Institutional versus Market Forces as the Prevailing Theoretical Paradigm

One major theme emerging from the 2003 edition of this book was the conflict between institutional and market forces, questioning which are the dominant drivers of health care organizations and which would ultimately prevail to set the agenda in health services research. Central to the debate was the concern that if market forces became the dominant paradigm, the relevance of institutional forces would be undermined, thereby calling into question their usefulness and legitimacy in the study of health care organizations.

For example, Scott (2003) argued that the legitimacy of institutional forces for the analysis of health care organizations was increasingly challenged as market forces represented by such factors as competition, revenue maximization, and efficiency have come to dominate the institutional forces of professional autonomy and control. Elements representing an institutional presence like physicians and other providers had gradually been replaced by a “managerial logic.” A number of scholars (Scott, 2004; Perrow, 2000; Hinings and Greenwood, 2002) attributed such displacement to the shift from academic sociology departments to professional schools (particularly business schools) as the crucible for organization theory formation. As Scott (2004) pointed out, that more organizational sociologists attend the annual meeting of the Academy of Management than attend the American Sociological Association annual meeting exemplifies a departure from broader theory-driven themes of equity and power to more applied, problem-focused lines of inquiry. Perrow (2000) had been among the most outspoken regarding the “rationalist business school drift” in organization theory development. He attributes the dominance of business schools in setting a theoretical and empirical agenda that has focused away from social impact issues to economic efficiency and strategic issues. His argument is echoed by proponents of critical management studies (CMS), a field that encompasses a wide range of perspectives that spurn traditional theories of management generated by business schools (Alvesson and Willmott, 1992). CMS acknowledges the migration of academics trained in sociology, history, philosophy, psychology and other social sciences to business schools in part as a response to changes in funding priorities in the 1980s. But CMS attempts to articulate its perspectives by providing ways of thinking beyond the current dominant theories and practices of management. However, as we note later in this chapter, not all view the emergence of problem-oriented health services research as a setback, given its compatibility with the evidence-based management movement that promotes the dissemination of best practices in health care management (Rousseau, 2006).

Central to this debate is whether these alternative ways of viewing health care organizations can coexist and even inform each other, or whether they are inherently irreconcilable. If the common ground between the two camps lies in the potential for institutional and market forces to complement each other, how is this interaction manifested, and how can it be used to form a better understanding of health care organizations?

Efforts to bring about this reconciliation in order to maximize the contributions of both perspectives also speak to the utility of adopting a problem-based approach (Wholey and Burns, 2003). For example, the complex relationship between institutional environments and economic markets was addressed by Rundall, Shortell, and Alexander (2004) in their formulation of a theory of physician-hospital integration. This study built on three streams of existing scholarship: “new” institutionalism, “old” institutionalism, and the theory of economic markets. They argue that while both institutional and market forces influence integration, the interplay between them will vary depending on characteristics of the local communities in which integration takes place. Thus, communities provide the context for the dynamic interaction between market forces and institutional forces, which in turn determines whether physician-hospital integration occurs and the form such integration takes.

Revisiting the debate over what is or should be the dominant paradigm, Scott (2004) described how organizations (including health care organizations) have experienced sequential changes cascading over time and the implications of these changes for the evolution of organization theory. First, in response to market forces, formerly fixed organizational boundaries became more permeable. In reaction to increased boundary permeability, strategies of internalization, such as horizontal and vertical integration, were replaced by strategies promoting externalization such as outsourcing, resulting in the subsequent downsizing of organizations. With employee job security diminished by downsizing, worker identification, trust, and loyalty to the organization also declined. Externalization also favors a more collaborative command-and-control system over vertical channels of hierarchical control, another factor diminishing paternalistic relationships in organizations. One implication of these changes for organization theory is the need to refocus on process (what organizations do and how they do it) as opposed to structure (roles and reporting relationships) in order to better capture the emerging interactive and interdependent nature of modern organizations. From a methodological perspective, Haveman (2000) prescribed the use of qualitative case studies that explicitly considered the transactional context in which organizational process occurs.

