Teamwork and learning: Two fundamental processes for safety
John S. Carroll ,
19 Sep 2014
Safe operations in organisations require both a formal managerial system and informal practices that enact and support the system. The formal aspect consists of a set of policies, procedures and practices often summarised under the heading of safety management systems.
- Safety policy
- Safety resources and responsibilities
- Risk identification and mitigation
- Standards and procedures
- Human factors based system design
- Safety training
- Safety performance monitoring
- Incident and problem reporting and investigation
- Information sharing internally and externally
- Continuous improvement
- Management of change
These formal aspects are necessary for safety performance, but not sufficient. To bring them to life in daily operations, leaders play a crucial role through their commitment to the principles of safety management and their efforts in implementing the formal systems in ways that allow everyone to perform their work effectively and safely. Two basic indicators for leaders to focus on in these efforts are well-performing teams and a general capacity to learn. Only by capitalising on the resources of everyone in the organisation for routine operation coping with exceptional situations and continuous learning can safety be assured and sustained.
Leaders support teamwork and learning by appreciating some general principles as well as the specific situation. Examples of contextual factors that strongly influence the way teamwork and learning should be implemented and promoted are the following:
- Tasks: Task complexity, task interdependencies, personal and process risks
- People: Formal qualification, demographic and psychological diversity
- Organisation structure: Distribution of decision authority, specialisation
- Technology: Degree of automation, technical coupling of work processes
- External relationships: Supply chain dependencies, regulatory regime
Furthermore, an overarching concern is the chosen approach for handling risk and the underlying uncertainties. While reducing uncertainty appears to be the most straight-forward approach, there are industries and work domains where uncertainty has to be retained and acknowledged or even increased in order to achieve an appropriate balance between stability and flexibility. Even in a nuclear power plant, where high levels of control and stability are needed, there is also innovation to continually improve technical equipment and organisational processes. Teams thus need to be capable of maintaining the highest quality in very routine work, but also of handling uncertainty during technical and organisational change. Similarly, there is a duality of learning, where in routine situations people's alertness to opportunities for learning needs to be fostered and in more complex situations, people may require support to not be overwhelmed and to trust their ability to learn.
Following, several principles for teamwork and learning as well as some considerations concerning the impact of contextual factors in managing safety are outlined.
Teamwork requires that people bring their knowledge and abilities to bear on solving tasks that are too complex for any single person to handle. By teamwork we include not only small face-to-face work teams that share sets of tasks but also larger organisational units that coordinate interdependencies across departments and other organisational boundaries. For this to happen,
- the team composition needs to reflect the capabilities needed for the tasks at hand
- team members must have a common understanding of what it takes to accomplish the task,
- team members need to feel safe to share their knowledge and speak up if necessary,
- team leaders have to be open to shared leadership as a way to make the best use of team member's resources
- the team is able to adapt to changing situations.
For most tasks, team composition needs to reflect the heterogeneous capabilities and viewpoints required for task accomplishment. Diversity concerning education, gender, age, cultural background or personal history can trigger stereotypes and personal conflict that may impede knowledge sharing. The benefits of diversity usually only emerge once the team has had sufficient shared experience to get to know each other as individuals. Often in high-risk teams this process is particularly difficult because teams come together in an ad hoc fashion and have to perform immediately, such as medical teams in large hospitals, aircrews or collaboration among traffic controllers and train drivers. Such teams must know how to build dependable relationships very quickly. Protocols for team briefing and debriefing can support this process.
In order to effectively cooperate, team members need a common understanding of task requirements, their respective roles, and relevant situational contingencies. If team members share the same professional training it is easier to build such a common understanding. On the other hand, if team members are too similar they may develop a shared, but wrong perception of situations, leading to premature agreement and faulty decision-making. Balancing the advantages and pitfalls of diversity can be facilitated by training, procedures, and practices that allow probing each other’s viewpoints. For example, many hospitals have established tumour boards, a multi-disciplinary team, to discuss the diagnosis and treatment of cancer before the medical specialists in charge make a decision.
The sense of feeling safe – also called psychological safety – implies the belief that it is acceptable or even desirable to voice concerns or admit mistakes. This is a crucial prerequisite for knowledge sharing and learning in teams. Team leaders can create psychological safety if they signal that they can be trusted and that they welcome all contributions by team members regardless of their status. Encouraging speaking up is important especially in complex and safety-critical tasks where premature decisions can have devastating consequences. Team interventions in support of psychological safety should reflect that reasons for not speaking up may vary among individuals and occupational groups. For cockpit crews, for instance, the desire to maintain a good team climate is important, while cabin crew members fear punishment. Moreover, lower status team members more often remain silent because they believe that speaking up will not make a difference.
Leadership should be shared, in that leadership functions are fulfilled not only by the formal leader, but by any team member based on individual competencies and situational requirements. For instance in effective medical teams and crisis management teams, it has been found that leaders navigate their teams through rapidly changing circumstances, delegating leadership tasks and taking them back as required by task demands. For flexible leadership to work well, it is important for everyone to have a clear understanding of who is responsible overall and what different capabilities exist in the team. Leadership becomes even more complex when it is not single teams that operate in high-risk environments, but rather multi-team systems, where leadership within and between teams needs to be distinguished. In such settings, the role of boundary-spanner becomes crucial, as for instance the purser in aircrews.
Teams need to be able to respond to different situational requirements by means of adaptive coordination, using both explicit and implicit coordination as appropriate. Explicit coordination is the deliberate and resource-intensive establishment of formal roles and responsibilities embedded in job descriptions, procedure manuals, and training, which then guides team decision-making. Implicit coordination, on the other hand, relies on shared assumptions and knowledge about the team, the task and the context, enabling team members to coordinate their action in an effortless manner with few demands on information acquisition and assessment. Because implicit coordination is less resource-intensive, it is considered most suited for very demanding tasks. However, if these tasks entail unexpected elements, explicit coordination may become necessary, indicating a delicate balance between stability and flexibility of team functioning. Heedful interrelating, that is the very deliberate taking of others' perspectives during action planning and execution, has been found to be effective in bridging these conflicting demands.
The modern world of work is infused with continual innovation and change, which therefore requires learning new capabilities for individuals, teams, and organisations. By learning, we refer to an increase in the repertoire of behaviours. As individuals move across jobs and companies, leaders are replaced, work is outsourced to contractors, companies are acquired and divested, surprises (accidents, near-misses, missed goals, falling behind competitors) occur, regulatory and legal requirements are rewritten, and new technologies are introduced, each individual and organisation is engaged in exploiting their current capabilities while exploring new (and uncertain) possibilities.
We know that organisations as well as individuals learn most commonly from repetition, or trial-and-error. Each failure is a test of existing knowledge and therefore an opportunity to improve, innovate, and coordinate better. Part of that learning from experience is learning from failure, but we are also learning to repeat what works well. For example, surgical teams become faster and make fewer errors as they perform more iterations of a new surgical procedure; factories increase production speed and decrease rework and other costs as they build their production history.
But this routinisation and perfection of a production process is based upon repetition and stability. Those activities that cannot easily be practiced or occur infrequently, such as responding to off-normal events and accident-prone situations, are much harder to learn. And, the learning process is more difficult when the pace of change is high, because the routines and personal work relationships are so frequently renegotiated, unlearned, and relearned.
Organisations have developed many tactics for increasing learning capabilities, including:
• Simulations, drills, practice fields, imaginative rehearsal
• Learning from and with others in communities of practice
• Benchmarking and sharing good practices across organizations
• Near miss reporting, performance tracking, dashboards of indicators
• Root cause analysis for identifying leverage points for change.
