Safer Surgery 134 attributes and characteristics associated with leaders are requisites of followers as well, referred to as good leadership or as good followership. Van Vugt (2006), who evaluated vast psychological literature on leadership as a database to test several evolutionary hypotheses about the leadership and followership in humans, found that leadership correlates well with initiative taking, trait measures of intelligence, specic task competencies, and several indicators of generosity, but he found no link between leadership and dominance. Burke and co-workers (2006, p. 302), who conducted a meta-analysis to examine the relationship between leadership behaviour in teams and behaviourally based team performance outcomes found that: Surprisingly, the results of the current literature review indicate the preponderance of empirical research conducted to date to examine team leadership has largely been grounded in traditional leadership theory. In other words, researchers are just transporting traditional theories of leadership to team settings. This is a concern because classical leadership theories have often been criticized for failing to fully appreciate and model the dynamism and complexities of team leadership (Kozlowski and Bell, 2002, Salas, Stagl and Burke, 2004, Ziegert 2004). In fact, most existing leadership theories are advanced as if ‘leader– follower relationships exist in a vacuum’. (House and Aditya 1997, p. 445) As current leadership theory and research predominantly refers to leader style, Avolio (2007) argues that future work needs to consider the dynamic interplay between leaders and followers. Presumptions about leadership and followership – such as leaders are more important than followers, followers are only doing what they are told, followers get their energy and aims from the leader – should be challenged also in theory and research regarding medical teams. Our general working hypothesis is that all team members are important for successful goal achievement and that the roles and interaction between sub-teams, professions, specialities, leader and followers have to be further explored in relation to patient outcome. Our specic working hypothesis is that training effectiveness will be better when these roles and interactions are identied for each sub-team, profession, speciality, leader and follower and used as targets in team training. In order to further elucidate the individual team members’ contribution to the collaborative teamwork and the interpersonal interaction between team members, we need an observational behaviour instrument that allows for rating all team members’ team skills. Leadership and Followership; Attitudes in Teamwork We consider leadership/followership, to be a signicant element of teamwork, in two aspects: attitudes and behaviours. Dickinson and McIntyre (1997, p. 25) explain team leadership: Involves providing direction, structure, and support for other team members. It does not necessarily refer to a single individual with formal authority over A-TEAM 135 others. Team leadership can be shown by several team members: explains to other team members exactly what is needed from them during assignment, listens to the concerns of other team member. Thomas et al. (2004, p. 160) comment, regarding observation of neonatal resuscitation, that: Leadership activities may include sharing of a mental model, assigning tasks, and sharing of information and opinion. This may be rated for any provider at the resuscitation. There is usually not a clear leader (either in deed or word), so the observability rating is often 0. Lack of leadership was not obviously detrimental to the process of care. Thus in a successful team, all team members, not only the formal leader but also the followers, embrace positive team attitudes such as team orientation, conict efcacy, shared vision, team cohesion, interpersonal relations, mutual trust, task- specic teamwork attitudes, collective orientation and importance of teamwork. Such leadership or followership attitudes facilitate a culture of non-negotiable mutual respect and trust so that effective teamwork can ourish. Leader Behaviour and Follower Behaviour The reasoning regarding leader and follower attitudes does not apply to the issue of leader behaviour and follower behaviour. In contrast to attitudes, if several members of a team simultaneously apply some of the identied ‘leadership‘ behaviours, there is an obvious risk of confusion and conict in the team, which jeopardizes team structure and process. A leader is recognized by his/her leader behaviour which might include taking the initiative to prompt discussions, making decisions, verbalizing plans and having the ‘last word’. A situation when a follower simultaneously behaves in that manner, or shows other leader behaviours, will confuse other members as to who is the leader. Tension arises and a prestige conict between team members might result which undermines, rather than enhances, the leader’s position, and demoralizes the team. Obviously, regarding certain behavioural elements there is a reciprocal relationship between leader behaviour and follower behaviour. Another recognized behavioural element, such as good communication, is identical for all team members. In agreement with Thomas et al. (2004), we seek behavioural markers that allow the rating of any team member’s team skills, whether as leader or as follower, or just as a good team member. Accordingly, in agreement with Murray and Foster (2000) and Ostergaard et al. (2004), we advocate using specied