Factors Associated With Preferences for Institutionalized Care in Elderly Persons

Abstract

Objective: To study the willingness to use institutional care versus home care in hypothetical situations of permanent disability and Alzheimer’s disease (AD). Method: A convenience sample of 484 adults aged 45 and older was obtained. Respondents were asked about their care preferences in case they were diagnosed with AD or became permanently disabled. Based on Andersen’s extended behavioral model, a large group of correlates was examined. Results: Institutional care was preferred in the situation of becoming sick with AD, whereas being cared for at home by the family was preferred in a situation of permanent disability. Fear of losing one’s independence and concern over burdening the family were associated with care preferences in both situations. Discussion: Although some similarities exist between the correlates for care preferences in the two distinct situations, there are noticeable differences. These are discussed and implications for practice are suggested.

Keywords

nursing home institutionalization Alzheimer’s disease permanent disability

Introduction

Demographic projections estimate a continuous decline in mortality during the next decades, accompanied by an increase in the number of older persons in general, and in the number of older persons with various health and functional problems in particular (Vincent & Velkoff, 2010). In this context, long-term care (LTC) and nursing homes will play a particularly important function (Katz, 2011). However, the percentage of older adults residing in nursing homes (NH) remains small (Lakdawalla et al., 2003), and despite the efforts of researchers to understand the reasons for NH placement, the phenomenon is still awaiting a clear explanation.

Research in this area has been largely focused on two distinct, although related, aspects. The first includes the examination of predictors of NH placement, usually by using prospective designs and based on population-based samples. Findings from this line of research show strong evidence that the main risk factors leading to institutionalization include increased age, functional and cognitive deterioration, dementia, previous nursing home placement, and multipharmacy (for a recent review on the topic, see Luppa et al., 2010).

Another important, although less epidemiological line of research in this field is aimed at examining the willingness of older adults to consider future institutionalization in a NH. These studies are based on motivational and cognitive assumptions, and attempt to identify individual factors that are potentially subject to change. Overall, these studies show high reluctance to enter a NH among older adults, with as few as 15% to 32% reporting themselves willing to enter a NH (e.g., Dubois, Dubuc, Raîche, Caron, & Hébért, 2008; Kim & Choi, 2008; Kim & Kim, 2004; Tse, 2007). In accordance with the epidemiological studies, this line of research found that health and functional problems (Dubois et al., 2008; Jang, Kim, Chiriboga, & Cho, 2008; Kim & Kim, 2004), as well as poor cognitive status and dementia (Kim & Choi, 2008; McCormick et al., 2002; Nieboer, Koolman, & Stolk, 2010), were associated with increased willingness to enter a NH. In addition, social factors (such as lack of social support and living alone; Chen, 2011; Guan, Zhan, & Liu, 2007; Nieboer et al., 2010), or having a close friend in a NH (Jang et al., 2008), were also identified as important determinants of willingness.

Findings regarding socio-demographic factors are inconsistent in both lines of research, with several studies reporting female gender (Chen, 2011), better financial status, and lower education (Guan et al., 2007) to be associated with greater willingness to move into a NH, whereas others found male gender (Kim & Choi, 2008), younger age (Kim & Kim, 2004), having fewer children, poorer financial status (Kim & Choi, 2008; Kim & Kim, 2004), and greater education (Kim & Kim, 2004) to be associated with greater willingness to enter a NH. Finally, one of the most important factors associated with greater willingness to enter a NH was positive attitudes toward NH (Chen, 2011; Tang, Wu, Yeung, & Yan, 2009) and greater knowledge of NHs (Chen, 2011).

Following these findings, the overall aim of the present study was to examine the determinants of preferences regarding NH care using the extended Andersen Model (Andersen, 1995; Bradley et al., 2002) as a theoretical framework.

Theoretical Context and Study Aims

According to Andersen’s behavioral model for health care utilization (Andersen, 1995; Andersen & Newman, 1973), and its current extensions (e.g., Bradley et al., 2002), individual factors that influence health-services utilization can be divided as follows: (a) predisposing factors, such as age, sex, and education; (b) need factors, such as the degree of physical and cognitive disability and the duration of the problem; (c) enabling factors, such as availability of support and financial resources; and (4) psychosocial factors, such as attitudes and beliefs regarding service use, knowledge of and familiarity with services (Bradley et al., 2002). This model provides a framework through which to understand the different factors that may affect the extent to which older adults are willing to consider utilization of health services, in general (Bradley et al., 2002), and moving into a NH, in particular (Kim & Choi, 2008).

In the present study, the willingness to move into a NH (vs. preferring to be cared for at home) was examined in relation to two hypothetical situations: becoming permanently disabled and being diagnosed with Alzheimer’s disease (AD). Whereas “aging in place”—meaning, residing in the community in old age is considered a desirable aging trajectory, and ample body of research was directed at examining enabling factors associated with the opportunity to age in place and studying the formation of supportive communities (e.g., Greenfield, 2014; Lehning, Smith, & Dunkle, 2013; Merrill & Hunt, 1990), severe situations, such as receiving a terminal diagnosis, might “tip the scales” and increase the willingness to choose NH placement over aging in place. Also, severe situation, where independence might be lost, might leave little choice but to accept NH placement.

The present study aims to explore the willingness to enter NH versus preferring to receive home care in two such severe (and terminal) situations. It expands current knowledge in several meaningful ways. First, it distinguishes between two permanent disability situations: permanent disability and AD. Although McCormick and colleagues (2002) have previously shown higher preferences for long-term care in the case of becoming sick with dementia, their study compared these preferences with a hypothetical situation of temporary disability (hip fracture), instead of a permanent disability. Therefore, it remains unclear whether the reactions to receiving an Alzheimer’s diagnosis are essentially different from other terminal states of disability. Second, in the present study, we also examined a wide range of variables that may explain the willingness to choose institutional care versus home care, with a special focus on a set of psychosocial factors that have not been previously examined in this context, such as social support and perceptions of risk and worry. Third, contrary to previous studies that restricted their examination of willingness to enter a NH to samples of older adults aged 60 and above (Dubois, 2008; Kim & Choi, 2008; Kim & Kim, 2004; Lee, Kovner, Mezey& Ko, 2001; McCormick et al., 2002; Tse, 2007), this study is based on a sample of community-residing adults aged 45 and older. This wider age range might provide us with meaningful insights, as attempts to change attitudes toward health care may prove more fruitful if directed to middle-aged adults than attempts directed at older populations. In addition, adults in this age range could play a significant role in shaping the care trajectory for their older parents, as well as for themselves. Finally, this is, to the best of our knowledge, the first study to examine this topic in Israel.

The Israeli Setting

In Israel, approximately 10% of individuals are aged 65 and older, and the percentage residing in a NH is very small—only about 3%, whereas the majority of older adults and elderly reside within the community (Brodsky, Shnoor, & Be’er, 2012). Two thirds of the number of total beds provided in LTC facilities in Israel are geared toward the needs of dependent and demented patients under the responsibility of the Ministry of Health, operating to almost full capacity (92% capacity). In terms of their demographic characteristics, NH residents in Israel are mainly female, Israeli residents who are cognitively impaired (Be’er, 2004).

Data collected between 1983 and 1999 reflect a significant 9% decrease in the percentage of NH placement in Israel. This trend is especially significant for the young-old cohort (65-74 years old), whereas findings suggest a slight increase in the percentage of the old-old cohort (75 and older) placed in NH (Clarfield, Rosenberg, Brodsky, & Bentur, 2010). Data from the United States corroborate with this Israeli trend, showing decreasing rates of NH placement (Clarfield et al., 2010), and reflecting the implementation of policies (both in Israel and in the United States) aimed at allowing older individuals to age in place. In the United States, special programs aimed at enabling transition of older people back into their homes after NH placement are successfully operating in recent years (Bardo, Applebaum, Kunkel, & Carpio, 2014). To the best of our knowledge, no such programs are currently being operated in Israel. In addition to its theoretical importance, findings from the present study might help decision makers considering such programs in Israel.

Method

Participants

A convenience sample of 484 Jewish community-residing adults aged 45 and above participated in the study. This age group was selected because of the relevance of the topic under study for it. All participants were unpaid volunteers who were approached by trained research assistants in three large cities in the northern part of Israel. Approximately half the participants were women, their mean age was 60, they had on average around 14 years of education, and the majority was married (85%). Further information on the characteristics of the sample and the study variables can be found in Table 1. Although random sampling is a desirable technique for minimizing bias, it was not feasible in this study because of financial constraints. However, comparing central socio-economic status (SES) characteristics of our sample with the characteristics of the Israeli Jewish population as reported by the Israeli Bureau of Statistics (2014) showed that the study sample was very similar to the Israeli Jewish population aged 45 and above, although there was a lower percentage of women (50% compared with 54%), a slightly higher level of education (14 years of education compared with 13 years) and of married persons (85% compared with 74%), and fewer children in our sample (2.7 compared with 2.9).

Table 1.

Characteristics of Study Participants and Main Variables (N = 408).

Main variables	n (%) or M (SD)
Predisposing factors
Age (in years)	60 (11.002)
Gender (male)	205 (49.6%)
Education (in years)	14.48 (3.39)
Religiosity
Secular	314 (77%)
Traditional	75 (18.4%)
Orthodox	19 (4.7 %)
Needs factors
Subjective memory	2.72 (0.93)
Cognitive status (range = 1-23)	2.96 (3.57)
Number of diseases (range = 0-17)	2.03 (2.14)
IADL difficulties (range = 1-5)	1.81 (0.45)
ADL difficulties (range = 1-5)	1.05 (0.21)
Enabling factors
Married	329 (85%)
Number of children	2.66 (1.21)
Social support (range = 1-5)	4.24 (0.76)
Financial status (range = 1-5)	3.48 (0.74)
Psychosocial factors
Negative attitudes toward NHs (range = 1-5)	3.19 (1.11)
Positive attitudes toward NHs (range = 1-5)	3.52 (0.71)
Familiarity with NHs (range = 1-3)	4.45 (0.76)
Familiarity with becoming disabled (familiar)	66 (16.1%)
Perceived risk of becoming disabled (range = 1-5)	2.77 (1.05)
Fear and worry of becoming disabled (range = 1-5)	3.25 (1.24)
Familiarity with AD (familiar)	61 (15%)
Perceived risk of becoming sick with AD	2.86 (0.86)
Fear and worry of becoming sick with AD	3.25 (1.24)

Note. IADL = instrumental activities of daily living; ADL = activities of daily living; NH = nursing home; AD = Alzheimer’s disease.

Analyses were performed on 408 participants who provided responses for all relevant questions. We tested for possible differences between respondents who completed and did not complete all the questions. We found that respondents who did not complete all the questions were slightly older (60 vs. 63 years of age), slightly less educated (13.08 years of education vs. 14.48 years of education), and a greater proportion of them were not married (40% vs. 19.6%).

