• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Towards an Organization With a Memory: Exploring the Organizational Generation of Adverse Events in Health Care

Posted on: Sunday, 19 June 2005, 03:01 CDT

The role of organizational factors in the generation of adverse events, and the manner in which such factors can also inhibit an organization's abilities to learn, have become important agenda items within health care. The government report 'An organization with a memory' highlighted many of the problems facing health care and suggested changes that need to be made if the sector is to learn effective lessons and prevent adverse events from occurring. This paper seeks to examine some of these organizational factors in more detail and suggests issues that managers need to consider as part of their wider strategies for the prevention and management of risk. The paper sets out five core elements that are held to be of importance in shaping the manner in which the potential for risk is incubated within organizations. Although the paper focuses its attention on health care, the points made have validity across the public sector and into private sector organizations.

Introduction

A cursory review of the media coverage around the UK's health- care system would lead many observers to believe that it is in a state of perpetual crisis. For health care, the problems raised by the press have included: problems around MRSA, full mortuaries and the storage of cadavers in unsuitable areas, the removal and storage of organs from cadavers without the permission of next of kin, the shortage of trained medical staff, the continued spectre of medical error and, finally, the problems of 'rogue' doctors. These events, while not directly related to each other and with many different causal factors and organizational settings, have prompted some media observers to argue that there is an overall sense of crisis within the health sector and that 'management' (in its various guises and including politicians) has played a major role in the onset of that crisis. What all of these events do illustrate, however, is the fact that there are considerable problems of control associated with the management of health care, and that its core processes around both treatment and intervention will inevitably give rise to a series of emergent properties. In turn, this emergence raises a number of questions about the role of managerial, as well as clinical (expert), assumptions in shaping the various precautionary norms that are in place within health-care organizations. On a closer examination of these problems, several issues emerge that centre on such factors, such as the role of knowledge within organizations, the processes of control, and even the nature of organizational behaviours themselves. These issues are clearly not restricted to the health-care industry but also have validity across the public sector and into private corporations. In addition, they are problems that are often faced in different countries and, therefore, not restricted to a particular ideological setting for public sector management.

In general terms, and especially given the constraints that surround a notion of safety within health care, the industry has a good safety record. In any one year, there will be several million treatments provided to patients, for which there is a positive outcome. There will inevitably be some outcomes that are negative and these are often due to the nature of the patients' problems and the scope that exists for effective intervention. There will be a smaller, but still significant, number of patient treatments that generate adverse outcomes in which the actions of an individual may have exposed a series of latent elements within the system that had the potential for harm. It is this group of activities upon which much of the media attention is focused and it is here that there is a search for culpability and blame, usually around a named individual. Yet many of these accidents are not simply the fault of one person but may arise out of a complex set of systems interactions that create the conditions in which a set of events can escalate quickly to cause harm.

The aim of this paper is to explore some of these issues around adverse events from an organizational-level perspective and the paper suggests that many of the problems around these 'events' occur as a result of the actions (or inactions) of management, rather than simply because of the actions of a 'rogue' or underperforming member of staff. The paper sets out a mosaic of organizational-level factors that serve to help generate a climate within which incidents rapidly escalate into accidents and also prevent organizations from learning effective lessons from the mistakes of the past. While the issues within the mosaic have been highlighted elsewhere,1,2 they have not been developed in detail and this paper seeks to move that initial discussion forwards.

To err is human?

There has been considerable attention by policy-makers on the manner in which errors in health care have resulted in the generation of adverse events. In the USA, for example, estimates have suggested that some 100,000 preventable deaths occur each year.3 The causal nature of such deaths is both variable and difficult to determine with accuracy, and there has been considerable discussion around the nature and extent of those preventable deaths.4 Barach and Small3 also suggest that there is a considerable amount of underreporting associated with adverse events and that this may be in the order of between 50% and 96%. In the UK, the publication of a pilot study, in June 2002 by the National Patient Safety Agency, suggested that there may be as many as one million errors per year and that one patient in 10 may be exposed to some form of medical error during treatment.5 The UK's Department of Health6 has concluded that a rate of adverse events exists which is in the order of 10% of admission per year and that these events cost the service over 2 billion. Even given the acknowledged ambiguity surrounding such data and the inherent problems that are associated with its reliability, it is clear that health care generates considerable potential for catastrophic failure as a function of its routine biomedical activities. Against this background, policy- makers and politicians have begun to explore the manner in which such adverse events can occur and have attempted to consider the possible preventative strategies that can be put into place to alleviate the problem.

The publication of 'An Organization with a Memory' in the UK6 and the US report, To err is human'4 have both marked the culmination of considerable and sustained attention by various health-care professionals into the processes by which failure has occurred within the health sector. More to the point, these reports have set a firm platform for subsequent policy initiatives aimed at reducing adverse events. These reports have also attempted to address many of the core issues that are involved in the processes of learning from adverse events and, in particular, the processes through which managers and organizations deal, or fail to deal, with the precursors of failure and the early warnings of problems. The rationale behind such an approach is that a failure to learn lessons from adverse events across the sector may lead to the incubation of such potential failures within a specific health-care setting. Managers, therefore, need to constantly ask the 'what if question, in terms of their own systems and processes, in order to deal with the manner in which the incubation of such failure occurs. Unfortunately, many of the issues that need to be dealt with in this process of failure prevention and learning are both difficult to define or, in some cases, to manage, and it is with an examination of this contentious area that this paper is concerned.

One aim here is to explore the manner in which the potential for adverse events can be incubated within organizations. Clearly, this is a broad topic and one that cannot be adequately covered in a paper of this nature. Our concern is, therefore, with macro-level issues rather than with specific issues of human error, failures with medical devices, and the range of unethical behaviours on the part of individuals. Our focus is at the organizational level, a perspective of analysis that is often ill understood and poorly researched, especially within health care. Obviously, this paper cannot provide neat prescriptions for such a potentially broad range of organizational malaise, but rather sets out an agenda for further discussion and consideration in the hope of spawning further research and discussion. Although this paper now progresses with a specific focus on health care, it is important to point out that the issues raised here will have a saliency across the public sector and into the private sector. It is our contention here that if managers are going to be effective in dealing with the problems of adverse events, then they will need to draw upon lessons from outside their own domains of control in order to challenge the ways in which they seek to manage risk. In order to provide a shape to that discussion, it is first necessary to set out the boundaries of the problem that we believe are important in framing the nature of the debate.

