Life care: new options for financing and delivering long-term care.

Continuing care retirement communities provide full insurance protection for and access to long-term care services. A new model, which retains risk pooling for long-term care and provides benefits and protections similar to continuing care retirement communities, is called life care at home. Life care at home combines the financial and health security of a continuing care retirement community with the freedom and independence of living at home and is affordable to a greater proportion of elderly people. The feasibility of this model is, in part, supported by the fact that nursing home use in the fully insured access-guaranteed continuing care retirement community is not that different from use among the elderly living in the general community.

Introduction

The number of elderly people needing long-term care because of functional impairment will continue to rise dramatically in the years ahead. The provision and finance of such care falls most heavily on the elderly themselves and their families. Nearly three-quarters of the noninstitutionalized disabled population receive all needed assistance from spouses and adult children. Moreover, Medicare coverage does not extend to custodial care, so that nearly 40 percent of nursing home payments are direct out-of-pocket expenditures. Although private long-term care insurance is emerging, few policies offer the type of protection that would ensure against the catastrophic costs of long-term care. Thus, many of the elderly who require long-term care services deplete their financial resources to pay for such care and then quality for Medicaid coverage. Given the constraints on Federal and State government financing for long-term care, there is an urgent need to develop and implement alternative models for the finance and delivery of long-term care.

One such model is the continuing care retirement community (CCRC). As a result of the considerable growth in the CCRC or life care industry throughout the last decade, the CCRC is receiving increasing attention as a method of providing and financing long-term care for the elderly (Tell et al., 1987a; Branch, 1987).

An important feature of CCRC's is that they combine the finance and delivery of long-term care services within a single organizational context and insure residents against the catastrophic costs of long-term care. CCRC's offer housing and related services that often include medical and preventive health care services and nursing home care. A typical community has about 200 apartments. Members usually pay an entrance fee and a monthly charge. Two-thirds of all CCRC's provide their contracting members with some level of insurance for nursing home care, and one-third are strictly fee-for-service. Specifically, about one-third of the communities offer lifetime coverage for long-term care on a pooled basis from residents' entry and monthly fees. Thus, residents who move into a nursing center pay no more than they paid when they were living in their apartments. These are called “Type A” communities. The other one-third of CCRC's provide a specified amount of nursing care at little or no additional cost to a member who moves permanently from an independent living unit into the CCRC's nursing home. After this specified amount, the resident pays for nursing care on a full fee-for-service basis. These are called "Type B" communities.

The Type C communities are strictly fee for service although access to nursing home care may be guaranteed. As of 1986, median entry and monthly fees for a two-bedroom unit were $65,000 and $800, respectively (American Association of Homes for the Aging and Ernst and Whinney, 1987). There is great variation in CCRC fees by age of the community and by contract type. The median masks somewhat the recent trend toward high cost CCRC's. In fact, CCRC's with entry fees above $150,000 and monthly fees in excess of $1,500 are not uncommon today. It is not surprising, therefore, that this retirement option as it is currently configured is only affordable to a small proportion of the elderly. One source estimates that less than 10 percent of all elderly persons could afford to join a CCRC (Cohen et al., 1987).

Continuing care retirement communities

Attitudes toward joining

People who join continuing care retirement communities (CCRC's) do so for a variety of reasons. The health guarantees embodying both access to care and insurance for care seem to be the element that most attracts people to join Type A communities. In a survey of over 1,400 residents and persons on waiting lists to two Type A CCRC's, access to medical care and access to long-term care services to maintain independence were cited most often as very important reasons to join the CCRC (Cohen et al., 1988a). Moreover, two-thirds of the respondents indicated that protection against the costs of long-term care was a very important reason in their decision to join a CCRC.

People considering joining CCRC's also have a number of concerns, many of which are related to the issues of lifestyle and location. In addition to high costs, other primary concerns about joining a CCRC include living near too many old people and being far away from shopping or family and friends (Cohen et al., 1988a). Having to move from the family home is seen as a major disadvantage of the CCRC. It is indicated from surveys that the elderly prefer to remain in their own homes and that most do not desire age-segregated living (Tell et al., 1987b). For many, the campus lifestyle is a disadvantage because it is seen as a restriction on independence or privacy.

