| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Department of Psychiatry, University of California, San Francisco
b Department of Family and Community Medicine, University of California, San Francisco
Correspondence: Morton A. Lieberman, PhD, Center for Health & Community, 3333 California Avenue, U.C. San Francisco, San Francisco, CA 94143. E-mail: mal{at}itsa.ucsf.edu.
Decision Editor: Eleanor S. McConnell, RN, PhD
 |
Abstract |
|---|
 TOP Abstract BackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
Key Words: Alzheimer's disease • Dementia • Placement • Caregiver burden
The considerable burdens of caring for an elder with Alzheimer's disease (AD) have been well documented (Schulz, Visintainer, and Williamson 1990
). The variety of programs (day care services, respite care, support groups) developed to mitigate such burdens are based on the assumption that the longer a family can manage the patient in the community without undue burden, the better the patient care and the longer the delay of nursing home placement (NHP). However, because dementia is a progressively debilitating disease, at some point in the course of the disease many health care professionals suggest that the family consider NHP. The assumption by both professionals and caregivers alike is that NHP will relieve family caregivers from many of the stressful and repetitive instrumental and emotional burdens of caring for a demented elder in the community.
Although there is general agreement about the advisability of NHP at some point during the course of the disease, surprisingly little is known about the subsequent effects of NHP on caregiver health and well-being. Does caregiver health and well-being improve after NHP, and, if so, are these improvements uniform among family members and across families?
Knowing more about the effects of NHP can help families consider this very difficult question in two important ways. First, in most cases, family members need to incorporate into their deliberations appropriate expectations about what will occur after NHP. If certain factors are not expected to change after NHP, families might alter their placement decision. Second, it is important for health care professionals to prepare family members for the instrumental and emotional changes that may or may not occur following NHP. Information provided by health care professionals prior to NHP can reduce later personal disappointment or surprise, and may mitigate the emotional sequella associated with "second guessing" the decision after NHP has taken place.
Despite the importance of NHP to families caring for a patient with dementia, there is only a limited empirical base to guide health care professionals and families through the decision-making process. This study addressed one aspect of this problem by evaluating family caregiver health and well-being over a 2-year period following NHP.
|
Background |
|---|
|
TOPAbstractBackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
found few differences in stress levels between spouses caring for patients with AD at home and those who had placed their ill elders in nursing homes. Ritchie and Ledesert 1992 reported that one year after NHP, one third of family caregivers expressed considerable distress, regardless of level of patient dementia, other associated patient pathology, and recency of admission. Zarit and Whitlatch 1992 reported that relatives who placed their ill elder in a nursing home experienced lower instrumental overload and tension than relatives who decided against NHP, but other indicators of stress, such as level of depression, did not change. Other reports (Novak and Guest 1992) have indicated that the caregiver's relationship to the patient qualified the linkages between NHP and changes in caregiver health and well-being, such that spouses generally experienced the greatest demoralization, compared to adult offspring and other family members.
With respect to longitudinal research, a study of a large community sample (Aneshensel and Pearlin 1993) found that NHP reduced family caregiver experience of being bound to the caregiver role (role captivity). King and Collins 1991 studied family caregivers' sense of general well-being and depression before and after NHP. Health and financial stresses were reduced following NHP for spouses, but stress increased for the patient's adult offspring. However, Pushkar, Gold, and Reis 1995 reported that caregivers' health and quality of life improved in families that decided on NHP, compared to those who kept the patient at home.
In general, both cross-sectional and longitudinal studies suggest similar findings. Studies focusing on caregiver instrumental burden show notable improvement following NHP. In contrast, studies focusing on caregiver health, well-being, depression, and life satisfaction indicate little change following NHP, although there is some variability of findings from study to study. It is not clear, however, whether the findings from both cross-sectional and longitudinal studies can be easily generalized to the large population of families that care for patients with dementia, or whether qualifications based on family caregiver gender, social class, position in the family, and such need to be taken into consideration. It may be that the lack of change in caregiver health and well-being following NHP may be accounted for, at least in part, by other caregiver, patient, and family factors that are not generally included in published research.
|
Factors That May Influence Change Following NHP |
|---|
|
TOPAbstractBackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
), amount of family help provided (Lieberman and Fisher 1999a), effect of caregiving on second-generation marriages (Lieberman and Fisher 1999b), utilization of community-based professional services (Lieberman and Fisher 1998), and decisions to institutionalize (Fisher and Lieberman 1999). The three groups of factors are: (1) demographic characteristics of the patient and caregivers, including age, relation to the patient, gender, income, and education (Branch and Jette 1982; Christenson and Moore 1994; Hogan, Thierer, Ebley, and Parhad 1994; Pruchno, Michaels, and Potashnik 1990); (2) severity of patient functional deficit (Bianchetti, Scuratti, Zanetti, and Binetti 1995; Wackerbarth 1996); and (3) characteristics of the family caregiving setting, including level of caregiver stress and burden (Collins, King, and Kokinakis 1994; McCoy and Edwards 1981; Pruchno et al. 1990) and availability of community resources (Cohen, Gold, Shulman, and Wortley 1993; Scott, Edwards, Davis, Cornman, and Macera 1997). These three groups of factors provide a context for observing variations in the effects of NHP on family members.
The central question of this research asked if NHP leads to improved family caregiver health and well-being over time. We studied the effects of institutionalization on several family members, not only the primary caregiver, who in this study was usually the spouse. We had previously shown that all family members' health and well-being were affected by the illness, that the marriages of adult offspring and in-laws were affected, and that the amount of formal and informal help utilized by the family was significantly related to family structure and dynamics (Fisher and Lieberman 1996; Lieberman and Fisher 1995a, Lieberman and Fisher 1998, Lieberman and Fisher 1999a).
We explored how the three groups of factors (characteristics of the family caregiving setting, demographic characteristics of the patient and the caregivers, severity of patient functional deficit) qualified the relationship between NHP and changes in family member health and well-being over time. For example, among characteristics of the caregiving setting, we expected that families that used professional services prior to NHP would display greater improvements in caregiver health and well-being following NHP than families that did not, because the former group would have more comfort and experience relying on external sources for caregiving than the latter group (Lieberman and Fisher 1999a). Among demographic characteristics, we expected that spouses would display less improvement in health and well-being following NHP than adult offspring and other family members because of their more intimate emotional involvement with the patient over time. Finally, we expected that caregivers of patients with high functional deficit would experience a greater improvement in health and well-being following NHP than caregivers of other patients because of the differences in associated caregiver burden.
|
Methods |
|---|
|
TOPAbstractBackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
Inclusion criteria for the study were: National Institute of Neurological and Communicative Disorders and Stroke diagnostic criteria for probable or possible Alzheimer's disease; at least one adult offspring residing in California; and the patient living in the community, either with a spouse, offspring, other relative, or alone. All patients and families meeting these criteria were referred by their respective clinic to the research team.
Of the 533 families referred, 198 did not fully meet inclusion criteria; of the remaining 335 families, 100 refused to participate, yielding a 70% acceptance rate for contacted families. The final sample consisted of 235 families, which included 89 spouses, 184 offspring, and 136 in-laws (married to offspring). The sample self-identified as 87.8% Caucasian and 12.1% ethnic minorities (Hispanic, African American, Asian/Pacific Islander). Using the centers' shared, computerized database, comparisons between those families that refused participation and those that agreed to participate yielded no significant differences in center location, reasons for seeking evaluation, patient gender, ethnicity, marital status, living arrangements, income, education, number of primary and secondary caregivers, number of problems, degree of cognitive deficit, level of dementia, and decisions to institutionalize within 2 years of evaluation. The sample appeared representative of the population served by the centers.
Patients were classified into two groups: those who were institutionalized within 6 months of clinic assessment (n = 47) and those who were not institutionalized for 2 years following clinic assessment (n = 135). Cases were omitted if the patient died within 2 years of evaluation (n = 13). To provide for sufficient post-NHP time to assess changes in caregiver health and well-being, cases also were omitted if the patient was institutionalized at the 12-, 18-, or 24-month follow-up (n = 40). Table 1 and Table 2 show patient and family member characteristics, respectively, for NHP and non-NHP cases. There was only one significant difference between patients and between family members in the NHP and the non-NHP groups among the variables listed in both Table 1 and Table 2 , a finding that could have resulted by chance, given the number of comparisons run: Family members in the NHP group reported significantly more patient troubling behaviors than family members in the non-NHP group (t = 3.13, p = .02).
|
|
Outcome Measures
Our primary goal was to evaluate the effect of NHP on family caregiver health and well-being over time. Three measures of health and well-being were included. First, a scale of common physical symptoms, derived from the Rand MOS (Ware et al. 1984), was used to assess common somatic complaints (headache, backache) that occurred regularly over the last 3 months (21 items, alpha = .93). Second, the Hopkins Symptom Checklist (Derogatis 1974), modified by Pearlin and Lieberman 1978, was used to assess anxiety and depression (23 items, alpha = .92). Third, the Well-Being Scale (Bradburn 1969) was used to assess general well-being (11 items, alpha =.94). Each of these scales was used in the multiple regression analyses reported below. In addition, we also constructed a single composite score for use in the mixed model analysis so that we could keep the number of comparisons to a minimum.
