APPENDIX B. HEALTH STATUS, INSURANCE, AND EXPENDITURES OF THE ELDERLY, AND BACKGROUND DATA ON LONG-TERM CARE Although the health status of the elderly appears to have been improving in recent decades, many elderly persons have conditions that require medical and long-term health care, sometimes in substantial amounts. Nearly all elderly persons have some insurance that protects them, at least partially, from the expenses arising from health care use. Many are well insured for their acute care needs--that is, for hospital and physician services. Others face greater risk of high out-of- pocket expenditures. This appendix reports on the health status, health insurance, and health care expenditures of the elderly. HEALTH STATUS By various measures, the health status of the elderly population has been improving over the years. For example, life expectancy at age 65 has increased from 13.9 years in 1950 to 17.2 years in 1989 (see table B-1). The improvements in life expectancy--or, alternatively, the declines in mortality rates--have been greater for females than for males. Morbidity indicators--such as the incidence of high blood pressure--also improved among those aged 65 to 74 years between the early 1960's and the late 1970's (see table B-2). TABLE B-1.--LIFE EXPECTANCY AT BIRTH AND AT 65 YEARS OF AGE, BY SEX, BY RACE, UNITED STATES, SELECTED YEARS 1900-90 [Remaining life expectancy in years] ---------------------------------------------------------------------------------------------------------------- At birth At 65 years At birth --------------------------------------------------------------- Year Both Female Both Female sexes Male sexes Male White Black ---------------------------------------------------------------------------------------------------------------- 1900\1\\2\...................................... 47.3 46.3 48.3 11.9 11.5 12.2 47.6 \3\33. 0 1950\2\......................................... 68.2 65.6 71.1 13.9 12.8 15.0 69.1 60.7 1960\2\......................................... 69.7 66.6 73.1 14.3 12.8 15.8 70.6 63.2 1970............................................ 70.9 67.1 74.8 15.2 13.1 17.0 71.7 64.1 1980............................................ 73.7 70.0 77.4 16.4 14.1 18.3 74.4 68.1 1984............................................ 74.7 71.2 78.2 16.8 14.6 18.6 75.3 69.7 1985............................................ 74.7 71.2 78.2 16.7 14.6 18.6 75.3 69.5 1986............................................ 74.8 71.3 78.3 16.8 14.7 18.6 75.4 69.4 1987............................................ 75.0 71.5 78.4 16.9 14.8 18.7 75.6 69.4 1988............................................ 74.9 71.5 78.3 16.9 14.9 18.6 75.6 69.2 1989............................................ 75.3 71.8 78.6 17.2 15.2 18.8 76.0 69.2 Provisional data: 1988\2\..................................... 74.9 71.4 78.3 16.9 14.8 18.6 75.5 69.5 1989\2\..................................... 75.2 71.8 78.5 17.2 15.2 18.8 75.9 69.7 1990\2\..................................... 75.4 72.0 78.8 17.3 15.3 19.0 76.0 70.3 ---------------------------------------------------------------------------------------------------------------- \1\Death registration area only; includes 10 States and the District of Columbia. \2\Includes deaths of nonresidents of the United States. \3\Figure is for the all other population. Source: National Center for Health Statistics, Health, United States, 1989, Hyattsville, Maryland: Public Health Service, 1990. TABLE B-2.--SELECTED HEALTH STATUS INDICATORS FOR PERSONS 65-74 YEARS OF AGE, BY SEX, 1960-62, 1971-74, AND 1976-80 [Percent of population] ---------------------------------------------------------------------------------------------------------------- Both sexes Male Female Health status ----------------------------------------------------------------------------------------- indicator 1960-62 1971-74 1976-80 1960-62 1971-74 1976-80 1960-62 1971-74 1976-80 ---------------------------------------------------------------------------------------------------------------- Borderline or definite elevated blood pressure\1\.......... 73.8 70.3 63.1 65.9 65.4 62.0 80.3 74.1 63.9 Definite elevated blood pressure\2\.... 48.7 40.9 34.5 40.5 36.4 33.3 55.4 44.4 35.5 High-risk serum cholesterol levels\3\ 37.3 31.3 27.2 20.8 19.9 18.1 50.8 40.0 34.3 Overweight\4\......... 34.6 31.5 32.7 23.8 23.0 25.2 43.3 38.0 38.5 ---------------------------------------------------------------------------------------------------------------- \1\Borderline or definite elevated blood pressure is defined as either systolic pressure of at least 140 mmHg or diastolic pressure of at least 90 mmHg or both based on a single measurement. \2\Definite elevated blood pressure is defined as either systolic pressure of at least 160 mmHg or diastolic pressure of at least 95 mmHg or both based on a single measurement. \3\High-risk serum cholesterol levels are defined by age-specific cut points of the cholesterol distribution. For 40 years of age and over, high risk is greater than 260 milligrams/deciliter. Risk levels defined by NIH Consensus Development conference statement on lowering blood cholesterol, December 10, 1984. \4\Overweight is defined for men as body mass index greater than or equal to 27.8 kilograms/meter\2\, and for women as body mass index greater than or equal to 27.3 milograms/meter\2\. These cut points were used because they represent the sex-specific 85th percentiles for persons 20-29 years of age in the 1976-80 National Health and Nutrition Examination Survey. Source: National Center for Health Statistics, Health, United States, 1985, DHHS Pub. No. (PHS) 86-1232, pp. 76- 79. Data are based on physical examinations of a sample of the civilian, noninstitutionalized population. Despite the trend toward improved health status of the elderly, their needs for medical and long-term care services remain substantial. First, greater life expectancy postpones the probable need for terminal illness care. (About two-thirds of the deaths in the United States are of the elderly. A recent study found that the 6 percent of Medicare beneficiaries who died in 1978 accounted for 28 percent of Medicare expenditures.\1\) Second, many of the elderly have one or more chronic conditions, many of which give rise to the need for continuing health care. Table 3 shows the incidence of several common chronic conditions among the elderly. Nearly half report having arthritis, about 40 percent report high blood pressure, and almost 30 percent report heart disease. The incidence of many chronic conditions is directly related to age and inversely related to family income. --------------------------------------------------------------------------- \1\J. Lubitz and R. Prihoda, ``The Use and Costs of Medicare Services in the Last Two Years of Life,'' Health Care Financing Review, Volume 5, 1984, pp. 117-131. --------------------------------------------------------------------------- Self-assessed health is a common method used to measure health status, with responses ranging from ``excellent'' to ``poor.'' Nearly 71 percent of elderly people living in the community describe their health as excellent, very good, or good, compared with others their age; only 29 percent report that their health is fair or good (see table B-4). Income is directly related to one's perception of his or her health. About 26 percent of older people with incomes over $35,000 described their health as excellent compared to others their age, while only 10 percent of those with low incomes (less than $10,000) reported excellent health. TABLE B-3.--SELECTED CHRONIC CONDITIONS PER 1,000 ELDERLY PERSONS, BY AGE AND FAMILY INCOME, 1988 ---------------------------------------------------------------------------------------------------------------- Age Family income --------------------------------------------------- Chronic condition All Less $10,000 $20,000 $35,000 elderly 65-74 75 and than to to and over $10,000 $19,999 $34,999 over ---------------------------------------------------------------------------------------------------------------- Arthritis.......................................... 486 445 550 608 452 471 397 Cataracts.......................................... 168 118 246 183 174 131 150 Hearing impairment................................. 315 274 381 308 364 259 314 Deformity or orthopedic impairment................. 161 151 177 182 179 136 140 Hernia of abdominal cavity......................... 58 54 64 72 67 46 51 Diabetes........................................... 92 95 88 98 101 76 71 Heart disease...................................... 296 272 334 346 324 269 257 High blood pressure................................ 373 373 374 472 396 345 321 Emphysema.......................................... 38 36 41 52 48 34 (\1\) ---------------------------------------------------------------------------------------------------------------- \1\Sample size is too small for reliable estimate. Source: U.S. Department of Health and Human Services, National Center for Health Statistics, Vital and Health Statistics: Current estimates from the National Health Interview Survey, 1988, Series 10, No. 173, October 1989. TABLE B-4.--SELF-ASSESSED HEALTH STATUS OF THE ELDERLY, BY FAMILY INCOME, 1989 [In percent] ---------------------------------------------------------------------------------------------------------------- Self-assessed health status\2\ All All health -------------------------------------- Characteristic persons\1\ status\3\ Excellent Very (thousands) good Good Fair Poor ---------------------------------------------------------------------------------------------------------------- All persons 65\4\............................... 29,219 100.0 16.4 23.1 31.9 19.3 9.2 Sex: Men......................................... 12,143 100.0 16.9 23.2 30.8 18.4 10.7 Women....................................... 17,076 100.0 16.1 23.0 32.8 20.0 8.1 Family income: Under $10,000............................... 5,612 100.0 10.3 19.4 29.7 25.0 15.6 $10,000 to $19,999.......................... 8,002 100.0 14.8 21.7 33.9 21.1 8.5 $20,000 to $34,999.......................... 5,242 100.0 20.2 25.7 32.5 15.7 5.9 $35,000 and over............................ 3,484 100.0 26.0 26.8 30.3 11.7 5.1 ---------------------------------------------------------------------------------------------------------------- \1\Includes unknown health status. \2\Excludes unknown health status. \3\The categories related to this concept result from asking the respondent, ``Would you say--health is excellent, very good, good, fair, or poor?'' As such, it is based on the respondent's opinion and not directly on any clinical evidence. \4\Includes unknown family income. Note.