Pathways to Assisted Living

Abstract

This article examines how race and class influence decisions to move to assisted living facilities. Qualitative methods were used to study moving decisions of residents in 10 assisted living facilities varying in size and location, as well as race and socioeconomic status of residents. Data were derived from in-depth interviews with 60 residents, 43 family members and friends, and 12 administrators. Grounded theory analysis identified three types of residents based on their decision-making control: proactive, compliant, and passive/resistant. Only proactive residents (less than a quarter of residents) had primary control. Findings show that control of decision making for elders who are moving to assisted living is influenced by class, though not directly by race. The impact of class primarily related to assisted-living placement options and strategies available to forestall moves. Factors influencing the decision-making process were similar for Black and White elders of comparable socioeconomic status.

The demographics of a rapidly aging population have fueled the demand for assisted living facilities (ALFs), a type of long-term care setting that bridges the gap between home and nursing home. In 2004, ALFs numbered 36,451 and served almost a million people (Mollica & Johnson-Lamarche, 2005), and the number of assisted living (AL) residents is expected to more than double, to 2 million, by the year 2030 (Assisted Living Federation of America, 2005).

The majority of AL residents nationwide are White (Hawes et al., 1995; Hawes, Rose, & Phillips, 1999; Morgan, Gruber-Baldini, & Magaziner, 2001; Sikorska-Simmons, 2006). Data indicate that higher poverty rates among African American elders contribute to their underrepresentation in AL (Ball et al., 2005; Belgrave & Bradsher, 1994; Brooks, 1996; Mui & Burnette, 1994). Most African Americans live in ALFs with 24 or fewer beds; almost all ALFs with majority African American populations have lower fees; and African American residents, compared to Whites, are more likely to have public financial support (Ball et al., 2000; Ball et al., 2005; Hawes et al., 1995; Morgan, Eckert, & Lyon, 1995; Mutran, Sudha, Reed, Menon, & Desai, 2001). Although African Americans also are less likely than Whites to reside in nursing homes (Gaugler, Kane, Kane, Clay, & Newcomer, 2003; Murtaugh, Kemper, Spillman, & Carlson, 1997; Wallace, Levy-Storms, Kingston, & Anderson, 1998), a higher percentage of older African Americans live in small urban ALFs than in nursing homes (Morgan et al., 1995). This finding may be attributable to the greater availability, financial accessibility, and cultural attractiveness of minority-owned and minority-run ALFs (compared to nursing homes) in large urban areas with significant minority populations (Ball et al., 2005; Morgan et al., 1995).

Most ALFs are privately paid for, with rates varying widely across sizes and types. The National Center for Assisted Living (2001) reported monthly private-pay fees ranging from $1,000 to $3,500, with an average of $1,873. A Metlife (2006) national survey found an average monthly cost of $2,968 for 2006. The majority of Americans cannot afford the larger, newer-model ALFs (Hawes & Phillips, 2000; Herd, 2001). Much lower fees ($423–$1,500) are found in small urban ALFs (Ball et. al., 2005; Morgan et al., 1995), and most poor residents live in ALFs with fewer than 16 beds (Stearns & Morgan, 2001). These homes are largely run by low-income older women caring for low-income older women (Herd, 2001). Because of limited public financing for AL, lower-income disabled adults, regardless of race, typically have the option of only small ALFs (also referred to as board and care homes). In 2004, 43 states had some type of Medicaid-reimbursed services in ALFs, but most public payments are low, and a relatively small number of beneficiaries are served in this setting—about 102,000 (Mollica & Johnson-Lamarche, 2005).

Despite the increasing importance of AL as a long-term care setting, little has been written about how people make decisions to seek this care option and how they negotiate the moving process. Our own study of African American elders in AL (Ball et al., 2005) compared the moving experiences of residents of small, low-income African American–owned ALFs to those of residents who moved to a large upscale facility targeting affluent African Americans. We found that elders’ pathways to both setting types were determined by multiple factors, including type and level of disability, financial resources, prior living arrangements, and family support. Underlying these more immediate factors were the life course experiences of racism and poverty shared by this cohort of elders.

Also common to these African American elders was their lack of control over moving decisions (Ball et al., 2005). For the majority of residents, the primary decision makers were family members (typically, adult children). These findings are similar to those in studies of White elders moving to AL (Ball et al., 2000; Hawes & Phillips, 2000; Kemp, 2008; Morgan, Eckert, Piggee, & Frankowski, 2006; Reinardy & Kane, 2003), congregate living facilities (Rossen & Knafl, 2003), and nursing homes (Holzapfel, Schoch, Dodman, & Grant, 1992; Reinardy, 1992).

However the process transpired, we found that the more control that residents had over the moving process, the happier they were with the move (Ball et al., 2005). Because moving is a process consisting of discrete stages (including recognition of need, deciding, preparing, moving, and settling in; see Young, 1998), control is a matter of degree. Each opportunity for control increased residents’ acceptance, even embracement, of the decision to move. Other research supports our finding that control over the decision-making affects adjustment and satisfaction with placement. Young (1998) found that residents who initiate their moves to retirement housing have the highest levels of satisfaction with both the decision and the overall moving process and so felt more at home in their new surroundings. Numerous studies of nursing home relocation have also demonstrated a positive relationship between control of decision making and resident well-being (Davidson & O’Connor, 1990; Forbes & Hoffart, 1998; Holzapfel et al., 1992; Mikhail, 1992).

We have found that among African American elders in AL, social class and access to material resources influence control over decision making (Ball et al., 2005). More affluent elders have greater choice over when and where they moved. Those who are poor have few resources to delay moves and can choose only the small, low-income homes. This earlier work is informed by the life course perspective, which views the life course as being shaped by one’s history and social context (Elder & Kirkpatrick Johnson, 2003; Settersen, 2003). The societal position of these African American elders, as defined by their race, class, age, and gender, continues to influence access to resources throughout their lives.

