Doctoral Dissertation Research in Economics: The Health Economics of Food Consumption and Obesity

肥胖已成为美国的主要健康问题。随着肥胖症的日益流行，以及越来越多的医学研究将肥胖症与其他疾病联系起来，了解肥胖症成为健康相关研究的主要关注点。 这种疾病为经济学提供了一个独特的机会，有助于正确理解健康，因为肥胖，比大多数其他疾病，最终是由一系列个人决定，所有消费者必须作出。肥胖，定义为过量的身体脂肪，是热量摄入或食物消耗与热量消耗（包括静息代谢和运动）之间不平衡的最终结果。这项研究旨在研究健康投资和结果与随后的医疗支出之间的联系。此外，为了更好地理解个人决定与肥胖之间的关系，本研究还将研究不同类型食物的需求模式，考虑到习惯的形成。本研究的第一个目标是研究在有健康保险的环境下，与肥胖相关的个人健康投资如何影响健康和医疗支出。首先，使用在国家健康和营养检查调查（NHANES）中收集的个人病史来估计风险模型，以计算给定个人体重史的疾病增加的概率。然后将这些结果与来自医疗支出调查的综合医疗支出数据相关联，以估计超重和肥胖的疾病财务成本以及这些成本如何在保险集团中分布。本研究的第二个目标是更深入地了解食物消费决策如何导致肥胖。该理论模型改编自贝克尔和墨菲的理性成瘾模型，允许习惯的形成与多种成瘾商品。该模型是通过构建一个伪面板使用NHANES跨越1971-2002年估计。这些数据不包括价格或支出，因此合并价格时需要限制对地理标识符的访问。这提供了这个丰富的健康数据集的经济推论。估计的结果将提供洞察不同类型的食物习惯的相对强度以及洞察不同类型的公共干预的有效性。这项研究的智力价值不仅来自于对经济学的理论贡献，而且还来自于从政策角度处理疾病的新方法。 这项研究的一个理论贡献是在一个模型中允许多种成瘾性商品的创新。具体来说，这种模式允许“好”习惯和“坏”习惯相互替代。以前的成瘾研究集中在一个单独的成瘾商品，或两个成瘾商品的商品是免费的。但是，考虑成瘾性替代品可能会为许多其他成瘾性商品的研究提供见解。此外，这项研究旨在将估计肥胖成本的研究与旨在支付这些成本的政策干预联系起来。可预防的医疗成本和食品税之间的联系表明，公共政策的方向应该更加集中，而不是仅仅因为肥胖不好而提倡更好的饮食。更广泛地说，这项研究将有助于公众了解肥胖如何影响健康，并提供最新的成本估计，从而有助于公众对超重和肥胖的辩论。此外，由于食物消费习惯可能很重要，研究的第二阶段将为决策者提供关于几项拟议政策有效性的具体指导。这项研究将通过在同行评审的经济期刊上发表向公众传播。

Obesity has become a leading health concern in the United States. The increasing prevalence of the disease, combined with the growing body of medical research linking obesity with other diseases, makes understanding obesity a primary concern of health-related research. This disease provides a unique opportunity for economics to contribute to a proper understanding of health because obesity, more so than most other diseases, ultimately results from a series of individual decisions that all consumers must make. Obesity, defined as an excessive amount of body fat, is the eventual outcome of an imbalance between caloric intake, or food consumption, and caloric expenditure, including resting metabolism and exercise. This research is designed to study the link between health investments and outcomes and subsequent medical expenditures attributable to the disease. In addition, this research will study the patterns in demand for different types of food, allowing for habit formation, in order to better understand the link between individual decisions and obesity.The first goal of this research is to study how individual health investments related to obesity impact health and medical expenditures in an environment with health insurance. First, a hazard model is estimated using individual medical histories collected in the National Health and Nutrition Examination Survey (NHANES) to calculate the increased probability of disease given individual weight history. These results are then linked with comprehensive medical expenditure data from the Medical Expenditure Panel Survey to estimate the financial cost of disease for overweight and obesity and how those costs are spread across insurance groups.The second goal of this research is to provide a richer understanding of how food consumption decisions can lead to obesity. The theoretical model is adapted from the Becker and Murphy model of rational addiction by allowing for habit formation with multiple addictive goods. The model is estimated by constructing a pseudo-panel using the NHANES spanning the years 1971-2002. This data does not include prices or expenditures, so restricted access to geographic identifiers is required for merging prices. This provides economic inferences out of this rich health data set. The results of the estimation will provide insight into the relative strength of habits for different types of foods as well as insight into the effectiveness of differenttypes of public intervention.The intellectual merit of this research comes not only from the theoretical contribution to economics, but also from its novel approach to dealing with the disease from a policy standpoint. One theoretical contribution of this research is the innovation of allowing for multiple addictive goods in one model. Specifically, this model allows for 'good' habits and 'bad' habits that are substitutes for one another. Previous studies of addiction have focused on one addictive good in isolation, or two addictive goods where the goods are complimentary. But, considering addictive substitutes is likely to provide insight into the study of many other addictive goods. In addition, this research seeks to connect studies estimating the costs of obesity with policy interventions designed to cover those costs. The connection between preventable medical costs and a food tax suggests a more focused direction for public policy than simply promoting better eating becauseobesity is bad.More broadly, this research will contribute to the public debate on overweight and obesity by contributing to the public understanding of how obesity impacts health and by providing the latest estimates of their cost. Furthermore, because habits in food consumption are likely to be important, the second stage of the research will provide specific guidance for policymakers about the effectiveness of several proposed policies. This research will be disseminated to the public through publication in peer-reviewed economic journals.