Determining boundaries: framing the issues

Any examination of adverse events in health care demonstrates that there are a number of important elements w\ithin organizations, that are, unfortunately, often both obscure and ill defined. Yet, these factors are central to the safe operation of the organization and its activities. Indeed, the very nature of management itself is often fraught with ambiguity and confusion, especially around those issues for which simple cause-and-effect relationships do not exist. In health care, such general problems of management and control are often exacerbated by the hazards involved in dealing with patients who, by definition, may be in a perilous state at the point of clinical intervention. Patients interact with the 'system' of health care at the point at which they themselves are often vulnerable and in a weakened state that requires treatment. The multiple actors and agents involved in the delivery and consumption of health care create considerable potential for the emergence of many of the precursors of adverse events.7,8 Thus, it can be argued that the treatment of patients, by its very nature, will generate the potential for adverse events. If we add to this potential for risk the problems that are associated with the further layers of organizational complexity that arise from resource constraints, impaired communication flows, and the vagaries of control around systems' emergence in complex nonlinear systems, then it becomes clear that health care also offers a significant challenge to many of the conventional theories that we hold about management.1,2

The complex mosaic of issues that is generated by this interaction does not easily lend itself to the simple, protocol- driven prescriptions for the prevention of organizational malaise that much of the more popularist management press often seems to extol. Instead, problems within health care can be seen to represent a challenge to much of the dominant theory within management and also to raise some fundamental questions concerning both the primacy of control within organizations and the processes by which decisionmaking is undertaken, especially around risk.7 Consequently, it could be argued that health care also exposes the limitations of contingency approaches to the management of risk and even raises questions about the role of management as a core function within organizations. Given the constraints of this paper, it is only possible to explore a small number of issues here. A matrix of issues concerning the relationship between health care and the potential for adverse events has been identified1,2 and it is these issues that will form the basis of our discussions here.

The first element within the matrix concerns the manner in which management as a process interacts with the practice of medicine. There are powerful structural constraints that impact on clinical practice and are often held to arise from the processes and practice of management. A central dynamic of this process concerns the notion of expertise within management relative to that within medicine. To what extent do 'experts' within health care allocate sufficient time to management issues (as opposed to clinical problems) and to what extent have these individuals been trained in the processes of management? Should managers be considered 'experts' within the processes of managing, especially when that activity impacts on issues of public health and the practice of medicine. Notions of expertise and the associated networks through which that knowledge is transmitted should, in theory, be important within the prevention of the conditions that can generate adverse events. Expertise and knowledge networks should allow for improvements in information flows and enhance decision-making. However, it is clear that many adverse events have occurred because experts were unable to see beyond the constraints of their own particular paradigms and disciplinary boundaries.

The second element in the matrix concerns the manner in which adverse events become incubated within organizational processes and activities and the ways that they escalate are not often discussed within management teams. Despite such 'neglect' within mainstream management thought, the triggers for adverse events and the manner in which they expose the 'pathways of vulnerability'9 embedded within the organizational system provide important challenges to the principles of control within management theory. Again, it could be argued that the very processes of managing may well inhibit the abilities of management to prevent and control adverse events.

A third dynamic concerns the ways in which organizational problems are evaluated and assessed. Our attempts at making sense of the world in which we work is an important, but often neglected, aspect of organizational life.10-12 Sensemaking, as a concept, is central to our understanding of organizational decision-making and the role of both cognitive and social processes therein. Challenging the dominant world view that prevails within organizations is essential if decision-making around adverse events is to be effective. However, before we can challenge those assumptions, we also need to make certain that we have an understanding of the processes by which incubation occurs and the associated dynamics of emergence. This is often an area in which organizations face considerable difficulties, especially in terms of dealing with the emergent dynamics of organizations and ensuring that management control systems are not bypassed or eroded.

Our fourth element, the notion and nature of behaviour within organizations, is also one that has not provided sufficient critical insight into the processes by which organizations operate in practice. The notion of the 'way in which we do things around here' has been a major factor in the incubation of the potential for adverse events and that dominant organizational paradigm often sits at the heart of incubation. The oft quoted, but often ill-defined, and poorly understood notion of an organizational culture serves to frame the behaviour of individuals working within an organization. It is here that notions of organizational behaviour - or more accurately, accepted organizational behaviours13,14 - is important. This notion of accepted behaviour is an important one within the context of our present discussion, as it is often based on perceptions of behaviour rather than on the reality of organizational life. Management theory has been slow to recognize the importance of organizational (mis)behaviour13 in shaping the way in which organizations function.

Finally, the apparent inability of organizations to learn effective lessons from adverse events is a generic problem. In health care, the plethora of agencies and professional bodies can create difficulties in the dissemination of lessons from events and this combines with such constraints as patient confidentiality and the previous lack of any major systematic reporting mechanisms to establish quite significant barriers to learning. The resulting mosaic of causal factors is shown in Figure 1 and it offers us a framework through which to discuss a series of issues facing health care.

Ultimately, our aim is to explore these issues in more detail and to suggest areas in which it might be possible to enhance current managerial practice with regard to failure prevention. While our intention is to raise issues for health care, the paper does not limit its analysis to health as a means of framing the discussion; instead, it seeks to adopt a more generic view of the issues and draws upon material from outside both the health care and, indeed, public management literatures. Inevitably, this paper will have its inherent limitations and these need to be outlined at the outset.

