Avoiding Patient Harm in Anesthesia: Human Performance and Patient Safety

Observation

Anesthesia professionals use observations to decide whether the patient’s course is on track or whether a problem is occurring; this is the first step of the decision making cycle. Data are observed and transformed by interpretation into information, followed by further interpretation into meaning. Data streams typically involve direct visual, auditory, or tactile contact with the patient, the surgical field, routine electronic monitoring, special (sometimes invasive) monitoring systems, contents of suction canisters and sponges, reading of reports of laboratory test results, and communications from other personnel. Loeb showed that anesthesiologists typically observe monitors for approximately 1 to 2 seconds every 10 to 20 seconds and that it usually took several observing cycles before they detected a subtle cue on the monitor. Management of rapidly changing situations requires the anesthesia professional to assess a wide variety of information sources. Because the human mind can attend closely to only one or two items at a time, the anesthesia professional’s supervisory control level must decide what information to attend to and how frequently to observe it (as later shown at CRM key point 14 “Allocate attention wisely”). Constant observation and interpretation of the different information systems is executed repeatedly throughout the course of an anesthetic regimen. The plethora of simultaneous data streams in even the most routine cases is a challenge. Vigilance, defined as the capacity to sustain attention, plays a crucial role in the observation and detection of problems and is a necessary prerequisite for meaningful care. Vigilance can be degraded by performance-shaping factors (see later section “Performance Shaping Factors”) and it can be overwhelmed by the sheer amount of information and the rapidity with which it is changing.

Verification

In the working environment of an anesthesia professional, the available, observed information is not always reliable. Most monitoring is noninvasive and indirect and is susceptible to artifacts (false data). Even direct clinical observations such as vision or auscultation can be ambiguous. Brief transients (true data of short duration) can occur that quickly correct themselves. To prevent them from skewing the decision making process and triggering precipitous actions that may have significant side effects, critical observations must be verified before the clinician can act on them. This requires the use of all available data and information and cross-checking different related data streams rather than depending solely on any single datum without sensible interpretation (as later shown in CRM key point 8 “Use all available information” and CRM key point 10 “Cross check and double check; never assume anything”). Verification uses a variety of methods, shown in Table 6.2 .

Problem Recognition

Having recognized a problem, how does the expert anesthesia professional respond? The classical paradigm of decision making involves a careful comparison of the evidence with various causal hypotheses that could explain the problem. This is followed by a careful analysis of all possible actions and solutions. This approach, although powerful, is relatively slow and does not work well with ambiguous or scanty evidence. Many perioperative problems faced by anesthesia professionals require quick action under uncertainty to prevent a rapid cascade to a catastrophic adverse outcome, and solution of these problems through formal deductive reasoning from first principles is just too slow. The process of problem recognition is a central feature of several theories of cognition in complex, dynamic worlds. Problem recognition involves matching sets of environmental cues to patterns that are known to represent specific types of problems. Given the high uncertainty seen in anesthesia, the available information sources cannot always disclose the existence of a problem, and even if they do, they may not specify its identity or origin. Anesthesia professionals use approximation strategies to handle these ambiguous situations; psychologists term such strategies heuristics . Stiegler and Tung give a detailed review of heuristics and other biases that affect problem recognition. One heuristic is to categorize what is happening as one of several generic problems, each of which encompasses many different underlying conditions (similarity/pattern matching). Another is to gamble on a single diagnosis (frequency gambling ) by initially choosing the single most frequent candidate event. During preoperative planning, the anesthesia professional may adjust a mental “index of suspicion” for recognizing certain specific problems anticipated for that particular patient or surgical procedure. The anesthesia professional must also decide whether a single underlying diagnosis explains all the data or whether they could come from multiple causes. This decision is important because excessive attempts to refine the diagnosis can be very costly in terms of allocation of attention. By contrast, a premature diagnosis can lead to inadequate or erroneous treatment. The use of heuristics is typical of expert anesthesia professionals and often results in considerable time savings in dealing with problems. However, it is a double-edged sword. Both frequency gambling and inappropriate allocation of attention solely to expected problems can seriously undermine problem solving when these gambles do not pay off or are not corrected in the reevaluation process.

Many of the issues that are related to problem recognition and cognition in general are discussed in more detail in the section on decision making further below, especially when dealing with the models of System I thinking and System II thinking and of recognition-primed decision making.

Prediction of Future States

Problems must be assessed in terms of their significance for the future states of the patient. Predicting future states based on the occurrence of seemingly trivial problems is a major part of the anticipatory behaviors that characterize expert crisis managers. Problems that are already critical or that can be predicted to evolve into critical incidents receive the highest priority (as later shown in CRM key point 15 “Set priorities dynamically”). Prediction of future states also influences action planning by defining the timeframe available for required actions. Cook and colleagues described “going sour” incidents in which the future state of the patient was not adequately taken into account when early manifestations of problems were apparent. One of the challenges known from research in psychology is that the human mind is not very well suited to predict future states, when things are changing in a nonlinear fashion. Under such circumstances, which are common for natural systems such as the human body, the rate of change is almost invariably underestimated, and people are surprised at the outcome.

