UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Predictors of registration with continuing care and the utilization of support services by family caregivers… Klassen, Lillian 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_fall_klassen_lillian.pdf [ 5.3MB ]
JSON: 831-1.0098863.json
JSON-LD: 831-1.0098863-ld.json
RDF/XML (Pretty): 831-1.0098863-rdf.xml
RDF/JSON: 831-1.0098863-rdf.json
Turtle: 831-1.0098863-turtle.txt
N-Triples: 831-1.0098863-rdf-ntriples.txt
Original Record: 831-1.0098863-source.json
Full Text

Full Text

PREDICTORS OF REGISTRATION WITH CONTINUING CARE AND THE UTILIZATION OF SUPPORT SERVICES BY FAMILY CAREGIVERS OF ALZHEIMER'S DISEASE PATIENTS  By Lillian Klassen B.A., The University of British Columbia, 1988  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF MEDICINE Health Care & Epidemiology Health Care Planning & Administration  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA August 1993 ©Lillian R. Klassen, 1993  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  1-1 caCk-k\A„Qp..se_^  The University of British Columbia Vancouver, Canada Date ^fr^■ t, el -3  DE-6 (2/88)  (^  , P14-4"j‘j-4-k 'he ari-co'`-  Abstract  This is a cross-sectional exploratory study designed to examine factors associated with the registration and use of Continuing Care (CC) support services by the caregivers and Alzheimer patients. Registration with CC was related to several of the need characteristics of both the patient and the caregiver. The patient's level of impairment as measured by the Functional Rating Scale (FRS) and the Present Functioning Questionnaire (PFQ) was significantly higher for those patients registered with CC. A greater degree of caregiver burden as measured by the Zarit Burden Interview was reported by caregivers of patients who were registered with CC. The Burden Interview was subjected to a factor analysis revealing 5 scales measuring different aspects of burden. When examined individually with respect to registration, three of the scales: emotional burden, social intrusion and demand burden surfaced as being significantly associated. It was noted that a total burden score can mask the specific aspects of burden which are related to the outcome. In a closer examination of those who were registered with CC, characteristics of patients and caregivers were examined, comparing those utilizing support services with those not utilizing services. The two groups did not differ significantly with respect to any characteristics. These findings suggest that caregivers may be prone to view registration with CC more as a last resort than a collaborative effort in the interim period. Possible attitudes and perceptions held by caregivers and patients are discussed as influential factors related to utilization practices. Implications for the Canadian Healthcare System and the role of healthcare professionals are also addressed. The need for a longitudinal study is noted due to the dynamic, yet progressive nature of Alzheimer's Disease and the ensuing burden connected with caring for a patient.  iii Table of Contents  Abstract ^ Table of Contents ^ List of Tables ^ Acknowledgements ^  ii iii vi viii  CHAPTER ONE: Introduction 1 Background to the Problem ^ 2 Statement of the Problem ^ 3 Purpose of the Study ^ 3 Theoretical Framework ^ 6 Research Questions ^ Significance of Study to Healthcare in Canada ^ 6 7 Definition of Terms ^ 8 Overview of the Thesis Content ^  CHAPTER TWO: Review of The Literature Introduction ^ History of Caregiving Studies Focus on Caregiver Stress ^ Focus on Service Utilization ^ Summary of Literature Review ^  10 10 27 36  CHAPTER THREE: Methodology Introduction ^ Research Design ^ Sampling Procedure ^ Measurements ^ Data Collection Procedures ^ Ethical Considerations ^ Data Analysis ^ Limitations ^  39 40 40 42 47 48 49 51  iv  CHAPTER FOUR: Analyses & Findings  52 Introduction ^ 53 Descriptive Analysis of Sample ^ Predisposing Characteristics Patients ^ 53 56 Caregivers ^ Enabling Characteristics ^ 56 Need Characteristics 59 Patients ^ Caregivers ^ 63 Correlation Matrix of Predictor Variables ^ 66 Comparison of Two Groups (Registered vs Not) ^ 73 75 Predisposing Characteristics ^ 75 Patients ^ 78 Caregivers ^ Enabling Characteristics ^ 80 82 Need Characteristics ^ Patients ^ 82 Caregivers ^ 86 Comparison of Two Groups (Users vs Non-users) ^ 87 Predisposing Characteristics Patients ^ 87 Caregivers ^ 88 Enabling Characteristics ^ 88 Need Characteristics 91 Patients ^ 93 Caregivers ^ Analysis of Zarit Burden Interview ^ 96 Findings Question 1. Which predictor variables are significantly related to caregiver psychological health,(burden, stress and depression)? ^104 Question 2. Which predisposing, enabling and need variables distinguish between those families who register with CC and those who do not? ^104 Question 3. For those registered with CC, which predisposing, enabling and need variables determine whether the family actually uses the services provided by the Continuing Care Division of the Ministry of Health of B C ^105 '  Question 4. Into what dimensions can the Burden Interview be separated to reveal unique associative predictive qualities for the outcomes of CC registration and the utilization of services? ^106  CHAPTER FIVE: Summary, Conclusions, Implications and Recommendations  V  108 Introduction ^ 108 Descriptive Analysis of Sample ^ Correlation of Variables with Caregiver Outcomes ^ 110 Comparison of Registered & Non-Registered ^ 113 116 Analysis of Zarit Burden Interview ^ Comparison of Users & Non-Users of Services ^ 117 120 Considerations for Healthcare ^ 122 Recommendations for Future Research ^  References ^  125  Appendices Appendix A Functional Rating Scale ^  138  Appendix B Present Functioning Questionnaire ^ 139 Appendix C Zarit Burden Interview ^  140  Appendix D Consent Form ^  141  vi  List of Tables 1.  Frequencies of Patient Predisposing Characteristics ^ 55  2.  Frequencies of Caregiver Predisposing Characteristics ^ 57  3.  Frequencies of Caregiver Enabling Characteristics ^ 58  4.  Frequencies of Patient Need Characteristics ^ 60  5.  Analysis of Patient Need Characteristics ^ 62  6.  Frequencies of Caregiver Need Characteristics ^ 65  7.  Correlation of Predisposing Characteristics ^ 67  8.  Correlation of Enabling Characteristics ^ 68  9.  Correlation of Need Characteristics ^  69  10.  Correlates of Caregiver Burden ^  71  11.  Correlates of Caregiver Depression ^  72  12.  Correlates of Caregiver Stress ^  74  13.  CC Group Comparison - Patient Predisposing Characteristics ^  76  14.  CC Group Comparison- CG Predisposing Characteristics ^ 79  15.  CC Group Comparison- CG Enabling Characteristics ^ 81  16.  CC Group Comparison- Patient Need Characteristics ^ 84  17.  CC Group Comparison- Patient Need Characteristics ^ 85  18.  User/Non-User Group Comparison - Patient Predisposing Characteristics ^  19.  Use Group Comparison- CG Predisposing Characteristics ^ 90  20.  Use Group Comparison- CG Enabling Characteristics ^ 92  21.  Use Group Comparison- Patient Need Characteristics ^ 94  22.  Use Group Comparison- CG Need Characteristics ^ 95  89  23. Coefficient Alpha for Five Scales of Burden Interview..100  vii  24.  Intercorrelations among the Five Scales of Burden Interview ^  102  25.  Five Scales of Burden Interview Related to CC Registration ^  103  viii  Acknowledgements I wish to thank the members of my committee for their help on this project. Dr. Clyde Hertzman was the supervisor for the committee. In addition to his positions in the Health Care and Epidemiology Department at U.B.C., he is a principal investigator for the B.C. portion of the Canadian Study of Health and Aging. Drs. Arminee Kazanjian and Holly Tuokko were the other Committee members. Dr. Kazanjian is Associate Director of the Centre for Health Services and Policy Research and has been involved in research of health personnel planning and the utilization of medical services. As such, she was able to provide assistance in the evaluation of service utilization. Dr. Tuokko is a neuropsychologist at the Clinic for Alzheimer Disease and Related Disorder, working with the multidisciplinary team in the assessment of the referred patients. Her comprehensive knowledge of assessment particulars was crucial in ensuring that the data was reliable. Mike Sorochan of Continuing Care was the contact person for the database at the Vancouver Health Unit. He directed me to Diane Bissendon, data consultant, at the Ministry of Health, Continuing Care Division in Victoria, B.C. She was the contact person responsible for providing the data regarding homemaker and adult day care services. Directors of the Meals-on-Wheels programs at the various centres of Continuing Care across B.C. cooperated in providing information regarding the use of Meals-on-Wheels over a 6 month period. I would also like to thank the following for their assistance through discussions concerning methodology and editing: The Rev. Patricia Backman, M.Sc., M.A., Dr. Kathryn Klassen, Ph.D., Dr. Joanne Miller, Ph.D. and Dr. Karen Tallman, Ph.D. In addition, Janine Weir, B.A. was essential in her assistance with the data entry and analysis.  1 CHAPTER ONE INTRODUCTION  Background to the Problem Dementia is a disease with numerous medical, social and economic ramifications. The aging of our population coupled with the increased prevalence of cognitive impairment for those over 85 (ie. 1% at age 65, 7% to 10% at age 75, 20% to 25% at age 85) will have serious implications for the healthcare system (Pfieffer, 1988). At present, the majority of those suffering from dementia continue to reside in the community, usually cared for by a family member or close friend. The impact of the disease therefore extends beyond the patient to the caregiver. Researchers have examined the effects of dementia, not only on the patients, but also on their family or other caregivers. The majority of the studies have used caregiver burden or depression as the outcome indicator. Perceived stress of the caregivers has been studied in relation to factors such as: age of the caregiver, (e.g. Brody, 1981), behaviour of the patient, (e.g. Pruchno & Resch, 1989), gender of the caregiver and the patient, (e.g. Barusch & Spaid, 1989), caregiver/patient relationship, (e.g. Young & Kahana, 1989), social network of the patient and caregiver, (e.g. Graham, 1989), and the stage of the disease, (e.g. Gilhooly & Whittick, 1989). Typically, studies involved rather small samples and produced varied results.  2 Inconclusive findings of such studies were understandable due to the fact that the studies were characterized by a diversity of outcome measures (Baumgarten, 1989). Other recent studies have considered the above mentioned factors, as well as the characteristics of the caregivers' physical and psychological health in relation to the utilization of healthcare services. Adult daycare for Alzheimer patients, (e.g. Sands & Suzuki, 1983), family-professional partnerships, (e.g. Silverstein et al., 1989), home support, (e.g. Lindsey & Murphy, 1989), and respite services, (e.g. Lawton, et al., 1989), have been evaluated in some settings to determine their effectiveness. The findings suggest that supports such as these could be effective in reducing the perceived stress of caregivers, allowing them to continue to base care in the community. These services are available in British Columbia, many through the Continuing Care (CC) Division of the Ministry of Health.  Statement of the Problem  A recent analysis of patient data from the Clinic for Alzheimer Disease and Related Disorders, University Hospital-UBC Site, indicated that over 60% of dementia victims residing in the community were not registered with CC, and therefore were not likely to be receiving any of the available support services (Scott, Kosarek & Tuokko, 1988). Many caregivers, for a variety of reasons are reluctant to relinquish care of the Alzheimer  3 patient to professionals in an institution or to the formal support workers in the healthcare system. Instead, they choose to continue to shoulder the care of their relative, often beyond healthful limits (Colerick & George, 1986). The problem that this study addresses is that caregivers of Alzheimer's disease patients are not using the support services available to them, at risk to their own health.  Purpose of the Study The objective of this study is to determine predictors of registration with and utilization of services from the Continuing Care Division of the Ministry of Health of British Columbia by Alzheimer patients and their caregivers.  Theoretical Framework A conceptual framework developed by Andersen & Newman (1973) has been used in studies examining service utilization. The framework categorized predictor variables of service utilization as enabling, predisposing or need characteristics of the patient. The Andersen model, generally used in evaluating and predicting service utilization, did not include characteristics of the caregiver or other informal support. This omission was striking since two decades of documented evidence has shown that families provided the bulk of care for the impaired aged (Bass & Noekler, 1987). Since typically one family member assumes major, if not exclusive responsibility for the elder's care, this  4  primary caregiver appears to be the logical focal point for incorporating informal supports into the Andersen Model. Therefore, for purposes of their study involving the impaired aged Bass & Noekler (1987) modified the Andersen model to include primary caregiver characteristics as well as particulars concerning the informal support network of the patient. The assumption underlying the inclusion of these variables was that the utilization patterns of the impaired elderly were influenced both directly and indirectly by the availability and characteristics of primary caregivers and could be modified by the amount of assistance offered to the elderly and their caregivers by secondary members of the informal care network. Therefore, the framework for examining predictors of service utilization included the informal support network and the predisposing, enabling and need characteristics of both the patient and the caregiver. Findings from Bass & Noekler's study indicated that caregiver need characteristics accounted for significant variation in whether or not services were used, while family enabling factors were the most important predictors of the amount of service used. Using the same framework, a recent study examined the predictor variables in relation to service utilization, distinguishing between groups using in-home services only, outof-home services only, both types of services or no services at all in the Cleveland and Pittsburg areas (Biegel, et al., 1990). The findings suggested that informal support facilitated service  5 use of out-of-home and both types of services. It was also shown that users of both types of services differed considerably from those who used in-home or out-of home services exclusively. Particular types of burden (emotional and relationship) were shown to differentially predict service use. As of 1991 a comparable study had not been conducted in Canada. In the present investigation, the Andersen framework as modified by Bass & Noekier (1987) was used to examine the variables distinguishing the AD (Alzheimer) patients and their caregivers who registered with the British Columbia Continuing Care Program and those who did not. Registration with CC entitles the care recipient to the formal support services deemed to be necessary and appropriate by the Continuing Care Division. The variables considered in this framework are listed below. 1. Predisposing Characteristics a) Patient -Age -Gender -Marital status -Household composition -Duration of illness b) Caregiver -Age -Gender -Relationship to patient 2. Enabling Characteristics a) Caregiver -Degree of involvement b) Family Support -Amount -Perceived lack 3. Need Characteristics a) Patient -Diagnosis of Alzheimer's disease -Severity Level indicated by the Functional Rating Scale  6  -Severity Level indicated by the Present Functioning Questionnaire -Rates of Decline (Overall, Cognitive, Social) -Concurrent Illnesses b) Caregiver -Stress -Depression -Burden score (Total & Individual Items)  Research Questions The nature of this study suggests at least two research questions, namely; 1.  Which predisposing, enabling and need variables  distinguish between those families who register with CC and those who do not? 2.  For those registered with CC, which predisposing,  enabling and need variables determine whether the family will actually use the services provided by CC?  Significance of Study to Healthcare in Canada Although there is a need for services (i.e., respite, adult day care and in home services), utilization has been surprisingly low (Caserta et al., 1987). Little is known about the utilization patterns of caregivers to dementia victims when services are available. The factors that influence service utilization need to be identified and examined (Ory et.al , 1985). Baumgarten (1989) stated that...  7 "if the caregivers are found to be at particularly high risk of certain health problems, then it may be necessary to define them as a target group for health interventions and services of various kinds. Furthermore, since the presence of a capable caregiver is often the crucial factor in keeping the demented patient in the community, preventing and treating health problems among the caregivers may reduce the likelihood of the demented patient being institutionalized."(p.1137) Recent research initiatives have revealed that our healthcare system would be economically advantaged by redirecting institutional costs into support services for those requiring ongoing care. The use of support services could be a preventative measure for those caregivers particularly at risk for physical or psychological illness. Indirectly, then, potential healthcare costs in relation to caregivers could also be circumvented.  Definition of Terms Alzheimer Patient - An Alzheimer patient was defined as having possible or probable Alzheimer Disease according to the NINCDSADRDA criteria (McKhann, et al., 1984) Those suffering from the  illness demonstrated varying degrees of cognitive and functional impairment. Caregivers - In this study, caregivers were the primary persons providing some form of care for the Alzheimer patient. They did not need to be living with the patient in order to be considered a caregiver. Burden  -  Burden was defined as "the physical, psychological or  8 emotional, social, and financial problems that can be experienced by family members caring for impaired older adults" (George & Gwyther, 1986) Continuing Care(CC) - Continuing Care is support offered for disabled adults from the ages of 19 and disabled and frail elderly individuals. Services provided by the Continuing Care Division (aka Long Term Care - LTC- Division) of the Ministry of Health in British Columbia range from in home help with household tasks, meal preparation, and personal care. In addition, CC offers Daycare programs, Meals on Wheels, respite care and the opportunity for placement into intermediate or extended care facilities.  Overview of Thesis Content Chapter II provides a review of literature addressing issues of dementia patients and their caregivers. Outcomes examined will include caregiver physical and psychological health as it is influenced by caring for a dementia patient, propensity to institutionalize the patient, and the utilization of formal support services as an aid in caring for the patient. The current study examines characteristics of the patient and the caregiver as they pertain to registration with CC and the utilization of services. Chapter III describes the methodology used to study the predisposing, enabling and need characteristics of the patient, caregiver and the informal network of family and friends. In addition, the measure used to  9 assess burden (Zarit, 1980) will be subjected to a factor analysis to determine specific aspects of burden which may be uniquely associated with the utilization of support services. Chapter IV summarizes the analysis and findings of the study and Chapter V consists of a discussion of the findings, the relevance for the current healthcare system and recommendations for further research initiatives.  10 CHAPTER TWO REVIEW OF THE LITERATURE  Introduction  During the past 20 years, there has been increasing research in two areas which could be applied to dementia victims: the impact of the illness on those providing care and the utilization of healthcare services by the care recipient.  History of Caregivinq Studies  Focus on Careaiving Stress Studies concerning the care of dementia victims have focused primarily on caregiver outcomes. They have examined possible determinants of caregiver stress, burden, sense of well-being or drug use. In 1989, Baumgarten summarized her comprehensive review of the caregiver literature to date which dealt with the negative effects on various aspects of caregivers' psychological and physical health. She limited her review to 28 studies (published from 1980 to 1989) which dealt specifically with the caregivers of dementia victims and included some measurable outcome in terms of physical, psychological or social well-being of the caregivers. Overall, the findings in the studies were contradictory and inconclusive. She suggested that the great variation in outcome measures as well as other methodological  11 shortcomings of the studies accounted for the findings. Studies since 1989 also produced inconclusive results due, again, to the various ways of operationalizing the outcomes and measuring the variables in question. Variables most often considered in the studies in relation to caregiver outcomes were caregiver's age, relationship of the caregiver to the patient, severity of the patient's illness, influence of social support, and gender of the caregiver. Less often examined were factors such as age of the patient, living arrangements, race or ethnic background and duration of the illness. Caregiver Age Of the studies examining caregiver age in relation to various caregiver health indicators, four studies indicated no association. Gilleard et al., (1984) examined psychiatric morbidity in a sample of 224 supporters of communityresident elderly persons (87% demented) attending or referred to geriatric day hospitals and noted no association. Gilhooly (1984) had the same result when examining caregiver age and mental health in his study of 37 supporters of community-resident demented patients attending day hospital. Fifty-four percent of these caregivers lived with the patient. In a study of 240 caregivers to Alzheimer patients recruited from the Alzheimer Society, Pratt (1986) found no significant association between caregiver age and perceived caregiver burden. Pagel and Becker (1987) observed similar results when studying caregiver age and depression in a sample of 68 spouse caregivers of Alzheimer patients referred by physicians, clinics and the Alzheimer  12  Society. Other studies indicated significant associations. However, results of these studies were inconsistent. In his study of the sample of 68 which Pagel and Becker (1987) had examined, Fiore (1986) found that older caregivers had more psychological symptoms than younger caregivers, but that physical symptoms and depression did not vary much with age. Fitting, et al., (1986) examined caregiver age together with gender in relation to burden and increased impairment and found that for a group of 54 caregivers, the younger wives and older husbands experienced more burden in relation to increased impairment of the patient. In an earlier study of a group of 129 supporters of psychogeriatric day hospital patients Gilleard et al., (1984) found that younger supporters^experienced more strain and burden than older supporters.