Figure 1 A mosaic of causality for adverse events2

At its most basic level, this paper should be seen as an early attempt to frame the issues relating to the incubation of adverse events and to broaden it into the wider literature on risk and crisis management. Consequently, much of the literature drawn upon here comes from academic disciplines outside medicine. Secondly, the issues raised here provide a baseline set of theoretical issues that will require empirical verification and, ultimately, translation into management practice. Finally, in complex adaptive systems, typified by health care, the sheer dynamism of the process needs to be framed in both space and time as well as in the cultural setting of the organization. Put simply, we have to question whether we can readily transfer learning from one situational context to another? These caveats need to be borne in mind when discussing the issues raised within this paper. As a means of framing such a discussion, we can now explore the five elements of our proposed mosaic of issues beginning with the nature of management itself.

Element 1: Management and medicine: incubating the potential for adverse events?

The reasons for an initial focus on the processes of management are quite straightforward. Management, it is argued, lies at the heart of the problem of organizational failure and, while most of the solutions to such problems are often portrayed as being technical, some 70% of failures are seen to arise from human intervention,15-18 which includes a considerable proportion that are shaped by management actions and inactions. At the same time, management, almost by definition, exists to cope with uncertainty and to deal with the problems that arise from emergence within socio- technical systems. Thus, the notion of control has been an important aspect of management theory and an assumption has long been made that the actions of individuals can be managed (and therefore controlled) in an effective manner. At the level where organizational members interact with external groups or over long time periods, such control strategies become much more problematic. Similarly, the issues relating to decision-making, where cause and effect relationships are often difficult to obser\ve or where powerful interests are involved, also generate issues for control. In some cases, the notion of efficiency expressed in terms of rapid decision-making and the production of goods and services with minimal resource demands - has overshadowed the process of effectiveness, with the latter often proving to be difficult to measure as a function of the non-linear nature of the process. The interactions between 'management' (as a process), the mechanisms and extent of 'control', and the tensions between efficiency and effectiveness will combine to create a set of assumptions about the ways in which the organization functions 'in theory' and may ultimately lead to failures 'in practice'.

Clearly, there are also difficulties here that are generally associated with the manner in which the management both makes decisions and implements them. Consequently, our focus will be on elements of those processes and herein lies an initial paradox. If management exists to deal with uncertainty - that is, to cope with the 'shocks' that impact on the routines of organizational activity - then the attempts of managers to control that uncertainty may well result in it being increased elsewhere within the system (as spatially defined) and across the decision timeline. As a result, short-term responses to problems may well generate long-term consequences for organizations to manage.

Despite its obvious role in control (especially in terms of the control of human elements of the organization), management has found it difficult to cope with the role of human intervention in accident causation (and especially its own role in the process). The result is that those who are deemed to be 'directly' involved in the final stage of an adverse event are often seen as the main culprits. Such a person-centred approach fails to take account of the broader issues of culture, communication, and latent management error in precipitating such events. In part, much of this narrow focus stems from the nature of management itself, the assumptions held by senior managers, and the bounded nature of their decision-making.

The dominant approach taken within management is that of contingency theory. Put simply, management is a process of planning and control and the management of adverse events is often seen as being dealt with by developing plans to cope with such eventualities. Management, therefore, seeks to develop plans to cope with (perceived) likely events and this is influenced by a number of factors. Hrebiniak,19 for example, observes that:

'There is one clear assumption underlying contingency approaches to the design and management of organizations: organizational structure and managerial practice depend on the situation. Organizational design and the process of management are not independent variables that can be described for all organizations. They are subject to the influence of various factors, most notably the technology and environment of the organization' (Hrebiniak, 'p. 108')19

This primary focus on order and control is clearly influenced by broader factors that are difficult to constrain - such as environmental factors and technological change. Within a closed system (which are invariably rare in practice), control might prove to be a relatively simple, and possibly achievable, goal. However, in those complex, open, non-linear systems that are typified by those dealing with health care, attempts at control are severely curtailed due to the number of network relationships and interactions that take place.

The interactions that occur within organizations can be seen to generate problems for management control at three levels.2 In the first instance, the sheer number of interactions is difficult to monitor and control. A typical acute hospital, with its various departments and 'client groups' (These would be wider than a patient grouping and would include, for example, other health professionals, social service [and possibly education] departments, primary care trusts and policy makers among other stakeholders), would have a multitude of interactions taking place at one time. Many of these interactions would not be systematically recorded or captured by management control systems. These networked interactions bring with them issues relating to information transfer, communication issues, and the risk of reputation damage. At any point in time, therefore, management may well be unaware of a range of interactions and relationships that, strictly speaking, fall within their supposed sphere of control. Secondly, the sheer scale of such interactions means that it is often difficult for managers to ensure that they occur in the manner that was initially intended and planned for within the 'optimal' operation of the activity. Gaps and fissures within control systems may emerge because of the difficulties involved in monitoring and adjusting such relationships.20 In practice, complex organizations may well be 'effectively' prevented from acting on feedback concerning performance because they are unable to respond to early warnings of problems and adjust their 'behaviours' accordingly. This is not simply a case of requiring greater analytical tools, but also reflects the need for more effective information capture and dissemination mechanisms. Finally, the more complex and numerous the interactions, the more likely we are to see the development of emergent properties - that is, outcomes and interactions that were not foreseen prior to the event and for which no control mechanisms exist. In a complex dynamic and open system like health care, the problems associated with emergence are likely to be considerable.

In addition to these factors, more general difficulties associated with the overall prescription of how to manage have also plagued managers and decision-makers. At the core of this problem has been the issue of what management as a process actually means in practice. While it is relatively straightforward to identify those individuals who are classified as managers within an organization, the question about what they do is much more problematic. In health care, the 'duality' associated with the 'medical manager' may well create a set of difficulties that confounds this problem still further.1,2 Putting this potential duality to one side, we can argue that the practice of management creates an interesting and complex set of issues with regard to adverse events.

There have been several attempts to capture the nature of the management process and the manner in which organizations function.21- 26 Despite such efforts, there is a need to recognize that most definitions of management will fail to capture its essence in practice rather than theory and, as such, will invariably be flawed. This will result in a failure and will fail to capture the richness that is inherent within the process.