Precompiled Responses

Once a critical event has been observed and verified the anesthesia professional needs to respond. In complex, dynamic domains, the initial responses of experts to the majority of events stem from precompiled rules or response plans for dealing with a recognized event. This method is referred to as recognition-primed decision making, because once the event is identified, the response is well known (see later section “Decision Making”). In the anesthesia domain, these responses are usually acquired through personal experience alone, although there is a growing realization that critical response protocols must be codified explicitly and taught systematically. Experienced anesthesia professionals have been observed to rearrange, recompile, and rehearse these responses mentally based on the patient’s condition, the surgical procedure, and the problems to be expected. Ideally, precompiled responses to common problems are retrieved appropriately and executed rapidly. When the exact nature of the problem is not apparent, a set of generic responses appropriate to the overall situation may be invoked. For example, if a problem with ventilation is detected, the anesthesia professional may switch to manual ventilation at a higher fraction of inspired oxygen (FiO ₂ ) while considering further diagnostic actions. However, experiments involving simulation have demonstrated that even experienced anesthesia professionals show great variability in their use of response procedures to critical situations. This finding led these investigators to target simulator-based training in the systematic training of responses to critical events.

Even the ideal use of precompiled responses is destined to fail when the problem does not have the suspected cause or when it does not respond to the usual actions. Anesthesia cannot be administered purely by precompiled “cookbook” procedures. Abstract reasoning about the problem through the use of fundamental medical knowledge still takes place in parallel with precompiled responses, even when quick action must be taken. This seems to involve a search for high-level analogies or true deductive reasoning using deep medical and technical knowledge and a thorough analysis of all possible solutions. Anesthesia professionals managing simulated crises have linked their precompiled actions to abstract medical concepts.

Taking Action/Action Implementation

Anesthesiologists need to share their attention among different cognitive levels, among tasks, and often among problems. The intensive demands on the anesthesia professional’s attention could easily swamp the available mental resources. Therefore, the anesthesia professional must strike a balance between acting quickly on every small perturbation (which requires a lot of attention) and adopting a more conservative “wait-and-see” attitude. This balance must be constantly shifted between these extremes as the situation changes. However, during simulated crisis situations, some practitioners showed great reluctance to switch from business as usual to emergency mode even when serious problems were detected. Erring too far in the direction of wait and see is an error that can be particularly catastrophic. Preparedness for active intervention in case of dynamically changing events is a key element of an anesthesiologist’s work. But how frequent is this requirement? According to the review of Wacker and Staender, adverse events in the perioperative period continue to be frequent, occur in about 30% of hospital admissions, and may be preventable in more than 50%.

At any time during an anesthetic regimen there may be multiple things to do, each of which is intrinsically appropriate, yet they cannot all be done at once. Simulation experiments have shown that anesthesia professionals sometimes have difficulty selecting, planning, and scheduling actions optimally.

A particular hallmark of anesthesiology is that the decision maker does not just decide what action is required but is often involved directly in the implementation of actions. Executing these actions requires substantial attention and may in fact impair the anesthesia professional’s mental and physical ability to perform other activities (e.g., when an action requires a sterile procedure). This is particularly an issue when other tasks have been interrupted or temporarily suspended. Prospective memory, one’s ability to remember in the future to perform an action (i.e., to complete a task) can be easily disrupted. In addition, anesthesia professionals engaged in a manual procedure are strongly constrained from performing other manual tasks or from maintaining awareness of incoming information.

Reevaluation

In order to successfully solve dynamic problems and to cope with the rapid changes and profound diagnostic and therapeutic uncertainties seen during anesthesia, the core process must include repetitive reevaluation of the situation. Thus, the reevaluation step, initiated by the supervisory control level, returns the anesthesia professional to the observation phase, but with specific assessments in mind (see also CRM key point 12, “Reevaluate repeatedly”). Only by frequently reassessing the situation can the anesthesia professional adapt to dynamic processes, since the initial diagnosis and situation assessment can be incorrect. Even actions that are appropriate to the problem are not always successful.

The process of continually updating the assessment of the situation and monitoring the efficacy of chosen actions is termed situation awareness . Situation awareness is a very interesting and important topic in analyzing performance and reasons for errors and is discussed in detail in a later section. Box 6.2 gives examples of reevaluation questions in order to maintain situation awareness.

Faulty reevaluation, inadequate adaptation of the plan, or loss of situation awareness can result in a type of human error termed fixation error . Fixation errors have been described in responses of professionals to abnormal situations. Avoiding and recognizing fixation errors in the field of anesthesia is covered more in detail in section “Introduction of the 15 Crisis Resource Management Key Principles.”

Supervisory Control

The supervisory control allocates the scarce resource of attention during multitasking, and oversees and modulates the core process. For example, determining the frequency of observation of different data streams, prioritizing diagnostic and therapeutic alternatives, actively managing workload, prioritizing and scheduling actions. Supervisory control actively manages the workload by (1) avoiding high-workload situations by anticipation and planning, (2) distributing workload over time or (3) over personnel, (4) changing the nature of the task to reduce work, or (5) minimizing distraction. More details regarding the active management of workload are touched on in the later section on CRM key point 5 “Distribute the Workload.”

