Health and Retirement: Pathways to Preservation and Improvement

说明(申请人提供)：退休对健康有何影响？由于个人寿命更长，需要保持经济活动直到晚年，以确保养老保险制度是可持续的，健康和工作之间的关系已成为主要的政策关切，并在最近的经济文献中得到了很大关注。综上所述，这篇文献发现退休对健康没有负面影响，并可能导致健康改善，尽管这一发现的规模和意义因研究和国家而异。虽然这些研究考察了退休和健康之间的因果联系，但没有一项研究进一步发现估计影响背后的机制。现有的大多数文献考察了退休、健康行为和健康投资之间的关系，但没有理解因果关系。这是因为这些研究往往忽略了这样一个事实，即退休是个人的选择，可能是由于健康状况下降造成的。虽然退休对健康的影响本身很有趣，但目前还不可能就如何影响因果关系得出政策结论。这项研究将进一步加深我们对退休后保持更好健康的因果途径的理解。我们将区分个人直接控制的行为(身体活动、社交网络、饮食、吸烟)和至少部分受医疗保健系统控制的行为(看医生、预防性护理)。我们将利用面板数据和/或工具变量方法，利用有关法定退休年龄的信息，评估退休对健康相关投资和行为的因果影响。我们建议使用最近在三种不同背景下对中老年成年人进行的三次统一的纵向研究的面板数据：八次健康和退休研究(HRS)(美国)；三次英国老龄化纵向研究(ELSA)(英格兰)；以及两次欧洲健康老龄化和退休调查(SHARE)(欧洲大陆)。这项研究将考察以下具体问题：1.退休期间哪些与健康相关的个人行为发生了变化？2.退休在医疗保健利用中扮演了什么角色？3.在解释退休对健康的影响以及国家之间的差异方面，哪些途径是最重要的？ 公共卫生相关性：虽然退休的因果健康效应本身很有趣，但目前还不可能就政策制定者或个人如何影响或改变因果关系得出政策结论。了解是什么驱动了这种因果关系，对于确定哪些行为和医疗保健利用模式可能会带来更好的长期健康非常重要。此外，了解潜在的因果机制可以帮助政策制定者了解他们应该将医疗保健政策的重点放在哪里：公共卫生信息、改变行为的计划，还是总体上增加获得医疗保健系统的机会。

DESCRIPTION (provided by applicant): How does retirement affect one's health? Because individuals are living longer and need to remain economically active until later ages to ensure that old age insurance systems are sustainable, the relationship between health and work has become a primary policy concern and has received much attention in the recent economic literature. Summarizing, this literature finds that retirement does not have a negative impact on health and may lead to health improvements, although the size and significance of this finding varies between studies and countries. While these studies examine the causal link between retirement and health, none of them go further to discover the mechanisms behind the estimated effects. Most of the existing literature that examines the relationship between retirement, health behaviors, and health investments does not get at the causal relationship. This is because these studies often ignore the fact that retirement is an individual choice that may be caused by declining health. While the health effect of retirement is interesting in its own right, it is currently impossible to draw policy conclusions about how to impact the causal relationship. This study will further our understanding of the causal pathways to maintaining better health in retirement. We will distinguish between behaviors individuals have direct control over (physical activity, social networks, eating, drinking, smoking) and those at least partially governed by the health care system (visits to the doctor, preventative care). We will make use of panel data and/or instrumental variable methods, exploiting information about statutory retirement ages, to assess how retirement causally affects health-related investments and behaviors. We propose to use panel data from three recently available, harmonized, longitudinal studies on middle-aged and elderly adults in three different settings: eight waves of the Health and Retirement Study (HRS) (United States); three waves of the English Longitudinal Study of Ageing (ELSA) (England); and two waves of the Survey of Health Aging and Retirement in Europe (SHARE) (continental Europe). This study will examine the following specific questions: 1. Which individual health-related behaviors are changing during retirement? 2. What role does retirement play in health care utilization? 3. Which pathways are the most important in explaining the effect of retirement on health and the differences found between countries? PUBLIC HEALTH RELEVANCE: While the causal health effect of retirement is interesting in its own right, it is currently impossible to draw policy conclusions about how policymakers or individuals can impact or change the causal relationship. Knowing what drives this causal relationship is important to determine what behaviors and health care utilization patterns may yield better health for the long-term. Also, understanding the underlying causal mechanisms can help inform policymakers about where they should place the emphasis of health care policy: on public health information, on behavior-changing programs, or on increasing access to the health care system in general.