^Haley et al., (1987) rated depression, life satisfaction and self-rated health with respect to the caregiver's age for 54 caregivers of community dwelling demented patients. They found a significant negative correlation between caregiver age and self-rated physical health, although they found no significant correlation between caregiver age and life satisfaction or depression. In their study with 131 spousal caregivers, Barusch & Spaid (1989) found that caregiver age was an important factor in predicting caregiver burden; specifically, younger caregivers reported greater subjective burden. In a study examining age, gender and relationship, Anthony-Bergstone et al., (1989) used the Brief Symptom Inventory to measure nine  13 dimensions of emotional distress: Somatization, ObsessiveCompulsive, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation and Psychoticism. They found that older women experienced more subscale elevations than younger women. Relationship Several researchers have focused on the  relationship between the caregiver and patient with respect to caregiver physical or psychological health. The majority of studies observed no effect of kinship tie on caregiver health. Gilleard et al., (1984) found that the type of relationship did not influence caregiver stress. Daughters and spouses did not differ in the feelings of burden they reported in a study involving 29 caregivers of community residing demented patients attending a research and training centre for the elderly (Zarit et al., 1980). Eagles et al., (1984) compared spouses with children in their levels of psychological morbidity and strain but found no difference. Two hundred and forty caregivers to Alzheimer patients were compared by their various relationships to the patients, but were not found to differ with respect to burden, morale or self-rated health. No consistent association was found between relationship of the caregiver to the patient and levels of well-being, perceived physical health or emotional health in a study of 58 family members of Alzheimer patients conducted by Quaghagen and Quaghagen (1988). In Gallagher et al.'s (1989) study of 158 caregivers the relationship of the caregiver to the care-receiver did not significantly affect rates  14 of depression as measured by the Schedule for Affective Disorders and Schizophrenia - SADS (Endicott & Spitzer, 1978). Gilhooly (1984) also found no effect of kinship tie on morale, although she did find a negative correlation between relationship and mental health. This indicated that the greater the kinship distance, the better the supporters' mental health. In contrast, Brody (1985) focused her study specifically on daughters and concluded that daughters of the patients are "women in the middle", caught in multiple roles which produced increased stress. The majority of studies, however, did not adjust for age or sex, thereby producing potentially confounding results. George and Gwyther (1986) did adjust for age and still found, in their sample of 510 caregivers of demented patients, that the spousal caregivers reported significantly more doctor visits than adult children or other relatives. They also reported lower levels of well-being, more stress symptoms, more psychotropic medication use, lower levels of affect balance, and less life satisfaction than the nonspousal caregivers. In a recent study by Anthony-Bergstone et al., (1985), it was noted that older women experienced higher levels of stress than older men or younger women. Considering the overlap of relationship and age, the women were split into relationship categories, and though not statistically significant, the wives were found to be especially vulnerable to the stress of caregiving. As part of a study examining burden for caregivers of demented and depressed patients, Liptzin et al., (1988) found in a combined analysis of  15 variance for the caregivers that the burden levels of the child caregivers were higher than those of spousal and other relative caregivers prior to the patient's admission into a geriatric unit. Severity of Patient Illness Studies considering the  correlation between severity of the patient illness and caregiver health have produced even more conflicting results. Deimling and Bass (1986) suggested that severity of the disease needed to be viewed in three dimensions: cognitive impairment, functional impairment and behavioral disturbance. Results from their study of 614 families living with and caring for an impaired elder show that cognitive incapacity had a less important direct effect on caregiving stress than disruptive behaviour and impaired social functioning. They offered several theories to explain this. One suggested that the increased level of care required for disruptive behaviour produced greater caregiver stress. A more compelling theory was that the patient was seen to be more accountable for disruptive behaviour than cognitive deficits which were seen as symptoms of the disease. This "inappropriate" behaviour resulted in more stress for the caregiver. However, even when the three dimensions of illness severity were considered, the findings were not consistent. Most studies showed no correlation between cognitive status and health of the caregivers (Zarit et al., 1980; Gilhooly, 1984; Scott et al., 1986; Haley et al., 1987; Gallagher, 1989). Contrary to their  16 expectations, in the Zarit et al. (1980) study, the extent of cognitive impairment of the patient was not correlated with the caregiver's level of burden. Gilhooly (1984) also expected to find the dependant's cognitive functioning and impairment level to be associated with low morale and poor mental health of the caregiver, but found no association. Scott et al., (1986) also found no association between the mental status of the patient and caregiver burden scores in her group of 23 primary caregivers. Using the MMS (Folstein et al., 1975) as a measure of the severity of patient cognitive impairment, Haley et al., (1987) noted no significant association between cognitive impairment and caregiver depression for 44 primary caregivers. In contrast, Eagles et al., (1987) found a significant correlation between cognitive status and caregiver stress as measured on the Relative Stress Scale (Greene, 1982), but not with the General Health Questionnaire-GHQ (Goldberg, 1972) measure of psychiatric morbidity. In another study conducted by Eagles et al., (1987) involving 274 caregivers, no significant association was found between cognitive impairment and scoring on the GHQ. However, an association between cognitive impairment and depression did appear for husbands only. In a longitudinal study conducted by Gallagher et al., (1989) of a group of 158 caregivers, depression was measured by the Schedule for Affective Disorders and Schizophrenia (SADS- (Endicott & Spitzer, 1978) and the Beck Depression Inventory (BDI- Beck et al., 1961). Cognitive impairment status of the patient was not found to be  17 related to caregiver depression as measured by the SADS and the BDI. Investigators examining patient functional status in relation to caregiver stress also discovered inconsistent results. Several studies found no significant association between functional status and caregiver depression or burden (Drinka et al., 1987; Gilleard et al., 1982; Pagel et al., 1985). In his examination of the correlates of feeling of burden for the relatives of impaired elderly, Zarit et al. (1980) found that, contrary to expectations, the level of functional impairment on Lawton's (1971) Physical and Instrumental Activity of Daily Living scales (PADL and IADL) were not correlated with the level of burden. Gilhooly (1984) also noted no significant correlation between the functional impairment level of the patient and caregiver morale, measured on the Kutner Morale scale (Kutner et al., 1956) or mental health, measured on the Older American's Resources and Services (OARS) Multidimensional Functional Assessment Questionnaire's mental health scale (Duke University Centre for the Study of Aging and Human Development, 1978). However, several studies have revealed varying degrees of association between functional status and caregiver outcomes. In a study involving 30 caregivers, Novak and Guest (1989) noted a moderately significant association between general functional ability of the care recipient and caregiver burden level. Functional decline was measured using Gurland's Criteria for Diagnosis and Severity of Dementia questionnaire (Gurland et al.,  18  1982) and caregiver burden was assessed using the Zarit Burden Interview. Wilder et al. (1983) conducted a survey with 162 key informants of community resident demented patients and found that limitations in the patient's activity were predictive of caregiver perceived inconvenience. Deimling and Bass (1986) noted a significant correlation between the patient's functional status and change in the caregiver's health, although not in depressive symptoms. By separating basic tasks from instrumental tasks listed as activities of daily living (ADL), Haley et al. (1987) found that whereas both types of tasks were not significantly associated with caregiver life-satisfaction or self-rated health problems, the ability to perform instrumental tasks was highly correlated with the caregiver's depression score. Fitting et al. (1986) examined the association separately for young and old husbands and wives and found the patient's impairment to be associated with perceived burden only for the younger wives and older husbands. In contrast, most studies examining behavioral disturbance in relation to the caregiver's physical and psychological health noted strong associations (Wilder et al., 1983; Coppel et al., 1985; Gilleard et al., 1984; Gilleard et al., 1982; George & Gwyther, 1986; Eagles et al., 1987; Deimling & Bass, 1986; Kiecolt-Glaser et al., 1987; Gilleard et al., 1984; Kinney & Stephens, 1989; Morris et al., 1989; Russell et al., 1989). In their study involving 586 caregivers in the Cleveland area, Deimling and Bass (1986) noted that disruptive behaviour had a  19 significant negative effect on the caregiver's physical health and levels of depression as measured by the Zung Scale (1972). Russell et al. (1989) also noted higher scores on the General Health Questionnaire (Goldberg, 1978) to be significantly associated with disturbing patient behaviour as measured on the 34-Item Problem Checklist (Gilleard, 1984). Kiecolt-Glaser et al. (1987) also used the 34-Item Problem Checklist and noted that greater behavioral disturbance was significantly associated with caregiver's level of depression as rated on the BDI. In their study of 131 spouse caregivers, Barusch & Spaid (1989) noted, with a stepwise linear regression model that behaviour problems were the most predictive of caregiver burden (accounting for 14% of the variance). Morris et al. (1989) studied 20 co-resident caregivers and used the BDI to measure depression. Perceived strain was determined using a seven point rating scale. Gilleard's (1984) Problem Checklist was used to measure the behaviour problems of dementia. The severity of dementia was found to be significantly correlated with both caregiver depression and perceived strain. Nevertheless, several studies observed no association between behavioral disturbance and caregiver outcomes (Haley et al., 1987; Zarit et al., 1980). In a study involving 29 caregivers, Zarit et al. (1980) noted that the frequency of behaviour problems as endorsed on a eight item check list was not correlated with caregiver burden. Social Support ^Studies examining social support  20 operationalized support in many different ways, which may account for the inconsistent findings. In quite a number of studies, no effect was determined between social support and caregiver wellbeing, psychological adjustment or depression (Gilleard et al., 1984; Gilhooly, 1984; Pratt et al., 1986; Fiore et al., 1986; Pagel et al., 1987; Gilleard et al., 1984; Robinson, 1989; Barusch et al., 1989; Pruchno, 1990). Gilleard et al. (1984) found that reported amount of contact time and levels of formal and informal outside support were not related to the scoring on the General Health Questionnaire (GHQ). Robinson used the Center for Epidemiological Studies Depression Scale (CES-D), (Radloff, 1977) to measure depression among 78 wives serving as primary caregivers to husbands with irreversible memory impairment. She determined the amount and type of social support received using the Inventory of Socially Supportive Behaviour-ISSB (Barrera et al., 1981). A regression analysis revealed that the level of social support had no significant effect on caregiver depression. In a sample of 44 spouse caregivers of Alzheimer patients, Fiore et al. (1983) considered social network in terms of "perceived upset" or "perceived helpfulness". She found that the perception of social network helpfulness did not relate to severity of depression. However, the degree of upset experienced by the caregiver in relation to network members in any area of support, either as a result of their negative input or their not meeting the person's expectations of support, clearly related to severity of depression. Pagel and Becker (1987) also considered the  21 individuals' overall satisfaction with their social network to be a function of both the positive and negative characteristics of the network. They found that the caregivers' degree of upset with their networks was strongly associated with lower network satisfaction and increased depression. Helpful aspects of the network did, however, interact with network upset in predicting network satisfaction and depression. Other investigators found that higher levels of support were associated with better caregiver outcomes (Zarit et al., 1980; Gilhooly, 1984; Fiore et al., 1986; George & Gwyther, 1986; Scott et al., 1986; Haley et al., 1987; Quayhagen & Quayhagen, 1988; Pagel & Becker, 1987). Gilhooly (1984) noted better physical health in relation to social support for the caregiver. George & Gwyther (1986) noted physical health, stress, affect balance and life satisfaction to be significantly related to the perceived need for more social support. In a study of 58 caregivers, Quayhagen and Quayhagen (1988) found that emotional and informational support as measured with a modified form of Cohen's Dimensions of Social Support Scale (Schaefer et al., 1981) was significantly correlated to caregiver well-being. Pagel et al. (1987) viewed social network helpfulness and network upset individually and found that whereas perceived helpfulness of the social network was not associated with caregiver depression, the amount of upset produced by the social network was positively correlated with depression. Family support was rated as "not enough", "enough", and "more than enough" in a  22 study conducted by Scott et al. (1986) The adequacy of support was coded by two raters using the caregivers' interviews. Both those receiving "not enough" support and those receiving "more than enough" support reported significantly greater burden than those receiving "enough" support. Gender Investigators found conflicting results regarding gender differences in caregiver burden or depression. Fitting et al. (1986) interviewed 54 spouse caregivers using the Burden Interview and the Minnesota Multiphasic Personality Inventory (MMPI- Greene, 1980) to determines levels of burden and various psychological states. In addition, two other measures were used: the Wiggins Depression Scale and the Ego Strength Scale. The results showed that husbands and wives did not differ significantly with respect to burden, but that wives scored significantly higher on the Wiggins Depression scale as well as the depression, hysteria and paranoia subscales of the MMPI. Husbands scored significantly higher on the Ego Strength Scale. In a 1982 study, Zarit found that women experienced higher levels of burden and psychological distress associated with caregiving than men. However, in a follow-up study in 1986, the difference between men and women was not significant. Gallagher et al. (1989) noted a significant effect of gender on the BDI. Specifically, wives and other female caregivers reported more dysphoria than did husbands or other male caregivers. In a study of 68 spouse caregivers, Pagel and Becker(1987) found that gender was correlated with caregiver depression. Female caregivers  23 scored slightly higher than their male counterparts. Barusch and Spaid (1989) studied 131 spouse caregivers and found that gender was related to caregiver burden as measured by the Zarit Burden Interview (1980). Specifically, women caregivers reported experiencing higher subjective burden than did men. In a study with 255 adult children who provided care to an elderly parent afflicted with Alzheimer disease, Chirboga et al. (1989) found that the gender of the caregiver was predictive of depression with the women more likely to report depressive symptoms as measured by a subscale of the Hopkins Symptoms Checklist-90 (HCL90- Derogatis & Cleary, 1977). Using SADS and BDI to measure depression, Gallagher et al. (1989) noted in their study of 158 caregivers, that caregiver wives showed a significantly higher proportion of clinical depression than did caregiver husbands. In a study in which age, gender and relationship were examined for a group of 184 caregivers, Anthony-Bergstone et al. (1989) found that older women also experienced more subscale elevations than their male counterparts. Gilhooly (1984) considered the gender of the patient as a possible correlate of caregiver morale and found that care of a female was associated with higher morale. Better mental health of the caregiver was also associated with care of a female dependant. Moritz et al. (1989) looked at the relation between the level of cognitive functioning in one spouse and indicators of depressive symptomatology in their partners. With the sample of 318 spouse pairs, they determined depressive symptomatology with  24 the Centre for Epidemiologic Studies Depression (CES-D) scale. (Radloff, 1977). Cognitive functioning was measured with the Short Portable Mental Status Questionnaire (SPMSQ- Pfeiffer, 1975). Their analysis showed that, in husbands, mean CES-D scores, as well as the proportion scoring >16 increased significantly in relation to declining cognitive function of the wives. The association was not significant between the wives' CES-D scores and cognitive function of the husbands. Although this appears to imply significant gender difference, the authors urged caution about the interpretation since the effect observed in men is well within the confidence limits of the effect observed in women, indicating the absence of a statistically reliable sex difference. Pruchno & Resch (1989) also examined the effect of greater spouse impairment on caregiver burden. Severity of impairment was measured by caregiver's responses concerning the extent to which 34 behaviours characterized the spouse during the past month. Burden was determined by two measures: a subjective response to a query regarding degree of burden and 17 selected items from those used in literature (Cantor, 1983; George & Gwyther, 1986; Zarit et al., 1986). While the findings showed no significant predictors for burden among husbands, for wives, burden was associated with greater spouse impairment. Patient Age ^Few studies have examined patient age in  relation to caregiver burden or depression. In Gilhooly's (1984) study of 37 caregivers, the age of the patient was not found to  25 be significantly related to the caregiver's morale, as measured by the Kutner Morale Scale. Using the Relatives' Stress Scale (RSS- Greene et al., 1983), Eagles et al. (1987) noted a correlation between the age of the dependents and the caregiver's stress level. Cohen & Eisdorfer (1988) looked at the depression levels (BDI) for 47 relatives of Alzheimer patients and found no significant difference associated with the age of the patient. Living Arrangement Fewer studies have examined the impact that the living arrangement has on the physical and psychological health of the caregiver. In her study of 37 caregivers, Gilhooly (1984) noted that co-resident supporters had slightly lower morale as well as slightly poorer mental health than non-resident supporters. She noted, however, that the co-resident supporters were also older, caring for dependents who were more impaired and were predominantly spouses. George & Gwyther (1986) looked at the relation between patient living arrangements and well-being indicators. They found that although physical health indicators were not significantly related to patient living arrangements, all mental health indicators differed significantly. Those caregivers residing with the patient reported greater use of psychotropic drugs, higher levels of stress symptoms and lower levels of affect and life satisfaction. This finding was similar to that of a 1988 study by Cohen and Eisdorfer in which it was found that living with the patient was significantly associated with caregiver depression levels (BDI). In contrast to these findings, in a study of 68 spouse caregivers, Pagel and Becker  26 (1987) found that living arrangements were not related to depression (BDI). In a recent study, Vitaliano et al. (1991) provided a theoretical framework for simultaneously examining the contributions of stressors, vulnerability and resources. Stressors, in their study, included the care recipient characteristics: cognitive impairment and behavioral ADLs (activities of daily living). Caregiver vulnerability was defined as demographics, health status and personality of the caregiver with respect to anger or expressed emotion. Resources included the social network, coping strategies and the caregiver's outlook on life or overall satisfaction with life. The outcome measured was caregiver distress which consisted of anxiety, depression severity and burden level. They expressed the model as a formula with exposure to stressors and vulnerability as the numerator and psychological and social resources as the denominator resulting in distress or burden. They used their model of distress with 69 dyads of AD patients and their caregivers to predict burden at follow-up 15 to 18 months after the first assessment. They controlled for the caregiver's initial burden score, the patient ADL functioning and the interaction of burden with other components in the model. They found that, in addition to care recipient functioning, caregiver vulnerability and resources were jointly important in predicting subsequent burden. They also pointed out that the interactions between vulnerability and resource, and stressors and baseline  27 burden needed to be considered in order to appreciate the dynamic nature of distress.  