Measures

Dependent variable: Care preferences

Participants’ care preferences were assessed in relation to two hypothetical and permanent situations: becoming permanently physically disabled and being diagnosed with AD. Similar to the study of McCormick and colleagues (2002), participants were asked to imagine themselves as being diagnosed with AD, or in a situation that required long-term care (according to the explanation provided, such situation might be a severe physical disability that makes everyday functioning very difficult, and stroke was offered as an example). For each of these hypothetical situations, participants were asked to choose the type of care they would be willing to receive. The types of care presented to them included the following: (a) being cared for at home by family members, (b) being cared for at home by paid caregivers, (c) being cared for in sheltered housing by caregivers in a nursing ward, and (d) being cared for in a NH by professional caregivers. Participants’ responses were exclusive and, for analysis purposes, were recoded into two variables: being willing to receive institutional care (combining care in sheltered housing/a nursing home) versus being willing to receive home care (combining care by family members/paid caregivers) in each situation.

Independent variables

These included Andersen’s extended model factors:

Predisposing factors

Participants were asked to state their age (in years), gender (male/female), number of years of education, and the level of their religiosity (with 1 indicating “secular,” 2 referring to “traditional,” and 3 being “orthodox”).

Needs factors

Cognitive status

Cognitive status was assessed using the Short Blessed Test (Kane & Kane, 2000). Participants were asked to name the current year, month, and time; to recall an address; to count from 1 to 20; and to recite the months of the year backward. The interviewer counted the number of mistakes the participant made (0-28). Hence, a higher score of this measure reflected decreased cognitive functioning.

Subjective memory problems

Participants were asked to assess their memory from 1 = very bad to 6 = excellent.

Number of diseases

Participants were asked whether (yes/no) their doctor had told them they suffer from various different health problems (a list of 17 problems), including high blood pressure; diabetes; heart attack; other cardiac problems; problems with the blood vessels, respiratory system, kidneys and/or the urinary tract, digestive system, and/or gall bladder; osteoporosis; stroke; rheumatism; depression; Parkinson’s disease; cancer; chronic pain; AD; or mental distress that was difficult to cope with alone. An overall score was calculated as the sum number of health problems that the participant reported suffering from.

Activities of daily living (ADL)

ADL were assessed using Katz, Down, Cash, and Grotz’s (1970) scale. Participants were asked to rate the degree of difficulty they have performing eight different activities, including moving around the house, showering, brushing hair/teeth or washing face, dressing up, putting shoes on/cutting nails, eating, moving from the bed to a chair, and using the bathroom. Each item was rated from 1 = no difficulty performing this activity to 5 = I am not able to perform this activity. A composite index for the degree of difficulty in ADL, ranging from 1 to 5, was calculated by averaging the items. Cronbach’s alpha reliability for this measure was .86.

Instrumental activities of daily living (IADL)

IADL were assessed using Fillenbaum’s (1985) scale. Participants were asked to rate the difficulty they have, from 1 = no difficulty to 5 = I am not able to perform this activity, performing eight different instrumental activities, including preparing meals, shopping, paying bills/running errands at the bank or post office, doing easy housework (making the bed, washing dishes), doing more difficult housework (washing floors, cleaning windows), going alone to the nearest bus stop, doing laundry, and using the phone. A total score was calculated by averaging the items, such that the scale ranged from 1 to 5 on IADL difficulties. The Cronbach’s alpha reliability score for this measure was .90.

Enabling factors

Social support

Social support was assessed using the Multidimensional Scale of Perceived Social Support (MSPSS; Zimet, Dahlem, Zimet, & Farley, 1988), which is a 12-item measure comprising three facets—support from friends, support from family, and support from significant others. There are four items per subscale, each with response options ranging from 1 (very strongly disagree) to 5 (very strongly agree). Sample statements included “I can count on my friends when problems arise,” “there is someone close to my heart who I can share my sorrows and joys with,” and “my family is willing to help me make decisions.” An overall index of satisfaction with social support was calculated by averaging all the items (range = 1-5). Cronbach’s alpha coefficient for the overall scale was .92.

Other enabling factors we accounted for were marital status (currently married/unmarried), number of children, and perceived financial status (rated from 1 = very bad to 5 = very good).

Psychosocial factors

Reasons for preferred type of care

In each situation, participants were asked to state to what extent (on a scale from 1 = not at all to 5 = to a great extent) their willingness was based on (a) their concern about losing independence, (b) their concern about being a burden to family members, and (c) financial concerns. We based these items on Keysor, Desai, and Mutran’s (1999) and Forbes and Hoffart’s (1998) findings.

Attitudes toward NHs

Participants’ perceptions about NHs were assessed using 11 items exploring different aspects of NH care, such as privacy, quality of the staff, and NH costs. The items were selected based on a review of the empirical research available (Guan et al., 2007; Schoenberg & Coward, 1997). Sample items included the following: “It is good to live in a nursing home because everyone is from the same age group”; “A nursing home is full of sick and strange people”; and “To reside in a nursing home involves high costs.” Each item was rated on a 5-point Likert-type scale, ranging from 1 = strongly disagree to 5 = strongly agree.

A factor analysis using VARIMAX rotation was performed, revealing two separate factors with eigenvalues of 1 and above. The first factor reflected negative attitudes toward NHs (their direction was reversed prior to the factor analysis) and explained 32.8% of the variance. The second factor reflected positive attitudes toward NHs and explained an additional 13.4% of the variance. Following these results, two overall indices (ranging from 1 to 5, with 5 indicating more negative attitudes in the negative attitude subscale and more positive attitudes in the positive attitude subscale) were calculated by averaging the items in each factor. One item referring to the high costs of living in a NH was dropped, due to very low correlation with the other items. It may be that the cost of living in a NH is not perceived as one of the purely negative aspects, as high costs might promise better quality, or it is not perceived as badly as the other items (referring to low level of privacy, being around old people, etc.). Hence, the positive and negative scales were comprised of five items each. The internal reliability of the factors was modest for the negative attitudes index (Cronbach’s alpha = .63), and good for the positive attitudes index (Cronbach’s alpha = .81).

