Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Depressive cognitive functioning among spousal caregivers of suspected dementia patients: application… O’Rourke, Norm 1995

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

Item Metadata

Download

Media
831-ubc_1995-0267.pdf [ 6.09MB ]
Metadata
JSON: 831-1.0054141.json
JSON-LD: 831-1.0054141-ld.json
RDF/XML (Pretty): 831-1.0054141-rdf.xml
RDF/JSON: 831-1.0054141-rdf.json
Turtle: 831-1.0054141-turtle.txt
N-Triples: 831-1.0054141-rdf-ntriples.txt
Original Record: 831-1.0054141-source.json
Full Text
831-1.0054141-fulltext.txt
Citation
831-1.0054141.ris

Full Text

DEPRESSIVE COGNITIVE FUNCTIONING AMONG SPOUSAL CAREGIVERS OF SUSPECTED DEMENTIA PATIENTS: APPLICATION OF THE HOPELESSNESS THEORY OF DEPRESSION BY NORM O'ROURKE H.B.B.A., Wilfrid Laurier University,  1986  Dip. Ed., University of British Columbia,  1992  A THESIS SUBMITTED IN PARTIAL FULFILLMENT O F THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Counselling Psychology) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April ©  1995  Norm O'Rourke,  1995  In  presenting this  degree at the  thesis  in  University of  partial  fulfilment  of  of  department  this thesis for or  by  his  or  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  representatives.  an advanced  Library shall make  it  agree that permission for extensive  scholarly purposes may be her  for  It  is  granted  by the  understood  that  head of copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department  of  C o u n s e l l i n g  P s y c h o l o g y  The University of British Columbia Vancouver, Canada  D  a  t  e  DE-6 (2/88)  i q q ^ A  P  r i i  18  ;  ABSTRACT  This  the  hopelessness  theory of depression among an older adult population.  Adapted from  the  study  provides  theory  of  the  learned  hypothesized to  exist  first  examination  of  helplessness, hopelessness depression as a specific depressive subtype  heterogeneous grouping of affective disorders. events  are  hypothesized  to  trigger  processes among predisposed persons. to attribute responsibility for negative  is  within a  Salient negative life  depressive  attributional  This entails the propensity events to stable and global  causes, leading to the generalized perception of hopelessness. Seventy were  recruited  One-time their  spousal caregivers  interviews  relative's  predetermined screening assess  within  outpatient  geriatric  assessment.  Caregivers  degree  of  between  association  the  be  construct  a  population.  grouped  within  a  study  was  of  hopelessness are  specific type  the  construct  to of  style. an  examination  caregiver  constrained cognitive  burden among caregivers. may  were  between  hopelessness and  analysis suggests the with  clinic.  Analyses were subsequently conducted  second focus of this  relationship  assessment  matrix on the basis of responses to two depression  hopelessness and depressive attributional A  patients  were conducted with caregivers at the time of  measures.  the  an  of suspected dementia  of  burden.  the This  processes associated  significantly  related  to  It is hypothesized that caregiver burden of  hopelessness depression within  this  The results of univariate a  strong  and  significant  depressive attributional  and multivariate analyses indicated  association between  processes.  hopelessness and  In contrast, depressed persons  who do not present as hopeless do not appear to attribute negative events to stable and global causes. indication  that  These findings provide the first  hopelessness effectively  differentiates  cognitive  functioning within this population of older adults. The related  construct  to  of  expressed  constrained  cognitive  reflect  despair  the  hopelessness also burden  set  objective  variables  epitomized  perceived  overwhelmed by this role.  among  by  appears  significantly  spousal caregivers.  The  by  may  hopeless  those  ideation  caregivers  who  are  This association appears over and above  related  to patient impairment  and duration  of  caregiving. The  sample  recruited  for the  current  study was  compared  against a randomly derived grouping of spousal caregivers from the Canadian Study of Health and Aging (CSHA). between  Demographic similarity  samples would suggest that caregivers recruited for  current study are representative of Canadian caregivers.  the  Based on  this finding, results from the current study can be generalized with greater confidence.  IV  TABLE OF CONTENTS  ABSTRACT  //  LIST OF TABLES  vii  LIST OF FIGURES  viii  ACKNOWLEDGEMENTS  ix  DEDICATION  x  CHAPTER 1 - INTRODUCTION Description of Problem  1  Extent of Problem  2  Purpose of Study  5  CHAPTER 2 - LITERATURE REVIEW Description of Alzheimer Disease  7  Diagnostic Criteria  10  The Study of Caregiver Burden  12  Primary Stressors of Caregiving  12  Secondary Stressors  15  Mediators of Stressors  16  Secondary Appraisal  18  Caregiver Outcomes  19  Depression Among Caregivers  21  Hopelessness Depression  23  Research Questions  27  v  TABLE OF CONTENTS  CHAPTER 3 - METHODOLOGY Subjects  32  Design  35  Data Collection Procedures  36  Instruments  38  CHAPTER 4 - RESULTS Interviewer Comparisons  53  Comparative Analyses  56  Preliminary Analyses  61  Hopelessness Depression Among Caregivers  67  Caregiver Burden as a Outcome of Hopelessness  77  CHAPTER 5 - DISCUSSION Hopelessness Depression among Caregivers  84  Caregiver Burden as a Specific Outcome of Hopelessness .91 Generalizability of Findings  94  Limitations of Study  95  Implications for Counselling and Psychotherapy  98  Future Research  REFERENCES  1 00  105  vi  TABLE OF CONTENTS  APPENDICES The Canadian Study of Health and Aging Population Projections and Anticipated  132 Prevalence  Estimates for Dementia in Canada  136  Burden Interview  137  Geriatric Depression Scale  139  Beck Hopelessness Scale  140  Attributional Style Questionnaire - Revised  141  Demographic Questionnaire  1 51  Marlowe-Crowne Social Desirability Scale  153  Edmonds Marital Conventionality Scale  155  Concurrent Validity of the Geriatric Depression Scale  156  Orthogonal Contrasts Among Caregivers of Demented Patients  1 57  Balanced MANOVA Comparing Attributional Style  160  Factor Analysis of the EMCS  1 63  LIST OF TABLES  Table 1.  Comparison of Responses by Interviewer  54  Table 2.  Descriptive Features of Derived Sample  55  Table 3.  Comparison of CSHA and Full Clinic Samples  58  Table 4.  Comparison of CSHA and Older Clinic Samples  60  Table 5.  Correlation Coefficients and Significance Levels Among Variables  Table 6.  62  Comparison of Responses on Dependent Measures by Gender  Table 7.  64  Orthogonal Contrasts Comparing Attributional Style Among Caregivers  Table 8.  69  Orthogonal Contrasts Comparing Attributional Style Grouped by Quartile on Dependent Measures  Table 9.  71  Multivariate Analysis of Variance Comparing Levels of Attributional Constructs Among Caregivers  Table 10.  74  Comparison of Features Between Depressed Groupings  Table 11.  78  Hierarchical Regression Analysis of Hypothesized Diathesis-Stress Burden Model  Table 12.  Orthogonal Contrasts Comparing Subjects Caring for a Demented Spouse  Table 13.  81  159  Multivariate Analysis of Variance Comparing Levels of Attributional Style Among Selected Caregivers.. 162  Table 14.  Factor Loadings of Edmonds Scale Items on Factors One and Two  165  viii  LIST OF FIGURES  Figure 1.  Hopelessness Model of Depression  25  Figure 2.  Centroids of Three Caregiver Groups  75  ACKNOWLEDGEMENTS  I would like to recognize the many people who have completion of this study an exceptional experience.  made  First of all, I  wish to acknowledge the assistance of my committee.  The subtle  but persuasive encouragement of Dr. Beth Haverkamp has kept me on track and consistently focused.  I would like to thank Dr. Richard  Young for stepping in at a difficult time.  In addition, I wish to  acknowledge the contribution of Dr. Holly Tuokko; not only as a former employer but also a mentor and friend. Completion of this thesis has been greatly aided by the staff of the Alzheimer Clinic.  I wish to especially thank Mr. Daniel Corrin  who volunteered to administer questionnaires. Most importantly, I must acknowledge the men and women who graciously agreed to be interviewed thesis  deals  with the  for this study.  clinical outcomes of  Though this  caregiving, I  remain  most impressed by those who cope well and effectively.  Part of the data reported in this study was collected under the auspices of the Canadian Study of Health and Aging (CSHA). This study was funded by the Seniors' Independence Research Program, administered by the National Health Research Development Program of Health Canada (Project No. 6606-3954 MC(S)).  I would like to acknowledge all of  the staff from across Canada who were involved in the C S H A , in particular, Mr. Richard Aylesworth who helped to make optimal use of this data.  In part, this study was also  supported by the Norcen/Superior Propane - Canadian Association on Gerontology Community Researcher Award received by the author.  DEDICATION  In memory, Violet Rose O'Rourke (1922  -  1991)  and my father, Norman, who cared for her.  Each is responsible for the welfare of others (Shavuot  39a)  1  CHAPTER 1 - INTRODUCTION  For the dreadful thing I feared has come upon me; and that which I feared has come to me.  I am not at ease,  nor am I at rest; nor am I quiet; yet trouble comes. (Job, Chapter 3: 25-26)  Imagine living in the midst of dense fog.  Once familiar faces  seem like those of strangers and actions performed a few earlier  are  quickly  forgotten.  Friends  and  family  moments  gradually  from view and even one's surroundings soon become unclear.  slip Even  one's spouse may be confused with a parent and the memories of children  soon slip away.  All the  while not knowing  amiss, the most basic of tasks become unmanageable. frightening even  to answer  bodily  functions  a phone, hygiene slowly  slip  from  anything  It becomes  becomes unimportant control.  is  Mortimer  and and  Schuman suggest roughly one person in five who reaches the age of 65 may end their days in this condition (1981).  Description of the Despite oblivion  of  the  Problem distress reflected  dementia  provides  in the  patients  security that caregivers are not afforded.  above  with an  description, ironic  degree  The range of  and behavioural disturbances caused by neurodegenerative  the of  cognitive disorders  2  such as Alzheimer disease create extraordinary  demands for those  closest to the patient (DeLongis & O'Brien, 1990; Zarit, Orr & Zarit, 1985).  Considering that as many as 83% of demented Canadians  continue  to  live  in  the  community  (Jeans,  Helmes,  Merskey,  Robertson & Rand, 1987), and that the average life expectancy is ten to twelve years after diagnosis (Davison & Neale, little  wonder  Fengler  and  Goodrich  (1979)  1990), it is  describe  family  caregivers as the hidden victims of dementia. The majority of these caregivers are spouses who themselves are elderly (Canadian Study of Health and Aging Study Group, 1994b; Haley, Levine, Brown, Berry & Hughs, 1987).  Although the patient  may look unaffected and free of physical impairment,  the strain of  caring for a demented adult may be akin to that of caring for an infant.  As noted by Colerick and George (1986), the primary reason  to institutionalize  a demented  of round-the-clock care.  relative is the  unrelenting  Mace and Rabins figuratively  demands  equate  this  role to a 36 hour day (1981).  Extent of the Problem Much of the concern regarding the prevalence of dementia is not simply due to the number of existing cases, but the extent to which the problem is expected to increase.  As dementia  afflicts  conditions  older  Alzheimer  persons,  disease  are  the  incidence  steadily  on  the  Palmertz, Andreasson & Svanborg, 1993). to improvements  of  rise  primarily such  (Skoog,  as  Nilsson,  In part, this may be due  in diagnostic effectiveness.  Yet,  as  populations  3  continue to age in western society, we can anticipate a substantial increase in the prevalence of dementia.  As described by Ineichen  (1987), this trend is akin to a rising tide. In Canada, the median age of the population can be expected to increase well into the next century. percentage  Based upon recent data, the  of persons over the age of 65 continues to increase  relative to other groups (Statistics Canada, 1992).  Though the total  population has increased by 50% since 1961, the proportion of elder Canadians has increased by 128% over this same period.  In fact, the  most recent census indicates that not only is the population aging, but the fastest growing segment is persons over 84 years. As noted by Jorm, Korten and Henderson (1987), the proportion of elderly with dementia increases exponentially with age.  We can,  therefore, anticipate an escalation in prevalence rates not only as a general trend toward an older population but also because of the expanding ranks at the upper end of this continuum. Many  studies  have  attempted  to  calculate  numbers of demented persons in this country. shown  considerable variability.  researchers'  use  of  different  In  the  existing  Yet the results have  part, this population  has  been  bases,  due  to  varying  methodologies and inconsistent definitions for inclusion. Among the strongest studies is the recent Canadian Study of Health and Aging (CSHA; Canadian Study of Health and Aging Working Group,  1994a).  This nationwide,  epidemiological study derived a  random sample exceeding 10,000 Canadians over the age of years.  Subjects were identified from provincial health  64  records in  4  all  provinces  predetermined Mini-Mental  (except cut-off  State  Ontario).  point  Those  on the  Examination  scoring  below  screening measure  (3MS); Teng & Chui,  a  (Modified  1987)  were  invited to undergo a full clinical examination.  On the basis of this  detailed  were derived  assessment (N  suggest  the  institutions  = 2,420), estimates  percentage who  of  currently  persons meet  in  dementia  the  which  community  diagnostic  and  criteria  ranges from 2.4% of those 65 to 74 years to 34.5% of those 85 years or older (CSHA Working Group, 1994a).  Appendix One provides a  more thorough description of the aims and methodology of the C S H A . The figures provided from the CSHA have been applied against population estimates to provide projections for anticipated of dementia in Canada.  cases  Based upon Statistics Canada estimates  (1994), Appendix Two shows how the prevalence of dementia will likely increase in coming years. persons  Not only are close to 300,000  now clinically demented, but by the year 2016  we  can  estimate that over half a million persons will suffer from a form of dementia in this country. It  is  important  necessarily prevalent  have form  Association, symptoms.  to  note  Alzheimer of  1994),  not  disease.  senile 70  that  these  Though this  dementia  other  all  (American  illnesses  also  persons is the  will most  Psychiatric  create  According to Katzman and Jackson (1991),  similar however,  Alzheimer disease accounts for roughly two thirds of all dementia cases  among  older  patients.  This  percentage  corresponds to  5  findings from the C S H A where Alzheimer disease accounted to 64% of all dementia cases (CSHA Working Group, 1994a).  Purpose of the Study The  primary  postulates  of  intent of this study has been to examine  hopelessness theory  Metalsky & Alloy, 1989). was  believed to exist  of  depression  the  (Abramson,  This hypothesized subtype of depression  in this population.  Spousal caregivers of  suspected dementia patients have provided an ideal subject pool to examine  this diathesis-stress model.  According to this theory, cognitive vulnerability a specific subset of persons. events,  Once activated by salient  depressogenic thought  perception  of  exists among  hopelessness  patterns and  lead  to  subsequently  a  generalized  to  depressive  symptoms (Alloy, Abramson, Metalsky & Hartlage, 1988). sought  to  identify  theoretically  should  distinct  attributional  distinguish  this  negative  This study  processes  population  which  from  other  depressed subjects and those who are asymptomatic. Among those who present as hopeless and depressed, the tendency to attribute negative events to stable and global factors should be evident measuring  these  (Abramson, Seligman & Teasdale, 1978). attributions  sought for this theory, interdependent.  at  one  point  in time,  was  which views each of these constructs as  According to  Abramson  et  reasonable to assess depressogenic attributions independently.  support  By  al.  (1989),  it  is  and hopelessness  6  As described by Alloy et al. (1988), a flaw of previous studies which to  have  tested  activate  sufficient  attributional  models is that stressors required  depressogenic attributional  magnitude.  Therefore,  it  style  have  not  been  is necessary for susceptible  subjects to view negative events as sufficiently severe to a  predisposing  tendency  Certainly  manipulation  ethically  problematic  of  for  depressive  this  within  contrast, an Alzheimer outpatient  cognitive  intervening an  of  functioning.  variable  experimental  activate  would  be  setting.  In  clinic has provided a population  of caregivers for whom depression was assumed to exist among a significant Eisdorfer,  percentage  according to  previous  findings  (Cohen  &  1988).  A further objective of this descriptive field study has been to examine the relationship between the construct of hopelessness and burden  among  theoretically  caregivers.  distinct  This  investigation  marks as  a  more  compared  precise to  and  existing  studies. Previous research has demonstrated depression is moderately correlated with expressed burden (Anthony-Bergstone, Zarit & Gatz, 1988;  Drinka,  Smith  Eastham, 1986). unclear.  &  Drinka,  1987;  Fitting,  Rabins,  The nature of this relationship, however,  Lucas  &  remains  In contrast, this study has allowed for an examination of  burden vis-a-vis depressogenic cognitions and hopelessness.  This  line of inquiry was chosen to ideally enhance understanding of the nature and etiology of burden among spousal caregivers.  7  CHAPTER 2 - LITERATURE REVIEW  ... there are two forms of death, not one.  In one form,  everything which holds us in the world, everything love,  may  remain  precious  until  the  last  we  instant.  Everything will stay as it is.  Faces will remain what  they have always meant to us.  In this form of death, life  holds all its beauty to the last second. Then there is the form of dying in which everything familiar  becomes strange,  unknown, loved  everything  true  becomes indifferent,  everything  known becomes  becomes false, everything  everything  pitiful  becomes  pitiless, everything compassionate becomes as hard as a stone.  This room will soon become a prison.  will be locked (Ignatieff, 1993,  The door  p 198-199).  Description of Alzheimer Disease Alzheimer  disease is the  most  prevalent  of  the  syndromes (CSHA Working Group, 1994a; Zee, 1993). disorder  of  later  life,  onset  is  generally  progressive course (McKhann et al., 1984).  dementia Primarily a  insidious  with  a  Incidence increases  exponentially with age, yet symptoms may appear as early as one's thirties (Katzman & Jackson, 1991). Characteristic functioning. generally 1993).  features  Though  entail  considerable  most apparent  is memory  impairment variability  of  occurs,  disturbance (Strub  cognitive what  is  & Black,  Other deficits may include speech, visuospatial perception,  construction, judgment, abstraction, praxis and personality changes (McKhann et al., 1984).  8  To date, there is no cure for Alzheimer disease. large  measure  unknown.  to the  fact  that  This is due in  its cause, or causes,  are  also  What is apparent, however, is that this disorder causes a  progressive deterioration of brain tissue over time.  Upon autopsy,  marked  as sulci have  atrophy  of the  cerebral cortex  is evident  widened, ventricles have become enlarged and gyri have become narrow and flattened (McKhann et al., 1984).  This occurs as a result  of the loss of neurons, or more specifically, the loss of synapses in the cerebral neocortex association areas and hippocampus (Kowall & Beal, 1988). Definitive  diagnosis  requires  the  identification  of  plaques and neurofibrillary tangles upon biopsy or autopsy. appear as small, round areas consisting of the  neuritic Plaques  remnants of lost  neurons and tangles are abnormal protein filaments within the cell bodies of neurons.  Greatest concentration is generally found in the  frontotemporal and parietal regions.  These accumulate through the  course of this disorder yet may be present 10 to 20 years prior to diagnosis (Selkoe, 1992). It would appear the presence of these features is a result of pathogenic  processes resulting from  abnormal  protein synthesis.  As of late, amyloid precursor protein (APP) has become the focus of studies of the molecular biology of Alzheimer disease (Katzman & Jackson, Alzheimer causal.  1991). patients  Its  presence within the  has  led  plaque and tangles of  researchers to  suspect its  role  is  9  Evidence  that  amyloid  may  play  a  critical  role  in  the  pathogenesis of Alzheimer disease comes primarily from studies of Down's syndrome. later  life  Also known as trisomy 21, individuals who reach  (35-40 years)  Alzheimer  patients.  present  In  many  addition  to  of the  impaired  same  features  cognition, it  as  would  appear all persons with Down's syndrome will develop the plaques and  tangles  characteristic  of  Alzheimer  disease  if  they  reach  middle age (Percy, 1993). Down's syndrome occurs as a result of an extra chromosome in position 21.  Markers for Alzheimer disease have also been located  on this location. coded  by  a  In fact, the amyloid (APP)  gene  located  on  discussed above is  chromosome 21.  This  has  led  researchers to consider the operation of a dominant gene in this position as responsible for the onset of Alzheimer disease.  This  genetic hypothesis is supported by the increased prevalence among first degree relatives of dementia Yet  even  among  does not occur. (1980), greater  patients.  monozygotic twins,  complete concordance  In research cited by Jarvik, Ruth and Matsuyama  concordance  among  monozygotic  twins  is  than for dizygotic probands (42.8% and 8%  significantly respectively).  Though this would suggest the operation of genetic factors, it is important  to  concordance  note  that even  is not absolute. a  among genetically This may,  operation  of  diathesis-stress  interaction  of both environmental  identical persons  therefore,  relationship  in  and genetic factors  the occurrence of this disorder (Mohs, 1988).  suggest  the  which  the  determines  10  Diagnostic  Criteria  According to the most recent Diagnostic and Statistical Manual of Mental Disorders (DSM IV; APA, 1994), diagnosis of dementia of the Alzheimer type (DAT) requires evidence of memory impairment in addition to at  least one other deficit  in cognitive functioning.  This may include abstract thought, impaired judgment, change (eg.,  or  other  aphasia,  disturbances  agnosia,  in  apraxia  higher or  personality  cortical  function  constructional  ability).  These disturbances must impede work or social activities and not occur as a result of a reduced level of consciousness (i.e., not delirium).  Other conditions which produce dementia-like symptoms  must also be excluded. The unipolar  most common condition misdiagnosed as dementia depression.  As discussed  this condition among the pseudo-dementia major  elderly  because  of  by Strub and  has often  similarity  in  Black (1993),  been  referred  presentation  depression and conditions such as DAT.  is  It  to as  between  is critical to  discern this difference, however, as depression among the elderly is most  often  responsive to  treatment  whereas  dementia  is  not  (Kaszniak & DiTraglia Christenson, 1994; Strub & Black, 1993). Few changes are evident between DSM IV and the previous diagnostic  manual  (DSM  lll-R;  A P A , 1987).  The  most  notable  revision is that impairment need no longer include remote memory. Also of note, greater emphasis has been placed on disturbance in executive function (i.e., frontal features).  Other changes include  greater delineation between Alzheimer disease and other dementing  11  illnesses (i.e., dementia revisions  were  intended  due to HIV to  create  disease). greater  In general,  compatibility  with  International Classification of Diseases (ICD-10) and the Institute  of  Disease  and  Communicative Related  McKhann et al., 1984) In 1984,  Disorders  Disorders  and  Stroke  Association  -  these the  National  Alzheimer  (NINCDS-ADRDA;  assessment criteria.  NINCDS-ADRDA devised criteria distinct from those  required for diagnosis.  Primarily for research purposes, these are  intended  to  distinguish  among  different  presentations  of  the  disorder.  In part, this was due to the high rate of false positives  based upon previous assessment criteria (McKhann et al., 1984). As earlier  noted, Alzheimer disease can only be  determined by biopsy or postmortem. receive  a probable  ADRDA criteria.  definitively  Thus living patients can only  or possible diagnosis according to  NINCDS-  The features of probable Alzheimer disease include  deficits in two or more areas of cognition, a progressive course and insidious onset. Between however,  do  one quarter not  Jackson, 1991).  have  and one third  such typical  of Alzheimer  presentations  patients,  (Katzman  &  Possible diagnosis may reflect less gradual onset,  plateaus in the course of the illness, only one area of cognitive impairment or the co-existence of other disorders. illness  may  be  identified,  it  is assumed the  impairment is due to Alzheimer disease.  