Life at the edge of chaos. (Health Care Applications of Complexity Science).

SINCE THE DEMISE OF THE CLINTON HEALTH PLAN, market-driven health care system evolution has been proceeding across the U.S. at a rapid pace. As the health care system undergoes unprecedented change, both clinical and institutional leaders find themselves under extraordinary pressure to create local and regional systems that offer high quality, cost-effective (i.e., high value) health care. In this context, clinicians are seeking improved ways of managing patients, and both clinical and administrative leaders of hospitals, group practices, and newly integrating systems are pursuing methods that promote high-value operational systems.

But a primarily economically-driven health care reform is generating serious questions regarding the traditional missions of both clinicians and the leaders of health care provider organizations. In the face of current pressures, can physicians genuinely pursue their desire to place patients first, economics second? Can institutional leaders follow their desire to create community-focused approaches to health improvement, and do so in an economically-viable manner?

Those who seek to answer these questions in the affirmative face sobering challenges. As they address pressures for higher quality and lower cost, where are today's clinicians and institutional leaders to look for guidance?

Recent discoveries from diverse fields, such as evolutionary biology, economics, physics, computer science, physiology, and mathematics--coupled with discussions by leading scientists across the boundaries of these fields--have much to offer clinicians and organizational leaders. Complexity science, a relatively new field of inquiry, forms the theoretical basis for this assertion. The pioneering findings now unfolding in this area have relevance at both a micro (the treatment of individual patients) and a macro (the organization and operation of health services) level.

Current literature reflects accumulating evidence that complexity theory has broad applications encompassing such seemingly diverse disciplines as clinical medicine and management science. Juxtaposing the case studies explored in this article--an individual patient and a medical staff reorganization effort--is both unique in the context of this literature, and illustrative of one of the features of complex systems: their fractal nature. Fractals, discovered by the mathematician Benoit Mandelbrot, are complex patterns that are self-similar on different scales.

Beyond such conceptual considerations, these case studies illustrate for both clinicians and leaders of health care organizations how complexity: (1) may be relevant, even helpful, as they consider difficult challenges in both patient and organizational management: and (2) might emerge as a synthesizing force as they face the extraordinarily complicated task of jointly creating integrated health care systems.

The machine metaphor

Listen to these words and phrases, and consider what images they bring to mind: "work-up." "diagnose," "fine-tune," "prescribe," "industry," "engineer," "design," "operate," "control," "check." Now consider such terms as "evolve," "adapt," "emergent," "self-organize" "diversity," "ecology." What images emerge?

Most medical practitioners and health care organizational managers likely draw upon the first set of terms as they deal with their day-to-day tasks. They derive this approach from their training, an educational and experiential process dominated by the metaphor of the machine--order, predictability, rationality, linearity. Complexity pioneers have observed the pervasive influence of the machine metaphor. Medical researcher and cardiologist, Ary L. Goldberger, MD, writing in Scientific American, notes, "The conventional wisdom in medicine holds that disease and aging arise from stress on an otherwise orderly and machine-like system--that stress decreases order by provoking erratic responses or by upsetting the body's normal periodic rhythms." (1)

Much good has come from this metaphor, which dates to Newton and the Industrial Revolution. But is this way of thinking adequate for many of the issues that clinicians and organizational leaders face today? Is this linear type of cause-and-effect approach optimal for dealing with patients suffering from complicated, interacting disease processes? Does it adequately give direction to those responsible for creating health care systems which work, and/or aid in uncovering new, cost-efficient ways to improve the health and well-being of communities?

Gareth Morgan, in his revised classic work on management. Images of Organization, suggests that the machine metaphor may not be appropriate to all settings:

"One of the most basic problems of modern management is that the mechanical way of thinking is so grained in our everyday conception of organization that it is often very difficult to organize in any other way." (2)

The concepts of complexity suggest promising new approaches, new ways of understanding a variety of situations we face every day. Its language reflects nature and life. Research has led to the discovery of common patterns of behavior in complex adaptive systems (GAS). John Holland, a founder of complexity and faculty at the University of Michigan and at the Santa Fe Institute, defines GAS as "systems composed of interacting agents described in terms of rules." He notes that these agents are "diverse in both form and capability" and that they adapt by changing their rules and, hence, behavior, as they gain experience. (3) Examples of GAS are found everywhere--ant and termite colonies, immune systems, the brain, economies, ecological systems, the Internet, as well the case studies described in this article.

Ary Goldberger suggests that nonlinear dynamics and complexity science may lead to advances in physiology and medical care. He notes the likely application of nonlinear dynamics to physiological monitoring in the near future and the discovery of new therapeutic approaches. He observes that, "Many disease states are marked by less complex dynamics than those seen under healthy conditions. The decomplexification of systems with disease seems a common feature of many diseases, as well as of aging." (4) As a result, diseases that exhibit such diminution of variability over time have been termed dynamical diseases.

Organizations, their leadership and management, have received growing attention from writers working from a complexity perspective. Ralph D. Stacey suggests that organizations should be viewed as complex adaptive systems: "Such network systems are ubiquitous in nature; not surprisingly, because we too are a part of nature, human interaction also sets up such systems. Each of us has a brain that is a complex adaptive system in which neurons are the agents. Each of us has a mind that is a GAS in which symbols and images are the agents. When we come together as a group we constitute a GAS in both a biological and a mental sense. It follows that all organizations are such systems." (5)

Complexity holds the real possibility of stimulating fresh insights and approaches to health and medical care--both its provision and its organization.

Complexity science is the confluence of several emerging fields in the sciences and mathematics: nonlinear dynamical systems theory: far-from equilibrium thermodynamics (originating in the work of the Nobel prize-winning physical chemist llya Prigogine); synergetics (from the work of the German physicist Hermann Haken and more popular in Europe and Russia): chaos theory: and research in complex adaptive systems (centered at the Santa Fe Institute). Because it is woven from strands coming out of these different theoretical heritages, the terminology and constructs can, at times, be confusing and ambiguous, a veritable "Tower of Babel." (6) Nevertheless, several common central concepts can be discerned.

Life at the edge of chaos

Perhaps the most significant tenet of complexity science is the idea of 'life at the edge of chaos." While specific facets of this concept remain under debate, complexity researchers argue that systems which are most adaptable and alive operate in a zone between order and disorder. Christopher Langton, faculty at the Santa Fe Institute, was the first to recognize that this zone was much more than a boundary, that there was another state besides order and chaos, and that this intermediate region was where a system was most alive. (7)

Stuart Kauffman, MD, also from the Santa Fe Institute, states, "To engage in the Darwinian saga, a living system must first strike an Internal compromise between malleability and stability. To survive in a variable environment, it must be stable to be sure, but not so stable that it remains forever static." (8) Ralph Stacey calls this the "zone of creativity at the edge of disintegration." He cites the similarity of this complexity concept to findings of psychodynamic theory. (5) Similar to the idea of the "edge of chaos" is a characterization of such systems as non-equilibrium or far-from-equilibrium. These terms refer to the way systems become fertile breeding ground for an emerging order when they are no longer bound to conditions maintaining the status quo or "equilibrium"--a state in which things are done in the same old way. (9)

Emergence of a new, higher order

Complex adaptive systems have multiple levels of organization, with agents at one level interacting and aggregating to create more complex behavior at a higher level. John Holland gives as an example the ant nest with the simple, stereotypical behavior of individual ants contributing to a colony that can adapt, evolve, and survive over long periods of time. "It is much like an intelligent organism constructed of relatively unintelligent parts," he says. (3) These agents or networks of agents are continually adapting and recombining in new ways as the system learns.