Management clearly has both an upside and a downside and it is the interaction between the two over time that gives rise to problems.8 For example, while planning is clearly a positive attribute of management, its role in dealing with adverse events has to be qualified. If planning is based on a series of false assumptions, or fails to cope with emergence, then it is clearly going to be limited in its abilities to prevent failure. Similarly, knowledge can also be based on a series of false assumptions and may reflect the broader issues of boundary limitations and the influence of powerful lobbies within the decision-making process. What emerges is a complex, dynamic web of interactions that will create fundamental difficulties for the management function, and which arises almost as a result of the workings of the management process itself. These elements of "management" can often be seen to revolve around the personal characteristics of the individual manager - their personality, training, and experience will all be important in shaping the relative importance of the various elements within their mode of managing. Similarly, the way in which the individual manager orientates his/her self to others within the organization and the effectiveness of the various teams that operate within the organization will all have an impact on the effectiveness of the process.27

With regard to the downside elements to management, it is important to note that the boundary between success and failure is often blurred and organizations can move from excellence to failure with, what often appears to be, apparent ease.8 Perhaps the most notable among these downside elements are problems surrounding the role of the assumptions and core beliefs held by senior managers,28,29 the impact of compliance and power,21,30,31 the binary views of many managers,32 and, finally, the nature of emergence and boundary limitations on the precipitation of failure9,15-17,33

In recent years, these concerns have taken on a new dynamic as it has become clear that organizational activities have the capability of causing considerable damage - not just within their immediate 'site hinterland', but across society as a whole and, increasingly, in an intergenerational manner.34-38 The response of 'management' to such wider concerns about hazards associated with their activities has often been manifested in ways that blame individuals (often those whose direct actions were perceived to have precipitated the failure) without considering the role of wider, less tangible issues in accident generation.

Health care is no exception to this process. Leape, for example, has criticized medical management for the generation of a dominant culture that is often blame-centred and in which errors are seen to involve a lack of attention or professionalism.39 Such a personcentred approach to adverse events has generally been held to suit the agenda of management - it allows individuals (rather than the wider system in which they work) to be identified, 're- educated' or removed. In more \recent years, a more critical perspective of error within organizations has seen the root cause of the problem as being embedded within both management systems and protocols and the overall strategic direction (including the dominant culture) of the organization. Leape also points to the significance of these wider processes by observing that there is a need to focus on root cause and systems level errors rather than on the unsafe acts themselves.39 Thus, failure can be seen to sit at the heart of the organization rather than as something that can be attributed to the actions of a rogue or inept individual. Indeed, there are some who have argued that such (mis)behaviour can also be generated by management actions.13 As a consequence of this realization, organizations (as well as senior managers within them) have been called to task for their actions in ways that hitherto would not have been considered. Evidence for such a shift can be found in the 'resignation' of Gerald Corbett, as the chief executive of Railtrack, following the crash at Hatfield in October 2000. More recently, the collapse of Enron in the USA has also raised questions concerning the role of senior management in organizational failure and collapse. It is likely that this sort of high-profile departure of senior managers will continue as their role in shaping the underlying culture of failure within organizations becomes more broadly accepted.

Another aspect of this changed perspective on "management" has been the recognition that risk and its control is not an objective process, which sits within the realm of science, but rather is something that is largely socially constructed, open to challenge and based on uncertain and often ill-defined knowledge.34'35 The public inquiry into the BSE crisis made it clear that managers and policy-makers should be more precautionary in their approach and also be willing to express their concerns and uncertainty to the organization's various stakeholders. The combined result of these factors is inevitably an organizational system in which there is a gap between the preventive measures and norms that management adopts to control the activities of the system, and the problems that emerge from the interaction of organizational elements.40 It is this relationship that Turner15-17 identified as a key factor in the manner in which the potential for adverse events can be incubated within organizations. Reason41-47 extends this argument and observes that the exposure of gaps within organizational defences can result from either active failures involving the behaviour of individuals, or latent failure due to problems relating to the management or the design of the system itself. Reason45 argues that the overall culture of the organization will influence and shape local climate conditions which will, in turn, allow errors and violations on the part of individuals to occur. At the same time, the actions (or inactions) of the management may create gaps within organizational defences4 or may allow emergence to go unchecked, thus creating 'pathways of vulnerability'.9,48,49 The end result is that violations and errors on the part of individuals expose a series of managerial and organizational weaknesses that, in turn, allow for the escalation of incidents into adverse events. As Vincent and Reason50 observe (p. 43):

'the crux of the matter is this: we cannot prevent the creation of latent failures; we can only make their adverse consequences visible before they combine with local triggers to breach the system's defences'.

The processes surrounding incubation and latent error form part of our second element of the mosaic.

Element 2: Incubating the potential for harm: control, emergence and latent conditions

'Given the volume and complexity of patient care provided by a modern healthcare organization, some serious lapses in standards of care are inevitable' (Donaldson, p. 218)51

We have argued thus far that one of the greatest challenges to face the management process is the prevention of failure. Given the interactions between patients, medical staff, administrators, external agencies, regional and national bodies, royal colleges and the like, one might expect to find significant fractures in organizational control systems within health care. As the quote from Donaldson illustrates, there is a degree of recognition that many of the conditions that underlie mistakes clearly exist within the provision of health care and one might argue that direct action is required by management in order to prevent such adverse events from emerging. The limitations of management with regard to issues of risk have also been widely discussed within the literature,8'45'46'52'53 although less so with reference to medicine until recently. As stated earlier, a central element of this process concerns the manner in which organizations can incubate the potential for failure as a consequence of their 'normal' managerial activities and decision-making processes. This process of incubation was first identified by Turner,15-17 who found that organizations developed precautionary norms for the control of risk primarily as a function of the core beliefs and assumptions of managers and operators working within that system. For health care, this raises some important issues such as: the manner in which professionals interact with each other, the processes by which new entrants to the profession are trained and supported, and the belief in the primacy of expertise and 'scientific' knowledge (as opposed to the experiential knowledge of patients). Each of these elements forms part of a patchwork of factors that interact together to generate fissures and fractures within controls and allow organizations to embed vulnerability within their procedures and processes.49

Incubation within health care is a more complex process that is compounded by the fact that the various organizations comprising the service operate within different cultures, work under different management styles, and deal with a diverse range of patient-centred problems. These environmental, organizational, and managerial characteristics interact together to create a 'matrix of emergence potential'2 in which interactions between the various elements occur at both a macro and micro level, and subsequently create those pathways of vulnerability,9 which then bypass organizational controls. Given the nature of the interventions with patients, the impact of the various groupings and their associated professional affiliations, and the impact of the culture of medicine as a profession, it is clear that complexity plays an important role in shaping the nature of incubation within the sector. Turner's incubation concept has been widely accepted and incorporated into the broad literatures on crisis management and human error.28'54'55 A major dynamic of incubation is the notion that it takes place in a manner that is generally unobserved by those who work within the organization.1 As a consequence, it is inevitably difficult to manage.