leader behaviours, follower behaviours and member behaviours as targets for team training to emphasize the complementary behaviours and to facilitate individual feedback to each individual team member. Safer Surgery 136 Behaviour markers for healthcare, many of them based on crew resource management (CRM) used in aviation (Gaba et al. 1994, Flin et al. 2003, Helmreich et al. 1999) and adapted to medical settings, such as ACRM (Anesthesia Crisis Resource Management; Howard et al. 1992), Revised teamwork behaviour matrix (Small et al. 1999), Team Dimensions Rating Form (Morey et al. 2002), OTAS (Observational Teamwork Assessment for Surgery; Carthey et al. 2003, Healey et al. 2004), ANTS (Anaesthetists’ Non-technical Skills; Fletcher et al. 2003, Flin et al. 2004a), EMCRM (Emergency Medicine Crisis Resource Management Behavioural Performance Evaluation; Reznek et al. 2003, Wallin et al. 2007), CARMA (Crisis Avoidance Resource Management for Anaesthetists; Flin et al. 2004b), BARS (Behaviourally Anchored Rating Scales; Shapiro et al. 2004), NOTSS (Surgeons’ Non-Technical skills in the operating theatre; Flin et al. 2006, Yule et al. 2008), Ottawa GRS (Ottawa Crisis Resource Management Global Rating Scale; Kim et al. 2006), CATS (Communication and Teamwork Skills; Frankel et al. 2007), are focused on the team leader alone (i.e., the anaesthetist or the surgeon) and/or the team (e.g., the crew). Only the instrument designed by Thomas et al. (2004) – University of Texas behavioural markers for neonatal resuscitation (UTBMNR) – is explicitly constructed for summative assessment of any team member’s team skills. In summary, we need a behaviour-based assessment instrument suitable for gauging all team members’ teamwork skills on an individual basis that is widely applicable and reects good practice across the healthcare professions, disciplines and medical environments, and that can be used as target for training, formative assessment for feedback and for summative evaluation of training. Description of the A-TEAM Scale Development of the A-TEAM Scale We used the decision loop (Figure 9.1) as a template for four behavioural categories: ‘Gathers information and communicates’, ‘Contributes to shared understanding’, ‘Makes collaborative decisions’, ‘Coordinates and executes tasks’, and added a fth category, ‘Takes a team member role’, which has three subcategories: ‘All members’ behaviour’, ‘Leader behaviour’ and ‘Follower behaviour’. From our experience these categories are easy for trainees to comprehend. We scanned the literature and sampled verbally dened behavioural elements from other scales and CRM programmes as generic behaviours of effective collaboration (Burke et al. 2004, Carthey et al. 2003, Dickinson and McIntyre 1997, Fletcher et al. 2003, Flin et al. 2004a, Flin and Maran 2004, Flin et al. 2006, Frankel et al. 2007, Healey et al. 2004, Howard et al. 1992, Kim et al. 2006, Moorthy et al. 2006, Morey et al. 2002, Morgan et al. 2007, Murray and Foster 2000, Ostergaard et al. 2004, Ottestad et al. 2007, Reznek et al. 2003, Shapiro et al. 2004, Small et al. 1999, Thomas et al. 2004, Wallin et al. 2007, Undre et al. 2007). We selected A-TEAM 137 behavioural elements meeting two criteria: a trainee should, voluntarily, be able to display the behaviour at work, and the behaviour should be observable for trainees. Behavioural elements related to the four categories – ‘Gathers information and communicates’, ‘Contributes to shared understanding’, ‘Makes collaborative decisions’, ‘Coordinates and executes tasks’ – were sorted to respective categories. In order to sort behavioural elements into the leader, follower and member sub- categories, we used our clinical experience of working in anaesthesia, intensive care and emergency medicine, and experience from acting as ‘consultant on call’ in hundreds of simulator scenarios with younger medical colleagues and nurses. The intention was that a senior consultant would stay in a follower position to give a second opinion and support to a younger, less experienced colleague. Behaviours by the consultant that immediately would dislodge this balance were categorized as leader behaviour. Examples of these behaviours were taking initiative to provide briengs, decision-making, conrming and verbalizing plans, and ‘having the last word’. Behaviours that stabilized the relationship, yet allowed use of all expertise available, were categorized as follower behaviours. These behaviours were such as giving the junior colleague suggestions to take initiative for briengs and decision-making, suggesting that he/she verbalize plans and let him/her ‘have the last word’ in order to support him/her as leader. The vast majority of behaviours sampled were both leader and follower behaviours and as such categorized as team member behaviours. For each category, we condensed the vast number of sampled behavioural elements so that the sum of them would be limited and not tax the observer’s, trainee’s or trainer’s, limited working memory, and hence easy to use as target for training, formative assessment and feedback. Using A-TEAM as Target for Training, Formative Assessment and Feedback The A-TEAM form should be presented and discussed with trainees and demonstrated by trainers before practice. During practice, non-active trainees can observe and grade active participants using an A-TEAM protocol, which enhances learning. All elements are directly observable; the text is short and in simple language. During feedback, trainers have the opportunity to further elucidate behaviour issues