Moderator Variables
Demographic Characteristics.
Caregiver gender, education (number of years of formal schooling), relationship to the patient (spouse, offspring, in-law), and the number of caregiving hours per week were assessed. Income was not included because of its high correlation with education (r = .66).
Severity of Patient Deficit.
Six measures were included, each assessed by the clinics as part of the initial evaluation: Level of cognitive disturbance was assessed by the Mini-Mental State Exam (Folstein, Folstein, and McHugh 1975); level of patient functioning was assessed by the center nurse during a home visit using the Blessed Roth Dementia Rating Scale (Blessed, Tomlinson, and Roth 1968); number of psychiatric symptoms was based on the center psychiatrists' rating of 26 symptoms, using DSM-III criteria (American Psychiatric Association 1980); number of neurological signs was based on the center neurologists' rating of 10 motor, 5 involuntary movement, and 6 primitive reflex signs, using a standard protocol; number of other current health problems or illnesses was based on the center physicians' rating of 42 disorders, using a standard protocol after a physical history and examination; and number of patient problem management behaviors was based on the center nurses' interview with a family caregiver, using a standard protocol (range 0–10). A principal components analysis of the six measures yielded one factor, accounting for 67% of score variance. A factor score from this single component indexed patient severity.
Characteristics of the Caregiving Setting.
Three measures were included. (a) Pre-NHP frequency of use of professional community services was included because it was thought to reduce caregiver stress and burden, thus potentially reducing the need for NHP. Service utilization was defined as the number of regularly used services provided by professional and/or community programs during the 12 months prior to NHP. A standard list of 26 services (respite care, support groups, case management, homemaker services) was employed, and frequency of use was recorded. (b) Caregiver distress was assessed by asking each family member to rate the patient's behavior on a subjective 5-point scale of "troublesomeness," based on their distress in dealing with 17 patient problem behaviors (wandering, getting lost, asking the same question over and over, hiding things, being suspicious or accusative; alpha = .83). (c) Hours of caregiving per week also was assessed.
Data Analysis
We used a mixed-models, repeated measures linear regression procedure (SAS PROC MIXED; Khattree and Naik 1995) to analyze the impact of six between-subjects and one within-subjects (time) effects on family caregiver health and well-being: patient group (NHP, non-NHP), patient severity, caregiver gender, caregiver education, relationship of caregiver to patient, and use of professional services. This procedure permits inclusion of both within- and between-subjects effects and enables examination of each main and interaction effect simultaneously and independently of the other effects. We used the composite health and well-being score in this analysis to keep the number of comparisons to a minimum.
Preliminary analyses indicated unacceptable levels of multicollinearity that could undermine the primary analysis. The mixed model analysis required that we drop caregiver distress because the variance attributable to it was already included in the remaining variables in the equation. One within-subjects effect was also included: time (baseline, 6, 12, 18, 24 months after NHP). Because of the number and complexity of the interaction terms (all 2-, 3-, 4-, 5-, 6-way interactions), only 2- and 3-way terms were reviewed.
|
Results |
|---|
|
TOPAbstractBackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
|
The analyses reported so far included NHP patients who were institutionalized during the interval between clinic evaluation and 6 months postevaluation. A 6-month cutoff was used to permit sufficient time (24 months) to elapse to assess reactions to NHP. It could be the case, however, that those patients who were institutionalized within 6 months of clinic evaluation were particularly problematic or in other ways differed from patients institutionalized at other times during the 2 years following clinic evaluation. Therefore, we returned to our original cohort and identified a group of 117 patients who underwent NHP any time during the 2 years following clinic evaluation, and 152 patients who remained with their families in the community during that period. The original analyses were then repeated, using health and well-being scores before and after NHP as the outcome variables. Three analyses were run with the same main and interaction effects outlined above. Because this was a larger sample than the one employed in the mixed model analysis, we had sufficient power to disaggregate the composite index of health and well-being into its contributing scores: somatic symptoms, anxiety/depression, and well-being.
Again testing the central hypothesis of the study, the two-way interaction of Group (NHP vs non-NHP) and Time was not significant in any of the three analyses (anxiety/depression, F = 1.04, df = 1, 253; somatic symptoms, F = 1.93; well-being, F = 2.40). Also, none of the three-way interactions among Group (NHP vs non-NHP), Time, and a qualifying variable reached statistical significance in any of the three equations. Thus, the same results occurred regardless of when NHP occurred and across all three health and well-being outcome variables.