Focus on Service Utilization Studies specifically addressing the repercussions of Alzheimer Disease over the past 20 years have generally focused on the impact of the disease on the patient, the primary caregiver and the family system. More recently, an increasing number of investigators attempted to address the issues of the family decision-making process. They related the symptomatology and characteristics of the vulnerable elderly in influencing family burden, institutional outcome or formal healthcare service utilization. Although the majority of these studies did not deal specifically with dementia victims, several investigators have recently examined the unique process for Alzheimer patients and their caregivers. Morycz (1985) studied 80 families, exploring whether the strain experienced by caregivers of relatives with Alzheimer's disease would be strongly related to the desire of the families to institutionalize their older members. He designed a conceptual framework which consisted of three sets of variables, relating patient, caregiver and environment. The three sets of complex characteristics interacted to produce the actual degree of family burden felt by caregiving kin. Translating this perceived burden into strain, he proposed that the strain was critical in the desire of kin to institutionalize patients with  28 Alzheimer's disease.^Therefore, the conceptual framework provided a way to view the decision to institutionalize. In addition to the demographics, patient variables included a degree of impairment of social, economic mental health, physical health and activities of daily living resources as measured by the OARS interview (Pfeiffer,1975). In addition, cognitive impairment was determined using the Mini-Mental State Examination (Folstein et al., 1975). Several scales were created for the study, including a Physical Labour Scale to determine physical needs associated with Alzheimer's disease and the Vigilance/Disruptiveness Scale to determine the number of functional or behavioral symptoms. Caregivers were given the Zung Self-Rating Depression scale (1965,1973) to determine their level of depression. Family burden was measured by a family strain scale and by a global burden question. Environmental factors considered were physical layout, financial considerations and the availability of social support. The desire to institutionalize (the dependant variable in Morycz' study) was measured by a Guttman-type scale which involved six questions concerning the issue of institutionalizing a relative. It was noted in follow-up interviews that the desire to institutionalize was significantly associated with actual institutionalization. Bivariate analysis showed family burden to be moderately associated with the desire to institutionalize. Objective stress as a combination of the scores on physical labour scale and the  29 vigilance/ disruptiveness scale was also moderately correlated with the outcome as was the problem severity score. Most strongly associated with the desire to institutionalize was global burden. Patient marital status and patient age were both significantly related to the desire to institutionalize. Specifically, caring for a married patient was associated with less desire to institutionalize. It was noted that the availability of social support had little relation to the desire to institutionalize. However, social support as measured by the ability to get someone to stay with the patient was negatively associated with family strain, both on the family strain scale and global burden question. The multivariate analysis indicated that family strain explained 19% of the variance in desire to institutionalize. Using a stepwise hierarchical regression, Morycz noted that a total of 38% of the variance was accounted for with four predictor variables; family strain, marital status of the patient, living arrangement of the patient and the physical labour required because of the patient's functional deficiencies. A separate stepwise regression on family strain revealed that the availability of support accounted for 53% of variance in strain levels. An additional eight percent of the variance was explained by the caregiver's perception of the problem severity, the patient's behaviours and functional deficits. Therefore, Morycz concluded that the strongest predictor of the caretakers' desire to institutionalize is family strain and the best set of  30 predictors for strain is the availability of social support. Colerick and George (1986) conducted a longitudinal study involving 209 caregivers to address the question of why some families continue to shoulder the burden of care, often beyond healthful limits, while others relinquish care to professionals. At Time One they assessed the patients, caregivers and support systems. Patient characteristics included demographics, the diagnosis, duration and severity of the illness as well as living arrangements and relationship to the caregiver. Caregivers were assessed with respect to their physical health (including the use of psychotropic drugs) and their stress level as determined by the Short Psychiatric Evaluation Schedule (Pfeiffer, 1979) as well as a 22 item Screening Scale (Langer, 1962). They were also asked about their participation in social and recreational activities and their perceived economic status. Social support at Time One consisted of the frequency of 11 types of social support and caregiving assistance from family and friends. At Time Two these were scored separately. Two logistic regression analyses were performed. The two groups were caregivers which continued to provide homecare for the patient or those who subsequently chose to institutionalize the patient. The first regression analysis was longitudinal, determining which of the factors measured at Time One predicted the membership into either group at Time Two. The second, cross-sectional analysis served to determine the extent to which group membership was related to patient and caregiver  31 characteristics measured at Time Two. A comparison of the means on the predictor variables indicated that those who subsequently institutionalized their patients were most often female, employed, and were among the youngest in the sample. They were more often children than spouses, had above average incomes and reported high levels of stress and dissatisfaction with time spent in recreational pursuits. This group also reported receiving more social support and having a greater knowledge of services available to them. At Time Two the group means indicated that although the caregivers who sought institutional placement for their relative reported an increased satisfaction with leisure, they experienced less support from friends than the other group. In addition, the caregivers who institutionalized their patients were more likely to use psychotropic drugs than at Time 1. The authors speculated that the social isolation and increased use of drugs could be an indication of lingering effects of pre-institutional burden. The logistic regression analyses at Time One was significant and suggested that two factors greatly reduced the probability of institutional placement: the relationship of the caregiver to the patient and the caregiver's need for caregiving assistance. Though not significant, the antilogged coefficient for patients' sex at Time One suggested that the odds of institutional placement were much greater for female patients. Since many of the female patients were being cared for by employed daughters this was understandable.  32 At Time Two, the coefficients from the logistic regression again showed the importance of the relationship. Those who relinquished care were also significantly more satisfied with time available for recreation, although those continuing homecare reported a higher level of general well-being. Obvious by their absence, patient characteristics (length of illness, and symptoms and their severity) did not emerge as a significant predictor of institutionalization. Caserta et al. (1987) conducted a study with 597 family caregivers to non-institutionalized dementia patients. They examined service utilization patterns and the factors which influence utilization. They limited their study to a review of three services since these were reported most frequently by caregivers: respite, daycare and in-home care. Factors which were considered as possible correlates of service utilization were patient characteristics, caregiver characteristics and social support. The patient level of functioning was measured using the Functional Dementia Scale (Moore et al., 1983). The Zarit Burden Interview (1980) was used to assess caregiver burden. The dimensions of informal support were measured by network size, frequency of contact with network members and perceived satisfaction with the support received. For their correlation, responders which reported using any of the three services were defined as utilizers. The nonutilizers were divided into two groups, those not yet ready to utilize and those not utilizing for other reasons. In their  33 comparison of those who utilized and those who did not utilize services for either reason, statistical differences were observed on functional dementia, burden and informal social support. Those not utilizing services because they felt they were not yet ready to utilize reported significantly lower burden and functional dementia but higher levels of social support. They were also significantly younger than caregivers in the other groups. Relationship to the patient was significantly different between the group not yet ready to utilize services and the other groups. A significantly higher percentage of those not yet ready to utilize services were spouses. This was interesting in light of the finding that spouse caregivers had slightly higher burden scores than non-spouse caregivers. In a study not limited to Alzheimer victims, Bass and Noekler (1987) studied 586 care recipients who were at least sixty years of age and lived with either a spouse or an adult child and who required help with one or more personal care tasks. Due to exclusion criteria, the caregivers interviewed for the study consisted of 52% spouses and 48% daughters or daughters-inlaw. The outcome measures were the use of in-home nursing over the six months preceding the interview and the number of hours of service. Factors considered in relation to the outcome consisted of predisposing, enabling and need characteristics of the patient and the caregiver. The model they used was a version of the utilization model developed by Anderson & Newmann (1973), expanded to include characteristics of the caregiver as well as  34 the network of social support. The predisposing variables included both the elder's and the caregiver's age and race as well as the relationship of the caregivers to the elders. Enabling characteristics examined were household income, and the involvement of family members measured by perception of supportiveness, number of helpers and the number of tasks which the helpers performed. Three aspects of the elder's health were considered as need characteristics: physical health, mental health and common impairments among the elderly. Caregiver need consisted of reported activity restriction, change in physical health and caregiving burden as determined by the number of difficult and tiring tasks noted. In addition, the caregiver's level of depression was measured using the Zung SelfRating Depression Scale. The first analysis of the predictors of use of services was done with a four-stage regression model in spite of the dichotomous nature of the outcome variable because of the equal numbers of those using and those not using services. Emerging as significant predictors of service use were the predisposing variable of race and two enabling variables: household income and the number of tasks performed by other helpers. Also significant were two elder need factors: presence of incontinence and paralysis as well as caregiver need variables; activity restrictions, health deterioration and perceived task burden. In total, however, these variables only accounted for 15% of the variance.  35 Since 51% reported using services, the amount of service use was calculated on that subsample. The four-step regression analysis predicting the amount of service used showed no predisposing or caregiver need characteristics as having a significant effect. Instead, several enabling variables emerged as significant: household income and, less significantly, the number of tasks performed by the helpers. Mental impairment was one elder need characteristic which was also a significant predictor of the amount of service used. In total, the regression model accounted for 20% of the variance. In a recent study, Biegel et al. (1990) focused specifically on 170 caregivers of mild to moderate non-institutionalized dementia patients. Using the modified version of Anderson & Newmann's conceptual framework (Bass & Noekler, 1987), they examined predisposing, enabling and need characteristics of the patients and their caregivers as predictors of utilization of services. Of the 170 caregivers, 83 reported using some services whereas 87 were non-users. They divided their services into inhome and out-of-home services. By their reports of use, caregivers fell into one of four groups: those not using any services, those using only in-home services, those using only out-of-home services and those using both in-home and out-of-home services. Of those reporting using services, percentages using the inhome services were as follows; Meals-on-Wheels (33%) in-home nursing (24%) housekeeping service (22%) and home health aide  36 (22%). Of the users, those using out-of home services reported using transportation (46%), adult day care (29%) and senior centre meals (17%). The patient need characteristics consisted of the impairment level as determined by the MMSE score and ADL dependencies with respect to seven items. Relationship strain and emotional strain were measured separately, with five and three items respectively. Using discriminant analysis, they found that the non-users and those using only in-home services reported inadequate informal supports, more emotional strain and patients with greater functional dependencies than those using out-of-home services or both types of services. Those using both types of services were more often female and reported higher levels of dyadic relationship strain than those caregivers who used only out-of-home services. The authors concluded that the findings showed that informal support facilitated the use of out-of-home services exclusively or in combination with the use of in-home services. They also noted that emotional burden and relationship burden differentially predicted service use, indicating that burden was a multi-dimensional concept which should not be measured univariately.  Summary of Literature Review The variables considered in the caregiving literature have not produced consistent results. This is not surprising in light  37 of the broad range of outcome variables considered and the different ways of operationalizing concepts related to patient and caregiver characteristics. Patient illness has been measured in numerous ways. In some studies illness severity has been assessed as a single state while in others it has been assessed using several separate dimensions, each rated individually. The impact of overall illness severity or of one particular dimension of the illness understandably related differently to outcomes of caregiver strain or health service utilization. Measurements to determine caregiver burden or depression levels also varied greatly. Both unidimensional and multidimensional scales were used to determine caregiver physical and psychological states such as strain, stress, depression, burden or well-being. This has produced wide variation in scoring and interpretation. Another complicating factor in the literature was the overlapping nature of the variables examined. For example, studies which examined the impact of kinship tie on caregiver strain or on the propensity to utilize formal support services were dealing with the confounding effect of age and sex. Unless these variables were held constant, the outcomes were less than conclusive. The nature of Alzheimer's disease also posed a unique challenge to investigators. Symptoms changed over time, impacting caregiver burden levels. Pruchno & Resch (1989) suggested that due to the nature of the symptoms over the course of the illness, caregiver burden did not follow a linear trajectory, but peaked  38 in the middle phases of Alzheimer's disease when behavioral disturbances became the most problematic. Emerging out of the literature on caregiving and use of support service is the sense that caring for a dementia victim is a complex and ever changing experience. It remains a challenge for investigators to capture the dynamic nature of the illness which impacts the caregiver level of burden and decisions regarding support services differently at the various stages of the illness. Baumgarten (1989) recommended more rigorous study designs, including features such as comparison groups, longitudinal follow-up and use of representative study samples. She also recommended that the changes over time in all variables be examined, noting the intricate association between these changes.  39 CHAPTER THREE METHODOLOGY  Introduction  The variables for consideration in this retrospective study were taken from those previously recorded by the Clinic for Alzheimer Disease and Related Disorders, U.B.C. (CADRD) and the Continuing Care Division of the Ministry of Health in B.C. Information at the Alzheimer Clinic is gathered by a multidisciplinaryteam consisting of a geriatrician, psychiatrist, neuropsychologist, neurologist, speech pathologist and social worker using standardized questionnaires which are either completed by the caregiver or used during interviews with the primary caregiver and for the patient. As a result, the data is considered to be reliable, and not reflective of the bias of particular interviewers. From this database, the predictor variables which I examined were categorized as predisposing, enabling and need variables, in keeping with the study framework chosen. As indicated in the section on background, the majority of these variables have been considered in relation to the level of burden experienced by the caregiver. More recently, studies have examined several of these in relation to the utilization of support services. Service utilization information was kept in a computerized database at the Continuing Care Division of the Ministry of  40 Health, Victoria, B.C. or at the Continuing Care Divisions in the various Health Units across British Columbia.  Research Design The study uses a retrospective cross sectional exploratory design. Data regarding patients and their caregivers gathered at the point of the first assessment at the CADRD was examined. In addition, the utilization patterns of CC services was analyzed for the six months prior to the respective dates of the assessments. Rather than testing an hypothesis, an exploratory stance was assumed in order to allow for relevant patient and caregiver characteristics to emerge as predictors of CC utilization.  Sampling Procedure The Clinic for Alzheimer Disease and Related Disorders at the University Hospital, U.B.C. Site, provides assessment for patients with memory or other cognitive impairment who are referred from around the province. Over the past 6 years, the Clinic has provided comprehensive assessment services to over 800 patients with suspected dementia, many in the early stages and most of whom reside at home in the community. Each patient registered at the Clinic is assessed by members of the multidisciplinary team. Demographics, relevant clinical and diagnostic information are entered into a centralized computer database as the clinic. The database also notes if each patient  41  is registered with CC. An investigation conducted by the Clinic (Scott, Kosarek & Tuokko, 1988) indicated that approximately 61 percent of Clinic patients living with a spouse caregiver had not been assessed by the Continuing Care Program and were not utilizing any formal support services. Therefore, in order to permit a comparison between those who utilized services and those who did not, two samples of relatively equal numbers were drawn from the Clinic files. Criteria for inclusion in the initial sample was a diagnosis of possible or probable dementia of the Alzheimer type, using NIACDS ADRDA criteria, and reported registration with CC. Patients who had died within 3 months of the assessment date were excluded. Because of the importance of the Zarit Burden Interview, the initial Clinic assessment date of March 1, 1989 was used as a starting point since the burden score was consistently noted as of that date. From March 1, 1989 to June, 1992 there were approximately 90 patients who had indicated registration with CC. For purposes of comparison, an equal number of names were then subsequently drawn of those not registered with CC, using the same initial date. The client files were then reviewed to ensure that all predictor variables were recorded. Those with several items incomplete or missing were excluded from the sample. Ensuing contact with the caregivers of the registered group to obtain consent revealed that several had incorrectly indicated registration. The resulting useable sample  42 consisted of 75 registered with CC and 83 not registered with CC.  Measurements The details of the predictor variables were obtained from the patient files at the CADRD. The information had been gathered by members of the multidisciplinary team, using both unidimensional and multidimensional measurements. The predictor variables and their measures are listed below:  Patient Predisposing Characteristics Age Gender Place of Birth Canada UK Other Unknown Marital status Single Married Widowed Divorced/Separated Household composition Lives alone Lives with spouse Lives with spouse & children Lives with children Other living arrangements Duration of illness Measured in months  43 Caregiver Predisposing Characteristics Age Gender Relationship to patient Spouse Child Sibling Other  Caregiver Enabling Characteristics Degree of Involvement (unidimensional measure) Constant and living with patient Daily but not living with patient Weekly Infrequent  Family Support Enabling Characteristics Amount of Support (unidimensional measure) Daily contact with family/friends Regular weekly contact with family/friends Regular monthly contact with family/friends Infrequent contact with family/friends No contact at all Perceived Lack of support from family/friends (unidimensional measure) Not at all Sometimes Frequently  Patient Need Characteristics Diagnosis of Alzheimers Possible Probable  44 Patient Need Characteristics, continued... Concurrent Illnesses Medical^e.g. Hypothyroidism Diabetes Mellitus Psychiatric^e.g. Alcoholism Depression Neurological^e.g. Parkinson's Disease Mentally handicapped Cardiovascular Accident None significant Rates of Decline (formulated measure) Overall Decline Formula=(FRS Total/Duration) Cognitive Decline Formula=(Sum of Cog FRS Subtotals /Duration) Cognitive FRS areas: memory, language, orientation and problem solving & reasoning. Social Decline Formula=(Sum of Soc FRS Subtotals /Duration) Social FRS areas: social & occupational, Functional Rating scores (FRS) Cognitive Subscales: Memory (1-5) Language (1-5) Problem Solving/Reasoning(1-5) Orientation (1-5) Social Subscales: Social/Occupational (1-5) Home/Hobbies (1-5) Personal Care (1-5) Affect Subscale (1-5) Total Score Range: (8-40) Present Functioning score (Total & Subscales) Personality Subscale (0-15) Everyday Functioning Subscale (0-12) Language Skills (0-9) Memory (0-15) Self-care (0-14) Total Score (0-65)  45 The Functional Rating Scale (FRS) was developed at the Clinic for Alzheimer Disease and Related Disorders to assist in establishing a diagnosis of dementia and monitoring patient changes in functioning over time. (Appendix A) The scale uses a multidimensional approach, focusing on 8 areas of psychosocial functioning; memory, social/occupational, home/hobbies, personal care, language, problem solving/reasoning, affect and orientation. These areas of functioning are consistent with those delineated as central to the diagnosis of dementia (e.g. DSM-3, ICD-9). The ratings were made on the basis of medical history, clinical examination and information from a collaborative informant. The multidisciplinary team rated the patient's level of psychosocial functioning in each of the 8 areas along a 5 point scale ranging from healthy (1) to severely impaired (5), allowing for a maximum total score of 40. This scale has been tested and has demonstrated very good interrater reliability. (Tuokko, et al., 1986) Scores for the subscales were evaluated as was the total score in relation to the outcome indicators. The Present Functioning Questionnaire (PFQ) was constructed at the Alzheimer Clinic by reviewing the research literature associated with problems of everyday functioning in the elderly (Appendix B). Sixty-five items were developed to reflect the deficits reported in this population and were organized into five problem areas: personality (15 items), everyday functioning (12 items), language skills (9 items), memory (15 items) and self-  46 care (14 items). Items were scored on a presence or absence basis, yielding a score out of a possible 65. This questionnaire has been tested and demonstrated adequate reliability and discriminant validity (Tuokko & Crockett, 1991). Analysis of the total as well as each of the 5 problem areas, was done in relation to the outcome measures. Caregiver Need Characteristics b) Caregiver i) Feeling stress (unidimensional measure) 1. Not at all 2. Sometimes 3. Frequently ii) Experiencing depression (unidimensional measure) 1. Not at all 2. Sometimes 3. Frequently iii) Burden score (Total & Individual Items) Consists of 22 items with Scores (0-4) Total Score Range = (0-88) The Burden Interview (Gallagher, et al., 1985) is composed of 22 items that assess the degree to which caregivers perceive their responsibilities as having an adverse effect on their health, personal and social life, finances, and emotional wellbeing (Appendix C). The measure provides a direct assessment of the caregiver's perception of the impact their involvement with the patient has on his or her own life. Prior studies have demonstrated good reliability and validity of this instrument. The 22 questions were answered on a scale of 0 to 4 (never to nearly always) allowing for a total score of 88 for the caregiver  47 experiencing maximum stress levels. As a measure of overall burden, Zarit Burden Inventory has recently been challenged as possibly masking "dimension-specific patterns of caregiving impact, precluding identification of the different antecedents or correlates of specific dimensions of caregiver burden" (George & Gwyther, 1986). Several researchers have designed multidimensional instruments in an attempt to uncover variability in caregiver strain (Niederehe & Fruhe, 1984; Kosberg & Cairl, 1986; Novak & Guest, 1989). Novak & Guest (1989) produced subscales for 5 dimensions of burden: timedependence burden, developmental burden, physical burden, social burden and emotional burden. Use of this scale revealed variability in the different dimensions of burden experienced by caregivers, in spite of similar overall scores. Although this study was restricted to data previously gathered using the unidimensional scale, analysis of individual items of the scale was done in the interest of determining specific burden aspects and their relation to the utilization of support services. In addition, a factor analysis was performed to explore the dimensionality of subjective burden.  Data Collection Details regarding the predictor variables were obtained through the database and the patient files at the Clinic for Alzheimer Disease and Related Disorders at the University Hospital, U.B.C. Site.  48 The outcome indicators (dependent variables) of interest were: 1. Registration with the CC Division of the Ministry Health in British Columbia. 2. For those registered with CC, record of the utilization of several services: Homemakers, Meals-on-Wheels and Adult Daycare. Permission to gather the information regarding the utilization of services for each of the patients indicating registration with CC, was obtained from the respective caregivers. To obtain permission, consent forms were mailed to their homes (Appendix D). Follow-up phone calls were then made to ensure that the forms had been received and that the intent of the request had been understood. Upon receipt of the signed consent forms, a list of the names, along with the consent forms and the dates of interest were sent to the Ministry of Health, Continuing Care Division and the respective Meals-on-Wheels centres across B.C. The Continuing Care Division compiled the information for homemaker service and adult daycare for the respective six month periods and forwarded a printout of the utilization details for each patient..  Ethical Considerations A request for ethical review was submitted to the Ethics Committee at the University of B.C. with a summary of the purpose  49 and objectives of the project. A copy of the consent form was included with the application. In the consent form, the participants were given a description of the study and the nature of their involvement. All participants were explicitly informed that they were not obligated to participate in the study and could withdraw or refuse to answer any questions without their future access to services offered by CC being jeopardized. They were also assured confidentiality, and told that they would be contacted by phone to answer any questions they might have concerning the study. Permission was granted from the ethics committee to proceed with the study.  Data Analysis  The statistical package, SPSS-X (1988) was used for all analyses. Descriptive analyses were performed for the sample, while parametric and non-parametric procedures were used for comparisons of the groups.  1. Analysis of the entire sample (N=158) A descriptive analysis was undertaken for the entire sample with respect to the predictor variables: predisposing, enabling and need characteristics of patients and caregivers.  2. Correlation Matrix The relationship between predictor variables and the outcome measures was examined using Pearson correlation and non  50 parametric correlation matrixes as appropriate.  3. Comparison of the two groups from the entire sample: Registered with CC (n=75) Not registered with CC (n=83). a. A descriptive analysis was done for each of the two groups of the predictor variables: predisposing, enabling and need characteristics of patients and caregivers. b. A comparison of the groups with respect to the predictor variables was undertaken using T-tests or Chi-squares, as appropriate.  4. Analysis of the 2 Sub-groups in the Registered sample: Utilizing services (n=20) Not utilizing services (n.55). Initially the intention was to divide the sample into 4 groups, those using in-home services only, those using out-ofhome services only, those using both in-home and out-of-home services and those not using any services at all. Due to the small numbers of those reporting use of each type of service, it was decided that, to be meaningful, those using any services at all would have to be considered as one group for comparison with those not using services.  a. A comparison of the user and non-user groups was performed using T-tests or Chi-squares as appropriate with  51 respect to the predictor variables: predisposing, enabling and need characteristics of patients and caregivers.  5. Burden of Care Analysis The data from the Zarit Burden Interview were used in the following analyses: a. Factor analysis with a varmiax rotation was performed on 22 items of the Burden Interview b. Factors were interpreted and labelled in terms of their item content. c. The reliability of the scales was assessed using Cronbach's alpha coefficient. d. The scales were analyzed in relation to group membership (registered vs. non-registered with CC.)  Limitations 1. Generalizability uncertain due to a sample from a diagnostic referral clinic. 2. Retrospective data used for patient and caregiver characteristics limited to variables previously recorded. 3. Causation undetermined in cross-sectional study. 4. Information re service utilization with CC limited to the six months prior to assessment at the Clinic.  52 CHAPTER FOUR RESULTS & DISCUSSION  Introduction  This chapter contains five sections. The first section provides the descriptive analysis of the sample of 158 caregiver and care recipient dyads. Pearson Correlation and non-parametric correlation coefficients for the predictor variables are discussed in the second section. Descriptive and comparative analysis is then discussed for the two groups which were formed based on their registration status with CC (registered or not registered). This analysis is supplemented with a regression analysis of the predictor variables on group membership. Fourthly, a closer look at the registered group compares those actually using services and those not using services. The initial analysis using T-tests and Chi-squares is also supplemented with a regression analysis of the predictor variables. Although a loglinear analysis was performed, it failed to converge, producing less than meaningful coefficients. Therefore, the results of the analysis are not presented nor discussed in this section. Finally, the measure used to assess burden (Zarit,1980) is analyzed to determine how subscales might uniquely predict the outcome of registration with CC.  53 Descriptive Analysis of Sample The final size of the sample consisted of 158 patients and their respective caregivers. Predisposing, enabling and need characteristics for both the caregivers and the patients were contained in computerized database. A descriptive analysis of patient and caregiver predisposing characteristics are discussed below and the frequencies are displayed respectively on Tables 1 & 2.  Predisposing Characteristics Predisposing Characteristics of Patients Predisposing characteristics of the patient entailed information concerning age, gender, place of birth, marital status, living arrangements and duration of the illness. The age of the patients ranged from 48 to 86 years with an average age of 72 (SD=7.2). With respect to duration of the illness, the minimum duration was 6 months with the maximum being 264 months or 22 years. The average term of illness was 64 months (SD=38). The minimum and maximum duration figures are misleading; in actuality, the period of time for 50% of the patients was less than 5 years. Only 4% of the sample had an illness term which exceeded 12 years. Fifty-eight percent of the care recipients had a duration of between two and five years. As displayed on Table 1, sixty-two percent (n=98) of the patients were females. Due to the many countries cited as place of birth, the sample was categorized into being born in Canada,  54 UK or another country. The majority of the patients were born in Canada (61% - n=97) while 11% (n=18) were born in the UK. Eleven percent (n=17) of the sample did not indicate their place of birth, leaving just 16% (n=26) with another birthplace. Only one person in the sample was single, with 76% (n=121) indicating that they were married (common-law arrangements were included in the married category); twenty-one percent (n=33) of the care recipients were widowed, while only 2% (n=3) were divorced or separated. Sixty percent (n=95) of the patients in the sample lived with a spouse, while 6% (n.10) lived with a spouse and children. Another 3% (n=5) lived with children, while 12% (n=19) indicated other living arrangements. The remainder of the sample,(18% - n=29) lived alone.  55 Table 1 Descriptive Frequency Analysis of Patient Predisposing Characteristics for the Sample (N=158) Gender  Frequency  Male^ Female^ Place of Birth Canada UK Other Unknown Marital status Single Married Widowed Divorced or /Separated  60 98  38% 62%  Frequency 97 18 26 17  61% 11% 16% 11%  Frequency 1 121 33  1% 76% 21%  3  2%  Household composition Lives Lives Lives Lives Other  Frequency Alone 29 with Spouse 95 with Spouse & Children 10 with Children 5 living arrangements 19  18% 60% 6% 3% 12%  56 Predisposing Characteristics of Caregivers Caregiver's predisposing characteristics consisted of age, gender and relationship to the patient. The age of the caregivers ranged from 23 to 87 years, with an average of 64 years (SD =13.6). As shown on Table 2 below, the caregiver sample was comprised of an equal number of males and females (n=79). Seventy-three percent (n=115) of the caregivers were spouses to the patients, while 22% (n=35) were children and only 5% (n=8) were siblings. More of the spouses were men (61 men to 54 women) while a greater percentage of the children were daughters (22 daughters to 13 sons). Of the eight siblings, five were brothers while 3 were sisters. There were no caregivers related to the patient in other ways, such as friends or distant relatives.  Enabling Characteristics Enabling^characteristics^pertaining^to^caregivers incorporate the degree of involvement the caregiver had with the patient. This degree of involvement was assumed to buffer the patient from the need for formal support services. As noted on Table 3 below, of the 158 caregivers, 77% (n=122) lived with the patients and therefore indicated that they were involved with them constantly. Twenty percent (n=32) reported not living with the care recipients. Of those not living with the patients, 46% (n=15) reported having daily contact with them, 50% (n=16) reported weekly contact and only 3% (n=1) indicated having ^ infrequent contact^(information ^was^missing  57 Table 2  Descriptive Frequency Analysis of Caregiver Predisposing Characteristics for the Sample (N=158)  Gender Male^ Female^ Relationship to patient Spouse Child Sibling Other  Frequency^% 79^50% 79^50%  Frequency^% 115 35 8 0  73% 22% 5% 0%  58 Table 3 Descriptive Frequency Analysis of Caregiver Enabling Characteristics for the Sample (N=158) Degree of Involvement Living with patient Constant Not living with patient Daily Weekly Infrequent (missing information)  Frequency 122  77%  15 16 1  10% 10% 1%  4  Amount of Support /Contact with Family /Friends Frequency Daily contact Regular weekly Regular monthly Infrequent contact No contact at all  89 63 3 1 1  56% 40% 2% 1% 1%  (missing information)^1  Perceived Lack of Support Not at all Sometimes Frequently (missing information)  Frequency 102 38 10 8  65% 24% 6%  59  for four caregivers of the sample). The amount of support received from family or friends in the form of contact frequency was considered an enabling characteristic. The perception of lack in this area by the caregiver was also regarded as a characteristic possibly serving to encourage the utilization of formal support services. Fiftysix percent (n=89) of the caregiver sample indicated having daily contact with family or friends. Another 40% (n=63) reported have regular weekly contact. Two percent (n=3) had regular monthly contact, leaving only 1% (n=1) with infrequent contact and 1% (n=1) with no contact at all. Although 65% of the caregivers indicated in the interview that they did not perceive lack of support from family and friends, 24% (n=38) indicated that they sometimes did feel this way, while another 6% (n=10) expressed frequently feeling that they were lacking in support (information was missing for eight of the caregivers). Need Characteristics Need Characteristics of Patients Need characteristics of the patients in the files consisted of the medical diagnosis of Alzheimer Disease, concurrent illnesses, and their scores on the Functional Rating Scale (FRS) and the Present Functioning Questionnaire (PFQ). Frequencies of diagnosis of Alzheimer Disease and concurrent illnesses are displayed on Table 4. Of the patients in the sample, 42% had been assessed with possible AD and 58% had been  60 Table 4  Descriptive Frequency Analysis of Patient Need Characteristics for the Sample (N=158)  Diagnosis of Alzheimers Possible^ Probable^  Frequency 67^42% 91^58%  Concurrent Illnesses  Frequency Medical^ 53^34% 24^15% Psychiatric^ Neurological^ 34^22% No significant^ 46^29% (missing information)^1  61 diagnosed as having probable AD (consistent with the NINCDS & ADRDA criteria for diagnosis of Alzheimer Disease). Thirty-four percent (n=53) of the patients were diagnosed as having noteworthy concurrent medical illnesses in addition to their dementia, such as hypothyroidism or diabetes mellitus. Psychiatric conditions such as alcoholism or depression were noted in 15% (n=24) of the sample, while 22% (n=34) were suffering from other neurological problems, such as Parkinson's disease, mental handicap or evidence of a cardiovascular accident (CVA). Twenty-nine percent (n=46) of the sample did not have any significant concurrent illness indicated in their file. Scores on the Functional Rating Scale and the Present Functioning Questionnaire were considered as need characteristics. The average, range and standard deviations of scores for the sample are contained in Table 5. The Functional Rating Scale was multidimensional, including scores for eight areas of psychosocial functioning as well as a total score based on team perception. The patients were rated by all members of the multidisciplinary team on a scale of 1 (healthy) to 5 (severely impaired) for each of the eight areas assessing their functional ability: memory, social and occupational, home and hobbies, personal care, language, problem solving and reasoning, affect, and orientation. The total score could range from 8 to 40. The mean score for the sample of caregivers in this study was 27.5 (SD=5.2) with a minimum score of 18 and a  62 Table 5 Descriptive Analysis of Patient Need Characteristics for the Sample (N=158) Functional Rating scale  Memory Language Problem Solving /Reasoning Orientation Social /Occupational Home/Hobbies Personal Care Affect Subscale Total Score  Mean  SD  Min  Max  4.0 3.1  .7 1.1  3 1  5 5  3.7 3.4  .8 1.0  2 1  5 5  3.6 3.4 2.5 3.8  .7 .8 1.2 1.0  2 1 1 1  5 5 5 5  27.5  5.2  18  40  Mean  SD  Min  Max  6.1 6.9 3.4 9.0 1.9  3.1 3.0 2.5 2.9 2.7  0 0 0 1 0  13 12 9 15 13  2  52  Present Functioning Questionnaire  Personality Everyday Function Language Skills Memory Self-care Total Score  63 maximum score of 40. The highest average score for the patients occurred in the realm of memory with a mean of 4.0 (SD=.7). Other means over 3 were affect (mean=3.8, SD=1.0), problem solving and reasoning (mean= 3.7, SD=.8), social and occupational (mean=3.6, SD=.7), orientation (mean=3.4, SD=1.0), home and hobbies (mean=3.4, SD=.8) and language (mean=3.1, SD=1.1). Personal care had the lowest mean of 2.5 (SD=1.2), indicating the least impairment for the sample in this area. Results on the Present Functioning Questionnaire also indicated the least problem with the area of self-care a perceived by the caregiver and reported in a structured interview. The PFQ consisted of 65 possible deficits related to dementia which were organized into 5 problem areas: personality (15 items), everyday functioning (12 items), language skills (9 items), memory (15 items) and self-care (14 items). Out of a possible total of 15, the mean number of deficits related to personality was 6.1 (SD=3.1) Everyday functioning (12 items) had a mean of 6.9 (SD=3.0) and language skills (9 items) had a relatively low mean of 3.4 (SD=2.5). As in the FRS, the average score for memory (15 items) was the highest with a mean of 9.0 (SD=2.9). The overall total PFQ scores ranged from 2 to 52 with a mean of 27.0, (SD=10.4). Need Characteristics of Caregivers Frequency of caregiver's feelings of stress or depression were need characteristics noted in the patient file. The Zarit Burden Interview was also recorded, reflecting individual item  64 scores and the total score. Frequencies for the unidimensional scales of stress and depression are contained in Table 6. The unidimensional measure of stress basically consisted of a single question: How often do you feel stress due to health, financial or interpersonal pressures? Response options were: not at all, sometimes or frequently. Thirty-five percent (n=55) of the caregivers indicated that they did not feel stress at all, while another 34% (n=54) reported sometimes feeling stress. Twenty-four percent (n=38) described frequently feeling stress (information was not obtained from the remaining 11 subjects of the sample). The rating for depression was also unidimensional, involving the question: How often do you experience depression? Again, the answer options were: not at all, sometimes or frequently. Fortyeight percent (n=73) indicated the they were not experiencing depression at all. Thirty-eight percent (n=60) stated that they sometimes experienced depression, while another 11% (n=18) admitted to frequent depression (seven caregivers did not provide answers to the depression question). The Burden Interview was filled out by the caregivers during their interview as well. It consisted of 22 items which were statements of burden-causing phenomenon related to caring for an impaired person. Answers for the individual items could range from 0 (never) to 4 (almost) all of the time. The total score then could range from 0 to 88. The scale was also  65 Table 6 Descriptive Frequency Analysis of Caregiver Need Characteristics for the Sample (N=158) Stress Frequency  Frequency^%  Not at all^ 55^35% Sometimes^ 54^34% Frequently^ 38^24% (missing information)^11  Depression Frequency  Frequency^%  Not at all^ 73^48% Sometimes^ 60^38% Frequently^ 18^11% (missing information)^7  66 unidimensional, producing one overall burden total. The average score on the Burden Interview was 27.7 (SD=14.9). The recorded scores ranged from 0 to 70, with 90% of the caregivers scoring below 49.  