The need for more qualitative studies of organizational process also emerged from a review of the value of market-based models for understanding health care organizations (Bazzoli et al., 2004a). Based on findings from the literature, this review concluded that efficiency gains, the relevant performance metric motivating vertical and horizontal hospital integration, are relatively small and temporary. Any savings that materialize appear to be a one-time occurrence as opposed to a bend in the cost curve (a continuous reduction in the rate of cost growth). They conclude that the reason for the failure of integration strategies to produce lasting positive efficiencies is related to fundamental gaps in understanding how these strategies are implemented. Methodological approaches with the potential to close this gap include observational case studies based on ethnographic and anthropological techniques in order to determine the important internal and external factors in successful integration and determine what commonalities exist across health care organizations. Observational case studies have been used to study hospital mergers (Barro and Cutler, 1997; Eberhart, 2001), integrated delivery networks (Coddington, Ackerman, and Fischer, 2000; Coddington, Chapman, and Pokoski, 1996; Coddington, Fischer, and Moore, 1994), and physician-hospital organizations (Gorey and Bannon, 1998; Greenberg, 1998). Since the full effects of integration could take years to materialize, the authors also recommend better temporal alignment between the qualitative case studies documenting existing process and the quantitative financial outcomes that may or may not manifest at a later date. Finally, since diverse stakeholder groups may differ in their assessment of potential benefit and harm from market-driven strategies like integration, the selective accrual of benefits should be included as weights in evaluating organization performance.

In summary, there is a growing consensus among organization theorists regarding the importance of the interplay of markets and institutions in understanding organizations. Subsequent developments in the institutional versus market forces debate acknowledge that economic activities are embedded in an institutional or social framework. Thus, when market outcomes are less than optimal, they could reflect the social context in which they are entrenched. For example, economic sociology views social institutions as complex entities in which informal, implicit institutional features interrelate with formal, explicit features in creating a coherent whole. This institutional complex is not a static optimal response to economic needs, but rather a reflection of a historical process in which past economic, political, social, and cultural features interrelate and have a lasting impact on the nature and economic implications of social institutions (Smelser and Swedberg, 2005). Using a game theory framework, proponents of economic sociology posit that initial social structures permit the emergence of particular self-enforcing economic and political institutions whose functioning further influences these structures (Greif, 1998). Thus, through the mechanisms of embeddedness and reciprocity, both market and institutional forces can influence organizational effectiveness. However, although there is consensus that both markets and institutions mutually affect behavior, there is less agreement on how these forces relate to each other and how their combined effect on organizations influences outcomes (Rundall et al., 2004).

Inside the Black Box: The Evidence for Evidence-Based Management in Health Care Organizations

A second major theme emerging from the 2003 compilation concerned the need to understand how organizational processes affect outcomes. Since that time, evidence-based management (EBMgt) has gained traction as a mechanism for process evaluation (Rousseau, 2006). EBMgt is defined as the systematic use of the best available evidence to improve management practice by deriving principles from research evidence and translating them into practices addressing organizational problems, thereby moving decision-making from intuition to scientific evidence (Briner, Denyer, and Rousseau, 2009). It entails making decisions through the conscientious, explicit, and judicious use of four sources of information: practitioner expertise and judgment, local context, a critical evaluation of the best available research evidence, and the perspectives of people who may be affected by the decision (Rousseau, 2007).

EBMgt is not new to organizational practice. The concept of applied research in management practice dates at least as early as the Hawthorne studies in the 1920s, and Chester Barnard (1938) promoted a version of it in his seminal writings. EBMgt gained traction in the 1990s in reaction to the success of evidence-based practice in medicine. Although examples of EBMgt practices were found in education and criminal enforcement practice, medicine was the first domain to institutionalize it successfully. Before the EBMgt movement, medical practice was as subject to variance as management practice was (Timmermans and Kolker, 2004). Knowledge translation was the vehicle used in medicine to overcome variations in clinical practice by emphasizing the movement of provider behavior toward acceptance of evidence-based practice.

Since evidence-based medicine (EBM) gained acceptance on the clinical side of hospitals and other health care organizations, health care management would seem to be a uniquely appropriate place to implement its management counterpart, EBMgt. However, EBMgt in organizations in general, including health care organizations, is underdeveloped, misunderstood, misapplied, and inconsistently implemented (Briner et al., 2009). As a result, the implementation of effective management practices in health care organizations lags far behind clinical applications, as evidenced by the wide variation in managerial practice patterns and persistent use of practices known to be ineffective. Although published articles on EBMgt have been increasing over time, scientific rigor has not improved, as most are based on opinion or anecdote. A systematic review of the literature on EBMgt found that none of the articles published directly addressed whether there is evidence that employing EBMgt improves organizational performance (Reay, Berta, and Kohn, 2009).

Since research demonstrating its impact on performance is virtually nonexistent, some question the merits of advocating for EBMgt in health care management (Arndt and Bigelow, 2007). However, the lack of evidence for EBMgt may be due to a divide between the academic and practitioner realm, exacerbated by the lack of consensus about what it is and who is doing it. For example, the term EBMgt is relatively new, so practitioners may not identify or recognize their engagement in it. Access to companies adopting change programs is often restricted, making it exceedingly difficult for outside social scientists to obtain a satisfactory sample (Barnett and Carroll, 1995). As a result, there are only modest advances in research to substantiate the use of EBM.