Unpredictability

The behavior of complex adaptive systems is said to be nonlinear, meaning that the interactions within the system yield behavior that is more complicated than the sum of the behaviors of the agents. Connections between cause and effect tend to be much less direct and obvious than thinking regarding simpler systems has suggested. While links between cause and effect may not totally disappear, several variables suggest that the long-term future of a complex adaptive system cannot be predicted, including:

1) the interaction of multiple agents whose individual behavior is constantly changing

2) the impact of random disturbances

3) the periodic aggregation of agents at higher, more complex levels of organization and behavior

The future of a GAS is thus unknowable, perhaps even bizarre. Many refer to this as the "butterfly effect," a term which derives from the famous and often misquoted presentation by MIT meteorologist, Edward Lorenz, titled "Does the flap of a butterfly's wings in Brazil set off a tornado in Texas?" (10) While this image compellingly captures the notion of unpredictability, Lorenz was specifically exploring how small changes in the initial conditions of complex physical systems--such as the weather system--can sometimes trigger huge changes over time. In complex adaptive systems, there are, of course, some limits to unpredictability.

In any given GAS, there will be some "structures" that limit emergent behaviors. For example, particular states of weather may be unpredictable--but the climate is predictable (on appropriate temporal scales). The climate serves to structure the weather to fall within a certain range of possible states, depending, for example, on the season or geographic location. It needs to be pointed out that some aspects of these systems remains predictable. For example, the mathematician Ben Goertzel distinguishes between state and structural unpredictability: chaotic systems are state unpredictable, meaning that the specific state of the weather a few days in the future is unpredictable, whereas the structure, or the range of possible states, is not unpredictable. In fact, the climate, as a "structure" which limits the possible states of the weather, is largely predictable. (11)

No central control

With multiple agents interacting and changing In parallel, it is readily apparent that no single agent, no matter how highly placed, is singularly "in control." Rather, overall control in complex adaptive systems is said to be highly dispersed. Any coherent behavior derives from the patterns of interactions which emerge from the competition and cooperation of the agents, all acting independently.

Impacting a complex adaptive system

If no one is in control and the long-term future is unknowable in a complex adaptive system, an obvious question arises: "How can one have an impact on such a system?"

There are powerful implications in the notion that every agent in the system can have an impact, sometimes--and particularly under conditions of instability--out of proportion to what seems "logical," at least from a linear, cause-and-effect perspective. (When a system is stable in various phases and/or regions, individual agents would be less likely to have so large an effect.) Complexity suggests new insights related to this potential influence of individual agents.

Viewing a patient, family, or an organization from a complex adaptive system perspective brings to light different features than a mechanistic orientation would illuminate. Ralph Stacey offers a simple two-dimensional construct (Figure 1) that allows a simultaneous consideration of both mechanistic and GAS perspectives. (12)

This construct suggests that traditional management, of both patients and organizations, is most applicable in situations characterized by a high degree of certainty and agreement. The lower left zone of Stacey's diagram is where rational planning and decision-making methods should be used and where measures, controls, and negative feedback are appropriate means of ensuring that intentions are realized. This is exactly the approach desired by the healthy patient undergoing elective joint replacement-a situation enjoying relatively high agreement and certainty regarding "the best way" to accomplish this procedure.

In science, author Kevin Kelly writes, "For jobs where supreme control is demanded, good old clockware is the way to go." (13) [Note: the authors define clockware as the "management processes that are rational, planned, standardized, controlled, measured, and efficient," and its contrasting term swarmware as the "process of exploring new possibilities--experimentation, trials, risk-taking, autonomy and freedom of agents, intuition, and working at the edge of knowledge and experience."]

But some situations are not defined by high degrees of either certainty or agreement. For example, a healthy total hip replacement patient poses to the clinician a much simpler challenge than one with multiple interacting diagnoses, poorly understood symptoms, etc. As Figure 1 suggests, more ambiguous situations demand a fundamentally different approach than those which enjoy high-certainty, high agreement regarding their basic nature. Stacey--and a growing cadre of complexity researchers--fundamentally challenge the pervasive machine metaphor, which has taught us to use clockware methods in all circumstances.

Complexity science helps us to appreciate that there is a space between the ordered stable region and its opposite, the zone of chaos, disintegration, and anarchy, where we are far from both agreement and certainty. It is in this "zone of complexity on the edge of disintegration" that promising approaches to complicated issues may be forthcoming. In Kelly's words, "Where supreme adaptability is required, out of control swarmware is what you want." (13)

As it receives attention from a growing array of scientists representing multiple disciplines, complexity is demonstrating its relevance to systems ranging from immune systems to human beings to complicated organizations. This emerging science is beginning to offer tentative strategies and methods for working with situations falling within Stacey's "zone of complexity, at the edge of disintegration." Among these methods, a common thread is becoming evident: both the physician treating a complicated patient, and the leader dealing in ambiguous environments might do well to foster conditions under which adaptability and creativity can emerge from within the system. In the language of Gareth Morgan at the VHA Leadership Task Force meeting In November, 1996, "Farmers don't grow crops, they create conditions in which crops grow."

Case studies

Case 1: A Patient with Dissociative Identity Disorder (DID)

Several medical researchers and clinicians have recently focused on the potential value of applying the perspectives of complexity science to both patients and patient care systems. Published examples include the fields of cardiac electrophysiology, (1) blood pressure control. (14) pulmonary function, (15) the mind-brain interface, (16) and clinical pathways. (17)

This literature suggests that complexity thinking may be useful in understanding and treating clinical conditions, including behavioral/psychiatric syndromes. In fact, because many patients suffer from conditions involving a complicated interplay of biological, psychological, and sociological factors, an approach that emphasizes nonlinear and complex adaptive (and maladaptive) interactions might constitute an ideal framework in which to formulate diagnostic and therapeutic intervention.

The first case study describes a patient with dissociative identity disorder (a "multiple personality"), and the customary treatment approach to this diagnosis.

Description of the patient

At the time of her presentation to therapist AH for a second-opinion consultation, SW was a 40-year-old divorced mother of two. During three weeks of inpatient treatment on a psychiatric unit, she remained dissociated much of the time, and was both suicidal and regressed. Important elements of her history included:

* a long history of dissociative episodes

* little memory of her high school experience

* polio as a young child

* physical abuse by her alcoholic mother during her childhood.

* fragmentary memories of sexual abuse as a child by a male, and clear memories of both physical and sexual abuse by her ex-husband.

Further, her dissociative episodes had worsened when she discovered that her (then) husband also had sexually abused their daughter. She decompensated completely after she lost legal custody of her daughter to her ex-husband (on grounds that SW was psychiatrically ill). AR agreed to become SW's therapist when she told him that her treating therapist had had sexual contact with her several months ago.