Growth in the market

Although less than .5 percent of the elderly are enrolled in CCRC's, the CCRC industry is rapidly growing. As of 1986, 100,000 to 200,000 elderly people lived in about 700 CCRC's (American Association of Homes for the Aging and Ernst and Whinney, 1987). Current predictions suggest that, by 1999, there will be 1,500 CCRC's with nearly 450,000 elderly residents, or about 1.3 percent of all persons aged 65 years or over (Tell et al., 1987a). Although these numbers in absolute terms are quite small, they reflect an increase of over 20 percent in the CCRC market. Moreover, the basic principles on which CCRC's rest, risk pooling for long-term care and guaranteed access to services, may be applicable to less costly models that could reach many more middleclass elderly. What CCRC's demonstrate is that risk pooling for all long-term care can be accomplished successfully with relatively small populations.

Trends in the industry

The life care industry, which first embraced the features of risk pooling for long-term care, seems to be moving away from these concepts. Newer CCRC's emphasize lifestyle and housing while offering little or, in some cases, no opportunity for insuring long-term care costs. The move away from fully insured life care models may be caused by a growing belief that the long-term care risk is not insurable. Experience has shown, however, that the risk can be shared across a relatively small elderly cohort, as has been done in many life care communities, some of which have been operating successfully for decades. Of the approximately 40 community failures to date, few if any, can be attributed to long-term care costs in an experience-related system. The majority of failures emerged from the fact that the communities offered contracts that limited the amount of increase to monthly fees that could be applied to cover increasing costs of services (Tell et al., 1987a). Had communities had the flexibility to increase monthly fees, these failures might not have occurred.

Most communities are successful in managing the risks associated with full coverage for long-term care. Entrance screens are used to ensure that only relatively healthy individuals enter the community. Moreover, the campus environment encourages the provision of informal support to residents. Finally, there are a wide array of formal services to help individuals live independently in their apartments.

Service utilization

Analyzed was longitudinal information about the utilization patterns of 3,316 residents from six Type A CCRC's, five of which opened after 1967. These communities offered full health-care guarantees for residents; that is, nursing home care was fully insured. Data collected from case files reflect residents' utilization histories from the opening date of the community for these five and from 1967 for the sixth CCRC that opened in 1960. Thus, the analysis of patterns of nursing home use is made in the context of a post-Medicare and post-Medicaid environment. Because the CCRC is a closed system, other changes occurring in the service system outside the CCRC, such as limitations on the number of nursing home beds, would have minimal effects on use within the CCRC.

Complete detailed information on all residents' use of health care services throughout their tenure in the community was available. The date, duration, and location of all resident movements to different care settings were recorded in a case file. The movement history includes transfers to hospitals, nursing homes, personal care facilities, independent living units, and deaths and withdrawals from the community. In essence, the data set afforded the opportunity to follow an initially healthy entry cohort for up to 15 years.

To make comparisons with the utilization experience of the elderly in the general population, information from other published studies was combined. The lifetime risk of nursing home entry among the general elderly population was taken from a lifetable analysis estimating the risk of nursing home entry for a 1977 cohort of Medicare beneficiaries (Cohen et al., 1986). The length of stay pattern to which the experience of CCRC residents was compared is based on the use of a 1976 admissions cohort (Meiners and Trapnell, 1984). Because most of the data on the CCRC sample is based on the utilization experience of residents during the middle to late 1970's, comparisons between the data sets are appropriate.

Determining nursing home entry risk

Studying nursing home use patterns in CCRC's is important for a number of reasons. First, such an analysis provides information on nursing home use in an insured environment. This is important to insurers and developers of new programs who are leery of the unknown liability that they may incur as a result of guaranteed access and full insurance coverage. Studying use in the CCRC affords the opportunity to learn about the effect of insurance-induced demand and the effects of a managed-care delivery system on lifetime use. Second, data from such an analysis may suggest alternative service roles or practice patterns for the nursing home that may result in an insured and access-guaranteed environment.

The transfer histories of both residents who have died before the last observation period and those who were still alive during the last observation period were recorded. Survival analysis, a statistical technique that evaluates the time interval between a starting event and a terminal event, was used to estimate risk. For the purposes of the analysis, the starting event was defined as “entrance into the CCRC” and the terminal event was defined as “first entrance into a nursing home.” The survival time was then estimated as the time between entry into the community and entry into a nursing home. These results are summarized in Table 1.

Table 1

Risk of nursing home entry for the elderly in the general population and continuing care retirement community (CCRC) residents

Age group	General population		CCRC residents
	Estimated lifetime risk	Sex-adjusted lifetime risk	Estimated lifetime risk
	Percent of population
65-69 years	35.6	38.6	55.0
70-74 years	36.9	41.9	59.1
75-79 years	40.1	44.5	65.1
80-84 years	41.6	43.4	67.0
85 years or over	38.8	39.5	82.7

SOURCE: (Cohen et al., 1988b and Cohen, Tell, and Wallack, 1986).

