Health Care in a Context of Civil Rights (1981)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 RACIAL DIFFERENCES IN USE OF NURSING HOMES Racial and ethnic variations in the use of nursing homes are a distinct feature of American health care. Elderly blacks* use nursing homes at lower rates than do whites. Various explanations have been offered, including differences in family networks and values, differences in geographic proximity to nursing homes, racial differences in survival and in numbers of elderly people, and racial discrimination. Although the pattern of less black use of nursing homes has been recognized for many years, the committee was unable to locate sufficient data for a definitive sorting out of the competing explanations. Notwithstanding the dearth of information for a direct assessment of discrimination in nursing home admissions, indirect evidence raises the possibility that discrimination may be widespread. However, little attention has been given to documenting it or to bringing civil rights enforcement activities to bear on it. Evidence pertaining to the racial difference in the use of nursing homes is reviewed in this chapter. Evidence regarding discrimination comes largely from data showing the inadequacy of competing explanations. To summarize this evidence, the low use of nursing homes by blacks is apparent in rates of nursing home use, not only in numbers (which may be affected by differential mortality rates). All indications are that the health problems and disabilities that create the need for nursing home care are at least as common among blacks as whites, and, although some differences in family living arrangements *The emphasis of this chapter is on the use of nursing homes by the black elderly. This is the ethnic minority about which the most concern has been expressed regarding possible discrimination in nursing homes, and it is the minority about which the most adequate data exist. Although there is a growing body of literature about the elderly in other minority groups, it is not adequate for a description of patterns of nursing home use and possible causes for patterns that are distinct to particular ethnic groups. The problems of explanation that are reviewed in this chapter are less difficult than the problems that would be faced in attempting a similar analysis of the use of nursing homes by any other ethnic minority group. 72

73 can be documented, there is no indication that partially disabled, elderly whites are less successful than blacks in securing needed assistance at home. Blacks probably suffer disproportionately from disincentives perceived by nursing homes in accepting Medicaid patients. In addition, there are indications from some states of a large racial difference in meeting the needs for nursing home care of elderly poor persons within the Medicaid population. Because the federal government, particularly through the Medicaid program, is the major source of money for nursing home care, and this federal involvement triggers the applicability of the Civil Rights Act's prohibition of discrimination, the lack of direct evidence about discrimination in nursing homes is rather remarkable. Similarly, although there is widespread agreement among knowledgeable persons that nursing homes have strong tendencies toward racial segregation, little direct documentation is available, despite all of the claims forms that are submitted for Medicaid and Medicare reimbursement. It is evident that the possibility, even the likelihood, of widespread patterns of discrimination and segregation in nursing homes has not been regarded by government as an important problem. INTRODUCTION There are about 18,900 nursing homes* in the United States in which more than 1,300,000 persons, mostly elderly, reside and receive care.2 Most (77 percent) nursing homes are privately owned, proprietary institutions,3 and some have religious affiliations. Data are not available on the countless boarding homes that provide some aspect of nursing home care, because few of these homes are licensed. Most come to official attention only when a fire or other disaster occurs. Although it has been known for years that minority groups make less use of nursing homes than do whites, some doubt may be raised about whether this pattern merits concern, because nursing home care is not an unmixed blessing. As Vladeck noted in his recent examination of nursing homes for the Twentieth Century Fund: [Nursing homes] have been described as "Houses of Death," "concentration camps," "warehouses for the dying." It is a documented fact that nursing home residents tend to deteriorate, physically and psychologically, after being *The National Nursing Home Survey includes several categories of nursing homes. The survey is not confined to facilities that are certified for Medicare and Medicaid reimbursment purposes as "skilled nursing facilities" or "intermediate care facilities." Not all homes in the survey actually provided "nursing services;" all, however, provided residents with assistance in activities of daily living.1

74 placed in what are presumably therapeutic institutions. The overuse of potent medications in nursing homes is a scandal in itself. Thousands of facilities in every state of the nation fail to meet minimal government standards of sanitation, staffing, or patient care. The best governmental estimate is that roughly half the nation's nursing homes are "substandard."4 Yet nursing homes meet important individual and societal needs for which adequate alternatives (stipends for family care, day care facilities, respite care, foster care, and so forth) do not widely exist. People reside in nursing homes because they are dependent upon others for some aspects of their care. According to the 1977 National Nursing Home Survey, more than 86 percent of nursing home residents require assistance in bathing, 69 percent require assistance in dressing, 52 percent require assistance in using the toilet room, 66 percent can walk only with assistance or are chairfast or bedfast, 45 percent have difficulty with bowel or bladder control, and 33 percent require assistance with eating.5 Fewer than 10 percent are dependent in none of the above activities, and almost one-fourth are dependent in all of them. For almost 80 percent of the patients, the primary reason given for residence in nursing homes is care needs stemming from poor physical health; other reasons included mental illness, mental retardation, behavioral problems, and social and economic reasons. For many persons a combination of factors is undoubtedly involved. The median age of patients is 81. Chronic conditions and impairments are common and varied, and include arteriosclerosis (48 percent of residents), hypertension (21 percent), stroke (16 percent), heart trouble (34 percent), chronic brain syndrome (25 percent), and senility (32 percent). A gross measure of the demand for nursing home care is provided by the remarkable growth of the nursing home "industry"—bed capacity tripled between 1954 and 1973 and has continued to grow.6 The fact that nursing homes are populated by people who need care does not demonstrate that such care is best provided in nursing homes. It is estimated that between 10 and 40 percent of the elderly placed in nursing homes could be better served, and at a lower cost to the community, were the necessary services available.7 But adequate domiciliary services are not available. In the face of concerns about cost and quality of nursing home care and the growth of the elderly population, interest in alternatives is growing. At present, however, public policy (as expressed by Medicare and Medicaid) continues to favor inpatient nursing home care. The fact remains that aging is often accompanied by the development of chronic physical and mental problems that produce a need for assistance. Nursing homes are an important source of such assistance, particularly when the needs for care exceed the capacity of dedicated family members and friends. There is little doubt that the need is genuine and not due to such factors as the shirking of family responsibilities, as seems to be widely believed. As Elaine Brody wrote, "Overall, the responsible behavior of families towards

75 older people has been so thoroughly documented that it is no longer at issue in gerontological research."® Much of the use of nursing homes in the past two decades can be accounted for by the growth of the over-75 population (and its substantial number of unmarried women without children) and by the deinstitutionalization movement in mental hospitals that once provided a home for many of the confused elderly.9 Thus, because nursing homes have a near monopoly on continuing care of the elderly, and are needed to meet the needs of a population that requires both medical care and assistance in activities of daily living, there is reason for concern about possible inequities in access. A second reason for concern arises from the flow of tax dollars to nursing homes and the consequences of this flow for different members of the population. Public policy decisions are responsible for many of the characteristics of the nursing home industry, both because public money is the dominant source of nursing home dollars and because of the regulatory web that accompanies those public funds. Three-fourths of nursing homes (containing almost 90 percent of nursing home beds) are certified to receive federal monies through the Medicare or Medicaid programs, and the federal dollar is the most important source of payment for nursing home care. Figures for 1976-78 show that the federal government provided 53 percent of the nursing home dollars, mostly through the Medicaid program.^ States also are a major source of funds for nursing home care through Medicaid matching funds. Because Medicaid programs are run by the states (with a substantial federal subsidy), state governments can be regarded as large purchasers of nursing home services. In the view of some observers, budgetary pressures at the state level bring the state's self-interest into conflict with the interests of persons needing nursing home care. This conflict is manifested in vigorous restraint of reimbursement rates (with a predictable impact on the availability of beds) and in less than vigorous attention to standards of quality.12 USE OF NURSING HOMES BY BLACKS Questions have long been asked about minority group access to nursing homes. For example, Senator Moss opened a set of hearings on long-term care in 1972 with the question, "Why are there no members of minority groups in nursing homes? It is a fact that comparatively few blacks, Asians, Indians, or Mexican-Americans are in nursing homes."^ His Subcommittee on Long-Term Care heard testimony about a variety of barriers to nursing home admission for racial or ethnic minorities. There was testimony about the problem of distance from nursing homes for Indians on Arizona reservations, language and cultural barriers for Asian Americans and Mexican Americans, and racism and economic and cultural barriers for blacks. The concerns expressed about the unmet needs of elderly individuals from different racial and ethnic minority groups was paralleled by frustration about