Traditional treatment approach

While the treatment of DID is anything but uniform, SW's two-year experience with her previous therapist appeared be something of a "hit and miss" approach, resulting in increasing dissonance, and, ultimately, a suicide attempt. Further, the sexual advance by this therapist had reinforced her experiences as a child and young adult. Clearly, she had been abused all over again, an important factor in her presenting in a decompensated state.

Recognizing that DID represents a challenging diagnosis for therapists, there are alternatives to "hit and miss." Many eclectic therapists, including AR, generally have followed the approach outlined by Ross involving three phases: (18) (1) a preparatory/mapping phase: (2) a middle "working through" phase; and (3) an integration phase. AH followed this broad outline during a three-year course of therapy. He introduced certain modifications from complexity perspectives.

During the first year of therapy. SW's awareness of three alternate personalities emerged:

1. a very young, virtually nonverbal child

2. an older, quite verbal girl who acted as the child's protector

3. a teenager who was named "the bitch" and who wanted to strike at others and inflict pain.

Within the trusting relationship during the mid-phase of this therapy, SW became aware of the underlying dynamics and roles of the dissociated "part-selves":

1. a morbid fear of abuse in the child

2. a desire in the older child (however fragile at times) to survive

3. rage and anger for having been hurt and mistreated in "the bitch."

During the integration phase, the dissociative episodes gradually stopped as SW began to understand the relationship between her affect and actions.

Case 2: Physicians reorganize the hospital medical staff

As health care provider organizations attempt to restructure, they often face situations that feature uncertainty and little agreement. The following case study describes the experience of a group of hospital medical staff leaders as they discovered the "zone of creativity at the edge of disintegration."

Description of the situation

Medical staff leaders of an academically affiliated, Midwestern community hospital were frustrated. They perceived that important issues were ineffectively addressed by a medical staff organization that encompassed no less than 26 committees. Further, physicians were feeling increasingly disenfranchised by the hospital administration's efforts at reengineering patient care processes in order to achieve cost reductions. Whereas physicians had previously exerted major influence on operational changes, decisions affecting their practices were increasingly made outside their realm of influence.

Finally, the medical staff executive committee (MEG) took action. They commissioned a task group to consider the governing structure and process of the medical staff. The group's charge was to come up with explicit recommendations for action.

Traditional approach

The task group proceeded according to the four traditional functions and presuppositions of management:

1. planning, based on the assumption that the future can be predicted accurately:

2. organizing, based on the assumption that task structures are to be imposed on organizational members;

3. controlling, based on the assumption that deviations from normative practices should always be diminished:

4. leading, based on the assumption that the leaders are "experts" and, accordingly, their goals and "visions" are sufficient for organizational direction and motivation. Members expected that the consulting firm they engaged would have definitive answers, particularly to the planning and leading aspects of their work. After all, weren't hospital medical staffs across the U.S. experiencing similar stresses, and weren't experts supposed to know the future and to have answers to issues of organizational efficiency and effectiveness? Wasn't it simply a matter of seeking out "best practices" from other institutions, and transferring this knowledge?

The task group members' implicit assumption was that they were asking questions in a realm where there existed a relatively high degree of certainty, and at least some degree of agreement regarding best methods; in sum, they were operating in the "lower left clock-ware" zone of Figure 1. Monthly meetings continued over approximately five months. The consultants, who attended each meeting and researched questions in the interim, kept detailed minutes. The group reached consensus on some issues:

* the functioning of medical staff departments, committees, and communication mechanisms was ineffective and inefficient

* related to this ineffectiveness, administration was filling the "action vacuum." and becoming increasingly influential in matters directly bearing upon clinical care (for example, downsizing)

* the medical staff was dealing with increasingly complicated issues, such as outpatient planning, newly forming economic entities (independent practice associations, physician-hospital organizations, group practice growth, etc.)

In addition, some progress was made on relatively non-controversial issues. The task group:

* affirmed the primary role of the medical staff as credentialing, quality assurance, enforcing rules and regulations, and professional education

* proposed that economic issues not be addressed directly by the medical staff

* acknowledged that the entire medical staff was the appropriate authority for "major policy decisions"

* agreed that medical staff leaders needed greater support from administration

But they were not achieving movement on what they had identified as the most important issues: (1) improved efficiency and effectiveness of structures and processes; (2) greater influence in redesigning clinical systems; and (3) dealing with external issues that were impacting the hospital.

It was increasingly evident that definitive, "expert" answers simply did not exist for many of these more crucial issues. As they queried the consultants, answers routinely fell short of the explicit direction, models, and solutions that the physicians had anticipated. Progress slowed, and finally came to a virtual halt. The psychiatrist chairing the task group was unsure of how to proceed. For her part, the CEO was becoming concerned that the hospital's financial support of this effort was returning minimal results.

The limits of past approaches

The traditional approaches to both SW's care and the medical staff reorganization effort achieved some progress, but generally left parties with the perception that better results might be achievable. This does not mean that traditional approaches were "wrong" in any absolute sense. But, as each case makes clear, these approaches were of limited effectiveness, and left some, if not all, parties dissatisfied. While these cases may, on the surface, appear "typical" of the problems clinicians and organizational leaders face, their effective solution involves complicated interacting variables, in areas where both clinical and management science remain incomplete.

In the past, those involved no doubt would have "muddled through," doing the best they could with available methods. However, the pressures which physicians and health care leaders now experience dictate that such an approach, no matter how well intentioned, may not be enough. Too often, the results may be unsatisfactory, reached at an unacceptable financial cost. Leaders who face extraordinarily complicated situations might well consider new frameworks.

Create "edge of chaos" conditions

From the concepts of complexity, what guiding principles emerge for clinicians and leaders? Cultivate/create "edge of chaos" or "far from equilibrium" conditions (in appropriate patient care and organizational situations). Stacey suggests that leaders work with several parameters to foster this "edge of chaos" condition. (5) Three of these parameters stem directly from the science of complexity--the pace of information flow, the diversity of the agents, and the number of connections among agents. If the complex adaptive system involves humans, he describes two additional parameters--the level of contained anxiety and the degree of power differentials.

Stacey's central notion is that these parameters can be used to help nudge a complex adaptive human system--the complicated patient, the organization in ambiguous circumstances--to the zone of creativity and adaptability. Implied is the importance of finding a proper degree or rate for each parameter. For example, a human system with insufficient diversity and connectedness, with low levels of anxiety, and a high concentration of power will tend to be static and stable; it may not progress. Conversely, too much information, diversity, too many connections, and extremely high levels of anxiety and little power differential correlate to a system Out of control--in the disintegration zone of

Figure 1.

For maximum creativity and effective innovation the mid-zone between stagnation and disintegration is the place to tune the organization. For the patient, this implies that the physician should exercise an artful amount of direction and anxiety containment, even as (s)he encourages initiative and responsibility-taking on the part of the patient. For the organization, it implies a proper amount of education/information and assurance, even as leaders challenge all organizational agents (i.e., individual people, work groups, teams, etc.) toward creative approaches to their tasks.

