organizational approaches to safety...the challenger launch decision. risky technology, culture, and...

IIOrganizational Approachesto Safety

Training in medicine largely focuses on individual excellence and achieve-ment. For those of us who spend much of our life in training, it may be naturalto assume that well-trained and highly motivated individuals, placed together inhealth care settings, will produce quality care. It may come as a surprise that in-stitutions and health care organizations themselves can either augment or sabo-tage the success of individuals and teams of health care providers. The conceptsof hospital, institution, and organization seem inanimate and impersonal, andhardly capable of influencing thought or action. This section explores the realitythat organizations have personalities and cultures that impact our ability tosucceed.

Unhealthy organizations impact individual behavior, sometimes in an inex-plicable manner. There is probably no more telling an example of this fact thanthe recent closing of Martin Luther King/Drew Medical Center in Los Angeles.The hospital, built out of racial unrest to care for the indigent and disadvantagedcommunity of South Central Los Angeles, had a long history of failed inspectionsand patient care concerns. In 2006, it lost federal funding and underwent a

81214_CH05.qxd 6/19/08 4:05 PM Page 23

radical restructuring. The account of the death of Edith Isabel Rodriguez onMay 9, 2007, at the Los Angeles hospital is a poignant example of the decay inpatient care (1). Lying on the floor of the Emergency Department in a pool ofvomit and blood, monitors reveal that the staff worked around her in apparentindifference even while bystanders called for help from 911 operators, beggingthem to take her to another hospital. It’s easy to find fault here with the individ-ual caretakers, the doctors, nurses, and janitor, but it requires a step back to seethe influence of environment on individual care and understand the role thatorganizational pathology and neglect had in shaping the individual responses.

The impact of organizational health can also be learned from areas outsidemedicine. Although the National Aeronautics and Space Administration (NASA)is viewed historically as a prototype for high reliability organizations, even itexperienced erosive forces that undermined the safety of its mission. The SpaceShuttle Challenger disaster has been described as an event rooted in normalizeddeviance; that is, daily purposeful decisions were made that allowed the organiza-tion to expand beyond the bounds of acceptable risk (2). Organizations may bepressured by production demands and limited by financial woes; certainly that istrue for many health care organizations. In such an environment, risk is some-thing that becomes a part of the daily routine. Without care, we may neglect theaspects of risk that can be managed, not just accepted as an inevitable part of ourwork. In her analysis of the Challenger disaster, Vaughan (2) notes that organi-zational influences largely caused the failure. She cautions us to avoid blamingindividuals, for, “as long as we see organizational failures as the result of indi-vidual actions our strategies for control will be ineffective, and dangerously so.”The organization influences how risk is viewed, impacts the behavior ofindividuals, and determines how effectively individuals at all levels of thehierarchy communicate. Another essential lesson that is gained from theChallenger incident is that reliability and safety are not qualities that areachieved, but rather sought and never fully satisfied.

In Chapter 5, Christianson and Sutcliffe introduce concepts from high relia-bility organizations: organizations that deal with complex and high-risk activitiesbut are successful in avoiding harm. They argue that success requires activeawareness of risk and specific intent to manage safety, both individually and col-lectively. Ideas from management and organizational literature offer a new lookat health care, stepping beyond our typical view of medicine and the manage-ment of any one patient to principles that influence how we can manage situa-tions and teams.

Westrum (Chapter 6) explores the influence of organizational culture on in-formation flow and, thus, the ability to solve problems. His description of gener-ative leadership explains how a generative culture allows individuals and teamsto function creatively and adapt to challenges in the heat of the moment.

Finally, Pimental (Chapter 7) highlights the features of organizations thatpromote success even in the face of excessive demands and unfavorable workingcircumstances. She argues that effective leadership provides vision and directionand promotes healthy relations.

24 SECT ION I I • ORGAN IZAT IONAL APPROACHES TO SAFE TY

81214_CH05.qxd 6/19/08 4:05 PM 4

These authors present a contrast to the harrowing accounts of Edith IsabelRodriguez and the Challenger disaster. Concepts from high reliability organiza-tions, healthy cultures, and relational leadership all contribute to organizationalhealth. Our authors present sound arguments that awareness and attention to thehealth and vitality of our health care institutions themselves is essential to oursuccess in caring for our patients.

R E F E R E N C E S

1. Landsberg Mitchell. Why supervisors let deadly problems slide. Los Angeles Times. December 9,2004:A1. Available at: http://www.latimes.com/news/local/la-me-kdday5dec09,0,6376506.story.Accessed February 17, 2008.

2. Vaughan D. The Challenger Launch Decision. Risky Technology, Culture, and Deviance at NASA.Chicago: The University of Chicago Press; 1996.

