Content uploaded by Sue Lukersmith
Author content
All content in this area was uploaded by Sue Lukersmith on Sep 17, 2020
Content may be subject to copyright.
Applying complex adaptive
system thinking to Australian
health care: Expert commentary
______________________________________________________________
Report for an Australian Commonwealth Government agency
Authors
Carmen Huckel Schneider (Deputy Director, Menzies Centre for Health Policy, University of Sydney)
Joachim Sturmberg (Conjoint Associate Professor, University of Newcastle)
James Gillespie (Associate Professor, Menzies Centre for Health Policy, University of Sydney)
Andrew Wilson (Director, Menzies Centre for Health Policy, University of Sydney)
Sue Lukersmith (Senior Research Fellow, CMHR Australian National University (ANU)
Luis Salvador-Carulla (Head, Centre for Mental Health Research (CMHR) ANU)
30 June 1016
Menzies Centre for Health
Policy
Faculty of Health Sciences
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 2
Table of contents
Applying complex adaptive system thinking to Australian health care: Expert commentary .......................... 1
Table of contents .................................................................................................................................................. 2
List of tables .......................................................................................................................................................... 4
List of figures ........................................................................................................................................................ 4
Acronyms and abbreviations .................................................................................................................................. 4
Executive Summary ............................................................................................................................................. 6
Conceptual overview ....................................................................................................................................... 6
Question 1: How does the concept of a complex adaptive system apply to the Australian healthcare
system? .............................................................................................................................................................. 7
Question 2: In considering the Australian healthcare system as a CAS, what agents exist at a macro,
meso, micro and nano level and what are the relationships between them? ............................................... 8
Question 3: In considering the Australian healthcare system as a complex adaptive system, where do
consumers fit and what are their relationships with other agents within the system? ............................... 8
Question 4: How can an understanding of the Australian healthcare system as a complex adaptive
system accommodate or support adoption of person-centred care? ............................................................. 9
Introduction and methods ................................................................................................................................. 10
Literature capture ........................................................................................................................................... 11
Structure of the paper .................................................................................................................................... 11
Conceptual overview ......................................................................................................................................... 14
Why take a systems perspective to health care? .......................................................................................... 14
Systems thinking approaches ........................................................................................................................ 14
Complex adaptive systems theory ................................................................................................................. 17
Question 1: How does the concept of a complex adaptive system apply to the Australian healthcare
system? ................................................................................................................................................................ 21
Systems thinking approaches ........................................................................................................................ 21
Complex adaptive systems theory ................................................................................................................. 25
Question 2: In considering the Australian healthcare system as a CAS, what agents exist at a macro, meso,
micro and nano level and what are the relationships between them? ........................................................... 29
Systems thinking approaches ........................................................................................................................ 29
Complex adaptive systems theory ................................................................................................................. 31
Question 3: In considering the Australian healthcare system as a complex adaptive system, where do
consumers fit and what are their relationships with other agents within the system? ................................. 33
Systems thinking approaches ........................................................................................................................ 33
Complex adaptive systems theory ................................................................................................................. 34
Question 4: How can an understanding of the Australian healthcare system as a complex adaptive system
accommodate or support adoption of person-centred care? ........................................................................... 37
Systems thinking approaches ........................................................................................................................ 37
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 3
Complex adaptive systems theory ................................................................................................................. 39
Appendix I: Complexity sciences conceptual framework .................................................................................. 40
The philosophy of Complex Adaptive Systems – Paul Cilliers ....................................................................... 40
Tackling the most difficult questions? – David Krakauer ................................................................................ 42
A general description of systems (nonlinear systems) – Russ Ackoff ............................................................. 43
Nominal Definition – Kevin Dooley ................................................................................................................. 43
Structure and dynamics of CAS are interdependent – Fritjof Capra .............................................................. 44
Appendix II – Cynefin Framework ...................................................................................................................... 45
Appendix III – Case study of a complex adaptive (health) system – the NUKA system .................................. 47
Background ........................................................................................................................................................ 47
Vision .................................................................................................................................................................. 47
Key Points .......................................................................................................................................................... 47
Appendix IV: Roles and responsibilities of agents at system levels ................................................................... 49
References ............................................................................................................................................................. 52
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 4
List of tables
Table 1: Comparing the terminologies and meanings of different types of systems ......................................... 18
Table 2: Simple, complicated, complex and complex adaptive subsystems in the Australian health system .. 28
Table 3: CAS features in the NUKA system ......................................................................................................... 48
Table 4: Agents and their interests in the health system .................................................................................... 50
List of figures
Figure 1: Main components of a complex health care system and items relevant for their analysis and
development of decision support systems (DSS) .................................................................................................. 15
Figure 2: Application of vortex model to the Australian health system ............................................................ 26
Figure 3: Distribution of health service use amongst ‘users’ ............................................................................... 35
Figure 4: Application of Cynefin Framework health and disease ....................................................................... 46
Figure 5: Application of Cynefin Framework to culture of safety ...................................................................... 46
Acronyms and abbreviations
ACGA Australian Commonwealth government agency
AHRQ Agency for Healthcare Research and Quality
CAS Complex Adaptive Systems
CDC Centers for Disease Control and Prevention
DSS Decision support systems
GP General practitioner
IHI-QI Institute for Health Improvement Quality Improvement
LHD Local Health District
NHS National Health Service
NIH National Institutes of Health
PHN Primary Healthcare Network
PPCHC Person- and people-centred healthcare
RCA Root cause analysis
WHO World Health Organization
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 5
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 6
Executive Summary
This expert commentary has been commissioned by the Australian Commonwealth government agency
(ACGA) and the Sax Institute to contribute to a series of consultation and discussion papers on its future
vision for the Australian Healthcare system.
The Report is structured to respond to specific questions posed by the Commission in the commissioning
brief. In the report, each question is addressed from the perspective of two broad approaches found in the
literature; 1) ‘systems thinking approaches’ in applied analyses of healthcare, and 2) ‘complex adaptive
systems theory’ as a global theory to understand health care.
This report has been specifically commissioned as an expert commentary rather than a systematic review of
the literature. We therefore began with published literature known to the authors of this review. This was
supplemented a Medline keyword search, grey literature and iterative discussions and commentary with
successive drafts a) within the core (author) group and b) in consultation with an external expert panel.
Conceptual overview
Contemporary health policy faces several wicked problems. Individual, isolated changes have foundered in
the face of this complexity. A broader systems’ perspective is required to understand how changes intersect,
affect one another or create new conflicts and synergies.
Complex systems are composed of many interacting components (agents) that are characterised by different
levels of variability, uncertainty and levels of organisation. Complex adaptive systems are a particular type
of complex organisation characterised by feedback – learning and self-adaptation leading to emergence of
new properties. That is, the agents of a complex adaptive system are able to learn from experience and adapt
to changes in the environment, and new agents and connections emerge.
Along a continuum, different types of systems can be classified as simple, complicated, complex (dynamic)
and complex adaptive systems.
• In
simple systems
, elements of the system interact in one-to-one relationships producing predicable
outcomes.
•
Complicated systems
have sophisticated configurations but highly predictable behaviours (e.g. a car or a
plane). Interactions between the elements in the systems are also linear and predictable.
•
Complex systems have two key characteristics, they self-organise without external control and exhibit
feedback resulting in newly created, i.e. emergent (at times unforeseen), behaviours.
•
Complex systems also tend to be open, loosely bounded, and influenced by its environment. Complex
systems are simultaneously a subsystem of a larger system and comprised of a number of subsystems.
•
Complex adaptive systems (CAS) are a special case of complex (dynamic) systems as they have elements
(agents) that can learn and adapt their behaviours to changing environments.
We use the term
systems thinking approaches
to refer to a broad set of perspectives, methods and
approaches developed in diverse areas of science, organisational management and engineering which are
increasingly being applied in health policy and planning.
Health interventions are one of the priority areas for systems thinking analysis.
Systems thinking can inform
health policy and program assessment and planning in that it elucidates how an intervention:
• couples and embeds within context and manifests itself in thinking and practice;
• changes relationships;
• displaces existing activities (which may account in part for intervention effect); and
• redistributes and transforms resources (material, informational, social, cultural).
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 7
Complex adaptive systems theory
seeks to make sense of how complex systems, such as social systems and
health systems operate and function. It seeks to define how, and why, an order exists within these systems,
despite the (complex) nature of relationships and the emergent (and therefore not precisely predictable)
nature of outcomes.
A healthcare system is a social system; described in the complex adaptive systems theory literature as
emerging based on common statements of purpose, goals and values.
• Values: refers to the ideals and customs of a system toward which the people/agents have an effective
regard.
• Having agreed purpose, goals and values defines the
driver
of the system; together they give rise to so
called “simple rules” that coherently direct the interactions within a CAS.
• “Simple rules” provide the necessary “safe space/freedom” to adapt agents’ behaviour under changing
conditions.
• Complex adaptive systems cannot be controlled, the direction of their development can be guided and
requires
leadership
that reinforces the organisation’s purpose, goals and values.
There is little doubt that health systems are complex, however, there is widespread disagreement as to
whether they act as complex adaptive systems in accordance with complex adaptive systems theory. This is
because health systems appear to struggle in response to changing demands (such as a change in the
morbidity patterns) and the defining purpose, goals, values, and central drivers remain elusive.
Question 1: How does the concept of a complex adaptive system apply to the Australian healthcare system?
Systems thinking approaches
Our knowledge of the characteristics and dynamics of overall health care systems is still in its early stages.
However, there are plentiful examples of tools, methods and applied analysis from various disciplines that
can be utilised for describing and understanding the components of health care systems.
System engineering
focuses on how to design, operate and measure complex systems over their life cycles, and how to analyse
and improve its efficiency, productivity, quality, safety.
Business analytics
has developed techniques that
incorporate modelling, knowledge management, expert knowledge, artificial intelligence and visualisation.
High risk industries
and sectors such as commercial aviation and extraction use systems thinking to
conceptualise adverse events. Decision support systems (DSS) and toolkits are used in
defence, business,
policy, education and healthcare
. DSS, including simulation models and agent-based modelling, play an
essential role in the development of knowledge-to-action strategies. Finally,
geographical information
systems
can be used to represent the distribution of agents, their connections and spatial relationships.
Few examples of the analytical techniques above use the full complex adaptive systems framework,
however, each of the approaches summarized above – and many more – are making valuable use of systems
thinking to develop analytical and policy tools to understand health systems.
Complex adaptive systems theory
Based on the principles that define organisational
complex adaptive systems in CAS theory
the Australian
health system has to be considered to be a CAS. It has:
• agents – (eg. health ministers, health financing organisations, hospitals, the various health
professionals, and individuals) and
• these agents interact and through these interactions, learn – e.g. in consultation, within a hospital,
between central bureaucracies and local health service units.
However, as the purpose, goals, values and simple rules of the Australian health system are not clearly
evident, the system as a whole has no identifiable unifying driver.
The lack of a unifying driver makes the
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 8
Australian health system look more like a “conglomeration of discrete units”, colloquially expressed as “
a
fragmented health system
”. Therefore, according to complex adaptive systems theory:
• Understanding the health system as an integrated CAS requires reaching agreement of an overarching
driver.
• Most adapted solutions for the problem of a subsystem will emerge from helping all agents of that
system to align their drivers with those of the whole system.
Question 2: In considering the Australian healthcare system as a CAS, what agents exist at a macro, meso,
micro and nano level and what are the relationships between them?
Systems thinking approaches
• Systems approaches to health care assessment and planning usually define agents, but do not use the
macro-meso-micro-nano schematic. In part, this is a response to the blurred lines of connectivity
and location of action in a complex system.
• Schemes using these levels can be complemented by the Donabedian quality of healthcare model –
which uses the categories of ‘structure’, ‘process’ and ‘outcomes’ – and extended by adding
geographic levels (country, local area, individual).
• (Social) network analysis and systems modelling techniques (multi-scale, dynamic, multi-level) are
increasingly being used to analyse connections between agents in a complex system.
Complex adaptive systems theory
This layered conceptualisation of the health system is consistent with the complexity notion of nested
systems found in CAS theory. There are obvious constraints on agents at any system level in taking a ‘whole
of systems’ perspective. This is because:
• Values, goals and drivers of different sub-systems may not be the same – or may even conflict
• It is difficult to determine the potential influence that action in one part of the system may have on
other parts of the system.
Strong goal-driven links between and across organisational levels would increase the robustness and
adaptability of the system.
Question 3: In considering the Australian healthcare system as a complex adaptive system, where do
consumers fit and what are their relationships with other agents within the system?
Systems thinking approaches
• Users (or consumers) are present in the system in many different roles. In such situations they are
frequently negotiating competing interests.
• Users’ response to interactions with other agents and events within the system depends on the
combined weight of those interactions, but also broader influences such as their personal social or
financial situation, social, political or religious settings and particularly their prior experience of
care.
• There is widespread agreement that users should be engaged in health care decision making. Key
approaches include ‘citizen juries’, co-design and co-production, collaborative care, self-support and
peer support.
Complex adaptive systems theory
• Complex adaptive system theory recognises individual-level agents as having similar properties:
they are autonomous, they interact with other agents in the system along multiple pathways, they
learn and they adapt their behaviour.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 9
• Full engagement of users is essential to achieve a user/patient-centred health system.
• Taking a whole of system perspective, users always play a key role in determining the purpose and
goals of a system.
• Users can play such a key role during phases of re-build, re-design or during slower pace change.
Question 4: How can an understanding of the Australian healthcare system as a complex adaptive system
accommodate or support adoption of person-centred care?
Systems thinking approaches
• The enabling and delivery of PPCHC, with its focus on customising care to the needs of
individuals, inevitably adds even greater complexity to the system.
• There is a great potential for widespread adoption of mapping and modelling techniques in the
Australian health system to aid health planning and policy to move the system closer to delivering
PPCHC.
• However, there is a need for caution. Even the most comprehensive models require some level of
aggregation of preferences, actions and consequences. Decisions based on models can be a step
closer to PPCHC, but are still not adapted entirely to the individual. Rather, models assist decision-
makers to become aware of how a system works and apply this knowledge in practice.
Complex adaptive systems theory
• To move the Australian health system towards a complex adaptive system that is coherent and goal-
delivering, a focal point (driver) is required to guide activities to achieve integration within and
across organisational levels of care.
• Taking a CAS theory approach, the main challenge from the whole of system perspective relates to
how to transform and solidify the values that sustain the system.
• System change requires leadership in finding the right driver for the health system. This could take
the form of a system-wide conversation about general expectations and approaches to healthcare
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 10
Introduction and methods
This expert commentary has been commissioned by the Australian Commonwealth government agency
(ACGA) and the Sax Institute to contribute to a series of consultation and discussion papers on its future
vision for the Australian Healthcare system.
The commissioning brief outlines the vision of ACGA to support the creation of a person-centred healthcare
system. In the portfolio of activities in moving towards this vision, the Commission proposes to adopt a
systems approach. It is envisaged that this perspective will offer a practical insight into the interplay of roles,
responsibilities and opportunities that exist within the system – providing a foundation for navigating the
enablers and constraints to person-centred care.
The commentary team has been engaged to offer independent expertise and experience to prepare this plain
English account of complex adaptive systems thinking. This paper has thus been prepared with the
following as its guiding purpose:
• To offer a clear, concise overview of the language and concepts found in Complex Adaptive Systems
(CAS) literature
• To demonstrate how these concepts perform when related to the Australian healthcare system – in
relation to 4 key questions posed in the commissioning brief.
