Long-Term Care for the Disabled Elderly:
Current Policy, Emerging Trends and Implications for the 21st Century
By Robyn I. Stone, DrPH
(Click here to submit comments and questions by e-mail to the author. Messages will be copied to Paul Cleary, Editorial Director.)


THE THREE LEGGED STOOL OF LONG-TERM CARE POLICY

 

To understand the implications of current and emerging trends in the demand for and supply of long-term care, it is important to recognize the three legged stool of long-term care policy-financing, delivery and training. These dimensions are essential to the development of sound and appropriate long-term care policy, and equal attention must be paid to all three in order to achieve that goal.

Long-Term Care Financing

Long-term care costs make up a relatively small but growing proportion of personal health care expenditures increasing from less than four percent in 1960 to over 11% in 1993 (Alecxih, 1997b). The financing of long-term care services is a patchwork of public (federal, state and local) funds and private dollars, primarily out-of-pocket. In 1995, approximately $106.5 billion were spent on long-term care; public resources accounted for 57.4% of the long-term care expenditures with Medicaid being the largest payer (37.8 %--21.1% federal and 16.7% state dollars) followed by Medicare (17.8%) and other federal and state funds (e.g., Veterans Affairs, Older Americans Act, Social Services Block Grant, state general assistance). Private insurance accounted for only 5.5% of the expenditures, with one out of three dollars attributable to out-of-pocket expenses (Figure 7) (National Academy on Aging, 1997). These estimates do not include monetization of the vast amount of informal care provided, including the opportunity costs of foregone wages (Stone & Short, 1990). One recent study of the value of informal caregiving reported that $196 billion a year are contributed to the U.S. health and long-term care systems by family and friends who provide care at home to people of all ages with chronic disability (Arno et al., 1999).

Medicaid

Medicaid, the federal/state health insurance program for the poor, is the major public program covering long-term care for the elderly and nonelderly disabled. Despite the public's tremendous interest in and personal demand for home care, Medicaid continues to exhibit a strong institutional bias; of the almost $50 billion Medicaid spent on long-term care services in 1995, $40 billion supported nursing homes and institutions for the mentally retarded (ICF-MRs) with the remaining $9.9 billion paying for home and community-based care. There has, however, been tremendous growth in the home and community-based care sector; while total Medicaid long-term care spending increased by 8.6% between 1993 and 1994, noninstitutional spending on home and community-based care waivers and the personal care option grew by 26%.

In contrast to the large federal role in financing acute care for the elderly, states are major financiers of long-term care, and there is tremendous inter- as well as intra-state variation in funding for both institutional and HCBC. Montana had the highest Medicaid nursing home expenditures per capita in 1994 while Arizona ranked lowest. Arizona also had the lowest per capita expenditures for home and community-based care while New York's spending ranked highest (Graves & Bectel, 1996). In fact, 35% of all Medicaid spending on home care in the U.S. in 1995 occurred in New York (Kenney et al., 1998).

While the institutional bias prevails in most states, there has been significant movement to level the playing field between nursing homes and home and community-based care options. Besides overall increases in home care spending, several states, most notably Oregon and Washington, have explicitly recognized nursing home placement as the setting of last resort, and have intentionally reduced the number of nursing home beds. In Oregon, for example, the ratio declined from 47 per 1000 elderly in 1982 to 35 per 1000 elderly in 1995. With an aggressive home and community-based care policy since the early 1980s, this state has successfully placed many seriously disabled elders and younger disabled in alternative assisted living facilities and adult foster homes. Oregon also supports a strong case management program which allows many disabled beneficiaries to remain in their own homes.