SECT ION I I • ORGAN IZAT IONAL APPROACHES TO SAFE TY 25

81214_CH05.qxd 6/19/08 4:05 PM Page 25

81214_CH05.qxd 6/19/08 4:05 PM Page 26

27

C H A P T E R 5

Sensemaking, High-reliabilityOrganizing, and ResilienceMarlys K. Christianson and Kathleen M. Sutcliffe

IntroductionSensemaking

What Is Sensemaking?How Is Sensemaking Relevant to the Emergency Department?What Enables Effective Sensemaking?

High-reliability OrganizingWhat Is High-reliability Organizing?How Is High-reliability Organizing Relevant to the EmergencyDepartment?What Enables Effective High-reliability Organizing?

ResilienceWhat Is Resilience?How Is Resilience Relevant to the Emergency Department?What Enables Higher Levels of Resilience?

ConclusionSummaryReferences

INTRODUCTIONAlthough the study of safety and reliability has a long tradi-tion within the management and organizations literature(1,2), it is only recently that organizational scholars havebegun to study safety and reliability in health care organi-zations (3,4). Similarly, it is only recently that health carepractitioners and scholars have begun to incorporate themanagement and organizations literature on safety and reli-ability into investigations of medical error and patient safety(5). This interdisciplinary interest is prompted, in part, by thepublic spotlight on medical error that has followed the pub-lication of the Institute of Medicine report, To Err Is Human(6), which has called for a greater focus on organizationalsystems.

In this chapter, we seek to strengthen the bridge betweenorganization and management theory and health care. We fo-cus on three key concepts—sensemaking, high-reliability or-ganizing (HRO), and resilience—that have emerged from theorganizational literature as being critical for understandinghow organizations and the people within them can detect andcorrect error early in its unfolding. Although these conceptsare distinct from each other, they are also interrelated. We in-troduce each concept, explain how each concept is relevant tothe emergency department (ED), and suggest ways in whicheach concept can help individuals organize to provide saferand more reliable care for patients.

SENSEMAKING

WHAT IS SENSEMAKING?

Sensemaking sometimes is described literally as “how peoplemake sense of events,” but sensemaking is more complicatedthan that. Sensemaking theory provides insight into how in-dividuals and groups of individuals notice and interpret whatis happening in their environments and how they translatethese interpretations into action (7). Sensemaking is a processthrough which individuals turn a flow of experiences and cuesinto words and understandings that then serve as a platformfor action (8). Phrased differently, people act their way intoknowing. Sensemaking involves constructing a plausible storyabout events and then testing that story. Fundamentally, sense-making consists of asking and answering two questions:“What is going on here?” and “What do I do next?” (8). Sense-making is especially salient in situations where uncertainty andambiguity are high.

Clinical diagnosis is an occasion for sensemaking. Diag-nosis involves an ongoing interaction between interpretationand action. Marianne Paget, a medical sociologist who stud-ied medical error, offers a particularly elegant description ofthe relationship between interpretation and action in diagno-sis and treatment:

The diagnostic and therapeutic process is a way of think-ing and acting out, or interpreting and experimenting withcare about cases. It unfolds as a sequence of activities beingacted out: as tests, procedures, plans, prescriptions, and ad-vice. The process is acted out in a double sense. A diagnosisis an interpretive act that tests the meaning of this particu-lar illness and of knowledge of human illness in this instance.It is also an interpretive act testing in acting as if it wereaccurate or plausible or revealing. A diagnosis, in otherwords, is not a diagnosis until tested. It is a hypothesis of adiagnosis to be acted on. In this same sense, a therapeuticplan is not the therapeutic plan. In fact, until it is tested, itis a hypothesis of an appropriate therapeutic plan about tobe acted on as if it were indeed the appropriate plan. A pro-cedure is a procedure being tested, presumed to be appro-priate until further notice (9, p. 52).

Studies investigating the process of clinical diagnosis oftenprivilege the decision-making perspective to the exclusion of thesensemaking perspective. In fact, sensemaking is sometimeslumped together with decision making, but we regard such

81214_CH05.qxd 6/19/08 4:05 PM Page 27

blurring as counterproductive. The medical literature usuallyframes the process of diagnosis in terms of a choice (decision)between different alternatives. However, we want to stress thatdecision making is only possible after sensemaking has alreadyoccurred. Why is this the case? Because sensemaking deter-mines whether the “entity” of a decision even gets noticed andincluded (or bracketed) in ongoing activity—thus, the notion ofsensemaking addresses events that happen long before peoplesuspect that there may be some kind of decision to be made (8).Whereas decision making presumes that individuals endeavorto select the alternative with the highest utility, sensemakinghighlights that individuals often encounter confusing situationswhere it may not be clear what choices exist and yet must takeaction despite ambiguous or incomplete information.

HOW IS SENSEMAKING RELEVANTTO THE EMERGENCY DEPARTMENT?

