Modelling the costs and benefits of interventions to prevent and reduce obesity
Key Objective:
To complete the first stage of developing a decision analytic model of interventions designed to reduce the poor health impacts of obesity
There is increasing concern, within Australia as in other countries, that the rising incidence of overweight and obesity will increase the future prevalence of chronic disease, increase premature mortality, and add to the costs of health service delivery. Governments are being lobbied to undertake population level interventions focused on overweight and obesity. However, there is as yet little evidence about the effectiveness and cost-effectiveness of possible interventions, particularly in terms of lifelong health outcomes. Controlled trial evidence of long term outcomes is difficult to accumulate for several reasons: it is not simple to control for all factors which influence an individual’s lifestyle; and long term can mean most of an individual’s lifespan. Consequently, estimating how a lifestyle intervention impacts on long term health outcomes and health care costs requires the development of an appropriate decision analytic model.
A decision analytic model will need to capture causality from the intervention to final health outcomes using the best available clinical or epidemiologic evidence. Modelling lifestyle interventions are complex as the model needs to accurately reflect the following, i) how a change in exercise leads to a change in BMI; ii) which leads to a change in risk factors, such as blood pressure levels and cholesterol; iii) leading to a change in habits that may or may not be sustained; iv) which lead to a long term changes in risk factor profile; v) which result in lower incidence of symptomatic disease; vi) and may result in less severe disease events; vii) and eventually will reduce premature mortality. The effect may vary by population sub-groups such as age and/or sex; and this must be incorporated in the model. Further, lifestyle factors have an impact on many chronic diseases and any one disease or condition is often associated with multiple risk factors. However, models often only focus on one risk factor or one disease, see for example (Liew, et al., 2006). It is obvious from this that the required model will be complex.
The overall objectives are:
• Gather economic model parameter data
• Develop economic model (“gold standard”)
• Build collaborations with other teams
• Further CHERE’s and UTS’s knowledge/reputation
Funding source
UTS Early Career Researcher Grant
CHERE staff
Jody Church, Richard Norman, Stephen Goodall
