Article Text

Download PDFPDF

Re-examining high reliability: actively organising for safety
  1. Kathleen M Sutcliffe1,2,
  2. Lori Paine2,3,
  3. Peter J Pronovost2,4,5
  1. 1Carey Business School, Johns Hopkins University, Baltimore, Maryland, USA
  2. 2Armstrong Institute for Patient Safety and Quality, Johns Hopkins Medicine, Baltimore, Maryland, USA
  3. 3Department of Medical Affairs, Patient Safety, The Johns Hopkins Hospital, Baltimore, Maryland, USA
  4. 4Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University, Baltimore, Maryland, USA
  5. 5Department of Surgery, Johns Hopkins University, Baltimore, Maryland, USA
  1. Correspondence to Dr Peter J Pronovost, Armstrong Institute for Patient Safety and Quality, Johns Hopkins Medicine, 750 E. Pratt Street, 15th floor, Baltimore, MD 21202, USA; ppronovo{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

In the 15 years since To Err is Human was published,1 the US healthcare industry has worked diligently to improve patient safety. Although progress has been made in reducing hospital-acquired conditions2 and, in some cases, rates of surgical mortality,3 healthcare has not achieved broad reductions for most patient harms. In recent years, healthcare has borrowed ideas from industries that have strong safety records, including teamwork and error reporting from aviation, and process improvement techniques from manufacturing. Healthcare's latest patient safety push is to encourage hospitals to become a ‘high reliability organisation’ (HRO).4

HROs have maintained remarkable performance despite complex and risky work. These ultrasafe organisations never set out to be HROs. As Rochlin5 observed: HROs ‘seek an ideal of perfection but never expect to achieve it. They demand complete safety but never expect it. They dread surprise but always anticipate it. They deliver reliability but never take it for granted. They live by the book but are unwilling to die by it’. HROs understand that reliability is an endless journey rather than a simple destination.

Evidence suggests that healthcare is starting to organise for higher reliability. Standardised protocols and checklists,6 preprocedural and postprocedural briefings,7 incident reporting and daily huddles,8 although imperfect,9 ,10 may hold promise for enhancing safer care. These types of activities may be part of an institution's master plan to create a comprehensive operating management system—an organisation-wide integrated approach to manage risk and to achieve safe and reliable performance—similar to the systems found in other industries such as oil and gas. Yet, we think it is more likely that these efforts represent piecemeal and fragmented initiatives adopted to solve particular problems. Regardless, high reliability remains elusive. One explanation is that organisations have failed to widely institutionalise high-reliability habits of thought and action.11 A second explanation is that low reliability persists because healthcare lacks a solid understanding of some fundamental underpinnings of highly reliable performance. Without a deeper, more nuanced understanding of these foundations, possible gains that can be made will not materialise or the gains made will be lost.

High-reliability operating logics

Highly reliable organisations design work systems to anticipate, contain and recover from mishaps.12 ,13 Like other high-performing organisations, they work hard to understand the nature of the work, create detailed operating procedures and contingency plans, and use the tools of science and technology (eg, failure mode effects analyses, or hazard and operability studies) to shape the behaviour of organisational members to avoid mishaps. Anticipating and preventing predictable risks and hazards through standard routines and protocols removes uncertainty, reduces information processing requirements and, thereby, decreases possible memory lapses, judgement errors, or biases that can lead to adverse outcomes.14 But there are limits to a logic of anticipation.

One limitation is that unvarying procedures cannot handle what cannot be anticipated.12–14 Given healthcare's complexity and variability (eg, patient genetics, socioeconomic status and health literacy), it is hard to craft procedures to anticipate all, perhaps even most of, the situations and conditions that might shape people's work.14 Even if that were possible, there are costs. Additional rules add complexity, which can decrease people's flexibility to contingently respond.14 Standardisation influences reliability because it creates operating discipline. Nonetheless, blindly following the rules can be deadly if it reduces the ability to adapt or to react swiftly to surprises. Anticipation (standardisation) certainly matters. But, as healthcare deals with patients who have different risks and respond to therapies differently, reliability also requires a logic of resilience and recovery.12 ,13 Indeed, resilience may be more important than anticipation. For example, Ghaferi et al15 found that hospitals with low patient mortality did not excel at controlling risks or preventing postsurgical complications. High and low-mortality hospitals experienced similar postsurgical complication rates. Low-mortality hospitals were more proficient than high-mortality hospitals at recognising and managing serious complications as they unfolded.