This article builds on this earlier work that examined the decision-making process of African American elders who were moving to AL, by comparing their decisions with those of White elders making similar decisions. We broaden our investigation of race and class by including ALFs with majority White-resident populations and those located in small towns, rural areas, and urban areas. Our aims were to increase understanding of how decisions are made to move to AL and how race and class influence the decision-making process. The increasing number and diversity of older persons who are choosing AL, coupled with the array of available AL settings, underscore the value of knowledge about the decision-making process. Such knowledge can be used by older people and their families, whatever their race and level of resources, to make the best possible choices.

Method

The data for this analysis come from two in-depth qualitative studies exploring independence and autonomy in ALFs, carried out between the spring of 1999 and the fall of 2001. A key research focus in both studies involved how decisions are made to move to ALFs. We conducted each study in five ALFs in Georgia, where they are termed personal care homes and broadly defined in statute to include a range of size and type. Georgia has 2,359 licensed ALFs with 28,512 beds (Georgia Office of Regulatory Services, 2007). Medicaid funding is available only in small homes (2–24 residents) through Georgia’s Home and Community Based Services Medicaid-waiver program. Available data suggest that Georgia AL residents resemble the national profile of an increasingly impaired and majority White and female population (Ball et al., 2000; Ball et al., 2005; Georgia State Health Planning Agency, 1993; Hawes, Rose, & Phillips, 1999).

Research Sample

We used purposive maximum variation sampling (Patton, 2002) to select the 10 ALFs that primarily served elderly residents and varied along the dimensions of resident race and socioeconomic status, availability of resources, size, and location. The 5 homes of Study 1 were small (6–18 residents) one-story structures located in the metropolitan Atlanta area. All were owned and operated by African Americans, and all served residents with low to moderate incomes; all or most residents were African American. Fees ranged from $430 to $1,500; Medicaid reimbursement was available in two of the homes. The 5 homes of Study 2 varied in size; race of residents, owners, and staff; geographic location; and ownership. Two homes were small (6–9 residents) and privately owned. One was located in a rural mountain setting and the other in a small town in a rural county. Fees at both were low ($870–$1,500), and both participated in the Medicaid-waiver program. All residents were Caucasian. A third home, converted apartments, was adjacent to a nursing home in an exurban community. It had 42 residents, all but one Caucasian, and fees were low to moderate ($1,200–$2,000). Two additional homes were urban purpose-built ALFs with high fees ($2,100–$5,650). One had 46 residents: It was located in an African American neighborhood, and all residents were African American. The other had 81 residents, almost all Caucasian. Both had dementia care units. Table 1 shows the pertinent characteristics of the 10 sample ALFs.

Table 1.

Characteristics of Sample Assisted Living Facilities (n = 10)

Location	Residents (n)	Race of Residents	Monthly Fees ($)
Urban	9	All Black	500–800a
Urban	8	All Black	500–1,500
Urban	8	63% Black	850–1,000a
Urban	18	55% Black	430–1,050
Urban	12	All Black	500–?
Rural,mountain	6	All White	870–1,500a
Small town	9	All White	870–1,500a
Exurban	42	98% White	1,200–2,000
Urban	46	All Black	2,095–3,645
Urban	81	94% White	2,300–5,650

^aHome participates in Medicaid-waiver program.

Table 2 summarizes the characteristics of all the residents who resided in the 10 homes during the data collection period for each study. The higher-than-usual percentage of residents who were under the age of 64 (15%) and male (30%) reflects the inclusion of the small homes that served residents with mental health problems and homes that participated in the Medicaid-waiver program. The combined 10-home sample is almost evenly divided between African Americans and Whites, with only three residents representing other ethnicities.

Table 2.

Characteristics of Residents in All Facilities (N = 201)

Characteristic	n	%
Age
18–44	7	4
45–64	22	11
65–74	20	10
75–84	88	46
85+	55	29
Sex
Male	60	30
Female	141	70
Race
African American	98	49
White (non-Latino)	100	50
Other	3	1
Health problems
Heart problems	59	32
Diabetes	35	19
Incontinence	51	27
Mental health diagnosis	42	23
Dementia	62	33
Depression	42	23

Note: In the three larger homes, data were derived from administrator reports of aggregate data; in the seven small homes, data were collected by means of interviews and record reviews. Owing to missing data in some categories, total sample size varies.

We used purposive sampling (Patton, 2002) to select residents and residents’ family members and friends for in-depth interviewing. We selected residents based on variation in age, functional status, health condition, race, and gender, all characteristics expected to affect their independence and autonomy. Family members and friends were those individuals most involved in residents’ care, as determined by direct observation of their activities and information provided by residents, home owners, and staff. The majority were children (22 daughters and 5 sons), but siblings, stepchildren, nieces, a nephew, a grandson, a mother, and two friends were also included. We selected relatives and friends of those residents who were interviewed and those who were not. Participants were recruited by telephone and through personal contact during facility visits.

Data Collection

In both studies, we collected data in each home in periods ranging from 12 to 17 months, utilizing participant observation, informal and in-depth interviewing, and review of residents’ records. Using detailed guides, we conducted in-depth interviews with 60 residents, 43 family members and friends, and 12 administrators. Interviews with residents and their family members and friends addressed the following: how and why moving decisions were made; who was involved in the decision making; what attitudes were held toward the move; and what the residents’ prior living situations were, including their life histories, health status, care needs, self-care practices, finances, social relationships, and informal support. We interviewed the owner or executive director in each home, as well as the marketing directors in the high-fee homes. Administrator interviews addressed facility policies and procedures regarding admission and marketing, services, and fees, as well as the administrators’ knowledge of residents’ health conditions, care needs, and experiences and attitudes related to moving. In general, the owners and administrators of the small homes, compared to those of the other homes, had more knowledge of residents’ moving circumstances. All formal interviews were tape-recorded and transcribed verbatim. We also collected information about the moving process, through informal interviews with residents, family members, friends, administrative and care staff, and two social workers. Informal interviews were recorded through detailed field notes. We obtained demographic and health data from residents’ personal records.