At 65 years of age, CCRC residents face nearly 1.5 times the risk of nursing home entry over their lifetime than do the elderly in the general community. The greatest difference in risk rates occurs for those 85 years of age or over. This may reflect differences in life-expectancy as well as age-specific rates of morbidity. CCRC residents are also more likely to have repeat entries to the nursing home (Cohen et al., 1988b). The average length of stay per admission among this CCRC population and among the general elderly population are shown in Table 2.

Table 2

Length of stay per admission for an admissions cohort of elderly and continuing care retirement community (CCRC) nursing home entrants, by age cohort

Length of stay	Entry age
	65-74 years		75-84 years		85 years or over
	General	CCRC	General	CCRC	General	CCRC
			Percent of population
Less than 1 month	29	66	27	53	24	43
2-6 months	31	19	26	27	30	27
7-12 months	10	5	12	8	11	11
1-3 years	16	7	17	9	21	16
3 years or more	14	3	18	3	14	3
			Average number of days
	419	129	497	159	425	231
			Median number of days
	56	13	66	25	123	48

SOURCES: (Cohen et al., 1988b; Meiners and Trapnell, 1984).

The length of stay per admission across all age groups is shorter in a CCRC than in the general community. Slightly less than one-half of the nursing home entrants in the general community have stays of less than 3 months, compared with nearly two-thirds of the CCRC study population. For both groups, the distribution of length of stay is skewed with only a few entrants staying a long time in the nursing home.

Taken together, these results indicate that lifetime utilization under a fully insured access-guaranteed system like the CCRC is comparable to lifetime use in the general population, at least through 85 years of age. In the CCRC, as in the general population, a small group of residents account for most of the costs. In the general population, about 15 percent of the elderly account for 90 percent of the costs; in the CCRC, however, about 20 percent of residents account for 90 percent of the costs (Cohen et al., 1988b). It is this small probability of incurring a catastrophic expense that makes nursing home care an ideal candidate for risk pooling in the general community as well as in the CCRC.

Explaining utilization differences

There are a number of factors that account for differences in utilization rates between CCRC residents and the elderly in the general population. Some of the more important factors include the following:

The probable differences in morbidity and mortality rates of the CCRC population that, in part, result from the use of entry screens.

The varying availability of substitute services.

The differing behavioral responses to financial and other incentives on the part of residents, managers, and referral sources in the CCRC.

The impact on utilization among the elderly in the general population of the nature of public financing (Medicaid) of nursing home care.

Nursing homes in these CCRC's seem to be used differently than those in the general community; there appears to be a much greater use of the homes for short-term, presumably recuperative, care than in the general community. This may indicate a more efficient use of hospital services because CCRC residents always have available to them a nursing home bed to which they can be discharged, whereas those in the general community do not. Or it could suggest greater use of CCRC-provided apartment-based services that can substitute for custodial nursing care. With respect to health maintenance organizations (HMO's), for example, the effects of prepayment on the use of services indicated that the style of medicine was less markedly hospital intensive and, consequently, less expensive (Manning et al., 1984). The same may be true for CCRC's: The integration of the nursing home with acute care and apartment-based services may encourage more cost-effective patterns of use. Thus, the way in which a delivery system manages substitute and complementary services, such as recuperative care and home-based care, can have a significant impact on nursing home use.

Developing less costly life-care models

Findings presented here support the feasibility of risk sharing for long-term care even in the context of guaranteed access to services. This is encouraging to those interested in developing off-campus models of CCRC's that provide insurance protection and access to the elderly living in their own homes. Tell et al. (1987c) developed one such model called “life care at home” (LCAH). This model is a variation on a CCRC, providing most of the same benefits to the elderly who prefer to remain in their own homes instead of moving to a central campus, but at a lower cost. LCAH combines the financial and health security of a CCRC with the freedom and independence of living at home because, if needed, members receive services in their own homes.

In essence, individuals join an LCAH plan by paying an upfront fee and monthly payments to the sponsoring organization. These payments guarantee them access to all needed services. Moreover, they are insured for most of the costs of the services provided by the plan. There are numerous possible sponsoring arrangements for plans such as a CCRC sponsor who could offer the plan as an extension of its campus option; an HMO sponsor; and a joint venture by different kinds of sponsors. This latter approach is currently being demonstrated in northeast Philadelphia. Jeanes Hospital and Foulkeways Retirement Community are sponsoring an LCAH plan that has already begun marketing to individuals.