Though a second primary  cause  of  12  The Study of Caregiver Burden A 1980 paper by Zarit, Reever and Bach-Peterson is among the most  cited  in  caregiver  research.  Up  until  this  point,  it  was  generally assumed burden was directly proportional to the level of impairment belief  of the Alzheimer  by  demonstrating  experience. grown  to  patient.  that  Yet this study refuted  burden  is  a  highly  this  idiosyncratic  In effect, this initiated an area of research which has become  gerontology  (Zarit,  one  of  1990).  the  most  Despite  widely  this  studied  interest,  topics  however,  in the  nature of perceived burden as well as its relationship to depression among caregivers, is still not fully understood. Based on the stress and coping model of Lazarus and Folkman (1984), Zarit (1990) has proposed a framework understanding of the demands of caregiving.  for the study and  This perspective views  specific demands or stressors as being mediated and cognitive features.  Situational  by  environmental  factors in this context  include  primary and secondary stressors, which are moderated by appraisal and mediating factors such as coping and social support. caregiver contextual  outcomes such as burden features  outweigh  the  In general,  and depression result  buffering  effects  psychological processes (DeLongis & O'Brien, 1990;  of  when  protective  Pearlin, Mullan,  Semple & Skaff, 1990; Zarit, 1990).  Primary Stressors of Caregiving Primary patient  stressors refer  interaction.  These  to the include  central the  features  demands  of  caregiver/  posed  by  the  13  patient's the  disabilities in conjunction with the contextual  relationship.  aspects of  For instance, this entails the specific tasks the  caregiver must perform as a result of the behavioural and cognitive deficits of one's partner. As noted by George and Gwyther (1986), the exclusive study of caregiving has often minimized the demands of this role.  Looking at  this  the  population  in  isolation  caregivers of dementia  tends  underestimate  patients experience  adults (Eagles et al., 1987). spouse with other  to  strain  relative to other older  For instance, a person caring for a  illnesses such as cancer or diabetes  mellitus  generally does not have to contend with wandering, lashing out or reversal of sleep patterns.  In part, this is due to impairment  cognitive functioning while physical capacity often  remains  of  intact  until the late stages of a dementing illness. In their comparison study with age-matched controls, George and  Gwyther  (1986)  suggest  caregivers  of  dementia  patients  experience three times as many stress symptoms, greatly reduced life  satisfaction,  lower  levels  of  higher  psychotropic drug  social activity.  use  and  As compared  to  substantially cohorts,  this  population is especially vulnerable to decreased well-being  in the  areas  results  of  mental  health  and social participation.  These  underscore the unique demands of caring for a demented relative. Research suggests it is not cognitive impairment per se which creates  excessive  disturbances. aggression  In  correlate  strain, fact,  but  behaviours  more  strongly  accompanying such with  as  behavioural  incontinence  perceived  burden  and than  14  functional  impairment  Gilleard, Whittick these  (Deimling  & Bass,  & Gledhill, 1984).  behaviours  precipitate  the  1986;  Gilleard, Belford,  It has also been suggested decision to  institutionalize  a  demented relative (Chenoweth & Spencer, 1986). Though  one  might  assume  the  extent  of  behavioural  impairment  increases with cognitive decline, this has been  in  studies  recent  (Teri,  Borson,  Winogrond, Fisk, & Kirsling, 1987). weak  correlation  between  &  Yamagishi,  1989;  In part, this may account for the  functional  burden (Baumgarten, 1989). neurodegenerative  Kiyak  refuted  impairment  and  perceived  Underlying the idiosyncratic course of  illnesses, the variability  with which  behavioural  problems occur suggests few consistencies exist. A  further  topic  related  to  the  contextual  features  relationship deals with the gender of the caregiver. are  generally  overrepresented  as  caregivers  in  of  this  Though women most  studies  (Baumgarten, 1989), comparisons have been made to determine if differences general, the  in  experience  exist  (Miller  results are conflicting (cf.  &  Cafasso,  1992).  Anthony-Bergstone et al.,  1988; Fitting et al., 1986; Quayhagen & Quayhagen, 1988). this  may  be  due  to  sampling  unrepresentative of all caregivers. that  reporting  differences  exist  In  methods  which  In part,  often  are  Though it has been suggested between  husbands  and  wives  (Verbrugge, 1985; Zarit, Todd & Zarit, 1986), few other conclusions can be made at this time. One feature related to strain among spousal caregivers is the quality of the premorbid relationship (Williamson & Schulz, 1990).  15  In  contexts  where difficulties  are  reported  prior to the  onset of  dementia, this relates to the perception of burden and depression (Zarit et al., 1986; Gilleard et al., 1984).  In part, this may be due to  the insidious nature of dementing disorders.  According to Woods,  Niederehe and Fruge (1985), it is not uncommon to attribute early disturbances  to  marital factors.  drawing causal conclusions. corroborate this finding. the  premorbid  Yet  caution  is required  before  Prospective studies do not exist to  It must also be noted that perceptions of  relationship  may  be  influenced  by  the  onset  of  disease (O'Rourke, Hayden, Haverkamp, Tuokko & Beattie, 1995).  Secondary Stressors Aside from the the  immediate  consequences of  caregiver's  life.  strain, secondary stressors entail  providing  These  care  include  in  other  restriction  of  domains leisure  of  the  activities,  financial strain and the loss of relationship with one's life partner (Woods et al., 1985). The stress which results from the degeneration of personality may  be  considerable  in  the  latter  stages  of  the  illness.  For  instance, it is not uncommon for a spouse to be confused with a parent  (Zarit  Alzheimer  et  al.,  disease  1985).  often  Though  prevents  the  insidious  caregivers  from  nature noting  of the  progressive change, these episodes strongly reinforce the extent of cognitive decline. A isolation  further many  area  of  study  caregivers  addresses the  experience.  degree  According to  of  social  numerous  16  studies, this is a primary correlate to perceived burden (Haley  et  al.,  &  1987;  Pearson-Scott,  Hyllyluoma, 1983). spouse  are  often  friends  and  family.  Roberto  &  Hutton,  1986;  Soldo  The continual demands of caring for a demented attenuated This  support system which  by  perceptions  includes restricted  otherwise  might  of  isolation  contact  ameliorate  from  with one's  the  frustration  and strain of providing continual care (DeLongis & O'Brien, 1990).  Mediators of Stressors The primary  mediators  proposed in the  have been coping and social support.  caregiving literature  The relationship  these factors and perceptions of strain and burden researchers  to  begin  to  clarify  the  complex  has  between enabled  interaction  among  contextual demands and these buffering factors. A consistent finding among longitudinal studies relates to the variable  nature  of burden.  Though this may be confounded by  behavioural changes over time (Teri et al., 1989), most suggest coping  ability  among  primary  course  of the  illness (Haley  caregivers et al.,  1987;  improves  through  Gilhooly, 1984;  the  Pratt,  Schmall, Wright & Cleland, 1985; Zarit et al., 1986).  This implies a  curvilinear  and  decline.  relationship  between  perceived  burden  cognitive  More precisely, as the demands of this role increase, the  caregiver's ability to cope gradually matches, then surpasses, these demands.  This corresponds to the  proposed by Haley and Pardo (1987).  adaptation  model of coping  17  According to Lazarus and Folkman (1984), a distinction exists between emotion-focused and problem-focused coping. refers  to  efforts  opposed  to  threatening effective  directed  means  at  regulating  directed  condition.  at  the  The former  emotion  constructively  itself  managing  this  the  Though both techniques are instrumental  in  coping, a problem-focused approach is believed to meet  the demands of caregiving more effectively (Zarit, 1990). is  as  a  more  resilient  strategy,  but  may  result  Not only  in  improved  management of the demented spouse (Niederehe & Funk, 1987). It has been suggested that reliance on these distinct coping patterns may change over time. upon  emotion-focused  More precisely, caregivers may rely  coping  in  the  initial  stages  yet,  as  familiarity with the condition increases, coping may become more problem-focused. as Niederehe account  If the latter is a more effective coping technique,  et al. suggest (1987), this change over time  for  the  curvilinear  relationship  between  burden  may and  functional impairment (Winogrond et al., 1987). The effectiveness of social support in mitigating likely  been the  original  most studied topic in caregiving  report  by  Zarit  al.  of family  visits.  suggested.  More  contact  se  per  but  recent  be  more  with  strongest  This study states burden is inversely related to frequency  may  interaction  the  network.  association  is  suggests  The  of  this  burden  (1980)  research.  predictor  Yet  caregiver  et  burden has  complex  research indicates  one's  support  than  initially  it is not volume  satisfaction with one's social network  of  which  18  relates most strongly to perceived stress and coping (Fiore, Becker, Coppel & Cox, 1986; Haley et al., 1987). Scott  et  al.  (1986) demonstrated  In fact, a study by Pearson-  that the  among their sample received the greatest extended In  most  burdened  group  amount of contact with  family. relation  to depression among caregivers,  with network satisfaction Becker, 1987).  has also be suggested (Pagel,  hypothesis  network  satisfaction  (Cohen  &  mediates  cognitions.  By interrupting  support  believed  depressive  Erdly &  Pagel and Becker (1987) provided support for the  buffering  are  an association  to  attributional  Wills, the  1985) effect  this relationship,  protect  the  processes to  by of  depressogenic  perceptions of social  caregiver all  demonstrating  from  facets  of  generalizing his/her  life.  This longitudinal study suggests depressogenic cognitions can exist among caregivers without leading to psychopathology so long as counteracted adequately  by perceived network satisfaction.  Secondary Appraisal The term secondary appraisal is drawn largely from the work of Lazarus and Folkman (1984).  This describes the process by which  people evaluate the adequacy of their resources vis-a-vis the threat posed  by  cumulative  stressors. perception  caregiving situation.  In  caregiving  research,  of  how  currently  one  this  entails  the  feels  about  the  The belief one's resources are adequate  or  inadequate  in this context may be a strong predictor of stability or  breakdown  of the caregiving relationship.  To date, this has been  19  most  objectively  institutionalize  gauged  a demented  with  respect  to  the  decision  to  relative (Zarit, 1990).  It is the conclusion of Gwyther and Spencer (1986) that the placement  decision is a function more of caregiver  than those of the patient. the  characteristics  Though most report the patient's level of  impairment  is  primary  (Chenoweth  & Spencer, 1986;  reason  to  place  one's  Shulman, Pushkar Gold,  relative Cohen &  Zucchero, 1993), the threshold of disruptive behaviours one is able to tolerant varies widely. high  initial ratings  place a demented  For instance, Zarit et al. (1986)  of burden  were predictive  of the  report  decision to  relative.  As discussed above, one's appraisal of social support is likely related  to the  decision to maintain  care within the  home.  In  a  longitudinal study by Colerick and George (1986), initial reports of social network satisfaction, oddly enough, were highest for those who later chose to place the demented patient.  It was a relative  decrease in social support (evident when later measured) which was most  strongly  related  to  the  decision to  institutionalize.  This  suggests changes in one's ability to cope may lead to a reappraisal of the residency decision.  Caregiving Outcomes According to Haley, Levine, Brown and Bartolucci (1987), the stress  and  coping  understanding  the  paradigm etiology  provides of  a  various  useful  framework  caregiver  for  outcomes.  Considering the idiosyncratic nature of perceptions such as burden  20  and depression (Duijnstee, comprehend  the  1992), this model allows one to better  cognitive  processes  which  lead  to  these  perceptions. As  noted  previously  caregivers  is  impairment  (Gilhooly,  processes  play  generally  a  in  this  discussion,  unrelated  1984).  causal  role;  It  to  would  model  (1984)  facilitates  patient's  seem  however,  defined the nature of perceived burden. Folkman  the  burden  level  various  research  among of  cognitive  has  not  fully  Though the Lazarus and  understanding  of  how  burden,  reduced morale or role strain may arise, it does not clearly define what burden is.  This conclusion is held by Vitaliano, Young and  Russo (1991) in their analysis of burden measures. these  instruments  reliably  identify  the  Though most of  condition,  the  construct  itself remains nebulous. In part, this is due to the varying manner in which researchers have  operationalized  burden  (Zarit,  1990).  Though  widely  recognized as a common experience among caregivers, definitional differences impede the ability to draw definitive conclusions. measure 1990)  most widely  used (Burden  Interview (Bl);  Zarit &  The Zarit,  taps each component of the stress and coping model  primarily strain  represents  and  conflict)  a  combination  and  secondary  of  Though  cognitive theory  secondary stressors (role  appraisal.  confirmed by factor analysis (Whitlatch,  yet  This  has  been  Zarit & von Eye, 1989).  has successfully identified  many  of  the  factors which lead to the perception of burden, the condition itself needs to be defined more succinctly.  21  Depression among Caregivers One consistent research finding is the degree to which burden correlates with measures of depression (Zarit, 1990). suggest the 1988),  level  Cohen,  may  be  Kennedy  overstated and  Though some  (Anthony-Bergstone  Eisdorfer  (1983)  believe  et  al.,  clinical  depression exists among a sizeable proportion of those who care for a demented relative. where the  This has been reported in independent studies  percentage  of caregivers who  range from half to over 80% 1988;  meet diagnostic criteria  (Barusch, 1988;  Coppel, Burton, Becker & Fiore, 1985;  Cohen & Eisdorfer, Drinka et al.,  1987;  Haley et al., 1987; Kiecolt-Glaser, Dura, Speicher, Trask, & Glaser, 1991; Rabins, Mace & Lucas, 1982). One depression Considering caregivers,  difficulty, within the it  this  however, population  extraordinary  is  is  defining (Becker  demands  reasonable  for  among a significant percentage.  &  faced  depressive  what  constitutes  Morrissey, by  1988).  many spousal  symptoms  to  exist  It could be argued, in fact, that it  would be more aberrant to be asymptomatic in this context. This relates to a further complication, as noted by Cappeliez (1988).  Diagnostic criteria for depression do not differentiate how  symptoms are expressed among older persons (DSM IV; APA, 1994), thus  failing  to  acknowledge  the  unique  characteristics  populations (Gallagher-Thompson & Thompson, in press).  of  older  From a  behavioural perspective, the demands and environmental conditions faced by the elderly must be considered (Teri, 1991).  Older persons  face  and  health  concerns, shrinking  social  networks  financial  22  constraints  unique  to this age  group.  In  addition, the  somatic  features of depression such as sleep disturbance occur commonly among non-depressed elderly  persons (Shaver  & Brennan, 1992).  These factors complicate the degree to which clinical diagnoses can be made with certainty (Cappeliez, 1988). However, identify  the  population.  the  research is rife with studies which  components Within  of  depressive  a cognitive framework,  study of various attributional the onset of affective  ideation  strive  among  disorders.  this  this has entailed  constructs which coexist or In their review of the  to  the  predict  literature,  Morris, Morris and Britton (1988) summarize findings which suggest certain  attributional  among caregivers.  processes  are  correlated  with  depression  More precisely, the tendency to perceive  the  strain and distress caregivers experience as enduring and likely to impact  all  other  facets  of their  measures of depression.  lives  appears  to  correlate  with  In separate studies, Morris (1986) and  Coppel et al. (1985) demonstrate an association between the extent caregivers severity  of  generalize  attributions  depression.  This  to  other  suggests  the  contexts  and  subgroup  the  whose  pessimism is not limited to the caregiving context are those more likely to become depressed. attributions  widely  The tendency to apply depressogenic  differentiates  who are asymptomatic.  depressed caregivers from  those  23  Hopelessness Depression  Bloody and bowed by the outrages of life, most human beings still stagger down the depression. downward  road, unscathed by real  To discover why some people plunge into the spiral of  real  depression one  must search  beyond the manifest crisis... (Stryon, 1990, pp. 39-40).  Based  on the  reformulated  theory  of  learned  helplessness  (Abramson, Seligman & Teasdale, 1978), the hopelessness model of depression between recent  provides a  attributional  style  developments  in  empirical  studies,  this  depressive disorders. (1988),  revised framework and  affective  cognitive model  describing the disorders.  science  recognizes  and  the  relation  Combining  findings  from  heterogeneity  of  According to Abramson, Metalsky and Alloy  hopelessness depression exists as a distinct subtype of  depression  with  contributory This  a  set  etiological  of  necessary  and  elements.  distinguishes hopelessness depression from  formulation.  For  instance,  the  revised  depressive cognitive processes are the onset of affective disorders. negative  taxonomy  inferences  depressogenic  across  attributional  model  the  1978  recognizes  that  but one factor contributing to  Defined as the tendency to make different  style  creates  contexts biased  over thinking  time,  a  among  certain populations (Alloy et al., 1988). Though it is believed some are predisposed to these thought processes,  these  first  must be triggered  by salient  negative  life  24  events  (Alloy, Abramson & Lipman, 1992).  component  suggests  predisposed persons.  cognitive  This diathesis-stress  vulnerability  exists  among  Once primed, depressogenic cognitive patterns  are theorized to activate a sequence of events in this causal chain (see Figure 1). In  keeping  with the  earlier  theory,  persons susceptible  to  hopelessness depression tend to ascribe negative events to global and stable causes. attributions  are  More precisely, this suggests negative causal  enduring (stable  contexts (Abramson et al., 1989).  over time) and generalized  over  Thus negative events are not seen  as isolated incidents, but indicative of one's future. The  perception  depressogenic expectancy response  of  hopelessness  attributional  that  desired  style. outcomes  in one's repertoire  is  the  Defined  as  will  occur and  not  will change the  result the  of  this  generalized that  likelihood of  no  these  outcomes, hopelessness accelerates this downward cognitive spiral. In effect,  this perception combines and expands upon global and  stable schemata for negative events creating depressogenic synergy between It  these  attributions.  is important  to differentiate  between  this  perceived  state  of hopelessness and circumscribed pessimism (Alloy et al, 1988). In  response  affecting  to  negative  causal attributions  events,  the  in other  latter  may  life domains.  arise In  without contrast,  generalized hopelessness creates a sense of powerlessness along with the belief one's future will be bleak. abnormal  to become  For example, it is not  distraught if one's spouse is  diagnosed with a  25  Figure 1. Hopelessness Model of Depression *  distal  i Negative j Life Events  Depressogenic Attributional Style: globality stablility  Adapted from Alloy et al. (1988).  proximal  Perceived Hopelessness  Hoplessness Depression  26  elementing  illness.  Only when this despair becomes wide  ranging  does depression occur (Fitting et al., 1986). In  contrast  to  most  diagnostic  research,  hopelessness  depression is grounded in theory as opposed to clinical observation. As such, hopelessness depression is defined by its etiology, not by a distinct  constellation  of  symptoms.  This  process-oriented  approach does not see the symptoms of hopelessness depression as markedly  distinct  from  other  affective  disorders.  What  distinguishes this hypothesized subtype of depression is the by which  it arises.  assessment  This enables  measures.  For  instance,  Attributional  Style Questionnaire  along  the  with  appropriate  instruments  researchers  (eg.,  to  revised  utilize  versions  means existing of  Peterson & Villanova,  developed  by  Beck  (1979),  the  1988), remain  in the study of hopelessness depression (Alloy et al.,  1988). To date, the hopelessness theory of depression has not been widely tested.  This is due primarily to its recent development.  few studies that exist provide equivocal support for the model. instance, Alloy et al. (1992) and Metalsky provided  supportive  McEvoy-DeVellis  and  research Bablock  findings. (1992)  The For  and Joiner (1992) have Yet  and  separate  Tiggemann,  studies  by  Winefield,  Winefield and Goldney (1991) have challenge the chronology of the model. The most recent published study by Spangler, Simons, Monroe and Thase (1993) attempted to differentiate patients hopeless within the population of depressed persons.  presenting  as  On the basis  27  of  symptom  attributional  presentation, style  levels  of subjects was  of  hopelessness  and  subsequently assessed.  the This  method appeared to appropriately categorize depressed persons on the  basis of hypothesized causal factors yet  symptoms between others to date, adults.  groups appeared  similar.  the  appearance  of  This study, like  all  employed a diverse grouping of depressed young  The hopelessness model has yet to be tested among older  persons.  Research Questions The framework  of this revised model suggests that a specific  etiology leads to the onset of hopelessness depression (Abramson et al., 1989). a  Once schemata are triggered among predisposed persons,  proportion  will  become  depressive symptoms. one can hypothesize depressed contrast,  operate  the  hopeless and  By approaching this model that all persons who  with  majority of  these  differences  are  present  with  retrospectively,  appear  depressogenic  persons who  depressed are likely invulnerable that mean  eventually  hopeless and  cognitions.  neither  to this diathesis.  In  hopeless nor This suggests  for the constructs of globality  and  stability  should be evident among distinct participant groupings. Related to this point, Alloy et al. (1988) state hopelessness depression is but one condition among the heterogeneous grouping of mood  disorders.  state  that  It  is, therefore,  depression  depressogenic thinking.  can  exist  consistent with the without  the  theory  operation  to of  For instance, depression which results due  28  to  organic  (Gilley,  1993)  or hormonal  Bradley & Davidson, 1982) cognitive  factors  (Manly,  McMahon,  need not require the operation of this  diathesis.  In this study, should the sample be representative of a diverse population of depressed persons, depression should be apparent with and without the presence of hopelessness.  This provides the first  hypothesis of this study.  H: 0  Among the proportion of caregivers who present as depressed, no  significant difference  will  be  evident  in  their  scores  on the hopelessness measure.  H: a  Among the proportion of caregivers who present as depressed, a  significant  difference  will  be  evident  in  their  scores  on the hopelessness measure.  Using measures of hopelessness and depression, this study has grouped subjects within a predetermined  matrix.  were formed  placed on the  in which each subject was  hopelessness  and  depression  scores  (neither  Four  quadrants basis of  hopeless  nor  depressed, hopeless and not depressed, depressed and not hopeless, hopeless and depressed). At this point,  no definitive  regarding subjects who appear chronology  between  onset  of  hypotheses have  been proposed  hopeless yet not depressed. hopelessness  and  The  hopelessness  29  depression is presently unclear.  These subjects may remain in this  state or eventually adopt depressive symptoms. For the purposes of this study, the theory can be tested with greatest  specificity  by  examining  the  attributional  constructs  among those who are hopeless and depressed as compared to the two  remaining  groups.  The  constructs of globality  and  stability  should be more evident among the former group as opposed to other subjects.  