UCLA

A typical American diet is comprised of too much high-calorie foods and insufficient fruits and vegetables. Many theories and models targeting unhealthy dietary behaviors focus exclusively on people as individuals, while behavior theories such as multi-level theory of population health emphasize the social component in human cognitive habits and behaviors, providing an alternative paradigm to understand dietary behaviors. This dissertation incorporates those behavioral theories into an agent-based simulation model (ABM) to generate insights about how an individual makes food choices in the context of social network and external food environment, for the purpose of simulating policy interventions that are potentially effective in changing unhealthy dietary behaviors in an adult population of Los Angeles County.

Chapter 2 describes the model structure, process overview and model settings. Chapter 3 focuses on empirical estimation of model parameters and tests the model using face validation and sensitivity analysis. Data from the 2007 Food Attitudes and Behaviors Survey and other empirical studies are used to estimate model parameters, and data from Los Angeles County Health Survey 2007 are used to validate the model predictions at baseline. Based on the validated model, Chapter 4 contrasts the potential effects of various policies on individuals' dietary decisions using model simulation. The model shows that a 20% increase in taxes on fast foods would lower the probability of fast-food consumption by 3 percentage points, whereas improving the visibility of positive social norms by 10%, either through community-based or mass-media campaigns, could improve the consumption of fruits and vegetables by 7 percentage points and lower fast-food consumption by 6 percentage points. Zoning policies has no significant impact on food consumption in a moderately-dense urban neighborhood since people have easy access to both healthy and unhealthy food outlets. Chapter 5 concludes that interventions emphasizing healthy eating norms have the great potential to create sustainable behavior change, and may be more effective than directly targeting food prices or regulating local food outlets. Agent-based modeling may be a useful tool for testing the population-level effects of various policies within complex systems.