• To demonstrate the practical benefits of complex adaptive systems thinking in a range of scenarios
when moving towards change.
During the preparation of this expert review two positions were found in the literature.
The first looks at the general problem of complexity, and the (often eclectic) theoretical and practical
approaches that start from the recognition that systems thinking is the starting point for understanding
contemporary health care. These essentially applied approaches to the use of ‘systems thinking’ and
‘complexity’ use various techniques to map relationships, identify feedback loops, find gaps in knowledge or
explain particular successes and failures within a system. The focus is on how core elements
within
a
complex system interact and thus the implications for policy drawn from this body of literature focus on
solving problems and progressively moving towards more people and person centred care
The second approach draws on ‘complexity science’, a more holistic theoretical approach with its origins in
various disciplines such as physics, biology and environmental science. We outline the key elements of
complex adaptive system theory and use metaphors to demonstrate how complexity science can be applied
as a global theory to understand how health care, as a complex (adaptive) system, can be defined and how it
functions. This approach focuses on the structure, drivers, and the overarching direction of a system
as a
whole
. Policy implications drawn from this body of literature support a complete shift in health
prioritisation, planning and decision making in order to steer the system in the direction of people and
person centred care.
Each section of the report is dedicated to a specific question posed in the commission brief. Under each
question, a subheading “systems thinking approaches” and “complex adaptive systems theory” is used to
reflect this division within the literature. It is important to note that this distinction is one of degree –
proponents of complex adaptive systems theory would accept most if not all the analytical methods found in
complexity approaches. However, they would situate these in a more thoroughgoing complexity science.
All the experts in the working group agreed on the need to move towards a systems-based approach in
health care policy and planning, as systems thinking can help us understand problems arising from either
the structure and/or interaction of agents. This approach has the capability to identify potential solutions to
complex problems that may otherwise remain hidden.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 11
The commentary team agrees with the definition of health care systems as complex systems as described by
the
Institute of Medicine
and reiterates that a deeper understanding of health care systems and their
behaviour is required:
Health care is complex because of the great number of interconnections within and
among small care systems
."."."
Health care systems are adaptive because unlike
mechanical systems they are composed of individuals - patients and clinicians who
have the capacity to learn and change as a result of experience. Their actions in
delivering health care are not always predictable, and tend to change both their local
and larger environment (1)
Specifically, in relation to the goals of the ACGA to navigate enablers and barriers to system change:
• Systems thinking refers to an emergent field of understanding that focuses on the relationships
between the parts of a system.
• Health care systems are complex systems and analysing their structure and function is useful for
policy planning.
• Systems thinking can help us understand problems and identify potential solutions to these
problems that are not obvious – because they lie in the dynamics of the system, rather than in
specific parts of the system.
• Adopting complex systems approaches can bring to light how events, or change, in one part of the
system affect other parts of the system.
Literature capture
This paper has been specifically commissioned as an expert commentary rather than a systematic review of
the literature. For this topic, we consider this appropriate. Systems science in healthcare is still emerging
and we draw upon sources from highly diverse fields which would be unlikely to be captured in any
meaningful way in a strict systematic search. We therefore began with published literature known to the
authors of this review. This was supplemented by:
•
Medline
search using the keywords “(complex adaptive system)*” and “(healthcare system)*”, period
to end of 2015
• Grey literature from reputable national and international agencies (e.g. WHO, King’s Fund, NIH)
• Iterative discussions and commentary with successive drafts
o Core group
o Consultation with external Expert Panel
• Review and revision of paper in light of expert input.
These papers were then drawn on to inform the commentary, distinguished from a traditional systematic
review in that:
1) database searches served to complement review team knowledge of seminal papers;
2) papers found were not subject to a systematic quality appraisal process - rather;
3) expert knowledge was sought to appraise the evidence in light of the questions posed by the
commissioning agency.
Structure of the paper
Each section of the paper responds to a specific question posed in the commissioning brief and sets out
approaches to the problem identified with complexity theory. It then applies the perspectives of complex
adaptive systems theory to the problem. It is important to note that the two approaches are not necessarily
incompatible. CAS theory (or science- a terminology adopted by expert commentators who adopted this
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 12
approach) uses many of the methods developed by analysts of complexity, but in in the context of an
encompassing theory, claiming application in domains of physical, biological as well as social spheres.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 13
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 14
Conceptual overview
Why take a systems perspective to health care?
“Waste in the health care system” refers to unnecessary health expenditure and unexplained clinical
practice variation, system inefficiencies, and costs and consequences of medical errors and other problems in
health quality and safety (2). Australia shows good health indicators at the macro level, but waste in the
health care system is a serious concern for the sustainability of our health care system. (3, 4) The increasing
waste in health care has been compounded by the increasing complexity in the patients treated in the
system, the interventions provided to these patients and the organisations providing care. Health care policy
and planning has struggled to keep up. Paradoxically, cutting edge technical and scientific improvements
coexist with outdated logistics and lack of integration. “We bank, shop, book taxis and airplanes... on our
smartphones. But for most people, when it comes to seeing a doctor or getting a blood test, we dial the clock
back 20 years” (5).
Individual, isolated changes have foundered in the face of this complexity. A broader systems’ perspective is
required to understand how changes intersect, affect one another or create new conflicts and synergies.
Systems thinking approaches
Systems thinking is an area in health research that analyses the parts of a defined system and their connections
in relation to the whole. Related approaches and terms found in the literature include ‘systems approaches’,
‘systems science’, ‘complexity analysis’ and ‘design thinking’. In this paper we use the term
systems thinking
approaches
to mean the broader set of perspectives, methods and approaches developed in diverse areas of
science, organisational management and engineering which are increasingly being applied in health policy
and planning (6).
Complex systems are characterised by different levels of variability, uncertainty and levels of organisation
that range from simple systems to highly complex ones.
A complex organisation cannot be replicated and
the capacity to predict its evolution and to generalise its outcomes is limited due to the properties of
complex dynamic systems, mainly non-linearity, time-dependency and context-dependency. Complex
adaptive systems are a particular type of complex organisation characterised by feedback – learning and self-
adaptation leading to emergence of new properties. That is, the agents of a complex adaptive system are able
to learn from experience and adapt to changes in the environment, and new agents and connections emerge.
Systems thinking approaches recognise these characteristics and shape research questions, methods and
analysis accordingly.
The systems thinking approach has long been applied in areas such as business, engineering, defence, and
public policy. Health care is a late adopter of this approach but the interest in health systems research and
planning is surging. Related areas such as global health, collaborative care, or health systems engineering are
attracting growing attention, and systems thinking is particularly relevant in health sectors characterised by
high complexity, such as chronic care, primary care and mental health care. Systems thinking is key to
designing and assessing integrated care and this approach has been adopted by WHO for the new strategy
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 15
“People-centred Integrated Care for All”. It has also been adopted by several health planning agencies such as
NHS Scotland and the Department of Health in the Basque country (Spain).
What are the building-blocks of information needed for systems thinking?
According to systems thinking approach, health systems should be described in terms of their boundaries,
drivers, agents, and connections.
•
Boundaries and context.
Systems thinking requires information on the hierarchy and the context of the
system that is analysed, particularly in relation to the boundaries of the system, the subsystems which
can be identified within the system, and the different levels of organisation where agents operate. As
complex dynamic systems are context-dependent and time-dependent, context analysis is a necessary
component of any analysis. Similarly, the understanding of the history of the development and
evolution of a system is also relevant to understand its current status.
•
Agents.
Agents in a health care system include patients, citizens, health professionals, research and
clinical units, hospitals, service providers, governments (government departments), advocacy
organisations, insurers, health related businesses, bureaucrats, peer groups and individual users.
•
Connections.
Connections describe the relationships between every agent in the system as the different
organisational levels at which they operate and that together characterise the level of complexity of the
system. An interaction is the occurrence of an action in the connection between two agents that
produces an effect. Health interventions are a particular type of interaction between patients and
professionals, or between the population and public health agencies.
•
Drivers.
Complex adaptive systems theory claims that identifying the ‘core driver’ of a system is critical
to understand the system and to guide change. However, there is a lack of critical information on the
drivers of the health care system, even at local or sub-system levels. Applying systems thinking to
healthcare systems is enabled where there are clear drivers, however analysis may need to proceed
knowing that drivers may vary, or even conflict, across parts of the system.
In order to apply systems thinking to the design and monitoring of health care it is important to gather
information on every one of the major components of the system: Boundaries, Drivers, Agents, and
Connections. (See Figure 1) (7);
Figure 1: Main components of a complex health care system and items relevant for their analysis and development of
decision support systems (DSS)
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 16
Application of systems thinking to analysis
Determining the effects and effectiveness of health interventions is one of the priority areas for systems
thinking, with significant implications for health policy and program evaluation, systematic review of
evidence, analysis of quality, safety and efficiency and for implementation. “Interventions are more effective
and sustainable when these complex and multilevel aspects are understood and considered” (8).
From a systems thinking approach health interventions can be defined as a particular type of formal
interaction between agents (most commonly users, health professionals, organisations and institutions); or as
occurring events in the system (9). Programs and interventions can also be conceived as evolving networks of
person-time-place interaction. This observation invites research on how an intervention:
a. couples and embeds within context and manifests itself in thinking and practice;
b. changes relationships—patterns of information giving and seeking, support, practical help, role
taking, skill use, decision-making, collaborating, competing, etc.;
c. displaces existing activities (which may account in part for intervention effect); and
d. redistributes and transforms resources (material, informational, social, cultural).
The capacity for an intervention to redistribute resources is its chief mechanism to address inequity, whether
the resources are taxes or new educational opportunities and skills. (10)
Systems thinking deliberately moves away from the most commonly used methods to determine the cause
and effects of interventions, or events, on specified outcomes. Established methods – such a linear regression
modelling – rely on assumptions about the variables in the causal relationship, such as their independence
and unidirectional cause-effect relationships. Systems approaches as outlined above, adopt various methods
to capture the complex nature of multi-directional, overlapping relationships, feedback loops, learning and
compounding effects of multiple changes within the system.
Key messages for policy
•
Systems thinking is an area in health research that analyses the parts of a defined system and their
connections in relation to the whole.
•
Complex systems are characterised by different levels of variability, uncertainty and different levels
of organisation that range from simple organisations within a system to highly complex ones.
Complex adaptive systems are a particular type of complex organisation characterised by feedback –
learning and self-adaptation lead to the emergence of new states with new properties.
•
According to systems thinking approaches, health systems should be described in terms of their
boundaries, agents, connections and drivers.
•
If the definition of core drivers and boundaries is lacking at meso level, then the capacity for
effective systems thinking is severely limited.
•
Health interventions are one of the priority areas for systems thinking which can elucidate how an
intervention:
(a) embeds within context and manifests itself in thinking and practice;
(b) changes relationships
(c) displaces existing activities (which may account in part for intervention effect); and
(d) redistributes and transforms resources (material, informational, social, and cultural).
Systems thinking deliberately moves away from standard methods of identifying the cause and effects of
interventions. It has adopted novel approaches to capture the complex nature of multi-directional,
overlapping relationships, feedback loops, learning and compounding effects of multiple changes within the
system.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 17
Complex adaptive systems theory
Complex systems theory
has arisen from two main schools of thought – general systems theory and
cybernetics (See Appendix I for an overview of the work of key scholars in these fields). As a theory it
provides a
model of reality
, not reality itself. However, it can have utility in solving particular everyday
problems.
We can use systems theory to distinguish between different types of systems. Along a continuum, they can
be classified as simple, complicated, complex (dynamic) and complex adaptive systems (summarised in Table
1).
In simple systems, elements of the system interact in one-to-one relationships producing predictable
outcomes. Simple systems can be engineered and controlled. They are closed to and therefore not influenced
by the external environment.
Complicated systems display some of the same characteristics of simple systems in that interactions between
elements in the systems are predicable, although any one element of the system may interact with multiple
other elements of the system. Relationships are still linear and outcomes are predictable. Generally
speaking, ‘complicated’ refers to systems with sophisticated configurations but highly predictable behaviours
(e.g. a car or a plane) – the whole can be
decomposed
into its parts and when reassembled will look and
behave like the whole again. They are also closed to and therefore not influenced by the external
environment.
Complex dynamic systems
have two key characteristics, they
self-organise
without external control and
exhibit
feedback
resulting in newly created, i.e.
emergent
(at times unforeseen), behaviours.
Complexity
is
the dynamic property of the system; it results from the interactions between its parts
.
The more parts
interact in a nonlinear way in a system the more complex it will be. Complex systems are also open, loosely
bounded, influenced by their environment. Such
fuzzy boundaries
entail some arbitrariness in defining a
system.
While any one system as a whole may be defined as a complex adaptive system, inevitably subunits are also
complex adaptive systems in their own right. Thus any defined complex adaptive system has to be thought
of as being simultaneously a subsystem of a larger system (or a suprasystem) and a suprasystem constituted
by a number of subsystems (this is known in the literature as the
nested structure of systems
).
Complex adaptive systems (CAS)
are special case of complex dynamic systems as they
have elements (agents)
that can learn and adapt their behaviours to changing environments. In the complex adaptive systems
literature the elements of the system are referred to as agents. Complex dynamic and complex adaptive
system behaviour is influenced by the
system’s history
, i.e. influences that have resulted in the current state
of a system have ongoing effects on future states.
The make-up of the complex and complex adaptive systems presents certain problems in terms of being able
to understand, describe and analyse them. While simple and complicated systems lend themselves to cause-
and-effect analysis, complex and complex adaptive systems require a mapping of relationships and drawing
of inferences that may be theory based or drawn from multiple sources of knowledge. (See Appendix II on
the Cynefin Framework (11) on understanding the function of CAS resulting from different strengths of
relationships between the structure of its agents).
Understanding the differences between types of systems is often the clearest way to define a complex
adaptive system. Table 1 summarises features of simple, complicated and complex systems and the language
used in the literature to describe them.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 18
Table 1: Comparing the terminologies and meanings of different types of systems
Types of Systems
Simple
Complicated
Complex
Mechanical systems
Complex (dynamic) systems
Complex adaptive systems
Structure of System
One-to-one relationships
One-to-many relationships
Many-to-many and
system-to-system relationships
(nested systems)
Outcomes
Highly predictable
Mostly predictable
Alter with history and initial conditions
Unpredictable/emergent
Linear
Non-linear and feedback
Complex - Chaotic
Outcome Patterns
A change in
x
results in a proportional change in
y
A change in
x
results in a
disproportional change in
y
Attractor patterns that may be appear chaotic
Control of System
Engineered
Laws of nature
Social “laws”.