In order to test the theory under these conditions, the  following hypotheses were proposed:  H: 0  Means scores for the attributional constructs of globality and stability  will not  who are  hopeless and depressed (HD)  same  H: 0  H: a  means  hopeless  (D-H)  hopeless  (-H-D).  O> g  be significantly  for  subjects  and  subjects  n )hd = ^ g - M d - h s  different  who  between  subjects  as compared to are  who are  these  depressed but  not  neither depressed  nor  = (M-g. M-s)-h-d  Means scores for the attributional constructs of globality and stability are  will be significantly  hopeless  and  different between  depressed  (HD)  subjects w h o  as compared to  these  same means for subjects who are depressed but not hopeless (D-H)  and  subjects  hopeless (-H-D).  who  are  neither  depressed  nor  30  H:  (fig, n )hd * (M-g. M d - h  H:  (M-g. M- )hd * (^g»  a  o r  s  a  S  Md-h  The second intent of this study has been to compare ratings of perceived include  burden  the  to each  attributions  of the of  depressive constructs.  globality  and  stability  These  as  well  as  hopelessness and overall depression. As discussed previously, burden is largely described as an outcome  of the  stress and  coping paradigm  in which  cognitive  processes are viewed as highly salient (Haley et al., 1987; et al., 1990; Zarit,  1990).  Perceived hopelessness is also viewed  largely as a cognitive phenomenon (Beck & Weishaar, 1989). as  a  construct  Pearlin  encompassing all  depressive  thought  If seen  processes  (Abramson et al., 1989), then hopelessness may be strongly related to burden within this population. As expressed by the theory, depression can also arise due to factors unrelated to cognition (Alloy et al., 1988).  Thus if burden is  seen largely as a cognitive process, it is theoretically consistent to speculate that this perception to  a  form  of  depression  thought processes.  H: 0  is more strongly correlated  assumed  to  arise  from  related  dysfunctional  This provides the final hypotheses of this study:  Perceived burden (b) among caregivers will not be correlated more strongly with hopelessness (h) than other depressogenic constructs [total depression (d), globality (g) and stability  (s)]  31  H  o  P bh = P bd = P bg = P bs  :  H:  Perceived burden (b) among caregivers will be correlated more  a  strongly  with  hopelessness  (h)  than  other  depressogenic  constructs [total depression (d), globality (g) and stability  H  a  (s)].  P bh > (P bd = P bg = P bs)  :  Within  applied stress and coping models (cf.  Pearlin et al.,  1990; Zarit, 1990), caregiver burden is believed to result when the demands  of this  resources. one's  patient  surpass a subjective  assessment of one's  In effect, burden results from constricted perception of  ability  context.  role  to cope  relative to the  demands  of the  caregiving  This is not simply an outcome of an objective appraisal of deficits  and  one's  resources,  but  appraisal of the context and one's future.  a  more  interpretive  It is hypothesized that  burden may arise because of depressive ideation arising from the demands of this role and the impact upon how one perceives other life  events.  Abramson activated This  may  caregiver  Similar et  by the lead  to  (1988),  the  hopelessness model  depressive  illness of operation  cognitive  one's spouse (i.e.,  depression may  be  outcome.  proposed  patterns salient  of depressive thought  burden as an eventual  hopelessness context.  al.  to  may  by be  stressor).  patterns  with  This specific form of  idiosyncratic to  the  caregiving  32  CHAPTER 4 - METHODOLODY  The  psychiatric literature on depression is enormous,  with  theory  etiology  after  theory  proliferating  as  concerning  the  richly as theories  disease's about  the  death of dinosaurs or the origin of black holes.  The  very number of hypotheses is testimony to the malady's impenetrable mystery (Styron, 1990, p. 77).  Subjects From  February  to  December  1994,  70  spousal caregivers  (23 men and 47 women) were interviewed for this study. represents initial  88%  study  caregivers  of those entering  criteria chose  (all  not  but  to  the  nine  be  Alzheimer  eligible  interviewed,  This total  Clinic who  participants). two  met  Three  interviews  were  discontinued due to poor English comprehension, two because of effective  anti-depressant  use,  one  due  to  cognitive  impairment  (i.e., lack of capacity for abstract thought), and one because of a pronounced  personality  disorder  (Cluster  B,  presentation  as  confirmed with Clinic Psychologist). Subjects As  for  caregivers  this  fall  study,  into  this  in effect,  were  category  due  derived to  indirectly.  their  relatives'  condition, they likely represent a highly heterogeneous population. The receives  Clinic for referrals  practitioners,  Alzheimer  Disease  and  Related  Disorders  from all regions of British Columbia.  neurologists and other  physicians direct  this facility in order to undergo a complete assessment.  General  patients  to  Both the  33  patient  and  a  caregiver  meet  with  professionals  from  disciplines over a two day period. These include geriatric  various medicine,  speech pathology, neurology, neuropsychology, social work, genetics and psychiatry. Collaboratively functioning Scale  among  is rated  (FRS);  Diagnoses  along various dimensions  Tuokko,  are  Communicative  disciplines, patients'  Crockett,  made  in  Disorders  Beattie,  keeping and  Stroke  -  the  validity  of  multidisciplinary Only caregiver  & Wong,  1986).  National  sample.  Alzheimer  (i.e., daughters, Frazier  and  who  a  determine  who  criteria  identified  protocol  and  (NINCDS-ADRDA;  exceeds  also  distinct  themselves  patients  excluded  daughters-in-law).  Monahan  represent  Disease  of  90%  in  settings.  community-residing  This  Institute  According to Katzman and Jackson (1991),  NINCDS-ADRDA  spouses for  of  Rating  Related Disorders Association assessment criteria McKhann et al., 1984).  level  (Functional  Horton  with  current  (1994), population  were other  As recently other as  cares for a demented  as  the  primary  included family  this  members  noted by  informal  in  Hooker,  caregivers  may  variables  often  personality  relative when  a spouse is  incapable or unavailable. It  was  caregivers  initially of  persons  Alzheimer  disease.  strongest  test  reflection,  proposed  of  however,  diagnosed  It the  that  was  participation with  assumed  diathesis-stress  this  criterion  was  be  probable this  would  model. deemed  limited  or  possible  provide Upon  to  the  further  unnecessarily  34  stringent.  For one reason, the diagnosis is not presented to patients  and their families until roughly two months after to their visit to the clinic (and subsequent to this interview for this study). It must also be noted that an initial diagnosis of not demented does not necessarily mean the patient  is cognitively intact.  The  insidious nature of neurodegenerative disorders may cause deficits to be subtly present several years before a diagnosis can be made (Selkoe,  1992;  Zee, 1993).  Within this facility, 38%  of a recent  sample of persons who did not meet criteria at the time of their first assessment were diagnosed with dementia  22  months  later  (O'Rourke, Tuokko, Hayden & Beattie, 1995). A final participation criterion related to the efficacy and use of  anti-depressant  medication  by  caregivers.  More  precisely,  subjects were excluded where this medication was deemed to be effective.  This  was  determined  by  their  responses  to  the  depression measures (i.e., two caregivers scoring below cut-off on both instruments were excluded).  In contrast, those who presented  as currently depressed were included along with other caregivers as it would seem the medication was ineffective.  Within the current  data set, one caregiver presentated as hopeless and depressed and a second presented as hopeless but not depressed.  These subjects  thus  anti-depressant  met  inclusion  criteria.  medication upon attributional study  for  National NHRDP  which Health  funding  has  (The  efficacy  processes is the topic of a current recently  Research Development  #6610-2140-55).  of  been  received  from  The  Program (Health Canada;  35  Design Eligible scheduled  subjects  were  appointments  identified  within  this  on  the  setting.  basis During  of  their  the  data  collection period, spousal caregivers were invited to participate this  study  (UBC  in  accordance with the  Behavioural  Sciences  approved  Screening  research  Committee,  University Hospital Screening Committee, 93.56).  in  protocols B93-0783;  This methodology  reflects a linear systematic sampling technique (Borg & Gall, 1989). Since the clinic  does  time not  in which reflect  a  patients  and their  caregivers enter  discernible pattern,  it  is believed  the this  sampling procedure did not introduce a significant confound. To test the primary hypotheses of this study, subjects were categorized within a 2 x 2 research matrix.  Dependent  variables  were scores on the Beck Hopelessness Scale (BHS; Beck , Weissman, Lester  &  Trexler,  1974)  and  (GDS; Yesavage et al., 1983).  the  Geriatric  Depression  Scale  Participants who scored above and  below the suggested cut-off points on each measure were placed in one of four quadrants  (neither hopeless nor depressed, N =  54;  hopeless and not depressed, N = 1; depressed and not hopeless, N = 10; hopeless and depressed, N = 5). It should be noted scores on these measures do not provide definitive depression.  indices  that  subjects  met  diagnostic  criteria  for  As noted by Robinson, Berman and Neimeyer (1990),  however, depression screening measures adequately correspond to formal  diagnoses.  More  precisely,  instruments  of  this  nature  possess concurrent validity as compared to psychiatric assessment.  36  Subsequent Attributional  to this categorization,  means  from  the  revised  Style Questionnaire (ASQ; Koch, 1985), were analyzed  among subjects from three of the  four quadrants.  The  primary  research question was addressed on the basis of mean scores for globality  and  stability  as  obtained  from  the  ASQ.  It  was  hypothesized that mean differences on the response measures would be evident among the various quadrants.  Data Collection Procedures During  the  clinic's two  day  assessment,  patients  undergo  neuropsychological testing for roughly an hour and a half. time,  the  caregivers were idle and available  At this  to take part in this  study. It  was  explained  that  Unlike the other interviews participate, this  it was  participation  entirely  voluntary.  in which the caregiver is requested to  explicitly  assessment were  was  stated  separate  that the  from  other  questions clinic  asked  activities  in and  intended solely for research purposes. With administered Nine.  written in the  The  consent order  Burden  obtained,  presented  Interview  (Bl;  questionnaires  in Appendices Four Zarit  &  Zarit  were through  1990)  was  administered during a separate assessment by a social worker blind to the hypotheses of this study. this instrument themselves. were they asked orally. work interview  Subjects were asked to complete  Only if questions remained  unanswered  It should also be noted that this social  was conducted either  two  days prior or two  days  37  subsequent to the administration of the remaining research battery. It is therefore  unlikely that caregivers would assume a connection  between their responses on the Bl and remaining questionnaires. Consistent with the theory being tested, the G D S and the BHS were administered prior to the A S Q in order to prime depressive schemata  (Beck  attributional activated  et  constructs  1979). of  It  was  globality  by these measures thereby  of the revised A S Q . related  al.,  to  hypothesized that  and  stability  increasing the  would  the be  effectiveness  This tactic is supported by the bulk of research  mood-congruent  memory  effects  among  depressed  populations (Mineka & Sutton, 1992). The  final  desirability  instruments  measures  administered  (Edmonds  Marital  were  the  two  social  Conventionality  Scale  (EMCS), 1967; Marlowe-Crowne Social Desirability Scale (MC-SDS), Crowne & Marlowe, 1960) as it was assumed that biased responding was  least  likely  to  occur  toward  the  end  of  the  interview.  Once caregivers had become familiar with the context and the types of questions asked, it was believed they would be less likely to respond defensively.  Administered as the final measures, elevated  scores on the E M C S or MC-SDS would suggest responses on previous instruments might be suspect.  If administered first, however, it is  conceivable scores would reflect elevated social desirability among persons who answered subsequent questionnaires more candidly. This demographic Particularly  rationale  also  questionnaire with  regard  to  justified toward  the  questions  administration end  of  the  of  the  interview.  regarding socio-economic  38  status and income, it was believed older participants would be more likely to provide this information once they had become comfortable in the setting and familiar with the Subjects were with this  all  but  method  consistent  individually interviewed  Burden  is optimal  with  assessment. collection  the  the  interviewer.  Interview.  for  vast  According to  research with older body  This interview  procedures  and orally  of  format  within  the  experience  presented  Fry  (1986),  populations and in  gerontological  is consistent with other  data  clinic  were  and  caregivers  familiar with this mode of interaction. Samples of likert scales were provided to subjects in order to facilitate responding on the A S Q .  This was a card the caregiver was  able to keep in view with each of the response alternatives and the corresponding number for each.  Instruments Burden  Interview  The Burden Interview was specifically designed for caregivers of  dementia  patients  (Bl; Zarit  & Zarit,  1990).  Though  more  heterogeneous measures exist (cf. Caregiver Strain Index; Robinson, 1983),  the  idiosyncratic  features  of  neurodegenerative  illness  necessitate an instrument designed to assess the unique demands of this caregiver role. As presented in Appendix Three, the Burden Interview asks a series of 22 questions regarding perceived strain in caring for the demented relative (Zarit & Zarit, 1990).  The degree to which the  39  caregiver would endorse each statement likert scale. two  This provides an added degree of sensitivity.  subscales exist  score  is generally  Beattie,  (personal reported  strain  & role strain),  (Hadjistavropoulos,  Though  a  cumulative  Taylor,  Tuokko &  1994).  Internal reliability coefficient at 0.88 1994b) and 0.91 1985).  is rated along a five point  has been  (Hassinger, (Gallagher,  Test-retest  reported  1985), 0.89  with Cronbach's alpha (CSHA Working Group,  Rappaport, Benedict, Lovett & Silven,  reliability  has been estimated  at 0.71  by Zarit  and Zarit (1990), however, no time frame was indicated.  Likely this  is  perceived  a  relatively  brief  period  considering  estimates  of  burden will fluctuate during the course of the illness. Zarit  and  established with  the  Zarit  (1990)  by correlating  total  score  Inventory (r = 0.41;  (r  report  a single =  0.71)  concurrent  validity  independent and  with  has  rating  the  of  Brief  been burden  Symptom  Derogatis, Lipman, Covi, Richels & Uhlenhuth,  1970).  Geriatric Depression Scale As noted by Seligman (1973), depression is like the common cold  of  Though  psychopathology the  focus of  --  at  once  considerable  familiar  research  decades, understanding is far from complete.  and  over  This is  true of depression among older persons (Gurland Jarvik, 1976; Salzman & Shader, 1978).  the  mysterious. past  four  particularly  & Toner,  1982;  40  Several consider  well-validated  the  environmental  older populations.  self-report factors  and  unique  exist,  yet  few  characteristics of  An exception is the Geriatric Depression Scale  (GDS; Yesavage et al., 1983). instrument  measures  As presented in Appendix Four, this  asks respondents to indicate if each of 30  are true or false.  statements  As opposed to measures with more complex  scoring systems (cf. Zung Self-Rating Depression Scale, 1965), this true/false format  is believed to be more appropriate for a broader  range of older adults (Scogin, 1994; Sheikh & Yesavage, 1986). The initial validation study suggests a score of 11 or greater on this instrument with a 95% 1982).  indicates the existence of possible depression  specificity rate or 5%  false  negatives  (Brink  Though a higher cut-off score of 14 virtually  rate of false  positives, sensitivity  (i.e., beta errors).  hundred  researchers  as  reduces the 84%  to  items  were  potentially  this sample. initially  chosen by clinicians and  differentiating  between  depressed older adults (Yesavage et al., 1983).  normal  were  (median  selected  correlation  which 0.68;  correlated range  0.47  best with the to  0.83).  study  coefficients administrations  by  range  Abraham from  (1991),  0.69  (39 week period).  to  30  total score  Split-half  Cronbach's alpha each indicate reliability coefficients of 0.94. separate  and  Administered to a  representative sample of persons over the age of 55 (N = 47), items  80%  For the purposes of this study, it was proposed a  score of 11 be used differentiate One  is reduced from  et al.,  and In a  Kuder-Richardson  reliability  0.88  repeated  over  18  Though the author suggests this  41  may  reflect  nature  a  degree  of  insensitivity  considering the  transient  of this condition, Yesavage (1991) notes depression is far  more stable among older persons. Concurrent validity has been establish vis-a-vis the  Hamilton  Rating Scale for Depression (HRS-D; 1967) and the Zung Self-Rating Depression Scale (SDS; 1965) according to Weis, Nagel and Aaronson (1986).  In fact, the G D S appears to differentiate among normal,  mild and  severely  depressed older adults  with equal  or  greater  sensitivity as compared to these other instruments (Yesavage et al., 1983).  As compared to persons previously assessed according to  Research  Diagnostic  Criteria  for  a  major  affective  disorder  (RDC; Spritzer, Endicott & Robins, 1978), it would appear the G D S provides a high degree of convergent validity (Weis et al., 1986). It is noteworthy that the G D S appears to hold less construct validity  as  compared  to  Research  Diagnostic  Criteria  other screening instruments (Yesavage et al., 1983). primarily (eg.,  to the  absence of items  the  This is due  measuring somatic complaints  sleep disturbance, weight loss,  symptoms).  than  cardiac and  gastrointestinal  During development, somatic items among the  initial  pool provided low inter-item coefficients (Brink et al., 1982).  The  relative effectiveness of the G D S as compared to other measures may thus result from the high percentage of asymptomatic elderly persons  who  experience  also of  endorse  depression,  sensitivity to instruments  these  items.  somatic  assessing the  among the elderly (Scogin, 1994).  Though  items severity  provide  integral  to  marginal  of this condition  42  A further correlation 1989).  strength of the G D S relates to its relative  with  social  desirability  response  sets  (Cappeliez,  Compared to the Beck Depression Inventory (BDI;  Steer, 1978), for instance, the G D S appears to be more among  older  adults.  This  suggests  depressed  lack of  Beck & effective  elders  more  accurately endorse items from the G D S .  Beck Hopelessness Scale It  has  become  hopelessness (Beck  increasingly  differentiates  & Weishaar,  apparent  subgroupings of  1989).  For instance, the  that  perceived  depressed  persons  risk of suicide is  greater among those who operate with this generalized perception of  despair  and  futility.  Cole  (1988)  suggests hopelessness  continues to be related to suicidal ideation even after controlling social desirability and depression. As presented in Appendix Five, the Beck Hopelessness Scale (BHS; Beck et al., 1974) this area.  has become the definitive  instrument  in  A total score is obtained from 20 statements for which  responses are gauged on a true/false format. positive and nine negatively keyed items.  This consists of 11  The entire scale can be  administered within 10 minutes in most instances. Psychometric research has suggested a cut-off score of 9 or greater  provides a reliable  moderately  distinction between  hopeless persons.  asymptomatic  As scores rise, the  severity  breadth of pessimistic cognitions are believed to increase.  and and  43  As  reported  reliability distinct  normative  populations.  (Beck & Steer,  of the  1992).  0.82  1988),  and 0.93  internal  among seven  Though test-retest  after one week; 0.66  sensitivity  (Dowd,  manual  coefficients range between  lower (0.69 the  by the  reliability  is  after six weeks) this may reflect  instrument  relative  to this  labile construct  A recent study reports Kuder-Richardson (KR-20)  coefficients of 0.92  among a diverse sample of 957 adults (Young,  Halper, Clark, Scheftner & Fawcett, 1992). As defined by Stotland (1969), hopelessness is a generalized pessimistic cognition encompassing negative and one's future.  attitudes  toward  This is the construct by which the  self  BHS was  developed and would appear to provide adequate content validity (Dowd,  1992).  Items were selected on the basis of clinical and  research experience and factor analysis would suggest this domain has  been  well  theoretically  defined.  consistent  The with  instrument  Beck's  own  also  appears  model  to  be  of  depression  reports  significant  (Owen, 1992). The  manual  concurrent (0.74  validity  &  with  Steer,  clinical  1988) ratings  of  hopelessness  among a sample of general practice patients and 0.62 among  suicidal was  (Beck  subjects).  0.86  (p  <  Inter-rater  .001).  reliability  Further  among  the  two  concurrent validation  judges  has  been  established between the BHS and the BDI (pessimism item of the BDI removed) with significance levels exceeding the .01% level. is  important  perfect.  to  note  correlations  are  significant  yet  are  It not  In the latter case, this would suggest the BHS is only a  44  secondary depression measure.  With coefficients between 0.44  0.74,  instruments  this  different  suggests  the  two  measure  and  related  yet  constructs.  Beck & Steer (1988) state the BHS is appropriate for adults between have  the ages of 17 to 80 years.  been  conducted with older  Though relatively few studies  populations,  the  few  available  support this contention (Hill, Gallagher, Thompson & Ishida, For instance, Abraham (KR-20)  across  18  (1991), reports mean  time  points  of  0.80  reliability  among  1988).  coefficients  elderly  subjects  (SD = 0.04). A  hopelessness  populations Fry, 1986) (Hayslip,  is  measure  available  specifically  (cf.  Geriatric  designed  for  older  Hopelessness  Scale;  yet it has received little support in subsequent studies Lopez & Nation,  consistency  and  poorly  1991).  defined  Based upon factor  modest  internal  these  authors  structure,  conclude the G H S "...has modest potential for clinical and research use with community-residing aged persons" (p. 504).  Though the  G H S may be suitable for psychiatric inpatients (Trenteseau 1989), this suggests the appropriate  for  & Hyer,  Beck Hopelessness Scale may be  non-institutionalized  older  persons  more  in contrast  to  Fry's earlier assertions (Hayslip et al., 1991).  Attributional The to  assess  original  Style Questionnaire Attributional  depressive  Style  cognitive  - Revised Questionnaire  processes  Abramson, Semmel & von Baeyer, 1979).  was  (ASQ;  designed Seligman,  Using a combination of  45  open-ended responses and fixed measures, the subject reports the cause of hypothetical events then provides ratings along a series of likert  scales  reformulated  to  tap  theory  attributional of  1978), this instrument  learned  dimensions.  Based  helplessness (Abramson  attempts to determine  on  the  et  al.,  the degree to which  persons perceive internal, stable and global factors as responsible for negative scenarios. primes  subjects  for  As an instrument which not only gauges but depressive  thinking  it  is  unique  among  cognitive measures (Parks & Hollon, 1988). The A S Q appears to discriminate between depressed and nondepressed subjects (Eaves & Rush, 1984; Persons & Rao, 1985), yet only a subset of depressed persons have been found to exhibit this purported  attributional  suggest the  style.  Though Carlton and  scale is less sensitive than  Hollon  specific, these  (1988) findings  support recent revisions to the theory (Abramson et al., 1989). As this attributional  style is now hypothesized to exist among  only a specific subset of depressed persons, subjects must first be identified as hopeless and depressed in order to test the existing theory.  In fact, the revised construct of hopelessness depression  makes a majority of previous research moot and academic. studies  are  beginning  (i.e., Alloy et al., 1992;  to  appear  specific to  the  revised  McEvoy-DeVellis et al., 1992;  Though theory  Metalsky et  al., 1992; Tiggemann et al., 1991), the validity of the A S Q remains to be conclusively determined. As university  the  original  populations,  ASQ many  was of  developed the  and  hypothetical  tested  among  contexts  are  46  inappropriate  for  older  adults  (Koch, 1985).  For instance,  the  statement "you go on a date and it goes badly" would be far less relevant to most persons in later years.  