The core elements of Turner's incubation processes are shown in Figure 2 and elaborated upon in Figure 3. At the core of incubation lie the issues of management core beliefs (and their inherent rigidity), the perceptions of managers concerning the nature and level of risk, and the manner in which they often seek to perceptually minimize the hazards that they face. Outside this inner core, there are a number of other issues that build on these central elements. The first of these concerns the nature of the information flows within the organization. Turner sees these problems in terms of the difficulties surrounding complaints from those who sit outside the management group (often seen as external critics, but increasingly in terms of internal 'dissidents') and the problems in information dissemination. In terms of the latter, there are obvious issues around the nature of knowledge, particularly under conditions of uncertainty, and the manner in which tacit knowledge is disseminated and made more explicit within the organization, especially across boundaries between expert groups. This is important in providing for an effective challenge to the dominant world-view that is held by senior managers and other organizational gatekeepers. The external issues, which comprise the final group, centre around the role of external agents within the incubation process - outsiders, the regulators, and those events and phenomena that distract managerial attention away from the core issues. While these elements of the incubation process can allow for the incubation of disaster potential, they can also impact on the organization's ability to learn from, and therefore prevent, such events from occurring, a point that will be returned to later.

Figure 2 Turner's model of disaster incubation adapted from Turner15,16

Figure 3 Core elements of incubation (after Turner15,16)

One of the central elements of Turner's model concerns the role of managers' core beliefs and values in shaping the agenda of 'risk- based' issues that they will consider as feasible. The whole notion of the 'worst case/most credible accident scenario', which is so often used in risk assessment, reflects the dominant paradigm of accepted failure modes and their associated effects. As a process, risk assessment is based on the analysis of data concerning the probability of various failure modes and their associated consequences. Such a process works well where there is a considerable amount of failure data for analysis. For 'unique' events, however, there is inevitably little or no previous experience of the failure mechanisms for managers to draw upon in their decision-making. On the basis of available evidence, therefore, there would seem to be little risk and, unless a precautionary approach is adopted, no action would be taken. Such an evidential basis of decisionmaking is a core eleme\nt of the rational-positivistic approach to management and can represent a significant barrier to learning. Unless evidence is available, which supports a challenge to the dominant paradigm, it is likely that pressure will be brought to bear on management to do nothing that undermines existing practice. Similarly, in medicine, criticism has been expressed concerning the reluctance of some clinicians to consider causal factors for disease that lie outside of their experiential base. In many respects, this gives us some insight into the manner in which core beliefs translate into precautionary norms, at both the individual and organizational levels. Such precautionary norms can provide a major barrier to any attempts at challenging the manner in which organizations work and, as a consequence, can also serve as a major barrier to effective early warnings and near-miss reporting. The challenge for management, therefore, should be to explore this process of incubation from a practical perspective and seek to mitigate its consequences in the process. Allied to this is the need to ensure that effective learning takes place within the organization - or, as Turner puts it, to ensure that full cultural readjustment occurs.

One issue that needs further consideration here is the manner in which we make sense of the complexity of adverse events. In particular, we need to consider the processes by which hindsight can distort our attempts at sensemaking for complex, emotionally charged, adverse events. The catastrophic failure of systems and the subsequent analysis of their main causal factors often bring with it a change of 'hindsight bias'. Had managers been aware of the key issues around causality at the time of the event, (of course) their actions would have been different. What is important, however, is that this hindsight is translated into effective organizational learning. It is here that further investment is required by managers in ensuring that the lessons learned are effectively incorporated into the organization so as to change the systems and procedures within it. Hindsight is undoubtedly one of our most important and possibly most costly sources of information, both in terms of the lives lost and the tangible and intangible assets that are damaged in the process of gaining that information. It follows then that, we should attempt to gain as great an understanding of such information as we can when it is presented to us and seek to use it as effectively as possible so the benefits of that information are maximized and any further unnecessary 'costs' associated with failure are kept to a minimum. Adverse events and, perhaps more importantly, near-miss incidents provide us with considerable learning opportunities for the management of risk. Unfortunately, however, organizations are often incapable of learning, and the process of 'full cultural readjustment' advocated by Turner is often absent in the aftermath of many adverse events. One significant barrier to this process is the manner in which we fail to make sense of the broader organizational processes that gave rise to the adverse event.

Element 3: Making sense of catastrophe: much ado about....?

The complex and opaque nature of modern organizations creates particular difficulties in terms of our limited abilities to make sense of the array of problems that face managers and operators. The sheer scale and scope of these problems often combine with their inherent interdisciplinary nature, to create risks that are typified by their speed of onset, the sense of interconnectivity within them, and the role of latent factors in shaping the dynamics of their emergence. This emergence often occurs in such a way and at a pace that causes major difficulties in sensemaking, interpretation, and response. The manner in which organizations deal with the initial emergent stages of an adverse event can also determine the processes through which an incident can escalate into a crisis. 9 The interpretation of events and the initial response to them can generate conditions that may bypass existing controls and create new emergent 'pathways of vulnerability'.48,49 Part of the problem here lies in the nature of organizational processes and the difficulties that exist for operators and managers to conceptualise the interactions that occur between elements of a non-linear system.