No controlling agent
Purpose, goals and values
define
simple rules
for
interactions
Properties of System
Self-organisation results in emergent behaviour
Complexity of systems increases with the rise in # of agents
Relationship to
environment
Closed
Open – loosely bounded
Relationship of
components/agents
Behaviour of
components/agents
Cause and effect repeatable,
predictable
Cause and effect are separated over
time and space
Cause and effect only coherent in
retrospect and are not repeatable
No cause and effect relationships are perceivable
(might or might not exist)
Analysis
Cause and effect analysis (reductionism}
Structure: mapping
Function: inference based on laws
of nature
Structure: mapping
Function: inference based on prior knowledge
Testing
Lab
Lab/Discrete event and or system
dynamics modelling
Lab/field
Agent-based modelling
Field trials
Generalizability
Yes
Yes
No
No
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 19
Complex adaptive systems theory: making sense of how complex systems work
Complex adaptive systems theory seeks to make sense of how complex systems, such as social systems and
health systems operate and function. It seeks to define how, and why, an order exists within these systems,
despite the highly interconnected and interdependent nature of relationships and the emergent nature of
outcomes stemming from these relationships. Despite these features, complex adaptive systems sustain a
high degree of order (only when they lose order do they shift into a purely chaotic state).
A healthcare system is a social system. Complex adaptive system theory describes social systems as emergent
based on their common statements of
purpose,
goals
and
values.
When a social system has clearly defined
purpose, goals and values, and these are adhered to by all its agents, the system will be ‘goal-delivering’ (12-
18). The larger and more open the system the more difficult it is to define, and the more difficult it is for
leaders to galvanise its agents to adhere to its purpose, goal and values. Appendix III contains a case study of
a complex adaptive system – the NUKA system – that has relatively well defined boundaries that
demonstrates how this concept can be applied in a real world example. (See Appendix III: Case Study – the
NUKA emergent health system).
In complex adaptive systems theory:
• Values: refers to the ideals and customs of a system toward which the people/agents have an
effective regard. Values are concepts that transcend contexts. They are universal within the system
(19).
• Having agreed
purpose, goals and values
defines the
driver
of the system; together they give rise to
the “operational instructions” that coherently direct the interactions within a CAS. These are
termed “
simple rules
” (20), and must not be contradictory.
• “Simple rules” reflect the core values of the systems.
Core values
are those that remain unchanged
in a changing world. If internalised and adhered to by all agents it results in the “smooth running”
of the system (18, 21-23).
• “
Simple rules
” provide the necessary “safe space/freedom” to adapt agents’ behaviour under
changing conditions. Adaptation is desirable; it fosters creativity and provides flexibility; it is the
prerequisite for the emergence of the system and the achievement of its goals (also referred to as
learning) (18, 21-23).
CAS activity results in patterned outcomes, based on purpose, goals and values within the
constraints of the
local context
. These outcomes, while not necessarily intuitively obvious, are the result of the self-organising
properties of a CAS, and overtime lead to “newly” emergent states e.g. while a call for controlling whooping
cough through public health measures by a health department provides the goal, each local public health
unit will implement it somewhat differently resulting in a unique outcome. The resulting outcomes, while
somewhat different, are “
mutually agreeable
”. These outcomes form a pattern that represents the system’s
adaptive abilities to achieve the desired overall outcome.
Importantly, goals, values and simple rules, and the way they play out within the system are unique to that
system. They cannot be transferred from one place to another as the local conditions that resulted in the
system outcome will be different, the reason why even proven innovations fail when transferred into a
different context (24).
Key messages for policy
•
A health system should be viewed not only in terms of its component elements/agents (eg. human
resources, financing, hospitals, clinics, technologies, etc.) but most importantly in terms of the
interactions between the agents (25).
•
There is little argument that health systems are complex, however, there is widespread
disagreement as to whether they function as seamlessly integrated complex adaptive systems in
accordance with complex adaptive systems theory.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 20
•
Health systems, being composed of agents that can learn and change their behaviours in light of
changing environments, have all of the defining features of a CAS.
•
However there are observed difficulties for the health system to respond to changing demands
(such as a change in the morbidity patterns) and in defining purpose, goals, values and central
drivers.
•
Complex adaptive systems cannot be controlled, the direction of their development can be guided
and requires leadership that reinforces each organisation’s
purpose, goals and values
and allows its
agents to act “independently” within the boundaries of the organisation’s
simple rules
(20, 26-30).
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 21
Question 1: How does the concept of a complex adaptive system apply to the Australian
healthcare system?
Premise: That the concept of complex adaptive systems is applicable to the Australian healthcare system,
and key characteristics can be identified
Systems thinking approaches
As noted above, we know little about the characteristics and dynamics of overall health care systems, and
practically nothing of the key properties of health care systems as complex adaptive systems (CAS).
However, there are plentiful examples of tools, methods and applied analysis for describing and
understanding main components of health care systems, their properties and the effects of interventions, or
events, on agents within the system.
Various disciplines have contributed to the applied approach of ‘systems thinking’ in health care. Many of
these approaches have functional applications that could be applied (or are already in use) for assessing and
planning health care. Here we review relevant contributions from six sectors: engineering, business,
defence, computer sciences, public policy and geography. We give examples where applications have been
used in Australia, or elsewhere in a way useful in the Australian context.
Systems approaches from engineering
Systems engineering focuses on how to design, operate and measure complex systems over their life cycles,
and how to analyse and improve their efficiency, productivity, quality and safety. Even though each system
is an integrated whole, systems engineering has developed methods for understanding complexity by
decomposing subsystems, specialized structures and sub-functions (e.g. microsystems) (31). There are
plentiful examples of the application of systems engineering to health care delivery; although it is more
common in the US and Europe than in Australia. These methods include Plan-Do-Study-Act (PDSA),
Situation-Background-Assessment-Recommendations (SBAR), stochastic modelling, House of Quality, and
statistical process control charts based on the lean method of six sigma steps: (1) Identify needs; (2) Define
requirements; (3) Specify performances; (4) Analyse and optimize; (5) Design, solve and improve; (6) Verify,
test and report (31, 32). Systems engineering has been adopted in the US to provide recommendations for
reducing the health care waste, and to improve its overall efficiency (33). An example of this approach is the
initiative promoted by Johns Hopkins University Medical Faculty, Applied Physics Laboratory and the
Whiting School of Engineering’s Systems Institute to couple systems engineering principles and best
practices with clinical expertise to improve understanding of the interactions among agents (clinicians,
patients, families, and other stakeholders), processes (institutional, regulatory, professional ethics, etc.), and
technology (medical devices and instrumentation) to formulate innovations and better patient outcomes
(34).
Systems approaches from business, economics and knowledge management
Some of the more advanced analyses of human organisations as complex systems come from the business
sector. Business intelligence uses qualitative and quantitative tools for explanatory and predictive modelling
to drive decision making. Business analytics uses a number of techniques that significantly improve the
capacity of traditional statistical techniques for data analysis under conditions of uncertainty, going beyond
multivariate analysis and data mining to incorporate modelling, knowledge management, expert knowledge
and artificial intelligence, and visualisation. Examples of their application in health care include Knowledge
Discovery from Data (KDD), Expert-based Cooperative Analysis (EbCA), outcome management analysis or
causality analysis using hybrid-based modelling (35, 36), and use of the lean approach to health logistics – on
which there is an extensive literature (37). The quality improvement model (IHI-QI) developed by the
Institute of Health Improvement in the US also incorporates systems thinking and system network analysis,
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 22
theory of knowledge, organizational psychology and system dynamics. The IHI-IQ model and the lean
method have recently been applied to the assessment of innovation in health care (38).
Hargreaves has reviewed the methods and techniques used for evaluating system change in health care(39).
She provides a clear differentiation between complicated and complex organisations. According to this
author computer simulation models of stocks, flows, and feedback; causality models; social network analysis,
and interrupted time-series analyses are useful techniques for the analysis of complicated systems, whilst
Geographic Information Systems (GIS), agent-based modelling, time-trend analysis and adaptive learning
systems, backward engineering or retrospective evaluation are suitable for the analysis of complex systems
(39).
Hybrid techniques derived from business intelligence and analytics were applied first in health economics in
the health care sector and then applied to system analysis in health care. Business analytics combined with
visualisation tools and mapping may be particularly useful for improving the analysis of large volumes of
data for policy and planning. An example of this approach is the development of an integrated information
system in the Northern Health region in British Columbia (Canada) to guide planning in this health care
system (40).
Systems approaches to knowledge transfer have drawn on management thinking about complexity to
identify the conditions under which change is likely in health organizations (41). Best and Holmes (2010)
have shifted the analysis of knowledge transfer from mechanical models of diffusion and dissemination to
build implementation models incorporating the interactions of evidence and knowledge, networks and
communications and leadership (42).
Systems approaches from defence, aviation and other high-risk sectors
High risk industries and sectors such as commercial aviation, the oil and gas industry and defence have
developed most rigorous methods to measure and improve safety performance. These have provided models
for improving performance in health care (43). The High-Reliability Organisation (HRO) is a safety
operational technology developed for dangerous and complex environments where an error can have fatal
consequences, such as aviation, nuclear power plants, or fighting wildfires. This approach was developed for
the US Navy’s nuclear-propulsion program, for contexts where agents (operators and users) “don’t have the
luxury of learning from their mistakes” (44). It develops a zero-defect culture based on the recognition that
highly technical operations depend on the interaction of systems, subsystems, agents and contexts. These
complicated/complex interactions give rise to deviations that must be corrected before they become fatal.
Winnefeld and colleagues enumerate 6 principles of HROs: integrity, depth of knowledge, procedural
compliance, forceful backup, a questioning attitude, and formality in communications. The US Agency for
Healthcare Research and Quality (AHRQ) has released operational advice for the transfer of this approach to
hospital management (45). The National Patient Safety Foundation (NPSF) has incorporated the HRO
approach into their guidelines to improve “Root cause analyses” (RCAs) at hospitals for exploring safety
events (46). SWARMing is an approach based in techniques from NASA (RCAs) and the US Veterans
Administration, which uses triage cards; and has been applied for system improvement by reducing adverse
events in the UK and the US (47). WHO has developed a similar conceptual framework for incident
reporting within the International Reporting and Learning Systems (RLS) Community of Practice. This
conceptualises adverse events as results of a complex interaction of agents and processes. Reduction requires
a comprehensive approach to this systems context. (48).
Another relevant contribution from the military sector, drawing on experiences of aviation safety, is the
analysis of the agents operating in a system as functional teams. The US Department of Defense (DoD) and
the AHRQ developed TeamSTEPPS, a teamwork model that offers a powerful solution to improving
collaboration and communication within health care settings (meso-level organisations). These teamwork
approaches have been one of the key initiatives within patient safety that can transform the culture within
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 23
health care. Teams are defined by the TeamSTEPPS model as “two or more people who interact
dynamically, interdependently, and adaptively toward a common, shared and valued goal, have specific
roles or functions, and have a time-limited membership”. A recent review of the implementation of this
systems approach in large health systems found ‘measurable improvements in teamwork, communication,
and patient satisfaction’ and a decrease in errors and averse outcomes. (49).
Systems approaches from computer sciences
Computer sciences have played a decisive role in the development of many of the advances in the different
sectors mentioned in this section. Since the 1960s decision support systems (DSS) and toolkits were
developed in defence, business, policy and education. The different approaches were summarised in the
1980s in a typology of DSS that included: 1)
File drawer systems
that provide access to data items; 2)
Data
analysis systems
that support the manipulation of data by computerized tools tailored to a specific task and
setting or by more general tools and operators; 3)
Analysis information systems
that provide access to a
series of decision-oriented databases and small models; 3)
Accounting and financial models
that calculate the
consequences of possible actions; 4)
Representational models
that estimate the consequences of actions on
the basis of simulation models; 5)
Optimization models
that provide guidelines for action by generating an
optimal solution consistent with a series of constraints; and 6)
Suggestion models
that perform the logical
processing leading to a specific suggested decision for a fairly structured or well-understood task (50). DSS
have had a huge impact in health care. They have played an essential role in the development of
knowledge-to-action strategies in systems thinking (see below), by incorporating new tools to model and
data-oriented systems, management expert systems, multidimensional data analysis, query and reporting
tools, online analytical processing (OLAP), Business Intelligence, group DSS, conferencing and groupware,
document management, spatial DSS and Executive Information Systems (50).
Simulation models can incorporate multiple levels of complex interactions. The Prevention Impacts
Simulation Model (PRISM) compares different interventions for reducing cardiovascular disease risks in a
controlled and systematic way). The comparison included the effects of interventions across diverse policy
domains – including clinical, mental health and behavioural, to start to represent the interactions
characteristic of complex systems in real life (51). In Australia, simulation modelling has been used to test
the combined effects of multiple interventions aimed at suicide prevention. This multi-scale dynamic model
mapped the structure of relationships between agents and events, encompassing both feedback and delays.
The model was used to enable desktop experimentation of policy scenarios. (52)
Marshall and colleagues provide a concise summary of the most commonly applied dynamic modelling
techniques as part of their work within the International Society for Pharmacoeconomics and Outcomes
Research. Systems dynamics (that models feedback, accumulations stocks, flows, and time delays), discrete
event simulation (that models queuing processes) and agent-based modelling (that models agency, dynamics,
and structure) integrate techniques from the above listed disciplines and have found widespread application.
The use of these models in healthcare has recently accelerated, although mostly for operations and logistics
in hospital settings such as scheduling, queuing and transportation. (53)
Systems approaches from public policy
Developments in public education and public policy deserve special attention. The development of Decision
Support Systems and knowledge-to-action strategies in public policy have also provided models for
evidence-informed health care policy. Microsimulation techniques have been increasingly adopted whereby
complex real-life events are simulated and the impact of policy change on the individuals that make up the
system are predicted.
An example of the potential applicability of new policy approaches to health care is Marine Ecosystem-
based Management (MEBM). It provides a guided approach for understanding complex processes and for
designing knowledge-to-action policy from a holistic, integrated approach to manage ecosystems, including
human (54). It is based on three principles: salience, credibility and legitimacy. This approach provides tools
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 24
to understand trade-offs to be made between ecological, economic and social sustainability criteria, the
diversity of cross-sectoral perspectives, values, stakes, and the specificity of each individual situation in
determining the outcomes of these trade-offs. The challenge of designing effective systems for linking
knowledge and action—systems that produce information that is perceived to be salient, credible, and
legitimate—is complicated if a system is perceived to be seriously lacking on any one of these dimensions,
its likelihood of producing influential information falls significantly.
The Interactive Systems Framework (ISF) for Dissemination and Implementation is a practical application of
the knowledge-to-action framework to health care policy. It has been developed to address the “how to” gap
that exists between scientifically determining what works, and moving that knowledge into the field for the
benefit of the public, particularly in prevention strategies. It has been applied by the US Centers for Disease
Control and Prevention (CDC) for designing prevention policies in highly complex areas. The ISF includes
activities or functions that are carried out by a variety of individuals in many different roles that make
dissemination and implementation possible. These activities include: (1) distillation (Prevention Synthesis
and Translation System—PSTS), (2) support (Prevention Support System—PSS), and (3) delivery
(Prevention Delivery System—PDS). By understanding the functions of these three systems and how they
interact, stakeholders (organizations, funders, researchers, and practitioners) can communicate better and
work together to disseminate information and more effectively implement prevention innovations. CDC has
developed the Rapid Synthesis and Translation Process (RSTP), using the exchange model of knowledge
transfer in the context of one of the ISF systems: the Prevention Synthesis and Translation System. It has
applied this to the design of health and integrated polices in violence (55) and suicide (56). There are other
approaches to rapid synthesis such as the Robert Wood Johnson Foundation’s (2009) six-step process in their
Synthesis Project (http://rwjf.org/pr/synthesisabout.jsp); or the exchange model adopted by the Canadian
Health Services Research Foundation to improve the evidence based decision making capacity of policy
makers (55).