Apart from an initial interview with each facility’s owner, manager, or director, we conducted all formal interviews after spending sufficient time to build rapport with participants and so determine their suitability for inclusion—at least 3 months in each home. Facilities varied in the type and number of participants. For example, in one small home, none of the eight residents who lived there during data collection were able to be interviewed, because of cognitive impairment. Instead, information about moving came from the home’s owner and the residents’ family members and friends. Although some residents whom we interviewed were unable to recall the precise details of their moves to AL, they were able to articulate their perceptions of their involvement in the process.

Data Analysis

Consistent with a grounded theory approach (Charmaz, 2006; Strauss & Corbin, 1998), a constant comparative method guided simultaneous collection, coding, and analysis of data. For the analysis on which we report in this article, we used a three-level coding process (Strauss & Corbin, 1998) to analyze interview transcripts and field notes from informal interviews that pertained to moving decisions. In the first stage, known as open coding, we examined all data, line by line, for emergent categories based on the research questions and the issues raised by the participants. We next developed codes in terms of their properties (attributes and characteristics) and dimensions (location along a continuum) and so applied them to interview data. For example, we identified family support as a key influence on decision making; then, we examined this category relative to who provided what, to whom, how, when, where, and how often. In the second stage of coding, axial coding, we related initial categories to other categories, or subcategories, through what Strauss and Corbin (1998) referred to as a paradigm model. This model links categories in a set of relationships denoting causal conditions, context, intervening conditions, action/interaction strategies, and consequences. For example, we linked family support to such categories as residents’ health status, race, and income; family history and care burden; decision-making patterns; and ALF options. As part of axial coding, we compared and contrasted data across participants and homes, using analytical diagrams and charts to facilitate analysis. Finally, we integrated and refined categories to form a larger theoretical scheme in the third type of coding—selective coding—in which we organized major categories around our core category (Strauss & Corbin, 1998): control of decision making. For many residents, data from two and sometimes three sources (residents, family members, and administrators) informed analysis of our key concept. In the few cases where residents and family members differed in perceptions of control, we attempted to reconcile the difference with a third source, or we reported the residents’ perspectives. We used the software package Ethnograph 5.0 as a way of sorting coded data segments to facilitate comparisons across facilities.

Protection of Human Subjects

Before entering the study homes, we obtained informed consent from the owners for general access to the homes to observe activities and interact informally with residents and staff. We then obtained consent from residents to review their written records, and we received consent from all interview participants. A few residents with mental illness and minimal cognitive impairment were interviewed. In these cases, facility administrators and family members were consulted to assess the residents’ ability to give informed consent. In the few cases of guardianship, family members or social workers gave consent for record review. The written consent form included an explanation of the purpose of the research, the procedures used, the time required, the extent of confidentiality, and any potential risks and benefits, and it assured participants that participation was entirely voluntary. All interview tape recordings, tape transcriptions, and other data were stored in a secure space, and numerical codes were used to identify participants. All names used in this article are fictitious, and we have altered other information (e.g., place of birth) to further disguise informants. Both projects reported on here were approved by the Georgia State University Institutional Review Board for the Protection of Human Subjects.

Findings

The Context of Decision Making

Disease and disability

All decisions to move into these 10 homes originated with a mental and/or physical disability that generated a need for care. About two thirds (65%) of the residents had disabilities that were primarily physical, mainly stemming from chronic diseases common in old age. Decline was typically gradual, but in cases of an acute event, such as a heart attack or stroke (about 7% of the total), it was abrupt. The remaining 35% of residents had mental disabilities. Most of these (27% of all residents) had some type of dementia. Others had mental illness—mainly, schizophrenia and depression. Both the type and the level of the disability determined care needs and so influenced moving decisions. Almost all residents with mental diagnoses lived in the low-fee homes. Level of impairment at the point of decision making varied widely, largely depending on the residents’ financial resources and the capacity of their support systems in their former living situations.

Socioeconomic status

The socioeconomic status of residents in these 10 homes was wide-ranging. In general, the fee level of the home in which they lived reflected their socioeconomic status. As a whole, residents of low-fee homes were poorer and less educated than residents of the two high-fee homes, regardless of race. About one fifth received supplemental security income, a clear indication of poverty, and only a few had income from sources other than social security or Veterans Administration benefits. In contrast, only one resident in the high-fee homes (out of 127 residents) received supplemental security income, a 60-year old man with cerebal palsy whose fees were paid by his sister. Although financial information was limited for residents of high-fee homes (owing to residents’ reluctance to divulge or their lack of knowledge), available data show that almost 40% had investment income that supplemented social security and other pensions. Furthermore, based on data from 56 residents, 45% of the residents of the low-fee homes were not high school graduates, and only 5 had college degrees. In the high-fee homes, all the residents whom we interviewed (n = 27) had graduated high school; 16 (59%) were college graduates; and 7 had postgraduate degrees.

Black residents and White residents were represented among the poor and uneducated, as well as the well off and highly educated. Harold Stamps, an African American resident of a small urban home, had no formal education, having spent his school years working “in the fields.” Later, he drove a taxi and did yard work. In contrast, Alma Burgess, who resided in the affluent home targeted to African Americans, was a retired college professor with a master’s degree. Corresponding disparity was evident among White elders who were residing in similarly juxtaposed ALFs. Margie Medlock, a White resident of the rural mountain home, who traveled as a child with her migrant farmworker family in tandem with the “picking season,” contrasts with Elsie Calhoun, a White resident of a high-end home who graduated from an Ivy League college and married a corporate executive. The socioeconomic status of these elders’ families also varied. Although Margie’s children prospered (one, a college professor; another, a business entrepreneur), Harold’s children remained poor.

Some variation in residents’ socioeconomic status was present within ALFs of similar fee levels. In the small homes, this variation reflected the practice of basing fees on what residents could afford. For example, the African American residents of one small, Black-owned urban home included two women who taught high school (both with master’s degrees), both of whom paid $1,500, and a man with mental retardation who received supplemental security income, who paid $500. None of the residents of the low-fee homes, though, could have afforded either high-fee ALF. Of the high-fee ALFs, the one where most residents were White had higher fees ($2,300–$5,650) compared to the one where all residents were Black ($2,095–$3,645). The director of the latter home stated that this facility’s fees were lower than those at facilities owned by the same corporation with majority White-resident populations because of the reality of the residents’ financial situations. Resident life history and financial data indicate that in these two homes, White residents as a whole were more affluent than Black residents.