However, individuals and teams must enact each of these learning activities, and therefore the good leadership and team practices discussed above again become important. Building upon practical experience and research studies in high-hazard industries, we offer several necessary conditions for effective learning:
- Top management exhibits visible commitment and involvement
- People at the sharp end are engaged and vigilant
- A supportive organisational culture respects and values the contributions of employees and avoids blaming individuals for problems
- Anticipatory learning focuses proactivelyon near misses and unsafe conditions without waiting for a catastrophe or regulatory demand for change
- A systemic or systems-based view of safety looks beyond a single cause (eg, technical error, human error) to consider multiple inter-related causes of failures
- Learning practices are context-dependent, varying with factors such as worker-management relationships, cultural expectations, internal capabilities and resources and external support and scrutiny.
The first essential for safety management is that top managers must convey their uncompromising commitment to safety in the face of competing priorities (e.g., cost, schedule). Top managers have to articulate the business case for safety for stakeholders inside and outside the organisation, and be positive role models. Safety performance has to be a key component of overall performance, a central topic of conversation, planning, and resourcing, and a legitimate way for careers to advance within the organisation. Setting challenging goals such as “zero errors” underscores commitment to learning (“always becoming safer”) rather than to compliance alone (“become safe enough”). A meaningful mix of safety metrics should capture safety outcomes and safety processes (eg workarounds, process upsets), with both quantitative metrics and qualitative examples and judgments.
People at the sharp end, including frontline employees and contractors, must be directly involved in every step of the learning process. Successful learning activities require resources in the form of engaged, committed and knowledgeable people. The value derived from incident investigations, benchmarking trips to other facilities, participation in industry events, and other learning activities depends upon involving the right people and giving them time away from competing job demands, thus signalling that learning is also “real work.” If insiders lack sufficient expertise, then insider-outsider teams can be a great way to transfer expertise (eg hire consultants, borrow people from other organisations). Those engaged in these learning activities learn about the work system, and also build interpersonal networks, develop mutual respect across realms of expertise, and begin to trust that management will fulfil their promises.
Effective learning requires an organisational culture that respects and values the contributions of all employees and does not attribute blame to individuals except in narrow circumstances of egregious conduct. Many learning activities entail “extra-role” behaviors that are not part of formal job descriptions, such as voluntary incident reporting or coaching others. These activities can be emotionally or interpersonally risky, exposing a worker to self-doubt, mockery, resentment for creating more work and even reprisals from co-workers and managers. A supportive culture ensures that people are treated with respect and fairness, and provides the “psychological safety” that speaking up about problems and opportunities is welcome; and that learning is seen as a strength rather than a weakness. Managers have to take the lead to actively model fair and respectful behaviour, and to ensure that decisions about safety are shaped by those with the relevant knowledge, not simply by those in authority.
Strong emphasis needs to be placed on anticipatory learning, in contrast to reactive learning where organisations give little priority to learning until a rare catastrophe occurs or regulatory pressures become unavoidable. The best way to anticipate problems is to learn vicariously from the failures and successes of others, by benchmarking, participating in industry knowledge-sharing events, utilising external experts and otherwise paying attention to events that have not (yet) happened to you. Additionally, learning from precursors and near misses is extremely valuable when the right precursors are identified and measured: For example, violation of separation between aircraft is an excellent precursor for airline collisions, but personal safety events are not an adequate signal of process or system safety vulnerability.
A systemic view supports investments in safety management systems that will enhance long term performance (and profits). A systems thinking approach or lack thereof becomes particularly obvious in incident or accident investigations: is the organisation ready for the potential of revelation that comes with seeking deeper, underlying, systemic causes of problems? For example, what appears to be an operator error in a chemical plant could be a symptom of multiple interacting issues: bad design of control room indicators; deficiencies in the hiring and training of operators; lack of resources to staff the control room with attendant overtime and fatigue; inability to report and fix known problems that have occurred repeatedly; or a cultural expectation for workarounds and “getting by”. Although “root cause analysis” was created with the intent of avoiding knee-jerk reactions such as blaming the front-line worker or making a quick fix that feels satisfying but does little to change the underlying causes, the label can create the misleading impression that there is a single root cause to be discovered. From a systems thinking viewpoint, root cause analysis must be appreciated for its potential to strengthen the safety management system and to engage managers and other stakeholders in a productive learning dialogue.