Correlation Matrix of Predictor Variables Correlations were performed between the predictor variables to note associations. Table 7 displays the correlations as calculated using the Pearson Correlation or Spearman Rank Correlation as appropriate. Among the predisposing characteristics, the expected strong associations existed between patient and caregiver gender, marital status and relationship. It was interesting to note that the duration of illness was not related to any of the other variables, including the age and gender of the patient. Correlations between the enabling characteristics are outlined below on Table 8. Enabling characteristics included degree of caregiver involvement, amount of family support and the perceived lack of support. The enabling characteristics were not strongly correlated with each other. Table 9 indicates the degree of correlation between the need variables. Many of the need characteristics of the patient and the caregiver were highly correlated. The Functional Rating Scale was positively related to the Present Functioning Questionnaire score (p<.001) indicating a significant agreement between the  ^  67 Table 7  Correlation of Predisposing Characteristics 1^2^3^4^5^6^7^8  1. Age of Patient 2. Patient Gender 3. Marital status  .03  -  -  * * *^* * .33^.22 ^  * 4. Household ^composition .11^.09^.10 5. Duration of illness^-.03 -.09^.04^.07 ^***^*** 6. Age of Caregiver^.27^.06 -.46 -.01 -.08 7. Caregiver Gender^.12 -.63^.05^.13^.07 -.23 ^* * * 8. Relationship ** to patient^.24^.31^.84^.10^.05 -.51^.09  Note: *^p < .05 **^p < .01 *** p 7 .001  68 Table 8 Correlation of Enabling Characteristics 1  ^  2^3  1. Degree of Involvement 2. Amount Support^.13 * 3. Perceived Lack -.03 .12 Note: p < .05  69 Table 9  Correlation of Need Characteristics 1  2  3  4  5  6^7  1. Diagnosis 2. FRS Total  .09  *** 3. PFQ Total  .03  .70  4. Concurrent Illnesses  .03  * .16  .11  .02  *** .28  *** .35  6. Caregiver Depression -.02  ** .20  *** .28  7. Burden Interview  *** .33  *** .49  5. Caregiver Stress  Note: * ** ***  p < .05 p < .01 p < .001  -.02  .08  .01  *** .51  .06  *** .54  *** .42  70 objective assessment of the multidisciplinary team the subjective perception of the caregiver surprisingly, the  FRS  (FRS)  and  (PFQ).^Not  was also significantly correlated with  caregiver stress, burden level (p<.001), and depression (p<.01). The  PFQ  was also highly associated with caregiver stress,  depression and burden levels (p<.001). In view of the concentration of earlier research on caregiver health outcomes such as caregiver burden, depression and stress, it was interesting to note the association between these characteristics and other factors in this study. Tables 10, 11 and 12 list the correlations between the measures of burden, depression and stress respectively. Caregiver burden was highly associated (p<.001) with: perceived lack of support, lower levels of functioning as indicated on the  FRS  and the  PFQ,  as well as caregiver sex.  Those reporting a greater perception of a lack of support also reported higher levels of burden. Caregivers of patients with lower levels of functioning as indicated on both the  FRS  and  PFQ  also experienced more burden. The correlations also demonstrated that women report a higher degree of burden than their male counterparts. Not surprisingly, burden was also significantly associated (p<.001) with caregiver stress and depression. A significance level of p<.01 was realized in a negative association between burden and the sex of the patient, indicating that caregivers of male patients were more burdened. Caregiver depression (see Table 11) was also highly  71 Table 10  Correlates of Caregiver Burden  Characteristics^Correlation Coefficient (rho) Patient Age^  -.01  Patient Gender^  -.22**  Marital status^  -.01  Household composition^.02 Duration of illness^ .10 Caregiver Age^ Caregiver Gender^  -.18* .27***  Relationship to patient^.01 Caregiver Involvement^-.03 Amount of Family Support ^.07 Perceived Lack of Support^.31*** Diagnosis of Alzheimers^-.00 Functional Rating Score ^.33*** Present Functioning score^.49*** Concurrent Illnesses^ .06 Caregiver Stress^  .55***  Caregiver Depression^.43*** Note:  p < .05 **^p < .01 *** p < .001  72 Table 11  Correlates of Caregiver Depression  Characteristics^Correlation Coefficient (rho) Patient Age^  -.07  Patient Gender^  -.22**  Marital status^  -.10  Household composition^-.02 Duration of illness^ .06 Caregiver Age^  .02  Caregiver Gender^  .18*  Relationship to patient ^-.09 Caregiver Involvement^-.17* Amount of Family Support^.24*** Perceived Lack of Support^.31*** Diagnosis of Alzheimers^-.01 Functional Rating Score^.23** Present Functioning score^.29*** Concurrent Illnesses^ .01 Caregiver Stress^  .54***  Caregiver Burden^  .43***  Note: p < .05 ^p < .01 ** *** p < .001  73 correlated (p<.001) with several predictor variables: caregiver stress, and higher patient ratings on the PFQ and FRS scales. Reports of less contact from family and friends and lack of support were also correlated with greater depression (p<.001). As with burden, caregivers of male patients indicated greater depression (p<.01) than those caring for female patients. Although caregiver sex was moderately associated with depression (r=.18, p<.05), with women reporting more depression than men, it was more highly related to stress (r..27, p<.001), with more frequent feelings of stress being reported by women. Caring for a male patient was also highly correlated with increased stress (r=-.26, p<.001). Increased stress was also significantly related to elevated FRS and PFQ ratings and perceived lack of support from the network of family and friends. Although not related to burden or depression, longer duration of the patient's dementia was significantly related to caregiver stress (r..21, p<.01). The three measures of caregiver health; stress, depression and burden were strongly related to each other as well (p<.001).  Comparison of the 2 Groups - Registered and Not-registered with Continuing Care Parametric and non-parametric analyses were undertaken, as appropriate, to compare the two groups with respect to the predisposing, enabling and need characteristics.  74 Table 12  Correlates of Caregiver Stress  Characteristics^Correlation Coefficient (rho) Patient Age^  -.06  Patient Gender^  -.26***  Marital status^  -.07  Household composition ^.05 Duration of illness ^ .21** Caregiver Age^ Caregiver Gender^  -.17* .27***  Relationship to patient ^-.01 Caregiver Involvement ^.01 Amount of Family Support^.22** Perceived Lack of Support ^.27*** Diagnosis of Alzheimers ^.01 Functional Rating Score^.30*** Present Functioning score ^.34*** Concurrent Illnesses^ .08 Caregiver Burden^  .55***  Caregiver Depression ^ .54*** Note: *^p < .05 **^p < .01 *** p < .001  75 Predisposing Characteristics Compared Between Groups In comparison of the registered and the non-registered group, significance levels were not reached with respect to any of the predisposing characteristics of patients or caregivers. Age of the patient came the nearest to significant levels (p= .095).  Predisposing Patient Characteristics The groups were compared with respect to patient age, gender, marital status, household composition and duration of illness. Those registered with CC were somewhat older with a mean age of 73 compared to a mean of 71 for Group 2 (p..095). Although intuitively it was expected that duration of illness would be significantly different for those registered with CC and those not registered, the registered group reported a mean of 68 months (approximately 5 1/2 years) compared with the 60 months (5 years) reported by those not registered. (p..161) The frequencies for the two groups of the remaining predisposing characteristics are summarized on Table 13. As shown, the registered group consisted of 43% (n=32) males and 57% (n=43) females compared to the non-registered group which held 34% (n=28) and 66% (n=55) females (p..248). Appreciably more of the non-registered group were married (83%) than those registered (69%). Those widowed in Group 1 represented 28% of the group compared to only 15% of those in group 2. The registered group held the only single person in  ^  76 Table 13 Comparison of Groups on Patient Predisposing Characteristic Frequencies  Gender  Group 1 Group 2 df Sign ^ Male ^32 (43%) 28 (34%) Female 43 (57%) 55 (66%) 1^.248  Marital status Single Married Widowed Divorced or /Separated  Group 1 Group 2 df Sign 1 ( 1%)^0 ( 0%) 52 (69%)^69 (83%) 21 (28%)^12 (15%) 1( 1%)^2( 2%)^3^.123  Country of Birth  Group 1 Group 2 df Sign ^ Canada 48 (64%) 49 (59%) ^ UK ^ 6 ( 8%)^12 (14%) Other ^ 13 (17%) 13 (16%) Unknown 8 (11%)^9 (11%)  Household composition Lives Alone Lives with Spouse Lives with Spouse & Children Lives with Children Other  Group 1  Group 2  df  Sign  4  .214  15^(20%) 14^(17%) 41^(55%) 54^(65%) 3^(^4%) 3^(4%) 13^(17%)  7^(^9%) 2^(^2%) 6^(^7%)  77 the sample, and had 1% indicating a divorced or separated status compared with 2% in the non-registered group. Since one-half of the cells had less than 5, the Yates continuity correction (Armitage & Berry, 1987) was applied, resulting in no significant difference between the groups. (p..161) Congruous with marital status figures, the group not registered with CC had 65% living with a spouse compared to the other group's 55%. An additional 9% of the non-registered group lived with spouse and children and 2% lived with just their children. In contrast, the registered group reported an additional 4% living with both spouse and children and 4% living with children only. Twenty percent of those registered reported living alone corresponding to the 17% of those not registered. Those indicating other living arrangements represented 17% of the registered group and 7% of the non-registered group. Again, applying the Yates correction for cells less than 5, the difference between the groups in household composition was not significant. (p=.214) Those registered with CC were also not significantly different from the non-registered group with respect to their country of origin. Although twice as many of those not registered with CC (12) had cited the UK as their country of origin compared to those registered (6), a near equal number in the registered group (48) and the non-registered group (49) named Canada as their birthplace. An equal number in the two groups (13) referred cited other countries as their birthplaces. Of the  78 subjects not providing information concerning country of origin, 8 belonged to the registered group compared to 9 in the nonregistered group.  Predisposing Caregiver Characteristics Caregiver characteristics considered were age, gender and relationship to patient. The mean age for the two groups was not significantly different (p..207) with the average age of 63 for those registered compared to 65 years for those not registered.  Frequencies for the two groups with respect to gender and relationship to the patient are outlined in Table 14. The characteristic emerging as a significant difference between the two groups was gender (p=.026) The proportions were virtually inverse with those registered with CC consisting of 40% female and 60% male in contrast to those not registered: 59% female and 40% male. Again, in keeping with marital status, those not registered with CC were more often spouses (78%) than those registered (67%). Almost twice as many of the caregivers whose care recipients were registered were children (29%) compared to those not registered (16%). Sibling caregivers represented 6% of those not registered and 4% of those registered with CC. Overall, however, the difference did not reach a significant level (p=.112)  79 Table 14 Comparison of Groups on Caregiver Predisposing Characteristic Frequencies  Gender  Group 1 Group 2 df Sign ^ Male ^30 (40%) 49 (59%) Female 45 (60%) 34 (41%) 1^.026  Relationship to patient Group 1 Group 2 df Sign Spouse^50 (67%) 65 (78%) Child^22 (29%) 13 (16%) Sibling^3 ( 4%) 5 ( 6%)^2^.112  80 Enabling Characteristics Compared Between Groups Enabling characteristics were limited to those of the caregivers and the support network. The caregiver.'s "degree of involvement", the objective amount of support from family and friends and the perceived lack of that support by the caregiver were the three enabling variables considered. Comparison between the two groups with respect to these three variables is displayed on Table 15. Degree of involvement by the caregiver emerged as the one enabling characteristic which differed significantly (p< .05) between the groups. The variable "degree of involvement" was defined by four levels of involvement: constant (in which the caregiver lived with the patient), daily (although the caregiver and patient did not co-habitate), weekly, or infrequently. Eighty-four percent of the caregivers whose care recipients reported not being registered with CC lived with and had constant contact with the patient compared with 70% of those registered. The groups were more similar in the numbers indicating daily contact (Registered - 9% , Not registered -10%). Caregivers from the registered group reporting weekly contact represented 17%, with 1% reporting infrequent contact. This was contrasted by the caregivers of the not-registered group whose percentage of those having only weekly contact consisted of only 4% and 0% for infrequent contact. The groups differed significantly (p=.023) allowing for Yates continuity correction.  81 Table 15 Group Comparison Caregiver Enabling Characteristics Group 1  Group 2  df  Sign  Degree of Involvement Living with patient Constant^52^(70%) Not living with patient Daily 7^(^9%) Weekly 13^(17%) Infrequent 1(^1%) (missing info) 4  70^(84%) 8 (10%) 3 (^4%) 0(^0%) 3  .023  Amount of Support Daily contact Regular weekly Regular monthly Infrequent contact No contact at all (missing info)  42^(56%) 28^(37%) 2^(^3%) 1^(^1%) 1^(^1%) 1  47 35 1 0 0  (57%) (42%) (^1%) (^0%) (^0%) 4  .577  48^(64%) 15^(20%) 8^(11%)  54 (65%) 23 (28%) 2 (^2%) 2  .073  Perceived Lack of Support Not at all Sometimes Frequently (missing info)^8  82 The groups were surprisingly similar in the amount of support they reported from family and friends. The amount of support was rated as contact, with levels of daily, weekly, monthly, infrequent and none at all. Slightly more of those not registered with CC reported daily (57%) and weekly (42%) contact than those registered with CC (56% and 36% respectively). Three percent of each group reported levels of involvement ranging from monthly to no contact at all. Obviously, no significant difference existed between the groups. (p=.577) Caregivers in the two groups did not differ significantly with respect to their perceived lack of support from the family and friend network. Feeling a lack of support was recorded as not at all, sometimes and frequently. Most notably different, 11% of those registered with CC reported frequently feeling a lack of support as compared to 2% of those not registered.  Need Characteristics Compared Between Groups Three of the need characteristics were significantly different between the two groups; patient scores on the Present Functioning Questionnaire and the Functional Rating Scale and the caregiver's level of burden as measured by the Burden Interview.  Patient Need Characteristics The characteristics related to patient need included the diagnosis of Alzheimer's Disease, concurrent illnesses, score on  83 the Present Functioning Questionnaire, score on the Functional Rating scale, and the rate of decline of the illness. Group comparison of the frequencies in diagnosis, concurrent illnesses are summarized in Table 16. Alzheimer's Disease in the patients was diagnosed as possible or probable. There was virtually no difference between the groups with respect to the diagnosis, with 41% and 43% of Group 1 and Group 2 respectively diagnosed with possible Alzheimer's Disease, leaving the remainder as probable cases (p=.922). Surprisingly, the presence of concurrent illnesses was not significantly associated with CC registration (p=.654). The groups did not vary significantly with respect to the incidence of concurrent psychiatric, neurological or other medical illnesses. The groups differed significantly (p<.01) on four of the five subscale scores on the Present Functioning Questionnaire as well as the total PFQ score (see Table 17). The problem area not yielding a significant difference between the groups was language functioning.^The means for Groups 1 and 2 were 3.6 and 3.3 respectively (p=.381). The problem areas which differed significantly between groups were personality, everyday functioning, memory and self-care. The total PFQ means were 30.0 for the registered group and 24.4 for the non-registered group (p..003). All eight areas of psychosocial functioning as measured by the FRS subscales and the FRS total score differed  84 Table 16 Group Comparison of Frequencies of Patient Need Characteristics Diagnosis of Alzheimers Group 1 Group 2 df Sign Possible^31 (41%) 36 (43%) Probable^44 (59%) 47 (57%) 1^.922  Concurrent Illnesses Group 1 Group 2 df Sign Medical^25 Psychiatric^13 Neurological^18 No significant^19 (missing information) 1  (33%) (17%) (24%) (25%)  28 11 16 27  (34%) (13%) (19%) (33%) 4^.654  85 Table 17 Group Comparison of Patient Need Characteristics Functional Rating scale Group 1^Group 2 Memory Language Problem Solving /Reasoning Orientation Social /Occupational Home/Hobbies Personal Care Affect Subscale Total Score  p  4.2 3.3  3.8 3.0  .000 .033  4.0 3.7  3.6 3.1  .011 .000  3.8 3.7 2.9 4.0  3.3 3.1 2.1 3.6  .000 .000 .000 .009  29.5  25.6  .000  Rate of Decline Group 1 Group 2 p Overall Decline Cognitive Decline Social Decline  .592 .205 .306  .623 .205 .329  .680 .989 .571  Present Functioning Questionnaire Group 1^Group 2 Personality Everyday Functioning Language Skills Memory Self-care Total Score  p  6.8 7.6 3.6 9.6 2.7  5.4 6.2 3.3 8.4 1.1  .006 .009 .381 .010 .001  30.0  24.4  .003  86 significantly between the groups. The areas which were less significantly associated with registration were language (p=.033) and problem solving (p=.011) The groups differed significantly in the area of affect (p=.009) The strongest difference between the groups lay in the other areas: memory, social & occupational, home & hobby, personal care, and orientation, as well as the overall FRS score with significance levels of p<.001. Not surprisingly, it was the patients registered with CC who had scored significantly higher than the non-registered group in all eight areas of the FRS. The examination of rate of decline included three different components: overall decline, cognitive decline and social decline. The formula used to arrive at the decline rates was the FRS scores/duration. The numerator to calculate Overall decline was the FRS Total score. The formula for Cognitive Decline included subtotals for memory, language, orientation, and problem solving & reasoning in the numerator, whereas the numerator for social decline consists of social & occupational, home & hobby, and personal care. Considering the highly significant differences between the groups with respect to the FRS scores, it was rather unexpected to find that the groups were virtually the same with respect to the various decline rates: overall decline (p=.680), cognitive decline (p=.571), and social decline (p=.989)  Caregiver Need Characteristics Three variables were considered as caregiver need  87 characteristics: stress, depression and burden. The unidimensional measures of stress and depression did not reveal any significant group difference (p=.357 and p=.213 respectively). Scoring on the Burden Interview (Zarit,1980), however, was significantly higher for the group registered with CC (total score=31) than the group not registered (score=24) (p.004).  Comparison of the 2 Groups - Users and non-Users of those Registered with Continuinq(n=75). Parametric and non-parametric analysis were used as appropriate to compare the two groups with respect to the predisposing, enabling and need characteristics. The groups did not vary significantly in any of the variables considered in this study.  Predisposing Characteristics Compared Between Groups Predisposing Patient Characteristics The groups were compared with respect to patient age, gender, marital status, household composition and duration of illness. They did not differ significantly with respect to any of these demographics variables. The slight difference in the age of the patients in the two groups was the reverse of the expected, with the average age patients using long term care being 72 compared to an average of 74 for those not using services. The duration of illness was somewhat longer for those  88 using services (74 months vs. 67 months), but the difference was not statistically significant. Table 18 summarizes the frequencies of the remaining predisposing patient characteristics. In both groups, the number of female patients outnumbered the males. The majority of both users and non-users were the married and lived with the spouse alone rather than with spouse and children.  Predisposing Caregiver Characteristics Caregiver characteristics considered were age, gender and relationship to patient. The groups were, again, noticeably alike with respect to caregiver demographics. The age was remarkably similar, with the average age of caregivers in the user group being 61 compared to an average of 63 for the group of non-users.  Frequencies for the two groups with respect to gender and patient relationship are listed in Table 19 below. Female caregivers comprised the greater percentage in both groups, as did the spousal relation over children or siblings.  Enabling Characteristics Compared Between Groups Enabling characteristics consisted of "degree of caregiver involvement", objective "amount of support" by family and friends and perceived "lack of support" by the caregivers. Table 20 summarizes the comparative scoring for the user and the non-  89 Table 18 Comparison of Frequencies of Patient Predisposing Characteristics for User and Non-User Groups Gender Male Female Marital status  Group 1  Group 2  df  Sign  7 13  25 27  1  .541  Group 1  Group 2  df  Sign  Single^0^0 Married^13^38 Widowed^6^14 Divorced or /Separated^1^0 Household composition Lives Alone Lives with Spouse Lives with Spouse & Children Lives with Children Other  2^.249  Group 1 Group 2 df Sign 7 8  8 32  1 0 4  2 3 7  Missing information 3  4  .242  90 Table 19 Comparison of Frequencies in Caregiver Predisposing Characteristics for User and Non-User Groups Gender Male Female  Group 1  Group 2  df  Sign  9 11  19 33  1  .510  Group 1  Group 2  df  Sign  13 6 1  36 15 1  2  .765  Relationship to patient  Spouse Child Sibling  91  user group for these three variables. The variable "degree of involvement" held four levels of involvement; constant (in which the caregiver lived with the patient), daily (although the caregiver and patient did not cohabitate), weekly, monthly or infrequent. Both the user group and the non-user group reported the majority of caregivers as having constant contact with the patient (ie. living with them).  The amount of support was rated as contact, with levels of daily, weekly, monthly, infrequent and none at all. The level of family support mentioned most frequently by both groups was that of daily contact. Feeling a lack of support was recorded as not at all, sometimes and frequently. Again, the groups were similar in that more of the caregivers from each group indicated not feeling a lack of support, with approximately 25% indicating that they experienced lack of support sometimes.  