Poor diet and obesity have been linked to increased risks for a number of chronic diseases including cancer and cardiovascular diseases. The proposed research will investigate the relationship between food (candy, baked goods, and chips), beverage (soda) and restaurant state sales taxes and food consumption patterns, diet quality, body mass index (BMI) and obesity prevalence among children, adolescents, and adults. The proposed project will build substantially on the previous literature which has linked food prices but not taxes to individual-level population data. These food and beverage sales taxes are on items that are usually high in either sugar and/or fat and given that most of these taxes already exist in a number of states (with substantial variation across states) they are likely to be starting points for potential policy changes. Thus, this research would provide policy makers with evidence on the extent to which implementing or increasing any of these state-level taxes is likely to affect food consumption behavior, diet quality and related BMI and obesity. All analyses will control for the local area availability of food stores and restaurants. We also propose specific analyses of low-income sub-samples and, in particular, we will examine low-income food stamp recipients and non-recipients separately because when purchased by food stamp recipients, such items are exempt from the tax. The specific aims of this project are threefold: 1) Examine the relationship between soda, candy, baked goods, chips and restaurant sales taxes and dietary patterns (intake of taxed food/beverage items and consumption of FAFH) and diet quality (total caloric intake, % fat and/or % sugar in diet, and overall healthy eating index); 2) Examine the relationship between soda, candy, baked goods, chips and restaurant sales taxes and BMI and obesity prevalence; and, 3) Examine the proposed relationships separately for low-income populations and assess the differences in tax sensitivity between low-income food stamp and non-food stamp recipients. To accomplish these aims, the proposed research will conduct secondary data analyses, using a unique combination of four types of data: 1) state-level food, beverage and restaurant sales tax rates; 2) local area outlet density measures of food stores and restaurants; 3) local area socioeconomic status drawn from the Census; and, 4) nationally representative population data on dietary intake and measured BMI for children, adolescents and adults from four data sets including three longitudinal surveys. To our knowledge, the proposed project represents the most comprehensive exploration to date of the contextual relationship between specific food, beverage and restaurant taxes and individuals' dietary patterns, diet quality and BMI. Given the serious public health risk posed by poor diet and obesity prevalence, this research can provide important information for policymakers and public health advocates about the potential effectiveness of implementing or increasing existing tax rates on soda, candy, baked goods, chips and restaurants.

In order to combat the obesity epidemic, it is important to understand the extent to which behaviors that affect the prevalence of obesity are modifiable through various factors. This project offers numerous contributions to the existing public health literature on the effects of economic contextual factors such as food prices/taxes on population diet and weight outcomes. A particularly novel aspect of this project is its comprehensive exploration of the relationship between specific food, beverage, and restaurant tax rates and dietary intake, diet quality and BMI/obesity. The importance of the relationship of these taxes will be examined directly on measures of BMI/obesity and indirectly on BMI/obesity via their effect on intermediary behaviors related to food consumption. The study also will control for local area access to different types of food stores and eating places. We will draw on four nationally representative survey data sets and we will be able to examine child, adolescent, young adult, and adult populations. By linking the economic contextual tax data to NHANES we will be able to examine the association between food, beverage and restaurant taxes and detailed food consumption patterns, diet quality, and weight outcomes. By linking the tax data with the ECLS- K, NLSY97 and PSID longitudinal data sets we will be able to examine changes in weight overtime and changes in a limited number of diet measures among children, adolescents/young adults and adults, respectively. Another important contribution of this study is that it proposes to provide evidence on the extent to which low-income food stamp versus non-food stamp recipients are tax sensitive and how we might expect the food consumption behavior, diet quality, and weight outcomes of food stamp recipients to change if the items bought with food stamps were no longer exempt from the taxes under examination in this study or if other forms of taxes such as excise taxes were introduced that would result in related price increases. This is particularly important if the aim of the tax increase is to change food consumption behavior and related health outcomes for the whole population. Given the serious public health risk posed by poor diet and obesity, the empirical evidence generated by the research proposed in this application can provide important information for policymakers and public health advocates about the potential effectiveness of implementing or increasing existing tax rates on soda, candy, baked goods, chips and restaurants to improve these important health outcomes.