related to coordination using A-TEAM. During training, observers and trainers use the following four grades for classifying team member behaviours: Poor: Behaviour that intentionally counteracts desirable team behaviour. In need of considerable improvement: Behaviour that neither counteracts nor fully contributes to desirable team behaviour. Good: Behaviour that really helps in solving the task and strives towards desirable team behaviour, but is not excellent. Procient: Procient behaviour, role model. • • • • Safer Surgery 138 The result of the development process is presented in Table 9.1. There are ve main categories of behaviours, and the rst, ‘Takes a membership role’, is divided into three subcategories. Each category and subcategory is dened separately. Examples of behavioural elements representative for each category are given for four grades in columns, read from left to right with increasing prociency. The grades ‘Good’ and ‘Procient’ are used as targets for training, while all grades are used for feedback and assessment of training outcome. During debrieng, advantages and disadvantages of using leader and follower behaviours can be discussed in relation to conicts and outcome for the (simulated) patient. Future research may elucidate in which situations leader behaviour, follower behaviour and member behaviour are of particular importance for patient outcome and integrated in training. Using A-TEAM for Summative Assessment All elements to be assessed for each category are directly observable, so no further assumptions/interpretations have to be made by the observers. Active verbs are used for describing observable behaviours. Global rating of an individual’s team skills and/or global rating of the complete team performance can also be registered if needed. Scoring Method in Summative Assessment For feedback during training, it is easier to ‘pinpoint’ the most important behaviours observed using four grades. However, for evaluation of training and for research purposes, it is statistically correct to handle these ordinal data as binary data. All raw data are converged into the following two stages: 0 = Poor and in need of considerable improvement. 1 = Good and procient. In summary, the A-TEAM instrument is designed for both training and research purposes. All team members are scored in terms of the occurrence and quality of certain behaviours. Hence, it will assist team performance assessment in a variety of real and simulated clinical events and can be used in the study of the relationship between the teamwork process and teamwork outcome, as well as for feedback during training of instructors and trainees. People with quite different educational backgrounds should be able to use the scale. Lipshitz (2005, p. 370) argued that pure factual reports: ‘without interpretation are either impossible or not meaningful’. Human beings do not simply perceive, but elaborate and interpret information and reect upon it (cognition) and so the observer is an interpreter of raw data. By using video recordings, good observational data can be obtained and the observer takes advantage of the serendipity. Discrepant observations are not A-TEAM 139 Table 9.1 The A-TEAM scale for assessment of individual team behaviour 1. Takes a team member role All members’ behaviour A team member participates actively in all relevant aspects of teamwork and task work, observes other team members’ performance and activities, and provides and accepts feedback and assistance. Poor I n need of improvement Good Procient Does not give or welcome any support. Does not acknowledge other members. Work on his/her own. Does not give or take advice gracefully. Support others only if challenged, not spontaneously . Supports in a reluctant way. Supports others, e.g., by helping when obviously needed. A ccepts and gives support directly. Acknowledges and involves others frequently . Asks quickly for help when needed. Protests clearly against inaccuracies. Leader behaviour A team leader takes initiative to provide structure and direct team- and task work. Poor In need of improvement Good Procient Does not identify himself/herself as person in charge of the situation. Makes no nal decisions. T ries to ‘do it all’. Does not take initiatives. Tries to make decisions and give orders but very unclear and vague. Takes initiative, but not in time when needed. Makes decisions, but not clear enough. G ives orders, but not directed. Makes nal decisions. Gives clear orders. Takes initiatives to e.g., short briengs; conrms and verbalizes decisions. A ccepts a non- leading role when appropriate. Follower behaviour A team follower supports the leader’s initiatives, and assumes assigned responsibilities and directives. Poor In need of improvement Good Procient Challenges the leader in an obstructive way. Stands back, takes a ‘hands-of f approach’. Demonstratively does not participate in briengs. Performs task duties, but only on demand. Does not support the leader verbally or in other ways. Supports the leader both verbally and in other ways. Takes a ‘hands-on approach’. Supports the leader. Takes a ‘hands on’ approach. C hallenges constructively . Requests and participates actively in briengs. Takes over leadership if required. Safer Surgery 140 2. Gathers information and communicates All team members actively gather and exchange information. Poor In need of improvement Good Procient Does not communicate at all. Masters through oral or body language. E ngages in unnecessary conversation. Makes inappropriate comments. C ommunicates vaguely, quietly or continuously without pause. Communicates in round terms, but not aimed at a team member. Does not give feedback. Communicates in round terms. Uses closed-loop communication. Uses SBAR format for briengs. Calls out critical information during emergent events. Makes good eye contact. U ses team members’ names. 