Perceptions of risk and worry

These included beliefs about risk, and worry and fear of becoming permanently disabled and of becoming sick with AD (Werner, Goldberg, Mandel, & Korczyn, 2013).

(a) Risk was assessed by a single question for each hypothetical situation—“How likely do you think it is that you will become disabled/will develop AD?” Each item was rated on a 5-point Likert-type scale, ranging from 1 = not at all likely to 5 = very likely. (b) Worry and fear were assessed by two questions for each hypothetical situation—“How much do you worry that you will become disabled/sick with AD?” and “How much do you fear becoming disabled/sick with AD?” Each item was rated on a 5-point Likert-type scale, ranging from 1 = not at all to 5 = very much. Because the correlations between these two items were very high for AD (r = .87, p < .001) as well as for disability (r = .82, p < .001), the mean of the two items was calculated and used as an overall score for worry and fear. For becoming disabled, this measure had a Cronbach’ alpha reliability of .89 and for becoming sick with AD, .93.

Familiarity with NHs

Participants were asked three yes/no questions to assess their familiarity with NHs. These included (a) whether they had ever visited a NH, (b) whether they had ever been institutionalized in a NH, and (c) whether they had friends or family members who resided in a NH. The sum of participants’ responses on these three items was calculated and used as a measure of familiarity with NHs.

Familiarity with being disabled and with AD

Participants were asked whether they had a family member who is disabled and whether they had a family member with AD (response type: yes/no). Thus, one item represented familiarity with being disabled, and one item represented familiarity with AD.

Procedure

Face-to-face interviews were conducted by trained interviewers, using a structured interview. All participants were recruited opportunistically from public places, such as university campus, places of work, and through personal contacts. The interviewers first briefly described the study’s topic and the purpose of the research. Prior to starting the process, the interviewers guaranteed participants’ anonymity and confidentiality, reminded them that participation was voluntary, and assured them that nonresponse on an item, as well as withdrawal, was acceptable. Finally, after granting consent, the interview process began. This procedure was approved by the Ethics Committee of the Faculty of Social Welfare and Health Sciences at the University of Haifa.

Statistical Analyses

Data were cleaned, coded, and analyzed using SPSS Version 21.0 (SPSS Inc., Chicago, IL, 2012). Descriptive statistics were used to describe the sample and the main variables. Due to the high correlation between ADL and IADL measures (r = .80), in the following analyses, only IADL was used. Statistical comparisons between the two hypothetical situations (being diagnosed with AD and permanent disability) were performed using McNemar and t tests. To address the issue of multiple comparisons (four comparisons using t tests and four using McNemar), we set a stricter p value to signify significance by dividing the acceptable p = .05 by 4 (Bonferroni correction). Accordingly, only when a p value was equal to, or smaller than, .0125 (.05/4 = .0125), results were considered significant. Last, logistic regressions contrasting those who preferred institutional care to those who preferred home care in each of the hypothetical situations were carried out. Because our sample size was not sufficient to test the high number of variables we were interested at, we tested each group of variables separately as an initial stage. We then constructed models consisting of those independent variables that were found to be significantly associated with our outcome in the previous step.

Results

Participants’ Care Preferences

Table 2 displays the percentage of participants reporting the different care preferences in the hypothetical situation of becoming permanently physically disabled, in comparison with the situation of becoming sick with AD. As can be observed, overall, and regardless of the hypothetical situation, a higher proportion of participants expressed their preference to be taken care of at home by paid caregivers than any other care arrangement. In addition, a statistically significant higher percentage of participants preferred institutional care (sheltered housing or a NH) in the hypothetical situation of becoming sick with AD (16.5% for sheltered housing and 15.0% for NH) in comparison with the situation of becoming permanently disabled (11.8% for sheltered housing and 6.1% for NH; p < .01 for sheltered housing and p < .001 for NH). In contrast, a significantly higher proportion of participants preferred to be cared for at home by a family member in the case of becoming permanently disabled (37.7%), compared with becoming sick with AD (25.2%; p < .001).

Table 2.

Comparisons of Preferred Type of Care and Basis of This Decision in the Hypothetical Situation That the Participant Became Disabled Versus the Situation That the Participant Became Sick With Alzheimer’s Disease.

	Hypothetical situations
	Permanently disabled	Sick with AD
Preferred type of care (in percentages)			McNemar’s test
At home by family*	154 (37.7%)	103 (25.2%)	p < .001
At home by paid caregivers	181 (44.4%)	175 (42.9%)	n.s.
By caregivers in sheltered housing*	48 (11.8%)	69 (16.9%)	p < .01
By caregivers in a nursing home*	25 (6.1%)	61 (15.0%)	p < .001
Basis of decision for choosing type of care (M and SD)			t test
Worrying about losing independence	3.99 (1.41)	3.91 (1.50)	t(407) = 1.51, n.s.
Worrying about the burden of care on family	4.10 (1.34)	4.19 (1.27)	t(407) = −1.85, n.s.
Worrying about financial issues	2.81 (1.58)	2.81 (1.61)	t(407) = 0.34, n.s.

Note. Results are considered significant when p < .0125. AD = Alzheimer’s disease.

Significant differences at a p < .0125 level.

When asked participants to rate the importance of each consideration to their choice of care. Participants expressed greater concerns about losing independence and being a burden on the family in both situations, whereas financial worries were relatively less pronounced. A one-way ANOVA with Tukey’s post hoc revealed that for a permanent disability situation, respondents gave similar weight to the independence and burden on the family considerations and less weight to financial issues, in comparison both with independence-related worries and burden-related worries (M difference 1.18, p < .001 and 1.29, p < .001, respectively). With regard to the hypothetical situation of becoming sick with AD, respondents indicated that becoming a burden on the family was the most important concern, and it was given a significantly (although only slightly) greater weight than independence (M difference 0.34, p < .05) and significantly greater weight than financial considerations (M difference 1.51, p < .001; Table 2).

Correlates of Participants’ Care Preferences

Logistic regressions were conducted to examine factors associated with a preference for institutional care over home care in each hypothetical situation (Table 3). The results indicated that in the hypothetical situation of becoming physically disabled, preference toward institutional care was positively associated with greater number of illnesses, more positive attitudes toward NHs, and basing the decision on the concern of burdening the family. Participants who based their decision on the concern of losing their independence were less likely to prefer institutional care, and so were participants with better financial status. When all these variables were included in the same model, only positive attitudes toward NHs, fear of losing one’s independence, and concern about burdening the family remained significant at the p < .01 level. The reported number of illnesses and financial situation became insignificant.

Table 3.

Logistic Regressions Predicting Preferred Care (Institutional Care vs. Home Care) in Two Hypothetical Situations (N = 408).

	Preferred type of care: Institutional care vs. home care (home care is the reference)
Predictors	If became permanently disabledEstimate (Wald 95% CI)	If became sick with ADEstimate (Wald 95% CI)
Predisposing factors
Intercept	−3.09** [−5.32, −0.86]	−3.33* [−0.52, −1.41]
Age	0.01 [−0.01, 0.03]	0.01 [−0.006, 0.03]
Female gender	−0.17 [−0.69, 0.34]	0.21 [−0.21, 0.64]
Education	0.07 [−0.008, 0.15]	0.11** [0.04, 0.18]
Religiosity
Tradition vs. secular	0.100 [−0.57, 0.77]	0.22 [−0.32, 0.77]
Orthodox vs. secular	−0.49 [−2.0, 1.01]	−1.17 [−0.26, 0.33]
Needs factors
Intercept	−1.13* [−2.16, −0.11]	−0.11* [−1.96, −0.26]
Subjective memory	−0.09 [−0.40, 0.21]	0.19 [−0.06, 0.45]
Cognitive status	−0.03 [−0.12, 0.05]	−0.11** [−0.18, −0.04]
Number of diseases	0.16* [0.02, 0.29]	0.16** [0.04, 0.27]
IADL difficulties	−0.33 [−1.03, 0.38]	−0.18 [−0.74, 0.37]
Enabling factors
Intercept	0.10 [−1.56, 1.78]	0.28 [−1.17, 1.73]
Married	−0.06 [−0.71, 0.59]	0.04 [−0.51, 0.58]
Number of children	−0.003[−0.21, 0.20]	−0.11 [−0.29, 0.08]
Social support	0.13 [−0.22, 0.47]	−0.14 [−0.42, 0.13]
Financial status	−0.48** [−0.79, −0.17]	−0.05 [−0.34, 0.23]
Psychosocial factors
Intercept	−4.50** [−7.36, −1.87]	−3.97*** [−6.05, −1.89]
Negative attitudes toward NHs	−0.08 [−0.39, −0.18]	−0.06 [−0.28, 0.16]
Positive attitudes toward NHs	0.54* [0.12, 0.97]	0.35 [−0.003, 0.70]
Decision based on concern of losing independence	−0.22* [−0.41, −0.04]	−0.18* [−0.34, −0.01]
Decision based on concern of burdening the family	0.44* [0.18, 0.70]	0.55*** [0.31, 0.80]
Decision based on financial considerations	−0.16 [−0.34, 0.02]	−0.05 [−0.21, 0.09]
Familiarity with NHs	0.23 [−0.14, 0.61]	0.18 [−0.12, 0.49]
Familiarity with becoming disabled	−0.46 [−1.26, 0.34]	—
Perceived risk of becoming disabled	0.14 [−0.21, 0.49]	—
Worry and fear about becoming disabled	−0.06 [−0.29, 0.17]	—
Familiarity with AD	—	−0.75* [−0.14, −0.07]
Perceived risk of becoming sick with AD	—	0.44** [0.13, 0.76]
Worry and fear about becoming sick with AD	—	−0.02 [−0.22, 0.17]

p<.05, **p<.01, ***p<.001

Note. To avoid over-fitting, each group of factors was tested separately. CI = confidence interval; AD = Alzheimer’s disease; IADL = instrumental activities of daily living; NH = nursing home.

In the case of becoming sick with AD, institutional care was preferred among those with higher education and better cognitive status, as well as among those with greater number of illnesses. Furthermore, participants who reported basing their decision to a lesser degree on concerns about losing their independence and to a greater degree on the burden of care for their family had a significantly higher chance of preferring institutional care. Finally, familiarity with AD was negatively associated with preference for institutional care, whereas the perceived risk of becoming sick with AD was positively associated with preferences toward institutional care. When all the significant variables were included in the same model, they all remained significant, with the exception of cognitive status that became insignificant.

Discussion

The primary goal of this study was to examine the contribution of predisposing, enabling, needs, and psychosocial factors associated with Israeli adults’ preferences regarding institutional care versus home care. Our study focused on two hypothetical situations: becoming permanently physically disabled and being diagnosed with AD.