76 the lack of data to document and explain, in a systematic way, ethnic patterns of disparity and unmet needs. However, a general picture of nursing home use by blacks has been available for some time. National studies have consistently shown low rates of use, although there is evidence of a slow increase.* The underrepresentation of minorities is most clearly seen when different age groups are compared. Table 15 presents data for all age groups over 65 and shows lower rates of nursing home use by non-whites than whites since 1963. Although the non-white rate has increased more rapidly than the white rate since that time, the most recent data still show a very substantial racial difference. Fifty per 1,000 white persons aged 65 and over were residents of nursing homes in 1977, compared with 30 per 1,000 for the rest of the population. Persons 85 years of age and over who are white are twice as likely as others of this age to be receiving nursing home care and the federal and state tax dollars that support it. Table 15 also shows that the lower black use of nursing homes is not simply due to differential mortality. Although it is true that the number of elderly blacks is smaller than it would be if white and black life expectancy at birth had been equivalent over the past 65 years, this fact does not affect the rates shown in Table 15, because they are stated in terms of the population aged 65 and older. Furthermore, at age 65 there is no longer much difference in life expectancy between whites and others, and in older age categories the mortality rate for whites exceeds that for blacks.18 Table 15 also shows that the racial pattern of nursing home use differs among different age groups. In 1977, among persons aged 65-74, the white rate of nursing home use was actually slightly lower than the non-white rate (14.2 versus 16.8 per 1,000). (Whether this pattern will continue as this cohort ages remains to be seen.) However, among those 75-84, the non-white rate was only 55 percent of the white rate, and among those above age 85, the non-white rate was only 45 percent of the white rate. In the latter category, one-fourth of the white population were residents of nursing or personal care homes as compared to only one-tenth of the rest of the population. *The National Center for Health Statistics (NCHS) 1969 survey of residents of "nursing and personal care" homes showed that 4.5 of the residents were from groups other than "whites."14 The 1973-74 National Nursing Home Survey showed blacks as 4.6 percent of the nursing home population, while "Spanish-American" patients made up 1.1 percent.15 Comparable figures from the 1977 Nursing Home Survey were 6.2 percent blacks and 1.1 percent "Hispanics."16 Among persons aged 65 and over, members of racial/ethnic groups other than "white" made up 9 percent of the U.S. population, but only 5.2 percent of the nursing home population in 1973-74.17 The 1977 survey showed persons from groups other than "whites" to constitute 6.8 percent of nursing home residents aged 65 and over.

77 Table 15. NURSING HOME AND PERSONAL CARE HOME RESIDENTS 65 YEARS AND OVER, ACCORDING TO AGE AND COLOR Year and Age 1963 65 years and over 65-74 years 75-84 years 85 years and over 1969 65 years and over 65-74 years 75-84 years 85 years and over Number of Residents 445,600 89,600 207,200 148,700 722,200 138,500 321,800 261,900 Number per 1,000 Population White1 All Other 26.6 8.1 41.7 157.7 38.8 10.3 5.9 13.8 41.8 17.6 9.6 1973-742 65 years and over 65-74 years 75-84 years 85 years and over 19773 65 years and over 65-74 years 75-84 years 85 years and over 961,500 163,100 384,900 413,600 1,126,000 211,400 464,700 449,900 11.7 54.1 221.9 48.1 12.5 61.9 269.0 49.7 14.2 70.6 229.0 22.9 52.4 21.9 10.6 30.1 91.4 30.4 16.8 38.6 102.0 Includes Hispanics. 2Excludes residents in personal care homes, 'includes residents in domiciliary care homes. SOURCE: Department of Health and Human Services, Health: United States, 1980 (Washington, D.C.: Government Printing Office, 1981) p. 496. Data are based on national surveys of nursing homes conducted by the National Center for Health Statistics in the years shown. Levels of Disability If the use of nursing homes was simply a function of degree of disability, there would be more blacks than whites in nursing homes; studies consistently show more disability among elderly blacks than

78 among elderly whites. Table 16, for example, presents responses from household interviews conducted by the National Center for Health Statistics in a national sample of the non-institutionalized population and shows that among those above age 65, blacks were more likely than whites to report limitations in activity and mobility and days of restricted activity and bed disability. * Blacks aged 65 and older reported an average of 25 days of bed disability in 1977 compared with an average of 12 days of bed disability for whites in the same age category. Half of the blacks above age 65 reported that they were limited in a major activity, compared with 36 percent of the whites. Twenty-four percent of the blacks reported limitation of mobility, compared with 17 percent of whites. Similar findings emerged from the 1975 National Survey of the Black Aged sponsored by the Social Security Administration and the Administration on Aging. Shanas summarized the findings on health status: 1. Although there is little difference between blacks and whites in their proportions of household and bedfast elderly, black aged, particularly black women, report more restricted physical mobility than do whites. Black women are far more likely than white women to report that they can go outdoors only with difficulty. 2. Capacity for self-care is less among blacks than among whites. Again, the greatest amount of incapacity is reported by black women. 3. Blacks are twice as likely as whites to report difficulties with common physical tasks. 4. Blacks are twice as likely as whites to report that they were giddy at least once during the week before they were interviewed. 5. Blacks are twice as likely as whites to report that they had spent time ill in bed the year before they were interviewed, and they are more likely than whites to report that they saw a doctor during the month before their interviews. 6. Blacks are twice as likely as whites to say that their health is poor and substantially less likely than whites to say that their health is good. 7. Blacks are twice as likely as whites to say that their health is worse than the health of other people their age.20 Although these findings have all of the limitations of self- reported survey data and cannot be assumed to reflect what would be found from physical examinations, this survey suggests that some of the physical difficulties that may lead to residence in a nursing home are more common among elderly blacks than among elderly whites. To the extent that poor health and the need for assistance in tasks of daily living define a need for nursing home care for non- institutionalized people, the need appears greater among the black elderly. Among nursing home residents there is little racial difference in various measures of need and dependency. The 1977 National Nursing Home Survey found that black and white residents of nursing homes exhibit similar patterns of dependency in six activities of daily

79 Table 16. SELECTED MEASURES OF HEALTH LIMITATIONS AND DISABILITY FOR THE POPULATION 45+ YEARS OLD, BY RACE AND AGE (Data exclude persons in institutions) 45-64 years Ratio: 65+ years Ratio: Type of Limitation Black White Black Black White Black or Disability (%) (%) to (%) (*) to LIMITATION OF ACTIVITY, 19771 White White Total 100.0 100.0 1.00 100.0 100.0 1.00 Limited in activity 29.6 22.5 1.32 54.9 42.0 1.31 Limited In major activity 25.8 17.9 1.44 50.9 36.1 1.41 Limited in amount or kind of major activity 15.6 12.0 1.30 23.6 19.8 1.19 Unable to carry on major activity 10.1 5.9 1.71 27.3 16.3 1.67 Not limited in major activity 3.8 4.6 0.83 4.0 5.8 0.69 Not limited in activity 70.4 77.5 0.91 45.1 58.0 0.78 LIMITATIONS OF MOBILITY, 19721 Total 100. O2 100.0 1.00 100. O2 100.0 1.00 Limited in mobility 8.6 4.4 1.95 23.7 17.0 1.39 Has trouble getting around alone 4.3 2.2 1.95 7.7 5.6 1.38 Needs help in getting around 1.6 1.0 1.60 8.3 6.5 1.28 Not limited in mobility 91.4 95.6 0.96 76.3 83.0 0.92 DAYS OF RESTRICTED ACTIVITY, 1977 Average number of days per year 35.5 23.1 1.54 59.6 36.4 1.64 DAYS OF BED DISABILITY, 1977 Average number of days per year 13.9 7.9 1.76 24.6 11.7 2.10 1nata refer to limitations due to chronic conditions. 2Data on limitation of mobility are for all nonwhite races. SOURCE: Administration on Aging, Characteristics of the Black Elderly—1980, DHEW Publication No. (OHDS) 80-20057 (Washington, D.C.: DREW, 1980). The sources of the data are National Center for Health Statistics, Vital and Health Statistics, Series 10, No. 96, Limitation of Activity and Mobility due to Chronic Conditions: United States—1972) No. 126, Current Estimates from the Health Interview Survey: United States—1977; and unpublished data from the 1977 Health Interview Survey.

80 0i to c c o a. 0) i0 -rf fcf 4* 4J •r* CO TO SB n c f 0 0 c o >H K < -H <M TT «£ 5 mm m 1 M Q i i ,g o) >, —i TO h £ O 0i 3 O K r-l CQ C U (0 M t) 0 •H *H O 'D ki i0 N ,U «J £ \ -C 4J CM tT. 0 « EH* C 14J 4J -O TO C O -H •« C rH O u, a. to v V m o Q * a m o 4J > W O1 -o EH c QJ s -H •iH 1 4J U 4J (Q >i -n C CO TO ,*-( r4 c a p> M r~ >, rH CO CO [0 l-l 0) ^ C, ^ 01 Oi •H XI rH CO TO 2 >H rH P ^ 2 O TO CO 14 .C M vc r. r- z s < o u o h w en *0 £ Q EH 55 Jz 01 u pq pq ft *Q ft) 0i O i tH H M 4J io c c o: u c O U ••H 0) TO C U 4i v* 4J 3 <. JTJ •H (0 *^ 05 4J ,H a io oi !"- o 3 -n D rH m r- IH h w o a to -rt a> 0) 10 C O (N tT. rH *-1 C -H PS < -n EH m tT tri m >. c? Q H c .2 > 0i hi 'o O -n CO C CO 0) TO CO v i 4» • 9 CO w en 1 .. *^ 0i ,H CO U 0) S !-- Q 3 -n a P'l *T CT Q a to • • • IB oi to c o* m CN en EH CC < -H C ri n ac ^_ M M O oi o u c 3 jy - CO C ~H fc in p <! 0) CO £ • » -2 IH -H n a £ f PJ a -H m in r- r« 4J a o 2 o- io 0i to c vc m to a §8 OK ^ *H 00 00 i- •H 4J C9t 4J (H ia r-f O 2 4J £ C rH fj n M 0i o o o o o o a O gg 'o 0i tH tH rH a tH I rH 1i o o o O w o < tx rH rH rH £ Qi tH (_) ItH C Q rtl OJ 4J 55 U 4J C w z O O M CM *^ •rf -H C ki U a, Bel j-^ C C "0 O 0i rH jJ ra i0 u a •o a C CU ij a a -) Jt c c a n u io a TO c o -< -H 0) TO "< rH I S f rH 0) CO m 4J JJ < < < M a »; rH >, o o » - • o ' o 0) U <H 4J C c O C 0) -rf C5 rH 0 U •H 0) JC TO T3 4J tj C -H -H -H u i « C ,u U a C TO TO -H TO 10 M C P. j: rH -n j EH U £ TO U • s • § Q

