Will You Need Long-Term Care?

Or will you be a care provider?

The Trump Administration’s flouting of the U.S. Constitution brings to mind Benjamin Franklin’s often quoted statement that “[o]ur new Constitution is now established, and has an appearance that promises permanency; but in this world nothing can be said to be certain, except death and taxes.” Taxes also may not be as certain as Franklin had thought with the recently passed budget resolution calling for $4.5 trillion in tax cuts over the coming decade.

But that aside, almost as certain as death and taxes is the likelihood that sometime during our lives we will be either caregivers or recipients of care. The problem is that it’s difficult to predict when and for how long we will be receiving or providing care or the level of care that will be needed.

This is especially problematic for women. They are much more likely to be thrust into a caregiving role and, if married to men, are much more likely to outlive them both because women on average live longer than men and because in heterosexual couples they are likely to be younger. As a result, when they do need care, they are less likely to have a spouse available to step in, which is one of the reason that 70% of residents of nursing homes are women.

Population-Wide Likelihood of Needing Care

It’s much easier to predict the need for care on a population basis than individually, but those population numbers can inform individual prospects of needing care. Researchers at the Center for Retirement Research at Boston College have crunched the numbers of various studies and in their report, “What Level of Long-Term Care Services and Supports Do Retirees Need?”, concluding that only about a fifth of retirees will need no assistance at all during their lives and a quarter will need extensive services. The remaining 60% of seniors will have either low (22%) or moderate care needs (38%).

But what do these categories mean? The researchers, Anek Belbase, Anqi Chen, and Alicia H. Munnell, look at both the level and the duration of care. In terms of level of care, they define low intensity as needing assistance with one “instrumental” activity of daily living (money management, cooking, shopping, communication, driving or other forms of transportation), moderate as requiring assistance with one activity of daily living (bathing, toileting, eating, dressing, transferring), and high as having dementia or requiring assistance with two or more activities of daily living.

In terms of length of care needs, they define short as up to a year, medium as one to three years, and long-term as more than three years. The result is the following chart:

Filling in the percentages results in this chart:

All of these percentages apply to individuals at age 65. Again, these figures are very useful for population-wide predictions, but less useful for individuals planning for their own futures.

Predictive Factors

However, the statistics revealed individual attributes that are somewhat predictive. Looking at whether a 65-year-old is married is somewhat predictive with those who are married at that age less likely to require extensive care later in their lives. For instance, married women have a 19% chance of needing no assistance during their lives as compared to 13% of unmarried men. Unmarried women have the greatest likelihood of requiring extensive (“severe”) long-term care at 27% as compared 22% for married women.

Level of education is even more predictive with just 9% of 65-year-olds who did not graduate from high school getting by with no assistance at all as compared to 18% of high school graduates and 22% of those who attended college. Consistent with these numbers, 28% of both those who did not complete high school and those who did require severe long-term care supports as compared to 20% of those who attended college.

Turning to race, Blacks fare worse than Hispanics and whites, with a third requiring severe long-term care as compared to 23% of both Hispanics and whites.

Not surprisingly, health at age 65 is extremely predictive of ultimate care needs. Just 5% of those who rate their health as fair or poor live out their lives with no assistance as compared to 30% of those who rate it as very good or excellent. Almost of a third of those in the first category fall in the severe category as compared to 18% of those in the second group.

Extrapolating from the Population Level to the Individual

Combining all these attributes, clearly a Black unmarried woman who did not graduate from high school and is in poor health at age 65 is much more likely to need extensive long-term care than a white married man with a college degree who reports being in good health at the same age. Unfortunately, these attributes are almost exactly misaligned with the ability to pay for care.

It should be possible to create a calculator for individuals to use to get some idea of their own likelihood of needing care. If it were combined with estimates of the cost of care in their region, it could be used in financial planning to provide some idea of how much money they can anticipate needing to spend. Such a calculator would not be conclusive since costs are largely determined by the setting where care is provided — home, assisted living or nursing home — and whether family members can provide unpaid care, but it would provide more guidance than appears to be available anywhere today.