Overall, our results revealed that in both hypothetical situations, only a small percentage of the participants were interested in institutional care—6% if they were to become disabled and 15% if they were to become sick with AD. These percentages are considerably lower than those reported in McCormick et al.’s (2002) study and might reflect cultural and structural differences in both countries and in their family-state division, in terms of responsibility toward elderly persons (Daatland & Lowenstein, 2005). Although the state provides services for the disabled and offers caregiving assistance, the Israeli society is characterized by close contact with the extended family and high caregiving commitment of families with regard to older relatives (Lavee & Katz, 2003; Walter-Ginzburg, Guralnik, Blumstein, Gindin, & Baruch, 2001). In addition, as noted, the Israeli welfare policy encourages community care over institutional care. It is possible that these two mechanisms operate simultaneously to direct public attitudes and preferences toward home care and away from institutional care. The fact that institutional care operates at almost full capacity (92%), and the consequent worry about the potential loss of privacy and impersonal care, may be another factor adding to the public’s unwillingness to utilize institutional solutions.

Despite these differences, both studies showed that institutionalized care was preferred to a greater extent in the hypothetical situation of becoming sick with AD, in contrast with the situation of becoming physically disabled. These findings might be a reflection of the considerable fear, anxiety, and concern associated with developing AD (Cutler & Hodgson, 2001; French, Floyd, Wilkins, & Osato, 2011; Suhr & Kinkela, 2007; Werner, 2002), and with the burden imposed on family members in such situations (Etters, Goodall, & Harrison, 2008).

Our results also revealed that although there were some common factors associated with care preferences in each of the hypothetical situations, there were also distinct factors.

Most interestingly, we found that psychosocial factors were among the most important predictors of preferring institutionalized care in both situations. This is an important finding, as it stresses the importance of potentially modifiable predictors to the type of care preferences. The concern over losing one’s independence and the concern over burdening the family were both significantly associated with care preferences in both situations, indicating that the private set of concerns is the most prominent indicator of behavioral intention.

As for the differences noted in the factors associated with preferred care in each situation, most noticeable is the decreased willingness to choose institutional care in the hypothetical situation of becoming sick with AD associated with impaired cognitive status. It is possible that the intimidating experience of decreased cognitive status makes the AD situation more readily perceived and threatening, and hence related to individuals’ decreased willingness to leave familiar surroundings in favor of institutional care. Interestingly, however, subjective memory was not associated with care preferences.

Education was related to a greater willingness to choose institutional care in the AD situation but not in situations of physical disability. It is possible that more educated people have a greater awareness of the cost of AD, and hence, AD is perceived as a greater loss that requires institutional care. However, this was not reflected in any significant correlation between education and perceived risk or fear of becoming sick with AD. Interestingly, familiarity with AD was negatively associated with preferring institutional care, whereas the perceived risk of being diagnosed with AD was positively associated with AD. It seems that experiencing the course of AD increases the reluctance to lose the familiar environment, whereas believing that AD is more probable increases the willingness to choose what is usually a less favored treatment trajectory.

We did not find gender, marital status, or the number of children to be significant correlates of willingness to enter a NH in either situation, as previous studies have (Chen, 2011; Kim & Choi, 2008; Kim & Kim, 2004; McCormick et al., 2002; Min, 2005). These gaps between this study and previous studies can be explained by our unique sample. It may be that gender differences are more pronounced in older populations—when women are more often left without a spouse. The lack of association between the number of children and care preferences may be a result of the differences between having at least one child to having none. In our sample, only five respondents had no children; hence, any comparison between these states is unfeasible.

Theoretical and Practical Implications

This study offers several theoretical and practical contributions. Theoretically, with regard to the Extended Andersen Model, our results lend support to the importance of psychosocial factors—most importantly, the motivations for care preferences. Perhaps due to the younger age of our participants, in comparison with other studies assessing the topic (Dubois et al., 2008; Jang et al., 2008; Kim & Choi, 2008; Kim & Kim, 2004; Min, 2005; Nieboer et al., 2010), the predisposing factors and health-needs factors played a relatively minor role in their care preferences. Besides its theoretical importance, the fact that psychosocial factors, rather than other factors, were the most prominent correlates to institutional care preference is of great practical importance. Decisions regarding future care (such as purchasing long-term care insurance) are mostly made during adulthood—before the actual need arises. Hence, campaigns aimed at increasing planning for future health care needs can benefit from targeting these psychological factors to try and shape LTC decision making.

The Israeli welfare system encourages the provision of LTC at home, which seems to corroborate with the preferences of the population. However, considering the aging of the population and the consequent increase in LTC needs, future efforts might benefit from targeting populations at risk for institutional care to help them overcome their resistance and change their attitudes to ease the transition.

Study Limitations

This study has several limitations. First, the sample was relatively modest in size (N = 408) and was not randomly selected and thus may have been subjected to selection bias. Second, participants were asked to estimate their willingness to enter a NH in hypothetical situations. This may not represent their actual willingness were they to encounter such circumstances

Third, because the hypothetical situations examined were defined in general terms (being diagnosed with AD and permanent disability), we cannot rule out the possibility that participants had imagined themselves in different situations. For example, one participant could imagine himself or herself in the early stages of AD, whereas another one could have imagined himself or herself in more progressive stages of AD. Thus, results should be interpreted cautiously, although it must be noted that a similar methodology was used in most studies assessing the topic (McCormick et al., 2002). Also, we believe that if we would have used a more detailed vignette describing advanced AD, the differences between the situations would have been greater. Thus, this limitation strengthens the probability that our findings regarding the differences err toward the conservative, and not vice versa.

Fourth, the relatively young sample used in this study may not be aware of the toll AD or permanent disability take, and thus could be more reluctant than older adults to enter a NH. Fifth, the survey method used did not allow respondents to explain their views and expand on their preferences. This is especially critical, given the importance and complexity of the topic. Finally, there is evidence that the older people themselves are often not included in the decision to move to a NH, as this decision is more often made by their children (Lai, Tse, & Lau, 2008). Hence, their willingness may not be the ultimate determinant of whether they are indeed institutionalized in a NH (especially, in the case of AD, when cognitive abilities may be compromised). The fact that our study included younger participants might help in overcoming this limitation, because younger adults are the ones who often make the decisions for their parents. However, we did not ask our participants to evaluate their future decisions with regard to their parents. Future research may expand the benefits of our sample by estimating this model among caregivers considering long-term care solutions for their older relatives, or among those who were recently diagnosed as suffering from AD or other terminal conditions and are considering the preferred care trajectory for the coming years. It is important to map the factors that are associated with future planning for a hypothetical situation and compare these with the factors playing a role in the decision process in a real disability situation.

Despite these limitations, the present study addressed a complex and understudied topic—the public’s preferences for entering a NH. In light of the demographic changes predicted for the future, increasing numbers of people worldwide can be expected to face important dilemmas regarding this topic. In addition, policymakers will be confronted with difficult decisions when trying to establish a legitimate and fair process of priority setting for the allocation of health resources. It is obvious that research about lay persons’ preferences about these issues could be helpful in making these difficult decisions.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Perla Werner is a Full Professor at the University of Haifa. She has been involved in gerontological research since 1978 and has contributed to the expansion of knowledge in the area of psycho-social aspects of Alzheimer’s disease and other dementing disorders. She has published over 200 papers in peer-review journals and has received numerous grants from prestigious sources such as the National Institute of Aging, the Israeli Science Foundation, the Alzheimer’s Association and more.Prof. Werner served as the Head of the Department of Gerontology between 2004 – 2006 and as the Dean of the Faculty of Social Welfare and Health Sciences between 2006 – 2010. She is currently the Vice Rector and head of the Center for the Research and Study of Aging at the University of Haifa.

Dikla Segel-Karpas is a visiting Fischbach post-doctoral fellow at the Department of Psychology, Brandeis University, Waltham, MA, USA. Before she was a post-doctoral fellow and a center coordinator at the Center for the Research and Study of Aging at the University of Haifa, Israel. She completed her PhD at the Faculty of Industrial Engineering and Management at the Technion - Israel Institute of Technology. Her main fields of interest are older workers and retirement, cognitive changes in older adulthood and old age, and the relationship between social and personality factors in adjustment to aging related changes.

References

Andersen

Newman

J. F.

(1973). Societal and individual determinants of medical care utilization in the United States. The Milbank Memorial Fund Quarterly: Health and Society, 51, 95-124.

Andersen

R. M.

(1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36, 1-10.

Bardo

A. R.

Applebaum

R. A.

Kunkel

S. R.

Carpio

E. A.

(2014). Everyone’s talking about it, but does it work? Nursing home diversion and transition. Journal of Applied Gerontology, 33, 207-226. doi:10.1177/0733464813505702

Be’er

(2004). A national census of residents of long-term care institutions, 2000, and trends in the institutionalization of older persons (Research Report 04-429). Jerusalem, Israel: Myers-JDC-Brookdale Institute.

Brodsky

Shnoor

Be’er

. (Eds.). (2012). The elderly in Israel—The 2011 statistical abstract (Catalogue No. M-70-12). Jerusalem, Israel: Myers-JDC-Brookdale Institute.

Chen

(2011). Knowledge and perceptions: Chinese older adults’ willingness to use institutional elder care (Gerontology Theses, Paper 22). Retrieved from http://digitalarchive.gsu.edu/gerontology_theses/22/

Clarfield

A. M.

Rosenberg

Brodsky

Bentur

(2010). Healthy aging around the world: Israel too? In Carmel

(Ed.), Aging in Israel: Research, policy and practice (Vol. 13) (pp.203-212). Transaction Publishers.

Cutler

S. J.

Hodgson

L. G.

(2001). Correlates of personal concerns about developing Alzheimer’s disease among middle-aged persons. American Journal of Alzheimer’s Disease and Other Dementing Disorders, 18, 9-20.

Daatland

S. O.

Lowenstein

(2005). Intergenerational solidarity and the family-welfare state balance. European Journal of Ageing, 2, 174-182.

10.

Dubois

M. F.

Dubuc

Raîche

Caron

C. D.

Hébért

(2008). Correlates of the perceived need for institutionalization among older adults without cognitive impairment. Gerontology, 54, 244-251.

11.

Etters

Goodall

Harrison

B. E.

(2008). Caregiver burden among dementia patient caregivers: A review of the literature. Journal of the American Academy of Nursing Practitioners, 20, 423-428.

12.

Fillenbaum

G. G.

(1985). Screening the elderly: A brief instrumental activities of daily living measure. Journal of the American Geriatric Society, 33, 698-706.

13.

Forbes

Hoffart

(1998). Elders’ decision making regarding the use of long-term care services: A precarious balance. Qualitative Health Research, 8, 736-750.

14.

French

S. L.

Floyd

Wilkins

Osato

(2011). The fear of Alzheimer’s Disease Scale: A new measure designed to assess anticipatory dementia in older adults. International Journal of Geriatric Psychiatry, 27, 521-528.

15.

Greenfield

E. A.

(2014). Community aging initiatives and social capital developing theories of change in the context of NORC Supportive Service Programs. Journal of Applied Gerontology, 33, 227-250. doi:10.1177/0733464813497994

16.

Guan

Zhan

H. J.

Liu

(2007). Institutional and individual autonomy: Investigating predictors of attitudes toward institutional care in China. International Journal of Aging & Human Development, 64, 83-107.

17.

Israeli Bureau of Statistics, retrieved from: http://www.cbs.gov.il/reader on March 2014

18.

Jang

Kim

Chiriboga

D. A.

Cho

(2008). Willingness to use a nursing home: A study of Korean American elders. Journal of Applied Gerontology, 27, 110-117.

19.

Kane

R. L.

Kane

R. A.

(2000). Assessing older persons: Measures, meaning, and practical applications. New York, NY: Oxford University Press.

20.

Katz

Down

T. D.

Cash

H. R.

Grotz

R. C.

(1970). Progress in the development of the index of ADL. The Gerontologist, 10, 20-30.

21.

Katz

P. R.

(2011). An international perspective on long term care: focus on nursing homes. Journal of the American Medical Directors Association, 12(7), 487-492.

22.

Keysor

J. J.

Desai

Mutran

E. J.

(1999). Elders’ preferences for care setting in short- and long-term disability scenarios. The Gerontologist, 39, 334-344.

23.

Kim

E. Y.

Kim

C. Y.

(2004). Who wants to enter a long-term care facility in a rapidly aging non-Western society? Attitudes of older Koreans toward long-term care facilities. Journal of the American Geriatric Society, 52, 2114-2119.

24.

Kim

Choi

W. Y.

(2008). Willingness to use formal long-term care services by Korean elders and their primary caregivers. Journal of Aging & Social Policy, 20, 474-492.

25.

Lai

C. K. Y.

Tse

M. M. Y.

Lau

L. K. P.

(2008). Placement appropriateness for seniors into long-term care—A neglected area of research. Asian Journal of Gerontology and Geriatrics, 3, 34-39.

26.

Lakdawalla

Goldman

D. P.

Bhattacharya

Hurd

M. D.

Joyce

G. F.

Panis

C. W.

(2003). Forecasting the nursing home population. Medical Care, 41, 8-20.

27.

Lavee

Katz

(2003). The family in Israel: Between tradition and modernity. Marriage & Family Review, 35, 193-217.

28.

Lee

Kovner

C. T.

Mezey

M. D.

I. S.

(2001). Factors influencing long-term home care utilization by the older population: Implications for targeting. Public Health Nursing, 18, 443-449.

29.

Lehning

A. J.

Smith

R. J.

Dunkle

R. E.

(2013). Do age friendly characteristics influence the expectation to age in place? A comparison of low-income and higher income Detroit elders. Journal of Applied Gerontology. Advance online publication. doi: 0733464813483210

30.

Luppa

Luck

Weyerer

König

H. H.

Brähler

Riedel-Heller

S. G.

(2010). Prediction of institutionalization in the elderly. A systematic review. Age and ageing, 31, 31-38.

31.

McCormick

W. C.

Ohata

C. Y.

Uomoto

Young

H. M.

Graves

A. B.

Kukull

. . .Larson

E. B.

(2002). Similarities and differences in attitudes toward long-term care between Japanese Americans and Caucasian Americans. Journal of the American Geriatric Society, 50, 1149-1155.

32.

Merrill

Hunt

M. E.

(1990). Aging in place: A dilemma for retirement housing administrators. Journal of Applied Gerontology, 9, 60-76.

33.

Min

J. W.

(2005). Preference for long-term care arrangement and its correlates for older Korean Americans. Journal of Aging and Health, 17, 363-395.

34.

Nieboer

A. P.

Koolman

Stolk

E. A.

(2010). Preferences for long-term care services: Willingness to pay estimates derived from a discrete choice experiment. Social Science & Medicine, 70, 1317-1325.

35.

Schoenberg

N. E.

Coward

R. T.

(1997). Attitudes about entering a nursing home: Comparisons of older rural and urban African-American women. Journal of Aging Studies, 11(1), 27-47.

36.

Suhr

J. A.

Kinkela

J. H.

(2007). Perceived threat of Alzheimer’s disease: The role of personal experience with AD. Alzheimer’s Disease and Associated Disorders, 21, 3, 225-231.

37.

Tse

M. M. Y.

(2007). Nursing home placement: Perspectives of community-dwelling older persons. Journal of Clinical Nursing, 16, 911-917.

38.

Vincent

G. K.

Velkoff

V. A.

(2010). The next four decades: The older population in the United States: 2010 to 2050 (No. 1138). Washington, DC: U.S. Department of Commerce, Economics, and Statistics Administration, U.S. Census Bureau.

39.

Walter-Ginzburg

Guralnik

J. M.

Blumstein

Gindin

Baruch

(2001). Assistance with personal care activities among the old-old in Israel: A National Epidemiological Study. Journal of the American Geriatrics Society, 49, 1176-1184.

40.

Werner

(2002). Assessing correlates of concern about developing Alzheimer’s dementia among adults with no family history of the disease. American Journal of Alzheimer’s Disease and Other Dementing Disorders, 17, 331-337.

41.

Werner

Goldberg

Mandel

Korczyn

A. D.

(2013). Gender differences in lay persons’ beliefs and knowledge about Alzheimer’s disease (AD): A national representative study of Israeli adults. Archives of Gerontology and Geriatrics, 56, 400-404.

42.

Zimet

G. D.

Dahlem

N. W.

Zimet

S. G.

Farley

G. K.

(1988). The multidimensional scale of perceived social support. Journal of Personality Assessment, 52, 30-41.