The twentieth century saw dramatic improvements in the quality of health care. Medical advances like the discovery of antibiotics and vaccines improved the treatment and prevention of infectious diseases. Chemotherapy and advances in surgical techniques made cancer a survivable disease. Public health initiative dramatically decreased death from modifiable causes like smoking and motor vehicle accidents. Maternal death rates from both obstetric and anesthesia causes were dramatically reduced. With these advances, however, came a significant increase in the complexity of delivering health care. By the end of the twentieth century and into the beginning of the twenty-first century, concerns about the safety of the healthcare system began to be raised. The system itself became a leading cause of patient harm and death. Hospital acquired infections caused by bacteria made more virulent by widespread use of antibiotics became major causes of patient harm. Clinician error became a leading cause of hospital death. This brought an emphasis on patient safety alongside the established focus on quality of care. The practices of obstetrics and obstetric anesthesia have led the way in this patient safety movement.

Most clinicians have now heard that tens of thousands of patients die each year in the United States due to medical error (1,2,3), that hundreds of thousands more are injured (4), and that the economic cost of these errors runs into the tens of billions of dollars. The recommendations that medical institutions adopt the teamwork concepts of Crew Resource Management (CRM) and use simulation to improve patient safety have been widely made (3,5,6,7). Sadly, these recommendations have not been well implemented, and now more than a decade after the Institute of Medicine (IOM) recommended these changes (8,9), adverse events due to medical error remain common (2,10). Many factors may influence the difficulty in adopting teamwork and simulation into medical practice. Leape suggested that the very culture of medicine, often associated with a “finger-pointing environment,” (11) is not conducive to a teamwork approach (8). A lack of understanding of what CRM is or what is needed to make it succeed almost certainly plays a role (12,13). (A thorough discussion of CRM and its utility in the clinical setting is provided later in this chapter.) Inherent differences between the aviation and medical industries are also likely to inhibit the proliferation of CRM. For instance, cockpit crews frequently are assigned to fly only one type of aircraft and may only fly a small number of flight paths. In contrast, medical staff—especially those in acute care areas like obstetrics—must care for whatever patients arrive on the unit. Machines (airplanes) respond much more consistently to a given set of parameters than do people (patients), who are likely to have their own, idiosyncratic reaction to nearly every treatment. The culture of aviation is very different from that in medicine. Most pilots support a relatively flat hierarchy, which then facilitates open communication and teamwork. Physicians are less supportive of a flat hierarchy (14). Cockpit crews are also more likely than medical staff to understand the negative impact that emotional stress, production pressure, and fatigue can have on performance (15). Finally, while the emotional impact of an adverse event or medical error can be devastating to medical staff (16,17), the caregiver does not directly suffer the impact of the error. The fact that a pilot may die due to his or her mistake adds a compelling reason to take any step necessary to prevent accidents.

The practice of obstetrics poses particular challenges to both the need for and implementation of teamwork and other patient safety practices. Most importantly, it is the only medical environment that can have a 200% mortality rate. Every action that a health care team takes to care for the mother may also impact the safe care of the unborn child. The birth of a baby is an inherently private and personal experience. Families may anticipate the birth experience in the same way they look forward to a wedding, graduation, or other significant life transition. Superimposing a team of previously unknown care providers may not fit these plans. Parturition frequently takes place literally behind closed doors in the privacy of a labor room. This is part of the private nature of the process, but it makes it difficult for team members to monitor the safe conduct of the birth. This is in stark contrast to the cockpit crew who sit within feet of each other and have ready access to outside experts (air traffic control) monitoring their performance. Finally, the physician leaders of the labor and delivery process are very different from pilots. They are frequently not on the unit. An obstetrician may be in his/her office throughout most of a parturient’s labor. The obstetric anesthesiologist may have duties in the main operating room and may come to the labor floor for only a few minutes to place a labor epidural. Teamwork is clearly difficult when much of the team is not physically present.

While estimates of the number of women and babies that die or are harmed due to medical error during obstetric care do not currently exist to the same extent that they do for the general medical population, available evidence does suggest that the patient safety crisis is as problematic in obstetrics and obstetric anesthesia as it is in the rest of the medical community. Labor and delivery is the most common cause for hospitalization in the United States, and cesarean delivery is the most common operation in the United States. Thus, the exposure to medical error is high for the parturient and her fetus. Perhaps 9% of deliveries are associated with a maternal or fetal complication (18). It has been estimated that up to 87% of perinatal adverse events are preventable, with poor teamwork protocol violations and unavailable staff being the common problems (19). Substandard care contributes to approximately 50% of maternal deaths, with poor communication and teamwork being primary factors in the substandard care (20,21,22). Up to 72% of neonatal adverse events can similarly be attributed to poor communication (23). Among obstetric cases that go to litigation, poor communication and teamwork is identifiable in 43% (24). Anesthesia, specifically failed or esophageal intubation, remains a leading cause of maternal death, and anesthesia care often indirectly contributes to poor maternal outcomes.

