Age Difference of Spouses and Long-Term Care

Rand Corporation, Santa Monica CA

Investigators

Abstract

National expenditures on long-term care currently total $100 billion annually in the US. The magnitude of these expenditures makes it imperative for us to understand past and future trends in long-term care demand. The aim of the proposed research is to examine how the declining age gap between spouses will affect future trends in the demand for long-term care. It has been well-established that the presence or absence of a healthy spouse is a major determinant of nursing home entrance: a disabled person married to a healthy spouse is about half as likely to enter a nursing home as a disabled person without a healthy spouse. In turn, the availability of healthy spouses may have been substantially affected by the shrinking age gap between spouses over the twentieth century. A woman born 1900 on average ended up with a husband who was 4.2 years older than she. The same woman born in 1950, however, would have ended up with a husband just 2.5 years older. We propose to show that this decline has two effects: (1) All elderly couples will tend to stay married longer, because they are more closely matched in age. (2) Today's elderly women will tend to have younger, healthier husbands, but today's elderly men will tend to have older wives. The first effect will reduce nursing home demand, while the second has an ambiguous effect. We propose to quantify both effects and estimate their overall effect on nursing home demand. First, we will use life table data from the Social Security Administration to compute the effect of the changing age gap on the prevalence of marriage among the elderly. Next, we propose to use data from the AHEAD study to estimate the probability of nursing home entrance by disability, marital status, and the age of a married person's spouse. As part of this analysis, we will use Census data to impute data on the ages of deceased spouses absent from the AHEAD. This analysis quantifies the effect of the changing age gap on the probability of nursing home entrance for married people. Using our estimates of the changing probability of being married, and the changing probability of nursing home entrance for married people, we then compute the total effect of the changing age gap on overall nursing home demand.