The ED simultaneously provides opportunities for sensemak-ing, as well as challenges to sensemaking. In the ED, patientsfrequently present with nonspecific symptoms—such as “weakand dizzy” or abdominal pain—that generate a long differ-ential diagnosis. Consequently, it can be difficult for physi-cians to make sense of why the patient is unwell. In additionto symptoms with multiple potential causes, diagnosis in theED is complicated by the fact that physicians and other healthcare providers often must work with limited information; pa-tients may be unable to communicate (i.e., are unconscious),may have altered mental states (i.e., have dementia or havetaken drugs or other psychotropic substances), or may haveimportant past medical history that is unknown to ED staff.Lastly, ED health care providers must function in a very com-plex and unpredictable environment: Work in the ED is un-bounded, involves caring for multiple patients simultaneously,is subject to high levels of uncertainty, is provided undersignificant time constraints, is associated with little or no feed-back, and affords little opportunity to practice risky procedures(10). Thus, the ED is a setting where diagnosis often requiresinterpretation and action under conditions of uncertainty andambiguity.

More generally, the sensemaking perspective is relevantbecause it is, in many ways, closer to the lived experience ofhealth care providers than the decision-making perspective.The complexity of patient care often goes unacknowledged,but is evident in the following quote, in which a medical res-ident describes an incident where an elderly gentleman wasadmitted for fever of unknown origin:

He came to the ED with a fever of unknown origin and fa-tigue . . . And they did a chest X-ray work up, urine, bloodtest, CBC, chem 7, it was negative [i.e., no source of infec-tion could be found]. They couldn’t figure it out. So, hewasn’t given antibiotics because they didn’t know the sourceof the fever . . . When they rounded on the patient in themorning, they repeated the chest X-ray. And it turned outthat there was pneumonia and they gave him antibiotics.But it was too late. So he became septic quickly and he died.So, I don’t know if he was given the antibiotic earlier dur-ing the night, 12 hours earlier maybe that could have savedhis life (28, p. 904).

Symptoms change and evolve over time. Sometimes thecorrect diagnosis is only apparent in hindsight. Moreover, aswe see in the quote above, clinical diagnosis is dynamic, fullof uncertainty and ambiguity, and, by its very nature, error-ridden work (9). Sensemaking underscores that diagnosis isoften not a linear process where physicians follow diagnosticalgorithms, but rather a messier process that involves patternrecognition (12,13,14), intuition (15), and trial and error learn-ing (16). In fact, as shown in the following quote, some or-ganization theorists go so far as to suggest that physicians maytreat patients on the basis of their symptoms without takingthe step of diagnosing very seriously:

Good doctors pay careful attention to how patients respondto treatments. If a patient gets better, current treatmentsare heading in the right direction. But, current treatmentsoften do not work, or they produce side-effects that requirecorrection. The model of symptoms-diagnoses-treatmentsignores the feedback loop from treatments to symptoms,whereas this feedback loop is the most important factor (16, p. 87).

Sensemaking theory also draws attention to the impor-tance of understanding how individuals coordinate andcollectively work together to discover and refine their under-standings. For example, when a patient shows up in the EDfor care, the different professionals caring for the patient cre-ate a shared interpretation (the diagnosis or differential diag-nosis) about the cause of the patient’s symptoms and thenengage in shared action (the clinical care plan consisting ofdiagnostic and/or therapeutic studies and interventions) totreat the patient’s symptoms. There can be disagreement andeven conflict around what the diagnosis or treatment ought tobe, but, because patient care is interdependent knowledgework, the different professionals’ interpretations and actionsinfluence each other as well as the patient.

WHAT ENABLES EFFECTIVE SENSEMAKING?

If sensemaking is about noticing cues or changes, interpret-ing them to create a plausible story and then acting to testthat story, then more effective sensemaking is about noticingsmall or subtle changes, creating more plausible stories, tak-ing action sooner, and monitoring the result of that actionmore closely. There are a variety of strategies that can be usedto enable more effective sensemaking. We group these strate-gies around the topics of noticing, interpreting, and acting.

Noticing that something has changed is often triggeredby surprise or a violation of expectations, i.e., a discrepancybetween what was expected and what is present. Getting bet-ter at noticing small changes in the patient’s condition ornoticing small errors is an important first step in preventingsmall changes or errors from escalating into large problems.However, noticing changes that prompt re-evaluation ofsensemaking is difficult because individuals tend to privilegeinformation that confirms what they already believe to be trueand, conversely, largely ignore information that challengeswhat they believe (17). How can one counter this tendency?

• Make tacit expectations more explicit so it is easier to spot vi-olations of expectations. For example, make predictions about


81214_CH05.qxd 6/19/08 4:05 PM Page 28

what sort of lab test results, x-ray findings, etc. should be pres-ent or absent if your working diagnosis for a patient is correct.

• Develop strategies to regularly step back and assess what’shappening. For instance, some EDs use systematic reviewstrategies to make sure that all ED patients are being re-evaluated at specific intervals (e.g., “running the board”).Other strategies include periodically reviewing individualcases during patient care and actively searching for discon-firming information.