Need Characteristics Compared Between Groups Patient Need Characteristics The characteristics related to patient need involved the diagnosis of Alzheimer Disease, concurrent illnesses, scoring on the Present Functioning Questionnaire (PFQ), scoring on the Functional Rating Scale (FRS), and the rate of decline. Alzheimer Disease in the patients was diagnosed as possible or  92 Table 20 Comparison of Frequencies in Enabling Characteristics for User and Non-User Groups  Degree of Involvement Group 1 Group 2 df Sign Living with patient Constant^15^36 Not living with patient Daily^2^4 Weekly^3^9 Infrequent^0^1^3^.286 (missing information)5  Amount of Support /Contact with Family /Friends Group 1 Group 2 df Sign Daily contact^12^28 Regular weekly^7^20 Regular monthly^1^1 Infrequent contact^0^1 No contact at all^0^1^4^.846 (missing info)^4  Perceived Lack of Support Group 1 Group 2 df Sign Not at all^13^33 Sometimes^4^11 Frequently^3^5^2^.847 (missing info)^6  93 probable. In both groups, a greater percentage of the patients were diagnosed as having probable Alzheimer's disease. The groups were also similar with both reporting a slightly greater number suffering from medical illnesses than neurological or psychiatric. The rates of overall, cognitive and social decline were also similar for the two groups. Comparison of the user and non-user groups with respect to scores on the FRS and PFQ are listed below. As the tables indicate, there was virtually no difference in the patient scores on the FRS or the PFQ. The only dimension of the impairment measures to come near to a statistical significance between the groups was the subtotal for memory on the PFQ (p..18).  Caregiver Characteristics Three aspects of caregiver needs were included in variables considered as caregiver need characteristics; stress, depression and burden. As Table 22 indicates, the unidimensional measures of stress and depression also yielded similar results for both groups. A greater number of caregivers in both groups reported that they sometimes felt stressed as opposed to feeling stressed frequently or not at all. Approximately half of each group indicated not feeling depressed at all, while less than 20% of both groups of caregivers reported frequent feelings depression.  94 Table 21 Comparison of Patient Need Characteristics for User and Non-User Groups Functional Rating scale Group 1 Memory Language Problem Solving /Reasoning Orientation Social /Occupational Home/Hobbies Personal Care Affect Subscale Total Score  Group 2  Sign  4.20 3.45  4.22 .948 3.29 .585  3.90 3.75  3.90 .986 3.69 .835  3.85 3.75 2.80 4.00  3.81 3.73 2.90 3.98  29.70  .820 .919 .760 .938  29.52 .898  Present Functioning Questionnaire Group 1 Personality Everyday Function Language Skills Memory Self-care  6.89 7.75 4.21 10.37 2.63  Total Score  32.19  Group 2 6.90 7.57 3.40 9.35 2.73  Sign .833 .825 .254 .187 .914  29.23 .385  95 Table 22 Comparison of Caregiver Need Characteristics for User and Non-User Groups Stress Frequency Group 1 Group 2 df Sign Not at all^6^15 Sometimes^9^19 Frequently^5^13^2^.940 (missing info)^8  Depression Frequency Group 1 Group 2 df Sign Not at all^9^27 Sometimes^7^16 Frequently^4^6^2^.640 (missing info)^6  96 Caregivers from both groups scored essentially the same on the Zarit Burden Interview with the user group scoring an average total score of 31.2 as compared to the non-user group's average total score of 32.  Analysis of Zarit Burden Interview The Zarit Burden Interview has been challenged by several investigators (Vitaliano et al., 1991, Pearlin et al.,1990, Kosberg et al., 1990). George & Gwyther (1986) suggested that "summary scores mask dimension-specific patterns of caregiving impact and preclude identification of the different antecedents or correlates of specific dimensions of caregiver burden" (p.253). In a recent study by Chirboga et al.(1989), the burden interview data was subjected to factor analyses in order to explore the dimensionality of subjective burden. Alpha factoring and an oblique rotation resulted in five interpretable factors which accounted for 61% of the original variance. The scales which resulted were: general burden (11 items), burden of guilt and uncertainty (4 items), burden of social embarrassment (3 items) burden of parental demand (3 items) and financial burden (1 item) Similarly, a factor analysis was undertaken with interview data in this study. Using a selection criterion of .30 or greater, 21 of the original 22 items loaded on the five factors. The five factors were interpreted and labelled in terms of their  97 item content. They were: Factor 1: Social intrusion (5 items) A higher score on this scale indicated that caregivers felt that they didn't have enough time for themselves, their relationships had suffered, they had lost their privacy, their social life had suffered and they were uncomfortable about having friends over. Examples: a. How often do you feel because of the time you spend with your relative you don't have enough time for yourself? b. How often do you feel your social life has suffered because you are caring for your relative?  Factor 2: Uncertainty burden (2 items) Higher scores indicated that the caregivers felt that they ought to have been doing more and doing a better job of taking care of their relatives. Examples: a. Do you feel you could do a better job in caring for your relative? b. Do you feel you should be doing more for your relative?  Factor 3: Emotional Burden (5 items) The caregivers scoring high on this scale experienced a range of emotions, including; embarrassment, anger, strain, exasperation and concern regarding their own personal health.  98 Examples: a. How often do you feel embarrassed over your relative's behaviour? b. How often do you feel angry when you are around your relative?  Factor 4: Demand Burden (4 items) High scores on this scale indicated that caregivers were feeling particularly taxed in their resources, both financial and time. They felt a loss of control over their lives and wondered how long they would be able to continue caring for their relative. Examples: a. How often do you feel stressed between caring for your relative and trying to meet other responsibilities for your family or work? b. How often do you feel that you don't have enough money to care for your relatives in addition to the rest of your expenses?  Factor 5: Dependency Burden (5 items) Caregivers scoring high on this scale indicated a heavy sense of responsibility. They felt that their relative saw them as the only one who could take care of them and yet they felt uncertain about what to do. They rated themselves as feeling overall burdened. Examples:  99 a. How often do you feel your relative is dependent on you? b. How often do you feel that your relative seems to expect you to take care of him/her as if you were the only one he/she could depend on? Internal consistency of the scales was assessed using Cronbach's alpha coefficient and is summarized on Table 23.  100  Table 23 Coefficient Alpha for the 5 Scales of the Burden Interview Scale Name  ^  Social intrusion  ^ Number ^ Coefficient of items Alpha  ^ ^ 5 .813  Need/Care ^ Uncertainty ^ Emotional ^ Demand Dependency  ^  2^.745 5^.776 4^.788 5^.745  101 Examination of the table indicates that the scales have acceptable reliability levels, with the alpha coefficients ranging from .75 to .81. Intercorrelations of the scales are presented below on Table 24. The scales most strongly correlated are those of social intrusion with the burden of dependency (r=.55) and social intrusion with emotional burden (r..51). Dependency burden is also significantly associated with emotional burden (r=.40, p<.001). The correlation between the emotional scale and uncertainty regarding need and care is also significant (r=.29, p<.001). The burden of demand exceeding resources is negatively correlated with all other scales. The highest negative correlations are between demand and the burden .of social intrusion (r.-.40) and the burden of dependency (r.-.42). To a lesser degree, the scale of demand is also negatively correlated with emotional burden (r.-.29). It is minimally, but also negatively associated with the burden of uncertainty. Scores on the 5 scales were analyzed in relation to the group membership (registered vs. non-registered with CC), revealing that three of the scales were significantly correlated with the outcome of registration with CC (see Table 25). Whereas the total burden score was significantly related to the outcome (p..004), emotional burden was the only scale with a significant correlation of p<.01. The scales measuring  102 Table 24 Intercorrelations among the five scales of the Burden Interview Scale^1^2^3^4^5 1. Social intrusion  -  -  2. Uncertainty.13 3. Emotional^.51*^.29*  -  -  4. Demand ^-.40*^-.07^-.29* 5. Dependency .55*^.20^.40*^-.42*  Note:^* p < .001  103 Table 25 Five Scales of the Burden Interview Related to Group Membership Group 1 - Registered with long term care Group 2 - Not registered with long term care  Scale Name  Mean Scores Group 1^Group 2 (n=75)^(n=83)  Social intrusion  6.37  4.61  .017*  Uncertainty  2.16  1.94  .484  Emotional  5.88  4.24  .008**  Demand  4.89  3.62  .025*  Dependency  11.45  10.21  .079  Total Score  31.33  24.46  .004**  Sign  104 social intrusion and the burden of demand were both significantly related to group membership (p<.05), whereas the scales measuring uncertainty and dependency burden were not significantly correlated with registration.  Findings Question 1. Which predictor variables were significantly related to each other? Answer: Because of the recent focus in caregiver literature regarding caregiver health outcomes, correlates of the caregiver stress, depression and burden were of greatest interest for this study. Increased caregiver stress, burden, and depression were significantly related to the sex of the patient and the caregiver. In particular, caregivers of male patients and female caregivers reported higher levels of stress, burden and depression. Greater levels of caregiver-perceived patient impairment as depicted by PFQ scores were also significantly correlated with caregiver stress, depression and burden (p<.001). FRS scores or professionally perceived patient functional status, were significantly related to burden and stress (p<.001) and caregiver depression (p<.01). Of the three caregiver outcomes, only the frequency of stress feelings was significantly associated with duration of the illness (r=.21, p<.01).  Question 2. Which predisposing, enabling and need variables distinguished between those families who registered with long  105  term care and those who do not? Answer: All predisposing characteristics of patients and caregivers failed to distinguish between the two groups. Age of the patient was most highly related (p= .095), with a slightly higher mean age for those registered with CC. A significantly larger percentage of the caregivers of those not registered with CC reported living with the patient (p< .05), the single enabling characteristic of significance. The groups also differed significantly with respect to three of the need characteristics; patient scores on the Present Functioning Questionnaire and the Functional Rating Scale and the caregiver's level of burden as measured by the Burden Interview. Ratings on the eight dimensions of the FRS were all higher in the group registered with CC. The total score was also significantly higher in the registered group. (p<.001). With the exception of language skills, all deficit areas in the PFQ as well as the total rating were significantly pronounced in the group of registered patients. (p<.01). Caregivers of patients not registered with CC reported feeling significantly less burdened overall than those whose care recipients were registered. (p<.01)  Question 3. For those registered with long term care, which predisposing, enabling and need variables determined whether the family actually used the services provided by the Continuing Care Division of the Ministry of Health of B.C.?  106 Answer: The groups did not vary significantly in any of the predisposing, enabling or need characteristics considered in this study. The duration of illness was slightly longer for those utilizing services, although the difference was not statistically significant. The small number of those using services (n=20) compared to those not using services (n=55) might partially account for these findings.  Question 4. Into what dimensions could the Burden Interview be separated to reveal unique associative predictive qualities for the outcomes of CC registration and the utilization of services? Answer: Through a factor analysis, the Burden Interview loaded onto 5 factors. Reviewing the content of the items in each factor, names were chosen to depict the various aspects of burden represented. The 5 factors are: Factor 1: Social intrusion (5 items) Factor 2: Uncertainty (3 items) Factor 3: Emotional (5 items) Factor 4: Demand (4 items) Factor 5: Dependency (5 items) Alpha coefficients for the scales were; social intrusion (.813), uncertainty regarding need/care (.745), emotional (.776), demand exceeding resources (.788), burden of responsibility (.745). Intercorrelations among the five scales are shown above on Table 24 reveal that social intrusion was not related to  107 uncertainty regarding need or care. On the other hand, social intrusion was highly related (positively) to both emotional burden and the burden of responsibility. Social intrusion was, however, negatively related to the burden of demand exceeding the necessary resources. Uncertainty about need or care levels was significantly correlated with emotional burden and to a lesser extent related to the burden of responsibility. Emotional burden was inversely related to the burden of excess demand, but strongly related to the burden of responsibility. Finally, there was a negative correlation between the burden of responsibility and the burden of demand exceeding resources. In an analysis of the individual scales in relation to group membership, three emerged as significantly related to being registered with long term care or not (see Table 26). Emotional burden was the most significantly related to registration outcome (p<.01) while the burden of social intrusion and the burden of demand exceeding available resources were also correlated significantly (p<.05). The correlation between burden of dependency of the patient and registration outcome was close to significance (p<.10) while the burden of uncertainty regarding need and care was not related to the outcome. The total score for the Zarit Burden Interview was significantly related to the registration outcome at a level of p..004. It was apparent from this analysis that the contribution of specific dimensions of burden are masked when confining the analysis to burden total.  108 CHAPTER FIVE SUMMARY, CONCLUSIONS, IMPLICATIONS AND RECOMMENDATIONS  Introduction This study was designed to examine the predisposing, enabling and need characteristics of Alzheimer patients and their caregivers which would predict their registration with Continuing Care of B.C. It also examined the differences in utilization patterns of those who were registered, ie. comparing the characteristics of those who used support services offered by CC with those who did not. In the interest of comparison with earlier research using caregiver outcome measures, this study also explored the relationship between the predictor variables. Finally, the Zarit Burden Interview was subjected to a factor analysis in the interest of extracting dimensions of burden which might uniquely predict the registration with and/or utilization of CC support services. This chapter provides an overview of the study, conclusions, possible implications for the healthcare system and recommendations for further research.  Descriptive Analysis of Sample The sample for this study was somewhat unique in that they either self initiated or were recommended to receive the comprehensive assessment available at the Clinic for Alzheimer Disease and Related Disorders at the University Hospital, U.B.C. Site. As such, they represent a segment of the population with some degree of information regarding their situation and the  109 availability of resources. All of the AD patients were residing in the community at the time of their assessment. The caregivers who accompanied them were either spouses, children or siblings. The sample consisted of 158 dementia victims and their caregivers. The average age of the patients was 64 years, with the majority being female who were either married or living common-law. Overall, only 18% of the patients were living alone. Eighty-four percent of the patients had been born in Canada or the UK, again suggesting a specific segment of the population. The sample was restricted to patients having been diagnosed with either probable or possible Alzheimer Disease, with the majority having a diagnosis of probable. The majority of the patients were also suffering with a medical, psychiatric or neurological condition in addition to the Alzheimer Disease. Patient scores on the measure of psychosocial functioning, the Functional Rating Scale indicated a mild-moderate level of impairment for the sample. On the Present Functioning Questionnaire, the average score implied substantial deficits for the group overall. The average duration of illness for the patients was 64 months. There was an equal number of male and female caregivers. This distribution was unexpected in light of the literature suggesting that woman are more often caught in the role of caregiving. The majority of the caregivers were spouses, while the others were either children or siblings. The majority of the children were daughters while 5 out of the 8 siblings were  110  brothers.^It was interesting to note that caregivers were exclusively close family members. Constant involvement was assumed for the 77% living with the patients, while nearly half of those not living with the patients indicated daily contact. The sample was therefore considered to be a group at high risk for stress and burden. Generally, the caregivers in the sample appeared to be receiving a fair amount of support, with more than half indicating daily contact with family or friends and another 40% indicating regular weekly contact. In spite of this, one third of the caregivers indicated occasional or frequent feelings of lack of support from their family or friends. One third of the caregivers denied feeling any stress, while the other reported either frequent or occasional stress. Nearly half of the caregivers indicated that they never felt depressed. Since these measures consisted of single question, the responses are suspect to minimizing. There was great variation in the burden scores for the group of caregivers; ranging from 0 to 70 (out of 88) yielding an average score of 28. These scores can be considered more reflective, due to the specific nature of the questions.  Correlation of Variables with Caregiver Outcomes Since the majority of published research initiatives have considered caregiver health variables as outcome measures rather  111 than actual utilization of services, correlational analysis was used to examine the relationship between other predictor variables and the caregiver health variables in this study: caregiver stress, depression and burden. Increased caregiver stress, burden, and depression were found to be significantly related to the gender of the patient and the caregiver. In particular, female caregivers of male patients reported higher levels of stress and burden and depression. This finding is consistent with other studies in which females have expressed greater strain or burden associated with caregiving. This could also be viewed in terms of possible living arrangements. In cases where the female caregivers are caring for a parent or older relative, they could possibly be living with them. More predictably, in situations where the female caregivers are married to the patients, they would be living with them. In previous studies, living arrangements have been shown to increase the levels of strain. There are several possible explanations for this phenomenon. One has to do with the traditional role assumptions inherent in the cohort group most represented in this sample. Women from the generation of those over 60 years have characteristically been assumed to provide the nurturing role in the home, whereas men have provided the monetary stability. If women then assume a caregiving role with a nurturing approach while absorbing aspects of responsibility unfamiliar to her, a heightened emotional response could be expected. Men, on the  112 other hand, approaching challenges in an instrumental way would tackle the additional responsibility in a "problem/solution" manner. An alternate explanation has more to do with the personal awareness or admission of stress, depression or burden. Again, inherent in the values of the generation of those over 60 is the notion of male strength and resourcefulness. The more recent inclination toward self disclosure and admission of fears and pain has not necessarily been adopted by the previous generations. This calls into question the limits of self reporting of depression and stress. Given the limitations of quantifying emotions, these gender differences in reported depression, stress and burden can be argued to be somewhat meaningless. Speculation of gender differences in depression, stress and burden would not be complete without a brief consideration of the patient being cared for. In cases where the male caregiver is caring for a spouse or a female relative, the role of the woman could conceivably impact the caregiving experience. Taking the notion of women as nurturers a little further, I would suggest that women might actually facilitate less stressful caregiving. Again, the subtleties of relational dynamics would seem to render this speculative at best. Greater levels of patient impairment as depicted by PFQ and the FRS scores were also significantly correlated with caregiver stress, depression and burden. Although earlier studies have  113 used various measures to determine severity, the findings are consistent with many, particularly those considering the behavioral disturbance aspect of the illness.  Comparison of Registered and Not Registered  The majority of the predisposing characteristics of the caregivers in the groups was also not significantly different. However, the proportion of female and male caregivers differed significantly between the two groups with more female caregivers indicating registration. Again, there could be several explanations for this finding. Women might be more inclined to request help; partially because an admission of needing help might not be as abhorrent to them as to men. In cases where the female caregiver is caring for a male, the problem of physical size might simply necessitate this step. Another explanation could simply involve the practical limitations of living arrangements. A greater percentage of caregivers not living with the patient were women. In addition, the correlation between marital status and caregiver gender indicated that less of the female caregivers were married to the AD patient. These two factors by definition limit the possibility for daily care, thereby prompting registration with outside assistance. This is consistent with the finding in other studies which have shown that the propensity to either institutionalize patients or rely on supports increases when daily support is not possible.  