Page images
PDF
EPUB

nursing home within 30 days of discharge. To examine how individuals pay for nursing home stays accurately, it is important to capture complete episodes of

care.

Table E-1 Prices Assumed in the Model for Services, by Source of Financing,

1990

The model assumes that nursing home lengths of stay, on an age and marital status basis, will remain constant over time.

Home Care Utilization-The model simulates both the use of formal home care services and the length of use. Home care utilization for purposes of this discussion includes any home health, homemaker, personal care, or meal preparation services for which the caregiver is reimbursed by the elderly recipient, Medicare, Medicaid, or funds from other public payers. These funds come from the Older Americans Act, state and local programs, social services, block grant monies, Veterans' Administration programs, and charity among other sources. Data from the 1982-84 NLTCS were used to estimate the use of home care. NLTCS data on the characteristics of persons in 1982 were used to predict utilization of home care services and length of use in 1984. Over time, the likelihood of using home care services and length of use are assumed to remain constant on an age/sex/disability basis for persons in the community.

Long-Term Care Prices

Long-term care prices, such as those for nursing home and home care, have a profound impact on the sources of financing for long-term care because many individuals deplete their assets and "spend down" onto Medicaid after long stays in a nursing home.

In the model, national averages are used for longterm care prices. For nursing home prices, daily rates are used for Medicare, Medicaid, and private pay. The Medicare rate is based on HCFA estimates of the average Medicare SNF per diem rate. The Medicaid daily rate is based on average SNF and ICF Medicaid payment rates in 1985 from the NNHS Facility File, weighted by the number of residents receiving Medicaid SNF or ICF payment in the 1985 NNHS. The private pay rate is based on average SNF and ICF private charges in 1985 from the NNHS Facility File, weighted by the total number of beds in a facility certified for SNF or ICF use. The 1985 Medicaid and private pay rates were adjusted to 1988 based on HCFA estimates of changes in nursing home prices over the period. Table E-1 shows the rates used in the model in 1990.

The annual long-term care inflation rate for nursing home and home care prices in the model is assumed

[blocks in formation]

to be 5.5 percent. This annual rate is equal to the projected rate of growth in compensation as specified in the 1989 Social Security Trustees' Report (4.0 percent increase in the CPI, 1.3 percent increase in real wages, and 0.2 percent increase in fringe benefits). This is equivalent to 1.5 percent annual real growth in long-term care prices over the long run. The rationale for using the growth in compensation is that it reflects the high proportion (approximately 70 percent) of labor costs in total long-term care service costs. It assumes that nursing home and home care wages and other benefits will increase at a rate comparable with that of the rest of the economy. It also assumes there will not be any change in the proportion of labor costs relative to other costs in nursing homes or home

care.

MODELING THE COMMISSION RECOMMENDATIONS

The Commission has proposed the establishment of a comprehensive public program to meet the needs of the long-term care population. The plan consists of three components, which together address the needs of institutionalized and community-based, severely disabled Americans. The first part of the proposal would provide social insurance for home and community-based care for severely disabled individuals. Under a second part of the proposal, persons who qualify for institutional care would be entitled to social insurance covering the first three months of their nursing home stay. Finally, for persons remaining in nursing homes beyond the three-month period covered under the social insurance, the plan would provide significantly higher income and asset protection than is available under the current Medicaid program.

The Brookings/ICF Long-Term Care Financing Model was used to simulate the cost of the proposed program and the program's impact on elderly Americans now and in the future. Estimates were made in 1989 dollars for the 1991 to 1995 period and for the 2016 to 2020 period. This section discusses in detail the various provisions of the proposal and their implementation within the framework of the model. 10

Social Insurance for Home and Community-Based Care

The proposal would provide public coverage of home and community-based care for severely disabled individuals of all ages. Specifically, this program would provide "hands-on or supervisory" assistance to persons having three or more ADL limitations or having severe cognitive impairment. Under the proposal, case managers would determine the number of hours of care and mix of services received by beneficiaries, subject to a budget. The budget reflects per capita amounts that vary by disability level. The types of benefits provided under this portion of the proposal would include skilled nursing services, therapy services, personal care services, homemaker chore services, grocery shopping and transportation, medication management, adult day care, respite care, and training and counseling services for caregivers. The plan would require beneficiary copayment of 20 percent of the lesser of costs of care or the national average cost of care. For individuals with incomes below 200 percent of the federal poverty level, however, the federal government would subsidize copayment requirements (see Table E-2).