It seems possible that when these control parameters are calibrated correctly they will drive a human system to find this space of creativity, where Kauffman says a system finds good enough solutions consistently and persists over time. (13)

Uncover and work with high leverage paradoxes

Gareth Morgan writes that edge of chaos circumstances manifest themselves as paradoxes or tensions between the status quo and alternative future states." (2) Without skillful intervention, patients and organizations are likely to continue in their patterns. In this context, both clinicians and organizational leaders might come to understand that their efforts to foster change will naturally encounter those who are more comfortable with, and may seek to hold onto, prior behaviors.

Linear models have tended to identify such behaviors as "resistance" often setting the stage for adversarial power struggles between "change agents" and "resistors." In fact, the whole idea of resistance needs to be revised in the light of complexity science. Instead of resistance," the notion of "attractor" from nonlinear dynamical systems theory an be understood as whatever it is that constrains the possible behavior of organizational members. (9) Utilizing the idea of "attractor" rather than resistance shifts the focus of organizational change from "overcoming" employee "resistance" (as if people were just like donkeys!) to eliciting commitment to more appropriate "attractors".

Using complexity, leaders might instead seek to employ "interventions that transcend the paradoxes or make them irrelevant." (2) The process might be likened to the context, such as redefining what some might term a "problem to be solved" into an "opportunity to be pursued." Morgan specifically cites the value of successful prototypes (e.g., innovative treatment approaches for patients, and structure or process interventions for organizations), rallying key opinion leaders not against prior behaviors, but instead in support of promising, new alternatives. It is useful to return here to the "butterfly effect" and remember that small changes in complex nonlinear systems can lead to large changes in the whole system.

Build a " good-enough" set of intervention objectives

Good enough suggests that aspirations for a system (e.g., treatment goals in patient care, and vision/purpose statements for projects or organizations) should provide direction, while simultaneously not being overly prescriptive or confining. Rather, skillful intervention should generate a space within which a system can explore, create, and organize.

In developing the topic of anxiety and creativity, Stacey references D.W. Winnicott's concept of the "good enough mother," who knows "how much to frustrate the infant and how much to hold him or her."5 Similarly, Morgan writes of "minimum critical specifications"--such elements as mission statement, guiding principles, boundaries, etc.--versus the typical organizational approach of "grand design" and "overcontrol. "2 In providing for such minimum critical specifications, while calling simultaneously for space for creativity and self-organization, Stacey and Morgan are aiming at edge of chaos, creative conditions.

Go for multiple options at the fringes and protect them

Because it is not possible to predict accurately the future of a complex adaptive system, it makes sense to foster multiple experiments. This approach tests what the system can produce or become, and explores how new possibilities might play out. Such actions, if they truly seek to move the patient or organizational system into new territory, also represent "on the edge activities." Because these experiments may be threatening to the status quo, they need to be protected.

For example, within the relatively protected environment of a few academic medical centers of the 1950s and '60s, a number of thoracic surgeons gained valuable experience in pioneering openheart techniques. Their innovations opened the way toward the scientifically validated methods that now save lives daily. Kelly posits that "A healthy fringe speeds adaptation, increases resilience, and most always is the source of innovations." (13) Similarly, Morgan emphasizes the importance of a "multiple option" approach, noting that such experiments "offer important insights on the negative feedback loops and defensive routines that sustain a dominant attractor pattern, and what can be done to help a new one emerge." (2)

Let direction emerge from action

This suggestion is closely tied to the notion of experiments, which must be more than thought experiments if one really wants to test what is possible and relevant. Kevin Kelly cites a number of researchers' work, concluding that populations that are exploring possibilities by changing behavior (i.e., learning via action) evolve faster than populations that do not. (13) In a similar vein, Stacey says that "vision and long-term plans are a fantasy (or illusion) and defenses against anxiety (or interpretation with hindsight) ...and that answers and direction emerge from action." (19)

Research has revealed that significant organizational change can emerge by taking advantage of accidents or unexpected events. (9) Indeed, this follows from the study of self-organization in complex systems in which new structures can come about through the amplification of random events.

Consult the shadow system

Stacey posits that the shadow network, the set of informal relationships that always exist in tandem with the legitimate, "official" network or hierarchy, is the locus of most creativity in human systems. While the legitimate system is focused on maintaining stability and the status quo, the shadow network has no such constraints. It is where innate human creativity can most naturally reside. Stacey also stresses the necessary. even complementary, roles of the legitimate and shadow systems in an organization. (5) How leaders address the natural tension that arises between these systems will critically influence whether this tension evolves as unhealthy and destructive, or healthy, and thus, creative.

The cases revisited

Case 1: The DID patient

The goals of treatment are to relieve pain and suffering, within a context of compassionate care, and if possible, facilitate a more adaptive response to noxious stimuli, for example, to promote healing and health. Complexity science suggests new approaches, especially to the second goal of helping patients such as SW achieve a more integrated, adaptive level of functioning. It does not replace traditional clinical treatment, but instead offers complementary approaches that encompass both new perspectives in conceptualizing DID and alternative interventions.

Emergence of a new, higher order

As previously noted, Holland defines adaptation (in the biological-behavioral sense) as the process whereby an individual "fits" him or herself into his or her environment. (3) As he postulates, this occurs as multiple agents--in the case of SW, physiologic, cognitive-behavioral, and affective stimuli--interact with one another, resulting in a series of adaptations. These adaptations might be either effective or counterproductive. For example, SW's regressed psychomotor functioning might be understood as a dysfunctional adaptation to powerful affective stimuli of fear. Given this understanding, her dissociations ("not being able to remember anything") emerge as a protective adaptation that serves the purpose of shielding her from conscious awareness of extremely unpleasant, harmful experiences.

Presumably, given sufficient time, individuals with DID would either adapt "well enough" to be effective in the here-and-now or, alternatively, evolve into an overall dysfunctional state. Therapy from the CAS perspective can be conceived as a way to positively influence the rules by which the various agents interact. The goal is to avoid the dysfunctional pathway in favor of the "emergence of a higher order--the effective adaptive state (healing).

Uncover and work with high leverage paradoxes

Gareth Morgan's concept of leveraging paradoxes is relevant to the overall treatment approach that AH pursued with SW. Morgan postulates that working with the critical 15 percent of variation within a system, and influencing this "high leverage" percentage correctly, can positively affect the entire system behavior pattern. (2) In the case of DID, this approach allows the therapist to avoid a repetitive retracing of earlier-life traumatic experiences, concentrating instead on the 15 percent of the paradoxes that appear to matter the most. AH concentrated on the exploration of two paradoxes: (1) SW's early-life sexual abuse, yet later-life repetition of sexually-abusive relationships; and (2) SW's flight into dissociative episodes as a way to reduce anxiety.

Many traditional psychotherapeutic interventions were involved in this treatment. However, framing the approach within a complexity science format allowed for both a directness of intervention with the patient and a powerful therapeutic direction. As noted, SW's maladaptation was to use repression ("avoidance") to an extreme. AH chose to focus on this "high leverage" element of SW's overall therapy as he sought to work with her toward more effective adaptation. Accordingly, he repeatedly gave her these messages (or "rules"):

1. together, you and I have 'to look the tiger in the eye;'

2. I realize from your self-reports that often you can see literally only one half of me. What is important is that you can learn to look inward and begin to see the whole picture;

3. your life experiences, however painful, define you. These experiences are telling you to listen. As you do that, we will be able to help you understand yourself better and become more functional.