As presented in Appendix  Six, Koch (1985) has revised the original version with items more suited to older adults. This revised A S Q relies on negative  hypothetical events as  these have been found to distinguish between depressed and nondepressed subjects more effectively In  contrast  to  the  reformulated  (Peterson & Seligman, 1984).  theory  of  learned helplessness  (Abramson et al., 1978), depressed subjects are not unique in their causal attributions  for positive events.  Koch (1985) incorporated  this finding in his revised version of the instrument. Two versions were administered adults.  The  contained  first  consisted of  seven.  coefficients (0.80  The  ten  former  vs. 0.70)  14  months apart  items  whereas  provided  to older  the second  stronger  reliability  and is the version proposed for this  study. As  with  most  studies  based  on  the  theory  of  learned  helplessness (Abramson et al., 1978), methods used by Koch (1985) to validate this instrument are now less appropriate.  For instance,  his research did not categorize subjects along the  continuum of  hopelessness. Koch (1985) attempted to determine the predictive validity of this  ASQ  subjects.  by  administering  five  unsolvable  By comparing attributions  anagrams  to  test  given after this exercise, it  was hoped a high correlation coefficients would result.  Statistical  47  significance  was  not  apparent  for  stability,  globality  nor  internality. Not only did this procedure fail to assess hopelessness among subjects, activate  it is also unlikely that this procedure was sufficient to depressogenic attributions.  According to  the  revised  theory (Alloy et al., 1988), the diathesis-stress component requires negative events to be of sufficient magnitude (Follette & Jacobson, 1987).  This has been a weakness of most studies of  attributional  style (Alloy et al., 1988). Though  it might  appear  Koch's  dissertation  invalidates  this  revised A S Q (1985), weaknesses are with the procedures employed as opposed to the instrument itself.  With respect to the theory of  hopelessness depression (Abramson et al., 1989), this questionnaire remains untested.  The original notions employed by Koch in revising  the A S Q for older adults remain valid. instruments  have  been  Though  more  current  developed (cf. Expanded Attributional Style  Questionnaire; Peterson & Villanova, 1988), few address an inherent bias toward younger populations. version  was  For these reasons, Koch's revised  used in this study.  As Alloy et  al.  (1988)  state,  the general format of this instrument remains the most appropriate method to gauge depressogenic attributions.  Demographic Questionnaire A features  questionnaire  was  of this sample  created  (see  to  examine  Appendix Seven),  the  demographic  including factors  such as gender, socioeconomic status and number of years married.  48  In order to group subjects for subsequent analyses, it was advisable to  demonstrate  statistically  along these domains.  significant  Otherwise  differences  did  not  exist  this would reduce the ability to  generalize results to a broader population.  Marlowe-Crowne According  Linden,  is  advantageous to include a measure to gauge social desirability  in  studies.  Defined  answers which make  Paulhus  Scale it  self-report  to  Social Desirability  as  one appear  a  and  Dobson  systematic  (1986),  tendency  give  more positively, these response  sets often act as a significant confound in face-to-face (Paulhus, 1991).  to  interviews  It is believed persons scoring higher on social  desirability measures are less likely to respond accurately on other measures socially  that gauge  sanctioned  studies using the  attributions  (eg.,  and  behaviours  suicidal ideation).  This  which  are  is apparent  not in  BHS according to Strosahl, Linehan and Chiles  (1984). In recent years, socially desirable responding has come to be viewed  as an  increasingly complex phenomenon.  In addition  to  conscious distortion, persons may also under report various beliefs and  behaviours  proposed  a  with  limited  two-component  impression management (an  honest,  yet  overly  awareness. model  which  Paulhus  (1984)  distinguishes  has  between  (conscious dissembling) and self-deception positive  self-presentation).  This  model  suggests under reporting on various instruments may not be solely intentional,  but  also  indicative  of  a  self-protective  psychological  49  stance.  As an example, Teusink and Mahler (1989), suggest denial  functions as a common coping strategy relative's  early  in the  dementing illness.  As noted by Paulhus and Reid (1989), the Social  onset of a  Desirability  remains the  most widely  Appendix Eight). impression  Scale  ( M C - S D S ; Crowne  &  Marlowe-Crowne Marlowe,  1960)  used measure of response biases (see  This 33 item instrument is believed to tap both  management  and  self-deception  domains,  likely  weighing more heavily on the latter. Concurrent validity has been established between the M C - S D S and  other  social desirability  measures  (Paulhus,  1991).  This  includes the Edwards Social Desirability Scale (1957) even though this measure gauges self-deception more heavily. Paulhus  (1991), alpha coefficients range  published studies.  Test-retest  reliability  from  As reported by .73  to  as reported  and Marlowe was .88 over one month (1960).  .88  within  by Crowne  Fisher (1967) reported  a coefficient of .84 over a one-week interval.  Edmonds Marital Conventionality Scale A second measure of response biases was included in this study  to  assess  the  candidness of  responses on  the  Burden  Interview.  Unlike the other dependent measures included in this  study, the  construct of burden is inherently  More  precisely, this construct is inextricably  relational  grounded within the  marital  union and  perceptions of one's spouse.  unitary  constructs  such  as  or dyadic.  hopelessness or  In  contrast  depression  to  which  50  involve the marriage  individual  and  appropriate  the  exclusively, burden  caregiving  is perceived within the  context.  It  was  thus  to include this second social desirability  deemed  measure  as  responses to M C - S D S may not reflect biased responding on the Bl. It  would seem plausible that a caregiver  monitor  responses  relative  to  his  or  her  might  defensively  relationship  while  responding more candidly on individual or unitary measures (or vice versa).  The former  may be particularly  cohort of elderly persons.  true among the  This generation  current  of spousal caregivers  was socialized within a culture which strongly valued the union;  especially older women  Gilligan, 1982;  (Chodorow,  Miller & Cafasso, 1992;  who constitute the majority  1978;  marital  Friedan,  1993;  Pearlin & Schooter,  1978)  of spousal caregivers (CSHA Working  Group, 1994b; Lawton, Brody & Saperstein, 1989).  Caregiving may  therefore  within this  group.  be viewed  as a self-defining endeavor  age  It is reasonable to speculate that many would experience  reticence reporting negative perceptions stemming from this role. The  Edmonds Marital  specific statements (EMCS;  1967).  Conventionality  Scale  asks  whether  regarding one's relationship are true or false  It is believed persons rarely endorse such items  without distorting the  context  in which the  marital  relationship is  presented (i.e., If my spouse has any faults, I am not aware of them). It  has  statistically  been  suggested  significant  role  marital  conventionality  in scores on both the  plays  a  Locke-Wallace  Scale of Marital Adjustment (Edmonds, Withers & DiBatista,  1972)  and the Relationship Inventory (Schumm, Bollman & Jurich, 1980).  51  It has also been shown to be an important factor in the assessment of  primary  relationships  among  specific  clinical  (Rychtarik, Tarnowski & St. Lawrence, 1989).  populations  Grigg (1994) recently  reported an alpha coefficient of .75 among alcoholic men and their wives. Weighted scores for the E M C S range from 0 to 89. original format contained 50 this scale to a 15  items, factor  item weighted  Though the  analysis later  questionnaire.  The  reduced  correlation  between the short, weighted version and the original long form was very  high (r = 0.99)  with reported  ranging from 0.80 to 0.93 administered  to  100  internal  consistency estimates  (Zweben, Pearlman & Li, 1988).  married  university  established a mean of 34 (SD = 30).  students,  this  Initially sample  From this study, the author  recommends a cut-off score of 20 (Edmonds, 1967); however, it is not entirely clear how this was determined. To obfuscate the instrument's intent, Edmonds suggests the 15  items  be  desirability.  mixed For this  with  others  study,  which  items  were  do  not  measure  interspersed  social  with  four  others selected from the Marital Status Inventory (Weiss & Cerreto, 1980) lead  (see Appendix Eight). respondents to  believe  It was assumed these items would the  questionnaire  assess the stability of the relationship.  was  intended  to  Though only four additional  items were included, the information solicited by these items is far more direct.  It was assumed this total was sufficient to convey an  impression of the scale distinct from the E M C S alone.  52  Though some authors suggest the E M C S should be used to screen subjects (Edmonds, 1967; Edmonds et al., 1972; Rychtarik et al.,  1989), this does not appear appropriate  construct(s)  measured  are  not  entirely  considering that the  clear  (Cappeliez,  1989).  Furthermore, a definitive cut-off score has yet to be identified.  Functional Rating Scale The Functional Rating Scale (FRS; Tuokko, Crockett, Horton & Wong, 1986) the  disability  of  patients.  designed to quantify along  a  series  occupational  care,  DSM  (APA,  (McKhann  et  of  eight  affect,  al.,  This  multidimensional  five-point home  and  language,  1994) 1983),  and this  (memory,  hobbies,  problem  NINCDS-ADRDA measure  was  impairment  scales  orientation).  assess  scale  levels of cognitive and functional  functioning,  personal IV  is used in this clinic to objectively  Beattie,  social/ solving,  Consistent diagnostic  provides a  rating  with  criteria of  full  severity while recognizing uneven rates of decline across various activities  of daily  living.  reliability  coefficients  Tuokko et al. (1986)  ranging from  .63  to  .93  report  inter-rater  and 94.2%  correct  classification as compared with neuropsychological tests.  These  findings support the reliability and concurrent validity of the F R S .  53  CHAPTER 4 - RESULTS  Most people in the grip of depression at its ghastliest are,  for  whatever  reason,  in  a  state  of  unrealistic  hopelessness, torn by exaggerated ills and fatal threats that bear no resemblance to actuality. on  the  almost  part  of  friends,  It may require  lovers, family,  admirers,  religious devotion to persuade the  an  sufferer  of  life's worth, which is so often in conflict with a sense of their own worthlessness (Styron, 1990, p. 76).  Interviewer Comparisons Each subject was interviewed  by the author  (N = 58)  or a  student volunteer (N = 12) who was blind to the hypotheses of this study.  A comparison of responses obtained by both interviewers on  all critical  measures  is shown  in Table  1.  A series of  indicates no significant difference exists between finding would suggest that interviewer responses derived for this study.  responses.  bias has not  t-tests This  contaminated  With this finding, all subjects  have been combined for subsequent analyses. All  caregivers  were administered Nine.  were  individually  interviewed  in the order presented  and  measures  in Appendices Four to  The time required to complete each assessment ranged from  40 minutes to 1.5 hours.  Table 2 provides a listing of descriptive  statistics for both patients and their caregivers.  54  Table 1. Comparison of Responses Obtained by Interviewers  Measure  Mean (SD)  T-Value  g  • author (N = 59)  6.5 (5.62)  0.53  0.60  • student (N = 12)  5.9 (4.17)  -0.15  0.88  0.70  0.48  0.08  0.93  -0.46  0.65  -0.18  0.86  0.78  0.44  Geriatric Depression Scale:  Beck Hopelessness Scale: • author (N = 59)  3.9 (3.26)  • student (N = 12)  4.1 (2.35)  ASQ:  Internality:  • author (N = 57)  41.9 (6.60)  • student (N = 12)  40.4 (7.55)  ASQ:  Stability:  • author (N = 57)  37.9 (10.2)  • student (N = 12)  37.7 (7.08)  ASQ:  Globality:  • author (N = 57)  36.9 (9.20)  • student (N = 12)  38.3 (10.3)  Edmonds SD Scale: • author (N = 58)  6.8 (4.21)  • student (N = 12)  7.1 (3.45)  Marlowe-Crowne SD Scale: • author (N = 58)  20.8 (5.20)  • student (N = 12)  19.5 (4.87)  55  Table 2. Descriptive Features of Derived  Feature  Sample  (N)  Mean (SD)  Alpha  Caregiver Age (70)  66.3  (9.81)  - -  Patient Age (70)  69.3  (8.38)  - -  Years Married (70)  36.9  (14.4)  - -  Total Family Income (69)  $ 36 ,870  Functional Rating Scale Total (69)  22.0  (5.61)  - -  Duration (Months) of Symptoms (69)  54.2  (36.2)  - -  Patient Depression Scale Total (59)  6.78  (5.48)  0.87  Caregiver Depression Scale Total (71)  6.34  (5.39)  0.87  Caregiver Beck Hopelessness Total (71)  3.96  (3.11)  0.77  Caregiver A S Q Internality (70)  41.7  (6.74)  0.34  Caregiver A S Q Stability (70)  37.9  (9.72)  0.77  Caregiver A S Q Globality (70)  37.1 (9.34)  0.74  Burden Interview Total (69)  27.2  0.94  *  (16.1)  (3)  (21,480)  Cronbach's alpha was also used as a measure of internal reliability with these  instruments though each employs a likert scale. consistency.  This likely inflates indices of internal  According to Anastasi (1988) a KR-20 would be more appropriate,  however, this calculation is not available in S P S S x version 3.0 ( S P S S Inc., 1988).  56  Comparative Analyses As patients pervasive  noted (and  relative  Gilley  (1993),  subsequently  selection  convenience, are  by  studies  their  biases.  involving  caregivers)  Most often  are  derived  as  Alzheimer subject  samples  to of  it is often unclear how representative these samples to  the  populations  (Dura & Kiecolt-Glaser, 1990).  from  which  they  are  drawn  For this reason, subjects recruited  for this study were compared against a corresponding sample of community-dwelling  spousal caregivers of  persons with  dementia  identified from randomly selected individuals in the Canadian Study of Health and Aging (CSHA; Canadian Study of Health and Aging Working Group, 1994a).  Appendix One provides a detailed overview  of the CSHA. In all provinces (except Ontario), persons over 64 years of age were  randomly  As these  selected  sources represent  from  computerized  the vast majority  health  records.  of persons in this  population, it is quite probable that the derived C S H A  sample is  highly representative of older Canadians. To facilitate  the comparative analyses, spousal caregivers of  community-dwelling  elders  were  identified  from  all  regions.  According to the protocol of the C S H A , caregivers were  identified  and interviewed when a subject scored below a predetermined cutoff  point  on the  screening measure  Examination (3MS); Teng & Chui, 1987). suspicion  of cognitive  impairment  (Modified  Mini-Mental  State  This would suggest credible  exists among this grouping of  C S H A patients, similar to those persons referred to this clinic.  57  It  must  be  noted,  however,  that  various  sampling  purposefully distorted the initial C S H A sample pool. all  regions  (Atlantic,  were equally  Quebec, Ontario,  represented  in spite  Prairies,  of  For instance,  British  population  biases  Columbia)  differences.  In  addition, older age groups were over-sampled. For  this  Coordinating population  reason, corrective Centre  from  weights  at  the  University  which  the  sample  were computed  of  was  Ottawa  to  drawn.  by  the  reflect  the  Based  upon  Statistics Canada data (1992), subjects were categorized by region, gender and age group in order to apply these corrective  weights.  Once applied, the C S H A sample is again reflective of the Canadian population. As shown in Table 3, Burden Interview totals for both Clinic and  CSHA  samples  significant difference 3MS scores.  do  not  differ  significantly,  in patient impairment  spite  of  a  levels as suggested by  Also of note, the gender composition between groups is  indistinguishable (chi square = .000, p = 1.00). majority of caregivers pretty good.  in  rated  their  health  In both samples, the  as satisfactory,  good or  On average, both groups had some secondary education  but the majority did not complete high school. Each measures.  sample The  completed  CSHA  separate  caregivers  were  self-report  depression  given  Center  for  1977)  and  the  Epidemiologic Studies Depression Scale (CES-D; Ratloff,  the clinic sample was administered the Geriatric Depression Scale (GDS;  Yesavage  distribution  of  et  al.,  responses  1983). on  both  To  facilitate  measures  comparisons, were  converted  the to  58  Table 3. Comparison of C S H A and Full Clinic Caregiver Samples  Feature  Mean (SD)  (N) *  T-Value  Caregiver Age: • C S H A (N = 113)  71.4  10.6  • Clinic (N = 70)  66.3  9.81  • C S H A (N = 114)  43.9  13.9  • Clinic (N = 70)  36.9  14.4  • CSHA (N = 102)  24.4  16.6  • Clinic (N = 68)  26.6  15.9  • CSHA (N = 114)  77.2  6.11  • Clinic (N = 70)  69.3  8.38  • C S H A (N = 106)  60.4  14.4  • Clinic (N = 67)  72.4  17.7  2.13  0.04  2.10  0.04  -0.56  0.58  4.27  .001  •2.91  .005  Years Married:  Burden Interview:  Patient Age:  Modified Mini-Mental  State:  The sample size shown for C S H A subjects reflects the total derived on the basis of selection criteria.  A smaller statistical N, however, was obtained subsequent to  application of corrective weights.  These smaller numbers are not shown as they do not  reflect the number of actual subjects upon whom descriptive statistics are based.  59  standard scores.  A 16 point score of a possible 38 point total was  used as a cut-off in the C S H A (CSHA Working Group, 1994b).  This  corresponds to a z score of .41 with 39% of the sample falling above cut-off.  A score of 11 or greater on the G D S categorized 20% of the  clinic sample.  This corresponds to a z value of  .93.  Though  different measures were used, this finding would indicate the C S H A sample was somewhat groups  suggests half  thresholds  between  more depressed.  The difference  a standard deviation the  respective  between  separates depression  groupings.  This conclusion  assumes a normal distribution of scores on both measures.  A closer  scrutiny of G D S scores, however, indicates a moderate degree of positive skewness (1.157).  This suggests caution is required in  comparing depression levels between groups. Also difference  of  note,  exists  t-tests  between  in  Table  groups  on  3  indicate  all  a  age-related  (i.e., years married, age of patients and caregivers).  significant variables  This is likely a  result of the exclusion of patients below 65 years of age by the CSHA. these  To compensate for this feature of the comparative data, analyses were  patients  recomputed solely with clinic caregivers of  over 64 years (48  of 70 patients).  Subsequent to this  revision, the distinctions between samples were markedly lessened (see  Table 4).  The remaining differences suggest that spouses  recruited within this facility appear to be caring for older patients who  are  more  impaired.  More  caregivers are no longer apparent. sample  derived  for  the  current  notable,  differences  between  These findings would suggest the study  may  be  representative  of  60  Table 4. Comparison of C S H A and Older Clinic Caregiver Samples  Feature  (N) *  Mean (SD)  T-Value  Caregiver Age: • CSHA (N = 113)  71.4  10.6  • Clinic (N = 48)  70.8  7.07  • C S H A (N = 114)  43.9  13.9  • Clinic (N = 48)  39.3  15.6  • CSHA (N = 102)  24.4  16.6  • Clinic (N = 47)  26.0  15.5  • CSHA (N = 114)  77.2  6.11  • Clinic (N = 48)  74.0  4.79  • CSHA (N = 106)  60.4  14.4  • Clinic (N = 46)  71.6  15.3  0.26  0.79  1.23  0.22  •0.40  0.69  2.45  0.02  •2.92  .005  Years Married:  Burden Interview:  Patient Age:  Modified Mini-Mental  *  State:  The sample size shown for CSHA subjects reflects the total derived on the basis  of selection criteria.  A smaller statistical N, however, was obtained subsequent to  application of corrective weights. These smaller numbers are not shown as they do not reflect the number of actual subjects upon whom descriptive statistics are based.  61  caregivers in this country.  Thus subsequent observations can be  generalized with greater confidence.  Preliminary  Analyses  Three scatter plots were computed comparing responses on caregiver burden, hopelessness and depression measures. SPSS  PLOT  coefficients  (SPSS  among  (see Table 5).  Inc.,  1988),  dependent  the  variables  significant was  Using  correlation  graphically  Also of note, no outliers were identified.  evident  This would  suggest that items were coded properly and that all cases can be included in subsequent univariate analyses. The Burden  distribution Interview  of  (Bl;  responses were Zarit  &  Zarit,  also 1990)  Hopelessness Scale (BHS; Beck et al., 1974). normal  distribution  skewness  (.502).  subscales  (personal  correlation  between  of  scores  Though  this  was  analyzed  evident  measure  and  the  with  the  no  Beck a  significant  two  reported  highly significant  subscales suggests a strong degree  dependence (r = .89, p < .001).  the  On the former,  contains  strain and role strain),  for  of  inter-  This coefficient is greater than that  reported by Hadjistavropoulos, Taylor, Tuokko and Beattie also obtained in this facility (r = .75, p < .001).  (1994),  For all subsequent  analyses, only the total Bl score will be utilized. The distribution of responses on the BHS indicates a degree of positive This  skewness  pattern  would  Hayslip et al. (1991)  more appear  pronounced  than  to support an  the  earlier  GDS  (1.550).  observation  by  regarding the sensitivity of the BHS among the  62  Table 5. Correlation Coefficients and Significance Levels Among  Variables  (N = 70)  Caregiver  Caregiver  Beck  Burden  Patient  Duration  Marlowe  Edmonds  Depression  Scale  Interview  Severity  Symptoms  -Crowne  Scale  0.64  0.55  0.21  0.11  -0.19  -0.31  .001  .001  0.08  0.39  0.12  0.01  1.0  0.43  0.23  0.29  -0.17  -0.27  .001  0.05  0.02  0.16  0.03  1.0  0.40  0.31  -0.09  -0.45  .001  0.01  0.45  .001  1.0  0.26  -0.05  -0.04  severity  0.03  0.70  0.75  Duration  1.0  -0.12  -0.29  0.35  0.02  1.0  0.24  Depression  Beck Scale  Burden Interview  Patient  Symptoms  Marlowe Crowne  Edmonds Scale  1.0  0.05  1.0  63  elderly.  Though  those who are relative  to  sample,  the  this  instrument  more severely  lower low  levels mean  of =  to  effectively  identify  hopeless, it may be less sensitive  this  (ji  appears  construct.  3.96)  and  Within  the  high  the  current  percentage  of  caregivers grouped at the lower end of the distribution of responses would appear to support this conclusion. Internal consistency for dependent measures was assessed by Cronbach's alpha.  As shown previously in Table 2, the alpha level  for the G D S among patients  and caregiver is identical  =  (3  .87).  Also of note, internal consistency appears satisfactory for the BHS (3 = .77), (3 = .73)  the Burden Interview (3 = .94),  as well as the  globality  and stability (3 = .76) subscales of the Attributional  Questionnaire (ASQ; Koch, 1985).  Of note, however, the internality  subscale appears to possess weak internal consistency (3 This  finding  supports  the  Style  limited  utility of  internality  =  within  .33). the  hopelessness model as proposed by Abramson et al. (1989). A comparison of response levels by gender was computed on the Burden Interview, GDS and BHS. significant  difference  was  evident  On the latter two measures, a  with female  higher on all measures (see Table 6). consensus  within  caregiving  research  caregivers scoring  This corresponds to (Miller  & Cafasso,  the  1992).  As noted in the DSM IV (APA, 1994), the incidence of depression is twice as frequent clear, however, a  reporting  1994).  among women as compared to men.  It is not  if this reflects an idiosyncrasy of gender or simply  difference  Factors  such  (Allen-Burge, as  Storandt,  socialization  among  Kinscherf & Rubin, this  older  cohort  64  Table 6. Comparison of Responses on Dependent Measures by Gender  Feature  (N)  Geriatric Depression • Women  Mean (SD)  T-Value  rj  7.3 (5.67)  2.80  0.01  2.35  0.02  Scale:  (N = 47)  • Men (N = 23)  3.8 (3.01)  Beck Hopelessness Scale: • Women (N = 47)  4.5 (3.42)  • Men (N = 23)  2.7 (1.85)  Burden  Interview:  • Women (N = 46)  29.1 (16.5)  • Men (N = 22)  22.3 (14.1)  *  1.65  0.10  Note, due to significant mean level differences on screening measures,  women composed all persons within the depressed groupings of the research matrix (Chi square = 10.15, p < .05).  65  (Friedan,  1993)  may  account  for  part  of  this  discrepancy  (Pearlin & Schooter, 1978). A further factor may be a distinction in the manner by which depressive  ideation  is tolerated.  In  research cited  by Linehan  (1993), men and women appear to have different coping strategies. When  clinically  distraction affective  depressed,  techniques  state  to  a  men  appear  whereas  greater  more  women  degree.  likely  to  ruminate  In  depression screening measures, therefore,  the  employ  upon  their  administration  the type of  of  statements  presented may be more consistent with the thoughts of women. Men, in contrast, may exclude these beliefs from current awareness and thus are  less likely to endorse items from these  measures  (Linehan, 1993). One further gender difference within the current sample is the disparity  between  reported  length of patient  symptoms.  Women  state their spouse has been symptomatic for 57 months (SD = 40) on average as compared to 48 months for male caregivers (SD = 25). As noted in the C S H A (CSHA Working Group, 1994b), the incidence of depression patient  symptoms.  