Version classiqueVersion mobile

Proceedings of the fourth Resilience Engineering Symposium

Erik Hollnagel
Éric Rigaud
Denis Besnard

Improving Resilience Through Practitioners’ Well-being: An Experience In Italian Health-care

Fabrizio Bracco, Andreina Bruno et Dimitri Sossai


Building and maintaining resilience in health care requires psychological and organizational attitudes that could be affected by the lack of worker well-being. Resilience requires the ability to give strong responses to weak signals, but, if well-being is threatened, workers are more committed to defend it, than detecting and monitoring weak signals in foresight. Malaise is a weak signal itself that, as well as leading to accidents due to fatigue, miscommunication, distraction, etc., blocks operators at a resource-saving cognitive level that prevents noticing and reporting further weak signals. We adopted the Skill-Rule-Knowledge model by Rasmussen as a framework to conceive resilience as continuous movement of workers along the three steps of the ladder. According to this model, we describe a research-intervention project carried out in a few Italian hospitals where trainees were enabled to develop a tool for detecting and monitoring malaise and threats to safety. Its potentials for reducing effects like distrust, resignation, cynicism, helplessness are discussed in light of a well-being-based resilience engineering.

Note de l’auteur

Note portant sur l’auteur1

Note portant sur l’auteur2

Note portant sur l’auteur3

Texte intégral

1 Introduction

1Worker well-being and safety in the health-care domain are two closely related concepts and together they can be linked to the wider concept of resilience. The lack of employee well-being affects effectiveness and mindfulness about risks. Breaking this negative loop may be possible if we are able to act upon the roots of such conditions, factors that erode well-being and increase the probability of risk behaviour.

2Statistics about causal factors of accidents in complex systems unequivocally show the human contribution not as lack of skill, but as miscommunication, inattention, physical and mental workload, poor situation awareness, bad decision making, ineffective action planning, inability to cope with stress, emotional load, and organizational dysfunctions (Reason, 1990; Dekker, 2005). All these factors have been generically classified as “human error” and could often be due to a lack of organisational well-being. We argue that clinical risk management should start from these issues in order to assure safety for the operators and patients. More specifically, we think that every safety oriented project should take into account the cultural and organizational environment in which it is supposed to be applied. Resilience is both a personal attitude and an organizational property: it is grounded on the concepts of mindfulness, proactive reasoning, flexibility and adaptability. These issues are not easy to develop and disseminate; they imply cognitive, emotional, motivational and organizational effort and we claim that such a cultural change would be difficult to boost in employees that are living and working in a stressful environment. Therefore, a resilient system should prioritize taking care of people’s well-being.

1.1 Well-being, safety and resilience

3Protecting and monitoring well-being for resilience’s sake is based on the link between safety and well-being. Therefore, since resilience requires the capacity to produce strong responses to weak signals that could lead to accidents (Hollnagel et al., 2006), we propose to consider well-being as a signal to be monitored. Noticing and responding to weak signals is based on strong commitments of operators and management. The workers should play a crucial role in noticing them, since they are at the front-line and have a direct experience of well-being and safety-critical situations. On the other hand, the management should implement the system so that the operators notice these signals, adequately collect them and respond to them. Resilience allows the organization to detect and respond to anomalies running at the base of Heinrich’s pyramid, but if well-being is threatened it is hard to be sensitive to weak signals.

4We claim that a lack of organizational well-being is a crucial factor for resilience since it is a weak signal itself that could go unnoticed if the system is committed strictly safety-related issues (i.e., operators’ training, maintenance, processes control, etc.), but that could have detrimental effects on safety attitudes. Moreover, a lack of well-being is a particular kind of weak signal, since it creates a kind of cognitive opacity that prevents noticing other weak signals, because operators could be more worried about actual threats to their condition than about hypothetically dangerous anomalies. Neglecting well-being as a value will affect also those properties relevant to resilient organizations (Wreathall, 2006):

  • -Top-level commitment: a lack of well-being will nullify top management’s efforts of valuing human performance, since operators could lose the sense of their activity;

  • -Just culture: a lack of well-being brings operators to isolate and close, feeling threatened and blamed for their actions: it increase selfishness and suspicion;

  • -Learning culture: a lack of well-being forces operators to rely on routines and habits. It builds resistance to change and to past experience optimization;

  • -Awareness: a lack of well-being blocks information circulation, since cooperation, trust and group-thinking are limited;

  • -Preparedness: a lack of well-being makes the organization unable to detect and manage weak signals in foresight, since operators are worried about their current problems, more that reporting hypothetically dangerous weak signals;

  • -Flexibility: a lack of well-being blocks people to their roles, functions, procedures, since they do not have psychological resources available in order to manage the unexpected;

  • -Opacity: among operators, a lack of well-being filters or blocks vertical communication, prevents the organization to know how the actual work is performed, how practitioners are working close to “the edge” and how they could respond in the case of unexpected events.