With SW, it took a year of frequently reinforcing these "rules"; but, gradually, over a three-year period, the effect was to allow SW to gain an emergent sense of a non-dissociating self. In fact, after about the first six months, she was able to "catch" herself during "phase transitions" (i.e., just as she was beginning to dissociate). According to complexity expert Christopher Langton, phase transition in the physical sciences is that state when matter exists as both a solid and a liquid. (20) As applied to DID, phase transition might be understood as the behavioral state just as a patient begins to dissociate--when the unstable sense of self exists as both order and chaos.

Cultivate "edge of chaos" conditions

Far from seeking to avert or abolish it, therapist AH worked with this unstable, "phase transition" state. He asked SW to concentrate on these "edge of chaos" experiences, with the hope that she could eventually forestall progression into chaos (i.e., dissociated episodes). This approach appeared successful. She was able to identify these "phase transition" events by an accompanying feeling of intense headache and fluttering eyelids. At first, this event was always followed by dissociative episodes for which she had no recall. But, with repeated attention to these phase transitions, she was more and more able to stay in the "here-and-now," yet recall more and more of heretofore unconscious memories.

As she gained more control during these episodes, she was progressively able to influence them toward a more effective adaptive state.

The ultimate test of a successful healing encounter--whether in general medicine or in psychiatry--is whether any sustained physiologic or behavioral change occurs. In the case of DID, the test is whether dissociative episodes stop and the person is able to maintain stability.

No central control

As this case illustrates, AH did not "cure" SW via the use of some "magic bullet" therapeutic intervention. He did not act as the "central controller" of the process. He instead assumed the role of an (admittedly very influential) agent in the emergent physician-patient system. His choice to "cultivate edge of chaos" conditions clearly required SW both to trust him and to have the courage to face the extreme discomfort of her "phase transitions." It was essential for her to assume an active role as "interactive agent" of the physician-patient system. In sum, the positive result was a true co-creation of both therapist and patient.

Figure 2 describes some additional aspects of the traditional and the complexity science approaches to clinical decision-making and treatment interventions. The reality is that we probably do best (both as diagnosticians and as healers) when we make use of both approaches. In terms of SW, her disabling dissociative episodes stopped after about two-and-a-half years of therapy. It has now been about five years, and she has maintained her progress.

Case 2: The medical staff task group

The impasse which the medical staff task group reached using traditional change methods opened the mind of its leader to alternative approaches. He chose to discuss his situation with MM, who had worked with the organization in previous years. and who had some understanding of the historical hospital-medical staff culture. MM agreed to meet with the task group for the purpose of exploring ways of becoming "unstuck," of moving forward with the project.

At this meeting, task group members came to understand that they were indeed exploring a subject--namely the "optimal" future medical staff organization--for which there existed neither certainty nor agreement. It became evident that health care restructuring throughout the U.S. was moving so fast, and the dynamics of this movement were so unstable, that no expert could suggest a best" structure for how any specific medical staff might optimally reorganize. Indeed, task group members came to view the health care system as moving rapidly from a lower left clockware" status toward an "upper right swarmware" zone. (Figure 1)

Regarding "best' structures to deal with such uncertainty, the then extant cliche was apt: "Once you've seen one reorganization model--you've seen one!" Even accreditation standards, the previous guide to medical staff organizational configuration, had become so non-prescriptive and open-ended as to be almost useless in drawing up a precise organizational structure. In effect, modern accreditation standards amounted to little more than Morgan's "minimal critical specifications." (2)

Task group members came to understand that they were, in fact, operating in a "low agreement-low certainty' environment. Yes, there were examples of medical staff reorganization in development. But the uncertainty inherent in the restructuring of health care delivery dictated that all such examples were essentially experimental-and thus, unproved. With this understanding, they decided to abandon their monthly meeting schedule. MM suggested as an alternative a single, one-and-a-half day exploratory retreat.

This shift was not just one of organizational process--from a series of meetings to a single retreat--but also of underlying assumptions. The meetings had implied a rational, linear process encompassing:

1. control by the task force chairman

2. learn from the experts

3. reach consensus or vote on "best" approach

4. draw up a plan

5. assign roles to implement the plan

6. recommend this plan to larger group (the medical staff)

In contrast, the retreat mode assumed (in random order):

* "minimum critical specifications" which might guide a process (in this case, accreditation standards, scientific principles of quality, the values of medical practice, etc.), but no precise expertise on "optimal medical staff structure"

* unpredictability: while the process designers certainly intended an outcome, which would advance the task force's objective, the exact nature of this outcome was not precisely predictable

* the collective ability of the "independently operating agents"--the physicians attending the retreat--to generate a useful work product

* the likelihood that one or a small number of key ("attractor") issues would emerge

* no central control; MM agreed to offer educational input (focused on "minimum critical specifications"--in this case, quality science and accreditation standards) and facilitation only. The task force chairman remained technically the "leader" of the process. (In fact, his place at the table used for the retreat implied no authority; a visitor would not have been able to identify the chairman from the physical space he occupied.)

The task group agreed to the retreat, and developed the following agenda:

* an interactive, content-based learning session, in which the consultant presented salient features of the changing health care system, and in particular emphasized the high levels of uncertainty and limited "expert knowledge which existed regarding their work

* a detailed discussion of perceptions and reactions to this material

* an exploration of basic values which the task group members held, and believed should characterize their future medical staff

* a "future purpose" (i.e., mission) exploration of the medical staff organization, without pre-conceived ideas about what this purpose should be

* brainstorming of actions which might encompass values and fulfill the perceived medical staff purpose

* an (unpredictable) action agenda as an essential outcome (perhaps a minimal "guarantee" to offer any group of people who are committing to such an intensive process)

Though several deviations occurred- consistent with the unpredictable element of dealing with complexity--the retreat followed roughly the agenda described above. The initial educational session added two important pieces of information that the task group had not previously accessed:

1. Modern quality science--the principles and practices underlying emergent best quality management practice"--was relegating traditional physician peer review to a position of diminishing importance in overall hospital quality management, including physician performance appraisal. As events evolved, this realization proved a major "leverage point," one that allowed the physicians to shift from a previously dominant, but now obsolescent model of quality (peer case review) to one emerging from information technology and continuous quality improvement applications to health care.

2. Accreditation standards regarding medical staff structure and process had recently changed dramatically--largely as a result of the influence of modern quality science. This factor enhanced members' sense that they were truly free to become more creative regarding their future structure and process.

As a result of these developments, the task group members reached an emergent understanding that focused upon two issues: (1) How might they link their aspirations and values into future structures and processes; and (2) how might they formulate their conclusions into recommendations that they could "sell," not only to their colleagues, but also to the CEO and board.

Since clear-cut answers to their questions regarding "optimal" structure and process were not immediately apparent, task group members were left to wrestle with how to best link this new information to explicit recommendations. The consultant's role--as only one of multiple agents in this emerging, complex adaptive system--reverted from information source to that of process facilitator, keeping the process on track, bringing out multiple views, offering insights and comments, pushing creativity, and generally managing the "boundaries," i.e., whatever aspects of group dynamics are necessary to contain anxiety, constructively manage conflict, and stay intact but creative as a group.

After one-and-a-half days, a somewhat drained group had hammered out eight explicit recommendations, of which two were of core, potentially transformational significance for the hospital.

1. That a larger task group, consisting of physicians, the CEO and her management team, and trustees form a task group to deal with major initiatives in hospital-physician collaboration.

2. That this task group define and create the role of a full time (physician) chief medical officer. This person would work with the medical executive committee and the CEO to foster improved medico-administrative function generally, liaison with third parties (physician-hospital organization, managed care, etc.), and help develop clinical systems.

What complexity concepts did this process represent, and how were the results different than the traditional approach? Figure 3 helps answer this question by contrasting traditional and complexity-oriented approaches to organizational change.

The experience of the task group illustrates how traditional management functions and assumptions are transformed in the light of a complexity perspective.

1. Since complex, non-linear systems have been shown to be vastly more unpredictable than simple, linear systems, the notion of planning as prediction of the future needs to be rethought.

2. Because complex systems have the capacity to self-organize into new structures, the idea of organizing as imposing task structures needs to be supplemented by the possibility of a spontaneous emergence of new structures.

3. The fact that self-organization results, in part, from incorporating deviations from normative behavior means that the management function of controlling as the dampening of departures from the norm must be revised.

4. Leadership expertise and vision is now seen as emerging not from a single or a few "experts," but instead from an entire working group; it is no longer the domain of a designated leader.

Cultivate edge or chaos conditions

The educational piece of the retreat design allowed the realization that "expert" solutions were non-existent. The participants were able to: give up "clockware" fantasies; move to a "swarmware" mode; and focus within their own group for sources of creativity regarding their future medical staff organization. In effect, they adopted a new attractor." The group already consisted of a diverse array of physicians--different specialties, senior as well as more junior staff members, and those in solo and group practice. The retreat process, which added a facilitator physician who stepped out of the expert role, allowed all members of the group to have significant influence.

Since the group now understood that there would be no "right" way to resolve their charge, a freer flow of information was allowed. In reaching consensus on their final eight points, a number of other ideas were raised, discussed, and ultimately discarded. Additionally, their realization that no definitive answers to their questions existed and that many other physicians/medical staffs were in exactly their situation offered some comfort. Even if they couldn't define the answers to their charge, the anxiety generated by their uncertainty regarding the future was at least somewhat contained.

Emergence of new, higher order

The group went into the retreat with a relatively simplistic notion of "strengthening the medical staff" in order to compete more effectively with the CEO for institutional power. The "swarmware" information flow eased this traditional competitive mind-set. The interdependency of the clinical staff and management team became clear. As participants grasped the multidimensional nature of modern quality management--in contrast to the relatively simple process of traditional physician peer review--they realized that whatever their desires for the medical staff as a part of the hospital, they would need the active support and participation of the CEO, her management team, and the board of trustees. In moving from an implicit "we-they" mode with administration to a genuine "us" orientation, the group was moving, in a truly evolutionary sense, to a higher level of function vis-avis the overall hospital leadership.

Unpredictability, letting direction emerge from action

The flexibly-structured retreat mode could guarantee neither a positive result, nor definitive answers. It could possibly have resulted in a "descent into disintegration--the loss-of-control nightmare that leaders might fear if they "let go' of traditional, clockware management. But the retreat design, which featured a tested general agenda, "ground rules" established by the group, and, perhaps most important, the sincere positive intention of participants predicted, at least, a constructive process, whatever the outcome.

An explicit final recommendation that illustrated the unpredictable and creative aspect of complexity-based processes was the emergent concept of the chief medical officer. In fact, this medical staff had previously been very suspicious of what some perceived as a "medical czar" within their ranks--or worse, a potential "administrative stooge." But as they considered a variety of issues in a "swarmware" environment, this group of diverse physicians reached genuine consensus. But notice that this consensus was emergent; it was neither an imposed demand, nor the kind of premature conformity that more accurately reflects the frustration to get something done.

The direction toward a chief medical officer--that this role was, indeed, the best solution to some of their own leadership needs--emerged from the "action" of the retreat process. The task group chairman captured the essence of this unpredictable result well: "We've opposed the medical director concept for years, and no expert consultant could ever have talked us into it. I can't believe that we came up with this recommendation ourselves!"

No central control

If a surgeon acting in the "command-and-control" mode, which appropriately characterizes the operating room, had chaired the task group, the outcome might have been quite different. But perhaps one of the strokes of luck was that a psychiatrist--a physician accustomed to ambiguity and incomplete control--served as chairperson. MM accepted his leadership responsibility seriously, but was also non-directive, and willing to encourage and work with his colleagues' views.

Consistent with this non-directive leadership, MM actively shunned the expert" consultant model, and, following the educational presentation, assumed a purely facilitator role. The lack of central control forced the multiple agents" of the system (the individual members of the task group) to develop a coherent behavior out of the "competition and cooperation of the agents, all acting independently." In fact, there was considerable disagreement and debate during the retreat on many issues. But in the end, there was full, genuine agreement on the final recommendations. This outcome represents a complex system undergoing self-organization toward a more coherent mode of being organized.

Build a "good enough" set of intervention objectives

When the task group originally convened, members held the expectation that explicit plans, perhaps actual blueprints regarding the future medical staff structure and governance, would be their work product. As the ambiguity of their environment began to take hold, however, their implicit objectives began to change. Though they were able to suggest explicit approaches to simpler issues, their major recommendations to commission an expanded task group and explore the chief medical officer position were really just interim steps toward more explicit plans for structure and governance.

Though their work output was less definitive than originally hoped for, it was "good enough" for the group to feel positive about their accomplishment--and to establish the conditions and frameworks for progress. As the chairperson said, "We fell short of our original expectations, but we now understand why these expectations were not realistic. And we feel very good about what we did accomplish."

Epilogue to the task group process

The task group met once following the retreat to review documentation and refine some of the language of its recommendations. It then presented its work to the medical executive committee. The MEC, after minor language tinkering, adopted all recommendations. In a remarkable show of support, many task group members stood up to express their support of both the group process and its recommendations. For a medical staff culture in which visible, verbal support of colleagues was not a norm, this phenomenon may well signify a shift toward a new, more mutually supportive culture.

As required by the bylaws, the task group presented its recommendations to the medical staff. Because of the task group chairperson's illness, a member of the group at the last minute "stepped into the breach," delivering an impromptu presentation of the work that she and her colleagues had accomplished. According to the traditional view of management, unexpected occurrences such as this illness present serious obstacles to progress. However, in complexity science, random events are seen as a crucial element in the process of self-organization.

In fact, the new structures that emerge are partly the result of the amplification, not diminishment, of random events. (9) Therefore, unanticipated events can be utilized in the ongoing generation of innovative organizational forms. In fact, the quality of this physician's "last minute sub" presentation was articulate and persuasive: the medical staff as a whole--approximately 100 physicians--adopted the recommendations without a dissenting vote. A more broadly based task group, inclusive of the board of trustees, CEO, and top management, is laying the groundwork for collaborative clinical-management redesign of health care processes.