LCAH insures enrollees against the catastrophic costs of long-term care and provides a care-managed delivery system to ensure access to needed services. LCAH differs from current long-term care insurance offerings in at least two important ways. First, the difference between LCAH and long-term care insurance is similar to the difference between an HMO and traditional health insurance. That is, in addition to financing needed long-term care, LCAH manages and provides care. Also, a greater emphasis is placed on home care services in LCAH, compared with most long-term care insurance policies that cover primarily nursing home care and offer little, if any, in-home benefits. It is more comprehensive than the social health maintenance organization (SHMO). The SHMO provides limited chronic care coverage for nursing home care, whereas LCAH provides comprehensive protection for both nursing home and home and community-based care. Unlike the CCRC, however, LCAH does not provide housing services. The costs of joining LCAH are about $7,500 in entry fees and between $180 and $225 a month. LCAH makes the life care concept available and affordable to a broader segment of the elderly population. Between one-quarter and one-half of all elderly could afford to buy into an LCAH program (Cohen et al., 1987).

LCAH incorporates a number of risk management techniques to reduce the risk of inappropriate use or overutilization of covered services. These techniques include specifying appropriate criteria to determine eligibility for enrollment and for benefits, creating benefit limits, employing cost-sharing techniques, and establishing a strong case management system.

There will be numerous challenges regarding the implementation of LCAH plans including designing a viable and marketable benefit package, avoiding adverse selection, securing the participation of high-quality service providers, and meeting regulatory requirements. Furthermore, it is unclear how the LCAH model will fit into existing health care and insurance regulatory structures. These and other issues will be addressed as plans for demonstrations of the various model types are further developed.

Joining life care at home

We assessed interest in life care at home through a combination of telephone and mail surveys to more than 4,000 income-eligible elderly throughout the United States. Consistently, more than one-half of the respondents indicated that LCAH would meet their current or future needs very well. Moreover, nearly one-quarter of respondents said that the costs of joining LCAH were reasonable. Finally, when interest is combined with ability to pay and eligibility for LCAH on the basis of health, somewhere between 10 percent and 25 percent of respondents could be considered viable prospects for joining a LCAH program. (Tell et al., 1987b; Cohen, 1987).

What people liked best about LCAH was its emphasis on services to help keep people living at home independently for as long as possible. Health guarantees and access to services were also assessed to be important features of the plan. Respondents' greatest concerns related to program cost and the ability of a sponsor to successfully implement the program.

These results probably understate the potential demand for life care models such as LCAH. As the elderly become more aware of their lack of coverage for and exposure to the catastrophic costs of long-term care, they are likely to demand such protection. To a large extent, current demand for all models of long-term care insurance is constrained by misperceptions about the limited extent of Medicare coverage for chronic care services.

Policy implications

Life care has the potential to protect older persons against the leading cause of catastrophic health expenditures in the United States. One life care model, the CCRC, offers an effective model with many years of experience at combining the finance and delivery of long-term care services. Its benefits to the elderly include financial protection; assured access to high-quality, long-term care; and personal security and maintenance of independent living. Lower cost models of life care have the ability to generate potential Medicaid savings by delaying or eliminating spend down (Cohen et al., 1988c).

States are beginning to play a more active role in facilitating the development of these programs. Some city and State governments have already begun to explore ways to offer subsidized or lower cost CCRC's for mixed-income populations. States can foster public acceptance of these programs by encouraging and assisting public information. The elderly need to be informed as to the limitations in coverage of Medicare and Medicaid for long-term care costs (American Association of Retired Persons, 1986). This will help create consumer demand for insurance models of life care. Also, States may want to create incentives for the purchase of products such as long-term care insurance with deep coverage or programs like LCAH or lower-cost CCRC's. For example, a Medicaid program may want to disregard for eligibility purposes an amount of liquid assets equal to some percentage of the premium or monthly fee paid to a comprehensive insurance program.

To assure the quality of products, States may want to certify certain long-term care insurance products, the purchase of which would entitle the purchaser to a tax benefit. Closely related, there is a need to remove barriers to the development of private markets for these products. One State, for example, prohibits life care contracts for health care. An important role for States is in setting minimum standards for products and eliminating legal barriers to their development. Finally, to assist program developers establish sound insurance programs, it is important to facilitate shared information on long-term care utilization and cost. Public investment in data development and acquisition is an additional important role for the public sector.