81 living (Table 17).21 The few differences show black nursing home residents to be slightly more dependent than whites. However, the racial differences in disability among nursing home residents are clearly smaller than such differences among the non-institutionalized population, which suggests that a smaller proportion of the disabled elderly black population than of the disabled elderly white population is admitted to nursing homes. To summarize, evidence from national surveys shows that although disability is more common among elderly blacks than among elderly whites, use of nursing homes is substantially higher for whites than for blacks. The probability that a disabled, elderly black person will be admitted to a nursing home appears to be much lower than the probability that an elderly white person will be admitted. Two principal explanations have been offered for the low rates of nursing home use among blacks. The first pertains to values and living arrangements that characterize the black family. The second pertains to the availability of nursing home beds for blacks and involves geographic and economic factors and the possibility of racial discrimination. Family Factors and Nursing Home Use It is frequently suggested that the relatively low use of nursing homes by both black and Hispanic elderly is at least partially due to certain values and characteristics of families in these minority groups. Regarding values, it is pointed out that racial and ethnic groups differ about such matters as the esteem in which the elderly are held and the extent to which the elderly play an active role in family life. For example, the sociologist Robert Hill says of the black family: With respect to family composition, it is very important to note that elderly persons have been a major source of stability for black families from slavery to present times. In fact, it is the role of the elderly that is primarily responsible for the strong kinship bonds in most black families.22 Patterns of exchange that have been described in urban areas may facilitate the community care of persons who need assistance in managing some of the basic tasks of life. This is described (though not in relation to the elderly) in the summary of Carol Stack's rich ethnographic account of the black family in an urban neighborhood: Black families in The Flats and the non-kin they regard as kin have evolved patterns of co-residence, kinship-based exchange networks linking multiple domestic units, elastic household boundaries, lifelong bonds to three-generation households, social controls against the formation of marriages that could endanger the networks of kin, the

82 domestic authority of women, and limitations on the role of the husband or male friend within a woman's kin network. These highly adaptive structural features of urban black families comprise a resilient response to the social-economic conditions of poverty, the inexorable unemployment of black women and men, and the access to scarce economic resources of a mother and her children as AFDC recipients.^ Wershow writes of a pilot study in Alabama in which he had expected to find elderly black women in non-institutional settings because of a child care role that they provided in their families. Instead, he found a number of elderly blacks, mostly females living in stable communities, were living at home because neighbors and church members helped in accordance with well organized plans. Most recipients of this neighborly aid have been pillars of the church and especially active in Ladies Missionary work, who are now reaping the fruits of their long years of faithful service by these church-organized volunteers. However, even if substantial resources of kin are available in neighborhoods, the needs for care and assistance that may develop in very old age may present very formidable demands. The evidence that values of mutual assistance play a substantial role in affecting black use of nursing homes is still scanty. Differing values are not the only familial explanation that has been offered to account for racial/ethnic variations in nursing home use. The National Nursing Home Survey (1977) showed less than half (44 percent) of nursing home residents to be receiving "intensive" nursing care.25 Such data often are cited to support the point that the use of nursing homes depends in part on the availability or non-availability of family members (usually the spouse or adult offspring) to provide care at home.26 Thus, the living arrangements of elderly and disabled persons may play a significant role in their need for long-term care, particularly in intermediate care facilities, and it is suggested that certain living arrangements found more commonly among minority groups may well facilitate home care for dependent elderly people. Without question, there are racial differences in the frequency with which certain living arrangements are found. Detailed data on the living arrangements of people aged 65 and older were collected in the 1968 Social Security Survey of the Demographic and Economic Characteristics of the Aged. Table 18 summarizes the data on race and living arrangements. Non-whites were less likely than whites to be living as married couples. However, white households (for both married couples and unmarried women) were much less likely than black households to include other relatives (most commonly children and grandchildren). The Census Bureau's Annual Housing Survey provides

83 Table 18. LIVING ARRANGEMENTS BY RACE: PERCENTAGE DISTRIBUTION OF AGED UNITS BY TYPE OF ARRANGEMENT, BY RACE OF UNIT, 1968 Nonmarried persons Type of Arrangement Married Couples Men Women Total Number White Negro White Negro White Negro (in thousands) 5,584 386 2,090 251 6,852 567 Total Percent 100 100 100 100 100 100 No relatives present 82 64 66 67 62 48 Alone 81 61 51 50 50 .37 With nonrelatives 1 3 5 15 3 8 In institutions * * 10 2 a 3 Relatives Present 18 36 33 33 38 52 Children 14 20 21 18 28 33 Under age 18 only 1 3 * 2 * No children 4 16 12 14 10 19 Grandchildren 4 17 11 9 10 21 Brother or sister 1 1 9 8 8 9 Parents * * * 2 * * Other relatives 4 11 17 15 16 22 *0.5 percent or less. SOURCE: Janet Murray, "Living Arrangements of People Aged 65 and Older: Findings from the 1968 Survey of the Aged," Social Security Bulletin (September 1971) p. 7. additional information on racial differences in living arrangements (Table 19). The black elderly are dispersed among a larger number of households, reflecting higher rates of separation and divorce. Thirty percent of black households include a person aged 65 and above, compared with 22 percent of white households and 16 percent of "Spanish origin" households. Almost all (95 percent) whites aged 65 and above live in households headed by a person aged 65 and above, while this is true for just over half (54 percent) of blacks. Finally, the Census Bureau's figures show that whites aged 65 and above are much more likely than either blacks or Spanish-origin persons to live alone. The percentage of the black aged who share a household with their grown children is larger than for whites, but blacks are only slightly more likely than whites to live near their children. Thus, a national survey of the aged conducted in the mid-1970s found that the portion of old people whose nearest child is either in the same household or no more than 10 minutes away is 59 percent for blacks and 52 percent for whites.27

84 U dp ^ dp dP 9i CN fO vo £ en CO N CM 0i VO -H in c CO O *P -H O O O O r- 'O CU (-3 O 0 o o o< t-~ rH ai en 0i 0 o o o G 0^ O C < 01 C ft ft •H H — JC •H > 0 CN o Oh m u £8 c £ 2 2 m in in fc •H 0 JO < in 00 rH o H 3 to ca < « SB O j * hi vo m ft X C 0) TJ T—( H O 0* C « a u a 1 K PQ dP dP Of dp o M X^ 9i ftji *S* »* g Si CO at in t- * g M 0) O o o 0 « a -•« W O o o o 0 in 'a S 9JB o o o o vo o ft ft ft ft 'a in rH o ^* l~~ a 0i C IO fry 00 00 f^ f~- PL] O 0i 'COG O CO CN ro on ^D — • ca <0 to i0 M ft ft H nil 3 0i ^ in rT-J H O x 0i m -— 1 H ft 2 SB dl VO o EH" 3 D 5- dp Of dP dP ea g X^ f) N O vo 2 A CM N CO rH 0i 1 •H H C *O 3 1 •iH fi a 3 C 'O fa c o in rH O o 4J VO O O O o a QJ a O O O 0 3 CO CD i0 T3 o o o 0 C • u 0i 0i 0i ft ft ft M --^ t3 ^ rn *% o o en CO <c ^ EH PQ r-t 4J 2 *fll O OB o f^ CO ~" 0 jj <o c 2 9i o (*) m *h 1 M ft ft ft CO 0 g 0i CO VO CN 3 in E Cfi CO H rH CO rH 3 C M C 1 ^j O •H « 0) X *£ BB A u pc CO 2 0i 0i pr* JC -H u 0 o O 0 43 o! a 4-1 0 o o o VH CQ o 'a o o o o o 2 rH ft ft. % % 0 ^~ H O 0) ^. o o vO ^* CO O T3 CO rH in rH 3 CO 8 < 0i nj iO CM ot VO <0 U — a PH 0i % ft 0) fH BB 3 X in n r- m M JJ O r^ VD 3 CO BB .£ 00 -H f M 9* c 0i H .C -H .. 4J io en w o « rH 0) j^ •H -H u <o H X! <0 EH 3 *J 1° Ski 10 ,H iH 01 CQ CQ CJ

85 Differences in living arrangements, however, do not necessarily imply differences in family resources available to the elderly. For example, in Shanas's national survey of the aged, there was virtually no difference between blacks (78 percent) and whites (76 percent) in reports of having visited with children during the previous week.^8 It appears that the racial difference in living arrangements may be an indicator more of economic or cultural differences rather than differences in the degree of social isolation of the elderly. Nevertheless, living arrangements that are found more often in the black family would appear to facilitate the home care of elderly persons who have become partially disabled. Yet there is some evidence that there may not be racial differences in whether partially disabled, elderly people are able to obtain needed assistance within their homes. Shanas's survey of the black aged focused particularly on patterns of assistance for elderly persons with various types of disabilities. The data are shown in Table 20. There was little overall racial difference in receipt of needed care from family members, although for whites this was more likely to come from a spouse (reflecting the racial difference in marital status) and blacks were more likely to be helped by relatives (other than children) living in or outside the household. Persons who had been "ill in bed" were asked about receipt of three types of help; whites were less likely to have received help in these circumstances. For persons needing three types of help (persons who were unable to care for their feet, persons unable to perform heavy household tasks, and persons having difficulty with meal preparation), blacks were slightly less likely than whites to have received help. Whites were more likely than blacks to have purchased help (from a podiatrist for foot care or a paid helper for household tasks). There was no consistent racial pattern in receiving help from unrelated persons outside of the household. Similar findings come from a study of "functional social networks" of black, Hispanic, and white elderly persons in New York City.30 Although there were differences among these groups regarding which sources of help were primary (for example, blacks were lowest in reliance on spouses, and Hispanics were notably high in the importance of children), those that mentioned no help sources were very few and were in roughly the same proportion among whites (6.6 percent), blacks (4.3 percent), and Hispanics (4.6 percent). Although the data from the Shanas study show living arrangements to be reflected in the sources of assistance received by the elderly, neither of these studies shows an overall, consistent racial difference in the ability of the elderly to obtain needed help outside of a nursing home. This casts doubt on the hypothesis that blacks are not found in nursing homes because they are more able than whites to obtain needed care at home due to living arrangements that are more typical of the black family. Firm conclusions about the role of family factors and living arrangements in explaining racial/ethnic differences in the use of nursing homes must await the conduct of additional studies in which the living arrangements of patients (with specified levels of disability or with carefully defined needs for

86 in 1| • in vo a* vo o to 0i 1 rH *H rH M rH H I m S k , r* *r u*i ft (MO «r Q A « M H rH rH W * 1 Jt U 4J 0) rH i•B 31 in o in ^ rt 5 * -0 O rH rH -H rH .»4 £ to 0) tt 0i H U *J tt ) 3 3 4J ^ -H CM W m v0 CM rH £ ^ n SO) rH -nS ° Sen o tn r- r- CD »-i M W m 4-J t/j OJ a 4J B CO ^ •H 3 3 o* oi o o -H * ^ vo m * « ra K E £ 0 H tt a CM rH * mO 0 0 £< ? g rH ffl a ,H -rl ft) T) O b V ,nO rH O N rH CM 8 @ • £ ^ < U | M Ju S * " S ". 0J jy a is a 2 S « J5 a a S c *i a\ n CM w r* i n S Q? 0 -H n 8 & 4J M r-t rH T-t a 4J m 9 ,^ r* 0 B U .D 4J S *1 £ rn 3 ^ 5S „ 8 in 0m v CM .w w co rH rH rH rH rH *H rH o I S S CO 4J £ § -* m ^ <r CM ^ ^ M W o x ' rH 1 ssl • *• in r- t£,, -T o • rH rH rH rH rH H rH C ,^ w 0) c ps ° O 0 ^ » * 0 CM *A « i § .-.. • £ rH *H tN rH iH s « 0 •l J£ a rH CM m r- m rH M £ 3 ,rl rH «i rH rH iH rH rH rH (M ^H *-4 QJ • CQ - CO ^ a 3 0 K en ° S ,rl C*l rH -r 00 VO r-- f rH rH rH H rH CM S • M 0 -H S* w O • ' ' <!1 f' •* Q O* H *O *"* M rH *N CM -H fe D S a i i 4J 9 O 0 Q V U *H Q 3 4J S S C0 -H 00 rH OV Ch VO m Jg f*i T f^ CM M -r S * O 0) • KB 5 5 CO rH S 2i U £| g,j a ^s ; jj j; C 4J 4J C i-l BlH«}4J tt4JC 8 -. mo-D o-a S'Q'o S* crj-o C-IQ 5 I ,M J= ft) ,-* *rj CL £: 01 OU 0 * -1 03-H tt V £ K4JU0*JCkl-rfa 0 CM *nc t- o aac VQ t ^ f ? ^ * *w U 0 C c .H « £ -D c .H a o a f • a'ua ^ o d > 4) x: o 4-* £3 "-*a. c n 0) UttrH 0) Xi 4JBr-l U b| Q i-, .o gj 3 S, B 3f CJ ^ a C c > 0! -* 44 C vuC *MDtfX U i0 < fM0 «TJ k, *-* rHuftt rH$ rH0fl •-!>a 00 00rH 0lO0l 0)0) 0)«ftI «ia« MA xa-n n *M 2 xc nc£ n £ I v 5

87 assistance) from different racial/ethnic groups are compared and related to familial characteristics, socioeconomic factors, and the expressed preferences of patients and caregivers. The role of values, family, and socioeconomic factors (for example, the resources to hire someone to provide assistance in the home) cannot now be distinguished from the role of the non-availability or the inaccessibility of nursing home facilities. Thus, even if it could be shown that elderly blacks with a given level of disability or need for assistance are more likely than elderly whites to stay with their families, the role of choice in this matter would remain unclear. That is, such a finding could be due to values or structural factors (such as multiple generations in the same household) that are more common in the black family, or it could be due to an attenuated range of options that may be available to many black families. That familial and residential factors may not explain low black use of nursing homes is reinforced by data on overall patterns of institutionalization among black and white elderly populations. The 1970 census data show that only 56 percent of the institutionalized elderly non-whites were in "homes for the aged," compared with 80 percent of institutionalized elderly white persons, while elderly non-whites were markedly overrepresented in mental hospitals and chronic disease hospitals (such as TB).31 The extent to which these trends reflect blocked access to nursing homes, rather than racial differences in health status, is unknown. Data on the extent of black residence in unlicensed boarding and personal care homes would also be instructive in evaluating the extent to which the low representation of blacks in nursing homes reflects blocked access. Less information is available on the role of familial factors in affecting nursing home use of ethnic minorities other than blacks. A recent Federal Council on the Aging staff report on elderly minorities suggests that the "bond between the natural support networks and older persons" is undergoing more stress among Pacific Asians and Hispanics than among blacks.32 However, little evidence was presented to show that this is true. The report also suggests that members of ethnic minorities may be particularly reluctant to use long-term care facilities, because of "fear of being removed from their cultural surroundings," and emphasizes the importance of cultural factors and language barriers in influencing ethnic minorities' ability to obtain needed services. These factors may have particular force on the long-term care context (as opposed to acute hospitalization) because of its residential aspect. Although the causal role of values and familial factors cannot now be assessed with certainty and may differ among various ethnic minorities, evidence examined by the committee suggests that, at least with regard to the black elderly, the most important factors underlying their low use of nursing homes pertain to the non-availability of beds. Beds may be unavailable either because they do not exist or because they are in some sense reserved for other people. Most nursing homes are private and can set their own rules for admission, such as an unwritten rule that you must be able to pay your own way for a year before going on Medicaid, with its lower reimbursement rates. The

88 factors underlying the non-availability of beds and the reasons why this, rather than values or familial factors may best explain the low rates of nursing home use among blacks, are examined in the next section. Factors Affecting the Availability of Nursing Homes for Minorities A number of factors may result in restricted availability of beds to minority group members. These include location of nursing homes, patients' ability to pay for care, and racial discrimination. Geographic Factors The location of nursing homes may negatively affect their use by minority groups in two ways. First, the lack of proximity of nursing homes to minority neighborhoods is a possible factor. Location near family is frequently a factor in selection of a nursing home.34 The National Nursing Home Survey (1977) showed that almost two-thirds of nursing home residents had visitors on a daily or weekly basis, usually from relatives. However, the committee did not locate any studies of the geographical patterns of nursing homes. However, given what is known about the availability of other types of medical care, it seems likely that nursing homes tend to be located away from predominantly black areas of cities. The same is probably true for the rural South. Second, nursing home beds tend to be in shorter supply in states with relatively high proportions of blacks. Scanlon, Difederico, and Stassen suggest that at least part of the national racial difference in nursing home use may be accounted for by "the concentration of black elderly in low income states, particularly the Southeast," that may not be willing to support a large nursing home population with public funds.36 About one-half of blacks in the United States reside in the South,37 which is the region with the lowest rate of utilization of nursing homes. Approximately 2.6 percent of the population aged 65 and above in the South reside in nursing homes, compared with 5 percent in New England and 4 percent in the Pacific states.38 The nature of any causal relationship between the concentration of blacks and the existence of nursing home beds is undoubtedly complex. Financial Factors The association between race and income in the United States is well established.39 Data from a recent national survey by the National Center for Health Statistics, for example, found family incomes of under $5,000 for 30 percent of blacks, 19 percent of Hispanics, 12 percent of Asian or Pacific Islanders, and 11 percent of whites.4^ The elderly black population is disproportionately dependent upon Medicaid for meeting the expenses of long-term care. The 1977 National Nursing Home Survey found that personal or family income was the primary source of support for 40.5 percent of white nursing home residents but for only 13 percent of black residents. Conversely, Medicaid was the primary source of support for 72.5 percent of black residents and for 46 percent of white residents.

89 Dependence on Medicaid has significant disadvantages for persons seeking nursing home care. In some states there have been serious delays in the process of making determinations of eligibility for Medicaid.41 This may result in patients remaining in hospitals who could be cared for in nursing homes. More important, because the demand for nursing home care exceeds the supply of beds in many, perhaps most, locations and because three-fourths of nursing homes maintain a waiting list, nursing home operators often have the opportunity to choose between a Medicaid patient and a private-pay patient when a vacancy occurs.42 Payment levels under Medicaid are controlled by each state and, because of cost-containment considerations, are generally lower than the rates charged by nursing homes to private-pay patients. The latter may eventually exhaust their resources and become Medicaid patients, but initially they can be expected to pay more than the Medicaid level. Indeed, the practice of charging non-publicly supported patients more than cost in order to make up for deficits resulting from low rates paid for publicly paid patients—cross-subsidization—is well known,4^ and it is often argued that a substantial representation of privately paying patients is essential to the viability of nursing homes. In Scanlon's words, "Cognizant of its oligopsony power as the largest purchaser of nursing home care, the government does not pay the market price. Instead, the state establishes a rate at which it will reimburse homes for care for an eligible person."44 Rational economic behavior of nursing homes is to accept as many private patients as possible, even if some of them will eventually transfer to Medicaid. An additional deterrent to the admission of Medicaid patients is the fact that their admission is accompanied by government-required review procedures designed to restrict unnecessary use of services. These procedures can put the Medicaid patient at a disadvantage compared with the private-pay patient who may be admitted without such review.45 In addition to reimbursement rates that make Medicaid patients less attractive than private-pay patients, providers face more paperwork with the Medicaid patient and have complained in some states about delays in receiving payment under Medicaid. These various circumstances appear to be reflected in behavior by providers that is not to the advantage of Medicaid patients. In the words of the New York State Moreland Act Commission, "the problem of discrimination against Medicaid-paid patients is apparent to virtually every knowledgeable person from whom the Commission has heard."46 In the IOM committee's experience, knowledgeable people continue to express certainty that Medicaid patients are discriminated against by nursing homes because of the factors already mentioned. Scanlon, an economist, assumes that proprietary nursing homes "operate as profit maximizers" and therefore "will want to discriminate between private- pay and Medicaid residents."47 He goes on to argue that non-profit nursing homes will behave similarly because of their own incentives to maximize income. The forms of discrimination that allegedly occur are not confined to a reluctance or refusal to admit Medicaid patients. There may also be refusals to admit private-pay patients who are likely to exhaust assets and go onto Medicaid relatively quickly, and there

90 are allegations that nursing homes that do not accept Medicaid patients sometimes divest themselves of patients who are shifting to Medicaid payment. There is nothing to force a home that does not accept Medicaid patients to retain a patient whose private funds have been exhausted. Such patients must shift for themselves, joining the line of Medicaid patients waiting (often in hospitals) for available Medicaid space in a nursing home. With the different levels of payment for nursing home care, economic discrimination against Medicaid patients is inevitable. Yet the situation and its differential effect on minority groups is not well documented. Records may be available through utilization review programs that would at least allow comparisons to be made among hospitalized patients in the amount of time spent awaiting nursing home placements. Such an approach was used in a recent study of the "hospital backup" problem of hospitalized patients awaiting placement in nursing homes, which was conducted by the Office of the Inspector General in DHHS Region 10.48 Characteristics of such patients at a sample of 57 hospitals were collected. For 66 percent of these patients, Medicaid was to be the initial source of payment for their nursing home care (6 percent would be self-pay patients), and Medicaid was the likely eventual source of payment for 88 percent. By comparison, the 1977 National Nursing Home Survey found 38 percent of nursing home patients to be self-pay (including family payment) and 48 percent to be Medicaid patients. This 48 percent includes an unknown, but presumably large number of persons who began on other forms of payment and went onto Medicaid after their Medicare or private insurance benefits ran out or their assets had been depleted. (The proportion of nursing home patients who are on Medicaid at the outset must be considerably smaller than 48 percent.) Thus, the study suggests that Medicaid patients make up a disproportionately large share of the pool of hospitalized patients awaiting nursing home placement. Virtually nothing is known about the characteristics of non-hospitalized persons who are seeking placement in a nursing home. A study of such persons would involve more primary data collection and some difficult sampling problems, but is necessary to a full understanding of the placement issue since placement from home involves different processes and actors than placement from a hospital. Adequacy of These Explanations It is likely that the low use of nursing homes by blacks may be due, in part, to successful adaptations within families and neighborhoods, the disproportionate location of blacks in states where nursing homes are in short supply, and their relative poverty and disproportionate membership in a class—Medicaid patients—that is itself discriminated against. However, examination of patterns of nursing home use reveals a number of other aspects that cannot be readily explained in terms of these factors. Among these patterns are some large racial differentials among Medicaid patients, patterns of racial segregation among nursing homes in some locales, and variations from place to place in the extent of black use of nursing homes.

91 Racial Differentials in Medicaid One way to assess the role of economic factors in the lower black usage of nursing homes is to examine variations among persons who have a similar economic position and who are using the same source of payment for care. Racial differences within Medicaid, the largest single source of payment for nursing home care, could hardly be attributed to economic factors. Statistics published by the Health Care Financing Administration about the Medicaid program have been the source of much concern about inequity in the program. However, data problems similar to those discussed in Chapter 2—the absence of data regarding the population from whom Medicaid beneficiaries are drawn and failure of many states to report utilization statistics by race— make it difficult to assess adequately the question of equity in nursing home expenditures under Medicaid. For example, the DHEW publication, "Health Status of Minorities and Low-Income Groups," includes a chart showing the racial distribution of different types of services under Medicaid. The ratio of whites to non-whites (per 1,000 beneficiaries) is shown to be 3.23 for "nursing homes" and 4.01 for intermediate care facilities.49 However, in the absence of data about either the characteristics of the pool from which beneficiaries are drawn or the age distributions of white and non-white Medicaid beneficiaries, interpretation of even such large apparent differences is speculative. The differences may be due, for example, to the greater frequency of the aged in the white Medicaid population. Nevertheless, these data are sometimes cited as evidence of racial inequity in nursing home care under Medicaid. However, some less ambiguous Medicaid data suggest that problems of racial equity exist regarding nursing home care. Several states that have significantly large minority populations do report their Medicaid utilization data by race. Although these states cannot be assumed to represent the United States as a whole, it is, nevertheless, instructive to examine their racial patterns in use of long-term care facilities under Medicaid. Data on racial patterns in Medicaid-paid nursing home use in this selected and diverse group of states is shown in Table 21. Data are presented separately for skilled nursing and intermediate care facilities; the fact that there are substantial variations among states in Medicaid benefits (particularly regarding skilled nursing) must be kept in mind because this greatly affects the absolute number of beneficiaries in various states. The available Medicaid data are provided for patients classified as white or "other," because they were collected on forms that offered only those choices. (This has undergone revision in accord with the more recent government-wide policy to standardize the collection of racial data.) For comparison, state data on the racial composition of the population above age 65 also are presented (more than 80 percent of nursing home residents are above age 65, according to the 1977 National Nursing Home Survey). Finally, data on the racial makeup of the poor population aged 65 and above in these states are presented and provide the most relevant point of comparison since this is the population from which the Medicaid-paid nursing home population is primarily drawn.

92 0) V .^.J -p 4J •H c £ rH •P^ 0 f coinoo^voaovovo 00 r» ft Q O M C cnr^voNcocft^rHrH o CM r» s £ 0 § tH «N l~ CN CM i-l ** rH ro g &' vo -H CO § r- 0) 0) av u O PQ M (~H) *J M E 2 u i0 ** e '9 1 -H O ininr^vorHoooochr- K CO r^VOOtfNlD'VVOCA vo m r» » «1 w | •H intrHCNl^ino»r-(CM ft ft CD o Q 0) rH Z 0) ro^* ^rHvOi-nfMtn CO ft g | •H -H £ iH ro rH r<C a CO to g M 8 M ft. CO « s j5 *D « at Q 9 •H Q) 4J H <o 4-> «l •H S U •H r^ e JC ^inoorooroCMin ft o m **j 9 1 0 O f rHVO00r-lmenvOCMrH rH o m U CJ 0 c CM CN r)- CO ,-('-< CM rH p 0) 1 0 A S ft 8 JJ « •H O M £ |4 f ra JJ 5 0 CO C 0) ^-j 0i U ft s O OH E • M (Q Pi -H 1 ,H r-r^roroeftOininer1 m 00 C4 !o H O CftrHCOCOCNIOCMiHvO e •H oocnrH^iHOrsiroto m O\ rH w g « SB 0) 10 rH ^ H rH CM CM H 00 « < « < H H fti i m Q ! 36 CO M 9 vo U-l 0 51 { N O 9 5 U H Q -p >i - i0 o^r^meftCMvor^vo N iH1 * 3 < 0i u 0 ^ n r-voooCnmcf>^*rHo 01 O N H CO S S! ,H CO CN rH CN CM rH T H u £ 4J _ 0) 2 Q g «> fti • e M nj CO p S 2 g § MH f O H M O 2 • U i ss w a m rH rHinoNvorHinr^in M ft *+ ,_^ § M M u a> f • 9 r~oOrHHrot~rot~o to " * 0 CN rH rH CN rH ft < M M pa 2^ * < , 'H ^ Q s 4 H ,fcCQOt* r*1 •O C * -H ,H n CN CO toiOMio it fi • a JCOig-H • O • B> • CO M QJ 0i CSOl<03rH-Htt) •H Q (0 rH JJ (0^rHO CCklUC <Ol0MCQ4->>l.CC CO cn (4 Q 10 0} • -Q flj 4J <<S55se5s;E ,H •H 0 EH CO S 2

