Complexity Theory and Decision Making in Health System

“ I never make predictions, especially about the future.” Sam Goldwyn Mayer

How Hazardous is Health Care Dangerous Safe 100K Health Care Driving 10K 1K Scheduled Airlines Total # of Deaths Chartered Flights 100 European Railroads Mtn Climbing 10 Chemical Manufacturing Nuclear Power Bungee Jumping 1 1 10 100 1K 10K 100K 1M 10M # Encounters / Death

Health Care as a Complex System Health care organizations can be viewed as complex systems (Institute of Medicine 2001; Plsek and Greenhalgh 2001; Sweeney and Griffiths 2002).

A C B X Sufficient Cause Cluster Effect

Simple Baking a cake is a simple problem. Simple problems lend themselves to a recipe approach. The process and results are generalizable; and while special skill at cooking is a plus, it is not essential for success.

K ~D M O A C B X

K ~D M O Y A C B J X G I

K ~D M O Y A C B J X G I P Z D X ~A

K ~D M O Y A C B J X G I P Z D X ~A A ~K

K ~D M O P Z G S R Y T N C V ~G A C B J X G I P Z D X ~A A ~K D U H T Q P L P Z D X ~A A ~K

Complicated Sending a rocket to the moon is an example of a complicated problem. Complicated problems are best dealt with using formulaic and expert-knowledge approaches. The overall problem can be mechanistically broken down into component parts (booster rocket, cabin environment, navigational equipment, etc.) and assigned to teams of experts who utilize the proven methodologies of their disciplines.

Complicated Rockets are similar in important ways, meaning that success with one rocket provides reasonable assurance of success with future rockets. When surprising events do occur, we can study these, build improvements into the system, and thus raise the probability of future success.

K ~D M O P Z ? S Y ? ? A ? B J X G ? ? P Z ? X A ~K

Complex In contrast to simple and complicated issues, an example of a complex issue is that of raising a child. Success in raising one child is no guarantee of success in raising another. Past experience, coupled with advice from experts, can serve as a starting point; but we know that simply applying the formula that worked before may not lead to success, and may even lead directly to failure because of the second child’s resentment at being treated this way.

Across all disciplines, at all levels, and throughout the world, health care is becoming more complex.

Newton's “clockwork universe,” in which big problems can be broken down into smaller ones, analyzed, and solved by rational deduction, has strongly influenced both the practice of medicine and the leadership of organizations.

But the machine metaphor lets us down badly when no part of the equation is constant, independent, or predictable.

Complex Adaptive Systems: Definition A complex adaptive system is a collection of individual agents with freedom to act in ways that are not always totally predictable, and whose actions are interconnected so that one agent's actions changes the context for other agents.

Complex Adaptive Systems: Examples Examples include the immune system, a colony of termites, the financial market, and just about any collection of humans (for example, a family, a committee, or a primary healthcare team).

Complex Adaptive Systems Fuzzy, rather than rigid, boundaries Agents' actions are based on internalized rules The agents and the system are adaptive Systems are embedded within other systems and co­evolve Tension and paradox are natural phenomena, not necessarily to be resolved Interaction leads to continually emerging, novel behavior Inherent non­linearity Inherent unpredictability Attractor behavior Inherent self organization through simple locally applied rules

Fuzzy, rather than rigid, boundaries In mechanical systems boundaries are fixed and well defined; for example, knowing what is and is not a part of a car is no problem. Complex systems typically have fuzzy boundaries. Membership can change, and agents can simultaneously be members of several systems.

Agents' actions are based on internalized rules In a complex adaptive system, agents respond to their environment by using internalized rule sets that drive action. In a biochemical system, the “rules” are a series of chemical reactions. At a human level, the rules can be expressed as instincts, constructs, and mental models. “Explore the patient's ideas, concerns, and expectations” is an example of an internalized rule that might drive a doctor's actions.

These internal rules need not be shared, explicit, or even logical when viewed by another agent. The mental models and rules within which independent agents operate are not fixed.

The agents and the system are adaptive Because the agents within it can change, a complex system can adapt its behavior over time. At a biochemical level, adaptive micro­organisms frequently develop antibiotic resistance. And adaptation within the system can be for better or for worse, depending on whose point of view is being considered.

