Understanding the dynamics of obesity prevention policy decision-making using a systems perspective: A case study of Healthy Together Victoria

Abstract

Introduction: Despite global recommendations for governments to implement a comprehensive suite of policies to address obesity, policy adoption has been deficient globally. This paper utilised political science theory and systems thinking methods to examine the dynamics underlying decisions regarding obesity prevention policy adoption within the context of the Australian state government initiative, Healthy Together Victoria (HTV) (2011-2016). The aim was to understand key influences on policy processes, and to identify potential opportunities to increase the adoption of recommended policies.

Methods: Data describing government processes in relation to the adoption of six policy interventions considered as part of HTV were collected using interviews (n = 57), document analyses (n = 568) and field note observations. The data were analysed using multiple political science theories. A systematic method was then used to develop a Causal Loop Diagram (CLD) for each policy intervention. A simplified meta-CLD was generated from synthesis of common elements across each of the six policy interventions.

Results: The dynamics of policy change could be explained using a series of feedback loops. Five interconnected balancing loops served to reduce the propensity for policy change. These pertained to an organisational norm of risk aversion, and the complexity resulting from a whole-of-government policy approach and in-depth stakeholder consultation. However, seven virtuous reinforcing loops helped overcome policy resistance through policy actor capabilities that were improved over time as policy actors gained experience in advocating for change.

Conclusion: Policy processes for obesity prevention are complex and resistant to change. In order to increase adoption of recommended policies, several capabilities of policy actors, including policy skills, political astuteness, negotiation skills and consensus building, should be fostered and strengthened. Strategies to facilitate effective and broad-based consultation, both across and external to government, need to be implemented in ways that do not result in substantial delays in the policy process.

Fig 1. Overview of analysis process of the Healthy Together Victoria case study.

Fig 2. Causal loop diagram of the Healthy Together Victoria policy process.

R loops (e.g., R1, R2 etc.) represent ‘reinforcing loops’ that magnify the effect of actors within the loop, including either positive or negative effects. B loops (e.g., B1, B2 etc.) represent ‘balancing loops’, in which feedback acts to stabilise the effect on the system. R loops and B loops were assigned sequential numbering as they were identified. A positive polarity (represented as ‘+’) indicates that as a cause increases, the effect increases, and as cause decreases, the effect decreases. A negative polarity (‘-’) indicates an inverse relationship between the two variables (i.e., as cause increases, the effect decreases). A dash sign (//) indicates some delay in effect, relative to the time scale of the remainder of the diagram. Solid lines indicate that data indicated an association between factors, with triangulation across both data sources and methods. Dashed lines indicate where there was some data to demonstrate an association between factors, however triangulation across data sources or methods could not be achieved. R1 = Policy actor capability to align policy interventions to decision-maker beliefs; R2 = Policy actor capability to align policy interventions to other government objectives; R3 = Stakeholder consultation as part of policy development; R4 = Policy viability; R5 = Evidence of policy impact; R6 = Evidence of implementation feasibility; R7 = ‘Softening up’ of key stakeholders. B1 = Organisational culture of risk aversion; B2 = Time need for effective consultation; B3 = Pressure from competing policy issues; B4 = Involvement of a diverse range of stakeholders; B5 = Time required to conduct a whole-of-government approach.