3. Contributes to a shared understanding of the situation In collaboration, all team members develop and maintain a common understanding of the situation, and have team situational awareness. Poor I n need of improvement Good Procient Does not share any information. Demonstratively does not contribute. Reports ndings unclearly. Gets distracted by non-essentials. Acts independently on symptoms without conrming with others. Reports ndings (positive and negative) in a clear and concise way. Does not try to explain things to others when obviously needed to. I nterprets information in a timely manner. Shares information and explains core events, thus making them understandable to the whole team. 4. Makes collaborative decisions In collaboration, all team members consider options, resources and risks to be able to take decisions. Poor I n need of improvement Good Procient Dees any attempts to decision and does not state an alternative. Demonstratively does not take part. Makes decision right away without discussing alternatives with the team though time and opportunity exists. C ontributes with ideas, but does not state any own opinion. R e-evaluates, discusses and takes other alternatives in consideration with other team members before reaching a conclusion. Actively drives the decision process forward. Table 9.1 Continued A-TEAM 141 always unreliable observations reducing the validity and reliability of ndings, but could be signs of originality and perceptiveness of the observer. As in most true discoveries, the observer can shift his/her focus from originally dened data collection to also include observation of new phenomena. Since the locus of exploratory observational studies is on discovery, several techniques should be used to broaden the set of observations and maximize the observers’ opportunity for uncovering interesting data. Roth and Patterson (2005, p. 382) have exemplied some good techniques: (i) broadly sampling domain practice (e.g., multiple trainees at several different levels of experience, many test sites), (ii) using multiple converging techniques (e.g., eld observations, interviews, questionnaires), and (iii) using several observers (who bring different conceptual frameworks for observation and interpretation). In future studies, A-TEAM could also be used in exploratory observational studies as contrasted with studies designed to test specic research hypotheses on human decision-making in complex medical domains. Application in the Operating Theatre Optimal surgical performance includes both technical and non-technical aspects. Regarding the latter, the equal importance of leadership and followership challenges the current dogma with respect to leadership. We outline a new hypothesis regarding the teamwork process in the operating theatre and identify the equal importance of leadership and followership, as well as the ability to engage in reciprocal rotation of these functions depending on the situation. There 5. Coordinates and executes tasks A team member coordinates his/her tasks in a timely and integrated manner with other team members’ activities, facilitating the performance of other members’ jobs. Poor I n need of improvement Good Procient Does not coordinate at all. Acts on his/her own. Stops others. Carries out inappropriate courses of action. Does not adhere to guidelines. Obstructs but does not stop. Does not change plans despite new information. C oordinates, but not for the whole team. Keep things progressing, but not in a clear and efcient way. Coordinates his/her activities with others. A dapts to any changes in the present situation. Prioritizes tasks. Tells others of plan for further care. A dheres to guidelines. Table 9.1 Concluded Safer Surgery 142 is a clear need for more awareness of these metacognitive aspects in the surgical community. Patient safety on the agenda facilitates this evolution. The A-TEAM instrument could be applicable in many different surgical challenges from simple surgical procedures to demanding and complex tasks such as liver transplantation, cardiothoracic surgery and spinal surgery where system failure is deleterious. The results should also be coupled to traditional performance variables and patient outcome. Regarding inability or failure of followership in the operating room (OR), Lingard et al. (2004a), for example, observed a failure in 30 per cent of communication events during surgical procedures. Of these failures, 36 per cent had observable consequences such as delay, tension among team members or procedural error. These results are supported by another observational study (Wiegmann et al. 2007) focusing on the effects of disruptions of the surgical process (e.g., communication failures, equipment problems). Their study found that surgical errors increased signicantly with increased disruptions and that teamwork and communication problems were the strongest predictors of surgical errors. Communication is a very important causal factor in errors made in surgery (Gawande et al. 2003). An observational study in paediatric cardiac and orthopaedic surgery found that effective teamwork was associated with fewer minor problems per operation (i.e., negative events that were seemingly innocuous but many of which contributed to