• Benchmark on atypical literatures, such as the literature onsituation awareness, which deals with how people can be-come more attuned to their environment (18), and literatureon error detection and correction (19).

Interpreting cues involves labeling (8), and labels are help-ful because they can help focus attention and direct action ina particular direction; however, labels can also be limiting.Choosing a more specific label—i.e., “probable cholecystitis”rather than “right upper quadrant abdominal pain”—can leadto premature closure, a narrowing of options before all possi-bilities have been explored (20). In addition, how an event islabeled has important implications for what can be learned inreal time and after the fact (21). Strategies for improving in-terpretation include the following:

• Keep labels broad, especially early in the process of diagnosis.• Appoint someone to be a devil’s advocate, which legitimizes

questioning and alternative interpretations. In academic set-tings, the attending physician may serve this role. Alterna-tively, taking seriously the questions of the patients and theirfamilies may also help improve interpretation.

• Cultivate a variety of interpretations—for example, encouragethe input of other professionals (i.e., ED nurses, respiratorytechs, pharmacists, etc.) involved in the care of the patient. Dif-ferent perspectives increase the requisite variety of the systemand help develop more nuanced and complex interpretations.

More effective sensemaking involves managing a paradoxof needing to take action yet remaining open to change andnot getting locked into a certain pattern of action. But it is easyto get stuck. One of the main barriers to remaining open tochange is that people are predisposed to make sense of theworld through cognitive shortcuts that can bias sensemaking.Researchers find—not surprisingly—that physicians are proneto the same type of cognitive biases as everyone else (22,23).Croskerry (24) reviews the cognitive shortcuts likely to causeproblems in the ED and suggests strategies to minimize bias.

HIGH-RELIABILITY ORGANIZING

WHAT IS HIGH-RELIABILITY ORGANIZING?

Studies of safety and reliability often focus on a subset of or-ganizations that operate in unforgiving environments rich withthe potential for error, where the scale of consequences precludeslearning through experimentation and where complex processesare used to manage complex technology in order to avoid fail-ures. These organizations are known as high reliability organi-zations and include organizations such as nuclear power plants,aircraft carriers, and offshore drilling platforms. High reliabilityconveys the idea that high risk and high effectiveness can coexist,

that some organizations must perform well under very tryingconditions, and that it takes intensive effort to do so. As a briefaside, the abbreviation “HRO” has traditionally been used to re-fer to high-reliability organizations; however, in this chapter, weuse “HRO” to refer to high-reliability organizing. We use thegerund form of the word (i.e., organizing not organizations) toemphasize our focus on the processes of organizing that enablehighly reliable and safe performance.

As evidence about the efficacy of high-reliability organ-izing (HRO) to promote safe operations has increased, so toohas evidence that respectful interaction and heedful interre-lating are necessary antecedents. Respectful interacting is a so-cial process through which one individual’s interpretation iscommunicated to another individual, and, through this com-munication, a shared interpretation is generated. Respectfulinteracting requires the presence of trust, honesty, and self-respect (25). The combination of trust, honesty, and self-respect makes it more probable that people will speak up aboutissues of concern, share their perspective, and ask others ques-tions about their interpretations. And whenever one or moreof these three components are missing, an adverse event ismore likely to occur.

Heedful interrelating is a social process through which in-dividual action contributes to a larger pattern of shared actionand in which individuals understand how their actions fit intothe larger action. More heedful interrelating means that, first,people understand how a system is configured to achieve somegoal and they see their work as a contribution to the system,not as a standalone activity. Second, people see how their jobsfit with other people’s jobs to accomplish the goals of the sys-tem. And third, they maintain a conscious awareness of howtheir work contributes to and how it fits with the work of oth-ers as they go about their duties, sometimes subordinating theirown goals to the goals of the organization in order to carry outthis shared action. Heedful interrelating is perhaps easiest to un-derstand by thinking about its opposite, heedless interrelating—when someone simply does his or her job while ignoring whatis going on around him or her. Heedful interrelating matters forsafety and reliability outcomes; for example, researchers foundthat aircraft carriers had fewer serious accidents when their crewswere more heedful in their relationships (26).

Respectful interacting and heedful interrelating generateshared interpretation and shared action, and are inextricablyand recursively linked. Respectful interacting and heedful in-terrelating are the psychological and behavioral processes thatform the relational foundation for HRO. We draw attentionto the importance of developing and enhancing health careprovider’s abilities to work effectively with their colleagues be-cause without a strong relational foundation, HRO is muchmore difficult to attain.

Respectful interacting and heedful interrelating have realconsequences for patient safety. Researchers are beginning tolink attitudes and behaviors to performance outcomes, such asmedical errors (27,28). When respectful interacting is absent—for example, when trust is lacking—health care providers of-ten don’t speak up about potential errors, either because theythink speaking up won’t make a difference or that speaking upmight harm their image or relationship with their supervisor(11). Vogus (29) demonstrated that higher levels of respectfulinteracting and heedful interrelating were associated with lowerlevels of medical errors and patient falls.