--Percentages may not add to 100 percent due to rounding. Source: National Center for Health Statistics. ``Current Estimates from the National Health Interview Survey, 1989.'' Vital and Health Statistics Series 10, No. 176 (October 1990). Data are based on household interviews of the civilian, noninstitutionalized population. CAUSES OF DEATH FOR THE ELDERLY\2\ In the United States, about 7 out of every 10 elderly persons die from heart disease, cancer, or stroke. Heart disease was the major cause of death in 1950, and remains so today even though there have been rapid declines in death rates from heart disease since 1968, especially among females. Death rates from cancer continue to rise in comparison to heart disease, especially deaths caused by lung cancer (chart B-1). In 1988, however, heart disease accounted for 40 percent of all deaths among persons 65+, while cancer accounted for 21 percent of all deaths in this age group.\3\ Even if cancer were eliminated as a cause of death, the average life span would be extended by less than 2 years because of the prevalence of heart disease. Eliminating deaths due to heart disease, on the other hand, would add an average of 5 years to life expectancy at age 65, and would lead to a sharp increase in the proportion of older persons in the total population.\4\ --------------------------------------------------------------------------- \2\This entire section is from Aging America: Trends and Projections, 1987-88 edition. \3\National Center for Health Statistics. ``Annual Summary of Births, Marriages, Divorces, and Deaths: United States, 1985.'' Monthly Vital Statistics Report Vol. 34, No. 13 (September 1986). \4\National Center for Health Statistics. ``United States Life Tables Eliminating Certain Causes of Death.'' U.S. Decennial Life Tables for 1979-81 Vol. 1, No. 2 (forthcoming). --------------------------------------------------------------------------- CHART B-1. DEATH RATES FOR LEADING CAUSES OF DEATH FOR PEOPLE AGE 75-84: 1950-89 The third leading cause of death among the elderly--stroke (cerebrovascular disease)--has been decreasing over the past 30 years. Reasons for this dramatic decline are not fully understood. Part of the decline may be attributable to better control of hypertension. Better diagnosis and improved management and rehabilitation of stroke victims may also be related factors.\5\ In 1988, cerebrovascular disease accounted for only 8 percent of all deaths in the 65+ age group. --------------------------------------------------------------------------- \5\National Center for Health Statistics. Health, United States, 1985. DHHS Pub. No. (PHS) 86-1232, Washington: Department of Health and Human Services, December 1985. --------------------------------------------------------------------------- Table B-5 shows the 10 leading causes of death for three subgroups of the older population. The factors which have led to reductions in mortality may or may not also lead to overall improvements in health status. If Americans continue to live only to about age 85, control of life-threatening disease could produce a healthier older population. But, if the life-span is increased dramatically in future years beyond age 85, the onset of illness may only be delayed, without an actual shortening of the period of illness. TABLE B-5.--DEATH RATES FOR TEN LEADING CAUSES OF DEATH AMONG OLDER PEOPLE, BY AGE: 1988 [Rates per 100,000 population in age group] ------------------------------------------------------------------------ Cause of death 65+ 65-74 75-84 85+ ------------------------------------------------------------------------ All causes.............. 5,105 2,730 6,321 15,594 --------------------------------------- Diseases of the heart........... 2,066 984 2,543 7,098 Malignant neoplasms............. 1,068 843 1,313 1,639 Cerebrovascular diseases........ 431 155 554 1,707 Chronic obstructive pulmonary diseases....................... 226 152 313 394 Pneumonia and influenza......... 225 60 257 1,125 Diabetes........................ 97 62 125 222 Accidents....................... 89 50 107 267 Atherosclerosis................. 69 15 70 396 Nephritis, nephrotic syndrome, nephrosis...................... 61 26 78 217 Septicemia...................... 56 24 71 199 All other causes................ 717 359 890 2,330 ------------------------------------------------------------------------ Source: National Center for Health Statistics. ``Advanced Report of Final Mortality Statistics, 1988.'' Monthly Vital Statistics Report Vol. 39, No. 7, Supplement (November 28, 1990). MEDICARE REIMBURSEMENT AND OUT-OF-POCKET LIABILITIES OF THE ELDERLY Tables B-6 through B-8 illustrate for 6 selected years how Medicare reimbursement, acute health care costs, and out-of- pocket liabilities of Medicare enrollees have changed. The years chosen are 1975, 1980, 1985, 1990, 1995, and 2000 (projected values). Constant 1990 dollar values were obtained using the CPI-U. The fastest-growing component of Medicare reimbursement is for benefits under the Supplementary Medical Insurance (SMI) program. For SMI, reimbursements increase at an annual rate of 13.3 percent, while the growth in total costs (including enrollees' share of costs) is 11.3 percent (see table B-6). As a result, the share of SMI costs reimbursed by Medicare increases significantly over the period--from about 64 percent in 1975 to about 74 percent by 1990. Through 1985, the growth in Medicare's share is due to the declining significance of the SMI deductible, so that more enrollees' costs were eligible for reimbursement. In the Hospital Insurance (HI) program, by contrast, the rate of growth in reimbursement is slower than the growth in enrollee's copayment costs. Consequently, the share of HI costs reimbursed by Medicare has decreased from 93 percent in 1975 to 91 percent in 1990. Overall, the share of costs reimbursed by Medicare has increased slightly. The percentage of costs paid by Medicare for services covered under Medicare was 82.2 percent in 1975 and 83.4 percent in 1990 (see table B-6). The other side of this--the share of costs paid directly by enrollees--is shown in the third panel of table B-7. Total direct costs plus Medicare reimbursement equals the total or 100 percent. TABLE B-6.--REIMBURSEMENTS AND OUT-OF-POCKET COSTS UNDER MEDICARE, SELECTED CALENDAR YEARS [Incurred costs per HI or SMI enrollee] ---------------------------------------------------------------------------------------------------------------- Annual growth 1975 1980 1985 1990 1995 2000 1975-2000 (percent) ---------------------------------------------------------------------------------------------------------------- In current dollars -------------------------------------------------------------- Hospital insurance: Reimbursement................................ $458 $906 $1,539 $1,959 $3,027 $4,385 9.5 Copayments................................... 34 66 117 188 246 329 9.5 -------------------------------------------------------------- Total...................................... 492 972 1,656 2,146 3,273 4,714 9.5 ============================================================== Supplementary medical insurance: Reimbursement................................ 184 402 763 1,298 1,951 3,309 12.2 Copayments................................... 83 138 246 394 544 864 9.8 Balance-billing.............................. 22 56 87 68 42 67 4.6 -------------------------------------------------------------- Total...................................... 289 597 1,096 1,760 2,537 4,240 11.3 ============================================================== Total Medicare reimbursement..................... 642 1,308 2,302 3,257 4,978 7,694 10.4 Total costs under Medicare....................... 781 1,569 2,752 3,906 5,810 8,954 10.2 ------------------------------------------------------------------------------------------ In constant 1990 dollars ------------------------------------------------------------------------------------------ Hospital insurance: Reimbursement............................................ 1,065 1,439 1,870 1,959 2,589 3,221 4.5 Copayments............................................... 79 104 143 188 210 242 4.6 ------------------------------------------------------------------------------------------ Total.................................................. 1,144 1,543 2,012 2,146 2,800 3,463 4.5 ========================================================================================== Supplementary medical insurance: Reimbursement............................................ 428 639 927 1,298 1,669 2,431 7.2 Copayments............................................... 193 220 299 394 465 635 4.9 Balance-billing.......................................... 51 89 106 68 36 49 -0.2 ------------------------------------------------------------------------------------------ Total.................................................. 672 947 1,332 1,760 2,170 3,115 6.3 ========================================================================================== Total Medicare reimbursement................................. 1,493 2,077 2,797 3,257 4,258 5,752 5.5 Total costs under Medicare................................... 1,816 2,490 3,344 3,906 4,970 6,578 5.3 ========================================================================================== Percent of costs paid by Medicare............................ 82.2 83.4 83.6 83.4 85.7 85.9 0.2 -------------------------------------------------------------------------------------------------------------------------------------------------------- Note.--1995 values are projected. The CPI-U was used to obtain constant dollars. Source: Congressional Budget Office (February 1993 baseline). TABLE B-7.--ENROLLEE COSTS UNDER MEDICARE, SELECTED CALENDAR YEARS [Incurred costs per HI or SMI enrollee] ---------------------------------------------------------------------------------------------------------------- Annual growth 1975 1980 1985 1990 1995 2000 1975-2000 (percent) ---------------------------------------------------------------------------------------------------------------- In current dollars ----------------------------------------------------------------- HI copayments................................. $34 $66 $117 $188 $246 $329 9.5 SMI copayments................................ 83 138 246 394 544 864 9.8 Balance-billing............................... 22 56 87 68 42 67 4.6 ----------------------------------------------------------------- Total direct costs...................... 139 260 451 649 832 1,260 9.2 Premium costs................................. 80 110 186 343 553 728 9.2 ----------------------------------------------------------------- Total enrollee costs.................... 219 371 637 993 1,385 1,988 9.2 ================================================================= Enrollee per capita income\1\................................... 5,158 8,431 12,767 15,454 19,141 23,074 6.2 ----------------------------------------------------------------- In constant 1990 dollars ----------------------------------------------------------------- HI copayments................................. 