The Governance of Health Systems: Comment on 'A Network Based Theory of Health Systems and Cycles of Well-Being'

Health systems research aims to understand the governance of health systems (i.e. how health systems function and perform and how their actors interact with each other). This can be achieved by applying innovative methodologies and concepts that are going to capture the complexity and dynamics of health systems when they are affected by shocks. The capacity of health systems to adapt to shocks (i.e. the resilience of health systems) is a new area of investigation. Social network analysis is a great avenue that can help measure the properties of systems and analyse the relationships between its actors and between the structure of a health system and the performance of a health system. A new conceptual framework is presented to define the governance of health systems using a resilience perspective.

T he paper written by Michael Grant Rhodes introduces a series of themes that are essential in innovative health system research: the dynamics of system, resilience of systems and the use of social network analysis (SNA) in health systems research (1).
In the field of international health, the dominant approach has been equilibrium thinking which consist of implementing similar and standard programmes in different countries as illustrated by the World Health Organization (WHO) strategy "Health for All" (2). Equilibrium thinking was translated into the rapid 'scaling-up' of pilot interventions, assuming that the success of small-scale projects can be generalised to large-scale programmes (3). The planning of this type of intervention is based on the assumption that all variables, including context and actors, remain unchanged during the entire implementation period (usually three or four years). Hence the generalised use of indicators, linear planning processes and rigid planning tools (for example, logical frameworks) which have been applied by most international donor agencies and non-governmental organisations. The assumption of contextual stability is questionable considering the rapid changes in the environment which modify our social networks and create new health challenges in a more globalised world, such as HIV/AIDS, pollution-related disasters or Avian Influenza (4). Past and current health strategies have been judged inadequate for highly dynamic socio-ecological systems and contexts (5). As suggested by Michael Grant Rhodes, New thinking on complexity and systems is needed to understand complex phenomena, analyse individuals' behaviour, cooperation relationships and dynamics between groups (6-8).
As Grant pointed out, the current frameworks and definitions on health system do not really reflect the interactions between actors and the ongoing adaptations within systems in response to the changing environment. If one looks at the applications generated by these frameworks, one notes that this was a clear response to the increasing demand from international donors for better accountability (9), and the need for appropriate health system performance assessment tools and methods (10). The current WHO definition of a health system was also perceived by some as being a constraining framework which restricted health policy makers to a normative and linear understanding of health systems, and prevented them from searching for innovative and alternative solutions to current complex global health needs and problems (11)(12)(13). In a globalised world, defining a health system has become more and more challenging due to the multiplicity of actors intervening on different scales, and the increasing interactions between global health policies and local health systems. Systems are not abstract entities. Because human actions dominate socio-ecological systems, the adaptability of such systems is mainly a function of the actions and decisions taken by individuals, networks and groups managing these systems (14,15). In other words, the resilience of systems closely relies on the adaptive capacities of actors to respond to stresses and shocks (16). Dynamic systems have various coping mechanisms depending on the nature of events or crises that affect health systems, as referred by Rhodes. In order to distinguish the types of events that affect dynamic systems, Bloom et al. proposed to classify events or crises that affect systems into two different groups: (i) shocks (transient disruptions), and (ii) stresses (enduring pressures) (5). Shocks are atypical and short-term interruptions to the 'normal' pattern of development (e.g. natural disasters, wars, debt) (17)(18)(19), whereas stresses are effects that play out over a far longer time-span than established policies can generally cover (5). Streefland showed that resilience is a key property of systems (20). Streefland described resilience as a permanent transformation and a process of contextualisation and adaptation that is linked to processes of survival and transformation (21,22). Resilience is a measure of the amount of change a system can experience whilst still maintaining the same controls on structure and function (21)(22)(23). This definition portrays resilience as a multidimensional and dynamic process (24). Health systems researchers need to develop tools to measure the level of resilience of health systems.
In health systems research, networks have implicitly been at the heart of health systems (25,26). In their own definition of a health system, Kohn et al. made even more explicit how social networks play a crucial role (27). They saw a health system as a network of actors who aimed to provide health care: "In health care, a system can be an integrated delivery system, a centrally owned multihospital system, or a virtual system comprised of many different partners over a wide geographical area" (27). The field of SNA provides an avenue for analysing and comparing formal and informal information flows in a system (28). Although SNA and health care have long been interconnected, SNA has never been applied to health systems research in lowand middle-income countries, which still remains a nascent field of investigation. SNA has proven, in other areas, to be helpful in understanding the nature of relations between actors within a system and how these relationships influence the structure of a system (29,30). Although applying SNA in systems research is not without its challenges, such as capturing the dynamics of systems and the effects of cross-scale events, innovative approaches have been introduced combining social network theories and other approaches, and can potentially generate new knowledge when applied to health systems (31,32).
To summarise these ideas on governance of systems, Lebel et al. proposed a conceptual framework to describe the main characteristics of "good" governance of social-ecological systems (33). This conceptual framework was adapted to be relevant to health systems. The "good" governance of health systems in relation to resilience is characterised by three main factors: (i) capacity to engage effectively with and handle multiple-and cross-scale dynamics, (ii) capacity to anticipate and cope with uncertainties and surprises, and finally (iii) capacity to combine and integrate different forms of knowledge (See Figure 1).

Ethical issues
Not applicable.
Capacity to engage e ectively with and handle multiple-and cross-scale dynamics Capacity to anticipate and cope with uncertainties and surprises Capacity to combine and integrate di erent forms of knowledge Resilience A measure of the amount of change a system can undergo and still retain the same controls on structure and function Governance of health systems