Systems are embedded within other systems and co­evolve The evolution of one system influences and is influenced by that of other systems. Our efforts to improve the formal system of medical care can be aided or thwarted by these other more informal “shadow systems.” Since each agent and each system is nested within other systems, all evolving together and interacting, we cannot fully understand any of the agents or systems without reference to the others.

Tension and paradox are natural phenomena, not necessarily to be resolved The fact that complex systems interact with other complex systems leads to tension and paradox that can never be fully resolved. In complex social systems, the seemingly opposing forces of competition and cooperation often work together in positive ways fierce competition within an industry can improve the collective performance of all participants.

Whereas conventional reductionist scientific thinking assumes that we shall eventually figure it all out and resolve all the unresolved issues, complexity theory is comfortable with and even values such inherent tension between different parts of the system.

Interaction leads to continually emerging, novel behavior The behavior of a complex system emerges from the interaction among the agents. The observable outcomes are more than merely the sum of the parts-the properties of hydrogen and oxygen atoms cannot be simply combined to account for the noise or shimmer of a babbling brook. The productive interaction of individuals can lead to novel approaches to issues. The inability to account for surprise, creativity, and emergent phenomena is the major shortcoming of reductionist thinking.

Inherent non­linearity The behavior of a complex system is often non­linear. For example, in weather forecasting the fundamental laws governing gases contain non­linear terms that lead to what complexity scientists have called “sensitive dependence on initial conditions,” such that a small difference in the initial variables leads to huge differences in outcomes.

Inherent unpredictability Because the elements are changeable, the relationships non­linear, and the behavior emergent and sensitive to small changes, the detailed behavior of any complex system is fundamentally unpredictable over time. Ultimately, the only way to know exactly what a complex system will do is to observe it: it is not a question of better understanding of the agents, of better models, or of more analysis.

Inherent pattern Despite the lack of detailed predictability, it is often possible to make generally true and practically useful statements about the behavior of a complex system. There is often an overall pattern.

Attractor behavior Complexity science notes a specific type of pattern called an attractor. Attractor patterns provide comparatively simple understanding of what at first seems to be extremely complex behavior. For example, in psychotherapy, clients are more likely to accept a counselor's advice when it is framed in ways that enhance their core sense of autonomy, integrity, and ideals.

Attractor behavior These are underlying attractors within the complex and ever changing system of a person's detailed behavior. Relatively simple attractor patterns have been shown in share prices in a financial market, biological systems (such as beat to beat variation in heart rate), human behavior, and social systems

Inherent self organization through simple locally applied rules Order, innovation, and progress can emerge naturally from the interactions within a complex system; they do not need to be imposed centrally or from outside. For example, termite colonies construct the highest structures on the planet relative to the size of the builders. Yet there is no chief executive termite, no architect termite, and no blueprint. Each individual termite acts locally, seemingly following only a few simple shared rules of behavior, within a context of other termites also acting locally. The termite mound emerges from a process of self organization.

In everyday life many complex behaviors emerge from relatively simple rules in such things as driving in traffic or interacting in meetings. While no one directs our detailed actions in such situations, we all know how to behave adaptively and end up getting to where we want to go.

Stacey Diagram Far from Agreement Level of Agreement Decisions will have varying levels at which the entire system agree that a particular effect is desired. The Degree of Certainty Decisions are more certain when the cause and effect linkages are well known. Close to Agreement Close to certainty Far from certainty

Stacey Diagram Far from Agreement Close to Agreement Where even enough hard evidence is not enough Far from Agreement Close to Agreement Where there is not enough hard evidence Close to certainty Far from certainty

Stacey Diagram Simple: Rational decision-making. Straightforward Far from Agreement Simple: Straightforward decision-making. Rational planning & control Simple Close to Agreement Close to certainty Far from certainty

Stacey Diagram Chaos Chaos: No discernible patterns. Disintegration Far from Agreement Chaos Chaos: Disintegration patterns. No discernible & anarchy Simple Close to Agreement Close to certainty Far from certainty