79 104 143 188 210 242 4.6 SMI copayments................................ 193 220 299 394 465 635 4.9 Balance-billing............................... 51 89 106 68 36 49 -0.2 ----------------------------------------------------------------- Total direct costs...................... 323 413 547 649 712 926 4.3 Premium costs................................. 187 175 226 343 473 535 4.3 ----------------------------------------------------------------- Total enrollee costs.................... 510 588 773 993 1,185 1,461 4.3 ----------------------------------------------------------------- Enrollee per capita income\1\................................... 11,998 13,386 15,513 15,454 16,374 16,951 1.4 ----------------------------------------------------------------- Percent of costs under Medicare paid by enrollees, by source of payment ----------------------------------------------------------------- HI copayments................................. 4.3 4.2 4.3 4.8 4.2 3.7 -0.7 SMI copayments................................ 10.6 8.8 8.9 10.1 9.4 9.6 -0.4 Balance-billing............................... 2.8 3.6 3.2 1.7 0.7 0.7 -5.2 ----------------------------------------------------------------- Total direct costs...................... 17.8 16.6 16.4 16.6 14.3 14.1 -0.9 Premium costs................................. 10.3 7.0 6.8 8.8 9.5 8.1 -0.9 ----------------------------------------------------------------- Total................................... 28.1 23.6 23.1 25.4 23.8 22.2 -0.9 Enrollee-paid costs as a percent of enrollee per capita income\1\......................... 4.3 4.4 5.0 6.4 7.2 8.6 2.9 ---------------------------------------------------------------------------------------------------------------- \1\From Current Population Survey, adjusted for underreporting. Note.--1995 values are projected. The CPI-U was used to obtain constant dollars. Source: Congressional Budget Office (February 1994 baseline). In constant dollars, HI copayments have increased the most rapidly between 1975 and 1995. However, between 1990 and 1995, premium costs are expected to rise the most rapidly due equally to copayments and premiums. In contrast, the cost to the enrollee from balance-billing has decreased significantly since 1985--a direct policy result of the participating physician program and the imposition of lower limits on balance billing. See table B-8 for deductible amounts and monthly premium amounts under Medicare. Enrollees are spending an increasing share of their income for health care. In 1975, about 4.3 percent of enrollees' per capita income went to cover their share of acute health care costs under Medicare. By 2000, enrollees will have to pay an estimated 8.6 percent of their per capita income to cover their share of costs under Medicare. Although direct household spending for health care by elderly households, that is households headed by a person 65 or older, as a share of household income has increased since the early 1970's, it has remained relatively stable in recent years. Chart B-2 illustrates direct household spending for health care as a percentage of household income before taxes for elderly and nonelderly households for years 1984 through 1992. In 1992, direct household spending for health care as a percentage of household income for elderly households was 11.9 percent, on average, up slightly from 10.6 percent in 1984. Over the same period, nonelderly households spent around 3.5 percent of their household income for health care. TABLE B-8.--COPAYMENT AND PREMIUM VALUES UNDER MEDICARE, SELECTED CALENDAR YEARS ---------------------------------------------------------------------------------------------------------------- Annual growth 1975 1980 1985 1990 1995 2000 1975-2000 (in percent) ---------------------------------------------------------------------------------------------------------------- In current dollars ---------------------------------------------------------------- Hospital insurance: Hospital deductible........................ $92 $180 $400 $592 $720 $916 9.6 Supplementary medical insurance: Annual deductible.......................... 60 60 75 75 100 100 2.1 Monthly premium\1\......................... 6.70 9.20 15.50 28.60 46.10 60.70 9.2 ---------------------------------------------------------------- In constant 1990 dollars ---------------------------------------------------------------- Hospital insurance: Hospital deductible........................ 214 286 486 592 616 673 4.7 Supplementary medical insurance: Annual deductible.......................... 139 95 91 75 86 73 -2.5 Monthly premium\1\......................... 15.57 14.61 18.83 28.60 39.43 44.59 4.3 ---------------------------------------------------------------------------------------------------------------- \1\The 1980 SMI monthly premium amount is the average of values for the first and second halves of the year. Note:--Values after 1990 are projected. The CPI-U was used to get constant dollars. Source: Congressional Budget Office (February 1994 baseline). CHART B-2. DIRECT HOUSEHOLD SPENDING FOR HEALTH CARE AS A PERCENTAGE OF HOUSEHOLD INCOME BY TYPE OF HOUSEHOLD, 1984-92 ANALYZING TRENDS IN MEDICARE SPENDING, 1967-98\6\ Between 1980 and 1985, total Medicare spending for hospital inpatient services grew at an annual rate of 14.6 percent. The estimated growth rate for 1985 to 1992 is 6.8 percent. The difference in these rates is due to changes in four separate trends: Medicare enrollment, admissions per enrollee, real expenditures per admission, and the general rate of inflation. --------------------------------------------------------------------------- \6\The following section borrows heavily from a memorandum prepared by Sandra Christensen, of the Congressional Budget Office, February 4, 1991. Updated April 1992. --------------------------------------------------------------------------- Reduced inflation contributes to the lower rate of growth in total Medicare inpatient spending. General inflation is estimated at 3.9 percent per year from 1985 to 1992, compared with 5.6 percent for 1980 to 1985. The growth rate for 1985 to 1992 would thus be about 1.7 percentage points higher at the previous rate of inflation. Real Medicare inpatient spending per enrollee removes the effects of changes in Medicare enrollment and general inflation from total Medicare inpatient spending (see table B-9). Since both enrollment and prices are almost always increasing, the growth of real per enrollee spending is slower than the growth of total spending. Real inpatient spending per enrollee grew at an annual rate of 6.4 percent between 1980 and 1985, and the estimate for 1985 to 1992 is 0.1 percent. The difference in these rates is due to changes in admissions per enrollee and real expenditures per admission. The number of Medicare enrollees grew at an annual rate of 1.7 percent between 1980 and 1985, and the estimate for 1985 to 1991 is about the same. Medicare enrollment thus makes no contribution to the observed difference in spending growth between the early and late 1980's. The trend in admissions per enrollee did change, however. In 1984, Medicare's peer review organizations were set up to monitor inpatient cases for appropriateness of treatment and site of care. Simultaneously, admission rates among the Medicare population--which had been increasing through 1983-- began to decline. Although admission rates inched up again after 1987, rates in 1989 for people age 65 or more (a proxy for the Medicare population) were still only 85 percent of rates in 1983. Perhaps Medicare's preadmission approval requirements for certain procedures, coupled with retrospective payment denials for care deemed inappropriate, encouraged physicians either to forgo some elective procedures for their Medicare patients or to move them to the outpatient sector. It should be noted that admissions for the non-Medicare population decreased for each year since 1981. Given this trend, some credit for lower admissions rates must go to changes in practice patterns and other factors not associated with Medicare policy. A reduction in real expenditures per admission makes the greatest contribution to decreased spending growth. This decline is primarily due to smaller increases in payment rates under PPS since the very large increases in the first 2 years (1984 and 1985). At the previous rate of increase in Medicare expenditures per admission, the estimated growth in total inpatient spending between 1985 and 1991 would be 12.0 percent per year, rather than 5.1 percent. The estimated real growth in spending per enrollee would have been 6.4 percent per year, rather than -0.3 percent. Costs in hospital outpatient departments have dropped relative to the previous trend, indicating that hospital inpatient costs have not simply been shifted to the outpatient sector. Savings relative to trend for hospital outpatient and home health services may in large part reflect unsustainably large rates of growth during the trend period from 1975 through 1980. Introduction of a new payment methodology (a blend of a fixed rate and the hospital's costs) for certain surgical procedures performed in outpatient departments tended to reduce costs somewhat, but this effect was partially offset by the shift of services from the inpatient sector. During the 1980s, Medicare's administrative agents implemented stricter standards for determining coverage of home health services (tending to reduce costs), but increased demand for services from patients discharged earlier from hospitals than they would have been prior to the prospective payment system would have worked to increase Medicare's spending for home health. Growth in spending for physicians' services has not slowed as much as hospital spending relative to previous trends despite the disproportionate impact on physicians of budget reconciliation bills. Apparently, growth in the volume of physicians' services has accelerated by enough to offset some of the enacted reductions in payment rates. Although not all of this growth was in response to fee cuts, growth in the volume of services was enough to completely offset the fee freeze in place from 1984 through 1986, but was insufficient to offset entirely the effects of subsequent fee cuts for ``overvalued'' procedures. Spending for skilled nursing facilities (SNFs) increased significantly. During the period from 1975 through 1980, real spending per enrollee for SNFs was falling. This trend was reversed during the 1980s. In 1988, growth in SNF spending accelerated sharply because of a revision in the manual used by administrative agents to determine Medicare coverage that greatly relaxed the definition of covered care to make it conform with legislative language. Growth in SNF spending further accelerated in 1989 under provisions of the Medicare Catastrophic Coverage Act, which briefly eliminated the requirement for a hospital stay prior to a covered SNF stay and which reduced the copayments required of enrollees for SNF stays. Table B-9 shows Medicare spending per enrollee in constant 1990 dollars where the CPI-U has been used to obtain constant dollars. The first column includes both Medicare benefits and administration. All other columns include spending on benefits only. TABLE B-9.