93 o in o> Ol rH CO TT O -J ro ^ CM CM vo vo vo m rH CO rH rH CM rn ^j- ,-j m 0) 4J * U 5 O vi hi 4J y^ 0 C B O ^j M CJ « vo kl T3 o r- [^ ^T ^j- O rH O m CM Ol C l~ Ol O 00 00 rH oh o r- r» rj <0 Ol rH O 00 CM Oi ^* VO ^ CO rH « * 'D a hi n in r- O rH C 4-> » i0 CM rH i0 C oo oi i0 o x rH kl •H rH O 4J iH 0) •H O *) -H 0i -H •H O lv ff. s c U i0 i0 - Ot m CM vc in t r~ oo • c vi a OQ T* frt Qj CM CM 0 CM ^ o in ^ £ rH rH rH rH CO CM rH CM 4J U JO 0i 0i i0 C kl o e a a •H 0 * • *•* £ C i0 ^ *O *J O -U 2 •H rH o w B i0 i0 H EC U 0) TJ Q •H &CJ • *Q »^H oo vo o CO CM Ol O *O c a •• r- IH o CO O rH O CO 0i rH a u . co oh in CM VO O Ol O 3S <^ " ^H *o • 04 VO CM o n r- vo oi hi rH . • e a CM rH CM rH rH f. rH 0) rH a •H S S3 . 8 >c 3 0 'O co en c o O 4J >! -H a1 ^H *^ nv n *j e 0) O M i0 -H 0i -H ^ 0) ki c Ji 1 C *J w oi a in r*~1 A U^ O -H £ c * H Q rH 00 VO rH O CO CO CO rH >, » iH rH 0 0i rH CO O VO CO o f: • C0 6 CO 3 'Q 0i rH rH ^T CM rH CM •H EH CJ co i^ o $ 01 D C -H 0) O1 > Ol c • *»* • O) O W CJ C M •H 0) IQ •• O •" 0 J= C 3 CO 4J VO r~ *J t0 •o rH * D1 0i ot c c O O C rH rH VW -H (0 C -^ -H O C*H c* r» .* co ^ o r» oh ^ 0 w «n " CO 4J « i0 -^ «M J|| , , — ' W JJ vo vo vo o oo m o r~ 3 (0 M C CM rH rH H O Vi & O a v to oi c a• £• a C vi CJ E rH 0i 3 >, ¥ 0*0 to -O O ffl £ i0 n *o •** o) <0 -H 0 • E rH 4J Q. c •H 4J U hi 0 W O <0 » 41 rH t0 - 0) 0 • hi 0i D] 2 o. £ -D W * D MH 32 4J v1 * t0 0i (0 4J 0I rH £ a •• -o •• S c Q i0 vi e * t) to 'H a 4J M H 0) PL) -H 0) O c " m t c 0i <K Vi 0i §> U Q. rj * JS O t0 O £ 0i U) -H 4J D" O *O •H « -H v i D B Di 4J C <0 C7^ 0i 3 C X M Vi O « 3 •-« 0 W H EH > ,o v, 5, c J-H rH -C .* 0 0 4J <0 0i D O Ol O 0) W iJ « * « * W D W OS

94 Table 21 shows that the black use of nursing homes under Medicaid is quite variable when considered against the racial makeup of the elderly poor population in the various states. For many states, the percentage of black Medicaid patients in nursing homes is quite similar to the percentage of blacks among the poor aged population. However, in a few states—most notably Alabama, South Carolina, and Mississippi—there are large differences. (There are a few counter- trends in a few other states where numbers are small.) In Mississippi, almost half of the elderly poor are black, but only about 20 percent of nursing home patients are black. In some states (for example, Maryland and Oklahoma) the racial composition of one type of nursing facility (skilled or intermediate care) reflects the characteristics of the poor aged of the state, while the other type of facility does not. This may be associated with state restrictions on benefits for one or the other type of nursing home, but it is not clear why the racial makeup of the two types of facilities differs. While the data do not explain the reasons for the variations within and between states, Table 21 does illustrate several important points. First, aggregate data for some states show a pattern that strongly suggests racial inequity in long-term care under Medicaid. For some reason, poor white elderly patients in these states obtain nursing home care at much higher rates than do poor black elderly patients. Second, the data show that states are not uniform in this regard; nursing home benefits in many states are provided in rough proportion to the needy population in those states. Thus, the low rate of nursing home use by blacks must be due to factors that vary from state to state. The fact that elderly black Medicaid patients in many states use nursing homes at roughly the same rate as elderly whites casts doubt on the suggestion that values and family structure generally underlie lower black use of nursing homes. Since it can be argued (based on the rates of disability and chronic illess among elderly blacks and whites) that use of nursing homes should be higher among blacks than among whites, it is still possible that familial factors have some effect on black use of nursing homes. However, since in many states the rates of nursing home use among poor blacks are as high as among poor whites, closer attention seems warranted in locales where blacks use nursing homes at substantially lower rates than do whites. Urban Variations in Nursing Home Use Although comparisons among cities and states in the relationship of white to non-white use of nursing homes can facilitate the identification of possible reasons for the overall racial difference in use of nursing homes, few studies exist that describe the overall patterns of use within specific geographic areas. Systematic data do not exist on the extent of variation from city to city in minority group use of nursing homes, but studies in Baltimore and Philadelphia suggest that there may be great variation. A study conducted in 1974 showed that the proportion of blacks in nursing homes both in Baltimore city and county exceeded the proportion of blacks in the 1970 census.50 In Baltimore city, where

95 blacks made up 25 percent of the population aged 65 and above, 32 percent of nursing home residents were black. By contrast, a study conducted in a similar city, Philadelphia, found that blacks made up only 13 percent of the nursing home population in 1978, compared with 20 percent of the population aged 65 and over in 1970.51 (No similar racial discrepancy was found in two suburban counties in the Philadelphia study.) The racial difference between the two cities apparently is not due to gross differences in Medicaid eligibility standards. Both states include the medically needy in their Medicaid programs, and Pennsylvania actually includes a larger percentage of the poor under Medicaid than does Maryland.5 It is unlikely that differences in values or family structures could account for such large differences in black nursing home use in the two cities. Thus, the explanation must be sought in such factors as racial attitudes, the supply of nursing home beds and their location, referral practices, and admission policies of nursing homes. Unfortunately, little information exists to enable the assessment of the way that these factors influence racial patterns in nursing home use in specific cities. Racial Segregation in Nursing Homes If racial discrimination affects the use of nursing homes by minority groups, it is reasonable to expect this to be manifest not only in low nursing home use among minorities, but also in patterns of segregation. That is, if discrimination by some nursing homes underlies the disproportionately small numbers of elderly blacks in nursing homes, it would presumably also cause those blacks who are in nursing homes to be concentrated in a limited number of such facilities. Thus, in addition to racial segregation itself being a cause of concern in a multi-racial society, the extent of racial segregation may itself provide an indication that discrimination is taking place, even though other factors may also contribute to patterns of segregation. If the nursing homes tend to be segregated in the same cities or states where there is also notably low black usage of nursing homes, this could not be readily explained in terms of familial factors or values. The coincidence of patterns of low use and segregation is consistent with the hypothesis that racial discrimination is at work. It is widely believed by persons familiar with nursing homes that they are characterized by a rather high degree of such segregation. Unfortunately, however, few data exist regarding the extent of racial segregation in nursing homes. There are historical reasons for concerns about segregation. Many facilities were established as private, non-profit institutions by religious or fraternal organizations (many of which are mono-racial) to take care of needy elderly members. Others grew out of the segregated "poor houses" of the South. Some of the factors that lead to the concentration of certain ethnic groups in particular nursing homes are quite understandable— persons who do not speak English well are undoubtedly more comfortable and better off in a nursing home with others of their own culture than in a home where they are linguistically and culturally isolated. The

96 same is true of persons who follow religiously imposed dietary restrictions. Nursing homes are in significant respects communities. It is not surprising that each seeks to create a harmonious group. Howeverr such rationales also can be used to justify racial discrimination. As in other areas of our society, racial and ethnic patterns may develop both because persons of similar economic and cultural backgrounds tend to cluster and because persons of different backgrounds may be actively excluded. Self-exclusion by minority groups may occur as well, either by choice or because of fear of mistreatment or abuse in an alien ethnic setting. The role that these various considerations and processes play in the racial/ethnic sorting of nursing home residents has received little empirical study. The permissibility of these processes, particularly when tax dollars are paying for care, has yet to be fully addressed. As in education, racial segregation in nursing homes may have an important impact on the resources available to the racial minorities. For example, in Kosberg's study of 214 nursing homes in the Chicago area, the adequacy of the treatment resources in facilities was negatively related to the percent of the facility's residents who were black. This largely reflected the resources that were available to the facility. Kosberg found a general pattern of few resources at institutions serving large numbers of poor (including black) patients because of the lower rates paid for their care. He also found that some nursing homes "managed to be rich in treatment resources" even though they had "sizable proportions of public aid recipients." These homes, he found, had raised the rates for private patients to make up for the money lost on public patients whose rates were below cost. Although there are obvious reasons for concern about racially segregated nursing homes, the matter has received little attention. The National Nursing Home Survey, conducted periodically by the National Center for Health Statistics, has not collected data that would allow measures, such as percentage of white patients, to be calculated for the separate nursing homes in the sample. The Nursing Home Survey's national estimates of minority group use of nursing homes are projected from data collected on the racial (and other) characteristics of only five sample patients per nursing home in the sample; thus, the data cannot be used as indicators of the racial composition of individual nursing homes and the extent to which minority group members are concentrated in a few nursing homes. The Census Bureau's 1976 Survey of Institutionalized Persons may have obtained data that could be used to examine the extent of racial concentration in nursing homes.5 Racial data were collected in interviews with residents of nursing homes; the sample was designed to include 10 residents in small institutions, 15 residents in medium-sized institutions, and numbers ranging from 15 to 40 in large institutions (where, incidently, blacks are disproportionately located according to the National Nursing Home Survey). However, no data have been published that show the degree of racial homogeneity or heterogeneity among residents (or staff members) of nursing homes. Despite the absence of national data on racial clustering in nursing homes, there are indications that it is common. Some studies,