Realist evaluation provides an empirical evaluation approach for the study and understanding of
programmes and policies. This technique assumes that knowledge is a social and historical product and that
programs are embedded in social systems. This is an increasingly popular approach in health systems
research seeking to understand why complex interventions work. Realist evaluation identifies the
mechanisms, contexts and the conditions enabling (or blocking) the transferability of interventions (57).
The guides for quality and reporting standards and training materials for realistic evaluation in health care
are currently under development (RAMESES project) (57).
Systems approaches from geography and context analysis
The boundaries of a system, its subsystems and clustered systems and the relationship with neighbouring
systems need to be formally defined and described before starting any system analysis. Paradoxically, the
majority of system literature in health care does not provide enough –if any- information on the
geographical location and spatial characteristics of the systems analysed. Geographical information systems
can be to represent and to understand the distribution of agents, their connections and spatial relationships
(58). Although relatively new, health geography has gained major attention in the last 15 years and has been
recently strengthened by the development of context analysis. Context refers to the totality of
environmental circumstances that comprise the milieu of human life. It has provided an integrative
framework to identify policy goals related to personal outcomes and to other indicators in healthcare. The
evidence gathered in observational/ecological studies in local areas together with the analysis of big data,
service use and costs have enabled analysis of care in a systems contexts (59, 60). Atlases of health care
provide decision support tools that combine context analysis of jurisdiction boundaries, agents operating in a
system, service delivery and network analysis. A growing number of examples of the use of these tools for
health system analysis have been published in Canada (40, 61); and Europe (62, 63). For example, the
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 25
mapping of food deserts, a geographical area devoid of access to (healthy) food has influenced outlet
licencing policy. (64)
While many of these approaches use some of the language of complex adaptive systems analysis, there are
few examples of empirical or operational analysis that uses the full CAS framework. However, each of the
approaches summarized above – and many more – are making profitable use of systems thinking to give a
better understanding, and develop analytical and policy tools, to understand health systems. At their
strongest, many of the tools and methods listed above have culminated in two most commonly applied
systems thinking tools for health care planning – mapping and dynamic system modelling.
Key messages for policy
•
Little research has been conducted that use a full complex adaptive systems framework to analyse
key properties of the Australian health care system as a whole (CAS).
•
There are plentiful examples of tools, methods and applied analysis from allied sectors that can be
applied for describing and understanding main components of health care systems; including from
engineering, business, defence, aviation and other high risk sectors; computer science; public policy
and geography.
•
Key applications of these methods include: risk management, predictive modelling, life-cycle
mapping, knowledge-to-action frameworks, social and system network mapping, decision support
systems, root cause analysis and safety operational technology.
•
Few examples of analysis techniques above use the full complex adaptive systems framework,
however, each of the approaches summarized above – and many more – are making profitable use
of systems thinking to give a better understanding, and develop analytical and policy tools, to
understand health systems.
Complex adaptive systems theory
Based on the principles that define organisations as complex adaptive systems in CAS theory the Australian
health system has to be considered to be a CAS. It has:
• agents – health ministers, health bureaucracy, health financing organisations, hospitals and
community offices, the various health professionals, health user groups and individuals (note:
agents in systems terms can be people, organisations, institutions etc), and
• these agents interact and through these interactions,
learn
– in consultation, within a hospital,
between central bureaucracies and local health service units etc.
However, the “operating principles” – the framework that defines the boundaries within which agents are
expected to act – are not readily evident. Tacitly:
•
purpose
– may be loosely described as helping people with illnesses
•
goals
– are most clearly defined as providing universal healthcare
•
values
– are not succinctly stated and evident to all agents
•
simple rules
– are not overtly defined.
As the
purpose, goals, values
and
simple rules
of the Australian health system are not clearly evident,
the
system as a whole has no identifiable unifying driver
. The lack of a unifying driver makes the Australian
health system look more like a “conglomeration of discrete units”, colloquially expressed as “a fragmented
health system”.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 26
Subsystems of the Australian health system
More commonly the health system is discussed in terms of it subunits, e.g. primary care, mental health, the
hospital, public health etc. As these subunits are CAS in their own right they can be analysed according to
the same principles. It is outside the scope of this review to describe individual units, sub-units, sub-sub-
units etc, however, it is clear that the agents of small functional units like an intensive care unit (65) at large
have a clear but
tacit
understanding of their
purpose, goals and values
and understand the
simple rules
that
guide their actions and behaviours.
Aligning drivers
The driver of a system provides the focal and unifying reference point for all agents in the system. While it
is acknowledged that subsystems require individual drivers to function, it is critical that for a system with
many subsystems to function seamlessly as a whole, subsystems need to align their drivers with the
overarching one that defines the whole system.
Current Australian health system most likely does not function as a CAS
The general observation that the Australian health system is fragmented and poorly integrated leads to the
conclusion that – as a whole – while having the structure of a CAS it does
not function
as a CAS. Using a
CAS-analysis approach Sturmberg, O’Halloran and Martin (66, 67) attempted to describe the Australian
health system in relation to the criteria of a “truly complex adaptive” system in accordance with complex
adaptive systems theory. The findings are represented in the health vortex (See: Figure 2) – the vortex
entailing the notion of the system’s focus and overall function; rising from the apex are the local primary
care services and hospital services, community level services, regional health services and at the top the
policy level.
Figure 2: Application of vortex model to the Australian health system
The left side of Figure 2 shows the fragmentation within the various organisational levels in the Australian
healthcare system (explored further in response to Question 2, below) resulting from differences in drivers
of subsystem units like: focus on patient’s health experience at the primary care level, disease focus in
hospitals, focus on public health priorities at the community services level, and a focus on global disease and
health budgets at the policy level. The vortex representation on the right shows a
seamlessly integrated
complex adaptive health system that assumes as its overarching system driver the “person’s health needs”.
As an overarching driver, levels across and within the health system would all align their activities towards
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 27
achieving the system’s “common goal”, but would do so autonomously in their own ways (subsystems
determine their own drivers).
The Australian health system incorporates subsystems of a simple, complicated, complex and complex
adaptive nature and is illustrated in Table 2.
Key messages for policy
•
Understanding the health system as an integrated CAS requires agreement of an overarching driver
that needs to be brought and kept in the forefront of all health system agents.
•
As most problems confronting health policy makers relate to subsystem issues, policy makers
should consider any problem in the context of the whole system.
•
Most adapted solutions for the problem of a subsystem will emerge from helping all agents of that
system to align their drivers with those of the whole system (addressing the interconnected nature
of system structure and function).
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 28
Table 2: Simple, complicated, complex and complex adaptive subsystems in the Australian health system
Types of Systems
Simple
Complicated
Complex
Mechanical systems
Complex (dynamic) systems
Complex adaptive systems
Examples in
community health care
Flu vacciantion
Managing a 2nd degree burn
Community Care for frail elderly,
linking care domains
Collaboration in GP clinic
Multimorbidity management
Examples in hospital
health care
Laboratory testing
Hip replacement surgery
Bed management in a hospital
Intensive care unit
Acute psychiatric unit
Staff management of a hospital
Examples in health sub-
system
Protocols, e.g. handwashing policy
Mental health
Managing an epidemic
Managing a natural disaster
Health financing
Structure of System
One-to-one relationships
e.g. nurse giving Flu shot
One-to-many relationships
e.g. surgeon managing a theatre
team
e.g. nurse unit manager ensuring
staff records every patient
incidents regardless how trivial
Many-to-many relationships
System-to-system relationships (nested systems)
Subsystems can have simple, complicated, and complex components
e.g. the value of services listed in the MBS (simple)
however the fees charged by individual doctors to different patients in their
practice is highly variable and depends on many factors (complex adaptive)
Outcomes
Linear
e.g. measured as percentage of
eligible population being vaccinated
Mostly predictable
e.g. measured as number and type
of in theatre complications
Alter with history and initial conditions
Unpredictable/emergent
e.g. contextual description of the SES characteristics of population groups, the
detailed description of delivery of an intervention and the observed outcomes
(e.g. harm reduction of illicit drug use)
Linear
Non-linear and feedback
Complex – Chaotic
Generalizability
Yes
Yes
No
No
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 29
Question 2: In considering the Australian healthcare system as a CAS, what agents exist at a
macro, meso, micro and nano level and what are the relationships between them?
Premise: That agencies in the Australian healthcare system can be mapped in terms of levels and inter-
relationships
Systems thinking approaches
Health systems operate in fuzzy boundaries with other systems and within boundaries set by geography and
levels of organisation. Wilson et al. (1995) stratified decision-making levels within health services,
identifying “micro” (between patient and clinician); “meso” (community level, including healthcare
services) and “macro” (governmental) (68). To enable the study of health systems at the level of service
delivery, the level at which users experience health care, “micro” should refer to “individual service” and
“nano” to the individual patients.
Berwick’s alternative scheme for conceptualising decision-making layers takes a more three dimensional
approach. It defines microsystems as the building blocks of a health-care delivery system where direct
interaction occurs between the patient and provider (69). Moving outward from this microsystem, the
delivery system encompasses more stakeholders at the mesosystem level (e.g., divisions of general internal
medicine, surgery, or nursing) and further out at the macrosystem level (e.g., hospital administrators, or
government regulators and policymakers) (31, 70).
This schema can be complemented by the Donabedian model as Tansella and Thornicroft (1998) have done
in the mental health care context. The model adds geographic levels (country, local area, individual) to form
a matrix, enabling a more holistic and systemic analysis of integrated care across the different components of
the system (71).
System levels and agents in applied analysis
Deriving useful learning means carrying out a thorough analysis involving quantitative and qualitative
methods with a range of different stakeholders (e.g. researchers, patients and healthcare professionals), at
different levels of the system (macro to micro), and considering the context of the system. A significant
proportion of the available literature in health systems and planning refers only to a limited number of
agents at organisational levels – and rarely are these analyses combined with contextual and spatial analysis.
We return to the building blocks of systems analysis to discuss these deficits in the literature and give some
examples where they have been used.
•
Boundaries.
We see the start of the use of systems approaches to analyse spatial contexts and the
boundaries of local systems. This has been undertaken by several health atlas projects, mainly at
local level (eg Fernandez et al, 2016 (72)). The recently published
Australian Atlas of Healthcare
Variation
(73), identifies areas of high variation in a series of key care performance indicators. It
describes for the first time health care provision across Australia and constitutes a major source of
information on service availability and care variability across the health system. It also provides a
detailed description of several subsystems and its agents.
•
Agents.
Systems approaches to health care assessment and planning usually define agents of interest
in the system, but rarely specifically stratify them according to system levels. In part, this is likely
to be a deliberate decision in recognition of the blurred lines of connectivity and location of action
in a complex system. ‘Locating’ an agent within a system level remains a conceptual exercise to a
degree, but a useful one all the same. It can assist in identifying key agents at the stage of mapping a
system, reducing the likelihood of ‘missing’ important agents. In Appendix IV we summarise
different agents that have influence within the system and stratify them according to the system
level where they exert direct influence and their interest within the system. For example, on the
macro level, policy makers in government, large firms, lobby groups and non-government
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 30
organisations exert influence. On the meso level, community, state and federal public institution in
various sectors plan and coordinate services. On the micro level health care services are delivered
and service providers interact directly with users and each user interacts with its direct social circle
comprising family and friends. On the nano level we see the individual.
•
Connections.
Social network analysis is emerging as one of the more popular tools for identifying
connections between agents at different decision making levels for assessing and planning health
care in Australia. For example, Nancarrow and colleagues mapped service integration for primary
healthcare patients in the Lismore GP superclinic. They defined ‘integration’ in terms of the
connections between agents at micro, meso and macro levels within the system, and assessed actual
connections by means of a social network analysis (Unicet 6) of patient referrals. (74). Systems
modelling has also been used in recent years to better understand the connections between agents
and use information about what is known about how agents react to environmental changes to
predict outcomes. Latkin et al. modelled HIV-related behaviours that influence HIV prevention and
detection on macro, meso and micro levels (75) and Maddox et al have examined social network
and their impact on tobacco use. (76, 77). Sadsad and McDonnell (2014) outline steps for creating
multiscale models on the example of stock levels in health services and include the determination
of systems levels as part of the process.(8) Predictive risk modelling has also become a widely
utilised method to predict future health service utilisation, although the majority of models
currently applied for this purpose do not take a systems thinking approach. (78)
•
Drivers.
There have been attempts to formally state the objectives and drivers of Australian health
care. In 2009 the Australian Government Department of Health and Ageing released a strategy
“with the direct intention of reaching the goal of Australia being the healthiest nation by 2020”.
Following this, health promotion and prevention may be regarded as priority goals in our system.
The 7 strategic directions included shared-responsibility, community engagement, the need to
influence markets and develop coherent policies as well as to “refocus primary care towards health
prevention” (3). The national health priority areas were focused in 9 health chronic conditions. The
concept of person centred health care was notable by its absence. On the other hand a number of
subsystems such as mental health have re-oriented their drivers towards person-centred care both
at national level and in several local areas.
In his integrative approach, Frenk emphasises that the analysis of systems dynamics is not possible if we do
not collect the basic information required on agents, connections, boundaries and drivers (79). Key authors
of this report are currently undertaking a program of analysis that brings together each of these elements in
an analysis of agents at various system levels in mental health. The “Integrated Atlas of Local Mental
Health” carried out by the University of Sydney has mapped the functional teams operating at the micro-
level in several health districts (PHNs/LHDs) in metropolitan Sydney and in the Far West in NSW (72).
Service availability, placement capacity and workforce capacity were mapped and an analysis
accessibility/service gaps/unmet needs in the local population undertaken. In the near future this
information will be completed with the social network analysis of the connections of the different agents
operating at micro, meso and macro level in the local system.
Key messages for policy
•
Decision-making levels within health services can be stratified into the following: “nano”
(individual); “micro” (between patient and clinician); “meso” (community level, including
healthcare services) and “macro” (governmental).
•
Schemes using these levels can be complemented by the Donabedian model adding geographic
levels (country, local area, individual) to form a matrix, enabling a more holistic and systemic
analysis.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 31
•
Systems approaches to health care assessment and planning usually define agents of interest in the
system, but rarely specifically stratify them according to system levels. In part, this is likely to be a
deliberate decision in recognition of the complex nature connectivity and location of action in a
complex system.
•
(Social) network analysis and systems modelling techniques (multi-scale, dynamic, multi-level) are
increasingly being used to analyse connections between agents in a complex system.
Complex adaptive systems theory
The question presupposes that the Australian health system is organised in functional units (subsystems) at
distinct organisational levels.
Organisational levels
Organisational theory views organisations as multi-level structures with macro, meso, micro and nano level
structures and functions (the roles and responsibilities at each level are detailed separately in Appendix IV).
In the context of the health system we typically think of:
• the macro level as dealing with policy and governance issues;
• the meso level managing regional health, community, social and infrastructure services;
• The micro level providing local/individual care delivery in the local community; and
• the nano level describing personal/organismic health and disease characteristics/functions (80).
This conceptualisation of the health system comprising multi-level structures is consistent with the
complexity notion of nested systems in CAS theory. Each level will have its own unique structures and
activities with well-defined roles and sets of well-defined rules. Whilst many of these levels/subsystems may
appear to work effectively when viewed in isolation, it is critically important to ensure that they seamlessly
contribute to the integrated function of the ‘whole system’. Only when agents and their various subsystems
act in accordance, with the system as a whole will it be ‘seamlessly goal-delivering’.