Finally, the effectiveness of specific safety management practices is context-dependent. For instance, a manager newly hired by an organisation that has little appreciation for safety, little in-house expertise, few resources for investigations, a culture of blame, and a history of failed implementations would have to work very differently from a manager in a more progressive, advanced, and resource-rich organisation. In the former case, an improvement strategy might start small, with ways to build credibility with small but rapid improvements, engaging a few highly-regarded workers who may have been frustrated by their organisation in the past, developing some easy-to-use anonymous reporting opportunities, simple analysis tools and metrics, and hazard identification and other training. It may be necessary to persuade people to think differently, by sending them on benchmarking visits, hiring a few skilled investigators or building insider-outsider teams, with the intent of spreading these skills rather than building a separate department of experts, and by coaching managers on how to create a sense of openness and trust. In contrast, organisations which already have a well-functioning learning system need to take the next step, whether that is focusing on lower level near-misses and precursor events, more systems thinking, risk-based prioritisation, or directly addressing “soft” problems.
Safety management is a challenging topic because so much of safety is invisible. Just as the management of innovation is distinct from the management of a production factory, the management of safety has distinct challenges. This is particularly the case when the hazards are low probability and high consequence. In such contexts, we cannot simply count accidents, because a single accident may destroy many lives and the reputation and viability of the company. We cannot simply manage by trial-and-error, because the costs of error are too high. Nor is it sufficient to announce a new mission statement or appoint a new safety executive. Instead, we must build a strong organisation that is capable of performing well while anticipating and avoiding most problems and responding quickly to signals that remaining risks could surface at any time. Such an organisation requires a strong safety management system, but that system must be complemented by effective teamwork and learning practices, supported by sufficient resources and leadership commitment. We have outlined a number of considerations that need to be taken into account by leaders in their endeavours to foster teamwork and learning in their organisations. While these considerations are not sufficient to ensure successful implementation, they can help to initiate organisational changes towards increased safety.
Argote, L. (2013). Organizational learning: Creating, retaining and transferring knowledge, 2nd ed. New York: Springer.
Bienefeld-Seall, N., & Grote, G. (2012). Silence that may kill: When aircrew members don’t speak up and why. Aviation Psychology and Applied Human Factors, 2(1), 1-10.
Carroll, J. S. & Fahlbruch, B. (2011). The gift of failure: New approaches to analyzing and learning from events and near-misses. Honoring the contributions of Bernhard Wilpert. Safety Science, 49, 1-4.
Carroll, J. S., Rudolph, J. W., & Hatakenaka, S. (2003). Learning from organizational experience. In M. Easterby-Smith & M. A. Lyles (eds.). Blackwell Handbook of organizational learning an knowledge management. Malden, MA: Blackwell, pp. 575-600.
Edmondson, A. C. (2012). Teaming: How organizations learn, innovate, and compete in the knowledge economy. San Francisco, CA: Jossey-Bass.
Goodman, P.S., Ramanujam, R., Carroll, J., Edmondson, A.C., Hofmann, D., & Sutcliffe, K. (2011). Organizational errors: Directions for future research. Research in Organiza¬tional Behavior, 31, 151-176.
Grote, G. (2009). Management of uncertainty - Theory and application in the design of systems and organizations. London: Springer.
Grote, G. (2012). Safety management in different high-risk domains – All the same? Safety Science, 50, 1983-1992.
Kolbe, M., Burtscher, M. J., Wacker, J., Grande, B., Nohynkova, R., Manser, T., Spahn, D., & Grote, G. (2012). Speaking-up is related to better team performance in simulated anesthesia inductions: An observational study. Anesthesia and Analgesia, 115, 1099-1108.
Leveson, N. G. (2011). Engineering a safer world: Systems thinking applied to safety. Cambridge, MA: MIT Press.
Reason, J. (1997). Managing the risks of organizational accidents. Aldershot, UK: Ashgate.
Weick, K. E. & Sutcliffe, K. M. (2007). Managing the unexpected: Resilient performance in an age of uncertainty, 2nd ed. San Fancisco: Jossey-Bass.
Download full article (PDF 275 KB)