97 focused on other issues/ simply take racial segregation as a given. Wershow's study of black and white nursing home residents in Birmingham and rural Alabama provides an example. In describing his sampling approach, he refers to the seven "predominantly black nursing homes" in the state and to their white counterparts. He indicates that the few whites in the black nursing homes are the "most isolated of the long-term state mental hospital 'discharges to the community' who seem to have been selected as the first white nursing home patients to 'break the ice' of racial desegregation in the predominantly black nursing homes." No information is provided about the number or characteristics of the black patients in the white nursing homes. Similarly, in her study of nursing home use in two East Coast cities, Schafft reported that there "remain racially identifiable hospitals and nursing homes. Repondents in the study referred to these institutions as 'black' and 'white'."56 Schafft indicates that, although she has wanted to study racially integrated nursing homes, she has found almost none in four cities with which she has become familiar—Atlanta, Georgia; Washington, D.C.; Wilmington, Delaware; and Richmond, Virginia. ^ Some scattered data are available about racial segregation in nursing homes. Race and admissions in 1978 to 133 nursing homes in Philadelphia, Montgomery, and Delaware counties in Pennsylvania were studied by two law students using records kept by the Pennsylvania Department of Health.58 In all three counties, blacks tended to be concentrated in certain institutions. For example, there were fewer than 3 black patients in 36 of the 62 nursing homes in Philadelphia County (there were no blacks in 18 of these homes); 95 percent of the black residents were in the remaining (45 percent) nursing homes. Blacks were particularly overrepresented in black-owned and in public and hospital-affiliated homes. Blacks were virtually absent from non-profit homes in Montgomery and Delaware counties. Data from the Baltimore study, though not presented in a way that allows for precise comparison with Philadelphia, suggest that a significant degree of racial clustering exists in Baltimore, although apparently less than in Philadelphia.5* (Data are not presented that would enable the calculation of quantified indices of segregation for more precise comparisons of degree of segregation. Such an index could be based on the average variation from the overall mean number of white or non-white residents.*) However, of the approximately 45 nursing homes in Baltimore city, 4 had no minority patients, and 7-10 others had very few; at the other end of the scale, 9 facilities had more than 50 percent minority patients. Apparently no other studies describe the degree of racial concentration in nursing homes or empirically explain patterns of racial clustering in nursing homes. The process by which persons and *One measure that has been used in several studies is the Index of Dissimilarity, which is defined as one-half the sum of the absolute differences between two percentage distributions.60

98 nursing homes select each other (and the degree to which, and circumstances under which, residents choose which nursing home they will use, as opposed to taking what is available) has received little study. Little is known about matters such as the extent to which patterns of ownership and location may explain patterns of nursing home use, the extent to which the location of minority nursing home residents represents their (or their family's) choice, and the factors that may influence that choice (including the range of options and information available). Racial Discrimination as an Explanation Most persons who have studied and written about black use of nursing homes believe that racial discrimination is a major explanatory factor for many of the racial patterns that have been described thus far. Testimony received by the committee strongly supports that point.61 Because of both the history of racial discrimination in the United States and factors related to the origins and ownership of particular institutions, the likelihood that racial discrimination exists approaches certainty. Racial discrimination is notoriously difficult to study directly (except in its most blatant forms) because it is usually not publicly acknowledged. The consequences of systematic discrimination, however, may show up in patterns such as have been described in this chapter—different rates of use of services, patterns of segregation, and so forth. Alternative explanations of racial patterns of nursing home use—demographic differences, differences in family values and living arrangements, and geographic and economic factors—appear to be inadequate to account for all the racial differences in nursing home use. If, as the committee believes, racial discrimination affects nursing home use, it is important to identify ways in which it may operate and to suggest some potentially useful lines of research. Discrimination may occur at many levels. At the federal and state levels, reimbursement practices reduce incentives to provide care for Medicaid patients, a population that is disproportionately made up of minority group members. State decisions on Medicaid coverage (for example, regarding inclusion of the "medically needy") have a strong effect on the ability of minority persons to obtain nursing home care. Recent analyses by economists suggest that governmental policies regarding the regulation of the bed supply in nursing homes may also work to the disadvantage of patients most in need of care.6^ Discrimination can occur at many points in the process by which persons are referred to and gain admission to nursing homes. Many nursing home admissions are arranged by discharge planners in hospitals, and decisions are made that match patients and facilities. Very little systematic information exists about the criteria and the practices used, such as racial "steering," in referral. Many admissions are arranged through relationships established between particular hospitals and particular facilities. Schafft noted that in the two cities she studied, racially identified hospitals remain, and "doctors who practice in the 'black' institutions and those who

99 practice in the 'white' institutions rarely cross over in placing their patients in nursing homes."63 Because of racial patterning in the use of hospitals, nursing homes can influence the characteristics of the persons who are referred through the choice of hospitals with which relationships are established. The source of referrals can narrow the population from which a nursing home is drawing. If nursing home vacancies were filled through a central registry, the opportunity for homes to exercise racial criteria in admission would be greatly reduced. The example of one city cited by Schafft suggests that a central registry of vacancies in nursing homes may have a significant impact on the racial patterning in nursing homes.64 In sum, it appears that nursing homes have some ability to control the characteristics of patients referred. The range and type of sources from which a nursing home accepts referrals can heavily influence the characteristics of those who seek admission, and admissions criteria (for example, with regard to Medicaid patients) of nursing homes may have the practical effect of limiting the access of minority groups. Many of these mechanisms are alleged to have been in play in Shelby County, Tennessee, according to litigation (Hickman v. Fowinkle, C.A. No. 80-2014, W.D. Tenn.) initiated in January 1980. There it is alleged that blacks are effectively denied admission to licensed, Medicaid-approved nursing homes and are relegated instead to unlicensed and unregulated boarding homes that do not provide the needed level of care. A number of practices are alleged to be involved, including giving preferential treatment to whites in admissions procedures, denying admission to applicants whose physicians do not have staff privileges at certain hospitals, and refusing to accept referrals from the county Department of Human Services.65 The evidence developed in connection with these allegations may help to illuminate some practices that tend to maintain racially distinct patterns, as well as to determine the legal acceptability of such practices. RESEARCH NEEDS More certainty about the influence of various factors on the use of nursing homes could be gained through the collection of additional data on a routine or sample basis. Priority should be given to the conduct of studies to better document (a) the extent to which race or ethnicity influences whether persons who need the services of nursing homes are able to obtain such care and (b) the extent of racial segregation in nursing homes and the processes that contribute to it. A variety of methods may prove useful in better documenting the characteristics and circumstances of partially disabled persons who do not reside in nursing homes. On the basis of census data and Medicare and Medicaid data, cities and states should be identified in which elderly persons from minority groups have a lower use of nursing homes. Efforts should be made to determine if race or

100 ethnicity influence whether persons who need nursing home care are able to obtain it. In some instances, information about persons who are awaiting placement in nursing homes may be available from hospitals, professional standards review organizations (PSROs), or local social service agencies. However, because little information is available about what happens to people who need nursing home care but are unable to gain admission, empirical research should be conducted to increase our understanding of how people cope with this situation. One useful model is provided by Shanas's Survey of the Black Aged (cited in this chapter), which provided information about the sources of assistance used by aged persons living outside of institutionalized settings. Better information is also needed about residents of unlicensed boarding homes, a population about which very little is known. It may be possible to sample such people by using addresses to which several Social Security checks are mailed (perhaps to persons who do not share a last name). There is also a particular need for more systematic information about the extent to which nursing homes are racially segregated. The possibility that widespread patterns of discrimination exist in nursing home admissions has received little attention, either as a topic for research or as an object of civil rights enforcement activities. Such information probably can be developed by making use of data that are already being collected for other purposes. Racial/ethnic information on Medicare or Medicaid claims and eligibility files provide a basis for describing the extent to which the racial characteristics of nursing homes in any locale depart from the overall racial/ethnic composition of the elderly in that locale. That is, do the nursing homes in a locale all have approximately the same racial/ethnic composition (and, if so, does this roughly parallel the racial/ethnic composition of the elderly population of the locale), or do the nursing homes tend to be mostly white or black? In locales where the nursing homes tend toward segregation, research is needed on the process by which people enter a nursing home. For patients (or families) who selected a nursing home, how did they do so? Do they believe they encountered discrimination? Other patients reach nursing homes through referral processes about which they may know very little. Better information about the operation of such referral processes is needed. What arrangements exist between institutions (such as particular hospitals and particular nursing homes)? To what extent can nursing homes control the characteristics of patients through the referral networks that they establish? To what extent do hospital discharge planners consider racial/ethnic factors in making referrals and how aware are they of informal policies of discrimination practiced by nursing homes? One source of information about racial/ethnic patterns in nursing homes is the data collected in connection with the annual certification procedures for nursing homes participating in Medicare and Medicaid. State certification teams are required by federal law (under guidelines issued in 1969) to conduct Title VI compliance