A number of obstacles to EBMgt implementation have been identified to explain why the gap between research and practice is so large. First, unlike medicine, management is not recognized as a profession with a common referent body of knowledge. In the United Kingdom, where articles on EBMgt have enjoyed wider circulation, there is still no consensus on proven best practices in management evidenced by the most fundamentally consistent findings from research. For example, although 46 percent of UK organizations surveyed in 2004 reported using psychometric testing routinely in hiring and recruitment, there is thus far no definitive evidence that the practice has resulted in improved recruitment and retention (Guest, 2007).

Second, managers are unaware of where to find support for EBMgt. Again, the experience in the United Kingdom, where there has been more publication of EBMgt-related research than in the United States, illustrates this problem. Less than 1 percent of human resource managers read the academic literature, in part because the timeliness of information published in journals is not conducive to the immediacy of most organizational problems (Guest, 2007). In addition, academic venues are not particularly accessible to practitioners. However, even when awareness of EBMgt exists, there is skepticism on the part of managers as to whether it represents more than the latest fad promoted by those with a vested interest in its dissemination. As Barnett and Carroll (1995) wrote:

Managers understand that the claims of consultants behind these programs (TQM, lean management, etc.) cannot be taken at face value. But, unfortunately, managers rarely put academics in position to pass judgment on program effectiveness using research designs and models… Generally speaking, consultants do not want objective external assessments of their impact, and they maneuver to avoid it

Again, the quality of research also needs to be more convincing. Managers need a stronger reason for implementation than expert opinion. Until that occurs, the most reliable source of information on best practices in management remains other managers facing similar issues in their own industries. (p. 231)

A third obstacle to EBMgt implementation relates to the time lag between the making of a decision based on EBM and evidence of successful outcome. This time lag blurs the linkage between cause and effect, allowing other interventions to influence outcomes.

A fourth obstacle to implementation echoes the resistance to EBM on the part of some physicians, who view the manifestation of EBM in practice guidelines as a threat to their professional autonomy. Similarly, managers may perceive that EBMgt imposes limitations on their prerogatives, marginalizing managerial discretion. The perception of personal adverse consequences stemming from management's attempts at innovation has roots in the past, when a perceived threat to managerial autonomy created resistance to the principles of Taylorism at the turn of the last century. Concerns regarding compromised managerial autonomy may create similar suspicions about the motivation behind EBMgt as to who in the organizational hierarchy stands to benefit.

Finally, managers are not formally trained in the fundamental principles of scientific evidence (Charlier, Brown, and Rynes, 2011). Lacking the training to do so, managers generally do not use their own organizations as laboratories for research by engaging in the equivalent of managerial clinical trials.

The case study method used in business schools does not incorporate EBMgt in that case studies emphasize the uniqueness of organizational context as opposed to the generality of principles. As Freek Vermeulen, associate professor of strategy and entrepreneurship at the London Business School, commented in the Financial Times:

There is a great divide in business schools, one that few outsiders are aware of. It is the divide between research and teaching. There is little relationship between them. What is being taught in management books and classrooms is usually not based on rigorous research and vice versa; the research published in prestigious academic journals seldom finds its way into the MBA classroom… Business schools largely fail in providing rigorous, evidence-based teaching. Instead, the separation between research and teaching causes their courses to rely largely on dangerously simplified generalizations at a time when corporate pitfalls—that recently laid our economy low—epitomise a need for more sound management in favour of popular fads. (Briner, 2011)

Overcoming these obstacles will require achieving a balance between teaching principles (cause-and-effect knowledge) and practices (solutions to organizational problems). For example, practitioner-research collaborations promoting knowledge transfer, such as the Cochrane Collaborative, established in 1993 to organize medical research information in a systematic way to inform EBM, enable innovations to disseminate and adhere (Rousseau, 2007). The Cochrane Collaborative in health care operates in over one hundred countries and engages more than twenty-eight thousand volunteer researchers and practitioners (Allen and Richmond, 2011). The systematic reviews it employs have been successful in promoting evidence-based medicine, but these methods need not be limited to this application. Rigorous, systematic evaluation of evidence through appropriate system reviews can also guide EBMgt in health care organizations.

Training in medical schools lends itself to evidence-based practice. One solution to advancing knowledge transfer in management would be to teach EBMgt in health care management education, as it is in medical training. Without training, managers are unlikely to practice EBMgt since they lack a foundation from which to do so.

There is also a need for more user-accessible venues for EBMgt dissemination. Web-based sites are often the first point of information access for managers, arguably making the Internet one of the more effective means of conveying EBMgt knowledge. There has been definite movement in this direction, furthering EBMgt implementation. In the United States, the Evidence-Based Management Collaborative, pioneered by Denise Rousseau at Carnegie Mellon University, among others, has made strides in addressing implementation concerns. Defining on its website (www.evidence-basedmanagement.com) its primary task as “designing the architecture and support practices for on-line access to best evidence summarized in ways practitioners and educators can readily use,” the collaborative posts EBMgt-related articles and other educational materials.

Finally, as found in the model for EBM in medical care, infrastructure supports are needed to allow for effective dissemination of EBMgt. In the United Kingdom, government advocacy of evidence-based policy and practice has encouraged dissemination in medical practice. The government has endorsed evidence-based approaches to setting government policy through the Centre of Evidence-Based Policy and Practices that supports diffusion of knowledge. Within health care provider organizations, Britain's national health system promotes evidence-based practice using the Cochrane Collaboration's recommendations as a standard for practice (Guest, 2007). Again, quoting Barnett and Carroll (1995):

Although the idea sounds ludicrous at first, one suggestion would be for the government or some other authoritative body to regulate the sale and promotion of organizational regimes much like the Food and Drug Administration does pharmaceuticals… The justification for such a drastic step could come not just from the revenues misspent on ineffective programs but from the potentially deleterious impact ill-designed programs could have on companies and individuals. (p. 232)

In summary, EBMgt shows promise to further our knowledge of how processes optimally link to outcomes and, in doing so, takes the lid off the black box of organizational management as called for in the 2003 Mick and Wyttenbach compilation. However, the following processes are needed to bring that promise to fruition:

• Establish the body of evidence for EBMgt by developing consensus on which EBM practices have clear value.
• Consider the use of more user-friendly venues, including electronic communication channels taking the perspective of the practitioner, in order to make information more accessible and meaningful. (See the website of the Institute for Healthcare Improvement for an example: http://www.ihi.org.)
• Establish collaborative relationships among researchers, practitioners, and educators.
• Introduce EBMgt into management curricula.
• Develop facilitating macroinfrastructure such as exists in the United Kingdom.

Paradigms Avoiding Oversimplification: The Promise of Complexity and Chaos Theory in Advancing Problem-Driven Research

The complexity of organizational interrelationships, a third major theme from the 2003 edition of this book, has again led to calls for the use of complementary or integrative theories that avoid oversimplification and challenge received knowledge by exploring new ways of envisioning health care organizations (Qiu, Donaldson, and Luo, 2012). Complementarities in organizational analysis across established perspectives such as institutional and exchange theories have already been noted. Newer perspectives, such as complexity theory, and a variant, chaos theory, were singled out as potentially conducive to a problem-based approach to health services research.

In the 2003 compilation, the case made for the use of complexity theory in the study of health care organizations was based on the argument that no single conventional theory is capable of capturing the unpredictable richness of organizational change and strategy (Begun, Zimmerman, and Dooley, 2003). Characterizing complex phenomena with a reduced and simplified set of explanatory variables does not do justice to the reality of organizational life or its interdependence with its environment. Rather, the density, complexity, and inherent dynamism of health care organizational life are better understood by frameworks that take these characteristics as given, such as complexity science, than through conventional reductionist theories.

This proposition reverberates in more recent work noting the striking lack of scientific evidence underpinning even some of the most popular models of change in health care organizations (Grol et al., 2007). Given the limited empirical evidence of the effectiveness and feasibility of individual theoretical approaches in predicting the antecedents and consequences of intended change, it may be more constructive to consider multiple theoretical perspectives simultaneously. Furthermore, since a variety of factors operating in different contexts (such as professional, social, organizational, or economic) contribute to organizational outcomes, hypotheses regarding effective change derived from multiple perspectives acknowledging contextual diversity are justified. Thus, basing research investigations on different theoretical approaches, including complexity theory, may prevent important factors from being overlooked (Grol et al., 2007).

However, while making the conceptual argument for incorporating complexity theory into health services research, the absence of methodological approaches for testing its premises was acknowledged then as it is now. Methodology considerations related to the use of complexity theory in health services research in the 2003 compilation was limited to the use of computer simulations, emphasizing the assumptions behind simulation essential to establish validity. Then as now, the feasibility of the use of multiple theoretical perspectives, particularly those incorporating complexity frameworks, in health services research still largely depends on overcoming methodological limitations.

