills, and the impact of these factors on outcome. Unlike trained observers, OR personnel judged disruptions to affect their colleagues more than themselves; surgeons perceived fewer team disruptions than did other OR team members.37 Nontechnical skills may need to be explicitly taught, because senior surgeons may or may not demonstrate better teamwork skills than those more junior, particularly in simulated crisis scenarios.35,36,38Teamwork MeasuresMany nontechnical skill measurement tools have been used (Table 1), but there is no single accepted instrument. Many are designed to measure nontechnical skills within a specific subteam (nurses, surgeons, anesthesiologists).49 Behavior rating systems must be valid (measure what they purport to measure), reliable (have good intraobserver and interobserver correlation), sensitive (detect differences in behaviors when they exist), and feasible (be easy to implement and be cost-effective).Table 1. Teamwork Assessment ToolsTools to Assess Teamwork Skills Within TeamDefinitionOTAS29,33,39–44Procedural task checklist centered on patient, equipment, and communications tasks ratings• Communication• Cooperation• Coordination• Shared leadership• Shared monitoringNOTECHS15,45–48Adapted from the aviation NOTECHS scale used in Europe• Cooperation/teamwork• Leadership/management• Situational awareness• Problem solving/decision making• ± Communication/interactionNOTECHS indicates Oxford Non-Technical Skills; and OTAS, Observational Teamwork Assessment for Surgery.Five measurement tools, each with its own strengths and weaknesses, have been designed for surgical team and subteam skills49: the Observational Teamwork Assessment for Surgery (OTAS),29,33,39–44,49 the Oxford Non-technical Skills (NOTECHS),15,45–48 the Non-Technical Skills in Surgery (NOTSS),50–52 the Anesthesia Non-Technical Skills (ANTS),53,54 and the Scrub Practitioners’ Non-technical Skills (SPLINTS).54a,54b Of these 5, NOTSS, ANTS, and SPLINT are designed to assess the individual nontechnical skills of surgeons, anesthesiologists, and scrub practitioners respectively, whereas OTAS and NOTECHS are specifically designed to assess team behaviors and skills.55 The OTAS includes a task checklist and a team behaviors assessment. It has good construct validity (ie, it actually measures what it appears to measure) and strong reliability between expert observers but weak reliability between expert and novice observers, which indicates that training of observers is required.41 The surgical NOTECHS was directly adapted from an aviation NOTECHS scale45 and measures skills in 4 domains (cooperation/teamwork, leadership/management, situational awareness/vigilance, and problem solving/decision making); some research teams have added communication/team skills.48 The NOTECHS has good reliability between expert and novice observers, has been used to show improvement in nontechnical skills after training, and has been used to show a significant inverse correlation between technical errors and nontechnical score.15,47 There is good correlation between the NOTECHS and OTAS scores when used in parallel47; both the OTAS and the modified NOTECHS have been found to be construct valid.47,56Surgical flow disruptions are correlated with adverse events in several studies but are defined differently in each study.20,37,57 Two tools have been proposed, namely, the Surgical Flow Disruption Tool (SFDT)57 and the Disruptions in Surgery Index (DiSI).37 Both have strong interrater reliability but have not been tested by other researchers.Outcome MeasuresPoor teamwork and poor nontechnical skills have been shown to adversely affect patient outcomes. Morbidity and mortality are associated with system failures,18 failures of coordination and communication,58 reported levels of communication,59 poor teamwork behaviors,12 unfamiliarity among cardiac surgical team members,21,60 and the number of minor events (disruptions) per case.22 Other studies have linked teamwork quality and behaviors to surrogates such as increased length of operation,23 number of technical errors in an operation,46 number of major errors,61 and stress levels of team members.62The ultimate desired outcome for any safety intervention is reduction in morbidity and mortality. Mortality in cardiac surgery is quite rare; thus, studies have to be very large to achieve adequate power to discern improvement in this measure. Neily and colleagues63 demonstrated a significant reduction in mortality with teamwork training but included 189 000 procedures at 108 Veterans Affairs hospitals to reveal a treatment effect.Because the safety climate of an institution correlates with communication errors, several studies have used changes in attitude toward safety or changes in team “emotional climate” as a surrogate of outcome to measure impact; these studies show training in nontechnical skills to be effective.64–70SummaryThe