End-of-Life Care in U.S. History

Approaches to caregiving for older Americans has changed dramatically over time. The colonial era was largely characterized by family- or community-provided support for the elderly towards the end of their life, whereas the early-to-mid 19th century saw the impoverished elderly shift into lower quality care as provided by poorhouses. The 20th century saw a general shift of end of life care for both well-off and impoverished elderly into nursing homes. Today, 70% of individuals living to age 65 are projected to use long-term care services at some point, and end of life and long-term care expenditures comprise a substantial share of state and federal budgets.
Past research in economics has argued both qualitatively (Posner 1997) and quantitatively (Costa 1999) that the elderly prefer to live independently if they can afford to do so. But if such preferences have been stable over time, what factors do we think have led to the dramatic changes in end of life care over time within the US?
Historians have stressed various social changes that may have reduced the role of family in caring for the elderly – e.g. improved employment opportunities in the manufacturing sector increasing geographic mobility of adult children (i.e. encouraging the exit of potential caregivers for elderly individuals). However, a variety of local, state, and federal policies over the course of the 20th century also seem like strong candidates to be drivers of these trends: state old age assistance laws (starting in 1915), the introduction of federal old age assistance (OAA) and Social Security, public assistance for hospital and nursing home construction under the Hill-Burton program and subsequent amendments, federal loan guarantees which subsidized the construction of nursing homes, Medical Assistance for the Aged, Medicare, and Medicaid, to name a few of the largest programs. Many of these programs likely affected both the quantity of available caregiving institutions (such as nursing homes) as well as the quality of care available in those locations. While the timing of these programs frequently coincided with dramatic time-series changes in end of life care, apart from a few specific examples – e.g. evidence from Costa (1999) and McGarry and Schoeni (2000) on how OAA and Social Security payments affected the probability of widows living independently – very little is known about the importance of these policies in driving the dramatic aggregate changes in end of life care over time.
This pilot project has three goals:
1. First, we aim to construct new historical data series with which to quantify changes over time in end of life care for the elderly in the US.
2. Second, our hope is that these new descriptive data series will inform and guide our decisions about where best to focus our attention in terms of understanding the drivers of these changes in end of life care, both on the demand side (e.g. changes in labor market opportunities for children of the elderly, which may have affected their willingness to provide in-kind care) and on the supply side (e.g. policies changing the costs of or incentives for facilities such as nursing homes to enter various markets).
3. Third, we intend to develop this work into some preliminary results that could – if promising – form the basis for a larger R01-type proposal.