This page has been left blank intentionally Chapter 23 Measuring the Impact of Time Pressure on Team Task Performance Colin F. Mackenzie, Shelly A. Jeffcott and Yan Xiao Introduction and Background Surgery often contains crisis-like moments, even for scheduled, elective procedures. Understanding team performance under time pressure provides a basis for developing counter-measures. However, prospective investigation to study team performance under stress is difcult, due to the unpredictable nature of crisis events. Time pressure is a hallmark of trauma care where breakdowns in coordination can result in errors which threaten the life of a patient and where decisions are frequently made with high levels of uncertainty (Xiao et al. 1998). Trauma care thus provides a window for us to examine team performance under time stress. Under time pressure, problems with errors of commission and omission are particularly apparent when potentially life-saving decisions and actions must be carried out dynamically within a few critical minutes. For example, cognitive errors during emergency care were found to be a signicant contributor to patient harm. Factors limiting the clinicians ability to make timely and correct diagnoses include multiple, concurrent tasks; uncertainty; changing plans; compressed work procedures; large workload and complexity (Xiao et al. 1996). In comparison to teams in other domains, surgical teams are often ad hoc in nature for a number of reasons: in teaching hospitals, trainees rotate monthly through different services; trainee staff may even come from other institutions to gain experience; requirements for specialty services vary from patient to patient; nursing and anaesthesia care providers may have different stafng patterns than their surgical colleagues. Under time stress, teams may have to split to take care of multiple patients simultaneously. Such changing team composition adds to the challenge of providing optimal team performance under stress, because team members may have never worked together before. In addition, surgical teams usually do not have formal training in coordination. Richer understanding of team performance under time pressure can benet patient safety by assisting in team training development and can boost individual and team competencies in non-technical aspects of care, such as task prioritization, leadership and decision-making. The nature of anaesthesiology care for trauma patients makes it a rich test bed for team performance measurement since within Safer Surgery 386 the resuscitation and operating room environment, activities and decision-making must be coordinated between anaesthesia, surgery and nursing professionals, who all have their own group cultures and team dynamics. Team performance depends on the ability to coordinate members’ capabilities and efforts. Special skills are needed to synchronize individual members’ activities especially when up to one third of the trauma centre deaths are reported as preventable and many can be attributed to non-optimal communication and coordination (Fitzgerald et al. 2006). Several strategies are available to coordinate teamwork, increase compliance with ‘best practices’ and improve patient outcomes including using established work procedures or standardized algorithmic approaches to assist decision-making and reduce errors. For instance, the protocol of airway, breathing and circulation (i.e., ‘ABCs’), used in resuscitating cardiac-arrest patients, species goal and action priorities for resuscitation personnel (American College of Surgeons 1997, Mackenzie and Lippert 1999). In emergency medical care (as in other medical settings) extensive on-the- job training and experience provide personnel with the ability to anticipate the needs of others and what they will do next without explicit (i.e., verbal or gestural) communication. Team members share the same workspace and event space (i.e., what is happening) and have continuous visual and auditory contact with one another. As a result, they can sense what others are doing or intend to do. A ‘wait- and-see’ strategy, for example, can be used in avoiding potential conicts among several people who have access to the patient and equipment (Xiao et al. 1998); however, task complexity in emergency medical care can reduce the effectiveness of coordination strategies (Xiao et al. 1996). This chapter examines the impact of time pressure on trauma resuscitation team performance, by using tracheal intubation as a model to contrast task performance at two levels of task urgency, emergency and elective interventions, depending on the clinical circumstances. The task of tracheal intubation is very relevant to understanding anaesthesia activities during surgery, as it is a task carried out after induction of anaesthesia in the operating room. Tracheal intubation includes rendering unconsciousness (using anaesthetic and paralysing drugs if the patient is awake or semi-conscious) to stop breathing efforts or patient resistance and allow placement of an endo-tracheal tube through the mouth and between the vocal cords. The ‘airway management’ achieved by tracheal intubation is a life- saving intervention when executed correctly in an emergency. However, it requires signicant technical skill and has risks (vomiting and aspiration, detrimental changes in vital signs including cardiac arrest) that may even signicantly outweigh the potential benets of improved oxygenation and ventilation. The complications associated with task accomplishment may, themselves, become more life threatening than the problem that the intervention was intended to remedy. Several methods were considered to collect data to examine the impact of time pressure on tracheal