On that front, there has been relatively little progress to date, although chapter 11 in this book is an important starting point directly addressing progress and potential in testing CAS perspectives. Although complex adaptive systems have been the focus of study across a variety of scientific fields over the past forty years, with few exceptions, there has been little systematic observational or experimental research in the context of health care. Complexity science has had its own journal, Emergence, since 1999, yet only two articles and one book review relevant to health care services have been published. Furthermore, neither article had an empirical component. One article adopts a “complexity perspective” to consider ethics in health care, noting that the 2001 Institute of Medicine report addressing medical errors, To Err Is Human, recommended that health care organizations be regarded as complex adaptive systems focusing on process as opposed to mechanistic structures (Mills, Rorty, and Werhane, 2003).

The second article promotes complexity theory as a different way of viewing mergers in the hospital industry (Zimmerman and Dooley, 2001), contrasting the limited insights gained from a mechanistic view of mergers with a more organic, loosely coupled view of why mergers succeed or fail. Complexity theory is proposed as an alternative to the deterministic Newtonian view of organizations as a machine, substituting the metaphor of the organization as a living biological organism. Thus, while a deterministic view would define mergers mainly in terms of their economic benefits, such as achieving economies of scale, complexity theory views mergers in terms of their potential for innovation, adaptation, and emergent synergies. The 2011 interview comments of David Ferrucci, the IBM scientist who designed the computer (Watson) that soundly defeated two expert human Jeopardy! players, illustrate these alternative organizational metaphors. When asked why Watson was not present at the interview to “speak for himself,” he noted Watson's limitations: “It's not an interactive dialogue system, so it can't conduct its own interviews. It would still be responding only from content it's been given and analyzed. [Watson] … could [not] go off and even approach a child's ability to do language, to move, to think, to interact” (Ferrucci, 2011, p. 104) Organizations, like children, are complex adaptive entities, and complexity theorists would argue that comparison to machines provides limited insight at best.

In Reengineering Health Care: The Complexity of Organizational Transformation (2002), Terry McNulty and Ewan Ferlie described the effort to bring about organizational transformation in a large British National Health Service teaching hospital, the Leicester Royal Infirmary, between 1992 and 1998. Their text used complexity theory to explain the unintended consequences accompanying the introduction of business process engineering. This case study documents the negative side of change, particularly the difficulty in overcoming clinician resistance to change, from a complexity perspective.

Although the use of complexity theory in health services research contexts has been sparse at best, one variation of complexity theory, chaos theory, has had a variety of applications in the field of nursing, including its use as a research tool in the development of managerial guidelines and for statistical modeling in the planning and delivery of nursing services (Haigh, 2008). For example, Anderson, Corazzini, and McDaniel (2004) used chaos theory to investigate the determinants of nursing home staff turnover and resident outcomes. In this study, nursing homes are conceptualized as complex adaptive systems engaged in nonlinear interactions. Organization effectiveness is modeled as a function of patterns of relationships that were tested empirically through the use of interaction terms. Chaos theory was also used to frame hypotheses in another study about how different management practices support favorable resident outcomes (Anderson, Issel, and McDaniel, 2003), and the study found that management practices supporting constructive self-organization (as stressed by complex adaptive behavior) were associated with better outcomes.

However, both these studies employed cross-sectional linear regression methods ill suited to capture adaptive behavior over time, which is fundamental to predictions derived from a complexity perspective. Thus, whether progress has been made in resolving the methodological issues surrounding the use of complexity theory in health care contexts is largely unaddressed by these studies. Indeed, whether complexity should be regarded as an organizing framework, a set of methods, or both has not been resolved. For example, Mills et al. (2003) juxtaposed complexity as an analytical approach with complexity as a perspective, coming down on the side of complexity as a way of framing ethics in health care organizations as opposed to a methodology for analysis.

While some conclude that a significant proportion of organizational behavior operates in complex or chaotic terms not subject to empirical verification, others continue to advance methods to test hypotheses drawn from complexity theory empirically. For example, Haigh (2008) applied a variation of a Malthusian model of unrestricted growth borrowed from population biology to predict outcomes in a hospital acute pain service based on chaos and complexity theory principles. Resnicow and Page (2008) also addressed empirical verification of complexity-derived propositions, noting the methodological issues that arise from using linear regression models to test predictions derived from complexity-based perspectives. For example, while the error term in a linear regression model may be interpreted as capturing complexity (unaccounted variance attributable to a chaotic component), it does not provide insight into the nature of complexity. In addition, linear modeling of complexity phenomena may entail use of higher-order interactions that are difficult to model, may be underpowered, and therefore are difficult to detect and hard to interpret (Anderson et al., 2004).