Modern organizations, by their very nature, are both complex and adaptive. It is true to say that this complexity ensures that there are few people, if any, within the organization who will have a detailed understanding of the nature of the system that is in operation across the range of the organization's activities. Even with modern methods of communication and data collection, the ability to process and disseminate such information impacts on our ability to make sense of the mass of data that is presented to us. New forms of data analysis may well mask the underlying uncertainty and possible ambiguity that exists within the data set and this may create a false sense of security for managers. Complex adaptive systems are typified by their ability to process large amounts of information in a manner that ensures that decisions can be made quickly. Any errors that are present within the data-set (or the analysis) will invariably impact quickly on the decision-making process. An additional problem can be seen in terms of the manner in which the schema of key decision-makers may shape their responses to particular sources of information. There has been a well- established body of research into the manner in which such information is processed, and the role of our 'mental maps' of complex events and situations in that process.56-60 The combined impact of these factors will be to shape the manner in which we interpret and make sense of the world in which we work. Sensemaking is an important facet of organizational performance and is clearly essential in shaping our response to risk.10-12,61 Clearly, sensemaking becomes particularly problematic when we are dealing with adverse events that can challenge our assumptions about the ways in which our organizations work. In essence, we have to overcome our assumptions about organizational controls (which will probably have failed) and the role that we might have played as an individual in that process. Such factors have been important in shaping a number of transport accidents in which the operators became confused and were subsequently unable to make sense of the events in which they were embroiled.9,11,27,61-63 In health care, the issues of sensemaking are clearly of importance, as a series of events involving medication errors, misdiagnoses, and the control of medical staff, all testify. Despite the importance of sensemaking within adverse events, it remains an under-researched area within health care.

Element 4: (Mis)behaviour and management: issues in organizational culture

Management theory is heavily populated with work that discusses the notions of organizational behaviour. Most of this work deals with issues such as motivation, the role of personality in team performance and the like, but critics have observed that little attention is given to issues of organizational misbehaviour, despite its prevalence within organizational life.13 Health care has been witness to some high-profile examples of such misbehaviour but, like all organizations, will also be subjected to (mis)behaviour that does not have malicious intent at its core. Indeed, one definition of misbehaviour frames it in terms of doing anything at work that you are not supposed to do.64 This would include a range of behaviours such as the breaking of rules and protocols, displaying negative attitudes to coworkers, and the approaches taken towards making decisions. However, it has been suggested that our role and position in the organization will determine the perception that people have about the classification of actions as misbehaviour, for as Ackroyd and Thompson13 comment:

'The key difference between managers and other employees.....is that employees are likely to have the results of their deviations from expected standards of conduct noticed and defined as misbehaviour.' (p. 3)

For health care, with its pronounced hierarchies and authority gradients, such a comment assumes a particular significance.

One other area of importance here is the role that management has in shaping the behaviour of others. While there is a growing acceptance that management can shape the potential for active error through the development of protocols and the culture within which people work, there are other ways in which management can shape the potential for misbehaviour. One important observation is that management control, while not proving to be particularly effective at eradicating many of the processes that underpin misbehaviour, can serve to shape and create the form of misbehaviour that takes place by generating motivational factors that spawn violations of protocols in order to 'get the job done'. Management has often also shown a reluctance to modify its own behaviour: 'there is little reason to think that it will eliminate misbehaviour, because it does not act on the tendency to misbehave itself'13 and one only has to consider the whole issue of management 'ethics' to find some support for such a claim. In health care, the hierarchies that are in place create a further barrier, as it is difficult, if not impossible, for junior colleagues to challenge the ethical stance of those further up the hierarchy61,65-67

While the media focuses attention on those elements of misbehaviour that are inevitably extreme,6,14,52,68-70 there is also a considerable amount of lower-order misbehaviour that is potentially problematic. This includes such factors as the breaking of protocols, the abuse of power, position, and authority, along with a range of judgemental errors, all of which \have the potential to generate adverse events within a clinical setting.4,14,71,72 Misbehaviour within health care also remains an area in which further research is needed if we are to understand more fully its relationship with creation of adverse events beyond that of the 'rogue' clinician. A first step on this path is to acknowledge that misbehaviour does occur, that it is widespread within health care (as in all organizations), and that the cause and effect relationships between misbehaviour and adverse events are not likely to be simple and transparent.

Element 5: Breaking down the barriers: barriers to learning and the incubation of failure potential

Our final element in the mosaic is concerned with the processes by which organizations can learn from both their own mistakes and the mistakes of others. A litany of crisis events over the last 20 years provides us with tangible evidence that there are a significant number of organizations that fail to learn the lessons from either their own or other people's mistakes. They also seem to be more than capable of ignoring explicit or 'near-miss' warnings that a hazardous situation exists. One such example includes the launching of the ill-fated space shuttle Challenger on 28 January 1986, where early warnings about the risks of failure were ignored.73 There are several difficulties that can hinder or even prevent such learning from taking place and this provides a significant challenge to management within health care, where the pressures to conceal one's mistakes and misbehaviour are considerably high.

One explanation as to why organizations sometimes behave in such a way is that they may have engaged in what Toft and Reynolds have categorized as 'passive learning'. This occurs when members of the organization are fully aware that current circumstances make them vulnerable to an unwanted event taking place, but decide not to change the situation because they underestimate the risk of injury associated with that hazard.74 As risk is, almost inevitably, a social construct, it is open to multiple, and often conflicting, interpretations. 5 Thus, our perceptions and sensemaking capabilities for such events are important in shaping the ways in which we react towards them both prior to an adverse event occurring and in its aftermath. In addition to the manner in which we deal with our perceptions of risk, there are also a number of other factors that influence our ability to learn effective lessons from failure, and these need to be explored briefly.

A number of authors have identified the importance of personal perceptions of invulnerability when explaining their reluctance to accept the risk of failure.28,29 It would appear that people sanitize potentially injurious events from their consciousness and, in so doing, develop the illusion of seeming to make the world a safer place. However, such a process also creates the conditions in which those individuals may well fail to recognize the significance of a whole series of warning signals and, therefore somewhat paradoxically, allow a dangerous situation to get worse. Denial and cognitive narrowing, therefore, are two powerful pathologies that allow incubation to take place.