intubation. Measuring the Impact of Time Pressure on Team Task Performance 387 Methodological Consideration of Studying Team Performance under Time- Pressure Direct observation is a valuable method for collecting information pertaining to team- and task-work performance when under time pressure. The structure and culture of healthcare organizations can act as a barrier to direct observation studies of task and team performance. However, provided free access, condentiality and consent issues are resolved, observation can identify the individual, team and organizational precursors to breakdowns in team communication and coordination that might be expected when a task is carried out under time pressure (Carthey 2002). Understanding the relationships between the processes of care, including environmental and interpersonal interactions and outcomes, is thus essential to developing a complete understanding of how task accomplishment is inuenced by time pressure. Research which adopts direct observation methods can be most useful, for example, for identifying recurrent and interrelated factors in communication during care delivery (Lingard et al. 2002, Shapiro et al. 2004). To adequately synthesize and collect data by observation, the observer should understand the care providers and team psychological support and cultural norms. The observer must also have knowledge about what is uncertain about the shared plan of action and why certain tasks take preference or need completion expeditiously. Difculty in understanding the situation in which trauma care specialists work is exacerbated by these intimate details that are often only gathered by being an experienced member of the team (Leonard et al. 2004). Christian et al. (2006) used observation to identify major systems features inuencing operating room team performance and carried out a quantitative analysis of factors linked to team performance and patient safety. Communication breakdowns, workload and competing tasks had a measurable negative impact on team performance. Observation of two different surgeries concluded that in both instances the results supported the concept that interventions designed to improve teamwork and communication may have benecial effects on surgical technical performance and patient outcome (Catchpole et al. 2008). Direct observation to measure aspects of surgical team performance has been carried out at a number of different levels of analysis. An example of a highly specic observation of an individual’s team skills, excluding the performance of other team members is called Observational Teamwork Assessment for Surgery (OTAS) (Undre et al. 2007, Healey et al. 2005). Undre and colleagues developed OTAS to simultaneously assess specic tasks among environment, equipment, provisions and communications by one observer. A second observer rates clinical behaviours against ve established components of teamwork: (1) cooperation; (2) leadership; (3) coordination; (4) awareness; and (5) communication among facets of the surgical process (Undre et al. 2007, see Undre et al. Chapter 6 in this volume). An assessment which attempts to account for the whole team is called the Anaesthetists’ Non-Technical Skills (ANTS) behavioural marker system (Fletcher Safer Surgery 388 et al. 2002, Fletcher et al. 2003b). ANTS comprises the following four skill categories: (1) task management; (2) teamworking; (3) situation awareness; and, (4) decision-making. It is the rst tool for non-technical skills training in anaesthesia and supports the need for objective measures of teamwork to appropriately inform real and simulated training initiatives (Fletcher et al. 2003a, see Glavin and Patey, Chapter 11, and Graham et al., Chapter 12 in this volume, for examples). More recently non-technical performance measurement has been carried out in surgical and operating room teams. Using task analysis assessment to identify technical errors and non-technical team performance measures for surgery suggests that it may be possible to avoid errors associated with lack of surgeon situational awareness by intra-operative briengs and workload management (Mishra et al. 2008). There is a role for both structured and unstructured observations to help capture and measure the complexity of team interactions. The rigour of these observations has been addressed by: employing strategies such as validating tools against best practice standards; using two independent observers to simultaneously collect data; and triangulating data using multiple data collection methods (Healey and Undre 2006). Additionally, further research is needed to ensure observation tools are robust and standardized, and, also importantly, remain consistent with current best practices (Undre et al. 2006). Carthey points out that direct observation, especially of the structured variety, may not be suited to all healthcare environments. Those environments which are dened by emergent rather than elective procedures – where there is unpredictable diverse case mix, larger size and the greater movement of staff around a wider area while treating patients – can create difculties for observers (Carthey 2002). It may be too great an attentional task to accurately record all teamwork communications due to the involvement of multiple actors and activities that can occur simultaneously (Wears 2000). Alternatives to Observation to Determine Time Pressure Effects Data collection is a major challenge, particularly in studies of emergency medical care because interesting care events are often unexpected. Real-time data collection is needed to counteract hindsight biases in retrospective construction of past events and to capture dynamically evolving emergency situations. Audio-video