To overcome some of the limitations of the use of linear statistical modeling of complex or chaotic phenomena applications, Resnicow and Page (2008) suggest the inclusion of complementary approaches incorporating nonlinearity into the design and analysis of research studies. Nonlinear models could be used to explore alternative mathematical relationships that account for variance attributable to complexity. For example, agent-based modeling has been used to predict the course of health care epidemics and climate change. This computer simulation technique uses “agents” (individuals following rules) to incorporate interindividual variability in change pathways, allowing for testable hypotheses. However, while these techniques account for complexity under prescribed circumstances, they may substantially limit the ability to develop generalizable statistical models of change. In addition to simulation, qualitative methods are suggested as a complementary approach for complexity-based research. For example, structured interviews could be employed to develop testable hypotheses exploring how and why organizational change occurs.

In summary, complexity theory and its variants still have appeal as a way of conceptualizing health care organizations. In particular, its emphasis on dynamic processes lends itself to EBMgt and other problem-based approaches that we advocate. However, methodological limitations still frustrate its application to testable hypotheses, a concern addressed in chapter 11.

Diversity and the Principle of Requisite Variety

In 2011, for the first time, Caucasian newborns represented the minority of births in the United States (Tavernise, 2012). For major parts of this country, diversity is the “new normal.” How is this affecting health care management and research on the effectiveness of provider organizations? Building on work sponsored by the Department of Health and Human Service's Office of Minority Health, the national standards for Culturally and Linguistically Accessible Services (CLAS) for health care providers were issued in 2001. The standards address several key domains of cultural competence:

• Diversity of staff and providers that reflects the relevant patient population
• System capabilities (data to track outcomes of subgroups) and care management that is culturally sensitive
• Effective interpreter services and cultural competency training for all management, staff, and providers

In recognition of the inherent threats to patient access, safety, and quality of care, the Centers for Medicare and Medicaid Services mandates four of the fourteen standards for participation in Medicare and Medicaid. In addition, accrediting bodies such as the American Association of Medical Colleges, the Joint Commission, and the National Committee for Quality Assurance have adopted some of the standards. Most widespread is the requirement that communications between providers and patients are accessible to people with limited English literacy. In a recent study of CLAS adherence in California hospitals, Weech-Maldonado and colleagues (2012) found that mean adherence to the full list of standards was about 68 percent, with substantial variation attributed to market and institutional factors. Their approach, combining predictions from resource dependence and institutional theories to account for variations in compliance, provides a rare contribution of organization theory to understanding the role of diversity in health care management. They recommend further research to examine adherence to the CLAS standards and patient experiences with care. Such work might well begin the process of understanding how cultural diversity works as a context for interpersonal as well as interorganizational behavior in health care.

The impact of diversity is a function not only of the patient population that a health care organization serves, but also of the caregivers who are increasingly themselves diverse in terms of culture and language. The National Council of State Boards of Nursing (2006) reported that in 2005, the United States had surpassed the United Kingdom in the number of foreign-born nurses entering its workforce, with nearly fifteen thousand passing the RN licensing exam. Aiken (2007) estimated from 2000 Census data that 8 percent of practicing RNs in the United States received their training in another country before migrating to the United States. The physician workforce in the United States since the late 1940s has tapped into the global market (Norcini et al., 2010). No longer a new phenomenon to most US communities, foreign medical graduates fill a disproportionate number of positions in primary care and internal medicine and serve in underserved areas. The diversification of language and culture, as well as training approaches across national and cultural boundaries, may pose significant challenges to providing safe and effective care within a provider organization. These challenges are likely to intensify exponentially as responsibility for patient outcomes extends beyond the provider organization's boundaries into diverse communities.

At a microlevel, to understand how diversity among patients and providers affects transitions in care, coordination across boundaries, and related contemporary challenges, researchers might examine how the principle of requisite variety operates in this context. Originating in early systems theory (Ashby, 1956), Morgan (2006) describes the principle of requisite variety as central to creating an organization capable of brainlike learning, change, and regeneration:

The principle of requisite variety … suggests that the internal diversity of any self-regulating system must match the variety and complexity of its environment if it is to deal with the challenges posed by that environment The principle of requisite variety is not just an abstract concept. It is vital management principle. If a team or unit is unable to recognize, absorb, and deal with the variations in its environment, it is unlikely to evolve and survive. The principle suggests that when variety and redundancy are built at a local level—at the point of interaction with the environment rather than at several stages removed, as happens under hierarchical design—evolutionary capacities are enhanced. Individuals, teams, and other units are empowered to find innovations around local issues and problems that resonate with their needs. This also provides a resource for innovation within the broader organization, as the variety and innovation thus experienced is shared and used as a resource for further learning. (pp. 109–110)