Former homes

Table 3 shows residents’ former living situations according to their race and ALF fee level. A number of differences can be seen between residents of the low- and high-fee homes. Forty-two percent of the residents were living alone in their own homes and apartments. About a third of those moved from out of town—all of whom were residents of the two high-fee homes, thereby reflecting their financial capacity to make long-distance moves, as well as their past mobility and that of their children.

Table 3.

Residents’ Former Living Situation by Race and Fee Level of Assisted Living Facility

	Low Fee		High Fee		Total
	Black (n = 43)	White (n = 50)	Black (n = 45)	White (n = 43)	n = 181
Former Living Situation	n (%)	n (%)	n (%)	n (%)	n (%)
Alone, in town	13 (30)	16 (32)	13 (29)	11 (26)	53 (29)
Alone, out of town	0 (0)	0 (0)	11 (24)	12 (28)	23 (13)
With spouse	0 (0)	2 (4)	5 (11)	10 (23)	17 (9)
Own home with family	0 (0)	1 (2)	3 (7)	1 (2)	5 (3)
Family’s home	18 (42)	17 (34)	4 (9)	1 (2)	40 (22)
Senior residence	2 (5)	1 (2)	3 (7)	3 (7)	9 (5)
Assisted living facility	4 (9)	11 (22)	6 (13)	5 (12)	26 (14)
Nursing home / hospital	6 (14)	2 (4)	0 (0)	0 (0)	8 (4)

Seventeen residents (9%) were living with a spouse. Most of these (n = 15) moved to the high-fee homes, and the majority (n = 10) were White. No married couples moved to the small homes, which typically cannot accommodate couples. Almost one fourth (22%) of residents moved from a family member’s home. The majority of these moved to the low-fee homes. Most were elders who had relocated to the home of a relative—typically, a child—but a few were adults with disabilities who had lived with a parent.

Fourteen percent of residents, representing all home types, moved from other ALFs. Some, because of mental illness, had lived in small low-income homes for most of their adult lives. Nine residents moved from independent senior housing, and two from skilled nursing facilities. Six, all African American, moved to one low-fee Black-owned home after discharge from the state psychiatric hospital. Residents’ prior living situations had bearing on their subsequent decisions to move to AL.

Control of Decision Making

As reported earlier for African American elders (Ball et al., 2005), control of decision making for this larger sample lay on a continuum. On one end of the continuum, residents exerted primary control; on the other, they had none. Between these extremes, decision making was shared with another person—typically, a family member. Our analysis revealed three decision-making types based on residents’ involvement in the process: proactive, compliant, and passive/resistant.

Proactive residents recognized the need to relocate; they made the decision; and they took independent steps to initiate moving. Only about a fifth (21%) of all residents assumed proactive roles. For the remaining residents, control (in varying degrees) was in the hands of others—children (60%), other family members (22%), social workers (15%), and friends (3%). Social workers were decision makers only for residents of the low-fee homes—mostly, residents with mental illness and no family support. Residents in the compliant group (46% of all residents) “went along” with the decision of another person or persons, many in the end admitting that the decision was “for the best.” Passive/resistant residents (31%) had virtually no role in decision making. They were either unable to participate because of cognitive impairment (passive), or they wholly resisted the move (resistant), some even after moving into the ALF. Most individuals exerting control believed that they acted in the dependent person’s best interests and had no other choice.

Table 4, which shows residents’ decision-making roles according to their race and ALF fee level, suggests that financial resources (regardless of race) bestow control. Of the 38 residents who were proactive (21% of the total sample), the majority (79%) moved to the high-fee homes, including 14 Black residents and 16 White residents. Although a minority of White residents (n = 8) who were moving to low-fee homes were proactive, all but one moved to the 42-bed ALF with somewhat higher fees ($1,200–$2,000), indicating their greater material resources. No Black residents moving to the low-fee homes were proactive.

Table 4.

Residents’ Decision-Making Role by Race and Fee Level of Assisted Living Facility

	Low Fee		High Fee		Total
	Black (n = 42)	White (n = 51)	Black (n = 45)	White (n = 42)	n = 180
Decision-Making Role	n (%)	n (%)	n (%)	n (%)	n (%)
Proactive	0 (0)	8 (16)	14 (37)	16 (42)	38 (21)
Compliant	23 (55)	29 (57)	16 (36)	15 (36)	83 (46)
Passive/resistant	19 (45)	14 (27)	15 (33)	11 (26)	59 (33)

For some residents, the availability of a desirable ALF influenced whether and when the moving decision was made, determining in some cases a resident’s level of control. Class and race influenced ALF options. Affluent elders had multiple ALFs with comparable amenities from which to choose. African Americans, though, seeking high-end homes where they would not be in the racial minority, had only one choice at that time. Low-income elders—particularly, those poor enough to qualify for supplemental security income—and elders with mental illness had especially limited options. Participation in the Medicaid-waiver program (which paid a portion of the ALF fees) increased the choices for some (9% of all residents). In the following section, we discuss patterns of decision making using case examples to enhance understanding of how race and class influenced the moving process. Table 5 summarizes these patterns.

Table 5.

Summary of Decision-Making Patterns

Proactive	Compliant	Passive/Resistant
Resident initiates moving     process and has primary     control.	Other person initiates the     process and has primary     control.	Other person initiates the     process and has complete     control.
Other persons have supportive     roles in locating and     selecting homes and     moving.	Resident involved in decision     making and is compliant.	Resident unable to participate     because of cognitive     impairment (passive).
The majority move to the     high-fee homes.	The majority move to     low-fee homes.	Resident is forced to move     (resistant).
Push-and-pull factors     influence decisions.	Push factors influence     decisions.	The majority move to low-fee.

Patterns of Decision Making

Proactive residents

Decision-making patterns of proactive residents differed from those of residents with lesser or no control. The majority (63%) of proactive residents were living alone or with a spouse, in their own home, in or out of town (see Table 6).