recording is an inuential tool that may assist such data collection. Those analysing video data can repeatedly examine activities in emergency medical and trauma care settings and extract detailed qualitative and quantitative data (Mackenzie and Xiao 2003). Additionally, high risk medical providers can review their own care through audio-video records and provide comments on covert mental processes cued by audio-video records. Such a cognitive approach to examination of real emergency medical events is a powerful tool to uncover many facets of the nature of emergency care work, including time pressure, clinician performance, diagnosis, Measuring the Impact of Time Pressure on Team Task Performance 389 planning, and communications during stressful tasks and identication of patient and practitioner safety. While observational studies illustrate types of errors and incidents in healthcare, they are not always effective at describing and analysing clinical conduct at the level of interactions within real complex clinical environments, such as detecting the inuence of time pressure on task performance in trauma settings. Use of video is more appropriate than observation as it overcomes many of the disadvantages that direct observation has by enabling capture of eeting events and communications and allowing repeated review and inter- rater reliability analyses of audio-video source material. Importantly video data allow the participant care providers to review and comment on the cognitive and uncertain aspects of the care that they provided, and the pressure they felt during task accomplishment. Video recordings can also be used in conjunction with direct observation to validate and supplement information collected by trained observers and other structured analysis. When video is compared to direct observation, video data can help overcome difculties in capturing subtle cues, eeting errors, brief utterances or team interactions and communications. There is a great need to understand what occurs in uncertain emergency medicine workplaces, where risky but life-saving procedures such as tracheal intubation are carried out, often in non-optimal circumstances (Mackenzie et al. 2007, Mackenzie et al. 1996b). Video data may be helpful to identify what Reason has termed, unsafe acts, pre-cursor events (Reason 1980), accident opportunities, latent and systems failures that traditional observational data collection methodologies may fail to capture. Video based analyses in complex and high-risk medical settings offers advantages over direct observation since it allows for: (1) ne grained analysis; (2) repeated reviews by multiple experts; (3) the ability of multiple observers without compromising access to the patient; and, (4) playback for clinicians whose care is recorded. Feedback to clinicians viewing their own care is potentially a most valuable advantage of video, providing systematic feedback and allowing team performance review. Such video review has the potential to advance the incorporation of non-technical skills training into healthcare and allow objective assessment of these skills (Moorthy et al. 2005). Also, in support of arguments elsewhere in this chapter, video records can identify problems and solutions to improve processes in emergency trauma care that are likely to be generalizable to other high risk medical domains (Mackenzie and Xiao 2003, Mackenzie et al. 1996b). Video data can also be used as input to root cause analyses or other quality improvement practices, and provide contextual detail, enable repeated raw data reviews by multiple reviewers and allow assessment of a critical event using measures developed subsequent to the video-recorded event. In this way, video analysis can be a powerful supplement to existing quality assurance methodologies for understanding the context and the proximal contributors to an adverse event. Safer Surgery 390 Data Collection to Assess Time Pressure Effects during Tracheal Intubation Task Audio-video recording was determined to be the optimal method to collect data on tracheal intubation. Before starting video data collection, non-structured interviews with subject matter experts (SME) among anaesthesiology faculty attending staff, were carried out to gain consensus about the task requirements and priorities. Interviews were conducted both at the trauma centre and a tertiary care hospital where mostly elective surgery was performed. From these interviews, some of which were audio-recorded and transcribed, the normative task model for tracheal intubation was derived and conrmed by SME’s review. Capture of audio-video records from cameras and microphones installed on the ceilings of trauma resuscitation bays and operating rooms of a trauma centre provided data to identify details of the task performance. Patient vital signs including heart rate, blood pressure, blood oxygen saturation and exhaled carbon dioxide were continuously recorded. Major components of video analysis included performance evaluation using the normative task analysis model as a performance template and completion of questionnaires based on this task model by SMEs during review of the video records. The questionnaires required completion of timing data for major task landmarks, subjective ratings of performance in tactic manoeuvres and decisions, and judgement of patient conditions. Aggregated results from the completed questionnaire produced the basis for further, often quantitative analysis. Two types of task settings were emergency and elective, which reect the urgency of decision-making and interventions. For tracheal intubation emergency was dened as within ten minutes of patient admission, identifying an urgent need to intervene rapidly to normalize