This proposal was implemented within the structure of the Brookings/ICF Long-Term Care Financing Model in the following way.

First, the number of users was modeled. Elderly persons with three or more ADL limitations or a similar level of cognitive impairment were assumed to be eligible for the benefit. The model does not identify these groups directly. It does, however, have data on persons with two or more ADL deficiencies. To estimate the number of persons eligible for the Commission proposal, the NLTCS was used to estimate the number of persons who would meet the proposal's eligibility criteria as a percentage of persons in the NLTCS with two or more ADL deficiencies. (To meet these criteria an individual must have three or more ADL limitations, severe cognitive impairment, or disruptive or dangerous behavior.) This percentage was then applied to the model's estimates of the number of persons with two or more ADL deficiencies.

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][subsumed][merged small]

The number of visits each user would receive was then modeled. The program establishes a budget within which the case manager allocates services. Estimates assume a maximum of 25 hours per week. If one assumes that each visit is four hours long (including travel time), then program beneficiaries would receive up to 27 visits per month. Current users of formal services were assumed to receive the same number of visits they do now, subject to the 27 visit per month maximum. New users of program services were assumed to have the same distribution of visits as current users of formal care. On the one hand, this may underestimate the increase in services because it assumes current users do not increase their use of services. On the other hand, it may overestimate the use of program services because it assumes that all new users receive services at the same intensity that current users do.

Program payment levels were then modeled. The program rates for home and community-based care used in the model are based on estimates of 1990 Medicaid costs per visit ($60.30 per visit), with an additional 20 percent for case management. An additional assessment cost of $100 per participant in 1989 dollars was assumed.

Beneficiary cost sharing was then modeled. No deductible applies to the benefit; however, a copayment of 20 percent of the cost of care is required. Individ

uals with income below 200 percent of the poverty level were assumed to receive subsidies toward their required copayment as indicated in Table E-2.

Finally, all persons qualifying for current programs who do not meet the eligibility criteria for the home and community-care benefit were assumed to continue to be served by these programs. This includes persons eligible for the Medicare home health benefit.

Three-Month Front-End Coverage of Nursing Home Stays

Under the proposal, all nursing home users would receive coverage for the first three months of their nursing home stay. Covered services include skilled nursing care and custodial care.

In modeling this part of the proposal, it was assumed that all entrants to nursing homes would receive the benefit (that is, current residents would not receive the benefit). A lifetime benefit limit of three episodes was also assumed. The required copayment of 20 percent of the cost of care was assumed to be subsidized at the rates in Table E-2 for beneficiaries with incomes below 200 percent of the poverty level. A payment rate in 1990 of $76.95 per day (125 percent of the current national average Medicaid rate) was assumed. This is about 10 percent less than the current private pay rate for nursing homes. Rates close to but less than the average private pay rate reflect the proposal's intent to ensure access to care while using the government's bargaining power to control costs.

Long Nursing Home Stays

The nursing home program provides income and asset protection for nursing home users of all ages who are judged eligible for nursing home care by a federally certified assessment agency. Covered services include skilled nursing care and custodial care. This coverage is effective through the entire length of a nursing home stay.

The proposal would protect $30,000 in nonhousing assets for single individuals and $60,000 for couples. Individuals having financial assets in excess of these amounts must contribute unprotected assets toward the cost of their care. Under the program, income projections include a housing allowance of 30 percent of income for at least the first year of a nursing home stay. For married beneficiaries, the housing allowance is extended up to one year after a spouse is no longer

in the community. In addition, the plan provides a monthly personal needs allowance of $100 for all beneficiaries. Community-based spouses of beneficiaries may retain any income up to the 200 percent of the federal poverty level for elderly couples. Any remaining income is contributed toward the cost of nursing home care received by the beneficiary. All of these provisions were modeled directly in the model. It was assumed that current residents would be eligible for the benefit.