101 review activities that include a one-day census of the race/ethnicity of patient populations. Although many (perhaps most) states apparently engage in such activities, little is known about how these censuses are used, whether the data in any states have ever been compiled so as to illuminate racial/ethnic patterns in the use of nursing homes, or whether the information has been used successfully in civil rights enforcement activities. Thus, a potentially useful source of information about the characteristics of nursing home residents may be readily available, at least in some states. CONCLUSIONS This chapter has reviewed evidence pertaining to the question of racial disparities in nursing homes. While the chapter demonstrates the difficulty of disentangling the possible causes of such disparities in nursing homes, the evidence suggests that discrimination may well be a factor in nursing home admissions. Research approaches are available that would help to further clarify the reasons underlying the racial patterns in nursing home use and the mechanisms that produce racial disparities. However, only if racial differences are seen as deserving serious attention can we come to understand adequately the impact of institutionalized mechanisms that lead to racial disparities in the American health care system. REFERENCES 1. National Center for Health Statistics, The National Nursing Home Survey: 1977 Summary for the United States, Vital and Health Statistics, Series 13, No. 43, (Washington, D.C.: Government Printing Office, 1979) pp. 209-210. 2. Ibid., p. 8. 3. Ibid. 4. Bruce C. Vladeck, Unloving Care: The Nursing Home Tragedy (New York: Basic Books, 1980) p. 3. 5. The National Nursing Home Survey: 1977. 6. American Health Care Association, Long Term Care Facts (Washington, D.C.: AHCA, 1975) p. 7. 7. Congressional Budget Office, Long Term Care for the Elderly and Disabled (Washington, D.C.: Government Printing Office, 1977); United States Senate, Development in Aging: 1979, A Report of the Special Committee on Aging (Washington, D.C.: Government Printing Office, 1980). 8. Elaine Brody, "Environmental Factors in Dependency," in A. N. Exton-Smith and J. Grimley Evans (eds.), Care of the Elderly: Meeting the Challenge of Dependency (New York: Academic Press, 1977) p. 91. 9. Vladeck, p. 16.

102 10. William J. Scanlon, "A Theory of the Nursing Home Market," Inquiry 17 (Spring 1980) pp. 25-41. 11. Robert M. Gibson, "National Health Expenditures, 1978," Health Care Financing Review (Summer 1979) p. 26. 12. Scanlon, p. 25. 13. United States Senate, Trends in Long Term Care, Hearings Before the Subcommittee on Long-Term Care of the Special Committee on Aging, August 10, 1972, (Washington, D.C.: Government Printing Office, 1973) p. 2439. 14. National Center for Health Statistics, Characteristics of Residents in Nursing Homes and Personal Care Homes for U.S. June-August 1969, Vital and Health Statistics Series 12, No. 19 (Washington, D.C.: Government Printing Office, 1973) p. 2. 15. National Center for Health Statistics, Characteristics, Social Contacts, and Activities of Nursing Home Residents for U.S., 1973-1976, Vital and Health Statistics Series 13, No. 27 (Washington, D.C.: Government Printing Office, 1977) p. 3. 16. The National Nursing Home Survey: 1977, p. 29. 17. Characteristics, Social Contacts, and Activities of Nursing Home Residents, p. 6. 18. Department of Health, Education, and Welfare, Health United States: 1979 DHEW Publication No. (PHS) 80-1232 (Washington, D.C.: Government Printing Office, 1980) pp. 440-443. 19. Blanch Spruiel Williams, Characteristics of the Black Elderly, No. 5 in the Administration on Aging series of Statistical Reports on Older Americans (Washington, D.C.: DHHS, 1980) p. 40. 20. Ethel Shanas, "National Survey of the Aged: Final Report," Unpublished report on Social Security Administration Project Number 10 P - 57823, (no date) p. 5. 21. The National Nursing Home Survey: 1977, p. 45. 22. Jacqueline J. Jackson (ed.), Proceedings: Research Conference on Minority Group Aged in the South (Durham, NC: Duke University Medical Center, Center for Study of Aging and Human Development, 1972). 23. Carol B. Stack, All Our Kin: Strategies for Survival in a Black Community (New York: Harper and Row, 1974) p. 124. 24. Harold J. Wershow, "A Pilot Study Comparing Certain Characteristics of Black and White Nursing Home Patients in Birmingham and Rural Alabama," Report to Administration on Aging, Department of Health, Education, and Welfare (Grant 93p-5734/04-01), (no date) p. 9. 25. The National Nursing Home Survey: 1977, p. 53. 26. William J. Scanlon, Elaine Difederico, and Margaret Stassen, Long-Term Care: A Current Experience and a Framework for Analysis (Washington, D.C.: The Urban Institute, 1979) p. 24. 27. Shanas, p. 16. 28. Ibid., p. 17. 29. Ibid. 30. Marjorie H. Cantor, "The Informal Support System of New York's Inner City Elderly: Is Ethnicity a Factor?" in Donald F. Gelfand and Alfred J. Kutzik (eds.), Ethnicity and Aging: Theory, Research, and Policy (New York: Springer, 1979).

103 31. Beth Soldo, "Accounting for Racial Differences in Institutionalized Placements," presented at the Annual Meeting of The Gerontological Society, November 1977. 32. Federal Council on the Aging, Policy Issues Concerning the Elderly Minorities: A Staff Report, Publication No. (OHDS) 80-20670, (Washington, D.C.: DHHS, 1980) p. 40. 33. Ibid. 34. Ethel Shanas, Personal Communication, August 5, 1980. 35. The National Nursing Home Survey: 1977, p. 59. 36. Scanlon, Difederico, and Stassen, p. 27. 37. Bureau of the Census, 1976 Survey of Institutionalized Persons: A Study of Persons Receiving Long Term Care, Current Population Reports, Special Studies, Series P-23, No. 69 (Washington, D.C.: Government Printing Office, 1978) p. 13. 38. Scanlon, Difederico, and Stassen, p. 30. 39. Bureau of the Census, pp. 19-59. 40. Health United States: 1979, p. 7. 41. New York Statewide Professional Standards Review Council, Inc., "Quarter-Billion Dollars in Medicare and Medicaid Money Wasted in New York State on Hospital Patients Awaiting Move to Nursing Homes," press release, May 20, 1980. 42. Scanlon; The National Nursing Home Survey: 1977, p. 13. 43. Jordan I. Kosberg, "Differences in Proprietary Institutions Caring for Affluent and Nonaffluent Elderly," The Gerontologist (Autumn 1973) p. 303. 44. Scanlon, p. 28. 45. New York State Moreland Act Commission, p. 8. 46. Ibid., p. 5. 47. Scanlon, p. 28. 48. Office of the Inspector General, Department of Health and Human Services, "Restricted Patient Admittance to Nursing Homes: An Assessment of Hospital Back Up. Draft Technical Report," July 1980. 49. Health United States: 1979, p. 264. 50. Office for Civil Rights, "Baltimore City and Baltimore County Nursing Home Referral Study," Unpublished report (DHEW, OCR, Region 3, 1974). 51. Vanessa J. Lawrence and Jill Duson Mirach, "Racial Discrimination in Nursing Home Admissions in the Greater Philadelphia Area," Health Law Project Library Bulletin IV (January 1979) pp. 20-27. 52. Janet B. Mitchell and Jerry Cromwell, Large Medicaid Practices: Are They Medicaid Mills? Health Care Financing Grants and Contracts Reports series, (Baltimore, MD: Health Care Financing Administration, 1980) p. 14. 53. Kosberg, p. 303. 54. Bureau of the Census, p. 363. 55. Wershow, p. 5. 56. Gretchen Schafft, "Nursing Homes and the Black Elderly: Utilization and Satisfaction," Proceedings of the Fifth North American Symposium on Long-Term Care Administration (Bethesda,

104 MD: American College of Nursing Home Administration, 1980) p. 106. 57. Gretchen Schafft, personal communication. 58. Lawrence and Mirach. 59. Office for Civil Rights, pp. 8-9. 60. Gary S. Kart and Barry L. Beckham, "Black-White Differentials in the Instnationalization of the Elderly: A Temporal Analysis," Social Forces 54 (June 1976) pp. 901-910. 61. Schafft; Susan Conner (National Citizens Coalition of Nursing Home Reform), Testimony before IOM Committee, Washington, D.C., June 6, 1980. 62. Judith Feder and William Scanlon, "The Shortage of Nursing Home Beds," Journal of Health Politics, Policy and Law 4 (Winter 1980). 63. Schafft, p. 106. 64. Ibid. 65. National Senior Citizens Law Center, "Race Discrimination in Nursing Homes," NSCLC Washington Weekly 4 (June 27, 1980) p. 6.