Resource Management

The highest layer of metacognition and control is known as resource management—the ability to command and control all the resources at hand to care for the patient and to respond to problems. This involves translating the knowledge of what needs to be done into effective team activity by taking into account the limitations of the complex and often ill-structured perioperative domain. Resources include personnel, equipment, and supplies, both in the immediate vicinity and, when necessary, throughout the various levels of the organization. Resource management explicitly demands teamwork and crew coordination. It is not enough for the anesthesia professional to know what to do or even to be able to do each task alone. Only so much can be accomplished in a given time, and some tasks can be performed only by other skilled personnel (e.g., catheterization lab). A key responsibility of the anesthesia professional is to mobilize needed resources and to distribute the relevant goals and tasks among those available. The details of this critical function are described in the section on “Crisis Resource Management.”

Primary Task Performance

The primary task performance measure assesses the subject’s performance on standard work tasks (e.g., cases seen, knots tied, etc.) as they are made progressively more difficult by increasing the number of tasks, task density, or task complexity. At first, the subject is able to keep up with the increasing task load, but at some point, the workload exceeds the ability to manage it, and performance on the standard tasks decreases.

Secondary Task Probing

Secondary task probing tests the subject with a minimally intrusive secondary task that is added to the primary work tasks. The secondary task is a simple one for which performance can be objectively measured. Reaction time, finger tapping, mental arithmetic and a vibrotactile device have for example been used for this technique as a secondary task. The anesthesia professional is instructed that the primary tasks of patient care take absolute precedence over the secondary task. Therefore, assuming that the secondary task requires some of the same mental resources as the primary task, the performance of the anesthesiologist on the secondary task is an indirect reflection of the spare capacity available to deal with it: the greater the spare capacity, the lower the primary workload. Depending on the secondary task response channels (manual, voice, gesture, multiple ways) there can exist channel interference. Controversy exists about whether these probes measure “vigilance” or “workload,” although the same techniques probably measure both aspects of performance. When probes occur infrequently, are subtle, have multiple response channels, and are performed with a low level of existing workload, they are more likely to measure vigilance; when they are frequent, readily detectable, require a manual response, and are performed during a high-workload period, they probably are more indicative of spare capacity and workload.

Subjective Measures

In subjective measures, individuals are asked, most commonly in retrospect but sometimes in real time, how much load they were or are under during actual work situations. A common and validated form to assess subjective workload is the NASA TLX form. Subjective measures usually complement objective measurements of external observations, since an anesthesia professional may subjectively underestimate the workload in settings in which objective measurements demonstrate a marked reduction in spare capacity.

Physiologic Measures

The final set of techniques for assessing workload consists of physiologic measures. Visual or auditory evoked potentials have been used successfully to assess mental workload, but this technique can be used only in a static laboratory environment. Heart rate (especially certain aspects of heart rate variability) and blood pressure are other physiologic measures that have been used, but there are challenges in reliable interpretation.

Teaching

Close interaction of experienced anesthesia professionals with inexperienced clinical trainees during actual surgical procedures is a standard approach to training. It can be hypothesized that teaching adds to the workload of the more experienced care provider who is simultaneously responsible for safe and efficient anesthesia care during the procedure. Weinger and co-workers found that teaching teams, involving one-to-one supervision of fourth-year medical students or first-month anesthesia residents by an attending anesthesiologist, had significantly slower response times to a warning light than non-teaching teams of attending(s) of similar experience. Response latency was highest during induction and emergence. This vigilance test was also a procedural (performance) workload assessment measure indicating increased workload and reduced spare capacity. They also found that workload density was significantly increased for teaching as opposed to non-teaching teams. In sum, intraoperative teaching increased workload and decreased vigilance, suggesting the need for caution when educating during patient care.

Delegation and Supervision

Experience suggests that the effect of delegation on workload varies depending on the nature of the task and how confident the delegating anesthesia professional feels about the capability of the person to whom the task is assigned. Delegation must be guided by the supervisor’s situation awareness and overall ability to process information on the patients’ condition. Several further aspects are presented in the work of Leedal and Smith.

Routine Events

Novice trainee anesthesia professionals were found to perform many of the same tasks as do more experienced personnel at specific phases of an anesthetic regimen, but take longer over tasks, show longer latency of response, and greater task workload than third-year trainees and experienced nurse anesthetists. While task density was generally highest for both subject groups in the period up to, during, and immediately after intubation, the more experienced trainees had higher task densities than the novices, suggesting greater competence among the former in carrying out multiple tasks in a short period of time. Those findings are in line with other studies, including the study of Weinger and associates that evaluated the mean response time of pressing a buzzer at the flashing of a red light (secondary task). The response time was markedly less than 60 seconds for experienced subjects in both the induction and post induction (maintenance) phases, but it was much higher for novice residents during the induction phase. One explanation for those findings may be that the reduction of workload depends partly on the degree to which tasks can become routine, thus freeing mental resources for other tasks. Leedal and Smith conclude concerning this matter in their review: “Experienced staff appear to show ‘spare capacity’ in performance during routine cases, which we suggest allows them an attentional ‘safety margin’ should adverse events occur ” (p. 708)

In contrast, a more recent study presented by Byrne and co-workers found there was limited evidence of a relationship between workload and experience.