CHAPTER 5 • SENSEMAK ING, H IGH - R E L IAB I L I T Y ORGAN IZ ING , AND RES I L I ENCE 29

81214_CH05.qxd 6/19/08 4:05 PM Page 29

HOW IS HIGH-RELIABILITY ORGANIZINGRELEVANT TO THE EMERGENCY DEPARTMENT?

From a theoretical standpoint, the ED shares many character-istics in common with the aircraft carrier flight deck settingwhere respectful interacting and heedful interrelating were firstdescribed (6,26). Both EDs and aircraft carrier decks are settingswhere there is high interdependence of various aspects of theorganizational system and where the organizational environ-ment is continuously changing, resulting in unexpected, high-variety disturbances. In the ED, highly reliable performance isnecessary, work is nonroutine, and there is a high level of in-teractive complexity. By interactive complexity, we refer to asystem property of high-risk organizations where interactionsbetween various processes can have hidden, unanticipated, orunintended consequences (30). Small problems that initially donot seem consequential can quickly become large problems.

WHAT ENABLES EFFECTIVE HIGH-RELIABILITYORGANIZING?

Studies of the best high reliability organizations (26,30–32)show that their nearly error-free performances result from aset of attitudes and practices that enable their members to pickup on problems earlier and to act on them before they growbigger (33). High-reliability organizing results from processesand practices aimed at (a) examining failure as a window onthe health of the system, (b) avoiding simplified assumptionsabout the world, (c) being sensitive to current operations andtheir effects, (d) developing resilience to manage unexpectedevents, and (e) understanding and locating expertise and cre-ating mechanisms for decisions to migrate to those experts.Collectively, these five processes focus attention on the veryperceptual details that are lost when people coordinate theiractions and share their interpretations (34).

What is distinctive about HRO is that there is a consis-tent effort to recapture detail. There is an effort to refine anddifferentiate existing categories, create new categories, and de-tect subtle ways in which contexts vary and call for contingentresponding.

To cultivate HRO, health care providers must pursuesafety as a priority objective, build in redundancy, decentral-ize decision making, shape culture towards reliableperformance, invest heavily in training and simulation, learnfrom close calls, aggressively seek to know what they do notknow, emphasize communication of the big picture and wherepeople fit into the big picture, and reward people who reportfailures. Some specific strategies include the following:

• Continuously evaluate failures, mistakes, near misses, andclose calls using an appropriate “after action review” proto-col. Winston Churchill’s debriefing protocol may be helpful:Why didn’t I know? Why didn’t my advisors know? Whywasn’t I told? Why didn’t I ask? (see 34 for a more detailedexplanation).

• Develop richer forms of communication. The STICC pro-tocol (15) may be useful in situations such as handoffs: Sit-uation (“Here’s what I think is going on”); Task (“Here’s

what I think we should do”); Intent (“Here’s why”); Con-cern (“Here’s what I think we should keep our eye on”);Calibrate (“Now, talk to me”).

RESILIENCE

WHAT IS RESILIENCE?

Resilience often is defined as “the capability of a strained bodyto recover its size and shape after deformation caused especiallyby compressive stress” or “an ability to recover from or adjusteasily to misfortune or change” (35). Definitions of resilienceevoke a variety of images, for example, a rubber band, whichhas been stretched but is able to return to its previous shape, ora ball bouncing back to a level higher than from where it hadbeen dropped, or a person who has been through a traumaticevent and is able to thrive in spite of difficult circumstances.

Resilience is a concept that applies to objects or individ-uals (36), as well as groups of individuals (37), and organiza-tions (38,39). Resilience is a capability that can be developed;being resilient in the past increases the likelihood that you willbe resilient in the future (38). Only in looking back over a pe-riod of performance can we evaluate whether an individual oran organization has been resilient (40). Evaluating resilienceinvolves comparing how an individual is performing with howthey performed in the past. It is a question of how well and towhat level do they bounce back?

The resilient response can be thought of as the opposite ofthe threat-rigidity response. When organizations respond tothreat in a rigid manner (i.e., they experience “threat-rigidityeffects”), they narrow information processing, increasingly cen-tralize and formalize control, conserve resources, and becomeunable to cope with large and novel challenges (41). In contrast,when organizations respond to threat in a resilient manner,they are able to sustain competent performance by broadeninginformation processing, loosening control, and utilizing slackresources (38). Resilience is characterized by two main capa-bilities: the ability to learn, particularly from mistakes and theability to quickly respond as events change over time and di-vert resources (knowledge, people, and equipment) to wherethe resources would be best used (38).