moderately this  increases  depressive  Though  correlated  coefficient  might  levels  significantly  is  with  duration  increasing  of  severity  are  only  in the current sample (r = .26,  p <  .05),  suggest the partially  and  severity  gender  attributable  disparity to  this  in  reported  difference  in  duration of caregiving. Analyses are not reported comparing these measures by age group.  A spurious correlation is believed to exist between severity  66  of depressive symptoms and the age of caregivers given that the age of patients is more likely the cause of this relationship (O'Rourke et al., 1994). spouses.  In general, the age of patients is similar to that of their As the incidence of dementia increases exponentially with  age (CSHA Working Group, 1994a; Skoog et al., 1993), older persons are  more  likely  to  be  caring  for  a  demented  spouse.  The  chronological age of the caregiver is thus secondary to differences in depression levels. their  It is likely to be the corresponding age of  spouses and the greater  levels of illness among the  latter  which account for this relationship (O'Rourke et al., 1994). The G D S was orally administered to patients as well as their caregivers in a separate interview.  This data was collected by a  member of the clinic staff blind to the hypotheses of this study.  As  noted in Table 2, the mean for patients as compared to caregivers was slightly higher. (Appendix Ten  Cronbach's alpha was identical for both groups.  provides additional  data  regarding the  concurrent  validity of the G D S as determined within this study). One reason the G D S was administered to both patients and caregivers  was  to  examine  the  relationship  between  their  respective depression levels.  In this study, the correlation between  the  nonsignificant (r  two  groups is low and  =  .04,  p =  .74).  This finding is in contrast to an earlier studies by Dura et al. (1987) and  Teri  caregiver patients.  and  Truax  (1994)  depression was The  the  where  a  affective  nonsignificant coefficient  significant state  of  obtained  correlate  the  to  respective  by the  current  67  study  provides  no support  for the  salience  within caregivers' social environment  of depressive  stimuli  (cf. Strack & Coyne, 1983).  Hopelessness Depression Among Caregivers First To  Hypotheses assess  the  viability  of  hopelessness depression  as  a  distinct subtype, it was first necessary to ascertain whether or not caregivers ideation.  presented  as  depressed  with  and  without  hopeless  This condition was satisfied among the current  sample.  Of 70 subjects recruited for this study, 15 (or 21.4%) scored above cut-off  on the G D S .  The majority of this grouping (or  10 of  15)  scored below cut-off on the BHS (i.e., depressed but not hopeless). This  would  suggest  characterized This finding  by  the  ideation  of  hopelessness in only  depressed a  minority  caregivers of  is  instances.  is consistent with the Abramson et al. (1989)  model  which hypothesizes hopelessness depression to be specific subtype among affective  disorders.  Second Hypotheses Each subject was primary  hypotheses  inclusion  criteria  measures  (GDS  of  and >  11;  placed into one of four cells to test the this  study.  cut-off  scores  BHS > 9),  On  the  basis  on  the  two  subjects  were  of  initial  depression  categorized  as  hopeless and depressed (N = 5), depressed but not hopeless (N = 10), hopeless depressed  but (N  not =  depressed 54).  It  is  (N  =  1),  or  neither  noteworthy that the  hopeless percentage  nor of  68  caregivers than  classified within depressed groupings  that generally  reported  Cohen & Eisdorfer, 1988;  in the  literature (cf.  Coppel et al., 1985;  Haley et al., 1987; Rabins et al., 1982). that the  (21.4%)  is less  Barusch,  1988;  Dura et al.,  1987;  This may be due to the fact  majority of subjects were recruited  relatively  early  in the  disease process (i.e., 80% at the point of initial patient diagnosis). A series of planned orthogonal contrasts were computed to assess differences in attributional the  four  cells  (Lomax,  assumptions were  made  style among subjects in three of  1992).  As  discussed  previously,  regarding  levels of stability  or  no  globality  among caregivers presenting as hopeless but not depressed (H-D). This a priori method adjusts to differences in cell size yet  allows  for only three separate contrasts in this instance (i.e., one less than the number of groups). applied  to  the  It should be noted that a value of zero was  hopeless but  not  depressed cell  (H-D)  for  each  orthogonal contrast where only one caregiver was categorized. It was  hypothesized that stability  and globality  levels  would  be statistically higher for the hopeless and depressed (HD) group as compared to those caregivers who were depressed but not hopeless (D-H)  and those neither  hopeless nor depressed (-H-D) (i.e., cells  one vs. two and four respectively).  It was also hypothesized that  those depressed but not hopeless (D-H)  and those neither hopeless  nor depressed (-H-D) would not differ from each other vs. four respectively). to examine  the  two  These comparative analyses were computed  salience of hopelessness relative to  style among caregivers.  (cells  attributional  69  Table 7. Orthogonal Contrasts Comparing Caregivers' Attributional Style (N = 70)  Grouping (N)  Stability (SD)  Globality (SD)  (5)  53.0 (5.61)  49.2 (3.03)  42.8 (9.07)  D-H (10)  39.1 (8.81)  38.1 (8.46)  37.9 (7.94)  H-D  42.0  42.5  40.5  36.1 (9.06)  35.6 (9.02)  42.3 (6.23)  H D  (1)  -H-D (54)  Internality (SD)  T-Value  p  Contrast 1:  2.81  0.01  Contrast 2:  -4.08  .001  Contrast 3:  0.95  0.35  T-Value  p  Contrast 1:  2.29  0.03  Contrast 2:  -3.34  .002  Contrast 3:  0.80  0.43  STABILITY  GLOBALITY  INTERNALITY  T-Value  p  Contrast 1:  1.32  0.20  Contrast 2:  -0.18  0.86  Contrast 3:  -1.81  0.08  Contrast 1: H o p e l e s s and d e p r e s s e d (HD) v s . depressed, not hopeless (D-H) Contrast 2: H o p e l e s s a n d d e p r e s s e d (HD) v s . neither h o p e l e s s nor d e p r e s s e d (-H-D) Contrast 3: D e p r e s s e d , not h o p e l e s s (D-H) v s . neither h o p e l e s s nor d e p r e s s e d (-H-D)  70  As  shown  globality  and  depressed Also  as  stability  (HD)  and  7,  are those  predicted,  between those  in Table  no  significant  evident  statistical  between  neither  those  hopeless nor  difference  in  differences hopeless  and  depressed (-H-D).  attributional  style  appears  subjects who are depressed but not hopeless (D-H)  who  are  neither  hopeless  nor  in  depressed  and  (-H-D).  The  differences between those who are hopeless and depressed (HD) as compared  to  those  depressed  but  not  hopeless  (D-H)  is  also  significant. These findings allow for rejection each  instance.  hopelessness  This  therefore  distinguishes  of the  suggests that  differences  caregivers within this sample.  null hypotheses in  in  the  construct  attributional  style  of of  Though no conclusions regarding  causality can be drawn, it would appear there is a highly significant relationship  between  hopelessness  and  attributional  style  as  proposed by the hopelessness model (Abramson et al., 1989). A  second series  of  contrasts  were  computed  using  liberal cut-offs for both depression measures (GDS >10; By grouping caregivers either both  measures,  somewhat subjects As  shown  the  altered. within  more  BHS > 4).  above or below the top quartile  composition  of  the  respective  cells  on was  This allowed for a less unequal distribution of the  in Table  respective 8,  significant  groupings differences  groups as hypothesized by this study.  (9,  8,  are  8,  and  evident  45).  among  Globality and stability levels  remain significantly associated with the construct of hopelessness.  71  Table 8. Orthogonal  Contrasts  Comparing  Attributional  Style  Caregivers Grouped by Quartile Between Two Depression  Among Measures  (N = 70)  Grouping (N)  Stability  (SD)  Globality (SD)  Internality (SD)  HD  (9)  46.7  (8.76)  45.8  (7.92)  41.8  (9.44)  D-H  (8)  37.0  (11.5)  37.0  (8.28)  38.5  (5.63)  H-D  (8)  38.8  (6.76)  41.5  (8.02)  42.6  (6.30)  36.1 (9.30)  34.6  (8.88)  42.1 (6.44)  -H-D (45)  STABILITY  T-Value  p  Contrast 1:  2.35  0.04  Contrast 2:  -3.32  .003  Contrast 3:  0.26  0.80  T-Value  p  Contrast 1:  1.85  0.04  Contrast 2:  -3.14  .001  Contrast 3:  0.73  0.47  GLOBALITY  INTERNALITY  T-Value  p  Contrast 1:  0.77  0.32  Contrast 2:  0.35  0.91  Contrast 3:  -1.36  0.18  Contrast 1: Hopeless and depressed (HD) vs. depressed, not hopeless (D-H) Contrast 2: Hopeless and depressed (HD) vs. neither hopeless nor depressed (-H-D) Contrast 3: Depressed, not hopeless (D-H) vs. neither hopeless nor depressed (-H-D)  72  A  final  series of orthogonal contrasts were computed only with  spouses caring for a demented spouse (see Appendix Eleven). In  place  multivariate  of  assessing  globality  and  stability  separately,  analysis of variance (MANOVA) was used to examine  the combined influence of attributional  constructs.  All caregivers  were initially selected for this analysis, yet it was first necessary to  re-examine response levels.  no univariate  outliers  Though earlier analyses indicated  exist within  this  data  set,  it  was  necessary to further analyze responses on dependent measures prior to multivariate analyses (Tabachnick & Fidell, 1989). Multivariate  outliers  were  identified  using the  Mahalanobis'  distance of each case to the centroid of all cases. degrees  of freedom  (i.e.,  the  a chi square value of 7.81  normality,  of  residual  linearity  independent  variables)  This provided a revised sample size  Subsequent to the  examination  of  three  necessitated the removal of 10 cases  prior to subsequent analyses. of 60 cases.  number  With  plots  removal of these subjects, an suggests  and multicollinearity  the  assumptions  of  had been satisfied.  Due to unequal cell sizes, a correction to the M A N O V A was required (SPSS  to  appropriately  Inc.,  hierarchical  1988).  The  preference  dependent variable.  partition  the  UNIQUE  resulting  method  exists within the  sum of squares  was  theory  chosen as regarding  no  either  Globality and stability are assumed to arise at  the same point within the hopelessness model (Alloy et al., 1988). Due  to  the  higher  inter-correlation  between  globality  stability (r = .57, p < .001), a more stringent alpha (3)  and  level was  73  chosen  prior to  univariate  analyses (Tabachnick  & Fidell, 1989).  This was done to control for an inflated type one error rate. d = .017 still F  2  (i.e., .05/3  differ  p <  shown  among  .001  groups  respectively).  in  criterion  procedure  2  = 4.80,  5 7  p <  .015;  Consistent with previous ;  Table  9,  multivariate  significant differences among groups. Pillais'  (F  remains nonsignificant ( F 2 5 7 = 0.67, p = .52).  analyses, internality As  independent variables), globality and stability  significantly  = 8.69,  5 7  Where  is  reported  here.  analyses  also  reveal  In place of Wilk's Lambda, This  more  conservative  was chosen because of unequal cell sizes as this method  is most robust to violations of MANOVA assumptions (Tabachnick & Fidell, 1989). of  This procedure indicates the combined significance  attributional  style  exceeds the  .01  level.  (Appendix  Twelve  recomputes these analyses with balanced cell sizes). Approached  alternately  (i.e.,  reversal  of  independent  and  dependent variables), discriminant function analysis was computed to determine  the percentage of subjects accurately grouped within  the three cells of the research matrix.  This multivariate  technique  is methodologically more appropriate than MANOVA as discriminant function  analysis  assumes  unequal  totals  within  dichotomous  groupings (Tabachnick & Fidell, 1989). This analysis appropriately grouped 83.3% basis of attributional is  markedly  greater  style (50 of 60 caregivers). than  would  (chi square = 18.9, p < .005).  be  achieved  of subjects on the This percentage by  chance  alone  Figure 2 shows the centroids derived  from independent measures for the three caregiver groups.  74  Table 9. Multivariate Analysis of Variance  (MANOVA)  Attributional Constructs Among Caregivers  Grouping (N)  Comparing  Levels of  (N= 60)  Globality (SD)  Stability (SD)  Internality (SD)  H D  (4)  50.3 (2.22)  55.0 (3.92)  39.8 (6.90)  D-H  (6)  38.2 (10.4)  35.5 (6.95)  39.7 (7.55)  -H-D (50)  35,7 (9.20)  36.3 (9.09)  42.2 (6.15)  Univariate  Between  Group  F Tests  with 2,57 df  Variable  Hypoth. S S  Error MS  F Value  Significance  Stability  1320.93  76.04  8.69  0.001  Globality  792.07  82.52  4.80  0.01  53.78  Internality  Multivariate  Test Pillais'  criterion  40.06  Test  of  0.67  0.52  Significance  Value  Approx. F  dj  0.30  3.24  6.0  Significance .006  75  Figure 2. Centroids for Three Caregiver Groups Derived from Analyses  Discrimimant  (N = 60)  2.00  m HD  1.00 Second Discriminant Function _ Q  -D  a  Q Q  -1.00  -2.00 -1.00  -0.60  -0.20  0.20  0.60  1.00  First Discriminant Function  Grouping  (N)  Mean Depression (SD)  Mean H o p e l e s s n e s s (SD)  H D (4)  19.0 (4.97)  11.8 (3.40)  D-H  (6)  12.2 (1.47)  4.5 (2.07)  -H-D (50)  3.9 (2.95)  2.8 (1.82)  76  For  each  equation The  of the three groups, the computed  appropriately  resulting  group.  place  a significant  classification  majority  centroids show the distance which  of subjects.  separates  each  In this instance, a maximum of two discriminant functions  are available as degrees of freedom are limited by the number of groups (i.e., df = 3 groups - 1). basis  of  exclusion statistical  internality of  Thus subjects were grouped on the  and stability  internality  technique.  scores.  compromised  It is apparent  from  It  the  is  unlikely  accuracy  of  that this  previous analyses that  mean levels of internality differ minimally among groups. To this concordance style. al.  point, of  each statistical  hopeless  ideation  procedure with  has supported the  depressive  However, no causal conclusions can be drawn.  (1989)  assert  that  globality  and stability  lead  attributional Abramson et to a  distinct  subtype of depression defined by hopelessness among predisposed persons.  Yet the design of the current study does not allow this  causal relationship to be examined. For  instance,  it is also conceivable that hopeless  ideation  results among those most profoundly depressed or as a consequence of severe contextual stimuli (i.e., patient impairment caregiving).  Should this be the case, hopelessness may not define a  distinct depressive subtype but exist as a severity current  data  set does not indicate  onset of depressive symptomology al.  (1989)  or duration of  model)  or represents  marker.  The  if hopelessness precedes the  (as proposed by the Abramson et an outcome  caregivers as a consequence of extraneous factors.  for a  subset of  77  As shown in Table 10, it would appear those hopeless and depressed present with higher G D S levels as compared to depressed persons who score below cut-off on the BHS. differences between levels,  patient  groups are apparent.  severity  similar.  This  variables  distinguish  and  suggests  reported  few  Aside from depression  symptom  differences  hopeless  However, no other  and  duration  other  than  depressed  appear cognitive  caregivers  compared to the larger grouping of depressed persons.  as  However, the  chronology of hopeless ideation can only be adequately assessed by longitudinal  research.  Caregiver Burden as a Specific Outcome of Hopelessness Third Hypotheses It  was  initially  hypothesized  that  burden,  as  a  cognitive  construct, would be more strongly correlated with hopelessness as compared  to  other  depressive  globality and stability). Table  5,  this was  measures  (i.e.,  total depression,  Based upon the coefficients presented in  not apparent.  Though burden  is significantly  correlated with hopelessness among caregivers (r = .43, p < .001), the  coefficient  between  depression and  Burden  Interview scores  reflects a stronger degree of relationship (r = .55, p < .001).  In part,  this  scores.  It  is  may  be  due  noteworthy,  to  the  however,  correlated with burden scores p = .01)  positive that as  skewness  of  hopelessness is compared  and stability (r = .31, p = .01).  therefore accept the null hypothesis.  to  BHS more  globality  strongly (r = .30,  In this instance, one must  78  Table 10. Comparison of Features Between Depressed Groupings  Measure  Mean (SD)  T-Value  Geriatric Depression Scale: • HD (N = 5)  17.8  5.07  • D-H (N = 10)  13.1  2.51  • HD (N = 5)  41.6  18.7  • D-H (N = 9)  45.8  16.0  • HD (N = 5)  23.6  5.32  • D-H (N = 10)  25.0  5.25  • HD (N = 5)  72.0  47.2  • D-H (N = 10)  57.3  46.6  • HD (N = 5)  71.0  7.94  • D-H (N = 9)  67.3  10.3  • HD (N = 5)  63.8  11.1  • D-H (N = 9)  60.7  13.2  2.45  0.03  •0.44  0.67  •0.48  0.64  0.57  0.58  0.69  0.50  0.45  0.66  Burden Interview:  Patient Severity (FRS):  Duration of Symptoms:  Patient Age:  Caregiver Age:  79  Subsequent analyses were undertaken burden  levels within the  depression.  paradigm  of the  to more fully  assess  hopelessness model of  This was based on the assumption that burden might  exist as a specific manifestation of hopelessness depression among spousal caregivers. Of note,  Burden Interview scores are significantly  with the Edmonds Scale (r = -.45, p < .001) dependent  variables  (see  significant  relationship  Interview  and  the  Table  exists  10).  between  Marlowe-Crowne  (r = -.09, p = .45).  It  correlated  as compared to other is noteworthy  that  responses on the Social  no  Burden  Desirability  Scale  In a separate study, O'Rourke et al. (1994) have  proposed the existence of a distinct dyadic response bias which is related strongly to responses on relation-based the Burden Interview.  measures such as  (Appendix Thirteen presents analyses of the  factor structure of the EMCS). The  significant  Interview  scores  procedures.  relationship  was  explored  between using  EMCS  and  hierarchical  Burden  statistical  Within a regression equation, E M C S scores were first  entered to control for the influence of response biases. A  regression  model  was  constructed  to  reflect  the  hypothesized chronology of the hopelessness model of depression (Abramson et al., 1989).  Subsequent to inclusion of E M C S scores  (to function as a covariate), a second block was entered including an objective  measure  duration  of  of patient's  symptoms.  This  functional status (FRS scores) and grouping  represents  the  objective  indices of caregiver burden and corresponds to concrete strain and  80  the demands of this role as operationalized within this diathesisstress model. The final block of variables corresponds to subjective factors hypothesized to arise as a function of the stress of caregiving. grouping,  composed  represents  the  of  stability,  constrained  globality  cognitive  set  and  This  hopelessness,  hypothesized  to  arise  among predisposed persons. Regression  analyses  were  performed  by  using  SPSSx  REGRESSION and SPSSx FREQUENCIES for evaluation of assumptions (SPSS scatter  Inc.,  1988).  plots  distributed  were  about  To test for the produced. predicted  As  scores,  independence of residuals, residuals this  were  normally  suggests  normality,  linearity, homoscedasticity and independence of residuals.  Because  of the extended period over which interviews occurred, the DurbinWatson  test  statistic  was  calculated  as  a  autocorrelation of errors over the sequence of cases. was  calculated  to  be  1.78  (not  significant  at  indicating no autocorrelation between adjacent cases.  measure  of  The value of d the  .01  level),  Though these  procedures did not result in the exclusion of further cases, it should be  noted  the  According  to  ratio  of  independent  Tabachnick and  variables  Fidell  (1989),  each  block  to  cases  however,  is  low.  this  ratio  exceeds a required minimum. As  shown  contributes  to  a  in  Table  10,  highly significant  of  variables  regression equation  Burden Interview scores (R = .73, p < .0001).  strongly predicting  After adjustment to  correct for the dyadic response bias (i.e., possible underreporting of  CO CO CD  CO  CO  "M  1  CM  CNJ  LO CM  LO  CO  <o|  I  CO CO CD CO  C2I  Q  CT> CO  00 O  CO  LO O  O  i -  i -  CO CO LO  TJ CD N CO CD -C  mi  »  LO "3-  CO O  CO O)  CO O  CM  CO  cr  T -  Q  > X  > CC c  CO  c  CD .Q o  CO  <  o 'co  CO •<3-  CO  o  CO  CD v_ D CD  V  CO  00 LO  CO  c*y  0 >  CO  LO  CO  i -  i -  co  "CO  LO  Jc  1  —  OS CM  1  V CL  Q  CD  1  TJ  CO  c  CO  o  CO  LU  o  CM  •tf o  CM 00  ZZl  r-— r-— W  W  CD  CD  > a  M —  CD  CO  co  i -  LO  •riCD TJ  i -  CO  o  "fr  o d i -  i -  O  n  1—  co  c  CD TJ L i  CD  CO  V  CO  ^  m to w  o  -»—»  »8  CO CM  CO  1—  o o  o q  c o  0 CO  1 .  Q_  CM  o  _C0 CD  o o  CO  DC  o  cc cr p  .a  c  CM  <  CO CO  o o  CO  T3 CD  o  c  00  CD c  w a> .o ro  CO  >  « a  CD >  co  ro  w w  .Q o  .Q  O  CO  ro  CD CD CL 0  1  CL  82  Bl  scores),  contribute  to  contribution model.  both the of  This  subjective  R  and  value.  It  each  grouping  would  indicate  factors  factors alone.  objective  exceeds  subjective is  noteworthy  is similar that  that  factors  the  that  within  this  relative  provided  As shared variance between  by  significantly the  relative  hierarchical  contribution  of  illness-mediated  groupings would be  claimed by the first block (i.e., objective variables), this suggests the subsequent contribution of attributional separate.  factors is distinct and  More precisely, cognitive factors included as the  final  grouping in the model provide a distinct and significant contribution to the prediction of burden scores. to  which  the  construct of  This would indicate the degree  burden  is  largely  influenced  specific  factors  by  the  perceptions of caregivers. The noteworthy.  relative  significance  of  is  also  As compared to severity ratings, the beta value for  duration of symptoms contributes much less to significance of the R square value.  It is likely there is a high degree of shared variance  between this latter variable and E M C S scores as this information is largely obtained from collateral informants (O'Rourke et al., 1994). Thus, within this hierarchical model, it is assumed that duration of symptoms contributes little to explained variance in burden scores above that initially provided by the E M C S . Consistent with previous analyses, the of stability  totals  relative  contribution  surpassed that provided by globality.  instance, however, stability emerged with a negative beta The  reason for this is not immediately  In  this  weight.  apparent as stability  was  83  positively correlated with Burden Interview scores (r = .31, p < .01). It  would  thus  appear  that  the  relative  contribution  of  stability  becomes inverse subsequent to the inclusion of one or more of the variables first entered into the regression equation. Most  notable  is  the  strong  and  unique  contribution  hopelessness relative to burden scores (B = .41). strongly  suggestive  ideation  relative  to  of  the  significant  caregiver  This finding is  contribution  burden.  This  of  of  depressive  relationship  also  supports the contention that caregiver burden may be a specific example  of  indicating  hopelessness depression. the  attributional construct  of  significant  style,  this  As with earlier  relation  between  hierarchical  model  hopelessness is  highly  analyses  hopelessness would  significant  suggest in  relation  expressed burden within this sample of spousal caregivers.  and the to  84  CHAPTER 5 - DISCUSSION  It is hopelessness even more than pain that crushes the soul. as  So the decision-making of daily life involves not, in  normal  affairs,  shifting  from  one  annoying  situation to another less annoying - or from discomfort to relative comfort, or from boredom to activity moving from pain to pain. briefly,  one's  bed  of  - but  One does not abandon, even  nails,  but  is  attached  to  it  wherever one goes (Styron, 1990, p. 62).  Hopelessness Depression among Caregivers The  results  of  this  study  strongly  support  aspects  of  the  hopelessness model of depression (Abramson et al., 1989).  The  construct  with  of  attributional  hopelessness thought  appears  patterns  to  linked  be  to  associated  the  onset  of  this  hypothesized depressive subtype among spousal caregivers. conclusion  is  supported  by  each  of  the  statistical  This  analyses  conducted. Initial orthogonal  contrasts assessed differences  in  globality  and stability levels among groupings divided on the basis of scores on both depression measures. construct  of  hopelessness  attributional style.  These analyses indicated that the appears  strongly  associated  with  This was not only evident among those who are  hopeless and depressed as compared to the non-depressed cohort, but  also within the  precisely,  the  full grouping  construct  of  of  depressed  hopelessness  persons.  would  More  appear  to  85  effectively  distinguish  depressed  persons.  patterns This  of  attributional  supports  the  notion  thought  among  of hopelessness  depression as a distinct subtype within a heterogeneous grouping of affective  disorders.  