Table 1 Potential strategies to overcome pathologies of failure7,8,49

Table 1 Potential strategies to overcome pathologies of failure7,8,49

In those individuals and organizations where an aversion to negative learning is strong, it may be difficult for learning to occur that draws upon the lessons learnt from mistakes elsewhere. For example, people often prefer to search for successful outcomes and then learn to copy that behaviour rather than identify actions or strategies that they should avoid repeating. This explains, in part, the prominence of 'template management'7'8 approaches to dealing with complex issues in organizations. There has been an alarming trend for managers to copy successful practices elsewhere, often without reflecting on the impact that local conditions and culture might have on the success of those strategies.

Another factor that adds to the creation of incongruent perceptions concerning risk is the strength of a person's beliefs about the 'facts' surrounding a particular situation. Such a process among senior managers can lead to the embedding of these beliefs within the culture of the organization and may lead to the creation of a crisis-prone organization as a consequence.28 This creation of a set of cognitive and emotional barriers to change can be a powerful process leading to cognitive blindness on the part of managers. These pathologies can come together in the process of groupthinking and have been held to be responsible for a number of examples of poor decision-making, resulting in catastrophic failure73,76-78

In reviewing the literature on learning, Smith and colleagues7'79 have identified a number of factors that seem to impact on the abilities of some organizations to learn effective lessons from near- miss events and accidents elsewhere. Within the context of incubation, it can be argued that such failures to learn will simply serve to reinforce the future potential for other adverse events through the continual incubation of the precursors of failure. The barriers listed in Table 1, along with potential strategies for intervention, provide a means of framing and surfacing potential causes of incubation within an organization, although much will depend on the ability of individual managers to reflect on the implications of these barriers for their own managerial practice. A detailed discussion of these issues is beyond the scope of our current discussions and, again, is an area that requires further research within a health-care setting.

Conclusions

Our discussions within this paper have identified a series of major obstacles to developing an Organization with a memory', as a means of preventing the incubation of adverse events. One of the key points to emerge from this discussion is the importance of situational context in the determination of incubation, Although it is possible to identify broad frameworks within which to evaluate failure potential, it is argued that each organization needs to ensure that it addresses the various pathologies within the context of its own environmental conditions and situational circumstances. A failure to take account of local conditions and problems will inevitably create difficulties in terms of the extent of the failure portfolio that the organization seeks to manage, Part of the dilemma for health care concerns' the very nature of the medical manager and the demands placed on them. The key skills required of such individuals in terms of both clinical and managerial competencies are extremely high. This, combined with the risks associated with the practice of medicine, the complexity of health-care organizations, and the processes of emergence within non-linear systems, all combine to create a complex and potentially hazardous managerial setting. The creation of pathologies of failure should be seen as an almost inevitable facet of organizational life within health care, and management needs to ensure that there is a constant search for means by which such pathologies can be removed. This would represent the first step at dealing with the mosaic of issues that underpin adverse events.

Finally, when evaluating the results of someone else's mistake, perhaps it would be useful if society as a whole were to reflect on the observation made by Turner16 in that:

'...if we are looking back upon a decision which has been taken, as most decisions, in the absence of complete information, it is important that we should not assess the actions of decision-makers too harshly in the light of the knowledge which hindsight gives us.' (p. 162)

If learning from failure is to become a significant and sustainable goal within health care, then the manner in which we apportion blame after an adverse event needs to be reassessed to take account of the role of wider organizational factors, including the manner in which management has contributed to the process of error generation.

Acknowledgements

The authors would like to thank two anonymous referees for their comments on an earlier version of this paper. However, all errors of omission and commission remain those of the authors.

References

1 Smith D. Management and medicine: strange bedfellows or partners in crime? Clin Manage 2002;11: 159-162

2 Smith D. Management and medicine - issues in quality, risk and culture. Clin Manage 2002;11:1-6

3 Barach P, Small SD. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting systems. Br Med J 2000;320:759-63

4 Kohn LT, Corrigan JM, Donaldson MS, eds. To Err is Human. Building a Safer Health System. Washington, DC: National Academy Press, 2000

5 Hawkes N. Hospital error figures held back in doubts over accuracy. The Times 2002; Section 6

6 DoH. An Organization With a Memory: Report of an Expert Group on Learning from Adverse Events in the NHS Chaired by the Chief Medical Officer. London: The Stationary Office, 2000

7 Smith D. Crisis as a catalyst for change: issues in the management of uncertainty and organizational vulnerability. In: BBA, ed. E-risk: Business as Usual. London: British Bankers Association/ Deloitte and Touch, 2001:81-8

8 Smith D. The dark side of excellence: managing strategic failures. In: Thompson J, ed. Handbook of Strategic Management. London: Butterworth-Heinemann, 1995:161-91

9 Smith D. On a wing and a prayer? Exploring the human components of technological failure. Syst Res Behav Sci 2000;17:543-59

10 Weick KE. Enacted sensemaking in crisis situations. J Manage Stud 1988;25:305-17

11 Weick KE. The collapse of sensemaking in organizations: the Mann Gulch disaster. Admin Sci Q 1993; 38:628-52

12 Weick KE. Sensemaking in Organizations. Thousand Oaks: Sage Publications, 1995

13 Ackroyd S, Thompson P. Organizational M\isbehaviour. London: Sage Publications, 1999

14 Smith D. Not by error, but by design -Harold Shipman and the regulatory crisis for health care. Public Policy Admin 2002;17:55- 74

15 Turner BA. The organizational and interorganizational development of disasters. Admin Sci Q 1976;21:378-97

16 Turner BA. Man-Made Disasters. London: Wykeham, 1978

17 Turner BA. The causes of disaster: sloppy management. Br J Manage 1994;5:215-19

18 Toft B, Reynolds S. Learning from Disasters. London: Butterworth, 1994

19 Hrebiniak LG. Complex Organizations. St Paul: West Publishing, 1978

20 Smith D. For whom the bell tolls: imagining accidents and the development of crisis simulation in organizations. Simulation and Gaming 2004