Additional Findings
Of the seven main effects (six between subjects, one within subjects), only one reached statistical significance (Table 3 ): Spouses reported significantly poorer health and well-being scores, compared to offspring and in-laws (F = 4.01, p = .02).
Because this was a study of change over time in caregiver health and well-being, we next explored the two-way interactions between each of the between-subjects effects and time for both NHP and non-NHP groups combined. These effects indicated how well each primary variable predicted change in caregiver health and well-being over time. Only one significant two-way interaction effect occurred: Time by Caregiver gender (F = 3.77, p = .02). For both NHP and non-NHP groups combined, the health and well-being of all female family caregivers declined over time more than the health and well-being of all male family caregivers.
We next explored the remaining three-way interactions with Time (both groups combined), and two reached statistical significance (Table 3 ): Patient severity by Caregiver gender by Time (F = 3.84, p = .02), and Patient severity by Relationship by Time (F = 2.81, p = .02). The findings for both three-way interaction terms were similar. At high levels of patient severity, the health and well-being of both male and female family caregivers declined equally over time. However, at low levels of patient severity, the health and well-being of female family caregivers declined more than the health and well-being of male family caregivers over time for both patient groups combined. In the second significant three-way effect, the decline in health and well-being over time was equal for spouses, offspring, and in-laws, but only at high levels of patient severity. The health and well-being of spouses declined over time more than the health and well-being of other family members at low levels of patient severity. Thus, severity of patient deficit qualified the original two-way interactions between Caregiver gender and Time such that the effects occurred only at low levels of patient severity.
|
Discussion |
|---|
|
TOPAbstractBackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
The findings also suggest that, overall, female family caregivers and spouses experience the greatest decline in health and well-being over time, compared to male caregivers and offspring and in-laws, respectively, regardless of whether or not the patient is institutionalized. These differences, however, are more pronounced for family members of patients with low disability than with high disability. It may be the case that a ceiling effect is reached at high levels of patient disability for all family members. The stresses of caring for a severely debilitated patient in the community may become maximized at some point in time.
How may we account for the absence of a significant improvement in family caregiver health and well-being following placement? Several speculations are worthy of consideration. First, the reasons that lead a family to decide on NHP are complex and may go beyond a simple explanation based on the problems associated with managing a deteriorating patient. Although studies of factors that contribute to a decision to institutionalize have found significant, positive relationships between patient severity and NHP, the size of the relationship is reduced significantly when studies are limited to samples of patients living in the community and cared for by family members. Thus, it is not patient severity alone, or perhaps most predominantly, that accounts for why many families decide on NHP in the first place and, by extension, how they react to NHP over time.
A series of studies that address the family context of patient management and decision making may provide a direction for understanding at least some of the dynamics underlying the effects of NHP on family members. Fisher and Lieberman 1999 used a longitudinal design with 164 patients with a diagnosis of AD and their three-generation families. They found that families that scored high on emotional closeness and on negative family feelings, and low on family problem-solving efficiency, institutionalized their ill elder at a significantly higher rate than families that scored in the opposite direction; this finding held even after controlling for a host of other potentially influential variables, including patient severity and demographics. These family characteristics accounted for substantially more variance in predicting NHP than both the severity of the patient's deficit and the amount of strain experienced by the caregivers. If some family caregivers' distress is a product of close family processes and relationships, then the solution offered by NHP is unlikely, in itself, to offer relief. Thus, the context of the family decision-making setting and the inherent history of positive and negative family relationships may account for greater variation in changes in caregiver health and well-being following NHP than the variables we explored in the present research.
This line of thinking is advanced by a study by Mittleman, Ferris, Shulman, Steinberg, and Levin 1996, who used a family psycho-education technique with families of patients with AD and then provided ongoing family care over time. The intervention included education about the disease and experience addressing the personal and family relationship changes that occur when caring for an elderly patient with an increasingly debilitating disease over time. Compared to controls, their findings indicated improved family member well-being over time and, most importantly, a significant delay in NHP. Again, qualities of the family context of caregiving played a significant role in family member health, well-being, and decisions to institutionalize.