--REAL SPENDING PER ENROLLEE [Fiscal years, in constant 1990 dollars] ---------------------------------------------------------------------------------------------------------------- Medicare Hospital HH & Physician Hospital Fiscal years Bft+Adm HI Bft SMI BFT inpatient SNF Hospice OPD & Lab Inp+OPD ---------------------------------------------------------------------------------------------------------------- ESTIMATES BY THE HEALTH CARE FINANCING ADMINISTRATION ------------------------------------------------------------------------------------------- 1967................ 648 470 134 449 18 4 3 130 451 1968................ 974 669 264 599 62 10 8 254 608 1969................ 1,138 792 293 721 62 12 13 276 735 1970................ 1,155 768 327 712 48 13 17 306 728 1971................ 1,194 818 315 779 32 12 21 289 800 1972................ 1,269 872 331 841 25 12 26 300 867 1973................ 1,244 861 324 831 23 12 25 294 856 1974................ 1,292 878 337 845 24 17 36 292 881 1975................ 1,479 1,030 382 989 27 21 54 321 1,043 1976................ 1,615 1,108 431 1,061 28 30 67 353 1,128 1977................ 1,756 1,208 484 1,156 28 35 80 393 1,236 1978................ 1,877 1,289 516 1,237 26 39 90 413 1,327 1979................ 1,941 1,317 557 1,264 25 42 98 445 1,362 1980................ 2,051 1,386 600 1,333 23 44 107 480 1,439 1981................ 2,216 1,501 650 1,447 21 46 117 521 1,563 1982................ 2,414 1,637 715 1,563 22 56 138 574 1,702 1983................ 2,565 1,710 795 1,620 24 66 152 642 1,772 1984................ 2,654 1,755 835 1,659 23 74 151 682 1,810 1985................ 2,876 1,921 885 1,821 22 80 159 725 1,980 1986................ 2,924 1,880 978 1,781 22 78 192 785 1,973 1987................ 3,002 1,829 1,110 1,735 23 72 215 894 1,950 1988................ 3,036 1,787 1,180 1,688 25 75 227 952 1,915 1989................ 3,133 1,845 1,215 1,691 75 81 241 972 1,932 1990................ 3,328 1,976 1,282 1,780 85 112 258 1,021 2,039 1991................ 3,303 1,918 1,315 1,695 70 155 267 1,047 1,962 1992................ 3,561 2,147 1,337 1,840 98 212 293 1,041 2,133 ------------------------------------------------------------------------------------------- ESTIMATES BY THE CONGRESSIONAL BUDGET OFFICE ------------------------------------------------------------------------------------------- 1993................ 3,734 2,286 1,371 1,889 134 264 319 1,050 2,208 1994................ 3,993 2,439 1,477 1,971 160 311 351 1,124 2,322 1995................ 4,216 2,546 1,593 2,032 171 345 387 1,204 2,419 1996................ 4,421 2,624 1,720 2,078 178 371 430 1,287 2,507 1997................ 4,865 2,926 1,860 2,307 198 424 476 1,381 2,782 1998................ 5,117 3,036 2,002 2,392 203 444 527 1,472 2,919 1999................ 5,410 3,176 2,154 2,510 207 463 588 1,562 3,098 2000................ 5,729 3,327 2,323 2,638 211 483 660 1,659 3,298 ------------------------------------------------------------------------------------------- AVERAGE ANNUAL GROWTH RATES (In percents) ------------------------------------------------------------------------------------------- 1975-80............. 6.8 6.1 9.5 6.1 -3.4 16.3 14.7 8.4 6.7 1980-85............. 7.0 6.7 8.1 6.4 -0.7 12.5 8.3 8.6 6.6 1985-90............. 3.0 0.6 7.7 -0.4 31.0 7.1 10.2 7.1 0.6 1990-95............. 4.8 5.2 4.4 2.7 15.0 25.2 8.4 3.3 3.5 1995-2000........... 6.3 5.5 7.8 5.4 4.2 6.9 11.3 6.6 6.4 ---------------------------------------------------------------------------------------------------------------- Notes.--Column 1 includes both benefits and administrative costs. All other columns include only benefits. The CPI-U was used to obtain constant dollars. Source: Congressional Budget Office (February 1994). From 1975 to 1985, total real spending per enrollee grew at an annual rate of 7.0 percent. From 1985 to 1990, there was a dramatic decline in the real growth rate in HI expenditures per capita due mostly to a drop in the inpatient hospital growth rate. This growth rate fell from 6.4 percent to -0.3 percent between the first 5 years of the 1980's and the subsequent years. While the outpatient growth rate increased slightly, the total real hospital spending growth rate declined from 6.5 percent annually to 0.7 percent between 1980 to 1985 as compared with 1985 to 1990. This decline in the hospital spending growth rate results in a 3.9 percentage point reduction in the total Medicare spending growth rate--a decline of 56 percent. If the total growth rate in Medicare spending continued between 1985 and 1990 at the same 7.0 percent rate exhibited between 1980 and 1985, total Medicare costs per enrollee would be $4,282 in 1991, or almost $1,000 per enrollee more than the actual estimate. This would imply additional Medicare spending of about $34 billion in that year. TOTAL HEALTH CARE EXPENDITURES FOR THE ELDERLY Expenditures for personal health care services for the elderly nearly quadrupled between 1977 and 1987, rising from $43 billion to an estimated $162 billion (see table B-10). Government programs (Federal and State) account for two- thirds of estimated 1987 spending for the aged (see table B- 10). The most significant of these programs is Medicare which pays for nearly half of the aged's health bill. Medicaid funds about 12 percent of the expenditures. Health insurance coverage of the elderly Table B-11 shows the sources of health insurance coverage for the noninstitutionalized population aged 65 and over in 1992. Over 95 percent of the aged population was enrolled in Medicare, and more than three-quarters of the Medicare enrollees had some form of supplemental coverage. Beneficiaries with incomes below the Federal poverty level were least likely to have supplemental coverage; those who had such coverage were more likely to rely on Medicaid. Higher income groups were more likely to obtain supplemental coverage through individually purchased medigap policies or through employer-based plans. Of those with incomes greater than 200 percent of the poverty level, 41.3 percent had employer coverage, compared to just 5.4 percent of those below poverty. (It should be noted that the Current Population Survey (CPS), on which table 11 is based, does not distinguish between primary and secondary sources of coverage. Some of the individuals reporting both Medicare and employer-based plans relied on the employer plan as their primary insurer, with Medicare functioning as a secondary payer.) About 3.6 percent of the elderly had more than oneP TABLE B-10.--PERSONAL HEALTH CARE EXPENDITURES FOR PEOPLE 65 YEARS OF AGE OR OVER, BY SOURCE OF FUNDS AND TYPE OF SERVICE, 1977, 1984, AND 1987 [In millions of dollars] ------------------------------------------------------------------------ Type of service Year and source of -------------------------------------------------- funds Total Nursing Other care Hospital Physician home care ------------------------------------------------------------------------ 1977 Total................ 43,425 18,906 7,782 10,696 6,041 Private.............. 15,669 2,319 3,323 5,424 4,603 Consumer........... 15,499 2,263 3,320 5,352 4,564 Out-of-pocket.... 12,706 927 2,147 5,264 4,368 Insurance........ 2,793 1,336 1,173 88 195 Other private...... 170 56 3 72 39 Government........... 27,756 16,587 4,458 5,272 1,438 Medicare........... 19,171 14,087 4,158 348 578 Medicaid........... 6,049 733 232 4,453 631 Other government... 2,536 1,767 68 470 230 1984 Total................ 119,872 54,200 24,770 25,105 15,798 Private.............. 39,341 6,160 9,827 13,038 10,316 Consumer........... 38,875 5,964 9,818 12,856 10,237 Out-of-pocket.... 30,198 1,694 6,468 12,569 9,467 Insurance........ 8,677 4,270 3,350 287 770 Other private...... 466 196 9 182 79 Government........... 80,531 48,040 14,943 12,067 5,482 Medicare........... 58,519 40,524 14,314 539 3,142 Medicaid........... 15,288 2,595 467 10,418 1,808 Other government... 6,724 4,920 162 1,110 532 1987 Total................ 162,000 67,900 33,500 32,800 27,800 Private.............. 60,600 10,100 11,900 19,200 19,500 Government........... 101,500 57,900 21,600 13,600 8,300 Medicare........... 72,200 47,300 20,300 600 4,100 Medicaid........... 19,500 3,300 500 11,900 3,700 ------------------------------------------------------------------------ Source: Office of Financial and Actuarial Analysis, Health Care Financing Administration as reported in Waldo, Daniel R., and Helen C. Lazenby. ``Demographic characteristics and health care use and expenditures by the aged in the United States: 1977-84.'' Health Care Financing Review, Fall 1984 No. 1, p. 1; and Waldo, Daniel R. et al. ``Health Expenditures by Age Group, 1977 and 1987.'' Health Care Financing Review, Summer 1989, Vol. 10, No. 4 and errata reprint Fall 1989, Vol. 11, No. 1, p. 167. TABLE B-11.--SOURCES OF HEALTH INSURANCE COVERAGE FOR THE NON-INSTITUTIONALIZED ELDERLY, BY RATIO OF INCOME TO POVERTY, 1992 [Population in thousands] ---------------------------------------------------------------------------------------------------------------- Individuals with family income-- Total ----------------------------------------------------------------------------- Under 100 percent 100-199 percent 200 percent of of poverty of poverty poverty or more ---------------------------------------------------------- Number Percent Number Percent Number Percent Number Percent ---------------------------------------------------------------------------------------------------------------- Total Medicare.................... 3,819 95.9 8,647 98.0 17,238 95.4 29,704 96.2 Medicare only................... 1,381 34.7 2,597 29.5 2,700 15.0 6,678 21.6 Medicare plus: Private supplement............ 961 24.1 3,395 38.5 5,739 31.7 10,095 32.7 Employer coverage............. 214 5.4 1,323 15.0 7,459 41.3 8,996 29.1 Medicaid...................... 1,099 27.6 855 9.7 413 2.3 2,367 7.6 CHAMPUS....................... 45 1.1 179 2.0 228 1.3 452 1.5 2 or more supplements......... 120 3.0 298 3.4 698 3.9 1,116 3.6 Insured through non-Medicare plan only............................. 42 1.0 89 1.0 678 3.8 809 2.6 Uninsured......................... 122 3.0 85 1.0 150 1.0 356 1.2 ----------------------------------------------------------------------------- Total....................... 3,983 100.0 8,822 100.0 18,065 100.0 30,870 100.0 ============================================================================= Percent of all elderly............ 