Although the terms resilience and reliability are used differ-ently in different literatures, they often are used to mean roughlythe same thing. However, the different terms come from dif-ferent traditions and, as a result, highlight subtly different as-pects of the same phenomenon. We use HRO to refer to theprocesses of coordinating action to achieve outcomes with a min-imum of error. We use resilience to refer to the outcome of thoseprocesses and to emphasize the maintenance of stable and com-petent performance under potential adversity or challenge.

HOW IS RESILIENCE RELEVANT TO THE EMERGENCY DEPARTMENT?

What might resilience look like in the setting of the ED? Re-silience means continued competent performance in the face ofa variety of threats, such as (a) external shocks or setbacks to thework of the unit, (b) internal strain or pressure generated within


81214_CH05.qxd 6/19/08 4:05 PM Page 30

the unit; and (c) adverse environmental and societal conditions(37). Put in more concrete terms, resilience in the ED wouldmean that the ED is able to sustain competent performance,without a decrease in the quality or timeliness of care providedto patients, despite various external shocks (i.e., community dis-aster requiring surge capacity in the ED), internal pressure (i.e.,overcrowding in the ED), or adverse environmental or societalconditions (i.e., high levels of ED provider burnout).

By its very nature, the ED poses a particular challenge toresilience. One of the main challenges to highly reliable andsafe performance in the ED is that staff may not be familiarwith each other—this issue is more predominant in teachinginstitutions where there are a number of staff who rotatethrough the department as part of their training and in smallerinstitutions that rely on locum tenens for staffing, especiallyafter hours, weekends, and holidays. Although the word teamsis often invoked when talking about groups of health care pro-fessionals working together to care for a patient, we argue thatteams may not be the most appropriate concept. Much ofthe literature on teams comes from the study of establishedteams—teams where members are familiar with each otherand meet together regularly to plan for an outcome for whichthey are jointly responsible. In contrast, there are frequentlysituations in the ED when the staff assembled to care for thepatient may never have met but are, instead, drawn togetheras a function of their roles, i.e., the trauma team (comprisedof various people from various services carrying the traumapager that day or shift) or the internal medicine team that ad-mits patients to the floor.

As such, interactions in the ED are often based onknowledge about what a person in a particular role—e.g., atrauma nurse or an attending ED physician—is capable ofrather than personal knowledge about the specific health careproviders. Roles can provide valuable clues about what to ex-pect from another person and are often useful in coordinat-ing work (42–45). Meyerson et al. (44) states, “If people intemporary systems deal with one another more as roles thanas individuals—which is likely because the system is built ofstrangers interacting to meet a deadline—then expectationsshould be more stable, less capricious, more standardized,and more defined in terms of tasks and specialties thanpersonalities.”

However, interactions based more on role-specific knowl-edge rather than person-specific knowledge are prone to a va-riety of problems. Lack of familiarity with coworkers can leadto problems with respectful interacting and heedful interre-lating (potential exception if “swift trust” can be established)(44). Lack of familiarity also means that it is difficult for peo-ple to know who the experts are or even who has what knowl-edge. There are often problems in getting groups of peoplewith diverse knowledge to share that diverse knowledge, par-ticularly when they are unsure of how others will react to theirsuggestions (46).

WHAT ENABLES HIGHER LEVELS OF RESILIENCE?

Resilience can occur in the face of dramatic threats, but isalso an everyday phenomenon in the face of day-to-day threatsand perturbations. High-reliability organizing processes aim

both at improving capabilities to anticipate problems and atimproving the ability to cope with errors once they are man-ifest. As Wildavsky (39) asserts: “Where risks are highly un-certain and speculative, and remedies do harm . . . resiliencemakes more sense because we cannot know which possible riskswill actually become manifest.”

Organizational resilience comes from stocks of generalizedcapabilities and uncommitted flexible resources that can be usedto cope with whatever unanticipated harms might emerge. Thisrequires investments in broad generalized training and retrain-ing as skills, such as teamwork, decay over time and need to berelearned and refreshed in order to remain effective (see 46 formore information). It also requires investments in improvingcapabilities to learn quickly, improvise, make do, adapt, andbounce back. But these skills are not necessarily embedded inindividuals. In fact, resilient organizations recognize that it isproblematic to rely on specific individuals and, instead, embedknowledge in roles, in practices, in routines, and in teams. Somespecific strategies to build resilience include the following:

• Encourage people to mentally simulate their work in orderto help them build capabilities to cope with disturbancesonce they appear. What activities lie upstream and down-stream from them? How can their work unravel? How candisturbances be corrected?

• Identify pockets of expertise and encourage people to self-organize into ad hoc networks to provide expert problemsolving when problems or crises appear.

• Encourage conceptual slack—a divergence in team mem-bers’ analytical perspectives and a willingness to questionwhat is happening rather than feign understanding.