114 This study also revealed that those indicating a lesser degree of involvement were more inclined to registration with CC. The groups were surprisingly similar in the amount of support they reported from family and friends. This is also consistent with several other studies in which additional informal support did not appear to reduce the level of stress and/or the need for formal support. By contrast, several of the Need characteristics for both patients and caregivers emerged as predictors of registration with CC. Patients varied significantly on their scoring of the Functional Rating Scale and the Present Functioning Questionnaire. It is evident that the more severe the patient's condition becomes with respect to manifestations of Alzheimer Disease, the more inclined the caregivers are to arrange assessment and subsequent registration with CC. It is interesting to note at this time that the duration of illness as a predisposing characteristic was not significantly associated with registration. This would imply that caregivers will continue to absorb burden over time until the illness reaches a level of severity which becomes overwhelming. There is a growing awareness of the danger which this could hold for caregivers. In their progressive absorbing of the burden related to their role, caregivers tend to minimize the impact on their own health and wellbeing in deference to the tragic plight of the patient. In a follow-up project, interviews with caregivers have revealed that many view their caregiving as an expression of  115 their love and support and feel that relinquishing the responsibility would be indicative of giving up or abandoning someone who would "do it for me" or who has "always been there for me". Unfortunately, this commitment often translates into a negative effect on the caregiver's health. A subtle aside to note with respect to illness severity is that of caregiver perceptions of memory impairment. On the FRS (as compiled by professionals), the dimension of memory was significantly more severe in those patients who were registered with CC. The subjective test for severity (PFQ) did not reveal this difference. This would suggest an unwitting compensation on the part of the caregivers with respect to the patient's condition. In areas of personality and personal care, the assessments of caregivers were consistent with those of the professionals. Again, it demonstrates that the more definable manifestations are admitted and finally acted upon, whereas the more subtle and wearying aspects are often downplayed by the caregiver. Research examining propensity to institutionalization has noted behaviour manifestations as a predictor. Earlier studies focusing on caregiver outcomes also found the severity level of AD as manifest in disruptive behavioral and the progressive need for personal care to be predictive of higher levels of stress, depression and burden. In this study, negative caregiver outcomes studied were shown to be associated with registration. These findings are then consistent with earlier research.  116 Of the three caregiver need variables, only burden emerged as being significantly different between the two groups with those in the registered group scored significantly higher. There are several possible explanations for the lack of difference in the levels of stress and depression. The first explanation lies in the unidimensional nature of the measures and the negative tone of the terms, possible prompting caregivers to minimize their experiences. This could serve to mask the subtle differences in levels of stress or depression between the groups. Another explanation could be the notion of a baseline of stress and depression inherent in caregiving. The measure for burden, on the other hand, captures various aspects of the burden experience with specific questions regarding feelings. These questions are more subtle in eliciting true responses resulting in more reflective scores.  Analysis of the Zarit Burden Interview As indicated in the previous chapter, the factor analysis of the Zarit Burden Interview resulted in the emergence of five factors; social intrusion, uncertainty, emotional burden, demand burden and dependency burden. Three of the scales were significantly correlated with the outcome of registration with CC: emotional burden, social intrusion and burden of demand. It was interesting to note that whereas the total burden score was significantly related to the outcome (p=.004) only the scale measuring emotional burden  117  reached a significance level of.01. Individual questions of the emotional burden scale were also examined. Those emerging as significantly related to registration touched on feelings of strain around the relative, the desire to leave the care of the relative to someone else and the feeling of personal negative health outcomes. This dimension of burden appears to capture the most predictive elements for registration with CC. Whereas the total burden score was significantly related with CC Registration, the closer examination of the five scales provides a clearer understanding of the aspects of burden which might finally prompt caregivers to reach out for assistance. There remains, however, the question of causation. If higher burden levels are related to registration with CC, is the burden prompting the registration or the registration enhancing the burden or is the burden existing in spite of the registration? Considering that registration with CC does not necessarily mean the utilization of services, the burden level associated with use of services should provide a better sense of the impact on burden.  Comparison of the Users and Non-Users of Services The purpose of the study had originally been to examine specific types of services (ie. in-home vs out-of-home) and examine which factors would differentially predict their utilization. Because of the small number of those using  118 services, a comparison could only be done between those using any services at all and those not using any services. The study was also restricted from any comparisons regarding amount of services used. In the further examination of those registered, no significant differences were found between the group using services and those not using services. Although this finding appears to discount the usefulness of services, it should be considered in light of the limited number in the sample. Nevertheless, it is an interesting finding to consider. A follow-up project has begun in which caregivers' perceptions of CC and the usefulness of services are being examined. Interviews with the caregivers are providing some helpful insights and providing significant challenges for the existing healthcare system. Caregivers generally describe CC and healthcare in "eventuality" terms. That is to say, they feel that when they are simply no longer able to take care of their relative, they will be forced to (unwillingly and sadly) place them into an institution. As the findings have indicated in this study, caregivers who are dealing with patients exhibiting greater levels of impairment and who were experiencing heightened burden were those who register with CC. In that sense, registration could be viewed as a desire to institutionalize rather than an indication of the desire for ongoing support services. When prompted, however, to consider assistance which they  .  119 would find helpful, caregivers invariably admit that they would benefit by having a person come in allowing them to get out for a period of time. Others say they would like to remain in their home with the comfort of knowing that someone else was taking the patient out for a day. They quickly add, however, that the patient has indicated the need for their constant presence and an aversion to strangers. Caregivers know that strangers are particularly distressing to a person with dementia and are reluctant to upset their relative. In addition, many caregivers state that they are hesitant to engage outside help because they feel that it is their responsibility to do all they can for their relative, stating "(he/she) would do it for me". The question still remains, however, why do those caregivers who have indicated using services not show a reduced level of burden in comparison to those who did not? This could better be addressed with a longitudinal study, noting changes in burden over time. It would seem reasonable to assume, however, that the provision of assistance should provide some relief of burden. On the other hand, one could argue that the strong correlation between level of impairment and caregiver burden might imply that by virtue of the progressive nature of dementia and the level of impairment reached by the time they use services, an impenetrable level of burden may have been reached. In the longitudinal study conducted by Colerick & George (1986) examining the propensity to and effects of institutionalization, it was shown that caregivers  120 increased their use of psychotropic drugs immediately following the institutionalization of the patient. Caring for a person with dementia has been termed "the ongoing funeral". If burden is viewed as an emotion connected with the relationship and grieving over the loss, it is understandable that practical support services would be inadequate in addressing the full degree of burden experienced by the caregiver. On the other hand, the questions in the Burden Interview most related to registration with CC dealt with issues which would appear to be addressed by support from outside sources. Perhaps, over time and with the gradual "letting go" of the full responsibility by caregivers, the effects of this support would be evident in reduced burden.  Considerations for Healthcare  This discussion would not be complete, however, without a consideration of the support services. Are these services appropriate and/ore adequate for Alzheimer patients? Generally, services provided by CC are directed at the client. In cases of dementia, patients are not able to recognize or appreciate the help provided. Caregivers, then, become the determinants of the effectiveness of the service. Caregivers can only realize relief in their burden levels if the service does not create additional strain for the patient and is adequate in its scope. The more typical services provided for clients by CC are limited assistance with housekeeping, personal care or meals on wheels.  121 Less frequently, provision is made for respite care or adult daycare. Often the caregivers do not feel the need for assistance with their home, needing rather the alleviation of their sense of their indispensability to the patient. Respite care is at a premium in the healthcare system, with a number of limited facilities providing beds for this purpose. The opportunities for daycare are increasing, however, many clients find unfamiliar group settings frightening. The option of having a person come in or accompany the patient for an extended period of time is currently not funded by CC services. There remains the task of bridging the gap between apparent need and appropriate solutions. If the caregivers carry the burden of caring alone beyond reasonable limits, the impact of their resulting ill health will be felt by the healthcare system as well. In summary, then, there are several factors playing into the utilization of services and their effectiveness. One is the attitude of the caregivers attempting to endure for the sake of commitment. Another is the reluctance of the patients to accept any outside help. Lack of information concerning the availability of assistance is also a key factor as well as the suitability of services for dementia victims and their caregivers. Prior to the assessment at the clinic, the primary access point to the healthcare system for most patients is the general practitioner. Reliance on the physician appears to translate into a perception of health focused on a diagnosis rather than  122 one of managing the condition in the most resourceful way. By virtue of orientation, physicians have traditionally medicalized health and as such have dealt less with the social aspects related to health. Although there appears to be a shift in health perspectives generally, the generation most affected by dementia appear to be caught in the transition. It would seem that to best address this dilemma, physicians may be compelled to help bridge the gap by encouraging patients and in particular their caregivers to expand their resource base and consider support services as a necessary preventative measure to circumvent negative health outcomes for the caregivers. In addition, the services provided need to be carefully reviewed in light of the unique needs connected with dementia and caregiver needs. An approach is needed which extends beyond the patient, addressing the health needs mounting in the caregiver. A Canadian study examining the physical and mental health of seniors was recently conducted (Canadian Study on Health and Aging, 1991/92). Findings from that study should provide some indication of the prevalence of mental impairment in the elderly across Canada. Information from their caregivers could also provide directives for service providers in the development of services to meet in relation to dementia victims and their caregivers in the community.  Recommendations for Future Research Considering the dynamic nature of dementia, a longitudinal  123 study observing caregiver burden over time in relation to the development of the illness should be undertaken. The five aspects of burden, as evident from a factor analysis of the Burden Interview should be examined individually to determined the impact of caregiver over the progression of the illness. As the stages of illness are examined in relation to burden levels, the amount and form of services most helpful for the caregiver in caring for the patient at each stage could be more clearly addressed. A longitudinal study would also be beneficial in determining  the impact of utilization of support services on the burden levels of the individual caregiver. This would allow for personal variability in response to the burden of caring for a relative suffering from dementia. Variables capturing attitudes and perceptions held by patients and caregivers should also be examined in order to understand additional reasons for choosing outside help or insisting on carrying on "as long as (one) is able". This information could prove helpful for counsellors dealing with caregivers who are inclined to sacrifice their own health for the sake of their relative. There remains an imperative to target those caregivers for whom a need clearly exists but who are still not utilizing services. Research is still needed to understand how interventions can be tailored to help specific subgroups of caregivers deal with specific problems in order to prevent them  124 becoming the hidden victims of dementia This remains the challenge for all those involved in the healthcare field.  125 REFERENCES  Adams, Trevor (1989). Dementia and Family Stress. Nursing Times, 85, (38), 27-29. Andersen, R.M., & Newman, J.F. (1973). Societal and Individual Determinants of Medical Care Utilization in the United States. The Milbank Quarterly, 51, 95-124. Anthony-Bergtone, C.R., Zarit, S.H., & Gatz, M. ^(1988). Symptoms of Psychological Distress Among Caregivers of Dementia Patients. Psychology & Aging, 3, (3), 245-248. Baldwin, B.A. (1988).^Community Management of Alzheimer's Disease. Nursing Clinics of North America, 23, (1), 47-56. Barnett, P. (1988).^Easing the Burden of Alzheimer's. RN, September, 91-94. Barusch, A.S., & Spaid, W.M. (1989). Gender Differences in Caregiving: Why do Wives Report Greater Burden? The Gerontologist, 29, 667-676. Bass, D.M., & Noelker, L.S. (1987). The Influence of Family Caregivers on Elder's Use of In-Home Services: An Expanded Conceptual Framework. Journal of Health and Social Behaviour, 28, 184-196. Baumgarten, M. (1989). The Health of Persons Giving Care to the Demented Elderly: a Critical Review of the Literature. Journal of Epidemiology, 42, 1137-1148. Biegel, D.E., Bass, D.M., Schulz, R., & Morycz, R. ^(1990). Predictors of Differential Service Use by Family Caregivers of Alzheimer's Disease Patients. Poster Session Summary, Prepared for the 43rd Annual Scientific Meeting of the Gerontological Society of America. Blake, M.^(1988).^Caregivers. B.C. Medical Journal, 30, (4), 268-271. Brane, G. (1986).^Normal Aging and Dementia Disorders Coping and Crisis in the Family. Neuro Psychopharmacological and Biological Psychiatry. 10, 287-295. Brody, E.M. (1981). "Women in the Middle" and Family Help to Older People. The Gerontologist, 21, (5), 471-480.  126 Brody, S., Poulshock, S.W., & Masciocchi, C.F. ^(1978) The Family Caring Unit: A Major Consideration in the Long-term Support System. The Gerontologist, 18, (6), 556-561. Brody, Elaine M., Saperstein, A., & Lawton, M. (1989). A multi-service respite program for caregivers of Alzheimer's patients. Journal of Gerontological Social Work 14, (1-2), 41-74. Buckwalter, K.C., Abraham, I.L., & Neundorfer, M.M. (1988). Alzheimer's Disease: Involving Nursing in the Development and Implementation of Health Care for Patients and Families. Nursing Clinics of North America, 23, (1), 1-9. Burdz, M.P., Eaton, W.O., & Bond, J.B.^(1988).^Effect of Respite Care on Dementia and Nondementia Patients and Their Caregivers. Psychology & Aging, 3, (1), 38-42. Burns, E.M., & Buckwalter, K.C. (1988). Pathophysiology and Etiology of Alzheimer's Disease. Nursing Clinics of North America, 23, (1), 11-29. Cantor, M.H. (1983).^Strain Among Caregivers: A Study of Experience in the United States. The Gerontologist, 23, (6), 597-604. Caserta, M.S., Lund, D.A., Wright, S.D., & Redburn, D.E. (1987). Caregivers to Dementia Patients: The Utilization of Community Services. The Gerontologist, 27, (2), 209-214. Cherry, D.L., & Rafkin, M.J. (1988).^Adapting Day Care to the Needs of Adults With Dementia. The Gerontologist, 28, (1), 116-120. Chiriboga, David, Weiler, P., & Nielsen, K. (1989). The stress of caregivers. Special issue: Aging and family caregivers. Journal of Applied Social Sciences 13 (1), 118-141. Choi, T., Josten, L., & Christensen, M.L. ^(1983).^HealthSpecific Family Coping Index for Noninstitutional Care. AJPH, 73, (11), November, 1275-1277. Clipp, Elizabeth. C. (1990). Psychotropic Drug Use Among Caregivers. Journal of the American Geriatrics Society, 38, (3), 227-235. Cohen, Donna, & Eisdorfer, Carl (1988). Depression in family members caring for a relative with Alzheimer's Disease. Journal of the American Geriatrics Society, 36 (10), 885-889  127 Colerick, E.J., & George, L.K.^(1986).^Predictors of Institutionalization Among Caregivers of Patients With Alzheimer's Disease. American Geriatrics Society Journal, 34, 493-498. Coughlin, T.A., & Liu, K. (1989). ^Health Care Costs of Older Persons with Cognitive Impairments. The Gerontologist, 29, (2), 173-182. Deimling, G.T., & Bass, D.M. (1986). ^Symptoms of Mental Impairment Among Elderly Adults and Their Effects on Family Caregivers. Journal of Gerontology, 41, (6), 778-784. Derogatis, L.R., Rickets, K.P., & Rock, A.F. (1977).^The SCL90 and the MMPI: A step in the validation of a new selfreport scale. British Journal of Psychistry, 128, 280289. Derogatis,L.R., & Cleary,P.A. (1977). Confirmation of the dimensional structure of the Symptoms Checklist-90: A study in construct validation. Journal of Clinical Psychology, 33 (4), 981-989. Duke University Centre for the Study of Aging and Human Development (1978).^Multidimensional Functional Assessment: The OARS Methodology, A Manual. Durham, NC: The Centre for the Study of Aging and Human Development. Eagles, J.M., & Gilleard, C.J. (1983). The Functions and Effectiveness of a Day Hospital for the Demented Elderly. Health Bulletin, 42, (2) 87-91. Eimer, M. (1989).^Management of the Behavioral Symptoms Associated With Dementia. Primary Care, 16, (2), June, 431-450. Eisdorfer, C., & Cohen, D. (1981).^Management of the Patient and Family Coping With Dementing Illness. Journal of Family Practice, 12, (5), 831-837. Endicott, J. & Spitzer,R. (1978) A Diagnostic Interview for Affective Disorders and Schizophrenia. Archives of General Psychiatry, 35, 522-525. Farran, C.J., & Keane-Hagerty, E. (1989). ^Communicating Effectively With Dementia Patients. Journal of Psychosocial Nursing, 27, (5), 13-16.  128 Fiore, J., Coppel, D.B., Becker, J. & Cox, G.B. ^(1986). Social Support as a Multifaceted Concept: Examination of Important Dimensions for Adjustment. American Journal of Community Psychology, 14, (1), 93-111. Fitting, M., Rabins, P., Lucas, M.J. & Eastham, J. ^(1986). Caregivers for Dementia Patients: A Comparison of Husbands and Wives. The Gerontologist, 26, (3), 248-252. Folstein, M., Folstein, S., & McHugh, P. (1975). Mini-mental State: a Practical Method for Grading the Cognitive State of Patients for the Clinician. Journal of Psychiatric Research. 12, 189-198. Ford, M., Fox, J., Fitch, S., & Donovan, A. ^(1986). Light in the Darkness. Nursing Times, January, 26-29. Francell, C.G., Conn, V.S., & Gray, D.P. ^(1988).^Families' Perceptions of Burden of Care for Chronic Mentally Ill Relatives. Hospital and Community Psychiatry, 39, (12), 1296-1300. Gallagher, Dolores, et al. (1989).^Prevalence of Depression in Family caregivers. Gerontologist. 29 (3-4), 449-456. Gendron, C.E., et al. (1986).^Skills Training With Supporters of the Demented. American Geriatrics Society Journal, 34, 875-880. George, L.K., & Gwyther, L.P. (1986).^Caregiver Well-Being: A Multidimensional Examination of Family Caregivers of Demented Adults. The Gerontologist, 26, (3), 253-259. Gilhooly, M.L.M. (1984). ^The Impact of Care-giving on Caregivers: Factors Associated with the Psychological Well-being of People Supporting a Dementing Relative in the Community. British Journal of Medical Psychology, 57, 35-44. Gilhooly, M.L.N., & Whittick, J.E. (1989). Expressed Emotion in Caregivers of the Dementing Elderly. British Journal of Medical Psychology, 62, 265-272. Gilleard, C.J., Gilleard, E. Gledhill, K. & Whittick, J. (1984). Caring for the Elderly Mentally Infirm at Home: a Survey of the Supporters. Journal of Epidemiology and Community Health, 38, 319-325. Given, C.W., Collins, C.E., & Given, B.A. ^(1988).^Sources of Stress Among Families Caring for Relatives With Alzheimer's Disease. Nursing Clinics of North America, 23, (1), 69-82.  129 Goodman, Catherine, & Pynoos, J. (1988). Telephone Networks Connect Caregiving Families of Alzheimer's Victims. Gerontologist, 28, (5), 602-605. Graham, R.W. (1989). Adult Day Care: How Families of the Dementia Patient Respond. Journal of Gerontological Nursing, 15, 27-31. Gray-Vickrey, P. (1988).^Evaluating Alzheimer's: The Importance of Being Thorough. Nursing, December, 34-39. Greene, V.L. (1983).^Substitution Between Formally and Informally Provided Care for the Impaired Elderly in the Community. Medical Care, 21, (6), 609-616. Greene, R.L. (1980).^The MMPI:An interpretive manual. New York: Grune and Stratton. Gregory, D.M. (1990).