Another recent finding indicates that more experienced anesthesia teams may be more likely to attempt to coordinate implicitly, without much overt communication, which makes them more reliant on accurate and shared understandings of the task and their teamwork.

Novice residents also spent more time speaking to their attending staff (11% of preintubation time) than did experienced residents or CRNAs. Experienced personnel observed the surgical field more than did the novices. Novices did take longer to complete patient preparation and induction of anesthesia, but it appeared that some of the extra time taken by novices working under supervision was offset by the efficiency of offloading other concurrent tasks to the attending anesthesiologist such that preintubation time was increased by only 6 minutes for novices.

Critical Events/Emergencies

Schulz and colleagues presented data where more experienced anesthetists (>2 years work experience) increased the amount of time dedicated to manual tasks from 21% to 25% during critical incidents, whereas the less experienced decreased from 20% to 14%. The less experienced anesthesia providers spend more time on monitoring tasks.

A study by Byrne and Jones looked at differences in the performance of experienced and less experienced anesthesia professionals during 180 simulated anesthesia emergency scenarios. The results showed significant differences only between the first and second year. As seen in other studies, significant errors occurred at all levels of experience, and most of the anesthesia professionals deviated from established guidelines.

A classic simulation study by DeAnda and Gaba investigated the response of anesthesia trainees and experienced anesthesia faculty and private practitioners to six preplanned critical incidents of differing type and severity. The incidents included (1) endobronchial intubation (EI) resulting from surgical manipulation of the tube; (2) occlusion of intravenous (IV) tubing; (3) atrial fibrillation (AF) with a rapid ventricular response and hypotension; (4) airway disconnection between the endotracheal tube and the breathing circuit; (5) breathing hoses too short to turn the table 180 degrees, as requested by the surgeon; and (6) ventricular tachycardia or fibrillation. For each incident, considerable interindividual variability was found in detection and correction times, in information sources used, and in actions taken. The average performance of the anesthesia professionals tended to improve with experience, although this varied by incident. The performance of the experienced groups was not better than that of the second-year residents (who were in their final year of training at that time). Many (but not all) novice residents performed indistinguishably from more experienced subjects. Each experience group contained some who required excessive time to solve the problem or who never solved it. In each experience group at least one individual made major errors that could have had a substantial negative impact on a patient’s clinical outcome. For example, one faculty member never used electrical countershock to treat ventricular fibrillation. One private practitioner treated the EI as though it were “bronchospasm” and never assessed the symmetry of ventilation. One resident never found the airway disconnection. The elements of suboptimal performance were both technical and cognitive. Technical problems included choosing defibrillation energies appropriate for internal paddles when using external paddles, ampule swap, and failure to inflate the endotracheal tube cuff that resulted in a leak. Cognitive problems included failure to allocate attention to the most critical problems and fixation errors.

Schwid and O’Donnell performed an experiment similar to those of DeAnda and Gaba and received similar results. After working on several practice cases without critical incidents, each subject was asked to manage 3 or 4 cases involving a total of 4 serious critical events (esophageal intubation, myocardial ischemia, anaphylaxis, and cardiac arrest). The anesthesiologists studied had varying experience levels. Significant errors in diagnosis or treatment were made in every experience group. The errors occurred in both diagnosis of problems and in deciding on and implementing appropriate treatment. For example, 60% of subjects did not make the diagnosis of anaphylaxis despite available information on heart rate, blood pressure, wheezing, increased peak inspiratory pressure, and the presence of a rash. In managing myocardial ischemia, multiple failures occurred. 30% of subjects did not compensate for severe abnormalities while considering diagnostic maneuvers. Fixation errors in which initial diagnoses and plans were never revised were frequent, even when they were clearly wrong.

Independent from professionals’ experience, Howard and colleagues found a substantial incidence of difficulties during simulated emergencies: managing multiple problems simultaneously, applying attention to the most critical needs, acting as team leader, communicating with personnel, and using all available OR resources to best advantage. Analysis of videotaped trauma and resuscitation cases has revealed inadequacies in the availability and arrangement of monitoring equipment, as well as nonexistent or ambiguous communication. Byrne and Jones evaluated the performance of anesthetists during nine emergency cases, showing that serious errors in both diagnosis and treatment were made and accepted treatment guidelines were not followed. Diagnosis of several common critical incidents, for example, anaphylaxis, was hard to name. As described earlier, once diagnosed, no structured plans or algorithms were used.