Possibilities

The patient care and organizational cases described illustrate how a complexity perspective brings insights to the fore and triggers useful new thinking and efficacious actions. It seems likely that CAS theory can and will contribute fresh understanding and creative, workable approaches to a host of other Important issues facing physicians in practice and in leadership positions.

Complexity science holds the promise of creating better approaches to patient care and treatment, and healthier, more naturally adaptive health care organizations, Health care professionals and leaders must devote themselves to serious learning about complex adaptive systems, with an appreciation that complexity is a major advance in our understanding of living systems, not a passing fancy.

FIGURE 2

Traditional and Complexity-Informed Clinical Approaches

Category                    Traditional

Requisite information base  * assumed to be contained
                              in the facts of the
                              history and lab results

Role of diagnostician       * deductive thinking
                              predominates--
                              assembling history,
                              X-ray images, and lab
                              findings yields diagnose



Definition of treatment     * emphasizes scientific
process, objectives           approach--target
                              biologic/molecular
                              intervention or surgical
                              "cure"

Role of treating clinician  * assumes a
                              director/control
                              position of the "expert"





Category                    Complexity Science

Requisite information base  * at best, facts are incomplete and
                              often full of paradox and
                              contradiction(s)

Role of diagnostician       * understand system by interacting
                              with it
                            * generally requires both deductive
                              and inductive thinking
                            * identify the context
                              contradictions
                            * look for paradox and
                              contradictions
Definition of treatment     * emphasizes healing over "curing"
process, objectives         * considers the importance of
                              context and metaphor in the
                              illness process and how best to
                              influence same

Role of treating clinician  * frames questions which explore
                              and guide the metaphor

                            * "nudges" the parameters of the
                              system via leveraging of the
                              paradox(es) to bring
                              about changes

FIGURE 3

Traditional and Complexity-Informed Organizational Approaches

Category                            Traditional

Guiding principle                   Seek conformity to plan
                                    or consensus


Requisite knowledge, models         Assumed to be "out there,"
                                    experts can access;
                                    creativity optional


Role of "outside expert"            Provide definitive
                                    information, direction



Definition of process, objectives,  Precise. Rifle metaphor:
end point                           "ready, aim, fire!" toward
                                    clear objective





Role of leader                      Direct, control, reduce
                                    conflict, set direction




Role of group members               Dialogue, discuss,
                                    incorporate knowledge,
                                    follow direction



Primary guidance source             External conditions, expert
                                    knowledge, "SWOT" * analysis





Category                            Complexity Science

Guiding principle                   Highlight differences of
                                    opinion, perception,
                                    attitude

Requisite knowledge, models         Knowledge, models imcomplete
                                    at best; potential exists
                                    within the system; creative
                                    component essential

Role of "outside expert"            Provide information "bits,"
                                    facilitate process toward
                                    emergent actions, work with
                                    edge of chaos conditions

Definition of process, objectives,  Imprecise, more
end point                           unpredictable. Auto-pilot
                                    metaphor: Ready, fire,
                                    aim! - toward "good enough"
                                    intention, constant, ongoing
                                    feedback, course correction
                                    necessary

Role of leader                      Frame questions, challenge,
                                    manage constructive conflict,
                                    tap shadow system, uncover
                                    divergent viewpoints, follow
                                    sometimes

Role of group members               Listen, seek understanding of
                                    diverse viewpoints, dialouge,
                                    discuss, create, seek novel
                                    approaches from differences,
                                    lead sometimes

Primary guidance source             Internal intention,
                                    interaction with other
                                    systems values, vision,
                                    natural creative tendencies,
                                    tools of innovation,
                                    creativity

* Strengths, Weaknesses, Opportunities, Threats- a framework that
assumes at least a reasonably stable context, and thus relatively
predictable future.

(In contrast, complexity assumes unstable conditions that, according to
Stacey, render predictions, to varying degrees, a fantasy status.)

References

(1.) Goldberger, Ary L. "Chaos and Fractais in Human Physiology." Scientific American. February, 1990, pp. 43-49.

(2.) Morgan G. Images of Organization. second edition. Thousand Oaks. CA: Sage Publications. 1997.

(3.) Holland, John H. Hidden Order: How Adaptation Builds Complexity Reading. MA: Helix Books, 1995.

(4.) Goldberger. Ary L. "Non-linear dynamics for clinicians: chaos theory. fractals, and complexity at the bedside." Lancet, Vol. 347, May 11, 1996. pp. 1312-1314.

(5.) Stacey, Ralph D. Complexity and Creativity In Organizations. San Francisco. CA: BerrettKoehler Publishers, 1996.

(6.) Goldstein, Jeffrey. The Tower of Babel in Nonlinear Dynamics: Towards the Clarification of Terms. In R. Robertson & A. Combs (Eds.). Chaos Theory in Psychology and the Life Sciences. Hillsdale. NJ: Lawrence Eribaum Associates. Inc., 1995. pp. 39.48.

(7.) Waldrop, M. M. Complexity: The Emerging Science at the Edge of Order and Chaos. New York, NY: Simon & Schuster, 1992.

(8.) Kauffman, Stuart At Home in the Universe. New York. NY & Oxford, England: Oxford University Press, 1995.

(9.) Goldstein. Jeffrey. The Unshackled Organization: Facing the Challenge of Unpredictability Through Spontaneous Reorganization. Portland. Oregon: Productivity Press, 1994.

(10.) Lorenz, Edward. The Essence of Chaos, Seattle, WA: University of Washington Press, 1993.

(11.) Goertzel, Ben. The Evolving Mind. Langhorne, PA: Gordon & Breach, 1993.

(12.) Stacey, Ralph D. 'Emerging Strategies for a Chaotic Environment." Long Range Planning, Vol. 16. April, 1996, pp. 182-189.

(13.) Kelly K. Out of Control: The Rise of Neo-Biological Civilization. Reading. MA: Addison-Wesley, 1994.

(14.) Wagner. C.D., Nafz. B. and Persson, P.B. "Chaos in Blood Pressure Control." Cardiovascular Research, 31: pp. 380-387, 1996.

(15.) Nelson, T.R., West. B.J., and Goldberger. A.L. The Fractal Lung: Universal and Species-related Scaling Patterns. Experientia, 46: pp. 251-254. 1990.

(16.) Morowitz, H.J. and Singer, J.L. The Mind, the Brain, and Complex Adaptive Systems. Reading, Mass: Addison-Wesley Publishing Company, 1995. pp. 25-43.

(17.) Priesmeyer. RH. "Chaos Theory and Clinical Pathways: a Practical Application." Quality Management In Health Care. 4(4), pp. 63-72, 1996.

(18.) Cohen. L., Berzoff, J. and Elm. M. Dissociative Identity Disorder. Northvale, NJ. and London: Jason Aronson. Inc.. 1995, pp. 413.434.

(19.) Stacey, Ralph D. Managing the Unknowable: Strategic Boundaries Between Order and Chaos In Organizations. San Francisco, CA: Jossey Bass Publishers, 1992.

(20.) Langton. CG 1986. 'Studying Artificial Life with Cellular Automata." In D. Farmer, a. Lapedes, N. Packard, and B. Wendroff (Eds.) "Evolution, Games, and Learning: Models for Adaptation in Machines and Nature." Proceedings of the Fifth Annual Conference of the Center for Nonlinear Studies, Los Alamos, NM. May 20-24, 1985. Amsterdam: North-Holland. pp. 120-149.