^Elderly Male Spouses as Caregivers: Toward an Understanding of Their Experience. Journal of Gerontological Nursing, 16, (3), 20-24. Gurland, B., Dean, L.L., Copeland, J., Gurland, R., & Golden, R. (1982)^Criteria for the Diagnosis of Dementia in the Community Elderly. The Gerontologist, 22, (2), 180-186. Haley, W.E., & Pardo, K.M. (1989). Relationship of Severity of Dementia to Caregiving Stressors. Psychology & Aging, 4, (4), 389-392. Haley, et al. (1987).^Stress, Appraisal, Coping and Social Support as Predictors of Adaptational Outcome Among Dementia Caregivers. Psychology & Aging, 2, (4), 323-330. Haley, W.E., Levine, E.G., Brown, S.L., Berry, J.W., & Hughes, G.H. (1987).^Psychological, Social, and Health Consequences of Caring for a Relative with Senile Dementia. American Geriatrics Society Journal, 35, 405-411. Hall, G.R. (1988). Care of the Patient with Alzheimer's Disease Living at Home. Nursing Clinics of North America, 23, (1), 31-46. Hall, G., Kirschling, M.V., & Todd, S.^(1986).^Sheltered Freedom- An Alzheimer'Unit in an ICF. Geriatric Nursing, May/June, 132-137. Harris, J., Gierz, M., & Lohr, J.B.^(1989). Recognition and Treatment of Depression in Alzheimer's Disease. Geriatrics, 44, (12), 26-30.  130 Harvis, K.A., & Rabins, P. (1989). Dementia: Helping Family Caregivers Cope. Journal of Psychosocial Nursing, 27, (5), 7-11. Haley, W.E., & Pardo, K.M. (1989). Relationship of Severity of Dementia to Caregiving Stressors. Psychology & Aging, 4, (4), 389-392. Haley, W.E., Levine, E.G., Brown, S.L. & Bartolucci, A.^(1987) Appraisal, Coping and Social Support as Predictors of Adaptational Outcome Among Dementia Caregivers. Psychology & Aging, 2, (4), 323-330. Hertzman, C., Pulcins, I.R., Barer, M.L., Evans, R.G., Anderson, G.M., & Lomas, J. (1990). Flat on Your Back or Back to your Flat? Sources of Increased Hospital Services Utilization Among the Elderly in British Columbia. Social Science Medicine, 30, (7), 819-828. Hirschfeld, M. J. (1978). Families Living with Senile Brain Disease. Dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Nursing Science, University of California, San Fransisco. Hughes, D.Y. (1987). Alzheimer's Disease and Psychiatric Nursing: Treating the Depression. Perspectives in Psychiatric Care, 24, (1), 5-6. Isaacs, B. et. al. (1972). Survival of the Unfittest: A Study of Geriatric Patients in Glasgow. London: Routledge & Kegan Paul. Johnson, C.L., & Catalano, D.J. (1983). A Longitudinal Study of Family Supports to Impaired Elderly. The Gerontologist, 23, (6), 612-618. Johnson, C.L. (1983). Dyadic Family Relations and Social Support. The Gerontologist, 23, (4), 377-383. Kapust, L.R. (1983). Living with Dementia: The Ongoing Funeral. Social Work in Health Care, 7, (4), 79-91. Kartman, L.V.K. (1983).^People Helping People Burnout. Activities, Adaptation & Aging, 3, (4), Summer, 49-57. Kay, D.W.K. (1989). Aging of the Population: Measuring the Need for Care. Age and aging, 18, 73-76. Kellerman, F. (1988). Alzheimer's Disease-The Silent Epidemic. New Zealand Nursing Journal, March, 29-31.  131 Kinney, J.M., & Stephens, M.A.P. (1989). Caregiving Hassles Scale: Assessing the Daily Hassles of Caring for a Family Member with Dementia. The Gerontologist, 29, (3), 328-332. Kinney, J.M., & Stephens, M.A.P. (1989). Hassles and Uplifts of Giving Care to a Family Member With Dementia. Psychology & Aging, 4, (4), 402-408. Kosberg, J.I., & Cairl, R.E. (1986). The Cost of Care Index: A Case Management Tool for Screening Informal Care Providers. The Gerontologist, 26, (3), 273-278. Kutner, B., Fanshel, D., Togo, A.M. & Langer, T.S. ^(1956). Five Hundred Over Sixty. ^New York: Russell Sage Foundation. Lawton, M.P. (1971). The Functional Assessment of Elderly People. Journal of the American Geriatrics Society. 19, 465-480. Lawton, M.P., Brody, E.M., & Saperstein, A.R. ^(1989).^A Controlled Study of Respite Service for Caregivers of Alzheimer's Patients. The Gerontologist, 29, 8-16. Lindsey, J., & Murphy, E. (1988). Dementia, Depression and Subsequent Institutionalization- The Effect of Home Support. International Journal of Geriatric Psychiatry, 4, 3-9. Liu, K., Manton, K.G., & Liu, B.M. (1985). Home Care Expenses for the Disabled Elderly.^Health care Financing Review, 7, (2), 51-58. Long, D. (1986). Alzheimer's Disease: Growing With Age. The New Zealand Nursing Journal, August, 16-19. Mace, N.L., & Rabins, P.V. (1984). Day Care and Dementia. Generations, 9, (2), 41-45. Mace, N.L., & Rabins, P.V.^(1981). The 36-Hour Day. Baltimore, MD: Johns Hopkins Press. Machin, E. (1980). A Survey of the Behaviour of the Elderly and their Supporters at Home. A thesis presented for the degree of Master of Science, University of Birmingham. Macmillan, D. (1969). Features of Senile Breakdown. Geriatrics, 24, 109-118. Maletta, G.J. (1988). Treatment Considerations in Dementing Illness. Clinics in Geriatric Medicine, 4, (4), 699-703.  132 Mann, S.H. (1986).^Practical Management Strategies for Families with Demented Victims. Neurologic Clinics, 4, (2), 469-478. Maslow, Kate (1990).^Linking Persons with Dementia to Appropriate Services: Summary of an OTA Study. Pride Institute Journal of Long Term Home Health Care, 9, (2), Spring, 42-50. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D. and Stadlan, E.M. (1984). Clinical Diagnosis of Alzheimer's Disease: Report of the NINCDS-ADRDA Work Group under the Auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology, 34, 939-944. Merriam, A.E. Aronson, M.K., Gaston, P., Wey, S., & Katz, I. (1988).^The Psychiatric Symptoms of Alzheimer's Disease. The American Geriatrics Society Journal, 36, (1), 7-12. Mohide, E.A., Torrance, G.W., Streiner, D.L., Pringle, D.M. & Gilbert, R. (1988).^Measuring the Wellbeing of Family Caregivers Using the Time Trade-Off Technique. Journal of Clinical Epidemiology, 41, (5), 475-482. Montgomery, Rhonda, Kosloski, K., & Borgatta, E. (1989). The Influence of Cognitive Impairment on Service Use and Caregiver Response. Special Issue- Aging and Family Caregivers. Journal of Applied Social Sciences. 13, (1), 142-169. Moritz, D.J., Kasl, S.V., & Berkman, L.F. ^(1989).^The Health Impact of Living With a Cognitively Impaired Elderly Spouse: Depressive Symptoms and Social Functioning. Journal of Gerontology, Social Sciences, 44, (1), S17-S27. Morris, L.W., Morris, R.G., & Britton, P.G. ^(1989). Cognitive Style and Perceived Control in Spouse Caregivers of Dementia Sufferers. British Journal of Medical Psychology, 62, 173-179. Morris, L.W., Morris, R.G. & Britton, P.G. ^(1988).^The Relationship Between Marital Intimacy, Perceived Strain and Depression in Spouse Caregivers of Dementia Sufferers. British Journal of Medical Psychology, 61, (3). Morrisey, E., Becker, J. & Rubert, M.P. (1990). Coping Resources and Depression in the Caregiving Spouses of Alzheimer Patients. British Journal of Medical Psychology, 63, (2).  133 Morycz, R.K. (1985). Caregiving Strain and the Desire to Institutionalize Family Members with Alzheimer's Disease. Research on Aging, 7, (3), 329-361. Motenko, A.K. (1989). The Frustrations, Gratifications, and Well-Being of Dementia Caregivers. The Gerontologist, 29, (2), 166-172. Munro, B.H., Visintainer, M.A., & Page, E.B. ^(1986). Statistical Methods for Health Care Research. J.B. Lippincott Company, Philadelphia. Novak, M., & Guest, C. (1989). ^Application of a Multidimensional Caregiver Burden Inventory. The Gerontologist, 29, (6), 798-803. Novak, Mark, & Guest, Carol. (1989).^Caregiver response to Alzheimer's Disease. International Journal of Aging and Human Development, 28, (1), 67-79. Orford, J., O'Reilly, P., & Goonatilleke, A. ^(1987). Expressed Emotion and Perceived Family Interaction in the Key Relatives of Elderly Patients With Dementia. Psychological Medicine, 17, 963-970. Ory, M.G. et. al. (1985). Families, Informal Supports, and Alzheimer's Disease. Research on Aging, 7, (4), 623-644. Ory, M.G. (1985). The Burden of Care: A Familial Perspective. Generations, Fall, 14-18. Pagel, M. & Becker, J. (1987). Depressive Thinking and Depression: Relations with Personality and Social Resources. Journal of Personality and Social Psychology, 52, (5), 1043-1052. Peplau, H. (1986). Light in the Darkness. Nursing Times, January 7, 26-29. Pfeiffer, Eric (1975). A Short Portable Mental Status Questionnaire for the Assessment of Organic Brain Deficit in Elderly Patients. Journal of the American Geriatrics Society, 22, 433-441. Pfeiffer, E. (1979). A Short Psychiatric Evaluation Schedule: A New 15-item Monotonic Scale Indicative of Functional Psychiatric Disorder. In the Proceedings of the BayerSymposium VII, Brain Function in Old Age. New York, Springer-Verlag. Pfeiffer, E. (1988). What's New in Alzheimer's Disease? Postgraduate Medicine, 83, (5), 107-115.  134 Pratt, C.C., Schmall, V.L., Wright, S. et al. ^(1986). Burden and Coping Strategies of Caregivers to Alzheimer's Patients. Family Relations, 34, 27-33. Pruchno, R.A. (1990). The Effects of Help Patterns on the Mental Health of Spouse Caregivers. Research on Aging, 12, (1), 57-71. Pruchno, R.A., & Potashnik, S.L. (1989). Caregiving Spouses: Physical and Mental Health in Perspective. American Geriatrics Society Journal, 37, 697-705. Pruchno, R.A., & Resch, N.L. (1989). ^Husbands and Wives as Caregivers: Antecedents of Depression and Burden. The Gerontologist, 29, (2), 159-165. Pruchno, R.A. & Resch, N.L. (1989). Aberrant Behaviours and Alzheimer's Disease: Mental Health Effects on Spouse Caregivers. Journal of Gerontology, 44, S177-182. Quayhagen, Mary, & Quayhagen, Margaret. (1988). Alzheimer's Stress: Coping with the Caregiving Role. Gerontologist, 28, (3), 391-396. Rabins, P.V. (1988). Psychsocial Aspects of Dementia. Journal of Clinical Psychiatry, 49, (5), (Supp), 29-31. Rabins, P.V., Mace, N.L., & Lucas, M.J. ^(1982). The Impact of Dementia on the Family. Journal of American Medical Association, 248, (3), 333-335. Radloff, Lenore S. (1977). The CES-D Scale: A Self-Report Depression Scale for Research in the General Population. Applied Psychological Measurement, 1, 385-401. Roberts, B.L., & Algase, D.L.^(1988). Victims of Alzheimer's Disease and the Environment. Nursing Clinics of North America, 23, (1), 83-93. Robinson, K.M. (1989). Predictors of Depression Among Wife Caregivers. Nursing Research, 38, (6), 359-363. Russell, V., Proctor, L., & Moniz, E.^(1989). The Influence of a Relative Support Groups on Carers' Emotional Distress. Journal of Advanced Nursing, 14, 863-867. Sands, D., & Suzuki, T. (1983). Adult Day Care for Alzheimer's Patients and Their Families. The Gerontologist, 23, 21-23.  135 Schaefer, C., Coyne, J.C., & Lazarus, R. (1981). The Health Related Function of Social Support. ^Journal of Behavioral Medicine, 4, 381-406. Scharlach, A., & Frenzel, C. (1986). ^An Evaluation of Institution-Based Respite Care. The Gerontologist, 26, (1), 77-82. Schlepp, S. (1989). Elder Care: Caregiving-Learning to Cope, Learning the Options. AORN Journal, 50, (2), 228-236. Scott, J., Kosarek, E. and Tuokko, H. Gender Differences Among Caregivers: The Subjective Experience of Women and Men Caring for their Elderly Demented Spouses. Paper completed as partial requirement for the degree of Bachelor of Arts (Honors), Department of Psychology, U.B.C., 1988. Scott, J.P., Roberto, K.A., & Hutton, J.T. ^(1986).^Families of Alzheimer's Victims: Family Support to the Caregivers. American Geriatrics Society Journal, 34, 348-354. Seltzer, B. et al. (1988). The Short-Term Effects of InHospital Respite on the Patient With Alzheimer's Disease. The Gerontologist, 28, (1), 121-124. Shanas, E. (1979). Social Myth as Hypothesis: The Case of the Family Relations of Old People. The Gerontologist, 19, (1), 3-9. Silverstein, N.M., Gonyea, J.G., & King, N.R. ^(1989). FamilyProfessional Partnerships for Addressing Alzheimer's Disease. The Gerontologist, 29, 830-834. Smith, C.W. (1988). Management of Alzheimer's Disease. Postgraduate Medicine, 83, (5), 118-127. Soldo, B.J., & Myllyluoma, J. (1983). Caregivers who Live with Dependent Elderly. The Gerontologist, 23, 605-611. Stephens, M.A.P., et al.^(1988). Stressful Situations in Caregiving: Relations Between Caregiver Coping and WellBeing.^Psychology & Aging, 3, (2), 208-209. Stetz, K.M. (1989). The Relationship Among Background Characteristics, Purpose in Life, and Caregiving Demands on Perceived Health of Spouse Caregivers. Scholarly Inquiry for Nursing Practice, 3, (2), 133-153. Stewart-Dedmon, M. (1988). Strain and Strategy. Nursing Times, 84, (17), 38-41.  136 Stoller, E.P., & Pugliesi, K.L.^(1989). Other Roles of Caregivers: Competing Responsibilities or Supportive Resources. Journal of Gerontology, Social Sciences, 44, (6), S231-S238. Swearer, Joan, Drachman, D.A., O'Donnell, B.F., & Mitchell, A.L. (1988). Troublesome and Disruptive Behaviors in Dementia: Relationships to Diagnosis and Disease Severity. Journal of the American Geriatrics Society. 36, (9), 784790. Teri, Linda, Borson, S., Kiyak, H., & Yamagishi, M. ^(1989). Behavioral Disturbance, Cognitive Dysfunction, and Functional Skill: Prevalence and Relationship in Alzheimer's Disease. Journal of the American Geriatrics Society,^37, (2), 109-116. Tuokko, H. & Crockett, D. (1991). Assessment of Everyday Functioning in Normal and Malignant Memory Disordered Elderly. In D.E. Tupper & K.D. Cicerone (eds): The Neuropsychology of Everyday Life: Issues in Development and Rehabilitation, Kluwer Academic Publishers, Boston. Weissert, W.G. (1986). Seven Reasons Why It is So Difficult to Make Community-Based Long-Term Care Cost-Effective. Whittick, J.E. (1989). Dementia and Mental Handicap: Attitudes, Emotional Distress and Caregiving. British Journal of Medical Psychology, 62, 181-189. Wilson, Holly. (1989). Family Caregiving for a Relative with Alzheimer's Dementia: Coping with Negative Choices. Nursing Research, 38, (2), 94-98. Yaffe, M.J. (1988). Implications of Caring for An Aging Parent. CMAJ, 138, February, 231-235. Young, R.F. & Kahana, E. (1989). Specifying Caregiver Outcomes: Gender and Relationship Aspects of Caregiving Strain. The Gerontologist, 29, (5), 660-666. Young, S.H., Muir-Nash, J., & Ninos, M. (1988). Managing Nocturnal Wandering Behavior. Journal of Gerontological Nursing, 14, (5), 6-12. Zarit, S.H., Anthony, C.R., & Boutselis, M. ^(1987). Interventions with Caregivers of Dementia Patients: Comparison of Two Approaches. Psychology & Aging, 2, (3), 225-232. Zarit, S.H., Todd, P.A., & Zarit, J.M.^(1986). Subjective Burden of Husbands and Wives as Caregivers: A Longitudinal  137 Study. The Gerontologist, 26, (3), 260-266. Zarit, S. (1989). Do We Need Another 'Stress and Caregiving' Study? The Gerontologist, 29, (2), 147-148. Zarit, S.H., Zarit, J.M., & Reever, K.E.^(1982). Memory Training for Severe Memory Loss: Effects on Senile Dementia Patients and Their Families. The Gerontologist, 22, (4), 373-377. Zarit, S.H. et. al. (1980). Relatives of the Impaired Elderly: Correlates of Feelings of Burden. The Gerontologist, 20, 649-660. Zung, W. (1973).^From Art to Science.^Archives of General Psychiatry. 29, 328-337. Zung, W. (1965). A Self-Rating Depresion Scale. Archives of General Psychiatry, 12, 63-70.  138  APPENDIX A  DIAGNOSTIC CRITERIA  FUNCTIONAL RATING SCALE Total Scott '40  DATIt OF =Ill ^  ONSET: ^  (sear etpla••sery iettnities)  AC1r AT MIMI  °ORATION/ Camber of mentims sinew first goateed). • First Assessment  Tomesis Assaaennat  SecesulAssominant  ?lids Looseness  IhiftAssamemenr Impairment at memory and those or mews of Lbw folloming abilities (sea angslanateary statements).  L 411CTLLSZILS  1  2  2' 4  3  5  1  4  3:  23451134512345113 .  .  -  PoseialiOcaapational.  '  "^.  Hose and lioeixtes  •  -:.  Personal Caro  Leetammie  Problem solvinni leasoning .  Affect .  .  Orientation  MEE DISOIDEtit WHICH MAT COMPTICA11CDIACNOS/Si (Cirels if sufficient basement for impairments) ;Neurological: Psychiatric: Medical: Ocher:  DIAGNOSIS OP ALZHETKER'S DISEASE: (See explanatory definition) First Assessment: Dace:^ Dx.:  Fourth Assessment Date:^ Dx.:  Second Assessment Date:^ Dx.:  Fifth Asssessaeat Date:^ Dx.:  5  139 APPENDIX B ?RESENT FUNCTIONING QUESTIONNAIRE Oepartment of Psychology UNIVERSITY HOSPITAL, MC SITE Informant: Relationship to Patient:^  Comments: Total Score- •^/ PersobalitV Being IRRITABLE: or ANGRY Being KISIGUUME. or DEPRESSED 3yinq Trssror -PANICXX Being APATHETIC • loins AGITATED or szersAcnmrs Being ANXIOOS or ...PRATO Stating that 'THINGS AREN'T REAL' Complaining of 'UPSETTING THOUGHTS' Being xocs.tssrrs • Being YOSPICTOOS • Being INSEESITEVE T3 OTHERS' PEELINGS Shawl:mg INAPPROPRIATE SMILES OR LADGETER Shoving DECREASED abaft' INVOLVE:NM TALXIING TO vixcisksr OTHERS Exhibiting ITERFPROPRENTE SEXUAL ACTIY/TrES EVervdaV*raskst 2rOblems PERFORMING HOUSEHOLD TASKS Problees HANDLING MONEY- • Problems SHOPPING Problem&•FTUDISG THEIR WAX INSIDE A ROUSE OR 30/LDING. Problems 7/HDING THE/A•WAY AROUND ?MILLAR STREETS . Problem& &MOQUE/NG SURROUNDINGS Problems RLOOGN/ZING TEM DATE OR DAY OP.WEEK Problems RECOGWITING THE TIME OP.DAy A10.KING.P.T• NIGHT AND TE/NRING IT IS DAY Problems READING•(except as ceased by poor vision) Problems PERFORMING Jos Problems ORIV/NG A CAR (1f could before) Lanousoe skills! Problems FINDING  012  WORDS TO EXPRESS RIK/HERSEL? tostsa sts/szx vocABOLxsz  Problems PROVOONCING WORDS. Problems UNDERSTANDING OTRERS More Lregoent1y-scusersc WORDS Problems CLIPPING ENDS OPT WORDS OR SENTENCES More !requently STUTTERING Problems FINDING NAMES TOR COMMON OBJECTS Problems FORMING ANY INTELLIGIBLE SPEECH  Memory functions: Problems AEMEMBERING PREVIOUS ACTIONS  ON TRZSAME DAY  Probless.XEMEMBERING PAST Lusty-sans . ASKING OCESTIONS REPEATEDLY (despite.ansvers) Problems RECOGNIZING :ACES Or OLD FRIENDS. OR ?AM= ProblemsRMOOGVIELRG NAMES 07•OLD TRIERDS OR ZANILY Problems REMEMBERING BZWLY—LSTRODOCED PERSONS Problems LEAVING STOVE BANNERS, WATER TAPS. AHD LIGHT' SNITOMOS TOREED Oft Problems mA.INTAINSNG. A TRADE orTrecocHT.  rroblems.coscrsrsuanto. .problems• Pm-Gni:maw s.razata. int / smog PLACED 0 15-Tmers zncmussmaur PROSTRATED over problems  resemnering or e" ,""i^g INCREASINGLY DETEYSIVE abbot problems remeMbariag Problems IMMEMBEGUNGI.MPORTAXT PERSONAL DATES SEEMINGLY UHAFOULE OP IMPORTANT CORRETT EVERTS DOES NOT KNOW OWN BAKE Self—care tanctioas, ^ EATING MESSILY WITH' SPOON ouLY mArIss MESSILY WITH SPOON AND WITH SOtt0 FOODS PAS TO BE ?ED SY SOMEONE :=.5r.. •Problems DRESSING, e.g., occIssasAL MISPLACED 30•TTONS -" Problems DRESSING, e.g., POTS ON - CLOTHES  re  XMOOG SECOE3=  .Prob/eMs:DRESSING. e.g., 'GRAZES TO DRESS ssir Problems with occAssosou= NETTS:PG BED Problems with yKEDGMFELY. WETTING BED DOC= racovrtszarr DOES SOT WASS HS-HAEMSHLT ryoocs ?cove SE spas= 3Y soilmoax ELSE CRCOKLMCI (combiao of hair, etc.) tamotoOsTs MOST 3E GROOMED BY SOMEONE ELSE wEEDS COSSTANT'SUPERVISION IN CARING rox.stis  —!  --  140 APPENDIX C  =IC 11701/7 '.:.`fr:LCV12%, /melt trivet I Lod, ro• otil t ratleo. or.^,IL ......^.^ .......... 3aSZOC:100fIslair La. a Ltat fsr sutras^raLLost lsoe• prowls esoenuso lasts oasis calrlog• ease .t law•C•alt peat•em... Altar heat^1.0.414.tt. low taros raw. torbia cast  7 Zs • .^7 • o• ,7 .7 7  •  wart- scour. eas0.7. .came?.. eta tzeaseatly, eat early clays. 'Mete arc be. runt^••••■•g amamattal.  are  O I s 1o. mew freak [bac roger Talon.. asks toe mows !olio cruse-^esreaat Se row foal that . - Osualuido at the, net Tear awe& rick your ralaCLro roe.^c Scow couorlo name ter yeuroellr Do. Too foal. aeramodria Iserwasue canoe. Ler. roar. ratan.. and^ee^rater radosants1111/tas ler yoor taaral-r or sock?  a a a  .32*  1. Op Too foal go.O.arroloadel. roar your. caLanwo's.  ^  bstarrior, yos /sal Larry . ober yea Are aseami year-  a  ^  relano./  a  •/, So you. :sal chat your relative. ourreotly •flour-a year ralaclooelalp . , Lila eclasr eastar ascata•ta at :C -2•04s a a ..ratio. rare  a  .03L  a a  7. Xt. yea afraid at vast Cho tucat• holds for your ralacloel  "T. Os y00- feel your [slat/aro is drOooloot .par 7oa1 .nitt  .7.  a a  ^ be Tee Zeal. Cts/1•04 ob... your arc •reseet yore  a  Q  z•lan..1 .025 •nO. 5e yor^your tssaltla tads ...flared. boaava* at your Lavolvase•t wins your r•lacluel  . t have as^ coca pc/vary It you ...AA LL'ca locaue• et yeare  It. 0e, you i••1 Chat !au dee  a  u.  bo you (sal coac your res.Lal 1.1.e• a•• auffara■  ^  a1Cay.a.t you ate tartar ler your colaclv•I .146  a  a  0  ^  a  13. Os you- loot araces•forta•i• about hartag friend, gluer a•o•us• al your ral•clu•1 It. Do roe !sal nee your raLan..a some to •Nto•Gt ^ 700 amt.. eh ,' Toe c. 444 ea-Val -et hi.T.AULS u LI  CI  LO. [sat •C your arougeoul  Girl se t til■AMICC^1.011.1• I I  LT. 0. yor Led 70.3. az.. Lear coo.zvI. 04 your Ltts• data your cal-o.n.o.' • Lit..._.( •L.I. Do Too rialaLroa could loot^ch. ear. rot yo•r ratan,. co eou.seau maul •1 34 Li. Do you te•LS me ta-Calat aheat V12..1 co 4. aeon roar ralanrel  a a a a a  Tear ralocloet  ..star tor your relanrel  a ^  00. bs you :sal youi••bsealt taa.lelar ...as ter  Z.L. be you (sal you toail le a letter iota is  ^  CI  Cal17 tea Os/ aloe cool/ da•soi ost LS. Do roe foal [Aar Iva !ea' to ^•Pear7 4. sac ler Tear^ oldi.noo  ,173.1.11, 1.41. CA 41.11 ^ bo• Too Coal teas Tea vi.12. S..  ^  a  boaran. hoe^ yaw. coal is .arise for 7.ur raLoclrol .002` O. :ter at all . I. A- Lion, Z. laociaracalr ). 0u.tte • alt ^Larrerely  •  7 .0S  •-•^.0 t tarte. S.l. Lei tort.t. 1. 3. (1.743). ^Ira... Cocas-rid., • (Aoaliable (ro. S.A. Lortc. beearcesoc •4 Cedt.rto.ai 11. 7•-■117 Stwlaa. 44.4•esaaa 0., •lersaec Sundial. ZI•ae reaneyi•soLa. Stair' Uhl ^ r. 0M ^ r Para.  ?•... 14a04. I.S.s.)  141 APPENDIX D  CONSENT FORM RE:  Research Project  - "Factors Affecting Continued Community Living of Patients with Dementia". Investigators - Drs. Beattie, Tuokko, Hertzman Research Associate - Lillian Peters, MSc student, UBC.  The purpose of this study is to examine the patterns of service utilization of dementia victims and their caregivers. We would like to gather information from caregivers of patients with dementia regarding the utilization of formal support services offered by the B.C. Long Term Program. Your experiences with and opinions regarding these services can assist in providing directives for the tailoring of existing services in order to make them more accessible and effective for caregivers. If you •agree to participate, you will be asked 12 questions concerning services you do or do not utilize. The anticipated time fnr this interview is approximately 5 minutes. If at any time during the interview, you would like to withdraw, simply request that the interview be terminated. A completed questicinaire will assume your consent. Your identity and that of the patient will be kept confidential in accordance with established health record procedures. Dr. Holly Tuokko, a neuropsychologist with the Clinic is overseeing the study and can be reached at 228-7031 if you have any further questions. Please sign below if you agree to participate in this study. Your signature indicates that: 1) You have understood the above explanation of this research and agree to participate in the project. 2) You give your consent for a brief interview to be conducted with the research associate of the project. S) You acknowledge receipt of a copy of this letter.  Name^  Signature  Witness^  Signature  Date  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items