These classic studies date back 25 years or more, yet the most recent studies remain completely consistent with older studies. In the 2017 Weinger and colleagues publication, a total of 263 consenting U.S. board-certified anesthesiologists (BCAs—i.e., physicians) participated in two, 20-minute, standardized, high-fidelity simulation scenarios of unanticipated acute events during existing Maintenance of Certification in Anesthesiology simulation courses. The scenarios were from among the following: (1) local anesthetic systemic toxicity with hemodynamic collapse; (2) hemorrhagic shock due to hidden retroperitoneal bleeding during laparoscopy; (3) malignant hyperthermia presenting in the postanesthesia care unit; and (4) acute onset of AF with hemodynamic instability during laparotomy followed by ST elevation myocardial infarction. Performance measurement rubrics were established in advance: a scoresheet of critical clinical performance elements, behavioral anchored ordinal scales of four nontechnical skills, ordinal scales for overall individual and team technical performance and nontechnical performance, and a binary rating as to whether performance met or exceeded that expected of a BCA. The results showed that critical clinical performance elements were commonly omitted, roughly in four broad areas of crisis management, failures to: (1) escalate therapy where first-line options did not work (e.g., using epinephrine or vasopressin when phenylephrine, ephedrine, or fluids did not sufficiently correct hypotension); (2) use available resources (e.g., calling for help when conditions have deteriorated appreciably); (3) speak up and engage other team members, especially when action by them was required (e.g., asking the surgeon to change the surgical approach when it is essential to effective treatment); and (4) follow evidence-based guidelines (e.g., giving dantrolene to a patient with obvious MH). The performance of approximately 25% of subjects was rated in the low portion of the various technical and nontechnical 9-point ordinal scales. In about 30% of encounters, performance was rated as “below the level expected of a BCA.”

Different Components of Human Factors

Derived from the SEIPS model and adapted by the authors, different components of HF are:

• ▪

  the behavior of individuals and their behavior/knowledge in regard to tasks (individual level)

• ▪

  the interactions with each other (team level)

• ▪

  the interactions of professionals with the organizational/sociocultural conditions (organizational level)

• ▪

  the interactions of professionals with the environment/workspace (environmental level) and

• ▪

  the interactions of professionals with technology/equipment (technology/engineering/design level)

Those five human factor components are necessary and sufficient to describe and understand the anesthesia professional’s entire work system from a HF perspective. The components interact with and influence each other, resulting in a large number of relationships between different levels. At times the components compensate each other (e.g., when professionals work faster to compensate for time pressure or when people collaborate to solve problems that are beyond an individual’s abilities). Other times the components resonate with each other and amplify their effects—for the good (e.g., having the right equipment for the task at hand) or the bad (e.g., lacking resources to solve a problem). Given that this chapter cannot deal with human factors in a comprehensive way, many topics can be touched on only briefly. Although the environmental and technology levels can be important—cognition is challenged when there is a power failure or a breakdown of key clinical equipment—in this chapter the focus lies on the most important aspects of human factors directly relevant to anesthesia professionals: the individual level, team level, and organizational level.

Different Ways to Categorize NonTechnical Skills

In general, NTS can be categorized into two broad areas: (1) cognitive and mental skills on the individual level, including decision making and situation awareness; and (2) social and interpersonal skills on the team level, including teamwork, communication, and leadership. Commercial aviation incorporated such nontechnical skills in the cockpit (later crew) resource management paradigm (CRM, 1980s and continuous later evolution) and the “NOTECHS” paradigm in the late 1990s. In anesthesiology, the anesthesia crisis resource management (ACRM) framework was introduced by Howard and co-workers in 1990 as an adaptation of aviation’s CRM. The ACRM approach categorizes NTS based on the five key elements of communication, situation awareness, decision making, teamwork (implicitly including leadership), and task management ( Fig. 6.5 ). The anesthesia nontechnical skills (ANTS) framework by Fletcher and colleagues was introduced in 2003; Flin and colleagues give an overview of its history, development, application, use, and emerging is issues. ^, The ANTS approach typically includes the four categories of situation awareness, decision making, teamwork (explicitly including leadership), and task management. Neither of these frameworks, and indeed no usable paradigm, can capture explicitly every important aspect of HF applied to anesthesiology. For example the management of the performance-shaping factors stress and fatigue are also part of the ANTS framework, but are implicit in ACRM. Communication itself is not an explicit skill element of the ANTS framework (which assumes that communication pervades each element), whereas in ACRM it is a specific skill that needs explicit mention and training. Of note, from their start in anesthesiology, nontechnical skills frameworks have been further adapted to several other medical fields, such as for surgeons and intensive care specialists.

Assessment of NonTechnical Skills

In response to the growing acceptance of human factors and nontechnical skills influencing medical performance, several rubrics to measure NTS have been developed. Gaba and colleagues in 1998 described the direct adaptation for assessment of NTS of a set of CRM-anchored ordinal scale markers from Helmreich et al. concerning aviation. In 2004 the ANTS framework was complemented by a behaviorally anchored rating scale in 2004. A new set of assessment scales for four markers of NTS was introduced by Weinger and colleagues in 2017 based on the fusion of aspects of a number of prior approaches.

A further, rather new approach for the assessment of nontechnical skills is the Co-ACT framework described by Kolbe and co-workers. The framework serves observing coordination behavior, especially in acute care teams like anesthesia, and consists of four categories, each category obtaining three further subelements that more specifically describe the NTS: (1) explicit action coordination with the elements instruction, speaking up , and planning ; (2) implicit action coordination with the elements monitoring , talking to the room (action-related), and providing assistance ; (3) explicit information coordination with the elements information request , information evaluation , and information upon request ; and (4) implicit information coordination with the elements gathering information, talking to the room (information-related), and information without request .