There is a wide range of agents in the system with diverse interests in the Australian health system (See
Appendix IV, Table 4 for some examples listed according to system level). Most notably their focus is often
limited and self-serving e.g. at the macro level a health minister has to mediate justifiable health delivery
demands against resource concerns, and in making those decisions is lobbied to act in the interest of vested
stakeholders, be it industry interests, citizens lobby groups or other non-government organisations.
Links across system level – the example of food regulation
There are understandable constraints on agents at any system level in taking a ‘whole of systems’
perspective, i.e. one that works coherently to achieve common purpose, goals and values. This is because:
1) Subsystems may adopt purpose goals and values out of alignment, or even divorced from those of
the whole system. As described above, the Australian health system as a whole has no identifiable
unifying driver, making the system more like a ‘conglomeration of discrete units’. Thus even if one
agent may act in accordance with the rules of their own subsystem, these may not contribute to a
common goal-orientation of the whole system.
2) Due to the systems highly distributed network nature, it may not be immediately clear, or
devisable, what course of action will be most promising to lead to the desired outcomes in any
another part of the system or for the system as a whole.
An illustrative example might be an agent on the macro level – a politician or senior bureaucrat responsible
for food industry regulation. The prevailing “whole of government” perspective may be reduced to short
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 32
term budgetary impacts, rather than consequences of regulation changes (e.g. limits on salt and sugar
content in foods) throughout the entire system.
This example alludes to “whole of system dynamics” e.g. changes to food regulation (a macro level activity)
will interfere with the status quo of food industry economics (meso level) (people employed, cost of inputs,
profit margins, taxes paid), the cost of foods available to consumers, but also the potential cost savings to the
health system due to better health of individuals and the community. The latter though, will have economic
implications to private sector health service providers. The potential need for fewer health services affects
health professional education providers and potentially affects the social service system should this result in
greater unemployment of health professionals in the community. At the nano level, community education
about healthier food choices might lead to a greater demand for healthier food, thus providing additional
incentives for food producers to change their production processes and product ranges. People might have
to rebalance their budgets; some higher spending on food may be offset against less spending on health
services. In addition, changes in food choices at the nano-level require adaptation in taste and the
acquisition of additional cooking skills.
A core problem for the Australian health system is its fragmented nature. This has arisen from the lack of
well-defined and promoted common goals, values, simple rules and thus drivers. Consequently, the
Australian health care system is unpredictable. Mapping, modelling and field trial methodologies provide
good starting points to understand these multiple subsystems of the health system.
Key messages for policy
•
Problems arising in any domain of the health system have implications across all organisational
levels.
•
Strong links between and across organisational levels increase the robustness and adaptability of
the system.
•
The fragmented nature of the Australian health system makes it difficult to act with a ‘whole of
system’ perspective because:
o
Values, goals and drivers of different sub-systems are not aligned – or may even conflict
o
It is difficult to derive the potential influence that action in one part of the system may
have on other parts of the system.
•
System mapping and modelling are group learning tools that may be applied
o
to advance whole of system thinking in the absence of common drivers and
o
to work collaboratively towards defining mutually agreeable purposes, goals, values and
simple rules.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 33
Question 3: In considering the Australian healthcare system as a complex adaptive system, where
do consumers fit and what are their relationships with other agents within the system?
Premise: That the consumer has a place in a complex adaptive systems framework.
Systems thinking approaches
Based on the available evidence, we cannot determine whether the Australian health system asserts all
properties of an integrated goal-oriented complex adaptive system. However, it is unquestionable that
consumers are key agents of the Australian health system. Here, we refer to ‘users’ of the system, rather than
‘consumers’ encompassing not just patients receiving episodes of care, but individuals in various roles within
the whole system.
Users are present in the system in many different roles, as patients, family members of patients, payer,
advocates, voters and citizens; and individuals can wear multiple hats simultaneously. In such situations
they are frequently balancing competing interests. These different roles bring different but overlapping
relationships with the other agents. Users ‘learn’ about the system from experiencing it themselves. They
form views about other agents from hearing about the experiences of others and from information supplied
by more or less credible sources including their health care providers and the media. Other agents, such as
doctors, health care managers, health insurers seek to modify the behaviour of users. Users’ response to
these interactions depend on the sum of those interactions but also broader influences such as their
personal, social or financial situation, social, political or religious settings and particularly their prior
experience of care. User responses to similar interactions therefore may vary depending on the influence of
other factors.
Probably one of the most critical factors for users is the information imbalance in health care decision-
making. Medical care is becoming increasingly complex, and the impacts of different diagnostic and
treatment technologies less straight forward. For example, our ability to detect early precursors or changes
of conditions such as dementia and other neurodegenerative conditions without being able determine the
specific implications for the individual introduces a new dimension of uncertainty for the patient. Similarly,
treatments may not offer cure but small increases in survival at the expense of significant side-effects, the
need for ongoing therapy or loss in quality of life. It is very difficult for an individual patient and their
family to understand the full implications and they are reliant on the expertise of the treating clinician to
interpret this. Understanding and interpretation is even more complex when the individual has multiple
conditions.
There is widespread agreement that users should be engaged in health care decision making. The spectrum
of user involvement will be as multiple as the number of relationships they have within the system.
However, if we accept the inherent characteristics of agents in complex adaptive systems, particularly that
in response to their multiple roles and competing interests, agents tend to adapt to each other’s behaviours,
and also that agents are intelligent and learn from their interactions, then it points us to some different
models of interaction.
The growing interest in involving the public in decisions about healthcare provision inspired the
development of ‘citizen juries’ to identify and bring community values – and ownership – into health
services. This could be a representative way of providing input into a complex adaptive health system
design. This model was adopted for health care reform in the US (NIH Consensus Development Program)
(81) and in Canada where consensus and state-of-the-science statements are prepared by independent panels
of health professionals and public representatives. Citizens' juries, whose members were randomly selected
from the electoral roll (rather than derived from user interest groups), were trialled in Western Australia a
decade ago (82) and recently with the National Disability Insurance Scheme (NDIS).
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 34
Other key approaches to user involvement in the change of the health system is the concept of co-design
and co-production. In practice, every day we make decisions intend to influence the design and operation of
the health care system. As all agents in complex systems learn and change as a result of their interaction,
then the concept of co-production has inherent attractions. A 2015 update reviewing areas in which user
representative input has been influential in health policy and program decisions identified 13 areas (83):
1) identifying consumer issues that needed to be addressed in policy and program design;
2) improving the design and targeting of health communication strategies;
3) influencing the content of medical education and training programs;
4) balancing the interests of industry sectors through presenting the consumer perspective on health
program resources and materials;
5) increasing the effectiveness of existing programs through improving access and targeting strategies;
6) influencing reporting requirements to ensure a consumer perspective is included and available to
influence government decision making;
7) raising the profile of existing programs with key bodies to gain their support and endorsement;
8) advocating for health literacy strategies to empower consumers;
9) influencing the content of key government health policies such as the National Medicines Policy;
10) contributing to a cultural shift to broader views on consumer participation in health policy and
program development;
11) influencing governance arrangements to ensure ongoing consumer input throughout an
organisation;
12) changing the rhetoric and terminology of health policies and communications to reflect consumer
experience; and,
13) achieving improvements in transparency of committee decision making processes.
Finally, collaborative care, self-support and peer-support are low-intensity complex interventions that
require a system approach for its design, evaluation, review and implementation (84) (85) Community-led
interventions have been implemented in Australia in some areas such as palliative care. LifeCircle conducts
a mentoring program in NSW, in which volunteer mentors provide support exclusively to primary carers of
terminally-ill patients, helping them to gather a support team and avoid burn-out, apart from other
community led programs in palliative care (86).
Key messages for policy
•
Users/consumers are key agents of the Australian health system.
•
Users are present in the system in many different roles, as patients, family members of patients,
payers, advocates, voters and citizens; and individuals can wear multiple hats simultaneously. In
such situations they are frequently balancing competing interests.
•
Users’ responses to interactions with other agents and events within the system depend on the
combined weight of those interactions, but also broader influences such as their personal social or
financial situation, social, political or religious settings and particularly their prior experience of
care.
•
User responses to similar interactions/interventions therefore may vary depending on the influence
of other factors.
•
Probably one of the most critical factors for users is the information imbalance in health care
decision making, particularly in light of increasingly complex diagnostic and treatment pathways.
Complex adaptive systems theory
Agents within the health system include not only institutions and organisations but also the whole
population:
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 35
o as patients, with specific needs requiring care;
o as users, with expectations about the way in which they will be treated;
o as taxpayers/service purchasers and therefore as the ultimate source of financing;
o as citizens who may demand access to care as a right; and most importantly,
o as co-producers of health through care seeking, compliance with treatment, and
behaviours that may promote or harm one’s own health or the health of others.
Users have different patterns of interacting with other agents of the health system at large. Users may be
involved with the system of primary care, prevention and health promotion, episodic treatment or acute
treatment. Only around 3.2% of users require secondary care and only 0.8% require resource intense
tertiary care in any given period, so users in our system should be conceptualised more broadly than patients
or users of certain services (87-89) (See Figure
3
).
Figure
3
: Distribution of health service use amongst ‘users’
Users are co-contributors to the health system
Taking a whole of system perspective, users (should) play a key role in determining the purpose and goals of
a system.
The case study presented in Appendix III – the NUKA system illustrates one example of how users can
contribute to defining the values, goals, purpose and thus driver of a health system. In that example a
national leadership (the United States Congress) legislated to create a geographically distinct subsystem for
Indigenous Americans (Alaskans). Users working within this relatively closed subsystem of the wider
American national and state health systems, were able to decide that they wanted a
person-centred
health
service that offered (
purpose and goals
):
• relationships with primary care providers
• being treated with courtesy, respect and cultural understanding, and
• access to care when needed
The community determined the vision of their health service (
values
). It should:
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 36
• contribute to the community’s physical, mental, emotional and spiritual wellness
• collaborate with the Native Community to achieve wellness through health and related services
• shared responsibility, commit to quality and family wellness.
• listen to people’s feedback and understand their needs AND explain the changes being made in
response to their feedback AND then communicate the organisation’s successes in delivering what
the community asked of it
As the example of the NUKA subsystem suggests, full engagement of users is essential to achieve a
user/patient-centred health system. But there are challenges in this approach, particularly in a system that is
fragmented, with a multicultural population, such as the Australian or broader US health system.
Engaging users in the design of health systems
Various approaches to increasing user involvement have been suggested in the literature, including:
• ‘Citizen juries’ as a way of providing public input into a complex adaptive health system
design(90). User consultation often will allude to non-health-professional aspects that contribute
to their health and health problems like social, housing, education, work and environment etc.
issues and needs to be considered in the re-design of the system.
• Organisational level multi-stakeholder involvement – locating of citizens at the centre of the
system surrounded by service delivery organisations and creating forums for joint decision making
at organisational levels (91). This design offers the greatest likelihood of achieving a truly
responsive, i.e. adaptive, health system; what happens at the higher levels is controlled by what
happens at the lower levels (the effect of bottom-up emergence).
• Centralised user forums, such as the Australian Consumer Health Forum that acts a gateway for
participation as well as lobby group for users at multiple system levels.
Key messages for policy
•
Agents within the health system include not only institutions and organisations but also the whole
population (and multiple population sub-sets).
•
Complex adaptive system theory recognises that while individual-level agents will have a set of –
potentially – wide-ranging properties, they will bring certain ones – autonomously – to the
forefront depending on context, and learn and adapt their behaviour accordingly.
•
Full engagement of users is essential to achieve a user/patient-centred health system. But there are
challenges in this approach, particularly in a system that is fragmented. It requires leadership
committed to engaging all involved in the process of defining purpose, goals, values and simple
rules and reinforcing these. Various approaches to increasing user involvement have been suggested
in the literature, including: citizen’s juries, organisational level multi-stakeholder involvement, and
centralised user forums.
•
Taking a whole of system perspective, users (should) play a key role in determining the purpose and
goals of a system - noting that according to complex adaptive systems theory, a goal-delivering
systems is one that has common values and drivers. Users can play a key role during phases of re-
build, re-design or during slower pace change. When users are involved in the design of the system,
they become owners of the system.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 37
Question 4: How can an understanding of the Australian healthcare system as a complex adaptive
system accommodate or support adoption of person-centred care?
Premise: That complex adaptive systems thinking is compatible with a vision to deliver care that prioritises
the consumer experience and values.
Systems thinking approaches
People and Person-centred health care (PPCHC) has been at the heart of recent attempts to improve the
quality and responsiveness of the health system. It requires a major shift from established modes of clinical
and administrative practice, making individuals, with their complex needs and preferences, the drivers of
health care.
Health systems exert a high degree of complexity. It consists of multiple complex systems, comprising a
large number of agents are connected along multiple pathways. Within the system, and sub-systems, the
effect of any one action or event is rarely proportional to cause. The enabling and delivery of person and
patient-centred care, with its focus on customising care to the needs of individuals, inevitably adds even
greater complexity to the system. (53)
PPCHC is a not only a desirable goal but necessary for good health care practice. In Australia, adoption of
PPCHC will mean moving even deeper into what Berwick describes as a health care system with the
simultaneous pursuit of the Triple Aim: improving both the experience of care, the health of populations,
and reducing per capita costs of health care. (92). Systems thinking tools such as dynamic simulation models
can assist in navigating these conflicting goals.
“In the context of health care delivery, a patient-centered approach requires an
understanding of the multiple and diverse determinants of health outcomes and patient
experience. Modelling these relationships and interdependencies at the system level
can provide a comprehensive view of the drivers that improve the quality of the patient
visit experience, such as shortened waiting times, quality of information, and access to
care. Care pathways can be designed to better reflect patient preferences for certain
subgroups, such as risk tolerance for therapies, the avoidance of adverse effects,
potential adherence to therapeutic regimens, or demographic characteristics and
medical history. In the complex interactions between doctors and patients, simulation
modelling may also yield insights into revealed versus stated preferences.”
(53)(Marshall et al 2015 . P 8).
PPCHC and has been a key component within the integrated care frameworks now actively pursued in
Australian health policy. This framework is closely related to system thinking and it has been extensively
reviewed in the accompanying expert report by the authors on person-centred health care. (93) The
perspective of the Australian health care system as a complex system is fully compatible with a person-
centred model. However, the goal of PPCHC must be located in a systems analysis. PPCHC needs a
complementary strategy to build integration across the care system. A narrower perspective loses sight of
the totality of needs and will increase inefficiencies, inequalities, unwarranted variation and waste in the
system.
PPCHC requires simultaneous change from the bottom up (e.g. individuals’ understanding of their health)
and top down (e.g. reallocating resources to enable providers to deliver needs-based care). The key learning
is that substantive change towards PPCHC will require systems thinking fit for purpose in the Australian
context. These points are developed in more detail in a separate expert commentary by the authors on
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 38
Person-centred care. That commentary identified the following facilitators in a system-wide move towards
PPCHC:
• Engagement with the person and people, shared management and decisions around health care
services
• Strong government and clinical leadership
• The integrated information systems and care pathways
• Inter-sector collaborations
• Focus on patient empowerment
For PPCHC to be achieved in a complex health systems system, specific actions, events and interventions
must integrate the level of individual practice (nano, micro) and organisational and whole system levels
(meso and macro). Systems thinking will aid in identifying the complexity of factors that influence a
person’s domains of health, such as environmental and personal factors and the relationships between
various components of the health care system.