Even when no complication occurs, the care provided during parturition is often substandard. Up to 85% of women with preeclampsia receive substandard treatment for their blood pressure (25), a statistic that would likely improve with better teamwork and more involvement by an obstetric anesthesiologist. Communication on the labor ward is often inadequate. Simpson found that obstetric nurses and obstetricians may communicate for only several minutes over the entire course of labor (26). Obstetric anesthesia handoffs are frequently short, interrupted by clinical care, and poorly structured (27). Poor communication and coordination of care has been identified in 43% of closed malpractice claims in obstetrics (28). Using in situ simulated eclampsia drills, Thompson et al. found that timely communication with senior obstetric staff was a recurrent problem (29). Similarly, Daniels et al. demonstrated that obstetric residents communicated poorly with their pediatric team members during a simulated emergent delivery. While 63% called for pediatric help during the simulated maternal cardio-pulmonary arrest, only 10% gave helpful information to the pediatricians when they arrived (30). Obstetric care providers often do not value patient safety initiatives (31). Finally, and perhaps most telling, many obstetric care providers do not grade their own institutions highly with regard to safety, and 30% would not want to be delivered in their own institution.

The final reasons to encourage both teamwork and simulation is to help identify areas of clinical weakness and to maintain skills in infrequently encountered clinical scenarios. Obstetricians are prone to making the same errors repeatedly when practicing uncommon events in the simulated environment (32). Up to half of the cases of failed tracheal intubation, the leading cause of anesthesia-related maternal death, are improperly managed (33). This is likely related to the decreased frequency of general anesthesia use in the obstetric population and thus a decreased exposure to the maternal airway (34,35,36). Better teamwork during these crisis events and the ability to practice them using simulation are likely to improve care and decrease adverse outcomes.

Despite all the challenges described above, and hopefully because of the great need, some of the greatest successes in both teamwork and simulation have occurred in obstetrics. The first literature demonstrating improvements in outcomes associated with both team training and simulation were published in obstetrics.

While CRM is not universally defined and no standard CRM training program exists, Salas et al. define CRM as a “family of instructional strategies that seek to improve teamwork in the cockpit by applying well-tested training tools (e.g., simulators, lectures, videos) targeted at specific content (i.e., teamwork knowledge, skills, and attitudes [KSAs])” (37). It has also been referred to as “Airmanship,” “Captaincy,” and “Crew Cooperation” (38). The roots of CRM date back to a NASA workshop in 1979 entitled Resource Management on the Flight deck (39). This conference was designed to help spread research indicating that poor interpersonal communication, decision making, and leadership were leading causes of air traffic accidents. The term Cockpit Resource Management was used at this conference to describe training processes designed to improve these interpersonal aspects of cockpit management. The first versions of this CRM focused largely on the psychological aspects of the captain’s managerial style (12,39). Since then, CRM has evolved through several generations to include a greater focus on the management of the cockpit crew, training of all crew members (flight attendants, dispatchers, maintenance staff, etc.), and a greater emphasis on the impact of human factors on error. The current (fifth) iteration of CRM, however, is the first to focus specifically on error management (12,39). Current CRM training teaches that error is inevitable (“normalizing error” (39), and that human performance is limited (40). Strategies have been developed to prevent error whenever possible, to trap error when it occurs, and finally to mitigate the impact of error when the first two fail (12,39). Specific tools and behaviors are included in CRM training to help accomplish each of these goals.