In contrast to the helplessness model (Abramson et al., which  implied  attributional findings  that all  style  from  unipolar  depressions were  current  study  support  Abramson et al. revised model (1989). roughly 33%  precipitated  (i.e., implied causality due to cognitive  the  the  1978) by  factors),  hypotheses  of  the  Among the current sample,  of depressed subjects appear to be classified within  the hopeless and depressed grouping. It is noteworthy the  proportion  pharmacotherapy  of  that this percentage  depressed  (Brown,  Sweeney,  1983; Georgotas et al., 1987). percentage  of  elders  depression  roughly corresponds to  who  do  not  respond  Frances, Kocsis  &  this  applicable entail  a  percentage  cases  to other substantial  of  may  not  hopeless and  populations number  Loutsch,  This might suggest that a consistent be  attributable  biological factors (i.e., possible etiology due to cognitive Should  to  depressed  considering the  wide  factors).  persons  (elderly or otherwise),  to  this  be  would  prevalence  of  depressive subtype  is  unipolar depression (APA, 1994). Support for this consistent model.  with  model as a distinct  equivocal  findings  for  the  earlier  helplessness  As discussed previously, studies investigating  attributional  style relative to the more global helplessness model often at conflicting conclusions (Parks & Hollon, 1988).  arrived  Some research  86  suggested differences depressed  persons  in attributional relative  to  Abramson, 1983; Cutrona, 1983). showed  no such differences  style were apparent  control  subjects  between  (Hamilton  &  The majority of studies, however,  (eg.,  Miller,  Klee  & Norman,  1982;  O'Hara, Neunaber & Zekoski, 1984). This may be due to sample size as well as the composition of previous study samples.  For instance, recruiting  cognitive therapy clinics (cf. Beck et al., 1979)  subjects within  may have lead to an  over-representation  of persons presenting as hopeless as compared  to those recruited  in other settings (i.e., those whose depression  was  not  amenable  to this  treatment elsewhere). patients  presenting  intervention  likely  would  have  sought  As a function of the relative percentage of  as hopeless, statistical  significance may  have  emerged.  In previous research, a significant association  attributional  style and depression may have been evident depending  on  the  percentage  and  respective samples.  size  of  hopeless  between  persons within  these  This point underscores the difficulty that can  arise when subjects are recruited within a single facility. Another  finding  from  the  current  study which  supports  the  hopelessness model is the lack of association between hopelessness and  perceived  (i.e., internality).  personal  responsibility  for  negative  events  This, again, is in contrast to the helplessness  model which hypothesized this construct to be integral to the onset of depression (Abramson et al., 1978). It would appear hopeless subjects in this sample see negative events  as  equally  attributable  to  themselves  and  others  or  87  circumstance. construct  On reflection, this would seem congruous with the  of  hopelessness  responsibility for negative  as  any  consistency  in  events would suggest the  forestall or prepare for these happenings.  perceived  possibility to  By definition,  it would  seem that hopelessness entails the perception of utter randomness in the occurrence of negative life events; the only consistency being that unpleasant events occur repeatedly,  with little or no  warning  and impact most domains of one's life. This  would  appear  consistent  hopelessness as a core construct. notion  a  control  In  (Rotter,  positive  of  views  entails  the  the  possibility  1966),  events,  the  and awareness could be applied to future  keeping  personal empowerment Because  that  For instance, if one claims to be responsible for physical  illness, this knowledge behaviour.  model  No matter how negative,  of personal responsibility for events  of change.  and  with  with the perceived  in theory,  construct  of  responsibility leads  internal for  locus of  both  to a sense of  negative  agency  or  incongruous with pervasive hopelessness.  incompatible  nature  of these  two  constructs, if  internality and hopelessness were to co-exist, cognitive dissonance would likely ensue (Cooper & Fazio, 1984).  One or both would have  to be addressed and redressed. What  is  apparent,  however,  internality appears to differ between not hopeless (D-H)  is  that  this  construct  of  those who are depressed but  as compared to those who are neither hopeless  nor depressed (-H-D).  A significant  difference  exists along  this  continuum among the former group caring for a demented spouse.  88  Yet, what is noteworthy  is that these depressed subjects appear to  see negative events as due to others or circumstance as opposed to themselves in contrast to the prediction of the helplessness model. This finding from the current study would suggest Abramson et al. (1989) are  correct to remove  internality from the  revised  (The low Cronbach's alpha for internality (3 = 0.33) globality (d = 0.73) lacks  and stability (3 = 0.76)  homogeneity  hopeless  within  persons, but  subjects,  the  the  also  construct  of  current  among  as compared to  indicates this construct  sample).  the  internality  model.  Not  remainder appears  only  of to  for  depressed have  little  relevance. Support univariate of  basis  of  grouped. this  hopelessness  (MANOVA),  and  significantly function  the  model  and multivariate analyses.  variance  stability  for  internality among  groups  (p  analysis also revealed attributional  In the  simultaneous scores  style,  provided  .001).  the  of  first  Finally,  globality, two  of  subjects  were  differ  discriminant  multivariate significance.  83%  On  effectively  grouped  subjects  as  the  appropriately  This result was again strongly significant (p < .005)  procedure  by  multivariate analysis  examination  indicated <  was  hopeless  as and  depressed (HD), depressed not hopeless (D-H), or neither hopeless nor depressed (-H-D) as a function of globality and stability levels. To a certain degree, the strength of these results challenges the proposed chronology of the hopelessness model. Figure 1 (p. 25), precede the  As shown in  Alloy et al. (1988) suggest three discrete stages  onset of hopelessness depression.  Salient  negative  89  events  (step  1)  are  believed  to  trigger  depressive  attributional  style among predisposed persons (step 2) which leads to perceived hopelessness (step 3) and eventually a depressive condition defined by hopeless ideation  (step 4).  The temporal  distinction  between  each of these stages would suggest persons can be identified each point along this continuum.  at  Findings from the current study,  however, do not support this notion. For example, only one subject in 70 recruited for this study presented as hopeless but not depressed (H-D). this  caregiver  was  taking  anti-depressant  included due to this elevated discussed in Chapter 3).  In this instance,  medication  and  hopelessness level (inclusion  was  criteria  It is conceivable that this medication may  have impacted her depressive state without a corresponding effect upon hopeless ideation.  The absence of persons presenting  hopeless but not depressed (H-D) pre-depressive model.  suggests it is unclear  hopelessness exists  as a separate  stage  Among predisposed persons, it is therefore  as  whether in  this  plausible that  the onset of depression is concurrent with perceived hopelessness. Also  significant  is  the  marked  difference  in  attributional  style between hopeless and depressed subjects (HD) as compared to those who are neither hopeless nor depressed (-H-D).  According to  the proposed chronology of the hopelessness model, it is likely that a proportion of persons in the latter group may eventually succumb to hopelessness depression.  In other words, the recency of onset in  their  may  spouses'  condition  have  triggered  depressogenic  90  attributional  style  though  hopelessness  has  yet  to  result  (i.e., between points two and three in the hopelessness model). According to this logic, the  neither  hopeless nor depressed  grouping (-H-D) should be composed of those who are asymptomatic (i.e.,  the  majority  who  will remain  at  this  point)  as well  as  a  percentage of persons at early onset for hopelessness depression. The marked degree of difference in attributional  style between this  group and those who are hopeless and depressed (HD), challenges the hypothesized heterogeneity  however,  of the neither hopeless  nor depressed (-H-D) grouping of caregivers.  For levels of globality  (t = 3.34,  p < .01) and stability (t = 4.08, p < .001), the difference  between  groupings  reflected  in the  Table  p. 70).  7,  hopeless  and  is  mean  highly  and variance  depressed  more  and  deviations (-H-D). little  mean (HD)  stability  is  that  is (see  level for  the  three  above the corresponding mean level for  globality  depressed  above  difference  approximately  hopeless nor depressed (-H-D).  pronounced for  hopeless  grouping  This  levels for both groups  For instance, the  standard deviations (SD) those neither  significant.  group  of those  levels (HD) neither  where is  This difference the  roughly  mean four  for  is the  standard  hopeless nor depressed  Assuming normal distribution of scores, this suggests very  overlap  between  their  respective  distributions.  words, these appear to be distinct populations.  In  other  This observation  does not appear to support the hypothesized heterogeneity  of the  neither hopeless nor depressed grouping (-H-D) as implied by the hopelessness model (Alloy et al., 1988).  91  This  would  either caregivers the  stress  of  suggest recruited  their  one  of  two  possible  for this study have  spouses'  condition  explanations:  been dealing with  long  enough  that  all  predisposed to hopelessness depression have progressed through all stages; or the steps leading to this condition are less discrete than Alloy et al. (1988) have proposed. distinction  between  theoretical  than temporally  Yet, these  with all  each  the  points  can  research.  in  the  model  is  more  autonomous points along a continuum.  discussion regarding  observations  longitudinal  of  For instance, it is possible the  only  the  be  fully  chronology of this addressed  by  model,  means  of  The methodology of this study only allows  observations to be made and hypotheses to be advanced.  Caregiver Burden as a Specific Outcome of Hopelessness This study has also provided preliminary and  distinct  conceptualization  of  caregiver  support for a novel burden.  Though  hopelessness does not appear more strongly correlated with burden as compared to overall  depression, subsequent analyses suggest  hopelessness is a salient among caregivers.  construct relative to expressed  The hierarchical regression model derived in this  study accounts for 53% of observed variance scores (p < .001). provided  by  burden  social  in Burden Interview  In addition to a significant inverse desirability,  both  objective  and  contribution subjective  groupings of variables significantly contributed to the prediction of burden scores (i.e., patient factors and depressive thought patterns).  92  This approach to the assessment of caregiver burden differs substantively from previous research. and  coping  models  traditionally  For instance, applied stress  view  burden  as  the  difference  between role demands and perceived resources (Pearlin et al., 1990; Zarit,  1990).  Though the latter suggests negative perceptions are  attributable to subjective factors, this need not always be the case. More precisely, a belief one is ill equipped to face the demands of caregiving may not be unrealistic. a  lack  of  instrumental  For instance, economic factors,  support, or caregiver  illness may  impede one's ability to cope and may thus be an entirely perception.  Similar  to  (Alloy & Abramson,  1 9 8 8 ) ,  life  be  situation  may  research  regarding  greatly objective  depressive  realism  negative beliefs concerning one's current  entirely  objective.  In  certain  instances,  depression is not a function of distorted perceptions, but an acute and  rational  contextual  awareness  of  one's  resources  relative  to  existing  stressors.  In contrast, the model proposed by this study documents the operation  of  a  distinct  subjective  component.  gauging perceptions specific to caregiving, the  As  opposed to  cognitive  patterns  assessed in this instance are separate from the caregiving context (i.e.,  globality,  respond within  to  stability  general  hypothetical  caregiver operation attempts  role  they  of this to  and  statements situations perform  restricted  separate  hopelessness). or is to  assessing  sufficiently more  cognitive  objective  Asking subjects  In  related  perceptions  distinct  effectively  set.  factors  their  effect, to  to  from  the  suggest  the  this  role  model  demands  93  from  the  constrained  thought  processes which  uniquely  and  significantly contribute to burden levels. A  further  distinction  exists  stress and coping models. (1984)  makes  between model,  no  between  assumptions  carries  and  traditional  The Lazarus and Folkman paradigm regarding  stressors and subjective however,  this  the  a  appraisal.  implicit  causal association The hopelessness  assumption that  subjective  factors are triggered by contextual stressors. This conceptualization of burden is not dissimilar from Beck's hypotheses suicidal  regarding the  ideation  (Beck  relationship between  & Weishaar,  1989).  hopelessness and Among depressed  persons, it is believed that a narrow perception of options allows suicide to be seen as a more viable alternative.  Any solution which  promises  the  to  provide  a  permanent  solution to  overwhelming  perception of current difficulties thus becomes more appealing. In a similar manner, burden may arise among caregivers as a result  of  hopelessness  cognitive set.  It  which  entails  this  same  constricted  is hypothesized that hopelessness limits one's  willingness to search for more effective coping strategies because of a pervasive belief that all attempts to manage the situation will be futile. of this  Thus, the caregiver resigns him or herself to the demands role  without attempting to cope more  effectively.  Even  when presented with new alternatives, pervasive hopelessness may cause caregivers to reject suggestions out of hand.  In this way,  hopelessness is believed to foster the sense that one is imprisoned  94  within  a  role  in  which  there  is  no  possibility  to  cope  more  effectively.  Generalizability  of Findings  Comparisons  between  the  current  sample  and  subjects  recruited in the Canadian Study of Health and Aging (CSHA) suggest caregivers  interviewed  in  population  from  they  which  clinic  sample  years  (i.e., to reflect  between  was  this  limited the  clinic  were  to  are  drawn.  those  Particularly  caring  protocol of the  groups were evident.  representative  for  patients  CSHA), few  of  the  after  the  over  differences  Though subjects recruited for the  current study appear to be caring for older patients who are cognitively  impaired,  these  64  caregivers  appear  similar  more  to  this  randomly derived, national sample. It must be noted, however, that the analyses comparing the respective samples are relatively article  by  between  Feingold  (1995),  groups lies in the  basic.  much tails  of  As discussed in a recent the  of their  bona  fide  respective  distributions.  Though measures of central tendency may indicate few differences, the  richness or unique character  lost by focusing on crude measures  in the  difference  significant  of samples may search for  be  statistical  significance. It would appear this discussion pertains large sample comparisons.  more specifically to  The relatively small number of subjects  recruited for this study and the C S H A comparative sample precludes analyses  in  this  detail  (i.e.,  few  cases  beyond  two  standard  95  deviations  of  the  mean).  The  point  is  well  taken,  however,  concerning the conclusions which can be drawn from a series of t-tests.  Though the  mean  and  variance  levels  appear  similar  between the current and CSHA samples, it can only be said that they appear similar and not identical.  Limitations of Study As noted previously, no conclusions can be drawn the and  causal link  hypothesized to exist  hopelessness.  Though the  between  model  regarding  attributional  suggests globality  style and  stability precede and lead to perceived hopelessness, the design of the present study has not permitted this to be addressed.  This  would require a longitudinal design. One alternative with the current data set might have been to conduct cross-sectional analyses. and  caregivers  by  initial  For instance, grouping patients  assessment  versus  subsequent  assessments may have permitted discrete comparisons. previously, one difficulty with this tactic is the of  persons  who  had  been  seen  small  previously  in  As noted percentage this  clinic  (i.e., reassessments provided only 14 cases or 20% of the current sample). A more serious factor precluding the use of cross-sectional analyses relates to the duration of the illness.  Initial assessments  within this facility do not necessarily occur early process.  in the disease  For instance, some patients first referred to this clinic  have been symptomatic for many years.  In other cases (particularly  96  where  the  relatively  patient  is younger),  the  onset of symptoms may  be  recent.  Another alternative would have been to group subjects on the basis  of duration  versus  of their  spouses'  subsequent assessment.  illness irrespective  However,  reported  of  initial  duration  of  symptoms itself is a suspect variable.  This information is largely  obtained  is  from  caregivers.  Not  only  correlated with dyadic social desirability  this  variable  (O'Rourke  strongly  et al.,  1994),  but various circumstances will make symptoms more apparent for some patients as compared to others.  For instance, an employed  person with young children will appear symptomatic sooner in the course of the illness as compared to a retired octogenarian with a consistent  routine  and  considerable variability patients  (Tuokko,  few  responsibilities.  exists in the  Also  of  rate of decline of  Vernon-Wilkinson,  Weir  &  note,  dementia  Beattie,  1991).  Because one patient has been symptomatic for the same number of years as another,  it cannot be automatically concluded that their  level of impairment  is similar.  A more pronounced limitation of the current study is the low number of subjects upon which conclusions are based.  As originally  proposed, it was anticipated that subjects would have been more evenly distributed along the hopeless and depressed continua.  Over  ten months, however, only five of 70 subjects were assessed as hopeless and depressed (HD). standpoint (i.e., the  relative  This is encouraging from a clinical mental  health  of caregivers  entering  this facility), yet it tempers the enthusiasm with which the results  97  of this study can be expressed.  One means to address this would  have been the use of statistical power analysis (Cohen, 1988). technique errors.  allows  for  a determination  of the  risk of Type  This II  (8)  In other words, power analysis provides a risk estimate of  accepting invalid hypotheses. This  discussion  research testing longitudinal outset.  underscores  a  fundamental  diathesis-stress models.  difficulty  Investigators  designs are faced with an important  in  not using  choice from  the  As it is necessary to indicate the operation of stressors as  well as the clinical condition to suggest predisposition, this can be approached in two distinct manners.  Heterogeneous samples can be  derived in which it can be documented that a salient negative has  occurred,  exclusively  or  of  researchers  the  target  can  recruit  population  and  subjects on the basis of past stressors. the  current  study  whereas  most  samples  attempt  event  composed  to  distinguish  The former was chosen for  research  has  approached  this  retrospectively. Certainly the former technique is not without problems. methodology groupings  is more  of  labour  subjects  procedures.  Yet  empirically  stronger  intensive,  and  it can  restricts  also  than  results  be  the  argued  alternate  in  disproportionate  choice that  of  this  quantitative  This  statistical  approach  is  methodologies.  As noted by Alloy et al. (1988), a primary short-coming of previous research stressors. events  in this area More  is a function  precisely,  were perceived  it  in the  of the  is unclear past.  It  salience of how  identified  importantly  must also be  these  noted  the  98  subjectivity (1985)  inherent  sought  unsolvable  to  in these indicate  anagrams  attributional  was  perceptions.  that  failure  sufficient  style among older adults.  For  to  instance,  solve  to  a  trigger  Koch  series  depressive  As discussed earlier,  results of this study were far from conclusive.  of  the  It could be argued  that not all elders would view this experience as overly important. This majority  suggests  studies  must  employ  stressors which  of persons would perceive as salient.  It is quite  that this requirement has been met by the current study. experimental accomplished.  perspective, however, Few  ethical  review  the likely  From an  it is uncertain  this  could be  committees  would  approve  research with proposed to deliver this type of stressor to subjects (with  or  without  informed  consent).  This,  therefore,  limits  researchers to recruiting subjects where these negative life events occur naturally or have already occurred (as in the present study). From the  outset, a selection bias results after which point it is  unknown how representative the sample remains as compared to the overall population.  Implications for Counselling and Psychotherapy This  study has provided the  first  indication that a distinct  depressive subtype might exist within this older adult population. As noted previously, roughly 33% of depressed persons from the current sample are distinguished by the construct of hopelessness and  corresponding  depressive  attributional  thought  processes.  Should subsequent research indicate that hopelessness effectively  99  distinguishes populations of depressed persons, this has significant implications for the practice of psychotherapy. It has long been recognized that cognitive interventions highly effective Miller,  in the treatment of depression (Hollon et al.,  Norman,  & Keitner,  1989;  Shapiro & Shapiro,  are 1992;  1982).  In  recent years, this assertion has also been supported in research specifically 1990;  with older  adult  populations  (Okum,  Olding &  Cohn,  Teri, Curtis, Gallagher-Thompson & Thompson, 1991).  was further Steffen  evident  (1994),  in a recent study by Gallagher-Thompson and  where  the  depressed family caregivers.  sample  interventions  superiority,  research  recent  was  composed  entirely  of  In contrast to the initial enthusiasm  for cognitive-behavioural more  This  based on their  comparative  has challenged this conclusion  (Gallagher-Thompson, Hanley-Peterson & Thompson, 1990; Scogin & McElreath, 1994). outcome  research  populations.  In part, this may be a function of conducting and  meta-analyses  within  heterogeneous  For instance, should depression exist as a grouping of  disorders as suggested by Abramson et al. (1988), it is more likely that the relative efficacy of any intervention will be masked. However,  should hopelessness depression exist as a distinct  depressive subtype, it is highly likely that this disorder would most amenable  to cognitive-behavioural  therapy.  be  As opposed to  mood disorders defined by their organicity (Gilley, 1993), a specific depressive subtype caused and characterized by depressive ideation is  more  likely  to  respond  to  an  intervention  challenges these depressive thought processes.  that  specifically  100  The results of the current study allow this to be extended to a discussion of burden among caregivers. been documented for the distinct  manifestation  population. significantly  As  of  to  support  has  hypothesis that burden may exist as a hopelessness  hypothesized,  related  Preliminary  burden  the  depression  construct  within this  of  within  this  hopelessness is  sample  of  caregivers.  This would suggest that a means to redress this phenomenon among caregivers might be to examine the beliefs which underpin hopeless ideation. Finally, it may also be possible to identify  persons at  risk.  Prior to the onset of hopelessness depression, those who display the  depressogenic diathesis  Within the framework more  effectively  might  be  of preventative  directed  to  identified  as  susceptible.  medicine, resources might be  caregivers  at  risk,  waiting for the onset of clinical depression.  as  opposed  to  For instance, at the  time of initial diagnosis, those with a history of mood disorders or those who appear vulnerable to depressive cognitions might receive greater assistance to aid the transition to the caregiver role.  Future Research It further  is hoped  results  study  the  populations.  of  from  the  present  hopelessness  study  model  will  within  encourage  older  adult  Though the current study suggests this hypothesized  depressive subtype exists among spousal caregivers, it has yet to be  determined  if  these  results  (or other) elderly populations.  can  be  replicated  within  this  For instance, it is yet uncertain if  101  similar  findings  patient  groups or those encountering other  events.  