(21.) Lane, D. & Maxfield, R. "Strategy under Complexity: Fostering Generative Relationships." Long Range Planning. Vol. 29, April, 1996, pp. 215-231."

(22.) Hurst, David K.. and Zimmerman, Brenda J. "From Life Cycle to Ecocycle: A New Perspective on the Growth, Maturity, Destruction, and Renewal of Complex Systems." Journal of Management Inquiry, Vol. 3, No. 4 December. 1994. pages 339-354.

(23.) Goldstein, Jeffrey. Beyond Lewin's Force-Field: A New Model for Organizational Change Interventions. In F. Massarik (Ed.) Advances in Organization Development (Volume 2). Norwood. NJ: Ablex Press. 1993, pp. 72-88.

(24.) Caulkin, Simon. "Chaos, Inc.," Across the Board, Jul/August, 1996 pages 32-36.

(25.) Petzinger. Thomas, "The Front Lines," columns. The Wall Street Journal. July 12, 1996. October 18, 1996, January 3. 1997. March 7. 1997. October 24, 1997.

RELATED ARTICLE: Complexity Resources

A growing list of researchers and authors are writing on the subject of complexity Among the most readable introductory works are those by journalists M. Mitchell Waldrop (7) and Kevin Kelly. (13) For those wishing to dig deeper, the works by Arthur, Axelrod, Holland, Kauffman, and Prigogine are recommended. One of the recognized centers for the scientific pursuit of complexity is the Santa Fe Institute in New Mexico. The Institute's web site covers the educational programs offered, summarizes research activities, provides a list of publications, and lists the researchers connected with the Institute. For organizations with a serious interest in the implications of complexity. The Institute sponsors the Business Network for Complex Systems Research. Work on the translation of complexity into medical and organizational practice is just starting.

For those interested in the medical and organizational implications of complexity, there are a few places to turn. Ary L. Golberger, MD, is the most notable medical researcher and writer. His works in Lancet (4) and Scientific American (1) are suggested. The Lancet essay contains a reference list of other medically oriented works. Two of Ralph Staceys works and Gareth Morgan's second edition of Images of Organization, (2) especially chapters four and eight, are good sources for those interested in leadership and organizations. Also highly recommended are the books and articles in the resource guide by Lane and Maxfield, (21) Hurst and Zimmerman (22) Goldstein, (6, 9, 23) Caulkin, (24) and selected "The Front Lines" columns by Thomas Petzinger in The Wall Street Journal. (25)

VHA, a national network of 1,400 community-owned health care organizations, is sponsoring an effort to bring the benefits of complexity to health care. The authors are part of this initiative which involves creating multiple learning networks for physicians, nursing leaders, and health care executives, action research to gain experience with complexity theory in a variety of settings and on myriad issues, and developing complexity related educational approaches and design principles and methods for use by health care personnel.

In addition to the references used for this article (please review reference section on page 20) the following publications offer additional background on complexity theory and its application in a variety of fields.

Arthur, W. B. "Increasing Returns and the Two Worlds of Business." Harvard Business Review, July-August, 1996, pp. 100-109.

Axelrod, R. The Evolution of Cooperation, New York, NY: BasicBooks, 1984.

Dooley, Kevin & Johnson, Timothy, TQM, Chaos, and Complexity. Human Systems Management, Vol. 14,1995, pages 287-302.

Huber, G.P & Glick, W.H., ed. Organizational Change and Redesign: Ideas and Insights for Improving Performance. New York, NY: Oxford Press, especially chapter by Weick "Organization Redesign as Improvisation" and chapter by Cameron, Freeman and Mishra "Downsizing and Redesigning Organizations," 1993.

Kauffman, S. A. "Antichaos and Adaptation." Scientific American, August, 1991.

Kauffman, S. A. The Origins of Order: Self Organization and Selection in Evolution. New York, NY: Oxford University Press, 1993.

Lewin, Roger: Complexity: Life at the Edge of Chaos. New York, NY: Collier Books, 1992.

Lindberg, Curt & Taylor, James, "From the Science of Complexity to Leading in Uncertain Times." Journal of Innovative Management, Summer, 1997, pages 22-34.

Nohira, Nitin & Berkley, James D. "An Action Perspective: The Crux of the New Management." California Management Review, Vol. 36, No. 234, Summer, 1994, pp. 70 - 92.

Prigogine, I. & Stengers, I. Order Out of Chaos: Man's New Dialogue With Nature. New York, NY: Bantam, 1984.

Web sites of interest

A wealth of resources is available on the world wide web. Below are some suggested sites. These sites can lead to many others.

http://www.bionomics.org - This is the site of the Bionomics Institute, a nonprofit educational organization devoted to helping people to understand the economy as a complex adaptive system. Resources from this institute's journal, conferences, and the writings of its founder, Michael Rothchild, are contained here.

http://www.brint.com/systems.htm - This site is the work of a University of Pittsburgh doctoral student. It contains a host of references, papers on business applications, and links to other web sites.

http://www.santafe.edu - This site is maintained by the Santa Fe Institute, one of the acknowledged centers of complexity science. It offers access to the scientific work, educational offerings, and background on SFI, its faculty, and business network.

http://www.calresco.force9.co.uk - This is the site of a non-profit organization, CALResCo, dedicated to promoting understanding of complex system sciences. It is full of useful information and links to other sites.

http://www.industrialstreet.com/chaos/metalink.htm - A wide array of general complexity and chaos offerings are found here--from bibliographies to pictorial representations of the Mandelbrot set to general educational offerings on chaos and complexity.

http://www.lissack.com/writings/ - A number of papers written by Michael Lissak on the implications of complexity for knowledge development and management are found here.

For a more comprehensive, annotated resource guide, e-mail Curt Lindberg at [e-mail]. call 609/395-7776. or write to the VHA. 68A South Main Street, Cranbury, New Jersey, 08512.

--Curt Lindberg, Alfred Herzog. MD. Martin Merry. MD. & Jeffrey Goldstein. PhD

Curt Lindberg, recently accepted a position as Senior Consultant for Complexity Management. He is responsible for researching and bringing promising new leadership and organizational concepts to VHA members. He previously served as President of VHA of New Jersey He can be reached at 609/395-7776 or via email at [e-mail].

Alfred Herzog, MD, is the Vice President of Medical Affairs at Hartford Hospital in Hartford, Connecticut and is in clinical practice in adolescent and adult psychiatry. He is a Professor of Clinical Psychiatry at the University of Connecticut School of Medicine in Farmington. He can be reached by calling 860/545-3501 or via fax at 860/545-3622.

Martin D. Merry, MD, is Associate Professor of Health Management and Policy at the University of New Hampshire and a Senior Advisor for Medical Affairs for the New Hampshire Hospital Association. He can be reached by phone or fax at 603/778-1531.

Jeffrey Goldstein, PhD, is a Professor in the School of Management and Business at Adelphi University in Garden City. New York. He is author of The Unshackled Organization: Facing the Challenge of Unpredictability Through Spontaneous Reorganization. He can be reached via email at [e-mail].