CONCLUSION

We suggest that sensemaking, HRO, and resilience are threeconcepts from the organizational literature that provide trac-tion on understanding how individuals can organize in waysthat result in safer and more reliable patient care. If sense-making is more about the interpretations and actions thatinfluence how people enact their way into knowing, thenHRO is more about how groups of people coordinate theirsensemaking interpretations and actions to work together inorder to detect and correct errors early in their unfolding.Resilience—the ability to maintain competent functioningdespite adversity—is the outcome of effective sensemakingand HRO.


SUMMARY

• Organizational and management literature offer threeconcepts that may be useful to improve safety in healthcare settings: sensemaking, HRO, and resilence.

• Sensemaking provides strategies to help health care work-ers successfully manage their clinical work in the error-prone environment of the ED, and make sound decisionsin the face of uncertainty and ambiguity.

81214_CH05.qxd 6/19/08 4:05 PM Page 31

R E F E R E N C E S

1. Roberts KH. Some characteristics of one type of high reliabilityorganization. Organization Science. 1990;1(2):160–176.

2. Perrow C. The organizational context of human factors engi-neering. Adm Sci Q. 1983;28(4):521–541.

3. Weick KE, Sutcliffe KM. Hospitals as cultures of entrapment: are-analysis of the Bristol Royal Infirmary. Calif Manage Rev. 2003;45(2):73–84.

4. Edmondson AC. Learning from failure in health care: frequentopportunities, pervasive barriers. Qual Saf Health Care. 2004;13(Suppl 2):ii3–ii9.

5. Berwick DM. Escape Fire: Designs for the Future of Health Care.San Francisco: Jossey-Bass; 2004.

6. Kohn LT, Corrigan JM, Donaldson MS, eds. To Err Is Human:Building a Safer Healthcare System. Washington, DC: NationalAcademy Press; 2000.

7. Weick KE. Sensemaking in Organizations. Thousand Oaks, CA:Sage Publications; 1995.

8. Weick KE, Sutcliffe KM, Obstfeld D. Organizing and the processof sensemaking. Organization Science. 2005;16(4):409–421.

9. Paget MA. The Unity of Mistakes. A Phenomenological Interpretationof Medical Work. Philadelphia: Temple University Press; 1988.

10. Eisenberg EM, Murphy AG, Sutcliffe KM, et al. Communicationin emergency medicine: implications for patient safety.Communication Monographs. 2005;72(4):390–413.

11. Blatt R, Christianson MK, Sutcliffe KM, Rosenthal MM. Asensemaking lens on reliability. Journal of OrganizationalBehavior. 2006;27:897–917.

12. Benner P, Hooper-Kyriakidis P, Stannard D. Clinical Wisdomand Interventions in Critical Care: A Thinking-in-Action Approach.Philadelphia: W.B. Saunders Company; 1999.

13. Dreyfus HL, Dreyfus SE. Mind Over Machine: The Power ofHuman Intuition and Expertise in the Era of the Computer. New York:Free Press; 1986.

14. Benner P. From Novice to Expert: Excellence and Power in ClinicalNursing Practice. Menlo Park, CA: Addison-Wesley PublishingCompany; 1984.

15. Klein G. Intuition at Work: Why Developing Your Gut InstinctsWill Make You Better at What You Do. New York City:Doubleday; 2002.

16. Starbuck WH. “Watch where you step!” or Indiana StarbuckAmid the Perils of Academe (Rated PG). In: Bedeian AG, ed.

Management laureates: a collection of autobiographical essays. Vol 3.Greenwich, CT: JAI; 1993:65–110.

17. Klayman J. Varieties of confirmation bias. Psychology of Learningand Motivation—Advances in Research and Theory. 1995;32:385–418.

18. Endsley MR. Toward a theory of situation awareness in dynamicsystems. Hum Factors. 1995;37:32–64.

19. Hofmann DA, Lee Z. Error proliferation in teams: factors influ-encing the identification and communication of errors. Workingpaper, 2005.

20. Croskerry P. The importance of cognitive errors in diagnosis andstrategies to minimize them. Acad Med. 2003;78(8):775–780.

21. Tamuz M, Thomas EJ, Franchois KE. Defining and classifyingmedical error: lessons for patient safety reporting systems. QualSaf Health Care. 2004;13(1):13–20.

22. Regehr G, Norman GR. Issues in cognitive psychology: impli-cations for professional education. Acad Med. 1996;71(9):988–1001.

23. Bordage G. Why did I miss the diagnosis? Some cognitive expla-nations and educational implications. Acad Med. 1999;74(10):S138–S143.

24. Croskerry P. Achieving quality in clinical decision making: cog-nitive strategies and detection of bias. Acad Emerg Med. 2002;9(11):1184–1204.

25. Campbell DT. Asch’s Moral Epistemology For Socially SharedKnowledge. In: Campbell DT, ed. The legacy of Solomon Asch:essays in cognition and social psychology. Hillsdale, NJ: Erlbaum;1990:39–52.