Challenges of the Assessment of NonTechnical Skills

The psychometric qualities of measuring NTS have been assessed by the developers of the ANTS system and were considered to be of acceptable level, whereas another study assessed the reliability after a 1-day training for raters and concluded that reliability was poor. A study from Denmark showed good psychometric qualities. Originally developed for educational purposes and used to discuss the NTS of anesthesiologists after training sessions in order to improve in NTS, Zwaan and colleagues assessed the usability and reliability for ANTS in research, finding that the ANTS system was reliable for the total score and usable to measure physicians’ NTS in a research setting. However, the investigators found a variation between the reliability of the different elements and recommend excluding elements in advance that are not applicable or observable in the situation of interest. It might not always be easy, however, to identify those elements that should be excluded. On the whole, the ANTS system appears to be a useful tool to enhance assessment of nontechnical skills in anesthesia and other medical fields further, and its careful derivation from an established system of nontechnical assessment in aviation (NOTECHS) may even allow some interdomain comparisons. However, more recently Watkins and colleagues directly compared ANTS with the system used in the Weinger and colleagues 2017 paper and showed that ANTS was more difficult to use but that the two systems otherwise achieved equivalent assessment results.

NonTechnical Skills: The Bad - The Good - The Variable

Good nontechnical skills (i.e., vigilance, efficient communication, team coordination, etc.) reduce the likelihood of active and passive errors and adverse events, and also increase human performance, whereas suboptimal NTS are expected to do the reverse. These effects are variable; for example, a person may communicate very effectively in one instance and fail to do so in the next challenging communication role.

A study from Denmark investigated the relationship between technical and nontechnical skills. Twenty-five video recordings of second-year anesthesiologists managing a simulated difficult airway management scenario were rated with the ANTS instrument and a score for the technical aspect for the procedure. In addition, written descriptions of the NTS performance were collected and content analyzed. The correlation between the two scores was not significant, but the content analysis comparing the NTS description in the best and poorest three scenarios identified what contributed most to good NTS. These were systematically collecting information, thinking ahead, communicating and justifying of decisions, delegating tasks, and vigilantly responding to the evolving situation. Poor NTS were related to: lack of structured approach, lack of articulating plans and decisions, poor resource and task management, lack of considering consequences of treatment, poor response to the evolving situation, and lack of leadership.

What is the Impact of Non-Technical Skills and Human Factors on Poor Performance in Medicine?

The impact of poor NTS on poor human performance in the medical field cannot be overestimated. Depending on the literature, it is estimated that up to 80% of all errors in medicine can be attributed to problems with NTS and human factors.

Even though one can argue on the one hand that due to progress in technology the findings and estimates of the pioneer study of Cooper and colleagues performed in 1978 (results relating up to 80% of incidents to human factors) have been outdated for a long time, on the other hand those figures are (1) comparable to findings in other dynamic and complex work environments and (2) newer studies as recent as 2015 still reconfirm those findings (references see below). In the following some illustrative studies are mentioned that link HF and safety challenges.

In 1993 an analysis of 2000 incident reports from the Australian Incident Reporting Study took place, investigating the incidents for relations to human factors and NTS. In 83% of incidents, human factor elements were scored by the reporters. Even though scoring by the incident reporters might not produce results as accurate as those of a systematic scoring by human factor experts and such text descriptions are of limited value for this type of data collection, it still shows the scope of the problem—having also in mind that voluntarily reported incidents only represent the peak of an iceberg, with many incidents not reported at all.

Fletcher and colleagues published a review of studies describing the influence of nontechnical skills in anesthesia in 2002, summarizing that it is clear “ that nontechnical skills play a central role in good anesthesia practice and that a wide range of behaviors are important […] [including] monitoring, allocation of attention, planning and preparation, situation awareness, prioritization, applying predefined strategies/protocols, flexibility in decision-making, communication, and team-working” [p.426].

As increasing evidence suggested that human factors like communication, leadership, and team interaction influence the performance of cardiopulmonary resuscitation (CPR), Hunziker and colleagues presented a study in 2010 reviewing the impact of human factors during simulation-based resuscitation scenarios. Similar to studies in real patients, simulated cardiac arrest scenarios revealed many unnecessary interruptions of CPR as well as significant delays in defibrillation—two outcome-relevant parameters. The studies showed that human factors played a major role in these shortcomings and that medical performance, at least in non-routine situations, depends among others on the quality of leadership and team-structuring.

Jones and co-workers only recently systematically reviewed the literature on the impact of HFs in preventing complications in anesthesia due to poor airway management and highlighted recent national reports and guidelines, including the 4th National Audit Project (NAP4). NAP4 (2011) was the first prospective study of all major airway events occurring throughout the UK, reviewing any complications resulting from airway management that led to either death, brain damage, the need for an emergency surgical airway, unanticipated ICU admission, or prolongation of ICU stay. 184 reports were reviewed. Subsequent in-depth analysis identified HFs as having been a relevant influence in every case, with a median range of 4.5 contributing HFs per case. HFs in the report included for example : casual attitude toward risk/overconfidence, peer tolerance of poor standards, lack of clarity in team structures, poor or dysfunctional communication including incomplete or inadequate handovers, inadequate checking procedures, failure to formulate back-up plans and discuss them with team members, failure to use available equipment, attempts to use unknown equipment in an emergency situation, heavy personal workload, lack of time to undertake thorough assessment, equipment shortage, inexperienced personnel working unsupervised, organizational cultures which induce or tolerate unsafe practices, no formalized requirement to undertake checking procedures, incompatible goals, and reluctance to analyze adverse events and learn from errors.