Table 6.

Residents’ Former Living Situation by Control of Decision Making

	Proactive (n = 38)	Compliant (n = 80)	Passive/Resistant (n = 59)	Total n = 177
Former Living Situation	n (%)	n (%)	n (%)	n (%)
Own home alone, in town	10 (26)	30 (38)	17 (29)	57 (32)
Own home alone, out town	5 (13)	9 (11)	7 (12)	21 (12)
Own home with spouse	9 (24)	2 (3)	5 (9)	16 (9)
Own home with family	3 (8)	2 (3)	1 (1)	6 (3)
Family’s home	2 (5)	21 (26)	15 (25)	38 (22)
Senior community	1 (3)	2 (2)	3 (5)	6 (3)
Assited living facility	7 (18)	13 (16)	5 (9)	25 (14)
Nursing home / hospital	1 (3)	1 (1)	6 (10)	8 (5)

Most decisions were intentional and considered, with the decision-making process extended over months or even years. In only four cases was the process compressed, all owing to a health crisis. Although these proactive elders had differing needs and resources for care, all had experienced declining health and felt overburdened in their former environments. Certain stressors were common, including uncertainty about the course of disability, loneliness, fear, and the demands of home maintenance. A combination of push factors influenced most decisions. Helen Brooks, an African American who at age 86 had been living in her home for over 50 years, provided a typical example:

Well, it took certain experiences at home to get me out, for instance, the falling, and I began to be fearful of being there by myself.… I got where I really couldn’t do the things that I had been doing in the home… and I decided that probably the best thing for me to do was to get somewhere else.

Similar factors pushed Elsie Calhoun to leave Florida and relocate to the high-fee, majority White-resident ALF near her son:

No, it was just the fact that I was almost 85 at the time. I was really tired of running a house. I lived in very hurricane territory down on the Keys. I was a little tired of coping with that. This seemed to be the best spot.

Comparable push factors motivated nine married elders to leave their homes. Three were the primary caretakers of their spouses (two with dementia). The decision of Alice Grant, who moved with her husband into the dementia care unit at the high-fee, majority White-resident ALF, illustrated the push of caregiver burden: “Yes, we were living in our own home, and then my husband got Alzheimer’s and I couldn’t take care of him and take care of the house. So we came here to live and he was taken care of.” Five of the nine proactive residents who had moved to AL as a couple had lost their spouses by the time of the study. Couples’ moving decisions were dependent on finding an ALF where they could be together.

Another common push factor involved the elder’s desire to relieve the family care burden, including the stress of worry. Eugene Ellis, who moved from his apartment in Ohio to the upscale African American ALF, explained:

Here, my daughter and my grandkids would not have to give me any daily attention. They could come whenever time would permit them, but during the interval, they would know that I was in a facility where I was secure, where I had attention, where if I had any emergencies, [staff] would respond.

Earl Harper, who had relocated from Detroit to his son’s house after a bout with severe depression, moved to escape the rule of a hired caretaker: “I was home by myself, and I had to have a baby-sitter. She read the Bible to me everyday for about a month, and I told [my son] to get me out of here.” Seven proactive residents moved from another ALF because of dissatisfaction with services.

Although push factors had greater influence, proactive elders, Black and White, also were pulled to relocate. A common pull factor for those moving from out of town was to be near children and grandchildren. For Eugene Ellis, the opportunity to know his great-grandsons was a draw. Elsie Calhoun gained the support of her son. Some elders also believed that AL would enhance their independence and autonomy, principally because of the more supportive physical environment relative to their former homes but also as a means of avoiding the restrictions of living with a child. Helen Brooks said, “I decided that the best thing for me to do was to find some-where to go where I would still have my independence, would not have to live with either my daughter-in-law or my other son.” Other pull factors included the availability of help in case of a medical emergency, location in or near a former community, and increased social interaction.

As noted, the majority of the proactive residents moved to the high-fee homes. These moves were likely inevitable, but they probably happened sooner because these elders had the resources to consider an ALF with desirable features. In the case of Eugene Ellis, his declining health and increasing care burden on his daughter pushed him to leave his apartment and home town; but without access to an ALF where he could find high-quality care among residents who shared his African American heritage and professional status, his decision would have been delayed, likely shifting him to the compliant category. For him, both race and class were decision-making factors. Earl Harper and Helen Brooks were also pulled by the racial and cultural commonality of this home, described by Helen as “the first of this type available to Blacks,” as well as by certain amenities. At the time of Study 2, it was one of the few large high-end ALFs in the Southeast, if not the nation, with a majority (in this case, exclusively) Black-resident population.

Lucille May, a low-income proactive White resident, moved from a nursing home to the small rural mountain home. Lucille’s unhappiness in the nursing home exerted push, but her past life had been hard, and she was pulled by the environment of her chosen ALF, depicted by her as the nicest place she had ever lived. Lucille’s access to the Medicaid-waiver program made relocation possible.

Although proactive elders initiated the moving process and made their decisions independently, with few exceptions they needed support from others (mostly, family members) to execute the moves. This support included help finding a facility, selling a home, packing possessions, and moving.

Compliant residents

The factors influencing decisions made for compliant elders were similar to those contributing to proactive decisions—worry about the health, safety, and well-being of the older person and the care burden of family or friends. Almost half (49%) of compliant residents lived alone in their own homes, and many had not contemplated moving. Some family members had little involvement in day-to-day care, and about one fourth lived in a different city, thereby heightening concern. Typical was Irene Blanton, White and age 84, who moved from Las Vegas to her high-fee home. She described her compliance:

My children didn’t want me there alone. My husband had died, and they used to worry about me all the time, and I used to have somebody take care of me for 4 hours a day. They didn’t think it was right that they had to be so far away. They’re busy kids, you know. I love my children so I’m just happy to be close to them. And I knew that since I was gonna have a great big operation, a knee replacement, that I would have the support.

When compliant elders lived alone in the same town as caregivers (38%), in many cases family members had assumed significant caregiving roles.

For eight compliant elders living alone, a health crisis led to the decision to relocate. These included five residents of one low-fee home who moved directly from the hospital to the one available ALF. For four of the five, social workers were the primary decision makers. None had supportive family members or the financial resources to consider other options.

Over a fourth (29%) of compliant residents lived with a family member. Most moved to the low-fee homes, and this group included Black residents and White residents. These family members had done their best to forestall the move to AL, for reasons that included a strong value to care for family, to abide by the elder’s wishes, and to save money. Most caregivers were employed, and the older person was left at home alone for long periods, creating situations like those of elders living in their own homes. An example was Bill Thomas, who moved from Alabama to live with a son after a stroke. Bill’s son worried about falls, poor nutrition, and his father’s lack of social outlets, and when he found the small Black-owned ALF near his own home, Bill agreed to move. Unlike Irene Blanton, neither Bill nor his son had sufficient financial resources to hire a paid caregiver. Bill’s access to the Medicaid-waiver program facilitated relocation.

Fifteen compliant residents (18%) moved from an independent senior residence or another ALF. Four residents of one low-fee home moved because their former ALF was closed by state regulators because of a history of poor care.

Most residents in the compliant group, like Irene and Bill, acquiesced gracefully, not wanting to buck or burden their caregivers, and some admitted that the extra support was welcome. All realized that their best and likely only strategy was compliance.

Passive/resistant residents

A third of residents had no control over the relocation decision. This group included residents who moved to all types of homes. Half lived in their own homes and apartments, most alone but a few with a spouse. Resistant residents had a strong value for remaining at home and so disputed the need to relocate. One was Elaine Pruitt, whose only child, Ben, lived in Los Angeles. Elaine believed that she managed fairly well at home with some paid help, but Ben worried about her safety. One weekend when in town, “out of the clear blue sky,” he took her to her high-fee ALF for “a visit,” and there she stayed. Elaine felt “put” into the home by her son—and in actuality, she was. A year later, she had moved none of her furniture and few of her clothes and continued to talk about returning home. “Home” represented Elaine’s identity and her life: “I don’t care what anyone says. Our little treasures that we have are in our homes; whatever the treasures are, they are in your home.” Elaine, one of the first African Americans to work for one of the utility companies, was proud of her accomplishments and her home.

Similar examples of resistance were found among affluent White elders and among less affluent elders of both races. Although in Elaine’s case, crisis lay dormant, in most cases family intervention was triggered by one critical event or a situation of mounting concern, often involving dementia. The words of Elizabeth Dodd’s son illustrate how decisions are made in such circumstances:

There was one climactic event, but there were also a series of smaller events that caused concern but didn’t push me over the edge. The event that finally pushed it over the edge is that my mother took off, either one night or one morning—we still don’t know which—in her car by herself and drove to Chattanooga. The Chattanooga police called her sisters and called me, asking what to do with this lady. We went into a complete panic. As soon as possible after that, she moved into the ALF. That incident made all of us realize that we can no longer allow her to do what she wants to do. She did not want [to move], and we let her control the circumstances until that event happened, and then we said, “We don’t care what she wants; she is going.”

Most family members appreciated the significance of usurping the older person’s autonomy and were loath to assume such drastic roles. One daughter explained her reaction:

When you put your parents in assisted living, you really are removing everything that they used to have control over—their car, their money, the place that they lived, their freedom. It is difficult for the people involved, like mother and dad, and it is difficult for whoever is acting as their caretaker because you don’t know [what to do].

Passive elders were unable to participate in decision making. Alice Hughes, who had an emotional breakdown after her husband’s death, illustrates one pattern involving such critical events:

When my husband died, I was wiped out. My kids didn’t want me by myself, so I went to their house for 2 or 3 days, waiting for them to put me in here. Sometimes I am sorry that I didn’t stick it out, but I didn’t have any choice. You bring up your children, and then they get older and they are the boss and you do what they tell you to. I am not unhappy here. It is just that everything is gone. I had beautiful furniture, and that is gone. All I have is the clothes on my back and these little pieces. It didn’t matter where I went. I was in a state of shock.

Individuals in the passive/resistant group included those whose family members had gone to great lengths to keep them at home—typically beyond the point of their ability to resist and their families’ endurance. Examples include Harold Stamps, whose daughter, sister, and niece rotated shifts in his home to provide round-the-clock care, and Tom Branch, whose daughter drove weekly to Alabama for over 2 years trying to keep him at home with paid caregivers. The divergent family support strategies and ALF choices (low-fee for Harold and high-fee for Tom) reflect the class differences of these two elders, both African American.

Class differences also were evident in the decision-making patterns of passive and resistant residents who were moving from a caregiver’s home—one fourth of residents in this group. All moved to the low-fee homes, and most interim moves were intended to be permanent. Maude Evans, an African American from south Georgia, moved into the home of her only child, Janine, when Maude’s behavior became erratic because of the progression of Alzheimer’s. Janine, who was married with a teenage daughter and two adult sons at home, worked full-time and had hoped to keep her mother with her until she could retire. Balancing work and multiple caregiving roles became too stressful, though, particularly after Maude became incontinent. Ultimately, Maude’s physician convinced Janine that her only choice was to place her mother in AL. Maude, a retired teacher, could pay the upper fee ($1,500) of her small ALF, but she could not afford the high-fee homes that Janine would have preferred. Janine and her husband, who were saving for their children’s college education, were unable to supplement Maude’s income. Similar scenarios played out for other residents of the low-fee homes: Had elders or their family members had resources that would have enabled them to move to high-fee homes, these decisions might have come sooner.

For five residents in the passive/resistant group, decisions were made by healthier proactive spouses who were the primary caregivers. In two passive/resistant cases, social workers were the decision makers. In one case (passive), a 79-year-old woman with dementia had been living with her mentally disabled nephew and had been having her meals “catered” from nearby fast-food restaurants. In the other case (resistant), the elder was being physically abused by her sister.

The value to care

Like Janine described above, many caregivers who controlled decision making for compliant and resistant elders experienced uncertainty and guilt. The words of another daughter, also African American, were typical of these decision makers: “I felt really guilty. I promised her I would never put her in a home, that I would let her stay home. I tried, and it just didn’t work.” Among the family members whom we interviewed, African Americans exerted greater effort to forestall moves; but our findings also indicate that the cultural value to care for elders at home is not universal and seems to be eroding with upward mobility. The words of one daughter express this clash of values:

Well, I’m just saying that, as a whole, it is not something that the African American people do. It’s usually, families stay with family.… I would like to even be able to spend more time with her, but I’m almost, well, 50.… I was a housewife and mother, the whole time that my children were growing up, and now that they are on their own—they completed college and all of that—I just feel like my life is opening.

Another African American daughter, one whose mother threatened to “chain herself” to her house if forced to move, rationalized her decision similarly: “I don’t feel like children should give up the days that God give them to devote to an elder parent like that.”

Finding a “proper” home helped to alleviate guilt in some cases, as expressed by this daughter, who moved her mother only after finding a small Black-owned home with the required private room:

I think it is a good solution for African Americans because it helps relieve you of that burden of guilt because you’re putting your loved one outside of your home, but it also allows you to have some comfort level because you know they’re being cared for, if it’s the proper home.

The majority of family members, Black and White, put forth significant effort to find what they considered to be the best home that their resources would allow, which in some cases was the deciding factor in their moving decisions.

Discussion

In this article, we explore the process of decision making surrounding moves to AL. We use in this analysis data from a diverse group of individuals who moved to a diverse group of ALFs. Although we identify multiple factors that influence why, how, when, and by whom decisions are made, our focus here is on the effect of race and class on decision making.

We found that control of decision making by both Black elders and White elders was on a continuum ranging from primary to none, and we identified three types of residents based on their decision-making roles: proactive, compliant, and passive/resistant. Proactive residents were the only ones with primary control, and less than a quarter of residents fit this category.

The push–pull model of relocation decisions (see Reed, Cook, Sullivan, & Burridge, 2003) is applicable to the decision making of proactive elders in our study. Push factors largely related to residents’ increasing disability, which impeded their ability to manage in their current environments. Lawton and Nahemow’s well-known ecological model of aging (1973) considers the interaction between the demands of the physical and social environment, or environmental press, and an individual’s personal abilities, or competencies. The model proposes that older persons better adapt to the problems of aging when they maintain a congruence between competence and press. These proactive elders chose to move to AL because they no longer could maintain that steady state in their own homes and neighborhoods. They sought in AL a better fit between their needs and the demands of the environment (Lawton, 1980; Moos & Lemke, 1994). AL research has shown that resident–facility fit is a key factor in residents’ ability to age in place in AL (Ball, Perkins, Whittington, Connell, et al., 2004; Bernard, Zimmerman, & Eckert, 2001).

Family care burden also influenced proactive moves. Black and White proactive elders actively sought to reduce the burden on their caregivers, as well as their own costs associated with this dependency. Studies of elders who are moving to retirement housing (Young, 1998), continuing care retirement communities (Groger & Kinney, 2006), and nursing homes (Groger, 1994) similarly found that reducing caregiver burden was a factor in elder decision making.

Reducing care burden also figured prominently in decisions made by caregivers. Relatives (usually children) of Black and White elders in high-and low-fee homes delayed moves to AL by providing significant support both in their own home and the elder’s. Class differences, though, were evident in the configuration of their caregiving roles. In almost all cases, it was residents, both Black and White, of the low-fee homes who had initially relocated to the homes of family members. With higher-income elders and family members, strategies typically entailed the use of paid caregivers and earlier moves to AL. Included among those caregivers who were experiencing burden were proactive spouses caring for less-abled partners. These couples were similar to the asynchronous spouses described by Kemp (2008) in her study of couples’ transitions to AL.

Although multiple factors pushed proactive elders to take action, pull factors, made available by “class,” also had influence. With few exceptions, proactive elders moved to the high-fee homes. Our findings indicate that the higher socioeconomic status of these residents granted them access to ALFs with features that they found attractive, thereby influencing them to take action while still able. More affluent elders were able to relocate from distant cities (to be near supportive family members) and to seek out ALFs in comparable communities and even former neighborhoods. Findings further suggest that race was also a decision-making factor for more affluent elders who sought racial and cultural commonality in the high-end ALF targeted to African Americans. Here, Cutchin’s conceptual model of geographic pragmatism (2003) is instructive. This model differs from the ecological model of aging (Lawton & Nahemow, 1973) in the greater emphasis placed on the social nature of thought and action. Hence, the pull factors for these elders include the ability to live in an AL setting where they might find creative ways of experiencing meaning in addition to having their care needs met.

When considering decisions to move to AL, some parallels can be made to migration decision making. Following a life course model, motivations for later-life migrations have been conceived of as amenity driven, assistance, or institutional moves (Litwak & Longino, 1987; Longino & Bradley, 2006). Our findings indicate that even for proactive elders, moves to AL cannot be considered amenity driven. A number of elders, though, had made assistance moves to the homes of family members before moving to AL. In addition, many of the moves from out of town directly to AL seem to fit the assistance, rather than institutional, category. Most of these elders relocated from their own locales because of the support that they expected to receive from family members who, before the elder’s move, had minor care roles. AL research has shown that, more so than in nursing homes, many AL residents continue to receive considerable support from family members (Ball, Perkins, Whittington, Connell, et al., 2004; Ball, Perkins, Whittington, et al., 2005; Kemp, 2008).

Our findings clearly show that control of decision making for elders who are moving to AL is influenced by class, though not directly by race. Among the elders whom we studied, the impact of class primarily related to the AL placement options and to the strategies available to forestall moves. Factors influencing moving decisions were similar for both Black and White elders of comparable socioeconomic status. A common argument for African Americans’ lower use of long-term care compared to that of Whites is that African Americans have stronger family ties and a tradition of caring for elders at home (Belgrave, Wykle, & Choi, 1993). Research comparing informal support of Blacks and Whites, however, has indicated a lack of accord on this issue. Although some studies have cited a greater reliance of African American elders on informal support (Jenkins, 2001; Mutran, 1985), others have found that White elders rely more on their families for in-home care (Mutran et al., 2001). Our findings suggest that structural factors have more weight than do those relating to culture in the lower use of AL by African Americans.