2 The Theoretical Frame: The SRK Model

5The theoretical frame of reference that we are following in order to notice and respond to malaise weak signals is that of a cognitive approach to resilience as outlined in Bracco et al (2008). In this approach the Skill-Rule-Knowledge (SRK) model by Rasmussen (1983) is the ground for the analysis of the behaviour of single operators and of organisations. The reasons to adopt this model are several: it has been validated in a number of contexts and has received a wealth of empirical support (Woods, 2009); it can be used to frame phenomena both at the single operator’s level, and at the organizational level; it is a clear model to be transferred during operators’ training and it is an easy and effective tool for self-monitoring and for organizational awareness.

6The aim of this approach is to allow the transfer of information at all levels of the SRK hierarchy. Resilience depends on the capacity people have in cyclically moving along this ladder while managing weak signals. We argue that if the operators are engaged in managing their malaise, they will devote to it the highest amount of resources, leaving spare resources that are too poor for the detection of weak signals. The highest number of accidents is due to Skill-based errors (Wiegman & Shappel, 2003) and, since it is a resource-saving modality, it is chosen when operators are experiencing overload, stress, fatigue, burnout, discomfort, and isolation. At this level the operators are not able to notice weak signals in the environment, and if they notice such signals, they are not ready to properly manage them since they have a passive attitude towards them and are engaged elsewhere. If the operators are not blocked at the S-level, they can be aware of the procedure they are following, thus moving at the Rule-level. At this level one should be thorough in considering environmental demands and adopting the proper response. It implies to pay more attention to the situation and to choose the adequate behaviour. But rules and procedures are abstract concepts that have to be adapted to the contingent situation, and operators cannot simply rely on blindly following them. This mind attitude lets them notice those weak signals that can affect their well-being and, with foresight, their safety. Noticing anomalies brings them to the Knowledge-level, since they are aware of what is going on, how much the procedure is adequate and how much it does not manage the anomaly. At this level the cognitive effort is high; it requires commitment, mindfulness, systemic mindset, and foresight. A lack of well-being will hardly favour such attitudes. Therefore, the operators’ well-being is the ground for mindfulness, which allows them to treat information concerning weak signals from the Skill level to the Rule level and later to the Knowledge level (i.e. operators do not simply perform routine tasks but also notice weak signals and are mindful of their effects on safety and well-being). The cyclical flow of information related to weak signals is accomplished by moving down from the K-level to the R-level and the S-level. At the K-level the single operator reports weak signals and calls the system for new resources to manage these anomalies. This level is effortful and the organization cannot spend too much time in such a cost-effective condition; it is necessary to think of how to learn from these weak signals and to reframe them inside the normal procedures (R-level). Descending from K to R-level means to enrich procedures with the new information provided by the operators’ mindfulness; it leads to the organization setting rules and procedures in order to manage the presence of the newly detected anomalies. Training and routine application of such reframed information, enriched by the weak signals, will descend at the S-level, which is more cost-saving and easy to implement. This whole process of descending from K-level to S-level has been defined as dynamic adaptation, since the system can change its procedures thanks to the mindful attitude of operators in detecting weak signals. Both processes, mindfulness and dynamic adaptation, are crucial for system learning, development and resilience. In short, well-being could be seen as the lubricant of the cognitive and organizational cyclical processes described with this model.

7This model can capture the essence of other approaches for improving safety culture (no-blame, flexibility, etc.), but, in addition, it has two main strengths: the SRK framework can describe phenomena both at the operators’ level, and at the organizational level, thus balancing a local and a global approach, a cognitive and an organizational view. Secondarily, it is easy to visualize and understand during training, providing operators with a useful tool for highlighting the crucial “dynamic non-events” related to safety (Weick, Sutcliffe, 2007). As happened for other models (e.g., the Reason’s Swiss Cheese, Edwards’ SHELL, Heinrich’s accidents pyramid, etc.), the fortune of a conceptual tool is a good balance of explanatory power and immediacy, usability, and effectiveness.

