Document View

Other available formats:

Care Arrangement Decisions for Frail = Older=20 Women: Quantitative and Qualitative Perspectives; [1]

Jenkins, Carol L.. Journal=20 of Women & Aging. New=20 York: 2000.=20 Vol. 12, Iss. 3/4; pg.=20 3

Abstract (Summary)

This study uses both quantitative and = qualitative=20 methods to investigate how a range of care arrangement decisions for = frail older=20 unmarried women are made. Quantitative data from the 1989 National = Long-Term=20 Care Survey provides information concerning factors that predict the = probability=20 of five categories of care arrangements, including self only care, = nursing home=20 care, informal only care, formal only care, and a mix of formal and = informal=20 care. Qualitative interview data provides information on what occurs = during care=20 arrangement decision-making processes. Results are combined to explain = the=20 choice of care arrangements for a small sample of chronically disabled = older=20 unmarried women. Results show that need factors, such as age and = disability, are=20 strong predictors of the need for assistance. Family members played a = central=20 role in determining care arrangements and often helped an older woman to = avoid=20 an unwanted care arrangement. The use of a broad measure of impairment = resulted=20 in high levels of disability for the sample participants. Both = disability status=20 and care arrangements were transitory in nature.

=BB Jump=20 to indexing (document details)

Full=20 Text (6788 =20 words)

ABSTRACT. This study uses both quantitative = and=20 qualitative methods to investigate how a range of care arrangement = decisions for=20 frail older unmarried women are made. Quantitative data from the 1989 = National=20 Long-Term Care Survey provides information concerning factors that = predict the=20 probability of five categories of care arrangements, including self only = care,=20 nursing home care, informal only care, formal only care, and a mix of = formal and=20 informal care. Qualitative interview data provides information on what = occurs=20 during care arrangement decision-making processes. Results are combined = to=20 explain the choice of care arrangements for a small sample of = chronically=20 disabled older unmarried women. Results show that need factors, such as = age and=20 disability, are strong predictors of the need for assistance. Family = members=20 played a central role in determining care arrangements and often helped = an older=20 woman to avoid an unwanted care arrangement. The use of a broad measure = of=20 impairment resulted in high levels of disability for the sample = participants.=20 Both disability status and care arrangements were transitory in = nature.

KEYWORDS. Disability, informal care, formal = care,=20 nursing home, caregiving, decision making

Demographic projections indicate that one = out of=20 about every five persons will be over 65 years of age by 2020; nearly 15 = percent=20 of these older individuals will be at least 85 years old (Aging America, = 1991).=20 Older persons' higher likelihood of ill health and disability often = results in=20 their need for help with activities of daily living (ADLs), such as = bathing and=20 dressing, or with instrumental activities of daily living (IADLs), such = as=20 housework and shopping. While recent research indicates a decline in the = prevalence of disability (Crimmins, Saito, & Reynolds, 1997; Manton, = Stallard, & Corder, 1995), growth in the absolute numbers of older = persons=20 is likely to increase the aggregate need for long-term care. This need, = when=20 met, will be done so by family members and friends, by the formal = service=20 provision system, or by some combination of both. Family members, = particularly=20 wives and adult daughters, provide the bulk of community care to frail = older=20 persons. They often do so in spite of other demands on their time and = energy=20 such as their own physical frailty, other family responsibilities, and = labor=20 market participation.

Medicare and Medicaid are the primary = payers for home=20 health and nursing home care; these two types of formal care together = accounted=20 for about 12 percent of all national health care expenditures in 1992 = (National=20 Association of Home Care, 1993). Continuing growth in expenditures since = that=20 time is of concern. Policymakers face the challenge of formulating = public=20 policies that continue to provide effective long-term care services = while=20 maintaining a balanced budget. At the same time, recent budget = initiatives=20 (e.g., tax credits for family caregivers) show some recognition that the = integral role families play in providing care needs to be recognized and = supported, particularly given recent changes in family structure that = can place=20 added burdens on family caregivers (e.g., increased numbers of single = parent=20 families, women's increased participation in the labor force). It is = important,=20 therefore, to understand the way in which older persons and their = families make=20 decisions about care arrangements. This study uses a combination of = quantitative=20 and qualitative methods to address the issue of care arrangement = choices.

BACKGROUND

This research is based on health service = utilization=20 and decision making frameworks developed by Andersen (1968), Janis and = Mann=20 (1977), and Gonyea (1987). Within this framework, easily measured = quantitative=20 variables (e.g., age, gender) combine with less easily measured = qualitative=20 variables (e.g., family relationships, attitudes about caregiving) to = form the=20 context of the decision-making process. Predisposing, or risk, factors, = such as=20 age, race, and gender, frame cultural and social conditions that = influence one's=20 personal preferences about, and opportunities for other factors such as=20 education, employment, marriage, and children. These factors in turn = help to=20 determine enabling factors, such as income and the presence of children, = and=20 need factors, such as health status. They also affect the way in which = family=20 members relate to one another and the world at large, and the attitudes = that=20 older parents and adult children have about giving and receiving = assistance.=20 Finally, they help to determine the need for and ability to access = formal=20 long-term care services.

Previous research focused on quantitative = factors=20 related to care arrangements shows that need factors, particularly older = ages=20 and the presence of disability, are associated with all types of care=20 arrangements, including informal, formal, and nursing home care (Kemper, = 1992;=20 Logan & Spitze, 1994; Tennstedt, Sullivan, McKinlay, & = D'Agostino, 1990;=20 Wallace, Levy-Storms, Kington, & Andersen, 1998; Wolinsky, Callahan, = Fitzgerald, & Johnson, 1992). The presence of familial and financial = resources enables older persons to have a range of options from which to = choose=20 a care arrangement. Unmarried persons, particularly women, are more = likely to=20 use formal services (Chappell, 1985; Logan & Spitze, 1994; Soldo, = Wolf,=20 & Agree, 1990). Merely having a child decreases the likelihood of = using=20 formal care and increases the likelihood of informal care (Kemper, = 1992).=20 Children's gender is important as well: the presence of daughters = increases the=20 likelihood of informal help (Soldo et al., 1990). Higher incomes = increase the=20 likelihood of using formal community care (Bass & Noelker, 1987; = Kemper,=20 1992; Logan & Spitze, 1994), and appear to be linked to a higher = likelihood=20 of nursing home care as well (Kelman & Thomas, 1990). Predisposing = factors=20 also play an important role in determining care arrangements. The = prevalence of=20 disability is substantially higher among older women than older men, = thus=20 increasing their need for some type of assistance (Crimmins, Hayward, = &=20 Saito, 1996). As a result, they have higher probabilities of using = informal,=20 formal, and nursing home care (Kelman & Thomas, 1990; Stone, = Cafferata,=20 & Sangl, 1987). Results on the effects of race or ethnicity on care=20 arrangements are mixed. Some studies show no race effects (Miller, = McFall, &=20 Campbell, 1994), while others show that minorities are less likely than = Whites=20 to use nursing home care (Falcone & Broyles, 1994; Salive, Collins, = Foley,=20 & George, 1993). The use of formal community care appears to be = associated=20 with minority status (Bass & Noelker, 1987; Miner, 1995) although it = is=20 unclear whether this result is due solely to racial or ethnic factors, = or is=20 confounded by socioeconomic status.

Studies of the qualitative aspects related = to care=20 arrangement decisions show that older individuals desire to participate = in=20 decisions relating to their care, even when they are not able to obtain = the care=20 arrangement they prefer (Wetle, Levkoff, Cwikel, & Rosen, 1988). = Active=20 participation in the process leaves them more satisfied with both the = decision=20 process and with the outcome (Coulton et al., 1989). At the same time, = they=20 prefer that family members, especially spouses and children, participate = in the=20 process as well. The inclusion of family members gives older persons a = sense of=20 support, and strengthens their feelings of control over the decision = (Prohaska=20 & Glasser, 1996). Family members provide the bulk of community care = for=20 older persons. As a result, they have a fundamental interest in = participating in=20 the care decision process. Many adult children, particularly daughters, = have=20 strong affective ties to parents that motivate them to provide = assistance to=20 frail parents (Horowitz & Shindelman, 1983). Other children provide=20 assistance to a parent out of a sense of duty; this is particularly so = when=20 dysfunctional relationships between children and parents have resulted = in a lack=20 of filial affection (Whitbeck, Hoyt, & Huck, 1994). Children = sometimes make=20 decisions related to caregiving without including the parent in the = process. For=20 example, Matthews and Rosner (1988) find that adult children often = determine=20 among themselves how responsibility for parent care will be divided. A = child's=20 birth order, or the perception that he or she is favored by the parent = over=20 other children, causes some siblings to assign primary responsibility = for the=20 parent to a specific child. In these cases, siblings provide only backup = or=20 sporadic assistance to the parent. Wolf, Freedman, and Soldo (1997) find = that a=20 child's hours of parent care decrease as those of siblings increase, = although on=20 less than a one-for-one basis. This indicates that children's decisions = about=20 providing care are made simultaneously and that children take the = choices of=20 other family members into account during the decision-making process.=20 Professional service providers are introduced into the decision process = when=20 formal care is needed. Guided by well established professional norms, = these=20 individuals are focused on protecting their clients's best interests, = which=20 includes safeguarding their decision-making autonomy, independence, and = safety=20 (Hennessy, 1989; High, 1988; Wetle, 1995). These issues take on added = importance=20 when surrogate decision-making is necessary, as is often the case when = an older=20 person's cognitive function has deteriorated (High & Rowles, = 1995).

The studies reviewed here make significant=20 contributions to our understanding of how care arrangements are chosen. = They=20 share some limitations, however, that constrain our ability to gain a = deeper=20 understanding of care arrangement decisions. In general, they have been=20 restricted to studying the choice of one specific type of care = arrangement such=20 as the choice of who will provide care (Matthews & Rosner, 1988; = Tennstedt=20 & McKinlay, 1989), the choice between informal and formal care (Bass = &=20 Noelker, 1987; Logan & Spitze, 1994), or the choice about entering a = nursing=20 home (Gonyea, 1987; McAuley & Travis, 1997). This limits our = potential to=20 identify significant predictors of a wide range of care arrangements, as = well as=20 themes that may be common across decisions related to different types of = care=20 arrangements. Qualitative studies of care decisions are based on the = perspective=20 of one, or at most two, of the decision participants. Reliance on only = one=20 participant's perspective of what occurs during the decision process = results in=20 an incomplete picture of care arrangement decision making. It can also = result in=20 missing the different perceptions of the process on the part of = participants=20 (Pratt, Jones, Shin, & Walker, 1989). This study addresses these = limitations=20 by using both quantitative and qualitative techniques to investigate a = full=20 range of care arrangements, while including the perspectives of the = various=20 participants in the decision-making process.

METHODOLOGY

Methods

The study presented here is the final part = of a=20 larger project studying the care arrangement choices of frail older = persons. A=20 wide array of care arrangements are investigated, including self only = care,=20 nursing home care, informal only care, formal only care, and a mix of = informal=20 and formal care. The first part of the project consisted of a = quantitative study=20 in which multinomial logit techniques (Liao, 1994) were used to examine = the=20 likelihood of the five different types of care arrangements, controlling = for the=20 various predisposing, enabling, and need factors of a nationally = representative=20 sample (n =3D 3,180) of frail older persons. In the second part of the = project,=20 qualitative data about the decision-making process were gathered during = a series=20 of in-depth interviews held with a small sample (n =3D 26) of older = women, family=20 members, and service providers. Content and thematic analysis techniques = were=20 employed to identify major themes associated with the decision = participants'=20 perceptions of how each care arrangement decision was made (Luborsky, = 1994). A=20 full description of the project, including the conceptual framework, as = well as=20 methods and results from the initial two stages, are provided in Jenkins = (1997).=20 The focus here is on the third stage of the project.

In this final stage of the project = quantitative data=20 obtained in earlier stages are employed to calculate the probability of = each of=20 the five care arrangements for the sample of older women who = participated in the=20 second stage of the project. The predicted probabilities are compared = with each=20 woman's actual care arrangement. Qualitative data gathered during the = in-depth=20 interviews is then used to discuss why some women's actual care = arrangements=20 differ from those predicted as most likely -- in essence, to explain the = way in=20 which observable, quantitatively measurable factors and unobservable, = less=20 easily measured qualitative factors combine to determine care = arrangements.

Sample

The sample who participated in the = interviews=20 consists of 26 individuals: eleven women, nine of their family members, = five=20 case managers, and one nursing home intake manager. Demographic, = disability=20 status, and care arrangement information for each of the older women is = provided=20 in Table 1. The women range in age from 65 to 88, with a mean age of 80. = Most of=20 the women are widows; only one has never been married and has no = children. There=20 is a wide range of disability states as measured by the number of ADLs = and IADLs=20 with which the women receive help. Among the community-dwelling women, = the two=20 whose sole strategy is self care are the least disabled, while those = women who=20 receive help from a mix of informal and formal helpers tend to be the = most=20 disabled. IADLs are not measured for nursing home residents since such = tasks are=20 routinely performed by staff members. Of the two women in nursing homes, = one has=20 a moderate level of ADL disability, while the other is severely = disabled.

RESULTS

The predicted probability of all five care=20 arrangements for each woman, as well as her actual care arrangement at = the time=20 of her interview, can be seen in Table 2. The care arrangement predicted = as most=20 likely for each woman, based on results of the logit estimation, is in = boldface=20 type. A discussion of all the cases is beyond the scope of this article; = therefore, case studies, presented in Exhibit 1, are provided for three = of the=20 women whose situations and decision processes are most representative of = the=20 entire sample.

Family members, particularly adult = children, played a=20 central role in the care arrangement decision-making process. While = often=20 participating in the provision of care themselves, they also played an = active=20 role in assisting their mothers to deal with the formal service system. = In Mrs.=20 H.'s case, for example, her daughters actively sought assistance from = the formal=20 care service system in order to avoid nursing home placement for their = mother.=20 Providers, by virtue of their perceived expertise, can hold the balance = of power=20 during the care arrangement decision process, particularly following a=20 hospitalization. Results demonstrate that family members' inclusion in = the=20 actual decision-making process influenced outcomes through the very = concrete=20 means of shifting some of the decision-making power away from service = providers.=20 As a result, nursing home care was avoided by some older women who then = received=20 formal community care, most often in conjunction with informal = assistance. In=20 several of the cases not described, the unavailability of family members = due to=20 geographical distance or personal illness resulted in the older women = being in a=20 care arrangement they did not prefer (e.g., nursing home care, formal = only=20 care). In a related finding, this intervention on the part of family = members=20 appears to relieve the inherent conflict facing professional service = providers=20 as they strive to balance their concerns about protecting both client=20 independence and client safety.

Family members also played a major role in = enabling=20 their frail older relatives to cope with severe states of disability; = Mrs. H.=20 and Mrs. P. serve as examples. Although they experienced difficulty with = most=20 ADLs and IADLs, they were both able to live by themselves with = assistance from=20 their children and, in Mrs. P.'s case, from her brother as well. At the = same=20 time, it should be noted that the definition of disability affected the = way in=20 which study participants were categorized. The disability measure used = for this=20 project as a whole was very broad, counting a person as having an ADL = disability=20 if he or she reported actually receiving help from another person, = needing=20 someone to stand by in case help was needed, or using any type of = special=20 equipment (e.g., tubrails, walker, raised toilet seat). The measure did = not=20 prove to be a reliable predictor for the sample of women who = participated in=20 this final study, however, particularly at the highest levels of = impairment.=20 Results of the logit model estimated in the first part of the project = indicated=20 that nursing home care was the most likely arrangement for older women = with a=20 high level of disability (5-6 ADLs). Of the three women in this study = who were=20 categorized as highly disabled, only one was in a nursing home. The = moderate=20 level of ADL disability (3-4 ADLs) showed similar results: of the five = women=20 categorized as having moderate disability, only two received care from a = mix of=20 helpers (the predicted outcome). It is apparent that factors such as the = availability and willingness of family members to provide informal care = played a=20 major role in the variance between predicted and actual results.

Finally, this study makes it clear that = both=20 disability status and care arrangements are transitory in nature, making = the=20 results obtained through the use of cross-sectional data extremely = sensitive to=20 the timing of data collection. Disability status was measured by the = number of=20 ADLs and IADLs with which a woman was having difficulty at the time of = her=20 interview. As a result, most of the assistance provided by other persons = in this=20 study was for IADL impairments, while ADL assistance was related = primarily to=20 equipment use. The fundamental reason for this result was the timing of = the=20 interviews. In the months preceding the study, four of the women had = been=20 receiving either informal or formal assistance with ADL impairments = following=20 discharge from hospital stays. As each woman's health and functional = status=20 improved, her need for care provided by other persons decreased until it = was=20 eventually discontinued.

Care arrangements had also changed, often = repeatedly,=20 for several of the women. Mrs. C., for example, was receiving only = informal care=20 at the time of her interview. She had been receiving homemaker = assistance, in=20 conjunction with help from her daughter, only a month before her = interview. Mrs.=20 R. and Mrs. M., both able to care for themselves at the time of their=20 interviews, had been receiving informal and a mix of care respectively = in the=20 weeks preceding their interviews. Table 2 shows that a mix of care is = the=20 predicted arrangement for Mrs. C. and informal care is the one most = likely for=20 Mrs. R. It is clear that a change in the timing of the interviews may = have=20 resulted in less variation between predicted and actual outcomes, = primarily due=20 to the transitory nature of many of the women's care arrangements.

Discussion

This study employs results from two = methodologically=20 diverse research models to investigate older persons' care arrangement = choices.=20 Three major results emerged. First, families play an integral role both = in=20 deciding on a care arrangement and in actually providing assistance to = their=20 frail older relatives. The children in this study actively participated = with=20 service providers in arranging for formal community care, often in = conjunction=20 with their own assistance. This lends support to the notion suggested by = other=20 researchers that children often serve as a bridge to the formal = community care=20 service system (Bass & Noelker, 1987; Logan & Spitze, 1994). It = is well=20 known that families provide the major share of community-based long-term = care=20 (e.g., Stone et al., 1987; Soldo et al., 1990); results of this study = confirm=20 that fact. At the same time, changes in family structure in recent = years, such=20 as the increases in single-parent families and women's increased = participation=20 in the labor market, are placing competing demands on family caregivers' = time=20 and energy (Ahlburg & DeVita, 1992). This suggests that policymakers = may=20 want to consider supporting programs that bolster families in their = efforts to=20 provide care to older relatives.

Second, the broad definition assigned to = disability=20 resulted in higher levels of disability for the study participants than = would=20 have been the case if a narrower measure were used. The measurement of=20 disability is an issue of great interest to academics and policymakers = alike.=20 The number of ADL impairments is a widely accepted measure of both = disability=20 and the concurrent need for assistance. It is often employed as part of = the=20 eligibility criteria for long-term care services, such as home health = care, or=20 as a ranking factor to assist case managers in making service allocation = decisions (Wiener, Illston, & Hanley, 1994). As such, it affects = both the=20 number of persons who might use formal long-term care services and = program costs=20 (Jackson, Burwell, Clark, & Harahan, 1992). The debate surrounding = the=20 definition of disability revolves around whether it should include = measures of=20 assistance, or whether it should measure any difficulty persons have = with=20 performing normal activities regardless of assistance (e.g., Verbrugge, = 1990).=20 To a lesser extent, it is also concerned with whether to measure only = personal=20 assistance, or whether to include equipment assistance. Six of the nine=20 community-dwelling women in this study reported ADL difficulties in the = moderate=20 (3-4 ADLs) to severe (5-6 ADLs) range. Five of these women also reported = that=20 they received no personal assistance with ADL tasks, instead coping = through the=20 use of special equipment (e.g., walker, raised toilet seat, bedside = commode).=20 This suggests that the inclusion of equipment assistance in any measure = of=20 disability is likely to greatly increase the prevalence of disability, = and, when=20 used to determine eligibility for programs, to also greatly increase the = number=20 of persons who could participate, as well as program costs. At the same = time,=20 some scholars have recommended that the measurement of disability should = be=20 related to health status, arguing that "disability is better measured by = functioning deficits due to health than by the use of assistance in the = older=20 population" (Crimmins et al., 1997: S70). Any discussion concerning the = most=20 appropriate use of limited long-term care resources needs to include = these=20 issues.

Finally, this study has clearly identified = the=20 transitory nature of care arrangements and, thereby, one of the = difficulties in=20 accurately predicting such episodes through the use of cross-sectional = data. The=20 inclusion of additional health-related information, such as health = status or=20 previous hospitalization, might enhance the predictive power of future = models.=20 Nine women were using some type of care besides self care; seven of them = had=20 experienced a hospitalization within the months preceding the study. Of = these=20 seven, six were essentially caring for themselves (with occasional = informal=20 assistance) before their hospitalization. That is, their use of care was = directly related to their health episode.

Some limitations of the study should be = mentioned.=20 Due to sample characteristics, generalizability is limited. Results = should be=20 interpreted with caution; at best, they indicate the potential behaviors = of very=20 specific types of older persons. The results remain pertinent to the = study of=20 care arrangements, however, given that a large proportion of the frail = older=20 population is made up of older unmarried women. Agreement between the = results=20 reported here and those of previous studies on care arrangements lend = credence=20 to this discussion, as well.

A second concern is attached to the data, = which were=20 gathered at two different points in time: data from 1989 were used as = the basis=20 for the logit model, while data gathered in 1997 were used to determine = the=20 probabilities for the interview sample. This problem was addressed in = part by=20 deflating the 1997 income and Medicare charges to 1989 dollars. The = study was=20 unable to account for changes in the policy environment that might = affect=20 service decisions made in 1997 relative to those made in 1989, however. = Changes=20 in the interpretation of Medicare home health care eligibility criteria = since=20 1989 have made this service more available, for example. Persons may be=20 receiving home health care today with the same characteristics as = persons for=20 whom it was not available in 1989. This concern is minimized since only = one=20 woman was receiving home health care, and her home health aides were = reimbursed=20 through a Medicaid waiver -- an option that was also available in her = state in=20 1989.

Finally, the simple way in which the mixed = category=20 (i.e., both informal and formal helpers) was identified did not provide = any=20 information about its composition. For example, it might be important to = know if=20 a mixed network consisted of primarily formal or informal helpers, or = which type=20 of helper(s) provided the greatest amount of assistance. Combined with = the=20 cross-sectional nature of the data, this inhibits our ability to study = other=20 critical issues, such as how the composition changes in relation to = changes in=20 the independent variables. For example, with increasing disability, is = there a=20 hierarchical progression from informal only care, to a mix of care, to = formal=20 only care?

This discussion suggests areas for future = research.=20 Qualitative data similar to that obtained in this study could be = gathered from a=20 larger and, importantly, more diverse sample. The research reported in = this=20 paper was focused on a specific type of person (i.e., chronically = disabled,=20 unmarried, older white women) with the purpose of at least partially = controlling=20 for these characteristics so that differences in outcomes would be more = clearly=20 related to the decision process. Studies that control for numerous=20 sociodemographic and economic factors show that the outcomes of care = arrangement=20 decision processes are different for various ethnic and racial groups = (Falcone=20 & Broyles, 1994; Miller, McFall, & Campbell, 1994; Miner, 1995). = We know=20 very little about how the decision process itself varies across = different groups=20 and how potential differences between groups might affect the choice of = care=20 arrangements. This type of information should prove invaluable to = researchers=20 and policymakers alike, particularly given the increasing diversity of = our older=20 population.

Given the transitory nature of caregiving=20 arrangements shown in this study, a next step might be a longitudinal = study of=20 care arrangement choices. This area has been addressed quantitatively. = For=20 example, Peek, Zsembik, and Coward (1997), using data from four waves of = the=20 Longitudinal Study of Aging (1984, 1986, 1988, 1990), show that older = persons in=20 their sample who were receiving care from a mix of formal and informal = helpers=20 at the beginning of a time interval were more likely to move to formal = only care=20 by dropping an informal helper. While quantitative models can address = such=20 topics as the changes in composition of caregiver networks or the risk = of=20 entering/leaving specific care arrangements (e.g., Wolf et al., 1997), = they=20 cannot provide information about how the decision process affects = outcomes. A=20 qualitative longitudinal study of care arrangement transitions, with = interviews=20 occurring at frequent intervals, could provide such information.

This study has used diverse, but = complementary,=20 methods to investigate care arrangement choices. As a result, we have a = more=20 complete picture of how care arrangements evolve. We have also seen that = qualitative data is useful in clarifying some of the unexplained = heterogeneity=20 inherent in certain quantitative models. Considering future projections = for=20 growth in the older population, and concerns about supporting families'=20 caregiving efforts while controlling long-term care expenditures, = information=20 such as that presented here may be useful to policymakers.

EXHIBIT 1. Case Studies

Case 1: Mrs. C.

Mrs. C. is an 85-year-old widow. She lives = alone in a=20 small apartment complex that provides subsidized housing for older = persons. She=20 has difficulty with bathing, getting around inside, and using the = toilet, and=20 uses special equipment (e.g., tubrails, raised toilet seat) to help her = perform=20 these activities. Her daughter assists Mrs. C. to manage her money and = drives=20 her on errands when necessary. Occasionally she also helps with = housework. About=20 a year and a half prior to her interview, Mrs. C. had a hip replacement. = While=20 still hospitalized, she and her daughter met with Mrs. C.'s physician = and the=20 hospital discharge planner to discuss her discharge plans. Everyone was=20 concerned about Mrs. C.'s safety since she lived alone. She insisted on = trying=20 to return home and was able to do so with her daughter's support. She = actually=20 went home for one night, although her daughter stayed with her "just in = case."=20 Mrs. C. realized immediately that she would no longer feel safe alone: = although=20 she had learned to walk quite well again, she was worried about falling = and=20 hurting herself. Her daughter suggested that she move in with her = family, but=20 Mrs. C. did not want to "be on top of them all the time." She decided to = move to=20 a senior housing complex, which is located about 10 miles from her = daughter's=20 home. While expressing some regret that she can no longer be "on my = own," she=20 stated with pride that she can "still make my own decisions."

Case 2: Mrs. H.

Mrs. H., a 77-year-old widow, lives alone = in her home=20 of over 40 years. About a year before participating in the study, she = had a=20 double knee replacement. Shortly after her return home from stays in the = hospital and a rehabilitation facility, she fell and broke her shoulder. = She now=20 experiences difficulty with all ADLs and IADLs and receives a mix of = formal and=20 informal assistance from at least six helpers, including two home health = aides,=20 a physical therapist, a friend, and her three daughters. She is also = monitored=20 24 hours a day through a Personal Emergency Response System -- a = transmitter=20 worn as a bracelet which sends a signal over the telephone lines to a = 24-hour=20 monitoring center whose professional staff can respond appropriately to = a need=20 for emergency help.

While receiving intensive physical therapy = at a=20 rehabilitation facility following her shoulder injury, the facility = discharge=20 planner and Mrs. H.'s physician recommended that she be discharged to a = nursing=20 home. Mrs. H. wanted very much to return home again; her daughters were=20 supportive of this desire even though they knew it would mean a great = deal of=20 involvement in her care on their part. They also realized that they = could not=20 provide all the care that Mrs. H. would need. One of her daughters = contacted the=20 local area agency on aging to discuss the problem with a case manager. = Working=20 in conjunction with personnel at the rehabilitation facility, the case = manager=20 was able to put together a package of home health care services funded = under the=20 state's Medicaid waiver. With the assurance that she would have the = necessary=20 care, Mrs. H. was able to return home.

Case 3: Mrs. P.

Mrs. P. is a 72-year-old widow who also = lives alone.=20 Her home is situated in a rural and sparsely populated area, although = her=20 68-year-old brother lives next door. Mrs. P. was hospitalized for = several months=20 during the year previous to her participation in this study as a result = of an=20 acute health care episode. She has difficulty with all ADLs except for = eating,=20 and also needs help with grocery shopping, laundry, heavy housework, and = meal=20 preparation. Mrs. P. copes with these difficulties through the use of = special=20 equipment (e.g., electric bed, shower seat, bedside commode, grab bars) = and a=20 mix of formal and informal helpers, including her brother, one of her = sons, two=20 friends, and a homemaker aide. While still hospitalized, her physician = and the=20 hospital discharge planner met with Mrs. P., her brother, and her oldest = son to=20 recommend that she enter a nursing home, at least on a temporary basis, = due to=20 her frail condition. Mrs. P. adamantly refused to consider this choice. = Her=20 brother, who lives next door to her, volunteered to be responsible for = her=20 well-being when she was alone. As a result, she was discharged to her = home,=20 initially receiving over 20 hours of home health care each week. As her = health=20 and physical stamina improved over the next few months, her need for = home health=20 care was gradually reduced; she now receives six hours of assistance a = week from=20 a homemaker aide.

TABLE 1. Demographic Data, Disability = Status, and=20 Care Arrangements for the Qualitative Study Sample

Marital Number of Number of

Age status children ADLs(a)

Mrs. R. 81 widow 2 sons 1

6 daughters

Mrs. M. 76 widow 2 sons 1

Mrs. W. 82 widow 1 son 5

stepchildren

Ms. J. 88 never none 3

married

Mrs. C. 85 widow 1 daughter 3

Mrs. B. 86 widow 2 sons 4

Mrs. H. 77 widow 3 daughters 6

Mrs. P. 72 widow 2 sons 5

Mrs. S. 84 widow 1 son 3

2 daughters

Mrs. F. 65 widow 3 sons 4

Mrs. L. 78 widow 1 son 1

1 daughter

Number Care

of IADLs(b) arrangement

Mrs. R. 1 self only

Mrs. M. 1 self only

Mrs. W. na(c) nursing

home

Ms. J. na(c) nursing

home

Mrs. C. 3 informal

only

Mrs. B. 4 informal

only

Mrs. H. 7 mix of

helpers

Mrs. P. 4 mix of

helpers

Mrs. S. 3 mix of

helpers

Mrs. F. 7 mix of

helpers

Mrs. L. 4 formal

only

(a)This is the number of activities of = daily living=20 with which a woman needed help; it includes eating, dressing, bathing, = getting=20 around inside the house, using the toilet, and transferring in and out = of=20 bed.

(b)This is the number of instrumental = activities of=20 daily living with which a woman needed help; it includes housework, = laundry,=20 preparing meals, shopping for groceries, managing money, taking = medicine, and=20 using the telephone.

(c)Data concerning IADLs was not gathered = from the=20 two participants who resided in a nursing home. Since most of the = activities are=20 performed by staff, nursing home residents would always report = disability for=20 these items.

TABLE 2. Care Arrangement Probabilities and = Actual=20 Care Arrangements for Each Sample Woman

Arrangements Mrs. H. Mrs. P. Mrs. L. Mrs. = S. Mrs. F.=20 Mrs.C.

Self Care .005 .005 .164 .021 .091 .017

Nursing Home .477 .521 .006 .135 .044 = .259

Informal Only .231 .220 .474 .345 .382 = .284

Mix .240 .198 .220 .398 .297 .340

Formal Only .048 .055 .136 .101 .187 = .100

ACTUAL MIX MIX FORMAL MIX MIX INFORMAL

ONLY ONLY

Arrangements Mrs.B. Mrs.W. Ms.J. Mrs.R. = Mrs.M.

Self Care .016 .002 .021 .122 .179

Nursing Home .245 .667 .179 .001 .005

Informal Only .286 .127 .236 .545 .471

Mix .354 .171 .338 .257 .206

Formal Only .099 .032 .225 .075 .138

ACTUAL INFORMAL NURSING NURSING SELF = SELF

ONLY HOME HOME CARE CARE

Note: The care arrangement predicted as = most likely=20 for each woman is in boldface type.

REFERENCES

Aging America: Trends and Projections. = (1991). U.S.=20 Senate Special Committee on Aging. Washington, DC: American Association = of=20 Retired Persons.

Ahlburg, D. A., & DeVita, C. J. (1992). = New=20 realities of the American family. Population Bulletin, 47, 2-43.

Andersen, R. 1968. A behavioral model of = families'=20 use of health services. Chicago: Chicago Center for Health = Administration=20 Studies, University of Chicago.

Bass, D. M., & Noelker, L. S. (1987). = The=20 influence of family caregivers on elder's use of in-home services: An = expanded=20 conceptual framework. Journal of Health and Social Behavior, 28, = 184-96.

Chappell, N. L. (1985). Social support and = the=20 receipt of home care services. The Gerontologist, 25, 47-54.

Coulton, C. J., Dunkle, R. E., Haug, M., = Chow, J.,=20 & Vielhaber, D. P. (1989). Locus of control and decision making for=20 posthospital care. The Gerontologist, 29, 627-632.

Crimmins, E. M., Hayward, M. D., & = Saito, Y.=20 (1996). Differentials in active life expectancy in the older population = of the=20 United States. Journal of Gerontology: Social Sciences, 51B, = S111-S120.

Crimmins, E. M., Saito, Y., & Reynolds, = S. L.=20 (1997). Further evidence on recent trends in the prevalence and = incidence of=20 disability among older americans from two sources: The LSOA and the = NHIS.=20 Journal of Gerontology: Social Sciences, 52, S59-S71.

Falcone, D., & Broyles, R. (1994). = Access to=20 long-term care: Race as a barrier. Journal of Health Politics, Policy = and Law,=20 19, 583-595.

Gonyea, J. G. (1987). The family and = dependency:=20 Factors associated with institutional decision-making. The Journal of=20 Gerontological Social Work, 12(1/2), 61-77.

Hennessy, C. H. (1989). Autonomy and risk: = The role=20 of client wishes in community-based long-term care. The Gerontologist, = 29,=20 633-639.

High, D. M. (1988). All in the family: = Extended=20 autonomy and expectations in surrogate health care decision-making. The=20 Gerontologist, 28 (Supplement), 46-51.

High, D. M., & Rowles, G. D. (1995). = Nursing home=20 residents, families, and decision making: Toward an understanding of = progressive=20 surrogacy. Journal of Aging Studies, 9(2), 101-117.

Horowitz, A., & Shindelman, L. W. = (1983).=20 Reciprocity and affection: Past influences on current caregiving. = Journal of=20 Gerontological Social Work, 5(3), 5-20.

Jackson, M. E., Burwell, B., Clark, R. F., = &=20 Harahan, M. (1992). Eligibility for publicly financed home care. = American=20 Journal of Public Health, 82, 853-856.

Janis, I. L., & Mann, L. (1977). = Decision making:=20 A psychological analysis of conflict, choice, and commitment. New York: = Free=20 Press.

Jenkins, C. L. (1997). Care arrangements = for frail=20 older persons. PhD dissertation, The Maxwell School, Syracuse = University.

Kelman, H. R., & Thomas, C. (1990). = Transitions=20 between community and nursing home residence in an urban elderly = population.=20 Journal of Community Health, 15, 105-22.

Kemper, P. (1992). The use of formal and = informal=20 home care by the disabled elderly. Health Services Research, 27, = 421-51.

Liao, T. F. (1994). Interpreting = probability models:=20 Logit, probit, and other generalized linear models. Newbury Park, CA: = Sage=20 Publications.

Logan, J. R., & Spitze, G. (1994). = Informal=20 support and the use of formal services by older americans. Journal of=20 Gerontology: Social Sciences, 49, S25-34.

Luborsky, M. R. (1994). The identification = and=20 analysis of themes and patterns. Pp. 189-210 in Qualitative methods in = aging=20 research, edited by J. F. Gubrium and A. Sankar. Thousand Oaks, CA: = Sage.

Manton, K. G., Stallard, E., & Corder, = L. (1995).=20 Changes in morbidity and chronic disability in the U.S. elderly = population:=20 Evidence from the 1982, 1984, and 1989 national long term care survey. = Journal=20 of Gerontology: Social Sciences, 50, S194-204.

Matthews, S. H., & Rosner, T. T. = (1988). Shared=20 filial responsibility: The family as the primary caregiven.Journal of = Marriage=20 and the Family, 50, 185-95.

McAuley, W. J., & Travis, S. S. (1997). = Positions=20 of influence in the nursing home admission decision. Research on Aging, = 19,=20 26-45.

Miller, B., McFall, S., & Campbell, R. = (1994).=20 Changes in sources of community long-term care among african americans = and white=20 frail older persons. The Journal of Gerontology: Social Sciences, 49,=20 S14-S24.

Miner, S. (1995). Racial differences in = family=20 support and formal service utilization among older persons: A = nonrecursive=20 model. Journal of Gerontology: Social Sciences, 50B, S143-S153.

National Association of Home Care. (1993). = Basic=20 statistics about home care 1992. Washington, DC: National Association = for Home=20 Care.

Peek, C. W., Zsembik, B. A., & Coward, = R. T.=20 (1997). The changing caregiving networks of older adults. Research on = Aging, 19,=20 333-361.

Pratt, C., Jones, L., Shin, H., & = Walker, A.=20 (1989). Autonomy and decision-making among single older women and their=20 caregiving daughters. The Gerontologist, 29, 792-797.

Prohaska, T. R., & Glasser, M. (1996). = Patients'=20 views of family involvement in medical care decisions and encounters. = Research=20 on Aging, 18, 52-69.

Salive, M. E., Collins, K. S., Foley, D. = J., &=20 George, L. K. (1993). Predictors of nursing home admission in a biracial = population. American Journal of Public Health, 83, 1765-1767.

Soldo, B. J., Wolf, D. A., & Agree, E. = M. (1990).=20 Family, households, and care arrangements of frail older women: A = structural=20 analysis. Journal of Gerontology: Social Sciences, 45, S238-49.

Stone, R., Cafferata, G., & Sangl, J. = (1987).=20 Caregivers of the frail elderly: A national profile. The Gerontologist, = 27,=20 616-626.

Tennstedt, S., & McKinlay, J. B. = (1989). Informal=20 care for frail older persons. In M. G. Ory and K. Bond (Eds.), Aging and = health=20 care: Social science and policy perspectives. London: Routledge.

Tennstedt, S. L., Sullivan, L. M., = McKinlay, J. B.,=20 & D'Agostino, R. B. (1990). How important is functional status as a=20 predictor of service use by older people? Journal of Aging and Health, = 2,=20 439-461.

Verbrugge, L. M. (1990). The iceburg of = disability.=20 Pp. 55-75 in The legacy of longevity edited by S. M. Stahl. Newbury = Park:=20 Sage.

Wallace, S. P., Levy-Storms, L., Kington, = R. S.,=20 & Andersen, R. M. (1998). The persistence of race and ethnicity on = the use=20 of long-term care. Journal of Gerontology: Social Sciences, 53B, = S104-S112.

Wetle, T. (1995). Ethical issues and value = conflicts=20 facing case managers of frail elderly people living at home. Pp. 63-86 = in=20 Long-term care decisions: Ethical and conceptual dimensions edited by L. = B.=20 McCullough and N. L. Wilson. Baltimore, MD: Johns Hopkins University = Press.

Wetle, T., Levkoff, S., Cwikel, J., & = Rosen, A.=20 (1988). Nursing home resident participation in medical decisions: = Perceptions=20 and preferences. The Gerontologist, 28 (Supplement), 32-38.

Whitbeck, L., Hoyt, D. R., & Huck, S. = M. (1994).=20 Early family relationships, intergenerational solidarity, and support = provided=20 to parents by their adult children. Journal of Gerontology: Social = Sciences, 49,=20 S85-94.

Wiener, J. M., Illston, L. H., & = Hanley, R. J.=20 (1994). Sharing the burden: Strategies for public and private long-term = care=20 insurance. Washington, DC: The Brookings Institution.

Wolf, D. A., Freedman, V., & Soldo, B. = J. (1997).=20 The division of family labor: Care for elderly parents. Journal of = Gerontology:=20 Social Sciences, 52B, 102-109.

Wolinsky, F. D., Callahan, C. M., = Fitzgerald, J. F.,=20 & Johnson, R. J. (1992). The risk of nursing home placement and = subsequent=20 death among older adults. Journal of Gerontology: Social Sciences, 47,=20 S173-S182.

For a complete reprint of this article = contact=20 Haworth Press by telephone (1-800-HAWORTH) or EMail=20 (getinfo@haworthpressinc.com). Copies are available from: Haworth = Document=20 Delivery Center, The Haworth Press, Inc., 10 Alice Street, Binghamton, = NY=20 13904-1580. Article copyright The Haworth Press, = Inc.

=20

Indexing (document details)

Subjects:	Aging,=20 Child = development%22)">Child development, Elder=20 care, Families=20 & family life, Human=20 relations, Interpersonal=20 communication, Life, = Lifetime= ,=20 Older=20 people, Parents=20 & parenting, Personal=20 relationships, Social=20 services
Author(s):	Jenkins, Carol = L.
Document = types:	Feature
Publication=20 title:	Journal=20 of Women & Aging. New York: 2000. Vol. 12, Iss.&nb= sp;3/4; =20 pg. 3
Source = type:	Periodical
ISSN:	08952841
ProQuest document=20 ID:	625163051
Text Word = Count	6788
Document = URL:

Document View	=AB Back=20 to Results	= ; =20	<=20 Previous Document 39 of 52 Next=20 >
Print=20 \| Email=20 \| Copy link \| Cite this \| Mark Document =	Translate = document=20 from:

Print=20 \| Email=20 \| Copy link \| Cite this \| Mark Document =	Publisher=20 Information
^ Back=20 to Top	=AB Back=20 to Results	= ; =20	<=20 Previous Document 39 of 52 Next=20 >