Requisite variety is unlikely to be achieved by a centralized environmental scanning or business development unit. Rather, it requires direct engagement between each unit and the key stakeholders in its environment through recruitment and hiring practices or outreach efforts. In the case of care coordination, the transaction among parties can be standardized only to a very limited extent given the clinical and contextual details that must be explicitly communicated. In parts of the country where diversity is most concentrated, as reflected in Kaiser Permanente's cultural competency investment in California, these efforts are recognized as strategically important. In communities where the impact of recent immigration is only beginning to be recognized, health care managers and providers may be inclined to view CLAS as just another part of the regulatory framework and seek ways to comply without investing heavily. In those settings, risk of harm to patients is perhaps greatest given inadequate interpretation systems and lack of awareness of risks posed by unarticulated cultural differences. While knowledge of the extent to which hospitals are meeting or exceeding CLAS standards for both institutional or resource dependence–driven reasons (Weech-Maldonado et al., 2012) is a valuable starting point in understanding how diversity is influencing health care, the operational mechanisms of managing diversity need to be considered at the team and individual levels as well. This suggests that macrolevels thinkers collaborate with microsystems thinkers for research models that explore the actual implementation of diversity management in context.

Interdependence in Care Transitions and New Perspectives on Work Design

Perhaps nowhere else is the field of organization theory addressing issues more critical to health services management than in the renewed (and quite new) thinking about the design of work. New perspectives go beyond task characteristics to address major shifts in the economy from a machine-driven workplace to one that is driven by service encounters. In health care, for example, the social aspects of work, including the interpersonal interdependencies among those providing the service and those receiving it, are central to care outcomes. How these interdependencies are best managed in particular contexts is at the forefront of new perspectives on work design theory (Grant and Parker, 2009). While economic theory helped us understand the role of information asymmetry in explaining the authority that the medical profession holds, a number of factors have complicated that observation. Mutual information asymmetry in the physician-patient relationship is now well established. It is increasingly recognized that advanced interpersonal communication and relationship-building skills are not a distraction from the work of providing care but rather a central part of the work. For example, portions of the MCAT examination required for medical school entrance now evaluate student communication and cultural competence skills (“The New MCAT,” 2012).

At the same time, patients face intense uncertainties and complex decisions when they access health care. Shifting policy incentives and consumer demand for help in navigating the complex maze of social systems that comprise health systems broaden the scope of what it means to provide care. In the context of all these uncertainties, health care professionals are now asked not only to be technically proficient but also to engage in systems thinking in terms of vigilance and proactive problem solving and to build relationships that will help those systems improve care. Thus, the “redesign of work design theory” (Grant and Parker, 2009) with a new focus on the social or relational aspects of work also makes explicit the ways in which workers enact their own jobs in uncertain environments. This should provide fertile ground for qualitative and longitudinal studies of health professional work. One way that nurses enact their work is through work-arounds that they create to compensate for system dysfunctions (Halbesleben, Wakefield, and Wakefield, 2008). Care navigators for cancer patients piece together care options and support systems, thereby enacting a care team for each client.

In hospitals, work design changes have already begun to address the complexity of information management between providers and across systems. One common approach to improving transitions in care is standardization in the form of checklists or protocols based on evidenced-based care. In its 2006 Patient Safety Goals, the Joint Commission on the Accreditation of Health Care Organizations required that all hospitals adopt a standard communication format. One approach is the SBAR communication format for transmitting information clearly across providers, which has been adapted for use in various clinical contexts (Haig, Sutton, and Whittington, 2006):

1. Situation: description of current issue
2. Background: brief description of preceding events
3. Assessment: key features of current state
4. Recommendation: suggested actions to take next

Relatively simple design features within electronic medical records (EMRs) have been shown to improve the likelihood that important information will flow with the patient. One approach reported the creation of two communication tabs in the EMR: an RN-to-MD tab and an MD-to-MD tab. Care guidelines and standardized communication processes are being developed to guide information flow among and between patients, nurses, and physicians (Knych, 2011).

While evidence-based practice is surely moving forward, unresolved questions remain regarding the changing nature of health care delivery as work. How and under what circumstances do health care providers move between adhering to best practice protocols and enacting their own work? Similarly, from the organization's viewpoint, what is the optimal balance between loose and tight coupling of roles and behaviors in alternative scenarios?

Coordination Structures: What Lies Beneath?

If it indeed takes a village to raise a child, it takes more than one village to deliver the health care quality that we expect as consumers. Intense interdependencies across myriad social institutions are not yet fully recognized, and they are rarely well managed. For example, the need to both promote postdischarge self-care management and activate support systems as needed requires health care managers to plan and negotiate with social service managers. More important, it requires clinical staff to talk with families, patients, and their community support systems. Hierarchy is likely a detriment to this kind of boundary spanning at the patient care level, but it seems that an explosion of coordination-focused boundary spanning is being heralded. What organizational forms and processes will support and optimize the burst in boundary spanning caused by increasingly complex exchanges among organizational forms?

Currently most observers attribute the endemic dangers of transitions in care to communication breakdowns. Communication within organizational contexts, however, is generally a manifestation of the coordinating structures employed in an organization, such as roles or procedures. Early theorizing about organizational coordination was based on the assumption that work systems and roles could be specified and controlled precisely enough to ensure predictable production outcomes regardless of other contextual variables (Fayol, 1917; Taylor, 1947). According to this school of thought, coordination across individuals, in well-designed systems, would be viewed as waste. Translated to the contemporary health care delivery environment, if “discharge instructions provided” is checked off on a medical record, then it is assumed to have occurred, and notation in a medical record represents fact. Building on this view, the scientific management school sought to standardize raw materials and procedures to reduce the need for coordination. Although increased focus on human reactions to the resulting routinization posed serious questions about the singular value of this approach, it was perhaps a more pragmatic manufacturing trend soon to be followed in medicine—specialization—that prompted work design theorists led by Woodward's (1970) contingency approach to embrace coordination and mutual adjustment among individuals and units as an important design element.

Despite the influential work of Lawrence and Lorsch (1967) on the importance of providing support for specialization (or differentiation) and simultaneously ensuring requisite coordination (or integration), research on this balancing act failed to develop. This was because the nature and technology of work as the drivers of contingency theory hypotheses were replaced by the perspective that the organizational environment mattered most in finding a good structural fit. Recently there has been renewed interest in work technology and structural contingency, however, and this may well be a fruitful avenue for addressing some of the profound challenges to providing care coordination within our fragmented and complex systems of care. The complexity of care management across providers and systems often requires active coordination and mutual adjustment. This is resource intensive and inefficient, and for these reasons, discharge processes and other care transitions have become threats to patients and providers alike.

In their review article on coordination in organizations that is both broad and deep, Okhuysen and Bechky (2009) identify three integrating conditions (i.e., mechanisms) that appear to reflect “the means by which people collectively accomplish their interdependent tasks in the workplace. Each of these conditions addresses some of the demands that the integration of specialized work imposes on the individuals performing the work and resolves some of the uncertainties created by interdependence” (p. 483). They identified accountability, predictability, and common understanding as the three conditions or mechanisms through which coordination functions, and they map these across five categories of coordination structures (plans or procedures and rules, objects and representations, roles, routines, and proximity) in a matrix that seems highly applicable to parsing the ways in which patient care suffers from poor coordination. Clearly many alternative approaches are being tested in different care transition settings, perhaps supporting the authors' contention that effective integrative processes emerge from local contexts rather than from globally conceived design mechanisms. For these approaches to become more widely adopted, however, research that examines the underlying mechanisms of accountability, predictability, and common understanding is needed.

Patient-centered medical homes and accountable care organizations have many choices before them that could be informed by research that applies this framework about the integrating mechanisms required for successful use of coordination structures and their contribution to performance. A multidisciplinary care planning committee, for example, is a structure for promoting coordinated care delivery, but it is not a guarantee that such performance will be achieved. Without deliberate attention to the development of mechanisms for and the actual processes of integrative work—accountability, predictability, and common understanding—the individuals in the group may not achieve a shared mental model that improves system performance. Challenges to these three conditions are ubiquitous in health care delivery. Specialization, status differentials, and other sources of divergent thinking do make it difficult for a team of health care experts to function as an expert team. Systematic training in teamwork is showing some promise in making operating room behavior more conducive to safe care (Weaver et al., 2010). Research on the impact of these types of interventions is still new, and it is rarely grounded in a multilevel, transtheoretical conceptual framework that will add lasting value to the field of practice. In pursuing research along this line, investigators will be tempted to settle for identification and codification of coordination structures such as rules, protocols, and roles because they are more evident than are the psychosocial conditions for real integration. Yet the opportunity for significant work in this area by digging deeper may be transformational.