nontechnical skills of individuals and teams affect patient safety.OTAS and NOTECHS have proven construct validity and reliability. Training of observers who use these instruments is strongly recommended for accurate results.Proposed interventions to improve nontechnical skills should be tested for their efficacy in improving skill before being implemented.Communication and TeamworkCommunication Within TeamsCommunicationCommunication is “the exchange of information between a sender and a receiver.”71 In the OR, multiple individuals communicate simultaneously. Unfortunately, communication skill has been measured as the worst of 5 aspects of teamwork behavior in the OR29; deficits in patient safety are frequently a product of breakdowns or delays in communication.72,73 Miscommunication can occur when the sender inaccurately encodes a message (eg, by using vague or incomplete language), when the receiver decodes the sent information incorrectly, or when the information is given at the wrong time or received by the wrong individual.72 Communication failures are common72,74,75 and were the most common cause of problems in a host of studies.16,21–23,58,76 Miscommunication has been implicated as the root cause of error and adverse outcomes in both general and cardiac surgery.13,18,20–22,59,77–80 It is worse when teams are unfamiliar with each other.21Communication failures in the OR are equally related to timing, content (erroneous or missing data), purpose, and audience (directed to or received by the wrong person).72 Effective communication is open, adaptable, accurate, and concise, and it is more likely to occur in supportive and safe climates.71 Open communication fosters seamless coordinated activities81; adaptable communication shows that team members are aware of and adapt to others’ workloads, and concise communication promotes efficiency.82The connection between effective communication and improved team performance/outcome has been shown in cockpit crews,83 navy teams,84 and surgical teams.81 A recent meta-analysis provided definitive evidence of the criticality of information sharing for effective team performance.85 Systematic literature reviews indicate that communication is a key feature of successful teams86 and is essential for high-quality patient care.87 Good communication enables and facilitates other fundamental team processes and states, such as coordination, cooperation, cognition, coaching, and conflict resolution.88CooperationCooperation is a critical element of teamwork as well and captures the feelings, attitudes, and beliefs that drive behavior. Attitudinal components began to be studied after several tragic aviation accidents were attributed to teamwork failures. Recognizing that the lack of teamwork skills (previously considered “nonessential”) created severe consequences, the aviation industry developed and implemented CRM (ie, cockpit or crew resource management) programs to improve teamwork.89Some of the most studied attitudes include collective efficacy (a collective sense of competence),90,91 team orientation (a preference for and belief in teamwork),92,93 cohesion (a commitment to the team, its task, or both),94,95 and mutual trust (a shared belief that all will contribute to and protect the team).96,97 Although data from cardiac surgical teams are lacking, other studies of dynamic, complex environments have shown that adaptive performance is critical. Psychological safety, team empowerment (the feeling that team members have the authority to control their work and environment), and safety climate are critical.98–101 Empirical research has shown that when teams have high levels of collective efficacy, members exert more effort and take more strategic risks, which leads to better performance and higher satisfaction.102,103 The level of trust within a team affects how much members monitor each other, how committed team members are to the organization, and performance.104–111CoordinationCommunication also enables the behavioral skills necessary for optimal coordination and team performance.112 Coordination requires effective communication and is essential for successful team performance. It is, essentially, “orchestrating the sequence and timing of interdependent actions.”113 Coordination can be established explicitly with synchronization and awareness or implicitly with covert sequencing and communication.71Implicit coordination entails a shared understanding of the task, the environment, and individual roles and responsibilities within the team. It allows members to anticipate each other’s actions and needs without explicit communication, which enhances efficiency.114–116A mutual team understanding allows team members to provide assistance, information, and feedback,71 which allows the team to modify structures and processes without detriment in performance.117 The ability to foresee is imperative for