Again, there is a great potential for widespread adoption of mapping and modelling techniques in the
Australian health system to aid health planning and policy to move the system closer to delivering PPCHC.
However, there is a need for caution. Even the most comprehensive models require some level of
aggregation of preferences, actions and consequences. Decision based on models can be a step closer to
PPCHC, but are still not adapted entirely to the individual. Rather models assist decision-makers to become
aware of how a system works and apply this knowledge in practice.
The use of modelling also faces barriers to implementation of use that should be taken into account.
Simulation modelling and mapping requires specialised skills and adequate data to populate them. Models
are also based on series of assumptions that can be subject to challenge and debate. The more sophisticated
the model (as is the case for dynamic system modelling), the more difficult it is to explain their logic, which
can meet with resistance from some policymakers. The greater the number of domains of interest, the more
sophisticated (and resource intensive) the mapping and modelling will need to be.
The Commission may consider the widespread uptake of simulation modelling for health service programs
and within the Australian Health System of the Australian Health System. The potential benefits of which
would need to be assessed on a localised basis. The SIMULATE Checklist, developed by Marshall’s group in
2015 may be one tool that can be used to judge those instances where the use of systems based simulation
models would be appropriate.
With adequate models of agents and connections and context, systems thinking can also be applied to
determine the relative strengths and weaknesses of interventions aimed to achieve change at any system
level.
Key messages for policy
•
PPCHC is a not only a desirable goal but necessary for good health care practice.
•
The enabling and delivery of person and patient-centred care, with its focus on customising care to
the needs of individuals, inevitably adds even greater complexity to the system.
•
PPCHC requires a major shift from established modes of clinical and administrative practice,
making individuals, with their complex needs and preferences, the drivers of health care.
•
Substantive change towards PPCHC will require systems thinking fit for purpose in the Australian
context.
•
There is a great potential for widespread adoption of mapping and modelling techniques in the
Australian health system to aid health planning and policy to move the system closer to delivering
PPCHC.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 39
•
However, there is a need for caution. Even the most comprehensive models require some level of
aggregation of preferences, actions and consequences. Decision based on models can be a step closer
to PPCHC, but are still not adapted entirely to the individual.
•
Rather models assist decision-makers to become aware of how a system works and apply this
knowledge in practice.
Complex adaptive systems theory
Moving the Australian health system towards a complex adaptive system that is coherent and goal-
delivering, requires a focal point (driver) to guide activities to achieve integration within and across
organisational levels of care. The authors of this report are convinced that a person-centred focus is most
appropriate for a health system that meets the needs of the Australian community, as:
• individuals experience the same disease in very different ways
• individuals’ circumstances surrounding the development of a disease have major implications for
their management, and
• not all diseases respond best to biomedical interventions, in many cases social and environmental
support results in
better health outcomes
for that person
The authors don’t underestimate the challenges in achieving a person-centred health system. Taking a CAS
theory approach, the main challenge from the whole of system perspective relates to how to transform the
values that sustain the current system. What needs to be facilitated is a system-wide conversation about the
general expectations and approaches to healthcare. Policy agencies have a key role to play in this
conversation by:
• encouraging local solutions to whole system problems
• giving attention to environmental factors that contribute to health – and ill-health
• acknowledging and leading discussion on what drives action in our health system
• promote multi-stakeholder involvement in health system – and sub-system, reform and design
• promoting key principles (e.g. People, Person and Patient centred care) at all system levels
Key messages for policy
•
There are major challenges to achieving a person-centred health system.
•
Evidence points to “person-centredness” being the right driver for a complex adaptive health
system
•
Taking a CAS theory approach, the main challenge from the whole of system perspective relates to
“how to transform the values that sustain the current system”.
•
System change requires policy leaders to engage all stakeholders in defining the right driver for the
health system
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 40
Appendix I: Complexity sciences
conceptual framework
In this appendix we quote large sections from key texts that, in our opinion, provide the most concise, yet
comprehensive foundations to complex adaptive systems theory.
The philosophy of Complex Adaptive Systems – Paul Cilliers
Excerpt from Cilliers 2013, p. 30 (94). References and footnotes from the original have been removed.
The notion “complexity” has been used in a somewhat general way, as if we know what the word means.
According to conventional academic practise it would now be appropriate to provide a definition of
“complexity”. I will nevertheless resist this convention. There is something inherently reductionist in the
process of definition. This process tries to capture the precise meaning of a concept in terms of its essential
properties. It would be self-defeating to start an investigation into the nature of complexity by using exactly
those methods we are trying to criticise! On the other hand, we cannot leave the notion of “complexity”
merely dangling in the air; we have to give it some content. This will be done by making a number of
distinctions which will constrain the meaning of the notion without pinning it down in a final way. The
characterisation developed in this way is thus not final – in specific contexts there may be more
characteristics one could add, and some of those presented here may not always be applicable – but it helps
us to make substantial claims about the nature of complexity, claims that may shift our understanding in
radical ways.
In the first place one should recognise that complexity is a characteristic of a system. Complex behaviour
arises because of the interaction between the components of a system. One can, therefore, not focus on
individual components, but on their relationships. The properties of the system emerge as a result of these
interactions; they are not contained within individual components.
A second important issue is to recognise that a complex system generates new structure internally. It is not
reliant on an external designer. This process is called self-organisation. In reaction to the conditions in the
environment, the system has to adjust some of its internal structure. In order to survive, or even flourish,
the tempo at which these changes take place is vital (see Cilliers, 2007 for detail in this regard). A
comprehensive discussion of self-organisation is beyond the scope of this chapter (see Chapter 6 in Cilliers,
1998 for such a discussion), but some aspects of self-organisation will become clear as we proceed.
An important distinction can be made between “complex” and “complicated” systems. Certain systems may
be quite intricate, say something like a jumbo jet. Nevertheless, one can take it apart and put it together
again. Even if such a system cannot be understood by a single person, it is understandable in principle.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 41
Complex systems, on the other hand, come to be in the interaction of the components. If one takes it apart,
the emergent properties are destroyed. If one wishes to study such systems, examples of which are the brain,
living systems, social systems, ecological systems and social-ecological systems, one has to investigate the
system as such. It is exactly at this point that reductionist methods fail.
One could argue, however, that emergence is a name for those properties we do not fully understand yet.
Then complexity is merely a function of our present understanding of the system, not of the system itself.
Thus one could distinguish between epistemological complexity – complexity as a function of our
description of the system – and ontological complexity – complexity as an inherent characteristic of the
system itself. Perhaps, the argument might go, all complexity is merely epistemological, that finally all
complex systems are actually just complicated and that we will eventually be able to understand them
perfectly.
If one follows an open research strategy - a strategy which is open to new insights as well as to its own
limitations - one cannot dismiss the argument above in any final way. Nevertheless, until such time as the
emergent properties of a system are fully understood, it is foolish to treat them as if we understand them
already. Given the finitude of human understanding, some aspects of a complex system may always be
beyond our grasp. This is no reason to give up on our efforts to understand as clearly as possible. It is the
role of scientific enquiry to be as exact as possible. However, there are good reasons why we have to be
extremely careful about the reach of the scientific claims we make. In order to examine these reasons in
more detail, a more systematic discussion of the nature of complex systems is required. The following
characteristics will help us to do this:
1.
Complex systems are open systems.
2.
They operate under conditions not at equilibrium.
3.
Complex systems consist of many components. The components themselves are often simple (or can
be treated as such).
4.
The output of components is a function of their inputs. At least some of these functions must be
non-linear.
5.
The state of the system is determined by the values of the inputs and outputs.
6.
Interactions are defined by actual input-output relationships and these are dynamic (the strength of
the interactions change over time).
7.
Components, on average, interact with many others. There are often multiple routes possible
between components, mediated in different ways.
8.
Many sequences of interaction will provide feedback routes, whether long or short.
9.
Complex systems display behaviour that results from the interaction between components and not
from characteristics inherent to the components themselves. This is sometimes called emergence.
10.
Asymmetrical structure (temporal, spatial and functional organisation) is developed, maintained
and adapted in complex systems through internal dynamic processes. Structure is maintained even
though the components themselves are exchanged or renewed.
11.
Complex systems display behaviour over a divergent range of timescales. This is necessary in order
for the system to cope with its environment. It must adapt to changes in the environment quickly,
but it can only sustain itself if at least part of the system changes at a slower rate than changes in
the environment. This part can be seen as the ‘memory’ of the system.
12.
More than one legitimate description of a complex system is possible. Different descriptions will
decompose the system in different ways and are not reducible to one another. Different
descriptions may also have different degrees of complexity.
If one considers the implications of these characteristics carefully a number of insights and problems arise:
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 42
•
The structure of a complex system enables it to behave in complex ways. If there is too little
structure (i.e. many degrees of freedom), the system can behave more randomly, but not more
functionally. The mere ‘capacity’ of the system (i.e. the total amount of degrees of freedom available
if the system was not structured in any way) does not serve as a meaningful indicator of the
complexity of the system. Complex behaviour is possible when the behaviour of the system is
constrained. On the other hand, a fully constrained system has no capacity for complex behaviour
either. This claim is not quite the same as saying that complexity exists somewhere on the edge
between order and chaos. A wide range of structured systems display complex behaviour.
•
Since different descriptions of a complex system decompose the system in different ways, the
knowledge gained by any description is always relative to the perspective from which the
description was made. This does not imply that any description is as good as any other. It is merely
the result of the fact that only a limited number of characteristics of the system can be taken into
account by any specific description. Although there is no a priori procedure for deciding which
description is correct, some descriptions will deliver more interesting results than others.
•
In describing the macro-behaviour (or emergent behaviour) of the system, not all the micro-
features can be taken into account. The description on the macro-level is thus a reduction of
complexity, and cannot be an exact description of what the system actually does. Moreover, the
emergent properties on the macro-level can influence the micro-activities, a phenomenon
sometimes referred to as “top-down causation”. Nevertheless, macro-behaviour is not the result of
anything else but the micro-activities of the system, keeping in mind that these are not only
influenced by their mutual interaction and by top-down effects, but also by the interaction of the
system with its environment. When we do science, we usually work with descriptions which
operate mainly on a macro-level. These descriptions will always be approximations of some kind.
These insights have important implications for the knowledge-claims we make when dealing with complex
systems. Since we do not have direct access to the complexity itself, our knowledge of such systems is in
principle limited. The problematic status of our knowledge of complexity needs to be discussed in a little
more detail. Before doing that, some attention will be paid to three problems: identifying the boundaries of
complex systems, the role of hierarchical structure and the difficulties involved in modelling complexity.
Tackling the most difficult questions? – David Krakauer
Excerpt from Krakauer 2015. (95)
One quite useful distinction that one can make is between the merely complicated and the complex. So the
universe is complicated in many parts; the sun is complicated, but in fact I can represent in a few pages of
formula how the sun works. We understand plasma physics; we understand nuclear fusion; we understand
star formation.
Now, take an object that’s vastly smaller. A virus, Ebola virus. Got a few genes. What do we know about it?
Nothing. So how can it be that an object that we’ll never get anywhere close to, that’s vast, that powers the
Earth, that is responsible in some indirect way for the origin of life, is so well understood, but something
tiny and inconsequential and relatively new, in terms of Earth years, is totally not understood? And it’s
because it’s complex, not just complicated. And what does that mean?
So one way of thinking about complexity is adaptive, many body systems. The sun is not an adaptive system;
the sun doesn’t really learn. These do; these are learning systems. And we’ve never really successfully had a
theory for many body learning systems. So just to make that a little clearer, the brain would be an example.
There are many neurons interacting adaptively to form a representation, for example, of a visual scene; in
economy, there are many individual agents deciding on the price of a good, and so forth; a political system
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 43
voting for the next president. All of these systems have individual entities that are heterogeneous and
acquire information according to a unique history about the world in which they live. That is not a world
that Newton could deal with. There’s a very famous quote where he says something like, I have been able to
understand the motion of the planets, but I will never understand the madness of men. What Newton was
saying is, I don’t understand complexity.
So complexity science essentially is the attempt to come up with a mathematical theory of the everyday, of
the experiential, of the touchable, of the things that we see, smell and touch, and that’s the goal. Over the
last 10, 20 years, a series of mathematical frameworks—a little bit like the calculus or graph theory or
combinatorics in mathematics that prove so important in physics—have been emerging for us to understand
the complex system, network theory, agent-based modeling, scaling theory, the theory of neutral networks,
non-equilibrium statistical mechanics, non-linear dynamics. These are new, and relatively, I mean on the
order of decades instead of centuries; and so we’re at a very exciting time where I think we’re starting to
build up our inventory of ideas and principles and tools. We’re starting to see common principles of
organization that span things that appear to be very different—the economy, the brain, and so on. So
complexity science ultimately seeks unification—what are the common principles shared—but also provides
us with tools for understanding adaptive, many body systems. And intelligence for me is in some sense, the
prototypical example of an adaptive, many body system.
A general description of systems (nonlinear systems) – Russ Ackoff
Excerpt from Ackoff 1994 (96)
What’s a system? A system is a whole that consists of parts each of which can affect its behaviour or its
properties. Each part of the system, when it affects the system, is dependent for its effect on some other part,
the parts are interdependent; no part of the system or collection of parts of the system has an independent
effect on it. Therefore a system as a whole cannot be divided into independent parts. This has some very
important implications that are generally overlooked. First the essential or defining properties of many
systems are properties of the whole which none of its part has. The performance of a system depends on
how the parts fit, not how they act taken separately.
Nominal Definition – Kevin Dooley
Excerpt from Dooley, 1996, p. 2-3 (97).
The basic elements of a CAS are agents. Agents are semi-autonomous units that seek to maximize their
fitness by evolving over time. Agents scan their environment and develop schema. Schema are mental
templates that define how reality is interpreted and what are appropriate response for a given stimuli. These
schemas are often evolved from smaller, more basic schema. These schemas are rational bounded: they are
potentially indeterminate because of incomplete and/or biased information; and they differ across agents.
Within an agent, schema exist in multitudes and compete for survival via a selection-enactment-retention
process.
When an observation does not match what is expected, an agent can take action in order to adapt the
observation to fit an existing schema. An agent can also purposefully alter schema in order to better fit the
observation. Schema can change through random or purposeful mutation, and/or combination with other
schema. When schema change it generally has the effect of making the agent more robust (it can perform in
light of increasing variation or variety), more reliable (it can perform more predictably), or more capable in
terms of its requisite variety (in can adapt to a wider range of conditions).
The fitness of the agent is a complex aggregate of many factors, both local and global. Unfit agents are more
likely to instigate schema change. Optimization of local fitness allows differentiation and novelty/diversity;
global optimization of fitness enhances the CAS coherence as a system and induces long term memory.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 44
Schema define how a given agent interacts with other agents surrounding it. Actions between agents
involve the exchange of information and/or resources. These flows may be nonlinear. Information and
resources can undergo multiplier effects based on the nature of interconnectedness in the system. Agent tags
help identify what other agents are capable of transaction with a given agent; tags also facilitate the
formation of aggregates, or meta-agents. Meta-agents help distribute and decentralize functionality,
allowing diversity to thrive and specialization to occur. Agents or meta-agents also exist outside the
boundaries of the CAS, and schema also determine the rules of interaction concerning how information and
resources flow externally.