12.9 28.5 58.5 100.0 ---------------------------------------------------------------------------------------------------------------- \1\Sample size too small for reliable estimates. Source: CRS analysis of data from the March 1993 Current Population Survey. source of supplemental coverage, such as both employer and individual medigap coverage, or both medigap and Medicaid. This figure does not include individuals who obtained multiple policies from a single basic coverage source, such as those who purchased more than one private medigap policy. About 1.2 million elderly persons did not report Medicare coverage in 1992. Of these, 809,000 had coverage from some other source. An estimated 25 percent of these are Federal annuitants who are covered through the Federal Employees Health Benefits Program (this estimate is based on unpublished data from the Office of Personnel Management). Approximately 356,000 persons aged 65 or over were without health insurance coverage in 1992. BACKGROUND DATA ON LONG-TERM CARE The phrase ``long-term care'' refers to a broad range of medical, social, personal, supportive, and specialized housing services needed by individuals who have lost some capacity for self-care because of a chronic illness or condition. Chronic illnesses or conditions often result in both functional impairment and physical dependence on others for an extended period of time. Major subgroups of persons needing long-term care include the elderly and nonelderly disabled, persons with developmental disabilities (primarily persons with mental retardation), and persons with mental illness. This section of appendix B focuses on the elderly long-term care population. The range of chronic illnesses and conditions resulting in the need for supportive long-term care services is extensive. Unlike acute medical illnesses, which occur suddenly and may be resolved in a relatively short period of time, chronic conditions last for an extended period of time and are not typically curable. Although chronic conditions occur in individuals of all ages, their incidence, especially as they result in disability, increases with age. These conditions may include heart disease, strokes, arthritis, osteoporosis, and vision and hearing impairments. Dementia, the chronic, often progressive loss of intellectual function, is also a major cause of disability in the elderly. The presence of a chronic illness or condition alone does not necessarily result in a need for long-term care. For many individuals, their illness or condition does not result in a functional impairment or dependence and they are able to go about their daily routines without needing assistance. It is when the illness or condition results in a functional or activity limitation that long-term care services may be required. The need for long-term care by the elderly is often measured by assessing limitations in a person's capacity to manage certain functions or activities. For example, a chronic condition may result in dependence in certain functions that are basic and essential for self-care, such as bathing, dressing, eating, toileting, and/or moving from one place to another. These are referred to as limitations in ``activities of daily living,'' or ADLs. Another set of limitations, which reflect lower levels of disability, are used to describe difficulties in performing household chores and social tasks. These are referred to as limitations in ``instrumental activities of daily living,'' or IADLs, and include such functions as meal preparation, cleaning, grocery shopping, managing money, and taking medicine. Limitations can vary in severity and prevalence, so that persons can have limitations in any number of ADLs or IADLs, or both. Long-term care services are often differentiated by the settings in which they are provided. In general, services are provided either in nursing homes or in home and community-based care settings. Nursing home care includes a wide variety of services that range from skilled nursing and therapy services to assistance with such personal care functions as bathing, dressing, and eating. Nursing home services also include room and board. All of these services are considered to be formally provided services, in that they require persons to pay the facility for care that is provided. Home and community-based care also includes a broad range of skilled and personal care services, as well as a variety of home management activities, such as chore services, meal preparation, and shopping. Home care services can be provided formally by home care agencies, visiting nurse associations, and day care centers. Home care is also provided informally by family and friends who are not paid for the services they provide. In contrast to nursing home care, which by necessity is formally provided care, most home and community-based care is provided informally by family and friends. Research has shown that more than 70 percent of those elderly persons living in the community and needing long-term care assistance rely exclusively on nonpaid sources of assistance for their care. The long-term care population Chart B-3 shows that an estimated 10.6 million persons of all ages require assistance with one or more ADLs or IADLs. About two-thirds of this total, or 7.1 million persons, are elderly. This is about one-quarter of the nation's elderly population. Another 3.5 million persons under the age of 65 are limited in ADLs and/or IADLs. Some of these persons have congenital or developmental conditions such as cerebral palsy or mental retardation. Others are disabled from traumatic accidents or the onset of chronic conditions such as multiple sclerosis.\7\ It should be noted that these estimates do not adequately measure the need for long-term care among young children, since ADL and IADL limitations are not appropriate measures of their disabilities. --------------------------------------------------------------------------- \7\A Call for Action, p. 91. --------------------------------------------------------------------------- Chart B-3 also indicates that the great majority of persons with ADL and/or IADL limitations live in the community. Of the total disabled population, 84 percent live in the community. The nursing home population amounts to only 16 percent of the total, with the elderly by far the greatest share of this group. Based on the projected growth of the elderly population in the future, major increases can be anticipated in the number of persons needing assistance with ADL and/or IADL limitations. Currently 32 million persons are 65 years of age and older. That number is expected to double to about 66 million by the year 2030. The 85+ population, the group at greatest risk of needing and using long-term care services, is expected to increase from 3.3 million persons in 1990 to 8.1 million in 2030.\8\ One study has estimated that the number of elderly needing assistance with ADLs and/or IADLs will grow from 7.1 million to 13.8 million by 2030, and the number requiring nursing home care will grow from 1.5 million to 5.3 million by that year.\9\ --------------------------------------------------------------------------- \8\U.S. Senate, Special Committee on Aging. ``Aging America: Trends and Projections.'' November 1989. Sen. Prt. 101-59, p. 4. \9\A Call for Action, p. 108. --------------------------------------------------------------------------- CHART B-3. PERSONS WITH ADL AND/OR IADL LIMITATIONS, 1990 THE NURSING HOME POPULATION\10\ --------------------------------------------------------------------------- \10\This material is drawn largely from ``Characteristics of Nursing Home Residents and Proposals for Reforming Coverage of Nursing Home Care,'' by Richard Price, Richard Rimkunas, and Carol O'Shaughnessy, CRS Report for Congress, No. 90-471 EPW, September 24, 1990. --------------------------------------------------------------------------- Demographic characteristics Analysis of the 1985 National Nursing Home Survey (NNHS) shows that the great majority of nursing home residents are 65 years of age and older. In 1985, 88 percent of residents were 65 years of age and older, and 12 percent were under the age of 65. As the top half of table 12 indicates, less than 5 percent of the total elderly population in the country were residents of nursing homes on any given day in 1985, and 0.1 percent of the under 65 population were residents in that year. Although in the aggregate less than 5 percent of the total elderly population was in a nursing home on any given day in 1985, younger and older age groups of the elderly show very different rates of utilization. Table B-12 and chart 4 show that about 1 percent of the 65-74 age group and about 6 percent of the 75-84 age group resided in nursing homes in 1985. For the very old, those 85 and older, however, the incidence rate increases dramatically. In 1985, 22 percent of the 85 and older group resided in nursing homes. This group accounted for 40 percent of total nursing home residents, and 45 percent of the elderly nursing home population. TABLE B-12.--NURSING HOME RESIDENTS AS A PROPORTION OF TOTAL POPULATION, BY AGE AND SEX, 1985 [All nursing home and U.S. population estimates in thousands] ------------------------------------------------------------------------ All residents -------------------------------------- Age Nursing home pop. U.S. pop. Percent ------------------------------------------------------------------------ Under 65......................... 173 210,197 0.1 65 to 74......................... 212 17,009 1.2 75 to 84......................... 508 8,836 5.7 85 and older..................... 597 2,695 22.1 65 and older..................... 1,317 28,540 4.6 -------------------------------------- Total........................ 1,490 238,737 0.6 ------------------------------------------------------------------------ Males Females ------------------------------------------------------------- Age Nursing Nursing U.S. pop. Percent home pop. U.S. pop. Percent home pop. ---------------------------------------------------------------------------------------------------------------- Under 65.......................................... 89 104,623 0.1 84 105,574 0.1 65 to 74.......................................... 81 7,475 1.1 132 9,534 1.4 75 to 84.......................................... 141 3,293 4.3 367 5,543 6.6 85 and older...................................... 112 769 14.6 485 1,926 25.2 65 and older...................................... 334 11,537 2.9 984 17,003 5.8 ------------------------------------------------------------- Total......................................... 423 116,160 0.4 1,068 122,577 0.9 ---------------------------------------------------------------------------------------------------------------- Note.--Figures are based on the number of current nursing home residents and U.S. Census Bureau estimates of the resident population. Figures do not reflect the likelihood of any individual being in a nursing home; rather these estimates indicate the percent of the total population that resided in nursing homes at a given point in time in 1985. Source: Estimates prepared by CRS using the 1985 National Nursing Home Survey, Current Resident File, and U.S. Bureau of the Census, Current Population Report, United States Population Estimates, by Age, Sex and Race: 1980 to 1987, series P-25, No. 1022, March 1988. These estimates are subject to limitations of the data and methods employed. Chart B-4 also illustrates that, among each of the age groups of the elderly, women were more likely to reside in nursing homes than men. For the elderly as a whole, women were twice as likely to be residing in nursing homes in 1985 as men (6 percent of women as opposed to 3 percent for men). The difference for men and women is particularly striking in the 75-84 and 85 and older age groups. Higher incidence rates for women, largely the result of longer life expectancies for women, mean a nursing home population that is predominately female. Chart B-5 indicates that 72 percent of nursing home residents were female in 1985. CHART B-4. SHARE OF RESIDENT POPULATION IN NURSING HOMES, 1985 CHART B-5. DISTRIBUTION OF CURRENT RESIDENTS, BY SEX, 1985 Studies have shown that persons without spouses are more likely to enter nursing homes than persons with spouses.\11\ Because many disabled persons often require a great deal of assistance, spouses are often the only person outside of nursing homes able to provide such intensive care. Chart B-6 indicates that, at admission, only 16 percent of nursing home residents were married. Of the remaining, 56 percent were widowed, 18 percent had never been married, and about 8 percent were either divorced or separated. --------------------------------------------------------------------------- \11\``Financing of Long-Term Care.'' Submitted to the Assistant Secretary of Planning and Evaluation, U.S. Department of Health and Human Services. Contract No. HHS-100-86-051, September 30, 1988. p. I- 9. --------------------------------------------------------------------------- Chart B-7 shows that, among the elderly, the proportion of residents who were married at admission decreases with age, and the proportion who were widowed increases. CHART B-6. DISTRIBUTION OF CURRENT RESIDENTS, BY MARITAL STATUS AT ADMISSION, 1985 CHART B-7. PERCENT OF CURRENT RESIDENTS, MARRIED AND WIDOWED, BY AGE, 1985 Number and type of ADL limitations of nursing home residents Chart B-8 presents data on the number of limitations in ADLs exhibited by nursing home residents of all ages in 1985. This figure shows that nursing home residents have substantial functional limitations. Seventy-eight percent of residents needed the assistance of others in two or more ADLs. Almost 55 percent of the nursing home population was severely impaired with four or more ADLs. Chart B-8 also shows that slightly more than 20 percent of nursing home residents were judged to have no, or only one, activity limitation. A review of the diagnosis classifications of residents by their number of ADLs shows that residents whose primary diagnosis was a mental disorder were disproportionately represented among the total number of residents who had no activity limitation. About 35 percent of those with no ADLs had a mental disorder as their primary diagnosis. Mental disorders include a wide range of disabilities, including dementias, psychoses, and mental retardation. Persons with mental disorders but without limitations in ADLs may be residents of nursing homes because they require supervision or because of the unavailability of other housing and social service arrangements in the community. CHART B-8. DISTRIBUTION OF CURRENT RESIDENTS BY NUMBER OF ADL LIMITATIONS, 1985 Chart B-9 presents data on the extent to which nursing home residents have various kinds of limitations in ADLs. The most frequently found limitation among residents was bathing, with 88 percent of residents needing the assistance of another person. The least prevalent ADL was in eating, with slightly more than one-third of residents needing assistance with this ADL. About three-quarters of residents needed assistance to dress and two-thirds needed assistance in getting out of a bed or chair (transferring). About half of all residents needed the assistance of others in getting to the toilet or in caring for an ostomy bag or catheter. In developing measures of functional limitations, researchers have found an ordered regression in functional abilities as part of the natural aging process. Loss of functioning begins with activities which are most complex and least basic, such as bathing or dressing. Functions which are least complex and most basic, such as feeding oneself, are retained longer. That is, persons are most able to retain their ability to feed themselves, but are less likely to retain their ability to bathe or dress without the assistance of others.\12\ In addition, persons who are the most severely impaired are least likely to be able to eat independently, and therefore are more likely to have limitations in all the other ADLs. This ordered regression in stages of functioning is reflected in the nursing home population. As shown in chart B-9, higher proportions of residents needed assistance in bathing or dressing than those who needed assistance in eating. --------------------------------------------------------------------------- \12\Katz, Sidney and Amechi Akpom. ``A Measure of Primary Sociobiological Functions.'' International Journal of Health Services, Vol. 6, No. 3, 1976. --------------------------------------------------------------------------- CHART B-9. PERCENT OF RESIDENTS REQUIRING ASSISTANCE OF ANOTHER PERSON IN PERFORMING ACTIVITY, 1985 Nursing home length of stay The profile of nursing home residents presented above suggests a fairly homogeneous population: largely very elderly, female, widowed, and very disabled. However, an examination of length-of-stay patterns among the nursing home population suggests a more diverse group of persons using care than might be suggested by demographic data alone. Analysis of discharge data from the NNHS shows at least two major users of nursing home care, as illustrated in charts B-10 and B-11. Chart B-10 portrays the distribution of persons discharged from nursing homes in 1984-85, according to their length of stay. Chart B-11 shows the distribution of days of care used by all discharged residents. It should be noted that the discharge file of the NNHS does not provide a comprehensive picture of the use of nursing home care by a single group of persons over time. As a result, estimates based on discharge survey data must be considered very general orders of magnitude of lengths of stay in a nursing home. Chart B-10 shows that most nursing home stays are relatively short. About 52 percent of persons discharged from nursing homes had stays of less than 90 days and about 63 percent of persons discharged had stays of less than 6 months. In contrast, 27 percent of persons discharged had long stays of 1 year or longer, and 17 percent had stays of 2 years or longer. The distribution of total days of care used by discharged residents is strikingly different. Chart B-11 shows that persons with stays of less than 3 months accounted for only 4 percent of days of care. Those with stays of less than 6 months accounted for 8 percent of all days. On the other hand, persons with stays of 2 or more years accounted for about 73 percent of all discharge days. In other words, persons with short stays accounted for the majority of persons discharged from nursing homes, but very few of the days of care used. Those with long stays accounted for relatively few of those persons discharged from nursing homes, but the bulk of days used. CHART B-10. DISTRIBUTION OF DISCHARGED RESIDENTS BY LENGTH OF STAY, 1984-85 CHART B-11. DISTRIBUTION OF TOTAL DAYS USED BY DISCHARGED RESIDENTS, BY LENGTH OF STAY, 1984-85 Status of nursing home residents following discharge Chart B-12 shows the distribution of residents by their status following discharge. In 1984-85, the largest share of persons--about 50 percent--were discharged from the nursing home to a hospital or other health care facility, including nursing homes (about 7 percent were discharged to another long- term care facility). About 28 percent of discharges were due to death in the nursing home. About 22 percent of the residents were discharged to the community. This mortality rate and the rate of return to the community may be conservative estimates. For example, 10 percent of those discharged from nursing homes to other health facilities died in these other facilities. Others are likely to have returned to the community. CHART B-12. DISTRIBUTION OF DISCHARGED RESIDENTS BY LIVING ARRANGEMENT AFTER DISCHARGE, 1984-85 The community-based long-term care population Chart B-13 below showed that the great majority of persons with ADL and/or IADL limitations live in the community. Almost 9 million persons of all ages, or 84 percent of the total population with ADL and/or IADL limitations, live in the community. The elderly represented almost 63 percent of this total. Chart B-13 shows the number and percent of elderly persons living in the community with ADL limitations by type of limitation, as of 1984.\13\ A total of 3.7 million elderly persons living in the community, or 14 percent of the total elderly population, reported some limitation in their ability to bathe, transfer, dress, toilet, or eat. The prevalence of these ADLs forms a hierarchy similar to that shown above in chart B-8 for the nursing home population. The most prevalent limitation was in bathing, with 10 percent of the elderly reporting difficulty with this ADL. The least common was in eating, with 2 percent of elderly persons reporting difficulty. --------------------------------------------------------------------------- \13\Rowland, Diane. ``Measuring the Elderly's Need for Home Care,'' Health Affairs, winter 1989, vol. 8, p. 42. --------------------------------------------------------------------------- CHART B-13. PERCENT OF ELDERLY IN THE COMMUNITY WITH ADL LIMITATIONS, BY TYPE OF LIMITATION, 1984 Chart B-14 indicates that 54 percent of the elderly population with any kind of ADL limitation in 1984 had two or more ADLs. This was about 2 million persons. Almost 22 percent has 4 or 5 limitations. The severity of impairment is not uniform in the disabled population. Among the 2 million persons with two or more ADLs, 1.1 million reported some difficulty and 0.9 million reported a lot of difficulty or inability to perform at least two ADLs.\14\ --------------------------------------------------------------------------- \14\Rowland, p. 43. --------------------------------------------------------------------------- CHART B-14. DISTRIBUTION OF ADL'S AMONG NONINSTITUTIONALIZED ELDERLY POPULATION HAVING ONE OR MORE ADL LIMITATIONS, 1984 Studies have shown that the great bulk of care provided to persons living in the community with ADL and/or IADL limitations is provided informally by family and friends who are not paid for the care they provide. Chart B-15 indicates that 70 percent of severely disabled elderly persons receiving long-term care in the community relied solely on informally provided care. Only 3 percent relied only on formal or paid care. More than 7 million spouses, adult children, other relatives, friends, and neighbors provided unpaid assistance to disabled elderly persons in 1984.\15\ Seven out of ten informal caregivers bear the major responsibility for care provided, and one of three is a sole provider. Three-quarters of all caregivers are female--wives and daughters of persons needing care. Research has shown that caregivers often reduce their work hours, take time off without pay, or quit jobs because of elder caregiving responsibilities. In addition, many caregivers are themselves elderly--one-quarter are between the ages of 65 to 74 and another 10 percent are 75 or older. --------------------------------------------------------------------------- \15\A Call for Action, p. 93-95. This discussion draws heavily on this report and research published by Robyn Stone, et al., ``Caregivers of the Frail Elderly: A National Profile,'' The Gerontologist, vol. 27, 1987. --------------------------------------------------------------------------- Use of formal, paid services by elderly persons living in the community is related to various characteristics of this group.\16\ Differences in functional status have been found to be strongly related to use of formal home and community-based care, with the likelihood of using any formal service increasing as levels of impairment increase. Age is also linked to the use of formal services, largely explained by the fact that age is associated with decreasing functional status. In general and in each age group of the elderly, more women use formal home and community-based care services than men. This is related to the longer life expectancies of women. Persons living alone are more than twice as likely to use formal services as compared to those living with other persons. In addition, the amount of money spent on home care services has been found to be directly related to income; that is, out-of- pocket expenses for home care increase substantially as median family income increases.\17\ --------------------------------------------------------------------------- \16\This material is drawn largely from Short, Pamela and Joel Leon, ``Use of Home and Community Services by Persons Ages 65 and Older with Functional Difficulties,'' National Medical Expenditure Survey, Research Findings 5, Department of Health and Human Services, Agency for Health Care Policy and Research, September 1990, p. 7-9. \17\Liu, Korbin, Kenneth Manton, and Barbara Liu, ``Home Care Expenses for the Disabled Elderly,'' Health Care Financing Review. Winter 1985, vol. 7, No. 2, p. 55. --------------------------------------------------------------------------- CHART B-15. SOURCE OF HOME CARE SERVICES FOR THE SEVERELY DISABLED ELDERLY POPULATION, 1989 Public and private spending for long-term care Table B-13 indicates that sizable public and private funds are being spent on long-term care services. For two major categories of long-term care services, nursing home and home care, total national spending amounted to almost $107.8 billion in 1993. This total is for all age groups using long-term care. By far the greatest portion of spending is for nursing home care. About $75 billion, or 70 percent of the total, was spent for nursing home care in 1993. Public programs paid about 60 percent of the Nation's total nursing home bill. Medicaid payments accounted for almost all of this amount. Medicaid is the Federal-State health program for the poor and for those who have become poor as the result of incurring large medical care expenses. In 1993, Medicaid spending for nursing home care amounted to 48 percent of total national nursing home spending. Table B-13 shows that private spending accounted for about $30 billion, or the remaining 40 percent of national spending. Nearly all private spending for nursing home care was paid directly by consumers out-of-pocket with income and/or accumulated resources. Private insurance coverage for long-term nursing home care is very limited, with private insurance payments amounting to 0.1 percent of total spending for nursing home care in 1993. Spending for home health care services amounted to $33 billion, or 30 percent of the total. Public programs accounted for about 72 percent of total home health care spending. Out- of-pocket payments accounted for almost all of private spending, private insurance again being very limited for this care. Most home and community-based care, as discussed above, is provided by family and friends who are not paid for the services they provide. Major Federal programs supporting long-term care Five programs represent the major source of Federal financial support available for nursing home and community- based long-term care--Medicaid, Medicare, the Social Services Block Grant (SSBG), the Older Americans Act, and the Supplemental Security Income (SSI) program. None of these programs supports the full range of long-term care services. Certain programs provide health services but exclude social services. Others provide strictly social services. Some have income eligibility requirements, others do not. Medicaid is the Nation's major program of financial support for long-term care, principally because of its coverage of nursing home care. Medicaid payments for nursing home care (excluding nursing homes for the mentally retarded) amounted to about 27 percent of total Medicaid spending in fiscal year 1991. Comparatively little funding is devoted to home and community-based care. Coverage of both nursing home and home and community-based services is restricted to those persons who have limited income and assets. In general, Medicaid rules limit eligibility to those persons who qualify for cash welfare assistance or who incur large health care expenses that deplete their income and assets.\18\ --------------------------------------------------------------------------- \1\\8\Most States extend Medicaid eligibility to persons who qualify for welfare benefits under the Supplemental Security Income (SSI) program. SSI requires that persons have assets that do not exceed $2,000 and income that does not exceed $446 per month in 1994. --------------------------------------------------------------------------- TABLE B-13.--ESTIMATED LONG-TERM CARE SPENDING FOR ALL AGE GROUPS, BY SOURCE, 1993 [Dollars in billions] ------------------------------------------------------------------------ Source of spending Amount ------------------------------------------------------------------------ Nursing home care: Medicaid............................................... $36.3 Medicare............................................... 5.7 Other Federal.......................................... 1.0 Other State............................................ 2.5 Out-of-pocket payments and other....................... 29.6 Private insurance...................................... 0.1 ------------ Total................................................ 75.2 ============ Home health care: Medicaid............................................... 7.4 Medicare............................................... 10.1 Other Federal programs................................. 1.6 Other State............................................ 4.5 Out-of-pocket payments and other....................... 8.9 Private insurance...................................... 0.1 ------------ Total................................................ 32.6 ------------ Total long-term care expenditures.................... 107.8 ------------------------------------------------------------------------ Source: Office of the Assistant Secretary for Planning and Evaluation, Office of Disability, Aging, and Long-Term Care Policy, Department of Health and Human Services. Medicare, the Federal health insurance program for the elderly and disabled, is focused primarily on coverage for acute health care costs and was never envisioned to provide protection for long-term care. Coverage of nursing home care, for instance, is limited to short-term stays in certain kinds of nursing homes, referred to as skilled nursing facilities, and only for those persons who demonstrate a need for daily skilled nursing care following a hospitalization. Many persons who require long-term nursing home care do not need daily skilled nursing care, and therefore, do not qualify for Medicare's benefit. As a result of this restriction, Medicare paid for about 7.6 percent of the Nation's expenditures for nursing home care in 1993. For similar reasons, Medicare pays for only limited amounts of community-based long-term care services, primarily through the program's home health benefit. To qualify for home health services, the person must be in need of skilled nursing care on an intermittent basis, or physical or speech therapy. Most chronically impaired persons do not need skilled care to remain in their homes, but rather nonmedical supportive care and assistance with basic self-care functions and daily routines that do not require skilled personnel. Three other Federal programs--SSBG, the Older Americans Act, and the SSI program--provide support for community-based long-term care services for impaired elderly persons. The SSBG provides block grants to the States for a variety of home-based services for the elderly as well as the disabled and children. The Older Americans Act also funds a broad range of in-home services for the elderly. Under the SSI program, the federally administered income assistance program for aged, blind, and disabled persons, many States provide supplemental payments to the basic SSI payment to support selected community-based long- term care services for certain eligible persons, including the frail elderly. However, since funding available for these three programs is limited, their ability to address the financing problems in long-term care is also very limited. Spending down for Medicaid coverage of nursing home care As discussed above, the Medicaid program is the major public source of support for the cost of nursing home care. Its spending for nursing home care is driven largely by its coverage of persons who are not initially poor but who become poor by depleting their assets on the cost of care. At an average cost of $35,000 a year, nursing home costs can quickly deplete the resources of an elderly individual, especially after prolonged stays, and these costs also exceed the monthly income of most persons. The depletion of financial resources on the cost of care and the movement from private payment for care to Medicaid coverage is referred to as the ``spend-down'' process. In 1991, Medicaid nursing home payments for elderly persons who spent down amounted to 60 percent of total Medicaid payments for all services for all elderly beneficiaries.\19\ --------------------------------------------------------------------------- \19\Spending down under Medicaid is a two-step process. First persons must meet the resources or assets test. The term ``resources'' generally refers to liquid assets such as cash on hand, savings and checking accounts, stocks and bonds, etc. In order to become eligible for Medicaid, the value of the individual's available resources must be less than a State-determined dollar standard, usually $2,000 for an individual without a spouse, the level used for the SSI program. Certain items, such as the house, are excluded as countable resources under SSI and Medicaid rules. Second, after an individual has depleted virtually all accumulated resources on the cost of nursing home care, or has transferred resources (for less than fair market value) prior to the time when eligibility could be denied because of the transfer, income standards are then considered. Most States have no absolute upper limit on income for applicants residing in nursing homes. These States have what are known as medically needy programs. As long as the applicant's current monthly income is insufficient to cover medical expenses, including the cost of care in the nursing home, the applicant can become eligible for Medicaid. Other States use a special income level to determine eligibility for persons residing in nursing homes. Like the medically needy, these persons have income in excess of cash welfare program standards. By Federal law, the special income level used by States can be no more than three times the basic SSI payment level, or $1,302 in 1993. This rule is known as the ``300 percent rule.'' --------------------------------------------------------------------------- Numerous studies have looked at Medicaid spend-down in the last 5 years. A recent review of these studies, ``A Synthesis and Critique of Studies on Medicaid Asset Spenddown'' by Adams, Meiners and Burwell, found that they generally use two different measures of Medicaid asset spenddown.\20\ One method measures the percentage of persons originally admitted to nursing homes as private payers who eventually convert to Medicaid prior to final discharge. This method is a measure of the risk to individuals of spending down to Medicaid over the course of their lifetimes, given the probability they enter a nursing home as private payers. --------------------------------------------------------------------------- \20\This material draws heavily on Adams, E. Kathleen, Mark Meiners, and Brian Burwell, ``A Synthesis and Critique of Studies on Medicaid Asset Spenddown,'' Office of the Assistant Secretary for Planning and Evaluation, Department of Health and Human Services, January 1992. --------------------------------------------------------------------------- A second method of measuring Medicaid spenddown examines the percentage of Medicaid residents of nursing homes who were not eligible for Medicaid when they were originally admitted. This method can be useful in capturing the proportion of State Medicaid expenditures for nursing home care that is accounted for by those who spend down. The review of spenddown studies, which use several different national and State-level data bases, found widely varying estimates of spenddown as measured by these two methods. According to the review, the critical factor explaining differences among these studies is the length of time that persons are studied. The proportion of persons spending down during a single stay is much lower than the proportion of persons who spend down over their entire lifetime, since half or more of persons using nursing home care have multiple stays. In general, studies using national data tend to show lower estimates of spenddown than do State studies that tend to observe people over longer time intervals. The review of spenddown studies found that between 20 and 25 percent of persons who originally enter nursing homes as private payers convert to Medicaid before final discharge. For this method of measuring spenddown, not enough State studies exist to determine the extent to which spenddown rates vary from State to State. On the other hand, estimates of spenddown as measured by the percentage of Medicaid residents of nursing homes who were not eligible for Medicaid when they were originally admitted vary considerably across States, reflecting variations in Medicaid eligibility policies across the States as well as other factors. Studies measuring spenddown according to this method have found spenddown rates of 27 percent for Michigan, 31 percent for Wisconsin, and 39 to 45 percent for Connecticut. Spenddown studies have also examined the length of time it takes for persons to spend down after nursing home admission. The results of these studies reveal that of those people who spend down, the majority spend down within a year of nursing home admission. This finding suggests that most people who spend down have limited assets when they first enter a nursing home. Certain State studies also show that people who spend down to Medicaid spend more time on Medicaid after converting to Medicaid coverage than they spend as private payers prior to conversion. The studies show that Medicaid-paid days account for at least 65 to 75 percent of all nursing home days used by those who spend down. However, the research also shows that, once eligible for Medicaid, people who spend down pay a greater proportion of total nursing home costs, through contributions of their income they are required to make before Medicaid makes its payment, than persons who are eligible for Medicaid at initial admission. As a result, people who spend down account for a somewhat lower percentage of total Medicaid expenditures than their percentage of Medicaid-covered nursing home days. Private long-term care insurance Private long-term care insurance is generally considered to be the most promising private sector option for providing the elderly additional protection for long-term care expenses. Long-term care insurance is a relatively new, but rapidly growing, market. In 1986, approximately 30 insurers were selling long-term care insurance policies of some type and an estimated 200,000 persons were covered by these policies. By 1987, a Department of Health and Human Services Task Force on Long-Term Care Insurance found 73 companies writing long-term care insurance policies covering 423,000 persons. As of December 1992, the Health Insurance Association of America found that more than 2.9 million policies had been sold, with 135 insurers offering coverage. (Note that this is a cumulative total of policies sold; fewer persons would be covered, due to failure to pay premiums because of death, a change in income, a decision not to continue coverage, etc.) Although growth has been considerable in a short period of time, the private insurance industry has approached this potential market with caution. Insurers are concerned about the potential for adverse selection in long-term care insurance, where only those persons likely to need care actually buy insurance. In addition, they point to the problem of induced demand for services that can be expected to be generated by the availability of new long-term care insurance. With induced demand, sometimes also referred to as moral hazard, individuals decide to use more services than they otherwise would because they have insurance and/or will shift from nonpaid to paid providers for their care. In addition, insurers are concerned that, given the nature of many chronic conditions, persons who need long-term care will need it for the remainder of their lives, resulting in an open-ended liability for the insurance company. As a result of these risks, insurers have designed policies that limit their liability for paying claims. Policies have been medically underwritten to exclude persons with certain conditions or illnesses. They have contained benefit restrictions that limit access to covered care. Policies also limit the period of coverage they offer, typically to a maximum of 4 or 5 years. In addition, most plans provide indemnity benefits that pay only a fixed amount for each day of coverage service. If these amounts are not updated for inflation, the protection offered by the policy can be significantly eroded by the time a person actually needs care. Today payment amounts can generally be updated for inflation, but only with significant increases in premium costs. These design features of long-term care insurance raise issues about the quality of coverage offered purchasers of policies. The insurance industry has responded to some of these concerns by offering new products that provide broadened coverage and fewer restrictions. One of the key issues outstanding in the debate on the role private insurance can play in financing long-term care is the affordability of coverage. The Health Insurance Association of America has reported that policies paying $80 a day for nursing home care and $40 a day for home health care with inflation protection and a 20-day deductible period and a 4-year maximum coverage period had an average annual premium in December 1992 of $1,597 when purchased at the age of 65 and $5,334 when purchased at the age of 79. Many elderly persons cannot afford these premiums. The insurance industry believes that affordability of premiums can be greatly enhanced if the pool of persons to whom policies are sold is expanded. The industry has argued that the greatest potential for expanding the pool of persons buying coverage and reducing premiums lies with employer-based group coverage. Premiums should be lower in employer-based group coverage because younger age groups with lower levels of risk of needing long-term care would be included, allowing insurance companies to build up reserves to cover future payments of benefits. In addition, group coverage has lower administrative expenses. As of December 1992, 506 employers offered a long-term care insurance plan to their employees. These employer-based plans covered over 350,000 employees, their spouses, retirees, parents, and parents-in-law. But just how broadly based employer interest is in a new long-term care benefit is unclear at the present. Many employers currently face large unfunded liabilities for retiree pension and health benefits. Also, many employers have recently experienced substantial increases in premiums for their current health benefits plans. Very few employers contribute to the cost of a long-term care plan. Most employers require that the employee pay the full premium cost of coverage. In contrast, the majority of medium and large sized employers pay the full premium cost of regular health care benefits for their employees.