While CRM does not represent a specific training program, a defined set of KSAs at both the team and individual level are commonly included in most CRM education. The most important attitude is that participants must believe that error, including their own error, is inevitable and that working as a team is likely to reduce these errors. They must be willing to advocate for safety whenever risk is identified and to listen to risk concerns, irrespective of hierarchy. Some of the early concepts are identified in Table 45-1. While these concepts exist to some degree in most CRM training, they may be
difficult to define or measure, and not intuitive to the learner. More recently, Salas (41) suggests that there is a set of “big five” team behaviors necessary for successful teamwork:

These behaviors are supported by shared mental models (the development and articulation of a shared vision regarding plans), closed loop communication (the use of specific communication techniques outlined in Table 45-2), and mutual trust. Other important concepts in CRM training include: Situation Awareness (the state of knowing the conditions of the team and environment that could influence the team’s performance), Conflict Resolution (the ability to rapidly and professionally resolve differences of opinion about appropriate actions when making or modifying plans), Team Structure (clearly identifying the members and the leader of the team so all members know their role and the role of the other members) (42). Finally, several team behaviors help to ensure that the teamwork KSAs occur. Team briefings before scheduled activities allow for planning, development of shared mental models, and defining of roles. Team meetings allow the entire team to come together at predetermined time or ad hoc to discuss general plans, concerns (current or anticipated), staffing problems (current or anticipated), and other issues that might impact team functioning. Team meetings help the team to maintain normal operation and prevent crisis. Debriefing after all activities, especially crisis intervention, allows the team to learn from both successes and failures, and thus to improve future performance.

The best way to teach teamwork has not been established and is controversial (43,44). Within aviation, CRM concepts are generally embedded into simulation scenarios that pilots must perform twice a year. However, options range from a high-fidelity simulation center, to didactic lectures, to the development of clinical protocols that help standardize the team behaviors. In reality, any teaching method will likely use a combination of techniques; those who endorse high-fidelity simulators generally include some didactic teaching, and classroom-based models often allow participants to practice teamwork behaviors in simulated scenarios. Some have advocated that a combination of methods may be best and may even act synergistically to teach teamwork and to bring the behaviors to the clinical environment (13,45). Irrespective of the teaching method, it is clear that CRM education should not be a single experience but must be part of an ongoing educational process (13,39).

Aviation and healthcare share much in common. They are both complicated, highly technical fields that depend on the well-coordinated interactions of multiple team members. Staff must often make decision with incomplete information,
and both must be able to quickly adapt to changes in the plans. Most importantly, both have historically worked with steep hierarchies, and poor performance can lead to fatal consequences. Despite this, there has been relatively little adoption of formal CRM-based training in medicine, although many of the concepts have been shown to be important to patient safety. Aviation-based team training has been shown to improve attitude toward patient safety among operating room staff (46,47,48,49). Surrogate measures have been shown to improve with teamwork processes, including operating room dosing of antibiotics and deep vein thrombosis (DVT) prophylactic medications (50), staff knowledge of planned surgical procedures (51), and efficacy of communication among operating room staff (48). Improvements in patient outcomes have been seen in the operating room (52,53) and emergency room (49) environment with formal team training.

A much larger body of literature demonstrates that even without full team training, the use of specific CRM-based team behaviors may improve outcomes in the general medical literature. Use of pre-procedure briefings has been shown to decrease perioperative mortality (54) and wrong side surgery (55). Use of structured, multidisciplinary rounds may decrease post-operative intensive care unit (ICU) admissions (56) and hospital (57) length of stay. Improved leadership is associated with better teamwork during resuscitation efforts (58).

Fortunately, obstetrics is one of the areas in which formal CRM-based team training has been implemented and assessed. Obstetric teamwork training itself, whether through simulation or classroom-based courses, has consistently been associated with improvements in staff attitudes toward patient safety and teamwork when assessed at the time of the training (20,28,59,60,61,62). The training is generally well received (20,59,62). These attitudes toward patient safety and teamwork can be translated to the clinical environment (24,60,61,63). Using the Safety Attitudes Questionnaire (SAQ) (64), Pratt et al. found that clinician attitudes toward patient safety were significantly higher among labor and delivery staff who had been through classroom-based team training than among those working on other units who had not been trained (60). The SAQ is a single page (double sided) questionnaire with 60 items and demographics information (age, sex, experience, and nationality). It assesses staff perceptions of safety in six categories (see Table 45-3). Gardner developed a 6-hour, multidisciplinary simulation course involving obstetricians, anesthesiologists, obstetric nurses, and midwives. Self-assessment of teamwork and communication more than 1 year after course completion demonstrated improvements in both areas. In addition, most clinicians felt that their clinical practice had changed because of the course (28). Finally, Haller et al. trained 239 obstetric nurses, physicians, midwives, and other labor and delivery staff in a 2-day, classroom-based, CRM-style course. Initial reactions to the course and evidence that participants learned the CRM concepts were both very positive. Surveys of the staff over the next year demonstrated improvements in attitudes toward patient safety, stress recognition, work conditions, and job satisfaction. Participants also reported improved availability of clinical information and “feeling part of a bigger family” (61).

