The empirical evidence currently available regarding which obesity-related policy interventions are most effective, and which behavioral contexts to target, is generally not causally interpretable. A proposed new method is designed to account for potential endogeneity and, therefore, will produce results that are causally interpretable and policy-relevant.
An econometric framework, combined with survey data, can be used to assess the effectiveness of alternative potential policy interventions for improving energy balance and health outcomes. The modeling framework will comprise two levels of obesity-related behavioral causality: (1) modeling energy-balance related lifestyle choices as functions of prospective policy lever variables, and (2) modeling energy balance as a function of the obesity-relevant lifestyle measures from the first component of the model. The model will be used to conduct a comparative effectiveness analysis of alternative policies via a database that matches individuals’ body weights and dietary, exercise, smoking, and drinking habits to county characteristics such as prices, store densities, nutritional education spending, and labor force participation rates. Policy-driven lifestyle outcomes include: carbohydrate consumption, fat consumption, protein consumption, smoking, drinking, and exercise. The results from these two levels of analysis will be combined to estimate the overall effects of policy changes on energy balance.