The Sentinel Event Report of the Joint Commission analyzed the root causes of 764 sentinel events (patient’s death, loss in function, unexpected additional care and/or psychological impact) reported voluntarily in 2014. Human-factor-related root causes, including among others communication and leadership, were the leading root causes of sentinel events and made up 65% of root causes. Even though the data are not specifically linked to anesthesia practice, it is very likely that there exist strong parallels.

One of the latest simulation-based studies, performed by Weinger and colleagues in 2017 and explained in detail earlier (section on assessing performance, “Performance as a Function of Experience”), also shows a broad variety of human-factor-related hurdles for anesthesiologists when handling emergencies.

What are the implications of the Impact of Humand Factors and NonTechnical Skills on Human Performance?

• ▪

  Balancing the importance of technical and nontechnical skills during medical education and training. While technical skills and medical knowledge have always been at the core of medical education and training, the importance of NTS has—despite long growing evidence—only recently been recognized.

• ▪

  Acknowledgement of the importance of human factors as individual and team. One of the first steps for individuals and teams in minimizing and mitigating human frailties and strengthening human strengths—and in consequence reducing medical error and its consequences—is by acknowledging human factors as a part of human performance and thus acknowledging human limitation. The World Health Organization (WHO) states:

“A failure to apply human factors principles is a key aspect of most adverse events in health care. Therefore, all health-care workers need to have a basic understanding of human factors principles. Health-care workers who do not understand the basics of human factors are like infection control professionals not knowing about microbiology.” (p. 111)

Multitasking and Multiplexing

Sometimes multitasking—doing several things at a time—can be successful, depending on the situation and the tasks at hand. Yet professionals need to be aware that it can be unsuccessful when the situation or the tasks change. When task density becomes high, simple linear performance of tasks becomes impossible; instead more complicated processes of both multitasking and multiplexing are needed. Multitasking—trying to perform two (or more) tasks simultaneously—is frequent in hospital settings, but is virtually impossible to execute if the same cognitive resources are involved. Conducting two or more relatively simple tasks is often possible, such as observing the surgical field while asking the surgeon a straightforward question. Even then there is a risk that one task will be dropped or degraded. However, adjusting an infusion pump while calculating the upcoming dose of antibiotic for an infant would be difficult to do in parallel.

In common use, the term multitasking sometimes refers specifically to media multitasking—everyday attempts to simultaneously work, listen to music, check email and text messages, and search the web. Studies have shown that frequent media multitaskers perform worse on psychology laboratory probes of actual multitasking than those who are infrequent media multitaskers. Many suggest that true multitasking is impossible and that almost always there is a decrement of performance on some or all tasks when multitasking is attempted.

The degree to which media multitasking, or laboratory tests, relate to the kinds of multitasking performed by personnel in professional settings of many data streams and many tasks is as yet uncertain; it seems likely that in anesthesia care there are limits to what individuals can safely do on their own.

While simple multitasking in perioperative settings may work, when tasks are complex it may in fact be inefficient. Rather than conducting two (or more) activities in parallel, humans often do things sequentially but shift rapidly between items or back and forth across many simultaneous threads. If the tasks use different cognitive resources, this is referred to as multiplexing. Multiplexing brings the challenges that people (1) have to refocus their concentration each time they switch and (2) are more susceptible to distractions and errors. This is especially true if the tasks at hand involve intensive real-time control as opposed to being cognitively automatic. There is evidence that health care personnel are not aware of the risks of decreased performance when attempting to multitask.

Douglas and co-authors recently reviewed the current literature concerning multitasking in the health care setting. They suggest that multitasking typically results in increased time of task completion, increased stress, risk of memory lapses, and subsequent errors and accidents. Those performance limitations occur more often if one is forced to multitask by the environment and/or if the different tasks compete for the same local cognitive resource. Their review located only two studies on the association between multitasking and errors in the health care setting although nothing definitive was found. Executing two tasks at the same time was found to carry the risk of decreased accuracy and efficiency and to have an increased reaction time to environmental stimuli.

Multitasking During Handovers?

One study observed different modes of patient handover from the OR to the post-anesthesia recovery unit in six hospitals. The researchers compared handovers of (1) simultaneous transfer of equipment and information (i.e., a nurse connecting the monitor while at the same time receiving verbal information) to (2) sequential transfer (equipment is first connected and lines sorted, then verbal information follows). The findings from 101 observed handovers showed that 65% of them took place simultaneously. Interestingly, simultaneous handovers were not significantly faster than sequential ones (1.8 vs. 2.0 min). In review of postoperative handovers Segall and colleagues recommended: (1) standardize/structure processes (e.g., through the use of checklists and protocols); (2) complete urgent clinical tasks before the information transfer; (3) allow only patient-specific discussions during verbal handovers; (4) require that all relevant team members be present; and (5) provide training in team skills and communication. More information on handoffs is given in the later section on “Communication.”