Stacey Diagram Chaos Complicated Simple Far from Agreement Close to Close to certainty Far from certainty

Stacey Diagram Chaos Complicated Simple Complicated Far from Agreement Close to Agreement Close to certainty Far from certainty

Stacey Diagram Zone of Complexity Far from Agreement Close to Close to certainty Far from certainty

The zone of complexity Langton has termed the set of circumstances that call for adaptive behaviors “the edge of chaos.” This zone has insufficient agreement and certainty to make the choice of the next step obvious (as it is in simple linear systems), but not so much disagreement and uncertainty that the system is thrown into chaos. The development and application of clinical guidelines, the care of a patient with multiple clinical and social needs, and the coordination of educational and development initiatives throughout a practice or department are all issues that lie in the zone of complexity.

Our learnt instinct with such issues, based on reductionist thinking, is to troubleshoot and fix things—in essence to break down the ambiguity, resolve any paradox, achieve more certainty and agreement, and move into the simple system zone.

But complexity science suggests that it is often better to try multiple approaches and let direction arise by gradually shifting time and attention towards those things that seem to be working best. Schön's reflective practitioner, Kolb's experiential learning model, and the plan­do­check­act cycle of quality improvement are examples of activities that explore new possibilities through experimentation, autonomy, and working at the edge of knowledge and experience.

Not all problems lie in the zone of complexity. Where there is a high level of certainty about what is required and agreement among agents (for example, the actions of a surgical theatre team in a routine operation) it is appropriate for individuals to think in somewhat mechanistic terms and to fall into their pre-agreed role. In such situations the individuals relinquish some autonomy in order to accomplish a common and undisputed goal; the system displays less emergent behaviour but the job gets done efficiently. Few situations in modern health care, however, have such a high degree of certainty and agreement, and rigid protocols are often rightly abandoned.

Some principles to assist decision making in the complex zone Use intuition and muddle through—Doctors frequently make what would be the best but not definitively the “right” decision on the basis of experience, evidence, and knowledge of the patient's story Experiment—Try different management options with patients, using an empirical trial of treatment or a plan­do­check­act cycle Minimum specification—Offer patients general goals, suggestions, and examples but do not attempt to work everything out for them—your tidy solution is unlikely to be compatible with all aspects of their lifestyle and values Chunking—Instead of trying to sort out every problem, try solving one or two; other solutions may follow naturally once a new pattern has emerged Use metaphors—Communication can be difficult when issues are complex. Using metaphors can often create a shared understanding—for example, “you seem like a tree bowed over by the wind” or “what does that last hypo remind you of?” Provocative questions—Ask questions that might throw light on basic assumptions, especially when the patient is “stuck”—for example, “if you got better, might this cause some problems for you?”

Examples of CAS The immune system, A colony of insects, The stock market, Families A flock of birds The Internet The weather The economy An ecosystem Health care organizations.

Complexity An emerging science which analyzes organizations from multiple dimensions, including biological models, rather than simple machinistic perspectives. Organizations act as complex adaptive systems and are therefore less predictable.

Complex Adaptive Systems (CAS) A collection of individuals, each of which is autonomous and free to act in unpredictable ways, and whose actions are interconnected such that one's actions change the context of the other individuals.

CAS Attributes Each element can change themselves; i.e., they can adapt. Systems are embedded within systems & their interdependency matters. System is nonlinear – small events may trigger huge effects. large events can have negligible effects. Complex behavior can emerge from a few simple rules.

CAS Attributes Not predictable in detail; forecasting is an inexact, yet boundable, art. Future is not just unknown, but unknowable. System co-evolves through constant tension and balance. Emergence of novelty and creativity is a natural state. Order can emerge without central control. CAS are history-dependent.

Adaptive Components Individual components can change/adapt. change will occur in response to alterations in the environment. behavioral adaptation is rapid-cycle learning from local experience. dependent upon the presence of diversity. necessary to maintain existence in an unpredictable environment. allows the entire system to adapt.