In addition to the federal Medicaid dollars that states match and the relatively modest sums available for personal care services through the Older Americans Act and the Social Services Block Grant, many states augment or create their own separate programs with state funds. Pennsylvania and New Jersey, for example, have relatively large state home and community-based care programs which are supported in large part with lottery revenues. A number of local communities have also been successful in raising funds for long-term care services. For example, Hamilton County, Ohio (the Cincinnati area) supports its disabled elderly through a county levy that the local area agency on aging (AAA) was successful in getting legislated (Council on Gain of the Cincinnati Area, Inc., 1997). In 1997, the AAA's Elderly Services Program spent $17 million for homemaker services, personal care, home-delivered meals, case management, adult day care and transportation for its frail elderly living in 88 neighborhoods throughout Hamilton County. The AAA in this local area was successful in convincing its citizens––elderly and nonelderly––that a levy for long-term care services was necessary given continuing cuts of federal funds, and that the dollars would benefit the entire community.

Medicare

Medicare has generally not been considered a major payer in the area of long-term care, and many have argued that one of the problems with elderly not being prepared for long-term care expenses is that they believe that Medicare will cover these costs. Medicare covers primarily acute care costs, with skilled nursing facility and home health care benefits intended as short-term coverage to meet the post-acute care needs of beneficiaries following a hospital episode. However, the belief that Medicare covers long-term care has more validity now than in the past. Through a series of regulatory and administrative changes since 1989, Medicare has come to support more long-term, nonskilled personal care (Komisar & Feder, 1998).

Medicare spending for home health services increased nearly tenfold between 1987 and 1995 (Kenney et al., 1998). A combination of a lawsuit and other administrative changes in 1989 led to a slackening of denial rates and looser interpretations of both definitions (e.g., homebound) and benefits (e.g., management and evaluation) by fiscal intermediaries. Most of the growth is attributed to an increase in the number of visits, particularly home health aide visits--the low tech, personal care type of services usually regarded as long-term care. Komisar and Feder (1998) estimated that visits per person served represented 49% of the growth in Medicare's home health spending between 1990 and 1996 (Figure 8). Furthermore, the 10% of the Medicare home health users with more than 200 visits in 1994 were responsible for 43% of the program's spending on home health care in that year (Figure 9). The length of the home health care episode has increased substantially, with a small but growing proportion of users receiving continuous care for two years or more. Recent research has found that these beneficiaries tend to be more ADL disabled than those with shorter, post-acute episodes and to be receiving more unskilled, home health aide visits (Komisar & Feder, 1998). These findings support the concern expressed by some policymakers that a small but relatively expensive subpopulation of Medicare home health users are receiving traditional long-term care through this program. It is difficult, however, to ascertain from the data the extent to which these individuals also have skilled nursing needs that truly warrant home health coverage. It is possible that the need for home health care has shifted since its inception in 1965 and that a more chronically disabled elderly population coming out of hospitals requires more unskilled personal care than in the past.

There is also some evidence to suggest that states are substituting or "maximizing" Medicare for Medicaid in order to reduce state costs for long-term care (Kenney et al., 1998). New York and Minnesota, for example, have explicit Medicare maximization policies. This state "Medicare maximization" behavior was substantiated through a series of case studies with state officials in which the interviewees acknowledged that there was budgetary pressure to help elderly Medicaid clients become eligible for Medicare home health benefits (Kenney et al., 1998). In many southern states, there is strong evidence of maximization; in Mississippi and Tennessee, for example, total Medicare spending on home health care in 1995 was, respectively, 31 times and 36 times higher than the Medicaid spending on home and community-based care. This maximization, however, is more likely to be attributed to for-profit home health agencies than to the behavior of state officials.

Congress and the Clinton Administration responded to the tremendous growth in the Medicare home health benefit by enacting provisions in the Balanced Budget Act of 1997 that significantly reduced payments to home health agencies, implemented an interim payment system with a new prospective payment system for reimbursement scheduled for 2000, and cracked down on fraud and abuse. Public policymakers chose to address the "problem" by trying to retain the post-acute nature of the home health benefit through significant restructuring of reimbursement. Others, myself included, believe that the federal government may have gone too far in its quest to "reign in" home health agencies. Many home health agencies, particularly the non-profits, have closed down due to insufficient funds to serve their clientele. The ultimate loser in this policy decision may be the disabled elderly who have significant long-term care as well as post-acute care needs.