Induced Demand

The proposed program is likely to increase costs for three reasons. First, many persons who currently are not eligible for long-term care services will be covered by the proposed programs. Second, those currently eligible for long-term care benefits will bear a smaller share of their costs directly, depending on the nature of the program. Third, with a reduction in cost to the individual, additional persons will seek services, and those currently receiving services may seek more services. The third effect is referred to as induced demand.

Concern about the impact of proposed financing changes on service use is understandable, particularly in light of the way services are used today. Under current methods of financing, a high percentage of long-term care services is provided voluntarily by rel atives. In addition, a large number of disabled elderly and nonelderly persons who do not receive services could be medically eligible if services were covered.

Unlike acute care, little research has been conducted on the effect of induced demand on the use of long-term care services. The primary reason may be that the effect is hard to measure. Because of the potential impact of induced demand on the use and costs of long-term care services, however, it must be considered as a component in the financing of these private and public proposals.

All the estimates of induced demand for long-term care services are derived from analyses of the demand for nursing home care. Studies use different methodologies, but the range and size of the price elasticities are narrow compared with those in the literature on the demand for acute care services. 11. 12 In general, the studies found highly elastic responses to price that range from a 10 percent decrease in price leading to an increase in demand from 7 percent to 23 percent. Regardless of which estimate is correct, even the lowest implies that private consumers of nursing home care are very responsive to price changes-that is, lowering prices will result in relatively large increases in the use of services.

Although the evidence from the demand for nursing home care literature suggests that use would increase a great deal, one should probably not use the literature estimates directly for five reasons. First, data from some studies are old and may not reflect current behavior, and the ability to generalize from others is limited. 13

Second, these studies are cross-sectional; therefore, the range of price elasticity estimates may reflect true differences in the populations sampled or simply differences in specification.

Third, according to recent surveys, the overwhelming majority of disabled elderly want to stay out of nursing homes.

A fourth issue is that price elasticities become less relevant when the nursing home bed supply is constrained by government regulation.

Finally, the program recommended by the Commission contains both home care and nursing home care provisions. Assuming that home care and nursing home care are to some extent substitutes, neither the nursing home induced demand effect nor the home care induced demand effect may be as large as the studies discussed might suggest.

Based on the estimate of the effective reduction in the price of care facing consumers, use of home care services is assumed to double under the Commission's proposal, and the probability of use is adjusted accordingly. Currently about 40 percent of those who would be eligible for care under the proposal use paid home care services at a point in time. In modeling the proposal it was assumed that this number will increase

to approximately 80 percent. This assumes all current users of formal care, and approximately two-thirds of eligible nonusers, would participate in the program.

Similarly, the decrease in the out-of-pocket cost of institutional care will increase utilization of nursing home services. In modeling the proposal it was assumed that the reduction in the price of nursing home care would increase utilization 20 percent.

Estimating the Cost of Long-Term Care for the Nonelderly

Data on the long-term care needs of the nonelderly are limited to estimates of the size of this population. 14 Precise estimates of severity of the disabilities, as well as current long-term care use or coverage from existing public programs, are not available. Therefore, the estimates of the net federal costs of providing long-term care services to the nonelderly were based on simple proportions based on the net per capita costs of the elderly applied to the nonelderly. Specifically, the net federal costs of providing home health care to the nonelderly were based on the per capita costs of the eligible elderly applied to the total number of nonelderly estimated to have any limitation in activities of daily living. Costs for the nonelderly population were estimated as 75 percent of this cost. Some of these individuals would not qualify as severely disabled, and others were expected to be receiving federal assistance through Medicaid, Medicare, or another public program.

To estimate the net federal cost of the nursing home portions, the net federal cost of the elderly population was simply increased by the current proportion of nonelderly nursing home residents.

Notes to Appendix E

1. Prepared by David L. Kennell, Lisa Alecxih, and Dhiren Patel of Lewin/ICF and Joshua M. Wiener of The Brookings Institution.