oxygenation and ventilation. Semi-emergency intubation occurred 11–59 minutes after admission. Elective tracheal intubation included those newly admitted trauma patients who required airway management one hour or more later. Findings Obtained from Video Analysis Illustrating Effects of Time Pressure To measure different effects on performance of the tracheal intubation task under elective and emergency circumstances two experienced faculty anaesthesiologists were instrumented with ambulatory vital signs recording devices (automated blood pressure cuff [BP], electrocardiogram [ECG]). An example of the ECG tracings and BP results is shown in Figure 23.1 which shows the ambulatory electrocardiogram (ECG) and blood pressure (BP) of an anaesthesiologist during elective and emergency intubations. During the elective intubation (top panel) blood pressure and heart rate are normal. The same anaesthesiologist two hours later dealing with an emergency intubation has a greater than doubling of heart rate (lower panel) and clinically signicant diastolic hypertension. Measuring the Impact of Time Pressure on Team Task Performance 391 Figure 23.2 shows heart rate/minute (HR) and blood pressure (BP) of an experienced anaesthesiologist obtained by ambulatory monitors. Changes are recorded in BP and HR with increased physical exertion (d), during resting (c), and during elective intubation (a × 4) and emergency intubation (a/b) of a trauma patient. There are clinically signicant increases in heart rate and systolic BP seen with emergency intubation (in comparison to elective intubation). These data show that the anaesthesiologist, despite being experienced, showed physiological signs of stress due to the time pressure of an emergency rather than elective tracheal intubation. The results of the task analyses of tracheal intubation are shown in Table 23.1, decomposed into a sequence of steps, in preparation for, during and after task completion. Figure 23.1 Ambulatory electrocardiograms (ECG) and blood pressure (BP) of an anaesthesiologist during elective (top panel) and emergency (lower panel) intubations Safer Surgery 392 Among emergency tracheal intubations, there was associated increased workload (many tasks to do at once), time stress both due to the severity of the patient injury and on occasion due to the surgeon calling for induction of anaesthesia and tracheal intubation, so that other tasks in the resuscitation process could proceed more rapidly. Compared to elective tracheal intubation, there was also more uncertainty of information about patient status because more than half of the emergency intubations were required due to unconsciousness or patient-induced challenges such as combativeness or intoxication. Errors in emergency intubation could be attributed to knowledge-based shortcomings, for example, drug dosages were erroneous in three emergency intubations (Mackenzie et al. 1996a). Even though the majority of video recorded tracheal intubations (23/42) were performed by experienced personnel (>5 years in trauma centre), there were a greater number of tasks omitted from the task analysis model in emergency than elective tracheal intubation. Figure 23.3 shows the comparison of task omission (among those tasks shown in Table 23.1) during elective (EL), semi-emergency (SE) and emergency (E) airway management in 42 patients of whom 15 showed emergency, 13 semi- emergency and 14 elective airway management. Twelve SMEs prioritized the tasks that were checked during airway management. On video analysis a low weighting was given if a high priority task was omitted. Omission of a low priority task considered to be task shedding was given a high weighting that would have Figure 23.2 Heart rate (HR) and blood pressure (BP) of an experienced anaesthesiologist obtained by ambulatory monitors (Holler) Measuring the Impact of Time Pressure on Team Task Performance 393 the effect of reducing the frequency of task omission. Despite weighting (grey background), there were still a greater number of tasks omitted in E than SE or EL airway management. Weighting decreased the percentage of tasks omitted in all three categories of airway management, indicating some task shedding occurred. Preparatory tasks Pre-oxygenation by facemask Head positioning Cricoid pressure applied correctly In-line neck stabilization Suction ready SpO 2, BP, HR all monitored pre-induction of anaesthesia and muscle relaxation IV running Drugs drawn up Stethoscope at hand A ssistance immediately available During Intubation equipment ready Re-oxygenation after 3 attempts Modication of technique between attempts Re-oxygenation if SpO 2 <95% Cricoid pressure maintained until cuff sealed Tube insertion distance checked Auscultation of both sides of chest by intubator Auscultation of upper abdomen After Listen to chest after ventilator connection CO 2 monitored within 2 mins of intubation Check neuromuscular blocker before non-depolarizer Check ventilatory parameters Table 23.1 Task sequence tracheal intubation where X = cross, SpO 2 = O 2 saturation, BP = blood pressure, HR = heart rate, IV = intravenous, CO 2 = carbon dioxide . measurement since within Safer Surgery 386 the resuscitation and operating room environment, activities and decision-making must be coordinated between anaesthesia, surgery and nursing professionals,. stress. Under time pressure, problems with errors of commission and omission are particularly apparent when potentially life-saving decisions and actions must be carried out dynamically within. called the Anaesthetists’ Non-Technical Skills (ANTS) behavioural marker system (Fletcher Safer Surgery 388 et al. 2002, Fletcher et al. 2003b). ANTS comprises the following four skill categories: