Building a Safety Program Using Principles of Resilience Engineering
Approach to Improving Safety
The traditional framework of safety is predominantly focused on systemically studying and preventing things that go wrong. The presence and absence of incidents, errors, and negative outcomes are, no doubt, indicators of system performance.
However, the absence of reportable incidents does not mean that everything is going well, and the occurrence of an incident does not always indicate substantive deviation from usual practice.(1) In the absence of a noticeable or concerning incident, such as an adverse event or near miss, we tend not to examine our everyday work practices to understand if they are bringing the system close to the boundary of safe performance; or how these practices may be contributing to safe performance. When an incident or error does occur, aspects of work are analyzed through the lens of the negative outcome. This analysis and the interventions that follow—new rules, training, or worse, disciplinary action—sometimes undermine the operational effectiveness of everyday work.
In reality, the risks, adaptations, and workarounds that are typically identified through retrospective incident analyses are always present, and—along with individual and team capabilities related to situation awareness, adaptation, and coordination—they serve to enable resilient functioning. The actions, processes, and relationships that enable or represent these capabilities mostly remain hidden, or are seldom studied, particularly in the absence of noticeable events. The field of resilience engineering focuses on learning proactively about how things usually go well and developing ways of supporting the same.(2) Rather than reduce variability and mitigate error, resilience engineering strives to enhance the system's capabilities identified within the resilience framework.
Resilience Engineering—Implications for Safety Programs
A key area of emphasis in resilience engineering is proactive learning about frontline work characteristics, specifically what enables and impedes people to succeed in everyday situations. Most empirical work in resilience engineering to date has centered on studying how resilience manifests in various systems.(3) These studies typically involve researchers observing and interviewing frontline workers.
However, developing a safety program calls for learning mechanisms that can be scaled longitudinally and at organizational levels. Health care organizations need to continually learn about dynamic work patterns while informing policymaking and design intervention at multiple levels. This requires a framework that involves learning from the workforce about how they deal with everyday challenges and constraints through adaptation, planning, coordination, knowledge sharing, and other capabilities. From an incident analysis standpoint, this framework implies shifting away from finding errors (especially as root causes) and toward incorporating insights gleaned from prior learning in that environment to provide better context to the analysis. The framework should include a multipronged approach to elicit information that is meaningful from a resilience standpoint. Some potential or implemented approaches to date include the following:
- Self-reporting tools: Hospitals can implement short survey reporting tools, which frontline caregivers and workers can use to share information about how things go well in their daily practice. As an example, a tool (recently piloted at a large multispecialty hospital) has been designed to elicit seminarrative information about adaptive practices, including workarounds, in anesthesia care.(4,5)
- Morbidity and Mortality (M&M) conferences: Current M&M conferences are typically centered on clinical or safety issues related to difficult cases or cases with negative outcomes. A resilience engineering approach could be to also discuss similar cases that were handled more effectively, highlighting the value of the decisions and adaptive maneuvers that are typically successful.
- Workshops and training: Shifting a hospital's approach to safety would require formal and informal employee training in the fundamentals of the resilience engineering approach—specifically, the need to focus on how safety is created on an everyday basis. The training should also communicate leadership's commitment to new forms of learning and expectations of staff in terms of information sharing.(6)
A further implication of resilience engineering is in the need to consider safety within a larger context of high-level organizational goals, which often have complex interrelationships. For instance, organizations under financial pressure often stress operational efficiency. However, a push for efficiency can lead to unanticipated effects on safety, as resources (such as staff time that provides extra capacity in times of need) are reduced. Workarounds developed by staff to cope with design shortcomings or the need to comply with regulations can become normalized behavior over time.(7) Adaptations can mask the increasing "brittleness" of the system, as the push for efficiency forces it to function at capacity. For instance, the push for maximizing operating room (OR) utilization can sometimes lead OR coordinators to overschedule or overcommit to add-on cases during a surgical shift. Most of the time, the OR staff manage such trade-offs individually and as a group without ill effects. Sometimes, however, the system is overwhelmed by multiple emergencies, requiring staff to be pulled away from scheduled assignments, throwing scheduled cases and patient wait times into uncertainty. Recognizing such drift toward boundary conditions and taking proactive actions to reinstate necessary buffers and resources is vital to facilitating resilient capabilities. This requires balancing of priorities across multiple stakeholders and levels of the organization, such as between different clinical, nonclinical, and administrative specialties and roles. Therefore, a key role of learning is to enhance integrated understanding and shared awareness between the operational and policymaking layers of the organization.
Challenges and Opportunities Ahead for Research and Implementation
Resilience engineering, as a field, is beginning to move from theory to practice. The following are some of the key challenges that influence the implementation of a resilience engineering-based safety program. First, to implement a resilience engineering approach, leadership must be committed to the framework and staff must participate with a sense of ownership. The Aviation Safety Reporting System (ASRS) serves as a model to emulate.(8) Its high response rates have been attributed to features such as confidentiality of reports and a feedback loop to communicate follow-up actions to respondents.(9) In health care, however, response rates of caregivers to self-reporting systems tend to be very low. Strategies that were instrumental in the success of the ASRS can be adapted for health care, as appropriate, in order to drive and sustain a healthy response rate. Second, more data collection tools and analytical frameworks are needed to actualize proactive learning at the organizational level. These then have to be catered to specific clinical environments. Few frameworks have been proposed—the Resilience Analysis Grid and the Functional Resonance Analysis Method being the most common.(10,11) However, the majority of resilience engineering research has focused on identifying how resilience manifests in different work systems rather than on building a comprehensive readily usable toolkit for application. Thus, we are not yet at a stage where hospital safety administration can comprehensively adopt an explicitly resilience engineering–based approach toward restructuring their organizational processes from policymaking to technology design.(12)
The concepts of resilience engineering hold significant potential to augment system-based efforts to improve safety. The application of this paradigm in health care will require that we explore strategies to effectively integrate these concepts into safety thinking and practice.
Sudeep Hegde, PhD
Department of Industrial and Systems Engineering
State University of New York at Buffalo
Ann M. Bisantz, PhD
Professor and Dean of Undergraduate Education
Department of Industrial and Systems Engineering
State University of New York at Buffalo
Rollin J. Fairbanks, MD, MS
Vice President of Quality and Safety, MedStar Health
Professor of Emergency Medicine, Georgetown University
5. Hegde S, Wreathall J, Hettinger AZ, Fairbanks RJ, Wears RL, Bisantz AM. Towards the development of a resilience engineering tool to improve patient safety: the RETIPS approach. Proc Hum Factors Ergon Soc Annu Meet. 2014;58:803-807. [Available at]
6. Chang S, Wears RL. Strategies to get resilience into everyday clinical work. In: Wears RL, Hollnagel E, Braithwaite J, eds. Resilient Health Care, Volume 2: The Resilience of Everyday Clinical Work. Boca Raton, FL: CRC Press; 2017:225-234.
7. Woods DD, Branlat M. Basic patterns in how adaptive systems fail. In: Hollnagel E, Pariès J, Woods DD, Wreathall J, eds. Resilience Engineering in Practice: A Guidebook. Farnham, UK: Ashgate Publishing, Ltd; 2011:127-144. ISBN: 9781409486534.
10. Hollnagel E. RAG—The resilience analysis grid. In: Hollnagel E, Pariès J, Woods DD, Wreathall J, eds. Resilience Engineering in Practice: A Guidebook. Farnham, UK: Ashgate Publishing, Ltd; 2011:275-296. ISBN: 9781409486534.
11. Hollnagel E. FRAM: The Functional Resonance Analysis Method: Modelling Complex Socio-technical Systems. Boca Raton, FL: CRC Press; 2017. ISBN: 9781409445517.
12. Hollnagel E, Braithwaite J, Wears RL. Making it happen—from research to practice. In: Hollnagel E, Braithwaite J, Wears RL, eds. Delivering Resilient Health Care. New York, NY: Routledge; 2018:210-216. ISBN: 9781138602250.