Shared Situation Awareness

SA is a concept that can be applied to the individual (personal SA, see above), but also the team (shared SA). Team SA is defined as “the degree to which every team member possesses the SA required for his or her responsibilities” (p. 39). Within different professions and different physical positions in the room, the situational awareness and interpretation of the same case can differ substantially within a team, due to variation in experience, involvement, focus point, interest, knowledge, or information. Of course, not every bit of information can be shared with the whole team; doing so would overwhelm everyone. Yet, creating and maintaining an overall shared mental model of the situation across all team members is critical for determining the best plan of action for patient care. Some team communication in the operating room that seemingly does not serve an obvious purpose, might, in fact, be a gauging of shared mental models and therefore shared SA. Anesthesia professionals, for example, routinely assess the progress and possible complications of the operation by perceiving nuances in the conversation of the surgeons.

Shared mental models predict good team performance and were, for example, shown to be related to positive performance in trauma teams, with team leaders and followers actively working toward an information exchange. Adapted from psychology and management literature, Schmutz and Eppich introduced the conceptual framework of team reflexivity to health care. They propose that shared SA is fostered by components of team reflexivity, namely (1) pre-action briefing, (2) in-action deliberation, and (3) post-action debriefing. Fioratou and colleagues challenge the common model of individual and team SA and instead introduced the model of distributed situation awareness (DSA) to anesthesia, arguing that cognition actually involves elements of the (physical) environment, for example, the values on a monitor.

A multicenter study from Denmark explored shared mental models of surgical teams during 64 video-assisted thoracoscopic surgery lobectomies to explore familiarity of the people involved in the operation with each other; mutual assessment of technical and nontechnical skills in the other persons present during the operation; assessment of the perceived risk for the patient from the procedure and from the anesthetic; and noting of problems in the present co-works task management. There was poor agreement between team members’ risk ratings showing limitations in how much the members of the surgical team share a mental model about the patient and the related risks. A follow-up study demonstrated the connection between relevant clinical markers and shared mental models.

Other aspects of situation awareness have already been introduced within the “core cognitive process model” in an earlier section. Box 6.2 gives examples of reevaluation questions in order to maintain situation awareness.

Ambient Noise and Music

The workplace of anesthesia professionals—predominantly, but not exclusively, the operating room—is a very complex physical and cognitive setting. Unless there is an unusual effort to reduce sound levels the routine noises of the suction, surgical equipment (electrocautery, pneumatic or electric power tools), and monitoring equipment yield levels considerably higher than in most offices or control rooms. Other sound sources such as conversation and music are controllable. The potential interference of noise with communication and situation awareness among personnel in the OR is particularly worrisome to those concerned with optimizing teamwork in this complex work environment.

The use of music in the OR is now widespread. Many health care professionals believe that music enlivens the workday and can build team cohesiveness when all team members enjoy the music. Others note that in some cases the volume level of music makes it harder to hear the rhythm and tone of the pulse oximeter and alarms as well as work-related conversation between team members. A laboratory study by Stevenson and colleagues determined that visual attentional loads and auditory distractions additively reduced anesthesiology residents’ ability to detect changes in pulse oximeter tone.

A controversial study by two social psychologists, Allen and Blascovich, suggested that surgeon-selected music improved surgeons’ performance on a serial subtraction task and reduced their autonomic reactivity (i.e., “relaxed” them) when compared with control conditions consisting of experimenter-selected music or no music at all. The methodology of this study has been criticized. In response to Allen and Blascovich, several anesthesiologists challenged the notion that the surgeon’s preference for the type or volume of music can or should override the needs of other members of the team.

Murthy and colleagues studied the effect of OR noise (80 to 85 dB) and music on knot-tying ability in a laparoscopic skill simulator. They found no difference in time or knot quality in the conditions tested and concluded that surgeons can effectively block out noise and music. The invited commentary that accompanied the article brings up important issues: What impact does noise have on other members of the surgical team? How does noise affect communication between team members? Does noise affect judgment? The question of the proper role of music in the OR has no simple answer. Clearly, optimal patient care is the primary goal. Some surgical or anesthesia personnel explicitly forbid any type of music in the OR. A more common approach of many OR teams is to allow any team member to veto the choice or volume of music if they believe that it interferes with their work.

Reading and Use of Mobile Electronic Devices

The observation that some anesthesia professionals have been seen to read journals or books casually during patient care led to a vigorous debate of the appropriateness of such activity. Although it is indisputable that reading could distract attention from patient care, a study by Slagle and Weinger in 2009 suggested that when the practice of reading is confined to low-workload portions of a case, it has no effect on vigilance. Many comments about the issue were related not to the actual decrement in vigilance induced by reading but rather to the impact of the negative perception of the practice and of those who do it by surgeons and by patients, if they were aware of it.

This issue has been greatly magnified by now-ubiquitous smartphones and social media. Nearly all clinical personnel (other than those scrubbed into surgery) have a smartphone immediately at hand. These provide great temptation for both pull activities—reading email, websites, or social media sites—and push activities, notification of new text messages, mail, or social media posts. The wide variety of channels of information makes it far more likely for people to spend time interacting with their phones than with newspapers, journals, or books. The content available is never ending. In their survey, Soto and colleagues give a current update of the usage patterns, risks, and benefits of personal electronic device use in the operating room.