3 The Model Implementation In Health Care

3.1 Safety and well-being in the health care domain

8This theoretical frame has been the ground for a project realized in the Italian health care domain, whose aim was to increase the commitment of operators and managers towards an organizational culture oriented towards safety and well-being, of both the operators and patients. Hospitals are one of the most critical work environment concerning well-being and safety (Dekker, 2005) and are good candidates for testing resilience engineering programs (Nemeth et al. 2008). Practitioners could be affected by stress, burnout, and workload; they could experience the pressure of balancing high technical skills together with the ability to communicate, decide, manage people and cope with unpredictable situations. In addition, they are at the centre of a critical intersection between advanced technology, complex organization, threatening environments and demanding clients. Therefore, our approach was not only about health prevention, but also health promotion. To reach this goal we implemented a research-training project that concentrated both on the dimension of safety culture and the professional practice. The project involved 30 operators/facilitators (nurses and physicians) from six Ligurian hospitals during 2010 and 2011. Each participant has been trained in order to involve the hospital workers in the implementation of the safety management systems and to disseminate methods and knowledge in their workplace. This “viral dissemination” commitment could reach up to 12.600 workers employed in the abovementioned hospitals.

3.2 The project and its outcomes

9The link between cultural and operational dimensions takes place by means of the circulation of a safety culture at the several levels of the functions and services. Safety is not implemented as a set of prescriptive rules but as a sense-making towards a systemic approach to well-being. Therefore, each element of the system has a key role for the achievement of the final result. Such a resilient organisation is characterized by the properties we listed above: ability to delegate, flexibility, sensitivity to weak signals, organizational learning, no-blame culture, and capacity to use information coming from the front-line. These principles ruled the project development and we aimed to stimulate participants at different levels. Thanks to an initial focus group and a constant monitoring of the experiences of the operators, we helped participants to work about their conception of well-being, and not about an abstract issue. In addition, we aimed at enhancing an internal locus of control that helped them treat weak signals at the Knowledge level, thanks to the capacity and responsibility to report and monitor well-being and safety issues. Moreover, applying mindfulness (from Skill to Knowledge) and dynamic adaptation (from Knowledge to Skill), will let practitioners develop a balanced locus of control, i.e., internal for what they can act on, and external for what others should manage and solve. This had the side-effect of reducing the learned helplessness and resignation: as happens in big organizations, bureaucracy and system complexity could boost a resignation about problem solving, but well-being and safety issues can be often faced at several levels, not only at the management level. Workers came to feel themselves as actors whose roles are not only passive. In addition, we openly stimulated a shared discussion about problems and about the several solutions that could be proposed in order to afford learning from such experiences. This Knowledge-level modality helps the organization to get a wealth of information from the front line; workers’ well-being is the engine for knowledge circulation. Finally, we outlined the training setting in order to let participants build a shared method according to their needs and habits.

10This method produced a tool that let participants clarify problems and their origins and the potential consequences in terms of well-being and safety. Moreover, they manage to clarify possible solutions and identify who is able to implement them. This tool is aimed at the reporting of criticalities acting against well-being, since practitioners clearly stated that resilience-oriented behaviours could be accomplished only in good enough conditions of well-being (e.g., lack of resources and role ambiguity could make the resilient action less feasible). To this aim, the middle management involvement has been crucial, since it has the function of linking the level of everyday, ordinary behaviours and the level of decision premises acting on well-being. This tool has the advantage of continuing the cyclical flow between workers and the organization, from Skill to Knowledge and back to Skill. It is composed by three sections:

  1. Anomalies detection (from Skill-level up to Knowledge-level): this section allows operators to signal problems, their frequency, and take into account the factors that produce them and the potential consequences. In addition, it stimulates the operator reporting a problem to propose a solution in order to give a personal contribution to the increase of the general safety and well-being. This enhances an internal locus of control and assumption of responsibility in process management.