Highly reliable organisations develop capabilities to detect, contain, cope with and rebound from the inevitable risks and hazards that are part of an indeterminate world,12 ,13 the healthcare world. The hallmark of an HRO is not that it is error free but that errors do not disable it. In a healthcare context, an HRO would not be error free, it would be harm free.

The abilities to prevent and manage mishaps before they spread and cause harm are generally traced to dynamic organising, particular reliability-enhancing habits of thought and action.11–13 HROs mindfully organise to improve collective alertness and awareness so that people will become aware of risks, harmful deviations and errors earlier in their unfolding, and will be able to contingently respond more quickly and more appropriately. Highly reliable performance is a moving target—it is a transient, dynamic, non-event that must be continuously re-accomplished and co-created by all organisational members.16 It is dynamic because safety is preserved by timely human adjustments; it is a non-event because successful outcomes rarely call attention to themselves. This can decrease vigilance and feelings of vulnerability, increase a propensity towards complacency and inertia, and decrease the quality of attention across the organisation. That is why establishing recurring daily habits of thought and action matter. But there is more to becoming highly reliable than standards and habits.

Social-relational foundations of high reliability

Much has been written about the distinctive operating styles and recurring organising processes and practices of HROs, but the necessary preconditions of high reliability have been virtually ignored by healthcare. High-reliability organising is enabled by particular human resource practices, a climate of trust and respect, and vigilant coordination of upstream and downstream work. These elements, accompanied by recurring daily practices, shape organisational cultures that enable people to recognise emerging problems earlier and to manage them more decisively. Such capabilities could serve patients well.

High-performance work practices

Work practices such as sensible selection, training, incentives and flexible work arrangements for employees have been used for years in many industries. Particular work practices are effective because they increase employees’ capabilities and motivate them to use it for the organisation's benefit.17 ,18 In healthcare, some work practices such as selecting and mentoring for interpersonal skills, continual training and fostering a learning orientation are critical to reducing errors and enhancing reliability (TJ Vogus and D Iacobucci. Creating highly reliable health care: how reliability enhancing work practices affect patient safety in hospitals. ILR (Review 2016, forthcoming). These practices fuel trust between interdependent colleagues who must engage in difficult, hazard-focused, stressful interactions.

Trust and respect

How people relate to each other at work can account for the kind of intelligence and safety they produce.13 ,19 Relationships and transient interactions are critical to mindful organising and reliable performance partly because these connections are moments when trust is negotiated, where discourse can sharpen or blunt one’s sensitivity to unexpected discrepancies and ambiguity about action options can be more or less resolved.13 When employees face situations where their private view is at odds with a majority view they sometimes feel threatened. This can inhibit them from speaking up and out about safety threats. In contexts where trust and respect are the norm, people are more likely to both communicate their interpretations to others, and through this communication, generate a clearer interpretation of the situation they face.13

Heedful inter-relations

All organisational members must pay heed to issues of interconnectivity, particularly how their activities inter-relate and how they work together with others to accomplish collective goals.13 ,20 All organisational members must understand what is happening both upstream and downstream. When people inter-relate their activities heedfully, they understand how a system is configured to achieve some goal and they see their work as a contribution to the system and not as a standalone activity. Also, they visualise the meshing of their job with other people's jobs in accomplishing the system goals. Finally, they maintain a conscious awareness of the overall goals of the system and how their own work connects with others as they perform their duties.20

Actively organising for a culture of high reliability

Healthcare organisations that repeatedly and continually organise in these ways are likely to achieve greater reliability than those that do not, in part because they create binding cultures and climates of safety.13 ,16 Healthcare widely recognises the importance of safety culture (and its surface manifestation in safety climate).21 ,22 Yet, there is often confusion about how these are shaped.

Our discussions with clinicians and other types of frontline staff reveal that people often view culture and climate as something outside and separate from themselves, as if they are passive pawns rather than potent players in creating them. But everyone plays a role. Safety culture and safety climate are, in part, by-products of organisational properties and inter-related organising processes and practices. Although culture and climate are critical to higher reliability and safer care, they are often discussed with insufficient richness to understand how they work.

Culture can be thought of as the frames of reference for meaning and action that encompass the skills, beliefs, assumptions, norms, customs and language that members of a group develop over time.13 ,16 Culture works unobtrusively by controlling people's decision premises—that is, their expectations about the consequences of particular behaviours (eg, risk-taking, procedure violation or unsafe behaviours, such as bypassing hand washing or reporting of errors).13 ,16 Safety culture leads to increased safety by fostering, with minimal surveillance, an efficient and reliable workforce sensitised to safety issues.21 ,22

Though we speak of organisational culture as a single construct, organisational cultures are rarely harmonious, integrated and shared; rather they are differentiated and fragmented with multiple subcultures.21 And, these subcultures are often more potent in guiding people's behaviour. One reason is because culture is acquired over time through socialisation, social learning processes and group problem solving. And, the stability of a group as well as the length of time a group has existed also play a role.13 ,22 But more importantly, safety culture and safety climate are enabled by organisational leaders through their actions and the management systems they create, are enacted by organisational members when they practice the organisation's safety policies, procedures and habits, and are continually shaped and elaborated over time as people continually reflect on practice through monitoring, analysis and feedback systems, and continuous process improvements.13 ,22 A culture of safety is shaped by leaders’ actions, particularly the extent to which they demonstrate a commitment to safety through the visions they create, the goals they set and communications that signal what is and is not important. Leaders influence culture also through other actions such as resource allocations, technology choices and availability, training expenditures, systemic policies and procedures (eg, selecting for interpersonal skills, incentive and accountability systems and care pathways), and information and reporting system design.13 ,22 The system's overall safety philosophy and practices are reflected in these activities, which, as we noted earlier, are called in some industries such as oil and gas operating (or safety) management systems. Healthcare lags behind other HROs in establishing a comprehensive and integrated programme such as this.

Enabling a culture of safety is not simply a top-down process. Bottom–up organisational discourse and communication by employees at all levels regarding the way ‘safety is handled around here’ is equally critical.13 ,16 Employees discern the hidden core values and assumptions that constitute the organisation's culture by actively making sense of the signals that organisational leaders send through their actions and by actively making sense of the organisation's operating system (eg, technology, practices, sets of rules and policies). When employees make sense of discrepancies between what leaders say they want and how they act (eg, differences in declared organisational policy and informal supervisory practice), they discern the organisation's collective unconscious values, beliefs and assumptions. Thus, culture is shaped through this ongoing social verification process.13 ,16 Safety culture is a dynamic process that is continually supported, shaped and co-created daily, which makes it hard to control. Consequently, culture is not an infallible form of behavioural control even though it is often alleged as a primary cause of myriad organisational disasters.

High-reliability organising alone will not cure healthcare's ills. Ultrasafe industries have designed safety into sector-wide systems, implemented wise regulations, and developed sophisticated and integrated tools and technologies that support workers, and measure and monitor risk and performance. Each of these components is underdeveloped in healthcare.

Still, high-reliability organising is a potent therapy that can improve patient safety. Such organising makes weak signals more salient and more amenable to adaptive action. By constantly adapting, tweaking and solving small problems as they emerge throughout the system, high-reliability organising prevents more widespread failures and improves safety. If healthcare is to benefit from the lessons of ultrasafe organisations, it must deeply and systematically work to establish the social and relational foundations of reliable performance and embed the practices that make them ultrasafe. Most of all healthcare must recognise that you never get safety completely behind you.


The authors thank Christine G Holzmueller, BLA and Vipra Ghimire, MPH for their review and assistance in editing the manuscript; both have granted permission to be acknowledged.



  • The OMS Framework provides a company-wide, integrated approach to manage risk (including effects, impacts and threats) to achieve benefits and to optimise performance.

  • Contributors All of the authors contributed equally in conceptualising and reporting of the content described in this paper. Relative to the contributors, CGH reviewed and edited the structure and flow of the essay, and VG proofed and edited the essay.

  • Competing interests PJP reports receiving grant or contract support from the Agency for Healthcare Research and Quality, the Gordon and Betty Moore Foundation (research related to patient safety and quality of care), the National Institutes of Health (acute lung injury research) and the American Medical Association Inc. (improve blood pressure control); honoraria from various healthcare organizations for speaking on patient safety and quality (the Leigh Bureau manages these engagements); book royalties from the Penguin Group for his book Safe Patients, Smart Hospitals; and stock and fees to serve as a director for Cantel Medical. PJP is a founder of Patient Doctor Technologies, a start-up company that seeks to enhance the partnership between patients and clinicians with an application called Doctella.

  • Provenance and peer review Not commissioned; externally peer reviewed.