The relationship between class and race, though, is undeniable. Poverty rates are higher among elderly African Americans (23%) than among White elders (8%) (Administration on Aging, 2006). The predominance of poverty and women among elderly African Americans and the greater likelihood of older African American women being unmarried and poor all contribute to differences in the health status of Blacks and Whites (Administration on Aging, 2006; Jackson & Perry, 1989). Compared to White elders, Black elders are in poorer health (Belgrave & Bradsher, 1994; Binstock, 1999) and have higher rates of chronic disease and disability (Manton & Gu, 2001; Wallace et al., 1998). Socioeconomic status contributes to the vulnerability of older African Americans. Disability rates are higher among less educated older persons (Manton & Gu, 2001), and disease and impairment are more common among those with low incomes (Jackson & Perry, 1989; Markides, 1989). This variation by race in the impact of aging—often referred to as the double jeopardy of being old and Black—is primarily the result of the experience of racism and cumulative disadvantage throughout the life course and into old age (Jackson, 1985; Markides, 1983).

Our findings have important implications for AL policy and practice. Numerous studies have documented the connection between control of decision making and adjustment and satisfaction with long-term care moves (Ball et al., 2005; Forbes & Hoffart, 1998; Mikhail, 1992; Rossen & Knafl, 2003; Young, 1998). Elsewhere (Ball et al., 2005), we have made recommendations for how facilities, family members, and elders themselves can improve elder control in moves to AL. These include ways to involve elders in decision making at each stage of the moving process (Young, 1998). The findings reported here indicate that our recommendations are not as useful for low- and moderate-income elders who are seeking AL care, because of the significant barriers to control exerted by structural factors. Although these barriers are clearly present for White as well as Black elders, the greater relative disadvantage of elderly African Americans points to a particular problem for this group.

Despite the fact that African American elders do not utilize AL to the same extent as do White elders, the need for AL care for this population will no doubt grow in the future. The number of older African Americans (8% of the 65-and-older population in 2000) is expected to increase by 147% over the next 30 years, compared to 74% for older Whites (Administration on Aging, 2006). Other minority groups are experiencing similar growth in the U.S. (Administration on Aging, 2006).

Our findings indicate that Medicaid funding increases poor and minority elders’ access to AL because it enhances their control over decision making. Although the majority of states have Medicaid coverage of AL services, most states limit the number of persons served by waiver programs, and only 14 states cover services under the Medicaid state plan (Mollica & Johnson-Lamarche, 2005). Coverage under state plans cannot be capped, which greatly increases numbers of beneficiaries. Waiver participation is rising in many states, including Georgia, which covered 3,455 AL residents in 2006, up from 2,851 in 2004 (Georgia Community Care Services Program, 2006). Only three states—Colorado, Oregon, and Washington—have shifted resources from nursing homes to AL (Mollica & Johnson-Lamarche, 2005). Another possible funding source for AL services for disadvantaged individuals is the U.S. Department of Veterans Affairs, which now is providing this level of care on a pilot basis (Hedrick et al., 2007). Support from such programs also increases the likelihood that small low-income homes will survive to continue to serve low-income and minority populations (Ball et al., 2005; Perkins, Ball, Whittington, & Combs, 2004).

Conclusion

AL is a relatively new long-term care arrangement, and many elders and families are just now becoming aware of it as a middle-ground alternative to home care and nursing homes. Their elder care decisions will continue to be fraught with conflicting needs and emotions, but AL may offer a solution to some of these elders and their families. Our findings suggest that income level is an important influence on the timing of decision making, the type of facility chosen, and the living and care arrangements preceding the decision to move to AL but that race makes very little difference for persons of similar social class. Although we acknowledge the greater likelihood that African American elders will have limited income—and choices—in their later life, those in the middle class appear to experience a decision-making process similar to that of their White counterparts. AL providers and family long-term care advisors should be alert to the greater relative importance of social class over race as an influence on the decision to move to AL.

Biographies

Mary M. Ball, PhD, is an associate research professor at the Gerontology Institute at Georgia State University. She is primarily a qualitative researcher. Her research for the past 11 years has focused on assisted living. Earlier work investigated residents’ quality of life, with an emphasis on issues of independence and autonomy and the experience of minority elders. More recent research has addressed job satisfaction and retention of direct care workers in assisted living. She is a fellow of the Gerontological Society of America.

Molly M. Perkins, PhD, is a research associate at the Rollins School of Public Health at Emory University in Atlanta, Georgia. She has been a co-investigator on two projects funded by the National Institute of Aging that have focused on workforce issues in assisted living. She is currently a project director of a study funded through the National Institute of Drug Abuse, investigating cognitive processes that current smokers employ when contemplating quitting and how they define dependence, tolerance, and relapse. Dr. Perkins is a health disparities scholar of the National Center on Minority Health and Health Disparities. Her research interests include minority aging, health disparities, substance abuse, and sexual risk behaviors.

Carole Hollingsworth, MA, is research coordinator for the Gerontology Institute. She has participated in qualitative research projects focusing on assisted living for the past 8 years. She has co-authored numerous scientific journal articles and one book, and she is co-editor of a forthcoming book.

Frank J. Whittington, PhD, is Associate Dean for Academic Affairs at George Mason University. Dr. Whittington received his doctorate from Duke University, and he is a fellow of the Gerontological Society of America. His publications include 8 books and over 50 articles and chapters on long-term care and health behavior of older people. His most recent book, coauthored with Mary Ball and four Georgia State University colleagues, is entitled Communities of Care: Assisted Living for African American Elders.

Sharon V. King, PhD, is an assistant research professor in the Gerontology Institute at Georgia State University. Her research focuses on relationships among multigenerational African American families, health disparities among ethnic elders, and grandparenting in African American, Chinese, and East African families. She is a minority fellow with the Gerontological Society of America and the 2005 winner of the David A. Peterson Award for her article about gerontology education in Kenya, in Gerontology and Geriatrics Education, the official journal of the Association of Gerontology in Higher Education.