effective teamwork and performance, especially in high-stress situations.71 Without coordinated behaviors, team members cannot ensure that actions and tasks are performed in synchrony without wasted effort.112For decades, research in the military and aviation has demonstrated that a team’s mutual understanding facilitates coordination and performance.114,115,120,121 Other studies show that teams with and without external pressures exhibit better performance when they have effective and efficient coordinating behaviors.122,123 Within medical teams, explicitly stating the team’s needs and goals or using team familiarity can build coordination skills and allow team members to develop clear expectations and understanding.71 Training in coordination and adaptation, providing information updates, and distributing responsibilities improves coordinating behaviors.115CognitionCognition is a shared understanding that arises from team interactions,124 which improves with repeated interactions.125 Cognition refers to the team’s collective knowledge about the roles, responsibilities, and capabilities of each member.82 The ability to anticipate team members’ needs enhances coordination and communication.126 A common understanding among team members enhances shared awareness of the surroundings, critical for problem solving in dynamic situations.117 Teams lacking in shared understanding have reduced coordination, which leads to poor performance.125,127Studies of team cognition in aviation and the military, as well as in laboratory studies with students, have shown that experienced teams and teams familiar with one another have better team cognition (eg, shared mental model) and better outcomes than inexperienced teams.21,60,128–131 Shared knowledge affects team behaviors and performance (reviewed by Mathieu et al132). Shared cognition improves team communication,133–136 learning and self-regulation,126,137–140 and coordination.125–127Within the medical domain, reflexivity training (ie, guided reflection of strategies used by the team),131,140 cross-training (ie, training on the tasks and duties of other members),126,141 and simulation-based team training142,143 have been discussed as effective interventions to improve team cognition. Improving the understanding shared among team members enhances coordination and performance.ConflictCommunication is pivotal for conflict resolution. Conflict, defined as discrepancies or incompatibilities among team members,144 can center on tasks, relationships, or processes.145,146 Conflict has been found to occur during the treatment of 50% to 75% of hospitalized patients,147,148 and this may be even greater in the OR, where ostensibly equal physician teams share in the care of a single patient.Conflict can have positive or negative implications.149,150 Task-based conflict improves group performance in the evaluation of nonroutine problems and in group decision making,144 but conflict also results in lower team member satisfaction, commitment,151 cohesion, and effectiveness.145 Unlike task-based conflict, relationship conflict has a profound negative effect on both performance and satisfaction and decreases members’ willingness to remain part of the group.151–153In the OR, conflicts are often poorly managed through avoidance, yielding, or competition, when collaboration and compromise would yield a better outcome.154 Collaboration and compromise are particularly difficult when there is status asymmetry, whereby one member has greater power or seniority, such as physicians with nurses or an attending physician with residents.147,155 Among OR personnel, 73% opined that disagreements in the OR are resolved appropriately, but 29% stated they would have trouble speaking up if they perceived a problem with patient care, and 41% felt unable to express disagreement.156 Behaviors that physicians perceive as decisive and necessary to achieve task goals may be viewed as harsh and demeaning by subordinates.157 Difficulty in seeing one’s own behavior as others see it is pervasive throughout OR and intensive care unit (ICU) teams.158,159 When watching videos of conflict scenarios, surgeons, anesthesiologists, and nurses rated the tension levels similarly but rated their own profession as having relatively less responsibility for creating or resolving the tension.160,161There are well-known approaches to conflict resolution in the literature (eg, the 7-step model, principle-based conflict resolution, advocacy/inquiry).144,146,162,163 Teaching conflict management to OR teams is important and possible.157,163 Effective techniques for conflict resolution are an important component of most team-training methods.63,164CoachingTeam coaching, defined as “direct interaction with a team intended to help members make coordinated and task-appropriate use of their collective resources in accomplishing the team’s work,”165 can be used to