Structure and dynamics of CAS are interdependent – Fritjof Capra
Fritjof Capra (physicist) (98) uses the
vortex
as the prototypical example to illustrate the structure and
dynamics of a CAS. Three aspects are of note:
• for a vortex to form and maintain itself it requires a focal point;
• every level within the vortex has unique dynamics; and
• following disturbances to the structural form of the vortex it will
re-establish itself autonomously through self-organisation as long
as the focal point is maintained.
These features are of particular importance to understanding health
systems and their dynamics.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 45
Appendix II – Cynefin
Framework
Kurtz and Snowden (11) developed the Cynefin framework to classify these dynamic patterns according to
the cause and effect relationships between agents – they can be tightly coupled, more or less loosely coupled
or entirely decoupled (Figure 4).
Tightly coupled cause and effect relationships produce highly predictable outcomes, the simple domain
where things are clearly known. Cause and effect relationships which include time delays result in
outcomes knowable to experts and define the complicated domain. The complex domain is defined by cause
and effect relationships that can only be understood in retrospect, and situations that have no obvious signs
of cause and relationship belong to the chaos domain.
Agents in human systems have unique identities and are able to change their behaviours in light of changing
circumstances, individually and/or collectively, i.e. they have adaptive capacities. Kurtz and Snowden (11)
developed the Cynefin model which defines the characteristics of systems based on the components’
relationship as “simple”, “complicated”, “complex” and “chaotic” and their relationships to each other. The
central space of “disorder” signifies those issues that require clarification through collective sense-making.
This model highlights that we can approach our understanding about a problem from various perspectives,
all of which are providing some insights but none of which is exclusively describing the whole in its
entirety. In addition, the Cynefin model allows a visual representation of the transitions between linear
(greater concern with content) and non-linear relationships (greater concern with context) between system
components, and the related degree of certainty (that which can be taught) and uncertainty (that which
needs to be learned) arising from their interactions
Two examples highlight the benefits of using the Cynefin framework in the medical context.
Structure and function in health and disease
Understanding structure and function in health and disease. Cells are well understood, and the detailed
structure and function of organs (a collective of cells) is well understood by experts. However, the
variability of organ function in health and disease is much less clearly understood as many internal and
external factors impact on the organised cellular function of the whole body. In acute disease there may be
dissociation of function and structure within and between organs that baffles even the most experienced
clinician (See figure below).
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 46
Figure 4: Application of Cynefin Framework health and disease
The culture of safety
Safety reflects the function of the system as a whole. While training and standard procedures are
prerequisite for
doing things right
,
doing the right thing
requires collaboration and adaptation in light of
changing circumstances. Breakdown of collaboration invariably results in adverse outcomes (See figure
below).
Figure 5: Application of Cynefin Framework to culture of safety
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 47
Appendix III – Case study of a
complex adaptive (health) system
– the NUKA system
Background
Southcentral Foundation is an Alaska Native-owned, nonprofit health care organization serving nearly
65,000 Alaska Native and American Indian people living in Anchorage, Matanuska-Susitna Valley, and
55 rural villages in the Anchorage Service Unit.
Southcentral Foundation NUKA System of Care (https://www.southcentralfoundation.com/nuka/) is a name
given to the whole health care system created, managed and owned by Alaska Native people to achieve
physical, mental, emotional and spiritual wellness.
Nuka is an Alaska Native word used for strong, giant structures and living things. The relationship-based
NUKA-System of Care is comprised of organizational strategies and processes; medical, behavioral, dental
and traditional practices; and supporting infrastructure that work together – in relationship – to support
wellness. By putting relationships at the forefront of what we do and how we do it, the NUKA-System will
continue to develop and improve for future generations.
Vision
A Native Community that enjoys physical, mental, emotional and spiritual wellness.
Key Points
• History
o For 50 years Alaskan health services were hospital based and run and controlled centrally by a
large bureaucracy from Washington
o Patients were treated as “beneficiaries” – weeks to get an appointment, emergency department
became default access point, no continuity of provider
o Staff were excluded from innovating service organisation and deliver
• Problems
o Disconnect between care for the mind and care for the body
o Departments and programmes acted independently
o Unhappy patients and unhappy staff
o Poor health statistics
• Change – 1998
o Southcentral Foundation, owned by Alaska Native people, takes over services for Alaska Native
people
o Community survey to elicit values, priorities and needs [
purpose, goals and values
]
§ Relationship with primary care provider
§ Being treated with courtesy, respect and cultural understanding
§ Access to care when needed
o Redesign of health services based on the community’s values and needs
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 48
§ Vision Statement
• A Native Community that enjoys physical, mental, emotional and spiritual
wellness.
§ Mission Statement
• Working together with the Native Community to achieve wellness through
health and related services.
§ 3 ‘‘key points’’ (or simple rules/operating principles)
• shared responsibility, commitment to quality and family wellness.
o Continuous improvement
§ listen to people’s feedback and understand their needs AND explain the changes being
made in response to their feedback AND then communicate the organisation’s
successes in delivering what the community asked of it
Table 3: CAS features in the NUKA system
What part of the system
Primary care system
Context
Rebuilding a person-centered culturally appropriate primary care system
The agents
NUKA community, Southcentral Foundation, doctors, nurses, allied health
professionals, social and community workers
Driver
shared responsibility, commitment to quality and family wellness
CAS-properties
Non-linearity
• Change to appointment system allows ready access to health service
• Change to appointment system allows continuity of care
Open to environment
• Major focus on social factors impacting on health and healthcare
delivery
Self-organisation
• Identification of key features for a new health service results in
reorganisation of the system’s structure and staff behaviours
Emergence
• Health centres transform to meet cultural needs and expectations
Pattern of interaction
• Guided by the three core principles of the organisation - shared
responsibility, commitment to quality and family wellness
Adaptation and Evolution
• User feedback guides improvement programmes
Co-evolution
• Health centres become community hubs and meeting centres
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 49
Appendix IV: Roles and
responsibilities of agents at
system levels
Excerpt from Martin and Sturmberg, 2006. (80)
(Note: references that were in the original document have been removed from this excerpt. Please see
original for full details).
Macro or policy-level:
Policy and financial frameworks need to address population needs as well as the
needs of vulnerable groups. The principle of optimal health for all citizens is central to policy innovations in
any model of primary health care. Currently prescriptive, “top down”, hierarchical and linear policy
approaches predominate. In “bottom up” approaches, general practice may advocate for patients and lobby
for strategies that provide considered multimodal frameworks in which all stakeholders work together to
develop locally appropriate solutions.
Meso or organizational and local-level:
Addressing health needs and health related determinants at a
regional/local level requires coordinated responses from both health providers and administrators. In order
to facilitate the evolution of new, locally relevant service models, it is important to allow key stakeholders
to operate in an open rather than heavily prescriptive planning environment. For example, the general
practitioner/family physician, thus has a developing organizational and knowledge brokerage role in
interdisciplinary and intersectoral care, and in the uptake of new technologies, while at the same time
maintaining the core principles of personalized care delivery. This requires the translation of research
knowledge, ensuring patients’ equity and access to timely health care, and the sharing and coordinating of
health care between the wide range of health care and non-health agencies.
Micro or individual-level:
Patients and their communities are the centre point around which care is
provided and organized. The effectiveness of the roles and responsibilities of general practice rests in the
consultation and the personalized interaction of the doctor/provider with an individual. The consultation is
the basic “production unit” in medicine – here decisions about resource consumption are negotiated
between the doctor and the patient. Yet roles and responsibilities in this area are evolving with care
delegation, new patient expectations, electronic information systems and internet medicine. Crucially there
is an increasing advocacy and leadership role to keep the patient (not a disease, a cost or a multidisciplinary
team) central to the health system and to ensure their core care remains continuous, coordinated,
relationship-based and located in primary health care.
Nano or organismic-level:
The level of health – subjective as well as objective – reflects the entire impact of
the forces influencing human health. Health perception, that is, the subjective experience of health or
disease, is the result of the person’s interdependence (a term coined by Ban-Yar) with his/her environment.
In other words, a patient’s experience of healthcare is an outcome reflecting the effectiveness of
consultations – nature and nurture, and the workings of the health system at large. However, in the end it is
the organism and its embodied experiences of mind, body and emotion that we label “health”. This is where
health care is directed and has its raison d’être. The judgement of primary health care success is ultimately
located at this level. Increasingly this is where the role and responsibility of general practice lies.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 50
Table 4: Agents and their interests in the health system
Level
Agents of Influence
Agents’ Interests
macro
• Government Policymakers
o Health
o Social services
o Social Infrastructure
o Environment
o Economics and finance
o others – education,
work & employment,
housing etc
• Private Enterprise
o Pharmaceutical industry
o Device makers
o Medical associations
o Health insurance
industry
• Citizen Lobby Groups
o Health User Forum
o Disease-specific support
groups
• Non-Government Organisations
o Research Councils
o
• Resource Allocation
o Determined by perceived priorities
o Has financial control over health
system
o Balanced Budget
• Market Share and Profits
o Getting new drugs developed and
accepted on formularies
o Financial interest of members
o Growing membership and market
share from public health system
• Getting greater resources for their specific
interests
• Getting greater resources for their specific
interests
meso
• Local community
infrastructure/environment
o Work
o Education
o Housing
o Roads
o Social infrastructure
o Open spaces
o Others
• Public hospital care
o Hospital departments
o Community outreach
services
• Private hospital care
• Dependent on cooperation with other interests
• Resource constraints
• Focused on specific tasks
• Shifting priorities with shifting government
agendas
• Resource constraints
o Compartmentalised according to
organ-system or technology
o Unstable workforce
o Staffing shortages
o High level of bureaucracy
o Performance based on throughput
• Return on investments
o Customer focus: doctors and specialists
o Performance based on maximizing
revenue per patient day
micro
• Health service delivery
• Primary
o GP-practice team, incl.
reception staff, nurses,
psychologists,
• Private enterprise concerns
o FFS-system of remuneration
o Competition between practices
o Resourcing according to income
generation potential
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 51
indigenous health
workers, others
• Pathology/Radiology
• Specialist
• Community
o Community nursing
o Physiotherapy
o Psychology
o Other allied health
professionals
o Over-servicing incentive
o Time = money, referral an easy option
o Fragmentary care
o Limited liaison with other health
professional providers
o Limited evaluation of health outcomes
micro
• Family, friends and social
networks
• Financial constraints
• Limited knowledge about patient care and
support
• Difficulties accessing community support
services
nano
• The person
• Concerned about their health experience, does
it limit desired levels of activity
• Safety of self-management
• Financial constraints
• Difficulties accessing community support
services
Appendix V: Members of Expert
Panel
Scholem Glouberman
Trisha Greenhalgh
Tim Holt
Holly J. Lanham
Carmel Martin
Di O’Halloran
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 52
References
1. Plesk P. Redesigning Health Care with Insights from the Science of Complex Adaptive Systems. 2001. In:
Crossing the Quality Chasm: A New Health System for the 21st Century [Internet]. Washington, D.C.: National
Academy Press; [309-22]. Available from: https://www.nationalacademies.org/hmd/Reports/2001/Crossing-the-
Quality-Chasm-A-New-Health-System-for-the-21st-Century.aspx.
2. Bentley TG, Effros RM, Palar K, Keeler EB. Waste in the U.S. Health care system: a conceptual framework. The
Milbank quarterly. 2008;86(4):629-59.
3. Australian Government. Australia: The healthiest country by 2020 In: Preventative Health Taskforce, editor.
Canberra, Australia: Commonwealth of Australia 2009.
4. Organisation for Economic Co-operation and Development (OECD). Health at a Glance. 2015 OECD Indicators,
Key Findings Australia. 2015 [updated 15 March 2016. Available from: http://www.oecd.org/australia/Health-at-
a-Glance-2015-Key-Findings-AUSTRALIA.pdf.
5. Potter B. "Heart-Sink" Patients: Tech Pioneers Battle Healthcare Inertia. Australian Financial Review. 2016 20-
21 February 2016.
6. Adams K, Hester P, Bradley J, Meyers T, Keating C. Systems theory as the foundation for understanding systems.
Systems Engineering. 2014 17(1):112-23.
7. Salvador-Carulla L. Informed Mental Health Decisions - A system approach to mental health planning.
Canberra Health Network: Primary mental health strategy reform group; Canberra2016.
8. Sadsad R, McDonnell G. Multiscale modelling for public health management: A practical guide. Discrete-Event
Simulation and System Dynamics for Management Decision Making: John Wiley & Sons Ltd; 2014. p. 280-94.
9. Hawes P, Bond L, Butler H. Knowledge theories can inform evaluation practice. What can a complexity lens
add? New Dir Eval 2009;124:89-100.
10. Hawe P. Lessons from complex interventions to improve health. Annual review of public health. 2015;36:307-
23.
11. Kurtz CF, Snowden DJ. The new dynamics of strategy: Sense-making in a complex and complicated world. IBM
Systems Journal. 2003;42(3):462-83.
12. Gottlieb K. The Nuka System of Care: improving health through ownership and relationships. International
Journal of Circumpolar Health. 2013;72:10.3402/ijch.v72i0.21118.
13. Begun JW, Zimmerman B, Dooley K. Health Care Organizations as Complex Adaptive Systems. In: Mick SM,
Wyttenbach M, editors. Advances in Health Care Organization Theory. San Francisco: Jossey-Bass; 2003. p. 253-
88.
14. Collins B. Intentional whole health system redesign. Southcentral Foundation’s ‘Nuka’ system of care. London:
King’s Fund, 2015.
15. Kaplan RS, Porter ME. How to solve the cost crisis in health care. Harvard Business Review. 2011;89(9):47-64.
16. Sturmberg JP, O'Halloran DM, Martin CM. Understanding health system reform – a complex adaptive systems
perspective. Journal of Evaluation in Clinical Practice. 2012;18(1):202-8.
17. Freedman LP, Schaaf M. Act global, but think local: accountability at the frontlines. Reproductive Health
Matters. 2013;21(42):103-12.
18. Bloom G, Wolcott S. Building institutions for health and health systems in contexts of rapid change. Social
Science & Medicine. 2013;96:216-22.
19. Barret R. The values-driven organization: Unleashing human potential for performance and profit. London and
New York: Routeledge 2013.
20. Eisenhardt KM, Sull DN. Strategy as Simple Rules. Harvard Business Review. 2001;79(1):107-16.
21. Kottke TE. Simple Rules That Reduce Hospital Readmission. The Permanente Journal. 2013;17(3):91-3.
22. Gilson L, Elloker S, Olckers P, Lehmann U. Advancing the application of systems thinking in health: South
African examples of a leadership of sensemaking for primary health care. Health Research Policy and Systems.
2014;12(1):30.
23. Plsek P. Appendix B: Redesigning Health Care with Insights from the Science of Complex Adaptive Systems. In:
Committee on Quality of Health Care in America - Institute of Medicine, editor. Crossing the Quality Chasm: A
New Health System for the 21st Century Washington, D.C.: National Academy Press; 2001. p. 309-22.
