“Scaling up” and “change” are two topics whose relationship is preoccupying me these days. I was delighted to discover an article that presents important insights by tying the topics to “complex adaptive systems”.
From what I’ve seen, “scaling up” strategies can be characterized in two extremes (with some mixing, of course):
My recent reading clarifies that the “roll-out” form is good in “simple” and “complicated” situations; the action learning form is appropriate for complex situations as distinguished by David Snowden. The roll-out top-down format requires the stable, predictable situations with few unknowns and linear power structures that are associated with simple and complicated situations; the action learning format is for situations where there are many unknowns, great variation in contexts and rapid change as is associated with complexity.
With the burdensome but clear name, Understanding pathways for scaling up health services through the lens of complex adaptive systemsis a great article that synthesizes others’ work, rather than drawing from first-hand research. In fact, it makes the point that has not found complex adaptive systems (CAS) thinking actually applied to health issues. The authors set the scene by describing problems with applying simple/complicated perspectives to CASs:
“People’s understandings of systems that are actually CAS are often over-simplified or erroneous, which creates problems for decision-makers who cannot control such systems through conventional means, often while being vulnerable to sudden changes in public opinion. It is not unusual for systems to show little response to many attempts to control them, or to change suddenly when a tipping point is reached. For example, many high-cost health investment projects have had little impact on people’s behavior or health status, in contrast to sudden changes that can occur in public opinion about smoking bans or in the demand for contraception.”
The real value of the article is that it makes two contributions to facilitate application.
Key Complex Adaptive System Phenomena
One of the article’s key contributions is its summary of five models of phenomena that are behind CASs. These demonstrate the problems with a roll-out approach taken in such circumstances.
1. Path dependence: history, culture and other contextual features can produce different outcomes even when starting from the same point.
2. Feedback: You can start from the same point, but during the process there are responses that affect the precursors of success to produce differing trajectories
3. Scale-free networks: In many issue arenas, some networks have hubs connecting many sub-groups of people/organizations, other hubs connect only a few (hence “scale-free”); large nodes can facilitate both connection and collapse of a system (think of the financial system).
4. Emergent behavior: new connections create new possibilities that are greater than the sum of their parts.
5. Phase transition: this is the “tipping point” factor: small changes and connections can suddenly “gell” to become dominant.
While people commonly focus on one or two of these phenomena — feedback and phase transition are, in my experience, the most popular — they all require consideration. Collectively these five phenomena produce a confusing array of action that is associated with complex adaptive systems and emergence; understanding them individually helps immensely in understanding complexity dynamics.
Application of Insights
The second contribution of the article is that it begins connecting these definable patterns (complexity being distinguished from chaos by such patterns) to both an action logic and methodologies with tools. The action logic is a play on the plan-act-learn one of traditional learning cycles and development strategies of simple/complicated issues.
Planning in CASs is predicated on the understanding that new, unforeseen opportunities and events will emerge that should be treated as an asset rather than a disruption in a planned set of outcomes mapped in advance. The process of learning, envisioning, clarifying and experimenting will “emerge” the outcomes, in contrast to engineering outcomes as in a roll-out processes.
Planning tools the article references include stakeholder analysis, network mapping, agent-based modeling and scenario planning. These “…deepen the understanding of path-dependent actions and consequences; …test their abilities to anticipate and adapt to changing conditions; (identify) feedback loops and emergent behavior; (and) identify neighborhood or place effects…”
The authors comment with respect to the next step of the action cycle:
“During the implementation and monitoring of scaling up ,,. an understanding of CAS would emphasize the importance of adaptation, learning and flexibility to emerging issues rather than the rigid following of initial plans.”
Tools identified for this stage include small-area statistical variation analyses, facilitating and analyzing dialogue, and time series analysis to identify patterns.
The last step, evaluation, is handled in an unsatisfactory way in my opinion. The authors recognize that unintended consequences must be addressed and point to problematic approaches such as creating unsustainable islands of excellence (is anything really an ‘island’?). However, they make no reference to the developmental evaluation of Michael Patton and others’ works that emphasize that in CAS situations “evaluation” and “learning” must be integrated – initiatives cannot wait for evaluations after-the-fact, given CAS interventions are typically lengthy and new learning must be integrated into actions on a very short cycle. Of course this does not deny the value of periodic reviews and assessments, but traditional simple-complicated evaluation methodologies in this situation are highly inappropriate in comparison to a more broad-looking assessment.
The authors conclude with particular reflection on the health field, but their remarks can be broadly applied: “The old assumptions have led to disappointed expectations about how to scale up health services, and offer little insight on how to scale up effective interventions in the future. The alternative perspectives offered by CAS may better reflect the complex and changing nature of health systems, and create new opportunities for understanding and scaling up health services. “