21 Etzioni A. A Comparative Analysis of Complex Organizations. New York: The Free Press, 1961

22 Elliott K, Lawrence P, eds. Introducing Management. Harmondsworth: Penguin, 1985

23 Dansereau F, Yammarino FJ. Overview: the many faces of multi- level issues. In: Yammarino FJ, Dansereau F, eds. The Many Faces of Multi-Level Issues. Oxford: JAI (Elsevier Science Ltd), 2002:xiii- xx

24 Kast FE, Rosenzweig JE. Organization and Management: A Systems and Contingency Analysis. 4th edn. New York: McGraw-Hill, 1985

25 Lilienthal D. Management: A Humanist Art. New York: Columbia University Press, 1967

26 Massie JL. Essentials of Management. Englewood Cliffs, NJ: Prentice-Hall, 1979

27 Smith D. Crisis management teams: issues in the management of operational crises. Risk Manage Int J 2000;2:61-78

28 Fauchant TC, Mitroff II. Transforming the Crisis-Prone Organization. Preventing Individual Organizational and Environmental Tragedies. San Fransisco: Jossey-Bass Publishers, 1992

29 Mitroff II, Fauchant TC, Finney M, Pearson C. Do (some) organizations cause their own crises? Culture profiles of crisis prone versus crisis prepared organizations. Indust Crisis Q 1989;3:269-83

30 Smith D. Corporate power and the politics of uncertainty: risk management at the Canvey Island complex. Indust Crisis Q 1990;4:1- 26

31 Mintzberg H. Structure in Fives: Designing Effective Organizations. London: Prentice-Hall International, 1983

32 Grint K. Fuzzy Management. Contemporary Ideas and Practices at Work. Oxford: Oxford University Press, 1997

33 Fortune J, Peters G. Learning from Failure - The Systems Approach. Chichester: John Wiley and Sons, 1995

34 Beck U. Risk Society. Towards a New Modernity. London: Sage Publications, 1992

35 Giddens A. The Consequences of Modernity. Cambridge: Polity Press, 1990

36 Erikson K. A New Species of Trouble. Explorations in Disaster, Trauma, and Community. New York: W.W. Norton and Company, 1994

37 Giddens A. The Third Way. Cambridge: Polity Press, 1998

38 Giddens A. Runaway World. How Globalization is Reshaping our Lives. London: Profile Books, 1999

39 Leape LL. Error in medicine. In: Rosenthal MM, Lloyd-Bostock S, eds. Medical Mishaps. Pieces of the Puzzle. Buckingham: Open University Press, 1994:20-39

40 Caiman K, Smith D. Works in theory but not in practice? Some notes on the precautionary principle. Public Admin 2001;79:185-204

41 Reason JT. An interactionist's view of system pathology. In: Wize JA, Debons A, eds. Information Systems: Failure Analysis. Berlin: Springer-Verlag, 1987:211-20

42 Reason JT. Human Error. Oxford: Oxford University Press, 1990

43 Reason JT. The contribution of latent human failures to the breakdown of complex systems. Philos Trans R Soc London 1990;B37:475- 84

44 Reason JT. Understanding adverse events: human factors. Q Health Care 1995;4:80-9

45 Reason JT. Managing the Risks of Organizational Accidents. Aldershot: Ashgate, 1997

46 Reason JT. Human error: models and management. Br Med J 2000;320:768-70

47 Reason JT. Understanding adverse events: the human factor. In: Vincent C, ed. Clinical Risk Management. Enhancing Patient Safety. 2nd edn. London: BMJ Books, 2001:9-30

48 Smith D. Mind the Gap! Exploring Emergence and Escalation in the Creation of Vulnerability in Organizations. Liverpool: Centre for Risk and Crisis Management, University of Liverpool, 2004

49 Smith D. Business (not) as usual - crisis management, service interruption and the vulnerability of organizations. Journal of Services Marketing 2005; forthcoming

50 Vincent C, Reason JT. Human factors approaches in medicine. In: Rosenthal MM, Lloyd-Bostock S, eds. Medical Mishaps. Pieces of the Puzzle. Buckingham: Open University Press, 1999:39-57

51 Donaldson L. Medical mishaps: a managerial perspective. In: Rosenthal MM, Mulcahy L, Lloyd-Bostock S, eds. Medical Mishaps. Pieces of the Puzzle. Buckingham: Open University Press, 1999:210- 20

52 Rosenthal MM. Dealing with Medical Malpractice: The British and Swedish Experience. London: Tavistock, 1987

53 Smith D. Beyond contingency planning - towards a model of crisis management. Indust Crisis Q 1990; 4:263-75

54 Weick KE. Foresights of failure: an appreciation of Barry Turner. J Contingencies Crisis Manage 1998;6:72-5

55 Perrow C. Normal Accidents. New York: Basic Books, 1984

56 Hodgkinson GP, Johnson G. Exploring the mental models of competitive strategists: the case for a processual approach. J Manage Stud 1994;31:525-51

57 Hodgkinson GP. Cognitive inertia in a turbulent market: the case of UK residential estate agents. J Manage Stud 1997;34:921-15

58 Hodgkinson GP. The cognitive analysis of competitive structures: a review and critique. Hum Rel 1997;50:625-54

59 Jackson M. Maps of Meaning. London: Routledge, 1989

60 Lynch K. The Image of the City. Cambridge, MA: MIT Press, 1960

61 Weick KE. The vulnerable system: an analysis of the Tenerife air disaster. J Manage 1990;16:571-93

62 Weick KE, Sutcliffe KM. Managing the Unexpected. Assuring High Performance in an Age of Complexity. San Francisco, CA: Jossey- Bass, 2001

63 Weick KE. Making Sense of the Organization. Oxford: Blackwell, 2001

64 Spr

Source: Health Services Management Research

More News in this Category