24. Seelos C, Mair J. Innovation Is Not the Holy Grail. Stanford Social Innovation Review. 2012;10(4):44-9.
25. Frenk J. The Global Health System: Strengthening National Health Systems as the Next Step for Global Progress.
PLoS Med. 2010;7(1):e1000089.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 53
26. Mintzberg H. Cover Leadership: Notes on Managing Professionals. Knowledge workers respond to inspiration
not supervision. Harvard Business Review. 1998;76(6):140-7.
27. Heifetz R. Leadership Without Easy Answers. Cambridge, Ma: Harvard University Press; 1994.
28. Heifetz R, Grashow A, Linsky M. The Practice of Adaptive Leadership: Tools and Tactics for Changing Your
Organization and the World. Boston, MA: Harvard Business Press; 2009.
29. Pye A. Leadership and Organizing: Sensemaking in Action. Leadership. 2005;1(1):31-49.
30. Fulop L, Mark A. Leading in healthcare - foregrounding context: The theory and practice of context –
Introduction to the Special Issue. Leadership. 2013;9(2):151-61.
31. Padula W, Duffy M, Yilmaz T, Mishra M. Integrating systems engineering practice with health-care delivery.
Health Systems 2014;3:159-64.
32. Franca V, Lima H. Systems Engineering and Quality Engineering: a systematic mapping study Australian Journal
of Basic and Applied Sciences 2015;9(11):990-4.
33. President's Council of Advisors on Science and Technology (PCAST). Report to the President. Better health care
and lower costs: accelerating improvement through systems engineering. In: President EOot, editor. Washington
DC. : President's Council of Advisors on Science and Technology; 2014.
34. Ravitz A, Sapirsteing A, Pham J, Doyle P. Systems Approach and Systems Engineering Applied to Health Care:
Improving Patient Safety and Health Care Delivery. John Hopkins APL Technical Digest. 2013;31(4):354-65.
35. Gibert K, Garcia-Alonso C, Salvador-Carulla L. Integrating clinicians, knowledge and data: expert-based
cooperative analysis in healthcare decision support. Health Res Policy Syst. 2010;8:28.
36. García-Alonso CR, Campoy-Muñoz P, M S-O. A Multi-Objective Evolutionary Algorithm for enhancing
Bayesian Networks hybrid-based modelling Computers and mathematics with applications. 2016; (in press).
37. Group Health’s MacColl Institute for Healthcare Innovation ipwRatCHCSNI. Integrating Chronic Care and
Business Strategies in the Safety Net. . In: Agency for Healthcare Research and Quality (AHRQ), editor.
Rockville, MD: AHRQ; 2008.
38. Scoville R, Little K. Comparing Lean and Quality Improvement. IHI White Paper. . In: Institute for Healthcare
Improvement (IHI), editor. Cambridge. Massachusetts2014. p. 12-3.
39. Hargreaves MB. Evaluating System Change: A Planning Guide. Methods Brief. Mathematica [Internet]. 2010
Accessed 28 February 2016. Available from: http://www.mathematica-
mpr.com/~/media/publications/PDFs/health/eval_system_change_methodbr.pdf.
40. Haque W, Urquhart B, Berg E, R. D. Using business intelligence to analyze and share health system
infrastructure data in a rural health authority. JMIR, Med Inform. 2014;2(2):e16.
41. Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O. Diffusion of innovations in service organizations:
systematic review and recommendations.
Milbank Quarterly
. 2004;82:581-629.
42. Best A, Holmes B. Systems Thinking Knowledge and Action: towards better models and action. Evidence and
Policy. 2010;6(2):145-59.
43. Hudson P. Applying the lessons of high risk industries to health care. Qual Saf Health Care 2003;12 (Suppl
1):p.i7-12.
44. Winnefeld Jr. JA, Kirchhoff C, Upton DM. Cybersecurity’s human factors: lessons from the Pentagon. Harvard
Business Review. 2015;September:87-95.
45. Hines S, Luna K, Lofthus J, Marquardt M, Stelmokas D. Becoming a high reliability organization: Operational
advice for hospital leaders. In: Agency for Healthcare Research and Quality (AHRQ), editor. Rockville: AHRQ;
2008.
46. National Patient Safety Foundation. mproving Root Cause Analyses and Actions to Prevent Harm Boston: NPSF,;
2015 [Available from: http://c.ymcdn.com/sites/www.npsf.org/resource/resmgr/PDF/RCA2_first-online-
pub_061615.pdf.
47. Li J, Boulanger B, Norton J, Yates A, Swartz CH, Smith A, et al. "SWARMing" to Improve Patient Care: A Novel
Approach to Root Cause Analysis,. Journal Quality Patient Safety,. 2015;Nov 4(11):494-501.
48. Larizgoitia I, Charlotte Bouesseau M, Kelley E. WHO efforts to promote reporting of dverse events and global
learning. Journal of Public Health Research. 2013;2(e29).
49. Sheppard F, Williams M, Klein V. TeamSTEPPS and patient safety in healthcare. J Healthc Risk Manag,.
2013;32(3):5-10.
50. Power DJ. A Brief History of Decision Support Systems., version 4.0, : DSSResources.COM, World Wide Web,;
2007 [Available from: http://DSSResources.COM/history/dsshistory.html.
51. Homer J, Wile K, Yarnoff B, Trogdon J, Hirsch G, Cooper L. Using simulation to compare established and
emerging interventions to reduce cardiovascular disease riks in the United States. Prev Chronic Dis
2014;11:140130.
52. Atkinson J-A, Page A, Wells R, Milat A, Wilson A. A modelling tool for policy analysis to support the design of
efficient and effective policy responses for complex public health problems. Implementation Science.
2015;10(1):1-9.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 54
53. Marshall D, Burgos-Liz L, Ijzerman M, Osgood N, Padula W, Higashi M, et al. Applying dynamic simulation
modeling methods in health care delivery research-the SIMULATE checklist: report of the ISPOR simulation
modeling emerging good practices task force. Value in Health. 2015;18(1):5-16.
54. Röckmann C, van Leeuwenb J, Goldsborough C, Kraana M, Pieta G. he interaction triangle as a tool for
understanding stakeholder interactions in marine ecosystem based management. Marine Policy. 2015;52:155-62.
55. Thigpen S, Puddy RW, Singer HH, Hall DM. Moving knowledge into action: developing the rapid synthesis and
translation process within the interactive systems framework. Am J Community Psychol,. 2012;50(3-4):285-94.
56. Wilkins N, Thigpen S, Lockman J, Mackin J, Madden M, Perkins T, et al. Putting program evaluation to work: a
framework for creating actionable knowledge for suicide prevention practice. Transl Behav Med. 2013;June
3,(2):149-61.
57. Greenhalgh T, Wong G, Jagosh J, Greenhalg J, Manzano A, Westhorp G, et al. Protocol—the RAMESES II study:
developing guidance and reporting standards for realist evaluation. BMJ Open. 2015;e008567.
58. Boulos MN, Roudsari AV, Carson ER. Health geomatics: an enabling suite of technologies in health and
healthcare. J Biomed Inform,. 2001;June 34,(3):195-219.
59. Bate P. Context is everything. . 2014. In: Perspective in Context [Internet]. London; [1-30]. Available from:
http://www.health.org.uk/sites/default/files/PerspectivesOnContext_fullversion.pdf.
60. Salvador-Carulla L, Symonds S. Health services use and costs in people with intellectual disability: building a
context knowledge base for evidence-informed policy. . Curr Opin Psychiatry,. 2016;Mar;29(2):89-94.
61. Lunsky Y, Klein-Geltink JE, Yates E, editors. Atlas on the Primary Care of Adults with Developmental
Disabilities in Ontario. Toronto2013.
62. Fernandez A, Salinas-Perez JA, Gutierrez-Colosia MR, Prat-Pubill B, Serrano-Blanco A, Molina C, et al. Use of
an integrated Atlas of Mental Health Care for evidence informed policy in Catalonia (Spain). Epidemiol
Psychiatr Sci. 2015;24(6):512-24.
63. Salvador-Carulla L, Amaddeio F, Gutierres-Colosia M, Salazzari D, Gonzalez-Callaro J, Montagni I, et al.
Developing a tool for mapping adult mental health care provision in Europe: The REMAST research protoocol
and its contribution to better integrated care. International Journal of Integrated Care. 2015;Oct-Dec.
64. Walker RE, Keane CR, Burke JG. Disparities and access to healthy food in the United States: A review of food
deserts literature. Health & Place. 2010;16(5):876-84.
65. Chen DTMDMPH, Werhane PHP, Mills AEMMBA. Role of organization ethics in critical care medicine. Critical
Care Medicine Organizational and Management Ethics in the Intensive Care Unit. 2007;35(2):S11-S7.
66. Sturmberg JP, O’Halloran DM, Martin CM. People at the centre of complex adaptive health systems reform.
Medical Journal of Australia. 2010;193(8):474-8.
67. Sturmberg JP, O'Halloran DM, Martin CM. Health Care Reform – The Need for a Complex Adaptive Systems
Approach. In: Sturmberg JP, Martin CM, editors. Handbook of Systems and Complexity in Health. New York:
Springer; 2013. p. 827-53.
68. Wilson R, Rowan MS, Henderson J. Core and comprehensive health care services: 1. Introduction to the
Canadian Medical Association´s decision-making framework. . Canadian Medical Association Journal,.
1995;152:1063-7.
69. Berwick DM. What 'patient-centered' should mean: confessions of an extremist. Health Aff (Millwood).
2009;28(4):w555-65.
70. Godfrey MM, Melin CN, Muething SE, Batalden PB, Nelson EC. Clinical microsystems, part 3: transformation of
two hospitals using microsystem, mesosystem, and macrosystem strategies. Joint Commission Journal on Quality
& Patient Safety. 2008;34(10):591-603.
71. Tansella M, A TG. Conceptual framework for mental health services: the matrix model. Psychol Med. 1998
28(3):503-8.
72. Fernandez A, Smith-Merry J, Gillespie J, Feng X, Astell-Burt T, Maas C, et al. The Integrated Mental Health
Atlas of Western Sydney Local Health District: Gaps and Recommendations. . Australian Health Review 2016;in
press (Accepted for publication 1st February 2016), .
73. Australian Commission on Safety and Quality in Health Care. Australian Atlas of Healthcare Variation. In:
ACSQHC, editor. Sydney, Australia Australian Government 2015.
74. Nancarrow SA, Avila C, Chamberlain D, Keast R, Bradbury J, Doran F, et al. Mapping service integration for
primary healthcare patients: Lessons from a regional GP Super Clinic. Australian Primary Health Care Research
Institute, 2015.
75. Latkin C, Weeks M, Glasman L, Galletly C, Albarracin D. A dynamic social systems model for considering
structural factors in HIV prevention and detection. AIDS and behavior. 2010;14(Suppl 2):222-38.
76. Maddox R, Davey R, Cochrane T, Lovett R, van der Sterren A. Study protocol - Indigenous Australian social
networks and the impact on smoking policy and programs in Australia: protocol for a mixed-method prospective
study. BMC Public Health. 2013;13(1):1-13.
APPLYING CAS THINKING TO THE AUSTRALIAN HEALTH SYSTEM | 55
77. Maddox R, Davey R, Lovett R, van der Sterren A, Corbett J, Cochrane T. A systematic review protocol: social
network analysis of tobacco use. Systematic Reviews. 2014;3:85-.
78. Kansagara D, Englander H, Salanitro A, et al. Risk prediction models for hospital readmission: A systematic
review. JAMA. 2011;306(15):1688-98.
79. Frenk J. The Global System: Strengthening national health systems as the next step for global progress Plos
Medicine. 2010;7(1):1-3.
80. Martin CM, Sturmberg JP. Rethinking General Practice - Part 2: Strategies for the Future. Patient-Centred,
Socially and Economically Responsible Primary Care and the Leadership Challenges. Asia Pacific Family
Medicine. 2006;5(5): www.apfmj-archive.com/afm5_2/afm48.pdf
81. Sorrell MF, Belongia EA, Costa J, Gareen IF, Grem JL, Inadomi JM, et al. National Institutes of Health consensus
development conference statement: management of hepatitis B.,. Hepatology. 2009;49((5 Suppl)):S14-S2.
82. Mooney GH, Blackwell SH. Whose health service is it anyway? Community values in healthcare. . Med J Aust.
2004;180(2):76-8.
83. Consumers Health Forum of Australia. Unique and essential: a review of the role of consumer representatives in
health decision-making Canberra2015 [Available from: https://www.chf.org.au/pdfs/chf/Role-of-Reps-Health-
Decision-Making-2015.pdf
84. Lovell K, Bower P, Richards D, Barkham D, Sibbald B, Roberts C, et al. Developing guided self-help for
depression using the Medical Research Council complex interventions framework: a description of the modelling
phase and results of an exploratory randomised controlled trial. BMC Psychiatry. 2008;8(DOI: 10.1186/1471-
244X-8-91).
85. Eysenbach G, Powell J, Englesakis M, Rizo C, 15; SAM. Health related virtual communities and electronic
support groups: systematic review of the effects of online peer to peer interactions BMJ. 2004;15:328((7449):
1166.).
86. Cancer Council NSW and LifeCircle. Cancer Council NSW and LifeCircle, Response to “Palliative care in NSW:
Palliative Care NSW policy statement discussion paper. 2012. [Available from:
http://www.cancercouncil.com.au/wp-content/uploads/2012/02/Submission_Palliative-Care_Jan-2012.pdf.
87. White K, Williams F, Greenberg B. The Ecology of Medical Care. New England Journal of Medicine.
1961;265(18):885-92.
88. Green L, Fryer G, Yawn B, Lanier D, Dovey S. The Ecology of Medical Care Revisited. New England Journal of
Medicine. 2001;344(26):2021-5.
89. Johansen ME, Kircher SM, Huerta TR. Reexamining the Ecology of Medical Care. New England Journal of
Medicine. 2016;374(5):495-6.
90. Mooney G, Blackwell S. Whose health service is it anyway? Community values in healthcare. Medical Journal of
Australia. 2004;180(2):76-8.
91. Lindstrom RR. Evidence-Based Decision-Making in Healthcare: Exploring the Issues through the Lens of
Complex, Adaptive Systems Theory. HealthcarePapers. 2003;3(3):29-35.
92. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. Health Aff (Millwood).
2008;27(3):759-69.
93. Lukersmith S, Huckel Schneider C, Salvador-Carulla L, Sturmberg JP, Wilson A, Gillespie J. Expert
Commentary: What is the State of the Art in Person-centred Care? Report for the Australian Commission on
Safety and Quality in Health Care (ACQSHC), 2016.
94. Cilliers P. Understanding Complex Systems. In: Sturmberg JP, Martin CM, editors. Handbook of Systems and
Complexity in Health. New York: Springer 2013. p. 27-38.
95. Krakauer D. Ingenious: David Krakauer.The systems theorist explains what’s wrong with standard models of
intelligence. In: Paulson S, editor. Nautilis. nautil.us2015.
96. Ackoff R. Speech at the Learning and Legacy of Dr W. Edwards In: Robyns hbCC-MaL, editor. 1994.
97. Dooley K. A nominal definition of complex adaptive systems. The chaos network. 1996;8(1):2-3.
98. Capra F. The Web of Life. London: HarperCollins Publishers; 1996.