26. Weick KE, Roberts KH. Collective mind in organizations:heedful interrelating on flight decks. Adm Sci Q. 1993;38(3):357–381.

27. Thomas EJ, Sexton JB, Helmreich RL. Translating teamworkbehaviours from aviation to healthcare: development of behav-ioural markers for neonatal resuscitation. Qual Saf Health Care.2004;13(Suppl 1):i57–i64.

28. Vogus TJ, Sutcliffe KM. The Safety Organizing Scale: develop-ment and validation of a behavioral measure of safety culture inhospital nursing units. Med Care. 2007;45(1):46–54.

29. Vogus TJ, Welbourne TM. Structuring for high reliability: HRpractices and mindful processes in reliability-seeking organiza-tions. Journal of Organizational Behavior. 2003;24:877–903.

30. Perrow C. Complex Organizations: A Critical Essay. 3rd ed. New YorkCity: McGraw-Hill; 1986.

31. Schulman PR. The negotiated order of organizational reliability.Administration & Society. 1993;25:353–372.

32. Sagan SD. The Limits of Safety: Organizations, Accidents, andNuclear Weapons. Princeton, NJ: Princeton University Press; 1993.

33. Weick KE, Sutcliffe KM, Obstfeld D. Organizing for HighReliability: Processes of Collective Mindfulness. In: Staw BM,Cummings LL, eds. Research in organizational behavior.Greenwich, CT: JAI Press; 1999:81–123.

34. Weick KE, Sutcliffe KM. Managing the Unexpected: ResiliencePerformance in an Age of Uncertainty. 2nd ed. San Francisco, CA:Jossey-Bass; 2007.

35. Merriam-Webster’s online dictionary. Available at: http://www.merriam-webster.com/dictionary/. Accessed February 7, 2007.

36. Bonanno GA. Loss, trauma, and human resilience: have weunderestimated the human capacity to thrive after extremelyaversive events? Am Psychol. 2004;59(1):20–28.

37. Worline MC, Dutton JE, Kanov JM, Lilius JM, Maitlis S.Organizing resilience by cultivating resources: a practice per-spective. Organization Science. In press.

38. Sutcliffe KM, Vogus TJ. Organizing for Resilience. In: CameronKS, Dutton JE, Quinn RE, eds. Positive organizational scholarship:foundations of a new discipline. San Francisco: Berrett-KoehlerPublishers, Inc.; 2003:94–110.

39. Wildavsky A. Searching for Safety. New Brunswick, NJ:Transaction; 1991.


• Effective sensemaking involves managing the paradox ofneeding to take action yet remaining open to change andnot getting locked into a certain pattern of action.

• High-reliability organizations function effectively in high-risk environments and successfully manage high-riskwork. Safe operations rely on respectful interactions thatencourage workers to speak up about issues of concernand encourage heedful interrelating, in which individualsunderstand how their work fits in the larger goals of theinstitution.

• Resilence of an organization is the ability to learn frommistakes and the demonstrated capacity to quickly and ef-fectively respond to threats. Resilence in the ED is mani-fest as competent performance despite adverse conditions.

• Resilient organizations recognize that it is problematic torely on specific individuals and, instead, embed knowledgein roles, in practice, in routines, and in teams.

81214_CH05.qxd 6/19/08 4:05 PM Page 32

40. Hale A, Heijer T. Defining Resilience. In: Hollnagel E, WoodsDD, Leveson N, eds. Resilience engineering: concepts and precepts.Burlington, VT: Ashgate; 2006:35–40.

41. Staw BM, Sandelands LE, Dutton JE. Threat-rigidity effects inorganizational behavior: a multilevel analysis. Adm Sci Q. 1981;26(4):501–524.

42. Baker WE, Faulkner RR. Role as resource in the Hollywood filmindustry. AJS. 1991;97(2):279–309.

43. Callero PL. From role-playing to role-using: understanding roleas resource. Soc Psychol Q. 1994;57(3):228–243.

44. Meyerson D, Weick KE, Kramer RM. Swift trust and temporarygroups. In: Kramer RM, Tyler TR, eds. Trust in organizations:

frontiers of theory and research. Thousand Oaks, CA: SagePublications; 1996.

45. Bechky BA. Gaffers, gofers, and grips: role-based coordinationin temporary organizations. Organization Science. 2006;17(1):3–21.

46. Bunderson JS, Sutcliffe KM. Comparing alternative conceptual-izations of functional diversity in management teams: processand performance effects. Acad Manage J. 2002;45(5):875–893.

47. Morey JC, Simon R, Jay GD, et al. Error reduction and per-formance improvement in the emergency department throughformal teamwork training: evaluation results of the MedTeamsproject. Health Serv Res. 2002;37(6):1553–1581.


81214_CH05.qxd 6/19/08 4:05 PM Page 33

organizational approaches to safety...the challenger launch decision. risky technology, culture, and...

Documents