The other Medicare growth area where the lines between acute and long-term care have become fuzzy is subacute care. There is no consensus about the definition of subacute care although it is described by proponents as a set of intensive and coordinated treatments and services provided to post-acute care patients for the purpose of minimizing or even bypassing expensive hospital stays. Whether subacute care is an innovative service delivery mechanism or a marketing strategy by sophisticated providers to repackage long-standing, post-acute care services (e.g., a skilled nursing facility, rehabilitation facility or home health service) is a subject of much discussion (Harvell, 1997). Subacute care may refer to certain types of services (e.g., rehabilitation (Singleton, 1993)); patients (e.g., those who no longer require acute services (Hyatt, 1993)); or levels of care (e.g., classified between acute hospital care and skilled nursing care (Gonzales, 1994)).

The use of Medicare nursing facilities increased substantially between 1990 and 1993, up from 22 per 1000 beneficiaries to 31 per 1000 beneficiaries (Alecxih, 1997a). Gage et al. (1997) found that increases in Medicare expenditures for these services were attributable in part to administrative and legislative changes and in part to the provision of subacute care services to more medically complex patients in nonacute care settings. The question remains whether subacute care is an innovative practice somewhere between acute hospital and skilled nursing care or, as Manard et al. (1995) concluded after a series of case studies, "old wine in new bottles."

Private Long-Term Care Insurance

Private long-term care insurance finances only a small proportion of the long-term care bill; in 1995, private insurance covered less than 6 percent of nursing home and home care costs (National Academy on Aging, 1997). The market has grown over the past decade, with the number of policies sold increasing from 800,000 in 1987 to almost five million in 1996. A 1997 Health Insurance Association of America (HIAA) survey indicated that the number of policies purchased increased by more than 600,000 in 1996 alone, the largest number ever of long-term care policies sold in one year (Coronel, 1998). It is important to recognize, however, that the five million estimate is the cumulative number of policies ever sold. The number in force is a fraction of those sold and could be even smaller, given the high lapse rate seen in this industry.

The HIAA survey reported some fluctuation in the number of companies marketing long-term care products, with 120 insurers selling long-term care insurance by the end of 1996 (Coronel, 1998). As of the end of 1996, approximately 80 percent of the 5 million long-term care insurance policies had been sold through the individual and group association markets. About one-third of the 1996 insurance carriers, however, sold policies in either the employer-sponsored or life insurance markets, up from 14 percent in 1988. These two markets also represented 20 percent of all long-term care policies sold as of 1996, up from less than three percent in 1988.

The Long-term care insurance markets vary widely across the country. The 1997 HIAA survey reported that by the end of 1996, half of all individual and group association policies had been sold in only nine states: California, Florida, Illinois, Iowa, Missouri, Ohio, Pennsylvania, Texas and Washington. Market penetration rates, as measured by the age-adjusted (65 years and older population) number of policies sold in each state, were highest in Iowa, Montana, Nebraska, North Dakota, and Washington.

Twelve companies represented about 80 percent of all individual and group association policies sold in 1996. All leading insurers offered plans that cover nursing homes, home health care, adult day care, respite care and alternative care services. A separate assisted living facility benefit was offered by 10 of the 12 top sellers. These companies offered plans with an annual five percent compounded inflation rate and with a nonforfeiture benefit. The average annual premium for the base long-term care product (four years of nursing home/home health coverage and a 20-day elimination period) purchased at age 65 was $980; costs rose to $1,321 with nonforfeiture protection, to $1,829 with five percent compounded inflation protection, and to$2,432 with both additional protections.

