Understanding Complexity

The Challenge of Complexity

In recent years it has become clear to people responsible for improving our health care system that the challenges they face are exceedingly complex. In fact, some have taken to calling these seemingly intractable issues “wicked problems” insofar as they are unique, difficult to delineate and impervious to established quality improvement methods. Welcome to the world of complex adaptive systems!

What is a complex adaptive system?
Complex adaptive systems (CAS) exist throughout nature and occur in many areas of human enterprise. They are made up of sub-systems and individual agents which are all in dynamic relationship, each influencing and being simultaneously influenced by the others. A key feature of such systems is their unpredictability. Global weather and economics are both complex systems.

Complex adaptive systems are self-organizing, constantly evolving and adapting to changes in their environment. The resistance of such systems to linear understanding and control has been compared to child-rearing. Rules can be effective but not always and not for all children. And strategies that work for raising one child do not ensure success if simply repeated with another.

Some principles of complex adaptive systems
Order is emergent and self-organizing
In a healthy CAS, control is distributed rather than centralized. Outcomes are unpredictable and emerge from interactions among individuals within the system through an evolutionary process of self-organization.

Receptiveness to change
Systems can be understood as operating within a spectrum from stasis to chaos. A system that is too tightly controlled cannot respond effectively to changes in its environment. Such systems are doomed to fail. A healthy CAS lies closer to the other end of the spectrum, and is open to change and new creative possibilities. The internet, with just high-level governance to ensure data security and operational viability, is an example of such a system.

Behaviour is governed by simple rules
Complex systems are not necessarily complicated and the rules governing system behaviour can be quite simple. Achieving positive change within a CAS can be helped if these simple rules or ‘leverage points’ can be identified. Flocks of birds, for example, are governed by three simple rules: follow the leader, maintain speed and avoid collisions.

Change occurs in response to new, meaningful information
A CAS changes when people within the system acquire new information that alters their understanding. To achieve lasting change it is not enough to tell people what they must do differently, they must first understand and accept the reasons for why change is needed, and then be allowed some latitude for implementing the change. The success of smoking cessation efforts in Canada is due more to educating young people about the health hazards of smoking than to limiting legal access to cigarettes.

Unpredictable outcomes
Complex adaptive systems are resistant to conventional change management methods. Social processes in a CAS are not linear and specific interventions rarely have direct, predictable effects. Unanticipated events can affect outcomes, so plans for a CAS must be flexible to allow adaptation to changing circumstances and unintended consequences. Democratic political systems are an example. Leaders are elected to implement policies, but must contend with the uncertainty of how things will turn out and must be ready to change course if things go badly.

Achieving positive change within a complex system requires:

  • Engaging the right people in dialogue at the right time to promote a shared vision and values and to foster collaborative relationships
  • Leadership distributed across the organization; encourage staff at all levels to lead and take risks in support of organizational goals
  • Effective, ongoing communication and rapid feedback loops
  • Building on a firm foundation of good management practices
  • Tolerance of difference and conflict
  • Encouragement of emergent solutions, risk taking and innovation
  • Knowledge of the ‘simple rules’ affecting system performance
  • Willingness to take measured action in the absence of perfect evidence
  • A supportive, learning environment
  • Accepting that plans must be flexible and course corrections will be required