A second line of reasoning concerns the important influence of caregiver bereavement. It is often assumed that many of the negative effects of caregiving are attributable to the emotional side effects of providing care. An additional explanation is that considerable distress among caregivers is attributable to a recognition of the impending "psychological" and physical death of a very close family member. Several studies (Garner 1997; Jones and Martinson 1992; Ponder and Pomeroy 1996; Walker and Pratt 1991) have suggested that an understanding of the consequences of caregiving for a demented relative may be better understood by considering the construct of "anticipatory grief." One study, comparing the spouses of NHP and non-NHP AD patients, found that although most caregivers reported a reduction in the burden of care following the institutionalization of their spouses, most also experienced considerable grief, depression, and guilt regarding the loss of a strong emotional attachment. This explanation is consistent with the fact that most distress experienced by caregivers is psychological, and that the stronger the attachment between the patient and the caregiver, the greater the caregiver experience of loss. This view also gains support from our findings, in that spouses experienced the greatest decline in health and well-being over time, compared to other family members.
Clinical Considerations
A recent study found that health care professionals base their suggestions that families consider NHP for patients with dementia on two general criteria: the severity of the patient's functional deficit, and the degree of family caregiver burden (Lieberman and Fisher 1998). The same report, however, also found that these two criteria are not major contributors to a family's actual decision to institutionalize. Rather, the quality of family relationships and the efficiency of family decision making play a far greater role in this regard. Thus, there is a disjunction between health care professionals and families regarding the reasons for suggesting and deciding on NHP. The findings of the current research extend those of this earlier report by suggesting that severity of patient deficit and caregiver burden also may not uniformly affect the outcome of NHP on caregiver health and well-being.
These findings can be translated into three sets of clinical considerations, when placed within the context of multigeneration families struggling to manage a major chronic health problem. First, the sources of distress that lead a family to consider NHP are multifaceted, and they may not be overtly obvious. It is important for health care professionals to evaluate the family setting carefully to identify the patterns and causes of distress that may lead a family to consider NHP, rather than to assume that the single cause is to relieve the stresses of caregiving. Likewise, reactions to NHP most likely are not related exclusively to pre-NHP patient severity and caregiver distress. Second, it is important to keep in mind that different family members experience the stresses of caregiving in different ways and to different degrees, and that reactions to NHP likewise are different. That is, the burdens of caregiving are not uniform among the several family members who provide care over time, both before and after NHP. Thus, it may be important to prepare family members, especially women and spouses, about the variety of reactions that may occur after NHP takes place. Third, the data suggest that emphasis be placed on considering caregivers within the broad family context of care, and understanding the burdens of caregiving as part of a set of family caregiving relationships (Fisher and Lieberman 1999; Lieberman and Fisher 1995b, Lieberman and Fisher 1999a, Lieberman and Fisher 1999b). This approach stands in contrast to considering only the primary caregiver.
These findings suggest that the frequent focus on patient severity, caregiver strain, and the other variables that are important in the caregiving process should be broadened. Future research should also address how characteristics of the family context of care affect post-NHP family caregiver health and well-being over time.
Practice Concepts
The Forum
Book Reviews
|
Acknowledgments |
|---|
Received for publication November 8, 2000. Accepted for publication July 30, 2001.
|
References |
|---|
|
TOPAbstractBackgroundFactors That May Influence...MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. K. Scherer, N. Scarmeas, J. Brandt, D. Blacker, M. S. Albert, and Y. Stern The Relation of Patient Dependence to Home Health Aide Use in Alzheimer's Disease J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2008; 63(9): 1005 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Williams, C. S. Williams, S. Zimmerman, J. Munn, D. Dobbs, and P. D. Sloane Emotional and Physical Health of Informal Caregivers of Residents at the End of Life: The Role of Social Support J. Gerontol. B. Psychol. Sci. Soc. Sci., May 1, 2008; 63(3): S171 - S183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Reid, N. L. Chappell, and J. A. Gish Measuring family perceived involvement in individualized long-term care Dementia, February 1, 2007; 6(1): 89 - 104. [Abstract] [PDF]
|
|
|
|
 |
|
 R. Newcomer, T. Kang, and C. Graham Outcomes in a Nursing Home Transition Case-Management Program Targeting New Admissions. Gerontologist, June 1, 2006; 46(3): 385 - 390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Spitznagel, G. Tremont, J. D. Davis, and S. M. Foster Psychosocial Predictors of Dementia Caregiver Desire to Institutionalize: Caregiver, Care Recipient, and Family Relationship Factors J Geriatr Psychiatry Neurol, March 1, 2006; 19(1): 16 - 20. [Abstract] [PDF]
|
|
|
|
 |
|
 R. L. Thompson, S. L. Lewis, M. R. Murphy, J. M. Hale, P. H. Blackwell, G. J. Acton, D. H. Clough, G. J. Patrick, and P. N. Bonner Are there Sex Differences in Emotional and Biological Responses in Spousal Caregivers of Patients with Alzheimer's Disease? Biol Res Nurs, April 1, 2004; 5(4): 319 - 330. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