Controversy around private long-term care insurance has raged over the last decade, with the private sector arguing that public programs will never be able to meet the demand and consumers and regulators expressing concern about affordability and fraudulent marketing practices. It is somewhat of an academic debate to argue what proportion of income or assets people will be willing or should be willing to spend for long-term care insurance (Friedland, 1990). One estimate suggests that a single person should have at least $40,000 in liquid assets to consider purchasing insurance (Polniaszek, 1997). A recent Consumer Reports article on long-term care ("How will you pay for old age?" 1997) suggests that only about 10 to 20% of the elderly can afford insurance and notes that premiums for two "adequate" policies bought at age 65 is $3500 per year or 13% of the median annual income of elderly married couples. Whether this is a high or low proportion of a couple's annual expenditures depends on such factors as the current and projected economic status of the couple and the nature of the competing financial demands.

Many have suggested that the only hope for private long-term care insurance to play a major role in financing these services is the development of an employer-based, group market where policies are sold to people at younger ages. Premiums for insurance policies sold through employers are lower than those for individual products because 1) employers offer more effective marketing to individuals at younger ages; 2) there are administrative and agent commission savings; and 3) there is potential for employers to use bargaining power to reduce insurers' profit percentages. Employer-based products also offer increased access to coverage due to less stringent screening criteria or absence of criteria (guaranteed issue). Furthermore, a group market offers increased ease and comfort of purchase due to fewer coverage decisions required.

According to the 1997 HIAA survey, 1,532 employers were offering long-term care insurance coverage to their employees and retirees by the end of 1996, up from 7 in 1988 and 1,260 in 1995. Over 500 employer-sponsored plans were introduced in 1995 and 1996; most of these plans were employee "pay all" plans with no employer subsidies.

Preliminary findings from a recent study of 39 employers representing 900,000 employees indicate that a majority of participants offered lighter underwriting or even guarantee issue policies to at least active full-time employees (The Lutz et al., 1999). Most also offered coverage to at least one group in addition to full-time active workers (i.e., parents/in-laws, spouses and retired employees). Nearly all employers used a single long-term care insurer and most offered three or fewer benefit amount options. All surveyed employers offered inflation protection and just over half offered some type of nonforfeiture benefit that would provide the purchaser with some level of benefits if the policy lapsed because of failure to pay premiums. All but two of the participating employers required the employer to pay the entire premium.

One interesting experiment in combining public and private policies is the Partnership for Long-Term Care, a demonstration program sponsored by the Robert Wood Johnson Foundation to promote the development of private funding sources for long-term care (Cohen, 1997; McCall, 1997). The Partnership, implemented in four states––California, Connecticut, Indiana and New York (Meiners & McKay, 1989)––uses private insurance to cover the initial costs of long-term care with Medicaid paying for services after the private insurance coverage is exhausted. Two models have been developed: a Dollar-for-Dollar Disregard model in California, Connecticut and Indiana, and a Total Asset Disregard model in New York. In the first model, consumers purchase an amount of private insurance coverage equal to the amount of assets they wish to protect. When the private benefits are exhausted, the amount of private insurance that was paid out for qualified services is disregarded in determining eligibility for Medicaid. The New York model requires that consumers purchase three years of private nursing home or 6 years of home care coverage, and all of the insured's assets are protected once the private benefits have been exhausted.

One of the major limitations of this private long-term care insurance evaluation (as is true for the development of the market itself) is the fact that there is frequently a significant lag time between the purchase of a policy and a claim. Consequently, the Partnership Demonstration has yet to provide empirical evidence as to the successes or failures of the project. Recent research, however, provides interesting information on the purchasers of these products (McCall et al., 1997). Partnership purchasers were older, had smaller families, and were much more highly educated than the comparison sample of individuals age 55 to 75 who were not covered by Medicaid. They were also more likely to be female, white, in reportedly good or excellent health, and in a relatively high-income bracket. The Partnership sample was also more likely than the comparison group to disagree with the statement "Medicare currently provides sufficient coverage for long-term care" and was much less likely to believe that government will pay for long-term care if they need it in the future. This study identified the Partnership purchasers as a self-reliant group, where decisions to purchase were based on maintaining independence and preserving income and choice, rather than the desire to leave an inheritance.

This discussion of how long-term care is financed in the United States underscores the complicated and confusing nature of our fragmented system. To summarize: