Choosing an Adult Foster Home or a Nursing Home: Residents' Perceptions about Decision Making and Control.
Key words: adult foster care; control; decision making; long-term care; nursing homes
Given the current emphasis on cost containment, quality of life, and alternatives to nursing homes for older people with long-term care needs, family-like settings (often called adult foster care programs) are an appealing policy option for cost and quality considerations. Yet compared with their knowledge about nursing homes, policymakers' and practitioners' knowledge about adult foster care programs, their clientele, and, in particular, the decision to enter such a residential setting is scant. This article explores the circumstances of the decision to move to a foster home rather than a nursing home through the retrospective perceptions of cognitively intact residents who had moved to each setting.
We take as a given that the decisions to move to any residential long-term care setting are complex and multiply determined and made in a social context where family members and others have potential to influence those decisions. This article also examines what residents consider important in seeking a care setting, aspects of the decision-making process, and the residents' sense of control over that decision. Of particular interest is how the decision-making process might influence resident satisfaction and social participation in either care site.
Social workers are involved as advisors to older people and their families as they make life transitions--and particularly to those making such transitions from hospital settings. Any systematic information about care decisions for options less known than the nursing home should help refine social work practice.
Nursing homes are well known to health and social services professionals as the long-term care service for elderly people that accounts for the vast majority of public funding. A nursing, home is defined, somewhat circularly, as a residential health care facility that meets the federal requirements for certification for payment under Medicare and Medicaid and any additional state licensure requirements. These standards have evolved since their initial enactment in 1965, particularly since the implementation of the Nursing Home Reform Act of 1987 (P.L. 100-203), which grew out of the Institute of Medicine (1986) report on the quality of nursing homes. Nursing home standards are organized in categories such as administrative practices, environmental concerns, quality of care, quality of life, residents' rights, and infection control. Federal regulations that articulate some minimum nursing standards and staffing standards for training of nursing assistants also have been enacted. Within these requirements, nursing homes vary in size and other attributes. For the most part, however, they are modeled on hospitals, with the preponderance of accommodations being shared and a predominance of nursing routines (Gamroth, Semradek, & Tornquist, 1995; Kane & Caplan, 1990).
Adult Foster Care
Although adult foster care in the United States has been traced back to colonial times, when boarding homes provided meals and laundry, foster care specifically for older adults with disabilities is a fairly recent development (Sherman & Newman, 1979). The confusing variation of names--for example, family care homes, residential care, "homes plus," supportive care, board and care--and regulatory definitions associated with this type of arrangement complicate investigation of adult foster care. The most recent national study of adult foster care (Folkemer, Jensen, Lipson, Stauffer, & Fox-Grage, 1996) standardized a definition to include targeting adults unable to live independently because of physical or mental impairments or disabilities and who need supervision or personal care; offering 24-hour supervision, protection, and personal care besides room and board; serving a designated small group (ordinarily from one to six); and providing a homelike, family-like environment with a caregiver often residing in the home.
Mehrotra and Kosloski (1991) identified four common assumptions underlying foster care placement: (1) Nursing homes have detrimental psychological and social effects; (2) a family setting is inherently superior to an institutionalized setting; (3) adult foster care homes represent a "least restrictive care environment"; and (4) foster care is a desirable part of a continuum of long-term care. They also noted unresolved issues, such as the need for appropriate targeting and program goal setting. In a survey to which 42 states and the District of Columbia responded, Folkemer et al. (1996) found that 26 states had definitions of foster care that met their more stringent definitions of giving personal care. Of these, most state foster home programs served fewer than 1,000 residents, all but seven served mostly publicly subsidized clientele, and only five states served a majority of private-pay residents. Twenty states supplemented the Supplemental Security Income benefit standard to support adult foster care, and eight used Medicaid waivers to support direct care in adult foster homes. As Hudson, Dennis, Nutter, Gallaway, and Richardson (1994) pointed out, few states collect data on the health and functioning of adult foster care residents.
Setting for the Study
Oregon is ideal for the present study of location decisions for two reasons: (1) Adult foster care is well established as a middle class as well as a low-income alternative form of long-term care (Kane, Kane, Illston, & Nyman, 1991), and (2) adults with substantial disability levels receive foster care in Oregon foster homes. Indeed, Folkemer et al. (1996) found that Oregon was one of only five states with a preponderance of private-pay residents in adult foster care and was the only state with a program of substantial size in comparison to its nursing home population. Thus, we knew there would be some overlap in characteristics of residents in foster homes and nursing homes in terms of functional and care needs.
Regulations define an Oregon adult foster home as a family-style home in a residentially zoned area that provides care to a maximum of five elderly or disabled people. Foster home caregivers are required to attend 18 hours of preservice training, supplemented by 10 hours of yearly training (Ladd & Hannum, 1992). Since 1982 care for the service component of adult foster homes (as opposed to room and board) in Oregon could be reimbursed under the Medicaid waiver for eligible residents. Indeed, the state stimulated the development of adult foster care as part of its effort to divert people from nursing homes. Initially the state imposed few regulations. But by the 1990s the homes were classified into three levels (depending on the training and background of the provider) that differ according to maximum impairment limitations of the adults they can serve. Staff members in adult foster homes provide personal assistance and supervision and, under Oregon's Nurse Delegation Act (Oregon Board of Nursing, 1993), may perform nursing tasks if a nurse trains them and certifies their competence (Kane, O'Connor, & Olsen Baker, 1995). Thus, most adult foster care programs administer medications to the residents and provide routine nursing services.
Public officials at state and local levels in Oregon view adult foster care as an alternative service that consumers can choose rather than a niche on the continuum for those not yet "needing" nursing homes. Funding imperatives, therefore, drive placement decisions less in Oregon than in other states, because multiple options are available under Medicaid for people with comparable disabilities. Information constraints for both consumers and their advocates and advisors of course may limit the extent to which less familiar alternatives are genuine options.
Although adult foster care in Oregon is a real alternative to nursing homes, it by no means replaces them. The populations in the two settings overlap, but a 1991 study (Kane et al., 1991) showed that the foster care population was less impaired on average. However, all foster care residents reimbursed by Medicaid must be eligible for nursing homes based on their functioning, and Kane and colleagues found that the privately paying foster care residents had more disabilities than the Medicaid clientele (Kane et al., 1991; Stark, Kane, Kane, & Finch, 1995).
The widespread acceptance of adult foster care homes in Oregon is underscored because more private-pay residents use them than Medicaid-supported clients do. In 1995 about two-thirds of Oregon's approximately 10,000 adult foster care residents were paying privately, a proportion that has characterized the program from the beginning, although the absolute numbers of foster care have grown steadily.
Because adult foster care in Oregon is a large, mature program serving people with a wide range of incomes and disability levels, it provides an opportunity to investigate issues surrounding the decision to enter a foster home compared to a nursing home.
In Oregon almost all nursing home beds are licensed as intermediate care facilities (ICFs). The distinction between ICFs and skilled nursing facilities (SNFs) disappeared for quality purposes with the regulatory changes of 1987, which eliminated any difference in standards for the two settings, although Medicare will cover only those deemed to need SNF care in facilities certified for skilled nursing care. However, the ratio between ICFs and SNFs and the reimbursement accorded to each has always been a matter of state policy. Weil before we began our study, Oregon had developed two classifications for ICFs based on acuity and had almost eliminated the SNF designation, which was reserved for a handful of residents in extremely grave medical circumstances. Therefore, in drawing our sample, we used ICF residents only to achieve more direct comparability with foster care residents.
Relocation studies typically are divided into studies about relocation within the community, from the community to institutions, and between institutions (Schulz & Brenner, 1977). The distinctions blur, however, as new options are added to the housing-care continuum, such as community care, retirement communities, and various types of supportive settings (Armer, 1993; Netting & Wilson, 1991). Most early studies investigated relocation per se, focusing on adverse outcomes, especially mortality and morbidity (see, for example, Aldrich & Mendkoff, 1963; Lawton & Yaffe, 1970; Pablo, 1977). More recent approaches study the predictors of adjustment and positive and negative outcomes (Armer, 1993; Coulton, Dunkle, Chow, Haug, & Vielhaber, 1988; Mirotznik & Ruskin, 1985; Pohl & Fuller, 1980; Reinardy, 1992, 1995; Lieberman & Tobin, 1983). There is no existing literature on why people choose one form of long-term care over another, including the relationship of social work practitioners to that decision.
On the other hand, decision making and control over the decision to move have received recent attention in the long-term care literature, particularly in studies of relocation to nursing homes. Associations have been found between control or perceived choice and such postadmission outcomes as adaptation and adjustment, satisfaction with posthospital care, psychological distress, satisfaction with nursing home services, participation in activities, higher physical functioning, decreased pain, reduced use of medications, life satisfaction, depression, and health status (Armer, 1993; Gallagher & Walker, 1990; Hard & Noelker, 1982; Pohl & Fuller, 1980; Reinardy, 1992, 1995; Rodin, 1986a, 1986b; Lieberman & Tobin, 1983). The theoretical frameworks used to explain the relationship between control and postadmission well-being differ. It is important for practitioners, however, to realize that control may stem from a perceived sense of "mastery" over the task--seeing oneself at the center of the decision--as well as from the use of coping and management skills such as working through the steps of the decision process and absorbing adequate information about the new environment and alternatives (Lieberman & Tobin, 1983; Schulz & Brenner, 1977). Taken together, the evidence suggests that social workers should support both approaches in helping their elderly clients maintain a sense of control over the residential moves that they make for health purposes.
In the present study, using an existing data set, we were able to obtain some direct perceptions of cognitively intact elderly respondents about the preferences and circumstances surrounding their move and the factors influencing the move. In addition, the data shed light on the themes found in the literature about the effects of decision and control on resident wellbeing. The study, therefore, affords an opportunity to learn more about the lived experiences of the clientele whom social workers serve at times of great personal crisis and disruption.
Respondents and Data Collection
We collected the data as part of a study to develop and evaluate the effects of Oregon's foster care program compared to its nursing home program on functional abilities to perform activities of daily living (ADLs) (Kane et al., 1991). We sampled 405 foster care and 402 ICF residents from four geographical areas, chosen to represent urban and rural populations (including the Portland area) and to reflect state characteristics such as bed-to-population ratios and both public and private-pay residents. We selected foster homes from the four geographic areas, and all of the residents in each sampled foster home were interviewed. We sampled ICFs proportionally to the 1CF supply in each of the four geographic areas and interviewed 10 residents from each ICF sampled.
We conducted in person interviews at the foster homes and the nursing homes. Because we were interested in the experiences as perceived by the older people themselves, this article considers only those sufficiently cognitively intact to be interviewed without proxy informants. The resulting sample for analysis consisted of 439 respondents, 260 adult foster care residents (or 64 percent of the original sample) and 179 nursing home residents (or 45 percent of the original sample), a difference that reflects the much higher levels of cognitive impairment measured among the nursing home residents. To make the determination of cognitive incapacity, the interviewers visited all the sampled residents in person and administered the Short Portable Mental Status Questionnaire (Pfeiffer, 1975); if this 10-point scale had more than four errors, family members and facility staff were interviewed as proxies for specific questions.
Variables Regarding Relocation. The analysis presented here relies particularly on questions about the resident's decision to move to the setting. These include the health, social, and economic circumstances surrounding the move (presented as lists of items to which residents agreed or disagreed about the presence of each); the characteristics of the setting that residents perceived as important in the decision to move to the particular setting (also presented as a list, including items such as location, activity programs, rehabilitation programs, price, and homelike setting); consideration of one or more other residential alternatives before moving; key actors, both professional and informal, influencing the move; residents' perceptions of their own control over the decision to move; and the residents' satisfaction with the relocation decision.
We measured degree of control by an ordinal question: "Ali in all, how much did you control the decision to move to this foster (nursing) home? Would you say, 'I had almost complete control over the decision,' 'I had some control over the decision,' or 'I had little or no control over the decision.'" To clarify further the closed-ended variable on control, we asked respondents an open-ended question: "Please tell me a bit about how it happened that you moved from -- to NAME OF PLACE."
Dependent Variables. Resident satisfaction and social participation were dependent variables for analyses of the effect of perceived control over the location decision. We measured satisfaction with services on an eight-item scale ([alpha] = .84) with four-point Likert ratings for each item. Respondents were asked how satisfied they were in the following areas: food, room, daily, care, daily activities, amount of time to see family and friends, medical care, physical safety, and the safety of personal possessions. Activities within the residence were measured on an eight-item scale ([alpha] = .71) with three-point ratings of 1 never, 2 = sometimes, or 3 = often for each item. The activities included were watching television, talking to other residents, reading books or newspapers, doing hobbies, chatting with staff, playing cards or other games, participating in activities organized by the foster care or nursing home, and helping with cooking or other housekeeping.
Covariates. The interview protocol largely used fixed-choice questions and for some variables used established scales to collect data on functional abilities, mood and outlook, cognitive status, and use of health services. We captured physical functioning in an ADL score derived from a six-item scale ([alpha] = .88) that mirrored the assessment that Oregon uses in its statewide Medicaid waiver program: The items were ability to move around both in and out of doors, to transfer to bed and chair, to bathe, to dress oneself, to use the toilet alone, and to feed oneself. Ali were measured on a three-point range, from "without help," through "need some help," to "depend completely" or "almost completely" on help.
We measured cognitive status by the Short Portable Mental Status Questionnaire (MSQ) (Pfeiffer, 1975). Mood and outlook was measured by a battery of questions adapted from a scale that has been developed by Ware and used in a large study of outcomes of nursing home residents (Kane, Bell, Riegler, Wilson, & Kane, 1983).
We used frequencies and percentages to describe demographic characteristics of both adult foster care and nursing home respondents and to summarize findings related to influences and circumstances surrounding the decision to move. The chi-square test and, where appropriate, chi-square-based measures of association were used to compare sample characteristics and findings between the two types of residents. Bivariate comparisons were made within each of the samples to determine whether residents' characteristics and the circumstances surrounding the move were associated with the degree of self-perceived control over the decision to move.
Control over the decision was the independent variable in regression models testing whether control over the move at admission was associated with satisfaction with services and degree of activity in the residence. Scores from the satisfaction scale and the activities scale were used as the dependent variables. We entered several other variables into the model to control for differences in social, physical, and mental status that might affect satisfaction or activity. These were gender, source of payment (public or private pay), marital status (married or not married), age, place admitted from (hospital, home, facility similar to current residence, or other) physical health status, functional status, cognitive status, and psychological mood. For physical health, variables indicated diagnoses in certain disease groups: cardiovascular diseases, musculoskeletal diseases, and diseases of the nervous system. Scores from the MSQ controlled for cognitive functioning, because some residents who were able to be interviewed were nonetheless somewhat cognitively impaired. We constructed two dichotomous variables from two ordinal questions in the mood and outlook battery. They indicate whether respondents answered that they "always or almost always" were depressed or that they "never or almost never" felt that there were things to look forward to.
Finally we performed content analysis of the respondents' verbatim comments on how they came to move to the care setting to gain a richer understanding of the residents' experience. We examined the content categories in relation to the degree of control that the respondents indicated in the categorical question.
Like many relocation studies, this study is based on a cross-sectional, retrospective view of the relocation experience, whereas postadmission experiences may influence the residents' perceptions of the preadmission process. Also, we have no baseline data to control for health and cognitive status at the time of the move. However, we did look for patterns that might suggest reconstructed perceptions of the move. Length of stay was entered into the equations to detect any association between time of residence and the effect of control on the dependent variables, and no such interactions were found. The results are generalizable only to the residents sufficiently cognitively intact to participate in or recall the relocation process.
Characteristics of Respondents
Respondents in nursing homes and adult foster care were similar in demographic characteristics: Most were female, currently not married, ages 75 to 84, and overwhelmingly white (97 percent, not in table). Most had lived at their current residence for under two years (Table 1).
Table 1 Respondent Characteristics In Foster In Nursing Care Homes Homes (N = 260) (N = 179) Characteristic n % n % Age (years) Less than 75 64 25 37 22 75-84 137 54 92 54 85 or older 54 21 41 24 Marital status Married 27 10 33 18 Not married currently 233 90 146 82 Gender Male 58 23 51 29 Female 199 77 127 71 Education 8 years or less 119 47 85 48 9-12 years 80 31 43 24 More than 12 years 56 22 48 27 Relocated from ** Own home 53 20 49 27 Hospital 30 11 69 38 Similar type of setting (a) 63 24 11 6 Different type of relocation setting (b) 114 44 50 28 Primary payment source Private pay 140 56 80 49 Medicaid 114 44 50 28 Number of ADLs requiring some help 0-1 221 85 96 54 2-3 22 9 26 14 4 or more 16 6 57 32 Length of stay (months) 6 or fewer 71 31 46 28 7-20 months 79 34 45 27 More than 20 83 35 76 45 Diagnosis Heart condition * 54 21 53 30 Stroke 50 19 34 19 Skeletal-muscular * 41 16 45 25 Dementia 39 15 37 20 NOTE: Numbers may not add to totals in some cases because of missing values. (a) The foster home residents coming from other foster homes and the nursing home residents coming from other nursing homes are counted in this category. (b) The nursing home residents coming from foster care and the foster care residents coming from nursing homes are counted in this category, and a few residents coming from institutions such as state mental hospitals. * p [less than or equal to] 05. ** p [less than or equal to] .01.
Three major areas in which the foster care residents and nursing home residents differed were place admitted from, physical health, and ADLs. The foster home residents were much less likely to have been admitted from hospitals (11 percent compared with 38 percent). On the other hand, 24 percent of foster home residents had been admitted from other foster homes, whereas only 6 percent of nursing home residents had moved there from other nursing homes. Also, more of the foster care residents came from other types of settings--mainly residential care--or residential care facilities (that is, board and care home with limited personal care services). It is striking, however, that many of the foster home residents had experienced nursing homes and nursing home residents had experienced foster homes immediately before their move to their present location. Had we inquired about all previous locations, we expect that even a higher proportion of respondents would have experienced both settings.
Although the foster care group included some people with serious disabilities, on average it had significantly less disability than the nursing home group. Across all ADLs, nursing home residents had more need for help, both "some help" (p < .001) and "complete help" (p < .001). They also were more likely to have been diagnosed with prevalent health problems such as heart disease or fractured hip. Fifteen percent of respondents in foster homes and 20 percent of respondents in nursing homes had a dementia diagnosis (even after removing the proxy interviews required for 35 percent of the foster care residents and 55 percent of the nursing home residents).
Forty-three percent of the foster care residents saw themselves as having had complete or almost complete control over the decision to move; 30 percent indicated some control; and 27 percent said that they had little or no control, a significant difference from the nursing home residents (p < .001) (Table 2). With these residents the pattern reversed itself: The largest category was made up of those who claimed they had little or no control.
Table 2 Respondents' Relocation Decision Variables In Foster Care In Nursing Homes Homes (N = 260) (N= 179) Variable n % n % Control over decision to move *** Almost complete 109 43 45 26 Some 78 30 45 26 Little or none 69 27 84 48 More than one home considered 85 36 71 44 Decision to move a good one ** 242 94 144 85 People influencing decision Family member 157 62 116 66 Physician *** 72 29 77 47 Home health nurse 18 7 18 11 Hospital discharge planner 22 9 24 15 Social worker 44 18 27 16 Case manager * 17 7 17 10 Circumstances before move Overnight hospital stay *** 115 46 112 64 Sickness requiring extensive care 55 22 49 28 Accident/injury 78 31 57 33 Falls ** 69 27 66 38 Memory loss 63 25 37 21 Serious new illness 53 21 37 21 Life changes before move Death of spouse 29 12 19 11 Death of caregiver 23 9 10 6 Caregiver moved away 28 11 11 6 Loss of income/financial difficulties 35 14 18 16 NOTE: Percentages may be slightly off because of missing values. * p [less than or equal to] .05. ** p [less than or equal to] .01. *** p [less than or equal to] .001.
Thirty-six percent of foster care residents had considered one or more other foster homes before choosing the current one, not significantly different from the portion of nursing home residents who had considered other nursing homes. Foster care and nursing home residents differed on whether they considered the decision to move to have been a good one. Although a high percentage of both groups agreed that the decision had been good, more foster care residents answered affirmatively than did nursing home residents (p [less than or equal to] .01).
For foster care residents, the two major groups influencing the decision to move were family members and Oregon Senior Services case managers (Table 2). Physicians were rated third. For nursing home residents, family members also were most likely to influence the move, but influence from physicians was more likely (p [less than or equal to] .001) and from case managers less likely (p [less than or equal to] .01) than was the case for the foster care residents. The difference highlights the alternative admission routes taken by the two groups; nursing home residents more frequently entered through the acute care system. Both foster care and nursing home respondents reported social workers, discharge planners, and home health-public health nurses as being far less likely to influence the decision.
For foster care residents, the most frequent change in health circumstances was an overnight stay in a hospital (46 percent), followed by the occurrence of an injury or accident (31 percent), frequent falls (27 percent), worries about memory loss (25 percent), sickness requiring extensive care at home (22 percent), and a serious new illness (21 percent). These percentages were similar to the nursing home group, although foster care residents were less likely to have experienced a hospital stay (p [less than or equal to] .001) or frequent falls (p [less than or equal to] .01).
There were no significant differences between foster care and nursing home residents regarding changes in life circumstances. For both groups, death of a spouse and financial difficulty were most frequently mentioned; 48 percent of foster care residents and 41 percent of nursing home residents had experienced none of the suggested events, including an unspecified "other" category. The primary issue is the comparability between those ending up in the two settings in terms of key factors that often precipitate a long-term care move. Although the numbers experiencing any event are relatively small, the key issue is the comparability.
Residents were presented with a list of 10 specific attributes and were asked if each was important in deciding to move to "this" home (Table 3). Supervision and safety (92 percent), home-like atmosphere (91 percent), privacy (82 percent), and flexible routine (76 percent) were most often mentioned by the foster care residents. Supervision and safety also had the highest rating by the nursing home group (86 percent), but this was followed by personal assistance care (81 percent), medical care (78 percent), and cost of care (69 percent). Home like atmosphere (p [less than or equal to] .001), privacy (p [less than or equal to] .001), and flexible routine (p [less than or equal to] .001) were more important to the foster care residents; medical care (p [less than or equal to] .05) and physical rehabilitation (p [less than or equal to] .01) to the nursing home residents.
Table 3 Factors Respondents Considered Important in Relocation Setting In Foster In Nursing Care Homes Homes (N = 260) (N = 179) Factor Viewed as Important n % n % Home-like atmosphere *** 223 91 96 53 Location/neighborhood 158 64 101 62 Personal assistance 191 76 132 81 Medical care * 169 68 127 78 Supervision 227 92 140 86 Privacy *** 204 82 92 58 Flexible routine *** 181 76 68 46 Cost of care 177 74 108 69 Organized activities * 80 33 70 44 Physical rehabilitation ** 67 28 64 41 None: Numbers may not add to totals because of missing data. * p [less than or equal to] .05. ** p [less than or equal to] .01. *** p [less than or equal to] .001.
Perceived Control over the Decision
The only demographic variable associated with perceived control was education. A low, positive association (p < .05; [delta] = .17) was found between years of education and degree of control for the foster care residents. For the nursing home residents, those admitted from acute care hospitals were more likely to indicate low control over the decision than those with other sources of admission (p < .005; Cramer's V = .24), whereas those admitted from their own homes were more likely to indicate high control (p < .02; Cramer's V = .20).
When control was cross-tabulated with the types of people perceived as influencing the decision, perceived control was related to family influence for both foster home and nursing home residents. Residents who were influenced by family members were less likely to indicate that they were in control of the decision (foster care: p < .02; Cramer's V = .18; nursing home: p < .004; Cramer's V = .25). For example, of foster care residents who indicated little or no control, 67 percent were influenced by family, and 33 percent were not. No associations were found between control and changes in life circumstances, but we found three associations for changes in health conditions prior to the move. Foster care residents who were aware of experiencing memory loss prior to the move were less likely to perceive themselves as having been in control (p < .01; Cramer's V = .20). And nursing home residents who had an overnight hospital stay or who had experienced a serious new illness also perceived less control (p < .02; Cramer's V = .21 and p < .01; Cramer's V = .22, respectively).
Surprisingly, no association was found between perceived control and consideration of more than one alternative home in making their decision. There was an association, however, between thinking that the decision to move was a good one and control. For both foster care and nursing home residents, those who said that the move was good were more likely to perceive themselves in control of the decision (p < .01; Cramer's V = .18 and p < .001; Cramer's V = .33, respectively).
Relationship of Control to Satisfaction and Activity
The regression model was initially run on all respondents (N = 439), but interaction was found between control over the decision and type of residence (foster care as opposed to nursing home). Subsequent runs for satisfaction as the dependent variable (Table 4) and for activities in the new setting as the dependent variable (Table 5), were therefore done separately for the foster care and nursing home residents. Ali four of the regression equations were significant (p < .01 to p = .000) with a moderate amount of variance explained ([R.sup.2]s from . 16 to .27). Control over the decision was significant, however, only for the foster care residents. In the nursing home group, grade school education (a dummy variable representing the questionnaire's categories of "completed grade school" and "did not complete grade school") was associated with higher satisfaction (p < .01), as was higher ADL functioning (p < .005). For the nursing home activity equation, those with greater disabilities, less education, and little to look forward to participated less in activities.
Table 4 Regressions for Scores in Satisfaction for Residents of Foster Care Homes and Nursing Homes (N = 439) Foster Care Homes Independent Variable [beta] Age -0.0119 Nervous system diagnosis -0.0078 Cardiovascular/musculoskeletal diagnosis -0.0325 Complete control -1.9570 **** (a) Some control -1.1361 * (a) ADL score 0.0942 Mental status score -0.0248 From same type facility -1.0983 From hospital -1.9539 * From own home -1.6488 ** Male -0.6440 Married -0.4421 Medicaid payment status 0.5748 Less than grade school education -1.0269 Depressed mood 1.0965 Little hope 0.3435 [R.sup.2] = 0.16 F = 2.218 *** Nursing Homes Independent Variable [beta] Age -0.0051 Nervous system diagnosis -0.3371 Cardiovascular/musculoskeletal diagnosis -0.6085 Complete control -1.3628 Some control 0.2581 ADL score 0.2233 *** Mental status score -1.0241 From same type facility 0.2184 From hospital 0.2311 From own home 0.2167 Male 1.0751 Married 0.6336 Medicaid payment status -0.6726 Less than grade school education -1.4364 ** Depressed mood 0.1714 Little hope 1.0917 [R.sup.2] = 0.24 F = 1.989 * NOTE: ADL = activity of daily living. (a) The beta is negative because low numbers represent higher satisfaction on the scale. * p [less than or equal to] .05. ** p [less than or equal to] .01. *** p [less than or equal to] .001. **** p [less than or equal to] .0001. Table 5 Regressions for Scores in Activity for Residents of Foster Care Homes and Nursing Homes (N = 439) In Foster Care Homes Independent Variable [beta] Age -0.0346 Nervous system diagnosis -0.5023 Cardiovascular/musculoskeletal diagnosis -0.0958 Complete control 1.2630 *** Some control 0.2095 ADL score -0.3375 **** Mental status score -0.1470 * From same type of facility 1.1888 From hospital 0.7973 From own home 0.8904 Male -0.2923 Married -0.6154 Medicaid payment status 0.3528 Less than grade school education 0.1105 Depressed mood -1.3424 Little hope -0.5719 [R.sup.2] = 0.27 F = 4.938 *** In Nursing Homes Independent Variable [beta] Age -0.0162 Nervous system diagnosis 0.1018 Cardiovascular/musculoskeletal diagnosis 0.8947 Complete control 0.1742 Some control 0.1062 ADL score -0.1383 * Mental status score 0.0191 From same type of facility -0.3185 From hospital -0.3446 From own home -0.0150 Male -0.3463 Married 0.4047 Medicaid payment status 1.6800 * Less than grade school education 1.1088 * Depressed mood -2.5946 Little hope -1.4599 * [R.sup.2] = 0.19 F = 2.04 ** NOTE: ADL = activity of daily living * p < .05. ** p < .01. *** p < .001. **** p < .0001.
For foster care residents, "almost complete" control predicted greater satisfaction with services (p < .005), as well as more activity in the home (p < .005), and "some" control predicted satisfaction with services (p < .05). Lower educational attainment also predicted greater satisfaction (p < .01), and residents admitted from their own home or the hospital were found to be more satisfied than those with other sources of admission (p < .01 and p < .05, respectively). For the activity equation, older age (p < .05), functional disability, (p < .001), and cognitive impairment (p < .05) predicted less activity, whereas those who had been admitted from similar homes tended to participate more in activities (p < .05).
Because the relationship between perceived control and the dependent variables was significant for the adult foster care residents, we ran an interactive model to test the hypothesis that control might have a greater effect on those who considered more than one home as an alternative. Considering more than one home had no effect on either satisfaction or activity. We also ran models to test whether place admitted from might interact with perceived control. One source of admission was found significant: the model testing interaction between control and admission from their own homes. Control was associated with higher satisfaction for residents who entered adult foster care from their own homes ([R.sup.2] = .18; F = 2.89; significance of home x high control = .05)
Content analysis of the open ended question on "how it happened that you moved here" helped to shed light on what the respondents meant by control.
Foster Care Residents
For foster care residents who claimed that they had almost complete control over the decision to move, a strong majority of answers, about 85 percent, fell within the major approaches to control described in our literature review: a sense of mastery or coping strategies. Some respondents emphasized mastery over the decision and let it go at that: "My family never tells me what to do," "I decided myself," "I wanted this home despite my daughter's suggestion to live in another home." In the majority of responses, however, residents did not speak directly to making the decision, but focused on the home as a better alternative or on the desirability of moving to this particular home (coping strategies): "Couldn't stand it at the nursing home," "Daughter helped me pick this one close to her home," "My wife was here," "Had met the owner and loved her."
The comments of foster care residents who claimed to have little or no control contrasted sharply with the first group. About 80 percent of the responses related to the notion of mastery, almost all suggesting that someone other than themselves was in control of the decisions: "My daughter made me get out," "My son just picked me up and brought me here." "I got kicked out of the other foster home." Although a few of the comments, fewer than 10 percent, suggested some coping strategies ("My wife is here."), most described a decision that they did not work through: "My daughter made all the arrangements," There was "no room at my granddaughter's so she arranged for me to come here."
The comments of foster residents who indicated "some control" over the move appeared to fall between the extremes described above. Most spoke about influence or help from family or others: "My oldest daughter did most of it," "My daughter said she knew of a nice home and we'll go out and see the place." About 20 percent dwelt on the reasons for the move without mentioning anyone's influence; about the same percentage spoke as if the arrangements had been made for them. If anything, this group appeared closer to those who claimed "no control" than those who indicated it was their own decision.
Nursing Home Residents
In comparison to the foster care group, there were no sharp contrasts across the three categories of nursing home residents: those with complete, some, or little control over the decision. And in general the notions of mastery and coping strategies were not illustrated as forcefully in their comments. For example, none of the remarks of the nursing home residents who fell in the high control category suggested a strong sense of mastery, as was the case for the foster care residents. In fact their answers appeared closer to those of foster care respondents who mentioned some control. There was control but it was qualified; many dwelled on help received from others in making the decision or on why the move was necessary. Again, in contrast to foster care residents, those who claimed little or no control on the fixed-item question did not focus on notions of being forced to move or being tricked into the decision. Rather their comments tended toward a simple acknowledgment that someone else, usually a family member, had made the decision: "My family put me here and said I'd like it," "My brother made most of the decisions while I was in the hospital."
The study shows substantial differences between interviewable residents in adult foster care and interviewable residents in nursing homes in the circumstances precipitating the move, the nature of the search, in perceived control, and in the attributes of the setting deemed important. The qualitative analysis shows dramatically that high proportions of residents in both settings perceived that they had been manipulated into the move or that the move had been inevitable with little opportunity for choice. Nursing home residents were flatter in the way they described this than were foster care residents.
Moving to a long-term care residence is an important life transition. The findings of this study suggest that in Oregon moving to an adult foster care home--at least in contrast to moving to a nursing home--can be an experience that allows many elderly people to see themselves in control and that such perceived control is associated with higher satisfaction and activity following the transition. The findings also suggest that, in making their decision, most older adults see foster care as a distinct alternative to nursing care, appealing to their needs for a home-like atmosphere, privacy, and flexible routines.
Why the finding that nursing home residents were less likely than foster care residents to perceive themselves in control of the move? The nursing home residents tended to relocate to long-term residence from the acute care sector, and admission from acute care was also associated with less control over the decision to move. Perhaps the acute care sector had its own momentum, channeling patients into nursing home care without the time or conditions that are necessary to encourage the older person to feel in control over the decision.
The study found an association between control and measures of satisfaction and activity for the foster care residents only, not the nursing home residents. Again, perhaps the large percentage of admissions from acute care hospitals in the nursing home group compared with the proportion of foster care residents making the decision at home accounted for this differential effect of control on the dependent variables. An additional explanation may be found in the nature of control and its association with adverse effects. Perhaps people hand over control to trusted others without negative impact, when they feel circumstances to be beyond their competence or control (Thompson, 1981). The comments of the nursing home residents, unlike the comments of the foster care residents, did not reflect the extremes of mastery over the move or being forced to move. Many mentioned that the decision was made by others, but without their protest. Patients in acute care settings, given the pressing conditions of health and circumstance, might have been more willing to delegate decision making to those whom they know and trust.
Social workers and nurses who work as hospital discharge planners and as community based case managers need to be more sensitive to the importance of facilitating the older person's sense of control over the move. This need is underscored by our finding that both foster care and nursing home residents view family members as providing a major influence on the move but one that appears to dampen rather than support a sense of control. Reliance on family members to make decisions may have negative effects on elderly people's later perceived well-being.
From a practice perspective, practitioners may need to give more thought to placing more emphasis on working directly with the older person rather than concentrating principally on family members, who are often more accessible. Notions that the "family is the client" may be counter productive. Although the implications of family systems theory are particularly salient when working with older adults, it is also most important to work with them as clients in their own right. Certainly it is necessary to deal directly with the feelings of helplessness that placement often engenders. It is also important to systematically encourage clients to consider the attributes of care settings that are important to them. Using lists such as those used in this retrospective study may be helpful to anchor such discussion.
It also seems that practice patterns have not yet caught up with the range of options available to seniors in a state like Oregon, or the dynamic value of their choices. Note that more than a few foster care residents had been in nursing homes and vice versa and that most of these foster care residents had moved from a different foster home. (One imagines that some of these moves reflect the resident's choice, taking advantage of the flexibility of the program to move to a preferred setting, whereas others might reflect rejection by the setting.)
Hospital discharge planning, as historically and currently too often practiced, seems much too limited a concept to reach and serve best the seniors who could benefit from help with relocation decisions. Our findings seem to call for a more flexible and continuous function of planning as opposed to a one-time and hasty discharge plan made from the hospital. Hospitalizations are short, postacute and subacute care occur in a variety of step-down locations, and posthospital location decisions need not be painted as irrevocable choices. Some assistance with reviewing and making new long-term care decisions may be needed for people in a wide range of long-term care locations. Perhaps this continuity can be found in case managers in public or private programs. Another potential is that personnel, including social workers, in managed care organizations can assist older people with the dynamic process of decision making related to their care regardless of their current physical location. At the very least, if nursing homes are used deliberately in lieu of hospitalizations or as sites for short-term rehabilitation, clients and family members must be made aware that nursing homes are not perceived as a final destination.
Practitioners have long had an uneasy feeling that the haste and sense of crisis associated with hospital discharge creates a poor circumstance for life decision making. Similarly, decision advisors will need to exert care that the current emphasis on clinical pathways and computerized decision trees for identifying the best placement for patients in managed care systems not omit the person from a highly personal decision. This study suggests that even in a state that has, through public policy, expanded the range of possible decisions for those who must relocate for long-term care, the way the older person participates in the decision can influence the postdecision outcomes.
In this cross-sectional study, we have incomplete data on the history of relocations for the respondents. Our data cannot tell us where in the sequence of decision making clients are more likely to see the decision as out of their control. Nor can the data tell us, once clients have lost control over the relocation process, whether they can regain that sense of control in subsequent moves. The current context of care seems to offer a wider range for residential decisions and more opportunities to make sequential decisions. Further clinical and health services research could help illuminate those patterns.
Original manuscript received October 21, 1996
Final revision received December 4, 1997
Accepted March 5, 1998
Aldrich, C. K., & Mendkofl, E. (1963). Relocation of the aged and disabled: A mortality study. Journal of the American Geriatric Society, 11, 185-194.
Armer, J. (1993). Elderly relocation to a congregate setting: Factors influencing adjustment. Issues in Mental Health Nursing, 14, 157-172.
Couhon, J. C., Dunkle, R. E., Chow, J., Haug, M., & Vielhaber, D. P. (1988). Dimensions of post-hospital care decision-making: A factor analytic study. Gerontologist, 28, 218-223.
Folkemer, D., Jensen, A., Lipson, L., Stauffer, M., & Fox-Grage, W. (1996). Adult foster care for the elderly: A review of state regulatory and funding strategies: Volume I (No. 9604A). Washington, DC: American Association of Retired Persons, Public Policy Institute.
Gallagher, E., & Walker, G. (1990). Vulnerability of nursing home residents during relocations and renovations. Journal of Aging Studies, 4(1), 31-46.
Gamroth, L. M., Semradek, J., & Tornquist, E. M. (Eds.). (1995). Enhancing autonomy in long term care: Concepts and strategies. New York: Springer.
Harel, Z., & Noelker, L. (1982). Social integration, health, and choice. Research on Aging, 4, 97-111.
Hudson, J., Dennis, D., Nutter, R., Gallaway, B., & Richardson, G. (1994). Foster family care for elders. Adult Residential Care Journal, 8(2), 65-76.
Institute of Medicine. (1986). Improving the quality of care in nursing homes. Washington, DC: National Academy Press.
Kane, R. A., & Caplan, A. L. (1990). Everyday ethics: Resolving dilemmas in nursing home life. New York: Springer.
Kane, R. A., Kane, R. L., Illston, L., & Nyman, J. (1991). Adult foster care for the elderly in Oregon: A mainstream alternative to nursing homes? American Journal of Public Health, 81, 1113-1120.
Kane, R. A., O'Connor, C., & Olsen Baker, M. (1995). Delegation of nursing activities: Implications for patterns of long-term care. (Report prepared for American Association of Retired Persons). Minneapolis: University of Minnesota.
Kane, R. L., Bell, R., Riegler, S., Wilson, A., & Kane, R. A. (1983). Assessing the outcomes of nursing-home patients. Journal of Gerontology, 38, 385-393.
Ladd, R., & Hannum, C. (1992). Oregon's adult foster care program--A model for community-based long term care for elderly and disabled. Community Alternatives, 4, 171-184.
Lawton, M. P., & Yaffe, S. (1970). Mortality, morbidity, and voluntary changes in residence on the well-being of older people. Journal of the American Geriatric Society, 18, 823-831.
Lieberman, M. A., & Tobin, S. S. (1983). The experience of old age: Stress, coping, and survival. New York: Basic Books.
Mehrotra, C., & Kosloski, K. (1991). Foster care for older adults: Issues and evaluations. Home Health Care Quarterly, 12, 115-136.
Mirotznik, J., & Ruskin, A. P. (1985). Interinstitutional relocation and its effect on psychosocial status. Gerontologist, 25, 265-270.
Netting, F., & Wilson, C. (1991). Accommodation and relocation decision making in continuing care retirement communities. Health & Social Work, 16, 266-273.
Nursing Home Reform Act of 1987, P.L. 100-203, U.S. Statutes at Large, 101, 1330.
Oregon Board of Nursing. (1993). Oregon administrative regulations 851-47-010(23). Portland: Author.
Pablo, R. Y. (1977). intra-institutional relocation: Its impact on long-term care patients. Gerontologist, 17, 426-435.
Pfeiffer, E. 11975). A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. Journal of the American Geriatric Society, 23, 433-441.
Pohl, J. M., & Fuller, S. S. (1980). Perceived choice, social interaction, and dimensions of morale of residents in a home for the aged. Research in Nursing and Health, 3, 147-157.
Reinardy, J. 11992). Decisional control in moving to a nursing home: Post-admission adjustment and weil being. Gerontologist, 32, 96-103.
Reinardy, J. (1995). Relocation to a new environment: Decisional control and the move to a nursing home. Health & Social Work, 20, 31-38.
Rodin, J. (1986a). Aging and health: Effects of the sense of control. Science, 233, 1271-1276.
Rodin, J. (1986b). Health, control, and aging. In M. M. Baltes & P. B. Baltes (Eds.), The psychology of control and aging (pp. 139-165). Hillsdale, NJ: Lawrence Erlbaum.
Schulz, R., & Brenner, G. (1977). Relocation of the aged: A review and theoretical analysis. Journal of Gerontology, 32, 323-333.
Sherman, S. R., & Newman, E. S. (1979). Foster family care for the elderly: Surrogate family or mini institution? International Journal on Aging and Human Development, 10, 165-176.
Stark, A., Kane, R. L., Kane, R. A., & Finch, M. 11995). Effect on physical functioning of care in adult foster homes and nursing homes. Gerontologist, 35, 648-653.
Thompson, S. C. (1981). Will it hurt less if I can control it? A complex answer to a simple question. Psychological Bulletin, 90, 89-101.
James Reinardy, PhD, is assistant professor, School of Social Work, University of Minnesota, 400 Ford Hall, 224 Church Street, SE, Minneapolis, MN 55414; e-mail: firstname.lastname@example.org. Rosalie A. Kane, PhD, is professor, Institute for Health Services Research, School of Public Health, University of Minnesota, Minneapolis.
|Printer friendly Cite/link Email Feedback|
|Author:||Reinardy, James; Kane, Rosalie A.|
|Date:||Nov 1, 1999|
|Previous Article:||Critical Action Research: A Model for Social Work Knowing.|
|Next Article:||The Bell Curve: Race, Socioeconomic Status, and Social Work.|