2. In the model, a Monte Carlo simulation methodology is used to simulate changes in an individual's status by drawing a random number between zero and one and comparing it to a fixed probability of that event occurring for an individual with a given set of sociodemographic characteristics. For example, the annual probability of death for an 85 year old noninstitutionalized female is assumed to be 0.03 (that is, three out of every 100 women age 85 who are not in a nursing home are expected to die each year). If the random number drawn by the model is less than or equal to 0.03 for this 85-year-old woman, then the individual is assumed to die in that year. If the number drawn lies between 0.03 and 1.00, then the individual is assumed to continue to live during that year.

3. Harold Fullerton Jr., "Labor Force Projections 1986-2000," Bureau of Labor Statistics Monthly Labor Review 110 (9) (September 1987): 19-29.

4. George Silversti and John Lukasiewicz, "A Look at Occupational Employment Trends to the Year 2000," Bureau of Labor Statistics' Monthly Labor Review 110 (9) (September 1987): 46-63.

5. David L. Kennell and John Sheils, Documentation of the Pension and Retirement Income Simulation Model (PRISM) (Washington, D.C.: ICF Incorporated, September 1986).

6. This occurs in every year except 1979, when the model simulation starts. In this year, a disability status is simulated for all persons age 65 and over, based on the prevalence rates estimated from the 1982-84 NLTCS. Sixty percent of individuals receiving Disability Insurance program benefits at age 62 are simulated to continue to be disabled when they reach age 65.

7. In the 1982-1984 NLTCS, disability was defined as the inability to conduct any of the activities of daily living or instrumental activities of daily living due to a health condition which had or would endure for 90 days or more. The measure used in the model is based upon the control card in the NLTCS and therefore includes persons

who require active human assistance, standby assistance, or assistance from special equipment.

8. Raymond J. Hanley, Lisa Maria B. Alecxih, Joshua M. Wiener, and David L. Kennell, "Prediction Elderly Nursing Homes Admissions: Results From the 1982-84 National Long-Term Care Survey," Research on Aging 12 (2) (June Liu and Kenneth 1990): 199-228; Korbin Manton, "The Characteristics and Utilization Patterns of an Admission Cohort of Nursing Home Patients," The Gerontologist 23 (1) (February 1983): 92-98; William Weissert and William Scanlon, "Determinants of Institutionalization of the Aged," Project to Analyze Existing LongTerm Care Data, Volume III (Washington, D.C.: The Urban Institute, 1983): 1-19; Leticia Vincente, James A. Wiley, and R. Allen Carrington, "The Risk of Institutionalization Before Death," The Gerontologist 19 (4) (August 1979): 361-367; A.S. Kraus, R.A. Spasoff, E.J. Beattle, D.E.W. Holden, J.S. Lawson, M. Rodenburg, and G.M. Woodcock, "Elderly Applicants to Long-Term Care Institutions," Journal of American Geriatric Society 14 (3) (Fall 1976): 117-25; Jay N. Greenberg and Anna Ginn, “A Multivariate Analysis of the Predictors of Long-Term Care Placement," Home Health Care Services Quarterly 1 (1) (Spring 1979): 75-99; Erdman Palmore, "Total Chance of Institutionalization Among the Aged," The Gerontologist 16 (6) (December 1976): 504-7.

9. Marc A. Cohen, Eileen J. Tell, and Stanley J. Wallack, "The Lifetime Risks and Costs of Nursing Home Use Among the Elderly," Medical Care 24 (12) (December 1986): 1162; Lawrence G. Branch and Alan M. Jette, "A Prospective Study of the Long-Term Care Institutionalization Among the Aged," American Journal of Public Health 72 (12) (December 1982): 1374.

10. We would like to acknowledge the direction and support of the assistant secretary for planning and evaluation, HHS, in the modeling work.

11. Barry Chiswick, "The Demand for Nursing Home Care: An Analysis of the Substitution Between Institutional and Non-Institutional Care," Journal of Human Resources 11 (3) (Summer 1976): 295-316; Alvin Headen, "InsuranceInduced Demand and the Hazard of Nursing Home Entry," North Carolina State University and Duke University Center of Demographic Studies, Faculty Working Paper No. 152, 1989; John Nyman, "The Private Demand for Nursing

« PreviousContinue »