  2. Problem setting and problem solving (understanding problems and designing solutions at the Knowledge-level, enriching procedures at the Rule-level): this section helps practitioners in organizing problem solving; it requires them to report who is involved in the process (doctors, nurses, etc.), how they carry on the problem analysis (staff meeting, personal discussion, etc.), what solutions are outlined, who is charged with implementing them (internal staff or external units), the level of general agreement and satisfaction for the proposed solution, and a timeline for solution monitoring.

  3. Solution process monitoring (a trigger for restarting the cyclical flow along the SRK ladder): in this section single operators and groups are involved in process monitoring and can track the steps towards a solution, assessing also the adequacy of intermediate outcomes and the possibility of further actions.

11This tool has the advantages of involving operators in a process of information flow; it gives them the right perception of responsibility and ability to intervene and modify their environment. In addition, it can become a database of already solved issues concerning safety and well-being, allowing other hospitals to learn from these experiences. It can also afford a clear monitoring of processes that are generally long and tortuous due to the organizational complexity of health care units. If a problem does not come to a solution, it is possible to find where and why this process stopped, thus helping operators to avoid falling again into resignation and helplessness.

4 Conclusions

12We developed a research-intervention project to enhance resilience through operators’ well-being. Resilience could be framed within the SRK model, according to which the proper circulation of workers and organization along the ladders enhances resilient behaviours. Malaise will cause workers to deal with weak signals at the wrong level, generally Skill, the most resource-saving modality, that has the limit of making them passive and blind to anomalies. The project ended with the proposal and the use of a tool for well-being and safety related weak signals, whose aim is to boost the correct circulation along the SRK ladder. The midterm aim of this project appeared to be particularly challenging and the participants are now committed to the project’s long term sustainability. They are asking managers how to accomplish this goal, for instance by activating a board of observers (cross-sectional to the several organisational contexts) about safety cultures in health-care. Such a need to sustain the long-term project is evident in the participants’ involvement of their colleagues and coordinators. The aim of such efforts is to continuously sustain the good practices realized by the participants, i.e. organizational problem management for safe productive process monitoring.



Bracco, F., Gianatti, R., & Pisano, L. (2008), Resilience engineering in Emergency Room operations: A theoretical framework. In E. Hollnagel, F. Pieri, E. Rigaud (eds.). Proceedings of the Third Resilience Engineering Symposium. Paris: École des Mines de Paris.

Dekker, S. (2005), Ten questions about human error, London: LEA Publishers.

Hollnagel, E., Woods, D., Leveson, N. (2006), Resilience engineering, concepts and precepts. London: Ashgate.

Nemeth, C.P., Wears, R.L., Woods, D.D., Hollnagel, E., Cook R.I. (2008), Minding the gaps: Creating resilience in healthcare, In K. Henriksen, J.B. Battles, M.A. Keyes and M.L. Grady (Eds.) Advances in patient safety: New directions and alternative approaches. Vol. 3. Performance and Tools, Rockville: AHRQ Pub. N. 08-0034-3.

Rasmussen, J. (1983), Skills, rules, knowledge; signals, signs, and symbols, and other distinctions in human performance models. IEEE Transactions on Systems, Man and Cybernetics, 13, 257-266.

Reason, J. (1990), Human Error, New York: Cambridge University Press.

Weick, K.E, Sutcliffe, K.M. (2007), Managing the Unexpected, John Wiley and Sons, Inc.

Wiegman, D. A., Shappel, S. A. (2003), A human error approach to aviation accident analysis, London: Ashgate.

Woods, D.A (2009), Rasmussen's S-R-K 30 Years Later: Is Human Factors Best in 3's? Human Factors and Ergonomics Society Annual Meeting Proceedings, Volume 53, Number 4, pp. 217-221(5).

Wreathall, J. (2006), Properties of Resilient Organizations: An Initial View. In E. Hollnagel, D. Woods, N. Leveson (Eds.), Resilience Engineering, Concepts and Precepts (pp. 275-285). London: Ashgate.


1 University of Genova, Department of Anthropological Sciences, Psychology Unit, corso A. Podestà, 2 – 16122, Genoa, Italy

2 University of Genova, Department of Anthropological Sciences, Psychology Unit, corso A. Podestà, 2 – 16122, Genoa, Italy

3 Director of the Health Safety and Prevention Department, Azienda Ospedaliera Universitaria San Martino, Largo R. Benzi, Genoa, Italy

© Presses des Mines, 2011

Licence OpenEdition Books

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search