Complex Imbedded Systems Systems are imbedded within systems. complex systems are parts of larger complex systems, and are made up of smaller complex systems. no component, including leaders, can act as though they are outside the system it is not the specific individuals that are the most critical, but the relationships between individuals

Non-linearity System is nonlinear. effects of small or large changes are not necessarily related to size or predictable large interventions may not achieve desired outcomes and may yield nothing or possibly the opposite outcome. there is a sensitive dependence to initial conditions - the "butterfly effect."

Non-linear Systems Inverse power law: the frequency of occurrence of a phenomenon is in inverse relation to its size. small waves are common, large waves are less frequent Common generative mechanisms both small and large waves are causes by the same generative mechanisms

Non-linear Systems Self-organized criticality - there is interdependence of agents in the system which creates tension over time Per Bak's classic sand pile experiment As grains of sand self-organize into a pile, a single new grain of sand can cause: nothing at all a small shift a large landslide

Non-linear Systems Confluence and concatenation BIG events, such as severe errors, often occur as a result of triggering events. these triggering events are the generative mechanisms leading to small events. BIG events are impossible to predict analyzing the cause of small events is the first step in preventing big events.

Non-linear Systems Tension and criticality - When there is sufficient tension and criticality in the system, a seemingly trivial event can serve as a powder keg

Simple Rules Complex behavior emerge from a few simple rules. complex plans are not needed, and may be detrimental simple rules are frequently unspoken, yet self-perpetuating within the system simple rules can be clarified by searching for subtle patterns and asking "Why?" five times.

Limited Predictability Not predictable in detail; forecasting is an inexact, yet boundable, art. Future is unknown and unknowable. need to analyze - trying to identify recurring patterns, the underlying simple rules and attractors forecasting tries to foretell how these patterns will yield outcomes into perceived future environment/conditions

Tension & Paradox System co-evolves through constant tension and paradox. in CAS, tension and paradox are natural. both sides of apparent contradictions are true and necessary. we may not need to resolve all the dilemmas of organizations. resolution of these dilemmas may be detrimental to long-term survival.

Emergence Novelty and creativity naturally emerge. CAS exist and thrive at the edge of chaos in an environment of uncertainty and rapid change, novelty and creativity are necessary for survival CAS have the adaptability which allow for the emergence of novelty and creativity. Dependent upon the presence of diversity Standardization smothers creativity.

Emergence Order can emerge without central control. CAS achieve order by reaching equilibrium, not stability. attempts to impose central control can have undesirable consequences. equilibrium is determined by the simple rules & attractors and the environment. changing the rules, attractors or environment may yield a new equilibrium.

Stacey Diagram Far from Agreement Close to Agreement Good enough planning–minimum specification, simple rules Multiple actions Experiment and tune system plan-do-check-act Scan for patterns Political decision making Negotiation and consensus building Compromise Listen to the shadow system Use intuition, muddle through Metaphors Use complex thoughts for complex systems Judgmental decision making Mission and vision based decision making Close to Agreement Plan and control Regulate Close to certainty Far from certainty

What is the fate of a complex system of decisions ? Far from Agreement Finding more cause components from already known cause clusters and discovering unknown cause clusters * * Close to Agreement Close to certainty Far from certainty

What is the fate of a complex system of decisions ? Far from Agreement * * Consensus building around shared Values and beliefs Close to Agreement Close to certainty Far from certainty

What is the fate of a complex system of decisions ? Far from Agreement * * Falling into chaos : Organizational Collapse, Revolutions, Death Close to Agreement Close to certainty Far from certainty

Stacey Diagram Many organizations are existing in all areas of the matrix at different times. Traditional management methods are effective in the Simple area. Management methods needs to be altered in the Complicated areas - negotiation / compromise mission & values based

Stacey Diagram If you try to use traditional management methods (plan & control) in the Zone of Complexity, you usually get unintended and unpredictable consequences. Complex Adaptive Systems can exist and thrive in the Zone of Complexity. PDCA cycle is an example of management in the complex zone allowing tuning, experimenting, and good-enough planning.

System Design The distinction between these systems is obvious, but frequently not taken into account when system is designed. When the human components respond in an unpredictable manner, they are labeled as being unreasonable or “resistant to change.” The designer then specifies behavior in greater detail via rules, guidelines, etc.

Thank You! Any Questions!