major problems), higher intra-operative performance (i.e., fewer key operating tasks were disrupted) and shorter operating times (Catchpole et al. 2007). Achievement of medical goals and patient safety are the main objectives for teamwork in medicine. In the operating theatre patient outcome can be quantied in mortality, morbidity, time to recovery, absence of iatrogenic injury i.e., a technically successful surgical procedure, time of low oxygen saturation etc. The team’s task work gives information whether the team has executed procedures, algorithms and routines as planned and expected. Task work can be related to patient outcome to provide information whether team performance is effective. The individual, interpersonal teamwork behaviour gives information on how a team member contributes to coordination of task work and to internal team structure and process. Assessment of team skills is necessary for feedback during training and as such a formative assessment. The aggregate of all team members’ behaviour gives information on the teams’ internal effectiveness and can be related to task execution (Lingard et al. 2004a, Catchpole et al. 2007, Ottestad et al. 2007) and to patient outcome (Catchpole et al. 2007, Wiegmann et al. 2007). Further analysis of the individual, interpersonal teamwork behaviour can provide information whether a certain subteam of members, such as surgical (Moorthy et al. 2005, Undre et al. 2007), anaesthetic (Undre et al. 2007), nursing (Undre et al. 2007), leader or followers, or a specic leadership style (Cooper and Wakelam 1999, Entin and Serfaty 1999, Klein et al. 2006, Marsch et al. 2005, Tschan et al. 2006), have favourable impact on task execution and patient outcome. Although team performance is a collective effort, there is lack of information on the impact of the quality of followers’ behaviour A-TEAM 143 on teamwork, team performance and patient outcome. Observational studies of teamwork have identied patterns of communication, coordination and leadership that support effective teamwork. However, only a few studies could establish a direct link between specic teamwork behaviours and clinical performance or patient outcome. Communication Patterns Supporting Effective Teamwork Among the most prominent themes in communication research in healthcare are the effects of interruptions and tensions on effective team functioning. For example, ethnographic studies showed that tensions in team communications in the operating room often evolve around the issues of time, safety and sterility, resources and work roles (Lingard et al. 2002 and 2004b). In medicine, CRM and simulation traditionally emerged within anaesthesia and intensive care. Human reliability from a team perspective is the focus. Within the different elds of surgical intervention, until recently, simulation was most commonly focused on technical skills i.e., manual dexterity of the operating surgeon. High delity and full procedural simulation in image guided surgery have clearly emerged as contemporary state-of-the-art training tools which are being embedded into curricula and as part of certication procedures in surgical education and training. Since metrics are inherent in image guided surgical simulators, the trainee gets evidence-based feedback and learning curves as well as prociency gains can easily be monitored. Extensive validation studies are convincing and demonstrate transfer of skills acquired in surgical simulators to real surgery performed on patients (Ahlberg et al. 2007, Seymour et al. 2002, Grantcharov et al. 2004). Factors underpinning technical performance such as visual spatial abilities, working memory and computer game experience have also been demonstrated (Enochsson et al. 2004, Hedman et al. 2006 and 2007, Kolga Schlickum et al. 2008). However, several initiatives now also focus on non-technical OR skills of surgeons and the whole surgical team (e.g., Moorthy et al. 2006). This evolution transforms simulation towards systems orientation and points out that the result of surgery is neither related to patient pathophysiology nor the surgeons’ technical skills alone but rather to the competence of the whole perioperative team. Traditionally, the OR team consists of the surgeons (attending staff and residents), the scrub nurse, the anaesthetist, the anaesthesia nurse and nurse assistants. As in all high hazard elds, it is clear and evident that communication skills are crucial for outcome. Increasing clinical complexity requires a reciprocal increase in communication skills not only within professions and disciplines, but also across professional culture barriers. A multidisciplinary and transprofessional approach is needed. As in other high risk industries, surgery requires a minimum amount of system redundancy. One major challenge is the unpredictable nature of surgery and the ability of the whole team to be able to shift from working in a routine mode to a state of emergency accommodating rapid and efcient . on’ approach. C hallenges constructively . Requests and participates actively in briengs. Takes over leadership if required. Safer Surgery 140 2. Gathers information and communicates All team. complementary behaviours and to facilitate individual feedback to each individual team member. Safer Surgery 136 Behaviour markers for healthcare, many of them based on crew resource management. desirable team behaviour, but is not excellent. Procient: Procient behaviour, role model. • • • • Safer Surgery 138 The result of the development process is presented in Table 9.1. There are ve main