The impact of team training on clinician teamwork behaviors has been poorly evaluated in obstetrics. While improvements in various teamwork behaviors have been demonstrated after training in other clinical environments (48,49), this has not been formally evaluated on labor and delivery units. This may be in part because measuring teamwork is an imprecise science and not easily done in obstetrics (65). Robertson et al. did evaluate the impact of simulation-based team training on teamwork behaviors among obstetric nurses, midwives, and attending and resident obstetricians within the simulator (66). The authors found that teamwork task completion improved from 24% to 40% in the first simulated scenario to 80% to 100% in the fourth. However, it is unclear whether the behaviors learned in the simulators were transferred to the clinical setting.

Improvements in patient outcomes are the ultimate measure of the impact of team training. A large, prospective randomized trial evaluating the impact of a classroom-based CRM course based on the MedTeams curriculum previously developed for the emergency room failed to demonstrate improvements in patient outcomes in obstetrics. The authors did find a 10-minute (∼33%) improvement in the time from decision to incision in emergent cesarean deliveries. Inadequate power, high staff turnover, and a short implementation time for the teamwork behaviors may have contributed to the negative results (18).

However, others have demonstrated improvements in patient outcomes associated with both classroom and simulation-based team training. Pratt et al. trained more than 220 labor and delivery staff in a classroom-based CRM teamwork course. In addition, the authors described a structured implementation process involving the use of templates, structured language, coaches, and three types of formal teams that helped to translate the behaviors to the clinical environment. The authors used a measure of adverse events called the Adverse Outcomes Index (AOI) (Table 45-4) (67). AOI, a weighted composite measure of ten adverse events, was used to measure the impact of CRM training on their labor and delivery unit. They found that obstetric complication rates and overall severity decreased by 23% and 13.2% respectively among more than 19,000 women who delivered after the implementation of teamwork. The severity of the outcomes among those who had an adverse event also decreased, suggesting improved team response to evolving events. They did not measure teamwork behaviors on the unit, and thus could not draw a direct causal relationship between the training and the improved outcomes (60). Similarly, Pettker described a multi-step process designed to improve safety on their labor and delivery unit. This included the development of clinical protocols, fetal monitoring certification, a safety nursing committee, and team training. The entire process required nearly 2 years to implement. The adverse event rate, as measured by the AOI, decreased by nearly 28% in the second half of the study period (see Fig. 45-1) (68). In addition, the percentage of nurses and physicians that reported a “good teamwork climate” increased from 16.4% and 39.5% respectively to 72.2% and 88.7%. Using a similar multi-step model that mandated CRM-based team training, Grunebaum et al. described a dramatic drop in the number of sentinel events on their obstetrics unit (69). Team training was the first of 19 individual safety steps implemented over a 6-year period. The unit experienced three to five sentinel events per year at the start of the program. This number dropped to three total events over the last five years (0.6/year). Shea-Lewis described a 43% reduction in the rate of adverse obstetric events after the implementation of a CRM-based team training curriculum in an intermediate-sized community hospital (70). Finally, in a landmark paper, Draycott et al. developed a 1-day course that combined didactic and simulation training in both teamwork behaviors and obstetric crisis management. All obstetric care providers at a large, urban
center were required to attend the course in multidisciplinary sessions. Evaluation of more than 19,000 deliveries during the 2 years prior to and 3 years after training demonstrated a 50% risk reduction in the rate of neonatal hypoxic ischemic encephalopathy (HIE) after the training (Table 45-5) (71). While it was unclear whether the improvement was related to better individual clinical care or better teamwork, the decrease in the rate of hypoxic brain injury is impressive.

As with the general medical literature, the use of elements of CRM-type teamwork behaviors without full team training has also been shown to improve provider care and patient outcomes in obstetrics. The implementation of an obstetric-specific rapid response team has been associated with improved outcomes (72). This effort used protocols to ensure role clarity, structured language, and simulation to allow the teams to practice. Similarly, Skupski et al. developed an obstetric rapid response team specifically for maternal hemorrhage. Guidelines helped define roles of the team members and improved communication processes were created. These efforts were associated with a decrease in maternal mortality due to hemorrhage (73). Clark described a system-wide approach to improving outcomes in obstetrics. This included many CRM concepts without formal training in teamwork. The authors described improved outcomes, fewer cesarean deliveries, and a drop in malpractice claims associated with these changes (Fig. 45-2) (21). The use of a shoulder dystocia protocol has been used to improve role clarity, situation awareness, shared mental models, and communication—all CRM concepts (74).