‘What if’ tool explores complex problems
TYPE Prevention Centre News
I was first exposed to the systems thinking literature several years ago and was quickly enticed by its interdisciplinary nature drawing on natural and social sciences, mathematics, engineering, philosophy, organisational theory and computer science.
I was attracted to the suite of quantitative and qualitative methodologies, each with their own advantage in addressing different problems. Mostly I was seduced by its promise to change the way we solve problems; through collaborative mapping of the underlying causes of a complex problem and uncovering of solutions that allow better targeting of limited resources.
But as I tried to explain my newfound enthusiasm for systems thinking to others I became acutely aware of the appalling abstractness of my explanations.
Serendipitously, I then met Dr Geoff McDonnell, physician and engineer, boundary rider, pioneer of multi-method simulation modelling, and generous mentor who convinced me that it was time to stop reading about how to ride a bike and put my feet on the pedals! With my primary interest in methods that might address some of the challenges of designing effective public health policy, Geoff directed me towards simulation modelling.
So what is simulation modelling and what promise does it hold for the prevention of chronic disease?
Simulation modelling has been successfully used in other sectors such as engineering, ecology, defence and business since the 1950s. It is a process of creating computer models that are simplified representations of the real world. These models allow us to map and quantify the underlying causes of disease trends.
Models can be used as a ‘what if’ tool, to examine in a risk-free and low-cost way the likely impact of different interventions or policy options (applied individually or in combination) as well as the comparative cost and health service implications of those options. For chronic disease, this is particularly important because of the many associated individual, social, cultural, economic and environmental risk factors.
Simulation modelling allows us to answer challenging questions in relation to chronic disease prevention such as:
- Which risk factors are more important than others in our context and how can we better target our investments in prevention?
- Where in the course of people’s lives should interventions be targeted to have optimal impact?
- What combination of interventions is likely to be most effective?
- What intensity of investment is required?
- What are the equity implications of a proposed policy?
Research evidence alone is often not able to answer all of these challenging questions. While in the past we have focused on how to get more evidence into public health policy, more recently we’ve started to ask a slightly different question: how can we put available evidence, data and expert knowledge to better use so that it can answer the challenging questions policy makers face.
This is where simulation modelling can be useful. It integrates diverse evidence sources into an analytic tool that can allow policy makers to systematically and rigorously explore the likely impact of different policy scenarios.
For example, health planners in Austin, Texas, in collaboration with the Centers for Disease Control and Prevention, the National Institutes of Health and others produced a system dynamics model that simulated some of the more significant processes driving chronic disease risk and prevention at a local level.
Their model allowed them to explore various intervention options (including improved access to primary care, junk food taxes, tobacco sales restrictions, social marketing for choices) and was used as a catalyst to engage stakeholders from public health, health care and not-for-profit groups, businesses and schools to establish policy priorities for prevention of chronic disease. Interacting with the model allowed stakeholders to see the likely consequences of policy scenarios in a convincing way, which encouraged them to align their own organisation’s strategies with the policy solutions. (Links to other examples not limited to the health sector can be found at Systems Dynamics Case Repository and AnyLogic.)
At the Prevention Centre, we think simulation modelling has the potential to advance our work, particularly in supporting strategic policy decisions for the prevention of lifestyle-related chronic disease. For example we’re developing a simulation model of alcohol use in Australia that will be used to simulate the effectiveness of a variety of approaches, both individually and in combination, to reducing alcohol-related harm. The model will address the problem of binge drinking as well as the long term harms from high average consumption.
We are also exploring how our projects can contribute to improving our models. For example, we hope that the results of a project developing new methods for the economic evaluation of prevention interventions can be incorporated into our models. We think this will deliver more accurate estimates of the costs and benefits of different policy options, and will better inform investment decisions for chronic disease prevention.
We also see that simulation modelling could be a tool for achieving consensus for public health action in Australia. We are testing a process that embeds modelling in a broader stakeholder engagement and consensus process in the hope that it might reduce resistance to policy decisions and better align stakeholder activities to support implementation of the agreed strategy.
Simulation modelling helps us make the most of the vast amounts of research evidence, data and expertise that already exists, it reduces the costs and risks of trialling policy options in the real world, and it offers promise in delivering to Australians more effective public health policy.