would  appear  among  persons caring  for  other  serious negative  life  Where possible, it would be ideal for future investigations  to employ longitudinal methodologies. This option now exists as an extension of the current study. Persons interviewed  in this clinic provide a representative sample  of caregivers which could be followed interesting  to  assess  the  relationship  over time. between  It  would  attributional  be  style  and hopelessness at separate points relative to patient factors. Patients  and  caregivers attending  this  clinic are  routinely  asked to return roughly 18 months following their first assessment. Though not all respond to this invitation, a percentage of families are tracked through the course of the illness. opportunity  to assess  the  natural  This provides an  history of their  through their careers as caregivers.  affective  state  As noted by Barusch (1988),  67% of caregivers report depressive symptoms at some point during the  course of their  would  allow  for  a  spouses' more  illness.  complete  A longitudinal and  thorough  investigation test  of  the  hopelessness model within this population. This also pertains to the study of caregiver burden.  Findings  from the current study would appear to further clarify the nature of this construct among spousal caregivers.  Operationalized within  the hopelessness model, caregiver burden is more fully defined as a cognitive construct superseding objective role demands.  It has yet  to be determined, however, if this finding will emerge within other  1 02  caregiver populations. within this  Also to be examined is the etiology of burden  framework.  Though  this  study  has  assumed burden  to  be  a specific  outcome of the hopelessness model, further research is necessary to clarify the chronology of this hypothesized causal relationship. It  has  been  manifestation However,  assumed of  that  caregiver  burden  is  a  specific  hopelessness depression within this population.  longitudinal  analyses will be  required to fully address  this assumption.  It is yet unclear if this restricted cognitive set is  triggered  spouse's  by the  illness, if  hopelessness precedes  the  onset of burden, and what other variables may be related to this phenomenon. A  further  placement  caregiver  George, ideation  of  decision.  institutionalize of  area  possible  As  noted  investigation previously,  the  a demented family member generally variables  1986).  versus those  Consistent  with  is believed to result in the  range of possible alternatives. institutional  relates  placement  of  of  the  earlier  patient  to  decision  to  is a function (Colerick  discussion,  &  hopeless  perception of a constrained  Hopelessness may lead to  one's  the  relative  relentless demands of the caregiving role.  as  a  solution  to  early the  Perceived hopelessness  may lead caregivers to seize upon any permanent solution which promises to resolve one's inability to cope. This  would  suggest  that  caregivers might forestall placement. hopelessness  and  the  placement  effective  intervention  with  Should a causal link between decision  be  established,  103  addressing this  perception  might  allow  caregivers  to function  longer periods.  This outcome would have tremendous  for  implications  for families and the health care delivery system. This  approach  is distinct from  previous  investigations.  As  opposed to tailoring services to better suit the needs of caregivers (O'Rourke,  Tuokko,  Klassen-Peters,  Rae  &  Beattie,  1994),  the  perception of hopelessness may create a bias where all additional support is viewed as futile.  According to this logic, it may first be  necessary to address this belief system so that additional is viewed  support  objectively.  A final direction for future research relates to the of dyadic social desirability.  construct  As noted previously, 25% of observed  variance in Burden Interview scores was accounted for by the E M C S . This  surpasses that  provided  by  both  objective  groupings within the hierarchical regression model. for the E M C S (3 = -.39)  and  subjective  The beta value  is surpassed only by that provided by the  hopelessness measure (B = .41).  As opposed to the assumption  reflected in the existing research literature, reported burden among this cohort of spousal caregivers appears highly susceptible to this dyadic response bias.  It has yet to be determined, however, if this  relationship would persist if burden were assessed in such a way as to  conceal  the  identity  of  caregivers.  This  significant  inverse  relation between the EMCS and burden scores must also be assessed among persons caring for other patient populations. This relational  study  provides  the  first  indication  that  response bias may exist distinctly from  a  dyadic  individual  or  forms  1 04  of  social  desirability  however, to determine measures. is  responding.  Further  study  is  required,  its impact upon reporting of other dyadic  Again, it has yet to be determined if the current finding  specific  to  expressed  burden  or  will  generalize  to  other  more  fully  emotionally laden, dyadic measures. Further  research  is  also  operationalize this construct.  necessary  to  As indicated by the current study,  the E M C S appears to have stronger internal consistency (3 = as compared to the MC-SDS (3 = .77). identified which will clarify the For  instance,  distinct  patterns  .86)  A factor structure may yet be  role and nature might  of this construct.  emerge  suggesting  that  spouses present an ideal image of their relationship separate from an ideal presentation of their partner.  Each construct may serve a  different function depending upon the demands of specific contexts. Overall, the  this  study  provides consistent findings supporting  salience of hopelessness within this population.  depression  and  burden,  the  construct  of  Specific to  hopelessness  is  significantly related to depressive thought processes and caregiver burden within this sample of spousal caregivers.  As a uniquely  cognitive variable, hopeless ideation appears to supersede objective factors related study  to the  provides  construct.  The  initial  problematic outcomes of caregiving. evidence  challenge  now  regarding  the  exists to  more  significant fully  This of  this  assess  the  function and etiology of hopelessness among caregivers and other populations.  105  References  Abraham,  I.  (1991).  The  Hopelessness Index:  Geriatric  Depression  Scale  and  Longitudinal psychometric data on frail  nursing home residents. Perceptual  and Motor  Skills,  72,  875-  880. Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. depression:  A  theory-based  Psychological  Bulletin,  96,  (1989).  subtype  of  depression.  358-372.  Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1988). depression:  Hopelessness  Hopelessness  An empirical search for a theory based subtype.  In R. E. Ingram (Ed.), Contemporary depression:  Treatment,  psychological  research  approaches  and theory.  New York,  to NY:  Plenum Press. Abramson, L. Y., Seligman, M. E. P., & Teasdale, J . helplessness in humans: of Abnormal  (1978).  Critique and reformulation.  Psychology,  87,  differences  in  the  sensitivity  of  Alloy,  9,  processes  L. Y.  perspectives. in  Guilford Press.  Psychology  443-445.  L. B. & Abramson, theoretical  (1994).  two self-report  depression scales in older depressed inpatients. and Aging,  Journal  49-74.  Allen-Burge, R., Storandt, M., Kinscherf, D. A., & Rubin, E. H. Sex  Learned  depression,  (1988). In  L.  B.  Depressive Alloy  realism:  (Ed.),  (pp. 223-265). New York,  Four  Cognitive NY:  The  106  Alloy, L. B., Abramson, L. Y., Metalsky, G. I., & Hartlage, S. The  hopelessness  aspects. British  theory  Journal  of  depression:  of Clinical  Psychology,  Alloy, L. B., Lipman, A. J . , & Abramson, L. Y.  Attributional 27,  history  Research, American  of mood  16,  Attributional Validation by  Cognitive  Therapy  and  391-407.  Psychiatric  statistical  disorders.  5-21.  (1992).  style as a vulnerability factor for depression: past  (1988).  Association  manual  of  mental  ( 1987).  Diagnostic  and  disorders.  (3rd ed., revised).  (1994).  Diagnostic  Washington, DC: Author. American  Psychiatric  statistical DC:  Association  manual  of mental  disorders.  (4th ed.).  and  Washington,  Author.  Anastasi, A. (1988).  Psychological  testing.  (6th ed.). New York, NY:  Macmillan Publishing Comp. Anthony-Bergstone, C. R., Zarit, S. H., & Gatz, M. of  psychological  patients.  distress  Psychology  Barusch, A. S.  (1988).  and Aging, Problems  spouse caregivers. Baumgarten, M. the  Journal  The  elderly:  of Clinical  caregivers  3,  and coping  health A  of  dementia  strategies  of elderly  28,  677-685.  of persons  critical  Epidemiology,  42,  review  of  giving the  of depression.  care  to  literature.  1137-1148.  Beck, A. T., Rush, A. J . , Shaw, B. F. & Emery, G. (1979). therapy  Symptoms  245-248.  The Gerontologist,  (1989).  demented  among  (1988).  Cognitive  New York, NY: The Guilford Press.  107  Beck, A. T. & Steer, R. A. (1988). San Antonio, TX: Beck, A. T.  K.  M.  Comprehensive  Hopelessness  Scale  manual.  Harcourt Brace Jovanovich, Inc.  & Weishaar,  Freeman,  Beck  M.  E. (1989).  Simon, L.  handbook  E.  Cognitive  Beutler  of cognitive  & H.  therapy.  In A.  Arkowitz  (Eds.),  therapy.  New York, NY:  Plenum Press. Beck,  A. T.,  Weissman,  A.,  Lester,  measurement of pessimism: of Becker,  Consulting J.  &  E.  Psychology, (1988).  reactions to chronic  as  by  Psychology Bland, R.  C,  spouse  and Aging, Newman,  3,  S. C ,  psychiatric  disorders  Psychiatrica  Scandinavica,  Borg, W.  R.  &  introduction.  White  (1974).  Journal  861-865. in  severe  caregivers  The  assessing  external stress  of Alzheimer  patients.  300-306. & Orn, H.  in  the 77  Gall, M. D.  42,  Difficulties  depressive-like exemplified  L.  The hopelessness scale.  and Clinical  Morrissey,  D. & Trexler,  elderly  in  (Suppl.338),  (1989).  Plains,  (1988).  NY:  Prevalence  Edmonton.  of  Acta  57-63.  Educational  research:  An  Longman.  Brink, T, L., Yesavage, J . A., Owen, L., Heersema, P. H., Adey, M., & Rose, T. L. Clinical Brown, R. P., (1983).  (1982).  Gerontologist, Sweeney, Age  as  endogenous Psychopharmacology,  Screening tests for geriatric depression. 1,  37-43.  J . , Frances, A., Kocsis, J . H., & Loutsch, E. a  predictor  depression. 3,  of  treatment  Journal 176-178.  response of  in  Clinical  108  Canadian  Study  Canadian  of Study  prevalence Journal, Canadian  on  Health  caring  Canadian Aging,  Cappeliez,  P.  Health  and  Working  Aging:  Aging  people  Working  with  and  Association  Group (1994b). Canada:  Study of Health and Aging. Canadian  Journal  Social  dementia  depression 9,  desirability  inventories  in  response the  set  elderly.  and Clinical  45-52. Some thoughts on the prevalence and etiology Canadian  Journal  on  experience  of  431-440.  Chenoweth, B. & Spencer, B. (1986). family caregivers. Journal  Dementia:  The  of the American  Geriatrics  Society,  267-272.  Chodorow, N.  (1978).  University Cohen, D.  methods  Medical  of depressive conditions in the elderly.  26,  (1994a).  470-487.  Cappeliez, P. (1988).  7,  Study  Canadian  and  for  (1989).  Gerontologist,  Aging,  Group  in  13,  self-report  Aging  899-913.  of  of  and  dementia.  150,  Patterns  of  of  Study  The  Health  &  The reproduction  of California  Eisdorfer, C.  of mothering.  Berkeley, CA:  Press. (1988).  Depression  in family  caring for a relative with Alzheimer's disease. American  Geriatrics  Society,  36,  Cohen, D., Kennedy, G. & Eisdorfer, C. family  caring  for  family  management.  a  relative  Journal  members of  the  885-889. (1983).  with  Clinical reality of the  Alzheimer's  The Gerontologist,  23,  disease  263.  and  109  Cohen, J .  (1988).  sciences  Statistical  (2nd  Associates, Cohen, S.  &  New  parasuicide  for  NY:  the  J.  college  Clinical  &  Lawrence  support  Bulletin,  student  L.  among  disease.  34,  social  Psychology,  George,  institutionalization Alzheimer's  Stress,  Psychological  two  and  98,  56,  K.  (1986).  caregivers  Journal  of  of  the  Journal  patients  American  caregivers  Therapy  and  Cooper, J . & Fazio, R. H.  Crockett, D.,  of  Alzheimer's  Research,  (1984).  in Experimental  Crowne,  D.  Journal  Consulting Cutrona, C. E. Journal  with  Geriatrics  9,  to  Social  Holliday, S.,  on  Aging,  independent Psychology, (1983).  of Abnormal  depression  disease  patients.  253-266.  A new look at dissonance theory.  P. & Marlowe, D.  desirability  of  Relationships  Psychology,  &  Koch, W.  17,  4,  24,  A multi-  populations.  101-109.  (1960). of  229-266.  (1985).  focus assessment scale for use with frail elderly Canadian  of  493-498.  in  Coval, M.,  the  Predictors  associated with coping reactions  spousal  and  131-136.  of cognitions  Advances  Erlbaum  310-357.  samples.  Coppel, D. B., Burton, C , Becker, J . , & Fiore, J . (1985).  Cognitive  behavioral  Hopelessness, social desirability, depression and  in  Consulting  Society,  York,  (1985).  hypothesis.  E.  analysis  Publishers.  Cole, D. A. (1988).  Colerick,  ed).  Wills, T. A.  buffering  power  A  new  scale  psychopathology.  of social Journal  of  349-354.  Causal attributions and perinatal depression. Psychology,  92,  161-172.  110  Davison, G . C. & Neale, J . M. (1990). Abnormal  psychology.  New York,  NY: John Wiley & Sons. Deimling,  G . T.  &  impairment  Journal  &  stress  D. M.  (1986).  of  Gerontology,  O'Brien, T. (1990).  and  coping:  Alzheimer's patients.  mental  An  families.  41,  778-784.  An interpersonal framework for  application  In M. A.  to  the  families  of  Paris Stephens, J . H. Crowther,  S.E. Hobfoll & D. L. Tennenbaum later-life  Symptoms of  among elderly adults and their effects on family  caregivers. DeLongis, A.  Bass,  (Eds.),  New York, NY:  Stress  and coping  in  Hemisphere Publishing.  Derogatis, L. R., Lipman, R. S., Covi, L , Richels, K. & Uhlenhuth, E. R. (1970).  Dimensions  Comparison Journal  of  a  of Consulting  Dowd, E. T. (1992).  of  clinical  outpatient versus  and Clinical  an  neurotic  pathology:  empirical  Psychology,  assessment.  34,  164-171.  Review of the Beck Hopelessness Scale.  Murphy, J . J . Kramer & J . C. Conoley (Eds.), The eleventh measurements  yearbooks.  In L. L. mental  Lincoln, NB: University of Nebraska  Press. Drinka, T. J . K., Smith, J . C ,  & Drinka, P. J . (1987).  Correlates of  depression and burden for informal caregivers of patients in a geriatrics Society,  referral clinic. Journal 35,  of the American  522-525.  Duijnstee, M. (1992).  Caring for a demented family member at home:  Objective and subjective evaluation of the burden. Jones  & B. L. M.  Research  Geriatrics  Miesen  and applications.  (Eds.),  Care-giving  In G . M. M. in  dementia:  London, UK: Tavistock/Routledge.  111  Dura, J . R. & Kiecolt-Glasser, J . K. (1990). research.  Journal  of Gerontology,  Sample bias in caregiving  45,  200-204.  Eagles, J . M., Craig, A., Rawlinson, F., Restall, D. B., Beattie, J . A. G. & Besson,  J . A.  supporters  of  Psychiatry, Eaves, G .  &  O.  (1987).  the  150,  The  demented  psychological  elderly.  Journal  of  British  (1984).  Cognitive patterns in symptomatic  and remitted unipolar major depression. 93,  Journal  of  Abnormal  31-40.  Edmonds, V. H. (1967). measurement.  of  293-298.  Rush, A. J .  Psychology,  well-being  Marital conventionalization:  Journal  of  Definition and  Marriage  and  the  Family,  DiBatista,  B.  (1972).  29,  681-688. Edmonds,  V.  H.,  Withers,  G.  and  marital  conservatism Marriage Edwards,  and the Family,  A.  L.  personality  (1957). assessment  &  34,  conventionalization.  Adjustment, Journal  of  variable  in  96-103.  The  social  and  research.  desirability  New York, NY:  Dryden  Press. Feingold, A.  (1995).  The additive effects of differences in central  tendency  and  variability  between  groups.  American  are  Psychologist,  Fengler, A. P. & Goodrich, N. (1979). The hidden patients.  important  in  50,  comparisons 5-13.  Wives of elderly disabled men:  The Gerontologist,  19,  175-183.  112  Fiore, J . , Coppel, D. B., Becker, J . & Cox, G. B. as  a  multifaceted  dimensions for Psychology, Fisher, G.  concept:  adjustment.  14,  (1967).  Examination  American  Journal  Social support of of  Psychological  dementia  The  Marlowe-Crowne  Reports,  patients:  Gerontologist,  20,  Social  26,  a  predictor  of  Personality  Friedan, B.  of  how  and  (1993).  Desirability  174. Caregivers  A comparison of husbands and wives. 248-252.  Follette, V. M. & Jacobson, N. S. (1987). as  Community  Normative and reliability data for the standard  Fitting, M., Rabins, P., Lucas, M. J . , & Eastham, J . (1986). for  important  93-111.  and the cross-validated Scale.  (1986).  people  Social  The fountain  Importance of attributions  cope  with  Psychology,  52,  of age.  failure.  Journal  1205-1211.  New York, NY: Simon &  Schuster, Inc. Fry, P.  S.  (1986).  elderly: Clinical Gallagher,  D.,  (1985). measures meeting  Assessment of  pessimism and  A geriatric scale of hopelessness. gerontology.  New York, NY:  Rappaport,  M.,  Reliability with of  Orleans, LA.  the  of  family  caregivers.  Gerontological  in  the  In T. L. Brink (Ed.),  Hayworth Press.  Benedict, A., Lovett, selected  despair  S. & Silven, D.  interview  and  Paper  presented  Society  of  self-report at  America,  the New  113  Gallagher-Thompson, (1990).  Hanley-Peterson,  D.,  Maintenance  psychotherapy Clinical  for  depression.  effects  of  58,  &  D.  & Thompson,  Journal  of  Steffen,  Clinical  A.  M.  (1994). and  Comparative  brief  Psychology,  62,  older  adults  in  theory  Beutler (Eds.), Foundations and George,  practice.  and  L. K. & Gwyther,  practice.  of  Psychotherapy  Theory,  research,  Oxford University Press.  L. P.  (1986).  multidimensional  examination  demented  The Gerontologist,  adults.  Journal  In B. Bongar & L.  of psychotherapy:  New York, NY:  dynamic  543-549.  Gallagher-Thompson, D. & Thompson, L. W. (in press). with  and  371-374.  cognitive-behavioral  and  brief  Consulting  psychotherapies for depressed family caregivers. Consulting  L. W .  of gains versus relapse following  Psychotherapy,  Gallagher-Thompson,  P.,  of  Caregiver well-being: family 26,  caregivers  A of  253-259.  Georgotas, A., McCue, R. E., Cooper, T., Chang, I., Mir, P., & Welkowitz, J.  (1987).  Clinical predictors of response to  in elderly patients.  Biological  Gilhooly, M. L. M. (1984). Factors  supporting  British  Journal  elderly  172-177.  the  demented  of Medical  Emotional mentally  with  a  Gilleard, C. J . , Belford, H., (1984).  22,  733-740.  The impact of care-giving on care-givers:  associated  people  Psychiatry,  antidepressants  relative  Psychology,  Gilleard, E., distress  infirm.  psychological  57,  the  the  Journal  community.  Gledhill, K.  supporters of  of  35-44.  Whittick, J . E. &  amongst  British  in  well-being  Psychiatry,  of  the 145,  114  Gilley,  D.  W.  (1993).  Alzheimer's (Eds.),  disease.  and affective  In R. W. Parks,  Neuropsychology  dementias. Gilligan, C.  Behavioral  of  disturbances  in  R. F. Zee, & R. S. Wilson  Alzheimer's  disease  and  other  New York, NY: Oxford University Press.  (1982).  women's  In a different  development.  voice:  Psychological  Cambridge,  MA:  theory  Harvard  and  University  Press. Grigg,  D. N.  (1994).  An  individual  and  couples  systemic  therapy  doctoral  ecological treatment  for  dissertation,  alcohol  of  recently  Annual  Review  of Gerontology  formats  of  (1982).  published  of the efficacy of  Unpublished  British  Depression studies.  of  experiential  dependency.  University  Gurland, B. J . & Toner, J . A. review  assessment  Columbia. in the elderly: A  In C. Eisdorfer (Ed.),  and Geriatrics.  New York, NY:  Springer Publishing Comp. Gwyther,  B. & Spencer,  family caregivers.  B.  (1986).  Dementia:  The Gerontologist,  26,  The experience of 267-272.  Hadjistavropoulos, T., Taylor, S., Tuokko, H., & B. L. Beattie. (1994). Neuropsychological deficits Geriatrics  and  deficits,  caregiver  Society,  42,  caregivers'  burden.  Journal  perceptions of  the  American  308-314.  Haley, W. E., Levine, E. G., Brown, S. L , & Bartolucci, A. A. Stress,  appraisal,  Aging, 2,  (1987).  coping and social support as predictors of  adaptational outcome among dementia caregivers. and  of  323-330.  Psychology  115  Haley, W. E., Levine, E. G., Brown, S. L , (1987).  Psychological,  caring  for  American  a  social  relative  Geriatrics  stress  and  with  senile  Society,  35,  Haley, W. E. & Pardo, K. M. (1987). to caregiver  Berry, J . W. & Hughes, G. H. health  dementia.  Journal  of  of  the  405-411.  Relationship  and coping.  consequences  of stage  of  dementia  Paper presented at the annual  meeting of the American Psychological Association, New  York,  NY. Hamilton,  E. W.  & Abramson,  L. Y.  (1983).  Cognitive patterns and  major depressive disorders:  A longitudinal study in a hospital  setting.  Journal  Psychology,  Hamilton, M.  (1967).  depressive  M.  whose  J . (1985).  of  Hill,  R.  D.,  rating of  173-184. scale  Social  for  primary  and  Clinical  of  Lopez,  institutionalization  Southern F.  aged  Gallagher,  D.,  patients  regarding over  doctoral  a  one  patient year  dissertation,  California.  G. & Nation,  of Personality  dementia  services  Unpublished  community-residing Journal  Journal  counseling  period.  University  a  Community-dwelling  sought  Predictors  follow-up  B.,  British  of  92,  6, 287-296.  relatives  care:  Hayslip,  Development  illness.  Psychology, Hassinger,  of Abnormal  P.  (1991).  persons:  Assessment,  A 57,  Thompson, L. W.  Hopelessness viable  in  construct?  498-505. & Ishida, T.  (1988).  Hopelessness as a measure of intent in the depressed elderly. Psychology  and Aging,  3,  230-232.  116  Hollon, S. D., DeRubeis, R. J . , Evans, M. D., Wiemer, M. J . , Garvey, M. J . , Grove, W. M.,  & Tuason, V. B.  pharmacotherapy Archives Hooker, K.,  in  of General  Frazier, L. D.,  coping  among  Gerontologist,  depression:  Cognitive therapy and  Singly  Psychiatry,  49,  and  in  of  spouses  with  Personality and dementia.  the behavioral  Hall, Ignatieff,  The  386-392.  Hopkins, K. D., Glass, G. V., & Hopkins, B. R. (1987). for  combination.  774-781.  & Monahan, D. J . (1994).  caregivers 34,  (1992).  sciences.  Basic  statistics  Englewood Cliffs, NJ:  Prentice-  Inc.  M.  (1993).  Scar  tissue.  Toronto,  ON:  Penguin Books  Canada, Ltd. Ineichen, B. (1987).  Measuring the rising tide:  cases will there be by 2001? 150, Jarvik,  L.  L.  Journal  of  Psychiatry,  Some  unanswered  193-200. F.  (1976).  questions. Jarvik,  British  How many dementia  F.,  Journal Ruth,  syndrome  Aging  and  V.,  of  and  depression:  Gerontology,  Matsuyama,  aging.  31,  324-326.  S. S. (1980).  Archives  of  General  Organic  brain  Psychiatry,  37,  280-286. Jeans, E. R.,  Helmes, E., Merskey, H., Robertson, J . M., & Rand, K. A.  (1987). Canada.  Some calculations of the prevalence Canadian  Journal  of  Psychiatry,  Jewish Publication Society of America translation text  of the  (3rd ed.).  Holy  Scriptures  Philadelphia, PA:  (1982). according Author.  32,  of dementia in 81-85.  Kethubim: to the  A  new  Masoretic  117  Jorm, A. F., Korten, A. E., & Henderson, A. S. (1987). of  dementia:  Acta  A quantitative  Psychiatrica  Kaszniak, A.  W.  diagnosis  Scandinavica,  & DiTraglia of  integration  dementia  76,  Christenson, and  depression  in  Washington, DC: Katzman,  R.  & Jackson,  516-  the  465-479. (1994).  Glaser, R.  depression. In M. Storandt & G. R.  older  assessment  adults:  A  of  dementia  clinician's  guide.  J . E. (1991). Journal  Alzheimer  of the American  disease:  Geriatrics  Dura, J . R.,  (1991).  Speicher, C ,  Medicine,  53,  E.  (Doctoral  Spousal  style  The  caregivers  of  and  NADPH  alterations  of the  questionnaire  dissertation,  Dissertation  revision  for  University  Abstracts  Kowall, N. W. & Beal, F. (1988). Y,  Society,  dementia  victims:  Psychosomatic  345-362.  (1985).  attributional  1984).  Basic  Trask, O. J . , &  Longitudinal changes in immunity and health.  R.  Differential  525.  Kiecolt-Glaser, J . K.,  Koch,  literature.  American Psychological Association.  clinical advances. 39,  of  G.  VandenBos (Eds.), Neuropsychological and  The prevalence  diaphorase  of  learned  an  helplessness  older  Wisconsin  International,  46,  population -  Madison, 1520-A.  Cortical somatostain, neuropeptide neurons:  in Alzheimer's disease.  Normal anatomy  Annals  of Neurology,  and 23,  105-114. Lawton, M. P., study patients.  Brody, E. M., & Saperstein, A. R. of  respite  service  The Gerontologist,  for 29,  (1989). A controlled  caregivers 8-16.  of  Alzheimer's  118  Lazarus, R. S. & Folkman, S. (1984). Gentry (Ed.),  Handbook  Coping and adaptation.  of behavioral  medicine.  In W. D.  New York, NY:  Guilford Press, Inc. Linden, W., Paulhus, D. L. & Dobson, K. S. (1986). styles  on  Journal  the  of Consulting  Linehan, M. M.  (1993).  personality Lomax,  report  R.  and  and  Cognitive-behavioral  Statistical the  somatic  Psychology,  New York, NY:  (1992).  education  psychological  and Clinical  disorder.  G.  of  Effects of response  54,  distress.  309-313.  treatment  of  borderline  The Guilford Press.  concepts:  A  behavioral  sciences.  (1991).  The 36-hour  second  course  for  White  Plains,  NY:  Longman. Mace, N. L. & Rabins, P. V. to  coping  for  dementing MD:  persons  illnesses  John  with  Alzheimer  and memory  Hopkins  University  loss  day;  A family  guide  disease  and  related  in later  life.  Baltimore,  Press.  Manly, P. C , McMahon, R. J . , Bradley, C. F. & Davidson, P. O. Depressive childbirth.  attributional Journal  McEvoy DeVellis, B. hopelessness expectancy. McKhann, G., Stadlan, disease:  Blalock, S.  (1992).  depression:  Drachman, D.,  Report  depression  Psychology,  &  M.  and  of Abnormal  Journal  E.  style  of Abnormal  the  245-254.  of  Psychology,  Clinical  following  Illness attributions and  role  Folstein, M.,  (1984). of  The  91,  hopelessness 101,  Katzman, R., diagnosis  NINCDS-ADRDA  Price, D.,  of  work  257-264.  Neurology,  &  Alzheimer's group  the auspices of the Department of Health and Human task force on Alzheimer's disease.  (1982).  34,  under  Services 939-944.  119  Miller,  B.  &  Cafasso,  Fact or artifact. Miller, I. W.,  The  Klee, S. H.,  nondepressed  Abnormal I.  W.,  W.  stressful  91, H.,  live events.  and  events,  Journal  of  of  depressed  I.  (1989).  Cognitive-  inpatients:  American  Journal  Six-  of  and  Psychiatry,  1274-1279.  C.  Psychological (1988).  Implications  for  Experimental and  hypothetical  & Keitner, G.  month follow-up.  disorders. R.  of  Depressed  78-81.  Mineka, S. & Sutton, S. K. (1992).  Mohs,  498-507.  (1983).  cognitions  and  treatment  twelve-  32,  & Norman, W. H.  tasks,  Norman,  Gender differences in caregiving:  Gerontologist,  Psychology,  behavioral  146,  (1992).  inpatients'  experimental  Miller,  L.  Cognitive biases and the emotional  Science,  the  use  models  of  of  animal  wellbeing  Neurobiology  New York, NY:  of  Unpublished  the spouse  Special Issue:  memory of Aging,  (1981).  The  dysfunction 9,  465-468.  epidemiology  of  Oxford University Press.  The psychological  master's  models.  age-related  Mortimer, J . A. & Schuman, L. M.  Morris, L. W. (1986).  65-69.  Memory impairment in amnesia and dementia:  neurodegeneration.  dementia:  3,  factors  affecting  emotional  of  dementia  sufferers.  caregivers  thesis,  University  of Newcastle-upon-  Tyne, Newcastle, UK. Morris, R. G., Morris, L. W. & Britton, P. G. (1988). the  emotional  sufferers.  British  wellbeing Journal  of  the  of Psychiatry,  Factors affecting  caregivers 153,  of  dementia  147-156.  120  Niederehe,  G. & Funk,  patients:  J . (1987).  Caregivers  presented  at  Psychological  styles  the  interaction  and  annual  Association,  Family  their  meeting  New York,  of  postpartum  predictive  factors.  depression: Journal  of  dementia  correlates.  Paper  of  the  American  NY.  O'Hara, M. W., Neunaber, D. J . , & Zekoski, E. M. study  with  (1984).  Prevalence, Abnormal  Prospective course, and  Psychology,  91,  457-461. Okun, M. A., Olding, R. W., Cohn, C. M. G. (1990). A meta-analysis of subjective  well-being  Psychological  interventions  Bulletin,  108,  among  elders.  257-266.  O'Rourke, N., Hayden, S., Haverkamp, B. E., Tuokko, H., & Beattie, B. L. (1995).  Quality  of prior  expressed  burden  dementia  patients.  British  among  Columbia,  relationship,  social  spousal  caregivers  Unpublished Clinic for  desirability,  manuscript,  Alzheimer  of  and  suspected  University  Disease and  of  Related  Disorders, Vancouver, B C . O'Rourke,  N., Haverkamp,  & Beattie, B. L. expressed dementia British  (1994).  burden patients.  Columbia,  B. E., Rae, S., Tuokko, H., Hayden, S., Response  among  spousal  Unpublished Clinic  Disorders, Vancouver, B C .  for  biases  as a confound  caregivers manuscript,  Alzheimer  of  to  suspected  University  of  Disease and Related  121  O'Rourke, N., Tuokko, H., Hayden, S., & Beattie, B. L. (1995). identification  of  Clock  Unpublished  Test.  dementia:  Predictive manuscript,  Columbia, Clinic for Alzheimer  validity  Early  of  University  of  the British  Disease and Related Disorders,  Vancouver, BC. O'Rourke, N., Tuokko, H., Klassen-Peters, L , Rae, S., & Beattie, B. L. (1994).  Service  presented  at  use  the  by  annual  persons general  with  dementia.  meeting  of  the  Paper Canadian  Association on Gerontology, Winnipeg, Man. Owen, S. V. (1992).  Review of the Beck Hopelessness Scale.  In L. L.  Murphy, J . J . Kramer &J. C. Conoley (Eds.), The eleventh measurements  yearbooks.  mental  Lincoln, NB: University of Nebraska  Press. Pagel, M. D. & Becker, J . (1987). Relations  with  Personality Pagel,  Depressive thinking and depression:  personality  and Social  and social  Psychology,  52,  M. D., Erdly, W. W. & Becker, networks:  We get by with  our friends. Journal  J.  resources.  Journal  of  1043-1052. (1987).  Social  support  (and in spite of) a little help from  of Personality  and Social  Psychology,  53,  793-804. Parks, C. W. & Hollon, S. D. (1988). Hersen practical  & A. S. Bellack handbook.  Paulhus, D. L. (1984). responding. 598-608.  Journal  (Eds.),  Cognitive  assessment.  Behavioral  assessment:  In M. A  New York, NY: Pergamon Press. Two-component of Personality  models of socially and Social  desirable  Psychology,  46,  122  Paulhus, D. L. (1991).  Measurement and control of response bias.  J. P. Robinson, P. R. Shaver & L. S. Wrightsman (Eds.), of personality CA:  and social  psychological  attitudes.  In  Measures San Diego,  Academic Press, Inc.  Paulhus, D. L. & Reid, D. B. (1989). into  two  components:  management.  Partitioning the Marlowe-Crowne  Self-deception  Unpublished  and  impression  manuscript, University of British  Columbia, Department of Psychology, Vancouver, B C . Pearson-Scott, J . , Roberto, K. A. & Hutton, J . T. (1986). Alzheimer's Journal Pearlin,  victims:  Family  of the American  Geriatrics  L, I., Mullan, J . T., Semple,  Caregiving  and the  and their measures.  stress  of Health Percy, M. E.  predictive  34,  caregivers.  348-354. M. M.  30,  (1990).  583-594.  The structure of coping.  Behavior,  Journal  19, 2-21.  Peripheral biological markers as confirmatory  tests  for  Alzheimer  J. Holland (Eds.), Alzheimer  Persons, J . B.  the  S. J . , & Skaff,  population and in Dow syndrome.  relationship.  Society,  The Gerontologist,  and Social (1993).  to  process: An overview of concepts  Pearlin, L. I. & Schooter, C. (1978).  or  support  Families of  disease  in  the  general  In J . M. Berg, H. Karlinsky, A.  disease,  Dow Syndrome,  and  their  Toronto, ON: Oxford University Press. &  Rao, P. A.  cognitions,  life  inpatients.  Journal  events  (1985). and  of Abnormal  Longitudinal depression  Psychology,  94,  in  studies of psychiatric 161-167.  123  Peterson, C.  & Seligman, M. E. P.  risk factor for depression: Review,  91,  (1984).  Causal explanations as a  Theory and evidence.  347-374.  Peterson, C. & Villanova, P. questionnaire.  (1988).  Journal  An expanded attributional style  of Abnormal  Psychology,  Pratt, C. C , Schmall, V. L , Wright, S. & Cleland, M. and  coping  Family  Psychological  strategies  Relationships,  of  34,  97,  87-89.  (1985).  Burden  caregivers to Alzheimer's patients. 27-33.  Quayhagen, M. P. & Quayhagen, M. (1988). Alzheimer's stress: Coping with the caregiving role. The  Gerontologist,  Rabins, P. V., Mace, N. L. & Lucas, M. J. dementia  on the  Association,  248,  Radloff, L. S. scale  family.  (1977). for  The CEDS-D scale: in  the  Measurement,  cognitive impairment  Robinson,  B.  1,  Journal Robinson,  L.  (1983).  Berman,  Psychotherapy  for  comprehensive  review  Psychological  American  Medical  A self-report depression population.  Applied  385-401.  Bulletin,  80,  (1989).  The prevalence of  38, J.  of  a  &  Neimeyer,  treatment of  108,  caregiver  strain  index.  A.  (1990).  344-348. S.,  the  Acta  303-309.  Validation  of Gerontology, A.,  the  The impact of  in an elderly Canadian population.  Scandinavica,  C.  (1982).  general  Robertson, D., Rockwood, K., & Stolee, P.  Psychiatrica  of  391-396.  333-335.  research  Psychological  Journal  28,  controlled 30-49.  of  R.  depression: outcome  A  research.  124  Roth, M., Huppert, F. A., Tym, E., & Mountjoy, C. Q. (1988). The  Cambridge  elderly.  examination  Cambridge, UK:  Rotter, J . B.  (1966).  for  R.  G., Tarnowski,  marital  Alcohol,  Psychological  K. J . & St.  adjustment  50,  C.  Monographs,  mechanisms Geriatrics W.  R.  L. S. (1989).  response sets on the alcoholics.  I.  between and  Society, R.,  in  Lawrence,  Journal  self-report  of Studies  on  24-29.  & Shader,  Relationship  Schumm,  the  Generalized expectancies for internal versus  Impact of social desirability  Salzman,  of  (1, Whole No. 609).  Rychtarik,  of  disorders  Cambridge University Press.  external control of reinforcement. 80  mental  CAMDEX:  (1978).  Depression in the  depression,  physical 26,  Bollman, S.  illness.  elderly:  psychological Journal  of  defense  the  American  253-259. R.  & Jurich, A. P.  communication or marital conventionality? the Relationship Inventory.  (1980).  A brief  Psychological  Reports,  Marital  report on 46,  1171-  1174. Scogin, F. R.  (1994).  Assessment of depression in older adults:  guide for practioners. Neuropsychological older adults:  A  In M. Storandt & G. R. VandenBos (Eds.),  assessment  A clinician's  of dementia  guide.  and depression  Washington, DC:  in  American  Psychological Association. Scogin,  F.  R.  &  treatments Journal  McElreath, for  geriatric  of Consulting  L.  (1994).  Efficacy of psychosocial  depression:  and Clinical  A qualitative  Psychology,  62,  review.  69-74.  125  Seligman,  M.  E.  Today,  P.  7,  (1973).  Fall  into hopelessness.  Psychology  43-48.  Seligman, M. E. P., Abramson, L. Y., Semmel, A. & von Baeyer, C. (1979).  Depressive attributional style.  Psychology, Selkoe, D. J . 267,  88,  (1992).  Journal  of  Abnormal  242-247. Aging brain, aging mind.  Scientific  American,  135-142.  Shaver, P. R & Brennan, K. A. (1991). Measures loneliness. (Eds.),  In J. P. Robinson, P. R. Shaver Measures  attitudes.  personality  and  and  & L. S. Wrightsman social  psychological  San Diego, CA: Academic Press, Inc.  Shapiro, D. A.  &  Shapiro, D.  outcome studies: Bulletin, Sheikh, J . I.  of  of depression  92, &  (1982).  Meta-analysis of comparative  A replication and refinement.  Psychological  581-604.  Yesavage, J . A.  (1986).  Geriatric Depression Scale  (GDS):  Recent evidence and development of a shorter version.  Clinical  Gerontologist,  Shulman,  K.  I.,  (1993).  Pushkar  and  study. 8,  165-173.  Gold, D.,  Clock-drawing  longitudinal Psychiatry,  5,  Cohen, C A., & Zucchero, C. A. dementia  International  in  the  community:  Journal  of  A  Geriatric  487-496.  Skoog, I., Nilsson, L , Palmertz, B., Andreasson, L. A., & Svanborg, A. (1993). olds.  A population-based  New England  Journal  study  of  of Medicine,  dementia 328,  in  154-158.  85-year-  126  Soldo, B. J .  &  Hyllyluoma, J .  dependent elderly. Spangler, D. L ,  (1983).  Caregivers who live with the  The Gerontologist,  Simons, A. D.,  23,  605-611.  Monroe, S. M., & Thase, M. E. (1993).  Evaluating the hopelessness model of depression: stress  and  symptom  Psychology,  102,  components.  Journal  Rationale  Psychiatry, SPSS  Inc.  35,  (1988).  of  Abnormal  592-600.  Spritzer, R. L , Endicott, J . , & Robins, E. (1978). criteria:  Diathesis-  and  reliability.  Research diagnostic  Archives  of  General  773-782. SPSS-X user's  guide  (3rd ed.).  Chicago,  IL:  Author. Statistics  Canada  (1992).  status.  (Catalogue  Services  Canada.  Statistics  Canada  Provinces 520). Stotland,  No.  (1994).  and  (1969).  Jossey-Bass  projections  1993-2016.  psychology  of  for  and  Canada,  San Francisco, CA:  Publishing. (1983).  private  and  social desirability  Journal  Supply  (Catalogue No. 91-  hope.  Social confirmation of dysphoria:  reactions  Social  Psychology,  to 44,  Strosahl, K. D., Linehan, M. M., & Chiles, J . A.  social  marital  Supply and Services Canada.  The  and  Personality  age, sex and  Ottawa, ON:  Population  Strack, S. & Coyne, J. C. Shared  of Canada:  93- 310).  Territories:  Ottawa, ON: E.  Census  depression.  Journal  of  798-806. (1984).  Will the  real  please stand up? Hopelessness, depression,  desirability, and of Consulting  the  prediction  and Clinical  of  Psychology,  suicidal 52,  behavior. 449-457.  127  Strub, R. L. & Black, F. W. neurology Styron, W.  (1993).  (3rd ed.).  (1990).  The mental  status  examination  in  Philadelphia, PA: F. A. Davis Company.  Darkness  visible:  A memoir  of madness.  New  York, NY: Random House, Inc. Tabachnick,  B. G.  statistics.  &  Fidell,  New York, NY:  Teng, E. L. & Chui, H. C. (3MS)  L. S.  (1989).  multivariate  Harper Collins Publishers, Inc.  (1987).  The  Journal  of Clinical  Examination.  Using  Modified  Mini-Mental State  Psychiatry,  48,  314-  318. Teri, L. (1991).  Behavioral assessment and treatment of depression  in older adults. behavior  In P. A. Wisocki (Ed.),  therapy  with  the  Handbook  elderly  client.  of  clinical  New York,  NY:  Plenum Press. Teri, L ,  Borson, S.,  disturbance,  Kiyak, A.  &  cognitive  Yamagishi, M.  dysfunction  (1989).  and  functional  Prevalence and relationship in Alzheimer's disease. the American Teri,  L,  Curtis,  (1991). adults.  Geriatrics  Society,  J . , Gallagher-Thompson, Cognitive  behavior  skill:  Journal  of  109-116. D.,  therapy  & Thompson, with  L.  depressed  W.  older  In L. S. Schneider, C. F. Reynolds, B. D. Lebowitz, & A. J .  Friedhoff (Eds.), Diagnosis life.  37,  Behavioral  Washington,  Teri, L. & Truax, P. patients: ratings.  DC:  The  of depression  in  late  American Psychiatric Press, Inc.  (1994).  Association  and treatment  Assessment of depression in dementia of  Gerontologist,  caregiver 34,  mood  231-234.  with  depression  128  Teusink,  J.  P.  &  Alzheimer's  Mahler, S.  (1989).  Disease. Hospital  Helping  and  families  Community  cope  with  Psychiatry,  35,  152-156. Tiggemann,  M., Winefield,  (1991). to  A. H., Winefield, H.  R. & Goldney,  The stability of attributional style and its relation  psychological  Psychology,  patients.  distress.  30,  Trenteseau, J . A.  &  British  Journal  of  Clinical  247-255. Hyer, L.  Journal  (1989).  of Applied  Hopelessness among later-life  Gerontology,  8,  355-364.  Tuokko, H., Crockett, D., Beattie, B. L , Horton, A., & Wong, M. February).  The use of rating  functioning  in  demented  annual  meeting  Society,  Denver,  Tuokko,  H.,  of  scales  patients.  the  to assess Paper  International  Kristjansson,  neuropsychological Health and  (1986,  psycho-social  presented  at  the  Neuropsychological  CO. B.,  &  Miller,  J.  neuropsychological detection of dementia:  of  R. D.  components  Aging.  Journal  of of  (in  press).  The  An overview of the  the Clinical  Canadian and  Study  Experimental  Neuropsychology. Tuokko, H.,  Vernon-Wilkinson, R.,  Weir, J . , &  Cued recall and early identification Clinical Verbrugge, L.  and  Experimental  (1985).  and evidence. 182.  of dementia.  Neuropsychology,  Gender and health:  Journal  Beattie, B. L.  of Health  13,  (1991).  Journal  of  871-879.  An update on hypotheses  and Social  Behavior,  26,  156-  1 29  Vitaliano, P. P., Young, H. M. & Russo, J . (1991).  Burden: A review of  measures used among caregivers of individuals with dementia. The  Gerontologist,  Weiss, I. K.,  31,  67-75.  Nagel, C. L. & Aaronson, M. K.  depression American Weiss, R. L.  scales to Geriatrics  &  the  old  Society,  Cerreto, M. C.  old  34,  (1986).  person.  Applicability of Journal  of  215-218.  (1980).  The Marital Status Inventory:  Development of a measure of dissolution potential. Journal Whitlatch,  of Family  Therapy,  C. J . , Zarit, S.  interventions meeting  of  G.  the  American  80-85.  & von  with caregivers.  Minneapolis, Williamson,  H.  8,  the  Eye, A.  (1989).  Efficacy  Paper presented at the  Gerontological  Society  of  of  annual  America,  MN.  M.  &  Schulz,  R.  (1990).  Relationship orientation,  quality of prior relationship, and distress among caregivers of Alzheimer's patients. Winogrond, I. R.,  and Aging,  of  caregiver  patient  Gerontologist,  27,  function  burden in  and  a  mental  The  therapeutic  setting.  The  336-339.  systems perspective. Generations, Health  502-509.  morale to Alzheimer's  Woods, A. M., Niederehe, G. & Fruge, E. (1985).  World  4,  Fisk, A. A., Kirsling, R. A. & Keys, B. (1987).  relationship disease  Psychology  Organization and behavioural  (1992). disorders.  10, The  Dementia:  A family  19-23. ICD-10  Geneva:  classification  Author.  of  130  Yesavage, J . A. (1991).  Geriatric Depression Scale:  depressive symptoms over time. Perceptual 73,  Consistency of  and  Motor  Skills,  1032.  Yesavage, J . A., Brink, T. L , Rose, T. L , Lum, O., Huang, V., Adey, M., & Leirer, V. O. (1983).  Development and validation of a geriatric  depression screening scale: Psychiatric Young, M. A.,  Research,  17,  A preliminary  An  Journal  of  37-49.  Halper, I. S., Clark, D. C ,  (1992).  report.  item-response  Scheftner, W.  theory  Hopelessness Scale. Cognitive  & Fawcett, J .  evaluation  of  the  Beck  Therapy and Research,  16,  579-  in  family  587. Zarit, S. H. (1990). caregiving  Conceptual research.  the Invitational  and  methodological  Paper  presented  Conference  on  issues at  the  meeting  of  Family  Caregiving  Research,  (1985).  The hidden  victims  Toronto, ON. Zarit, S. H.,  Orr, N. K.,  Alzheimer's  & Zarit, J . M.  disease:  Families  under  stress.  of  New, York, NY:  New York University Press. Zarit, S. H., Reever, K. E., & Bach-Peterson, J. the The  impaired  elderly:  Gerontologist,  Zarit, S. H., Todd, P. A.  20,  Correlates  26,  of  Relatives of  feelings of burden.  649-655.  & Zarit, J . M.  husbands and wives as caregivers: Gerontologist,  (1980).  260-266.  (1986).  Subjective burden of  A longitudinal study.  The  131  Zarit, S. H. & Zarit, J . M. (1990). Checklist  and  The Memory  the Burden  Interview.  Pennsylvania State University,  and Behavior Unpublished  Problems manuscript,  Gerontology Center,  University  Park, PA. Zee, R. F.  (1993).  disease.  In  Neuropsychological R. W.  Parks,  R.  functioning  F. Zee,  in  Alzheimer's  R. S. Wilson  Neuropsychology of Alzheimer's disease and other New York, NY: Zung, W. W. K. General Zweben, A. advice  (1965).  A self-rating depression scale. 12,  Perlman, S. and  conjoint  83,  Archives  of  63-70.  & in  Li, S. the  (1988).  899-916.  A comparison of brief  treatment  The results of the marital systems study. Addiction,  dementias.  Oxford University Press.  Psychiatry, &  (Eds.),  of British  alcohol Journal  abuse: of  132  Appendix One The Canadian Study of Health and Aging  The Canadian Study of Health and Aging (CSHA) was a nationwide investigation of the health status of persons over 64 years of age in Canada. status,  one  In addition to obtaining a general survey of health  of  the  goals  of  this  study  was  to  determine  prevalence of dementia, in various forms, across the country.  the This  focus emerged in response to the increase in life expectancy, the relationship between age and the prevalence of dementia, and the impact of dementia prevalence on the health care delivery system. The  CSHA  also studied  risk factors such  caregivers of persons with dementia.  and the  role  of  In response to the complexity  of the methodological issues involved in studying the epidemiology of  dementia,  the  CSHA  protocol  for  a  multi-centre,  multi-  disciplinary study was drafted by the Coordinating Study Centre at the  University  of  Ottawa's  Department  of  Epidemiology and  the  Laboratory Centre for Disease Control of the federal government. Early in 1989, 18 centres across Canada agreed to participate in developing and implementing the protocol.  Working groups were  established to guide the development of the six components of the research  protocol:  examination;  screening for  cognitive  impairment;  neuropsychological assessment; the  physical  risk factor study;  the study of the impact of dementia on caregivers; and the ethical aspects of the study (Tuokko, Kristjansson & Miller, in press).  133  Appendix One (Cont.) The Canadian Study of Health and Aging  The C S H A  involved two distinct populations of persons over  64 years of age in Canada: community-dwelling institutions.  and residents of  All participants were required to be fluent  in  either  English or French and were assessed in their preferred language. For the community sample, a two-phase approach to persons with dementia was taken.  identify  First, a screening interview  was  conducted on an age-stratified random sample of persons from five geographically  defined  regions  (i.e.,  Maritimes,  Prairies and British Columbia; N = 9008). interview,  the  Modified  Mini-Mental  administered (3MS; Teng & Chui, 1987). 78/100 on the  3MS  Quebec,  Ontario,  As part of the screening State  Examination  All subjects scoring below  and a subsample of persons scoring 78  greater were invited to attend the clinical component of the (N = 2339).  was  or  CSHA  Those who could not complete the 3MS were referred  for clinical evaluation (Tuokko et al., in press). For the institutional sample, all selected residents (N = were invited to take part in the clinical component of the without undergoing the screening examination.  1817) CSHA  This approach was  adopted because of high prevalence of dementia in institutions made this unnecessary (Bland, Newman & Orn, 1988; Robertson, Rockwood & Stolee, 1989).  1 34  Appendix One (Cont.) The Canadian Study of Health and Aging  The clinical component was designed to confirm the presence of cognitive 3MS)  impairment  in those screened positive  and to allow for a further  impairment  was confirmed.  evaluation;  physical  differential  (<  78  diagnosis if cognitive  This consisted of four parts:  examination;  on the  laboratory  blood  nurse's  work;  and  neuropsychological assessment. The nurse's evaluation included re-administration rudimentary signs,  measures  height,  subject's  weight  history,  of vision and and  cognitive  from a collateral informant H of the  Cambridge  medication ands  hearing, use.  functional  recordings of Information  status  (usually a family member)  Examination  for Mental  of the 3MS,  was  on  vital the  obtained  using section  Disorders (CAMDEX;  Roth, Huppert, Tym & Mountjoy, 1988). During the physical examination,  the physician evaluated  the  general appearance of the subject, examined the head, neck, limbs, chest, and cardiovascular system and evaluated central  reflexes,  coordination. basis of the  peripheral  neuromuscular  The physician made a preliminary information  and by the nurse.  the  primitive  responses  and and  diagnosis on the  collected during the physical  examination  Laboratory blood work was done for subjects  suspected of having dementia or delirium.  135  Appendix  One (Cont.)  The Canadian Study of Health and Aging  Those attaining 3 M S scores of 50 or above during the nurse's evaluation  were  administered  a  standardized  neuropsychological  battery by a trained psychometrician (i.e., technician trained administration). diagnosis  on  The the  psychologist basis  of  made  information  a  in test  neuropsychological collected  by  the  psychometrician and by the nurse (Tuokko et al., in press). In  case  conferences  typically  psychologists,  nurses  and/or  diagnosis  derived  taking  was  information.  attended  by  psychometricians,  relevant  Subjects were classified using a three-stage  process:  (APA,  all  consensus  clinically  DSM-III-R criteria  into account  a  physicians,  1987), differential diagnosis based  upon  DSM-III-R, NINCDS-ADRDA (McKhann et al., 1984), ICD-10 criteria for depression, Alzheimer disease and other dementias The the  clinical assessment resulted following  in classification of subjects  categories:  No cognitive loss Cognitive loss but not demented Dementia: Alzheimer type (probable,  possible)  Vascular Mixed vascular + Alzheimer Other  specific dementia  Unclassified  respectively.  dementia  (specific)  into  136  Appendix Two  Population Projections and the Anticipated Prevalence of Dementia in Canada *  Age 65-74  Group  75-84  85 +  Total  1 999  51,780  144,200  149,937  345,917  2001  52,346  152,480  164,186  369,013  2006  55,166  168,842  199,928  423,936  201 1  64,488  176,612  242,570  483,670  201 6  81,413  189,133  275,379  545,925  B a s e d upon Statistics C a n a d a data (1994) a n d dementia estimates provided by the C a n a d i a n Study of Health a n d Aging ( C S H A Working G r o u p , 1994a).  137  Appendix Three Burden  1 Never  2 Rarely  3 Sometimes  Interview  4 Quite Frequently  1 1.  Do you feel that your s p o u s e asks for more help than s/he n e e d s ?  2.  D o you feel that b e c a u s e of the time you s p e n d with • your s p o u s e you don't have enough time for yourself?  3.  D o you feel stressed between caring for your s p o u s e a n d trying to meet other responsibilities for your ° family or w o r k ?  4.  Do you feel e m b a r r a s s e d over your spouse's behaviour?  5.  Do you feel angry when you are around your s p o u s e ? °  6.  D o y o u feel that your s p o u s e currently affects your relationship with other family m e m b e r s or friends in a negative w a y ?  7.  A r e y o u afraid of what the future holds for your spouse?  '  8.  Do you feel your s p o u s e is dependent upon you?  °  9.  Do you feel strained when you are around your spouse?  '  10.  Do y o u feel your health has suffered b e c a u s e of your involvement with your s p o u s e ?  '  11.  Do you feel that you don't have as much privacy as y o u would like b e c a u s e of your s p o u s e ?  °  5 Nearly Always  2  3  4  5  •  •  •  •  °  °  °  1 38  Appendix Three (Cont.) Burden  1  2  3  Never  Rarely  Sometimes  Interview  4 Quite  5  Frequently  Nearly  Always  1 2. Do you feel that your social life has suffered because you are caring for you spouse? 13. Do you feel uncomfortable about having friends over because of your spouse?  °  °.  14. Do you feel that your spouse seems to expect you to take care of her/him as if you were the only one s/he could depend on? 15. Do you feel that you don't have enough money to care for your spouse in addition to the rest of your expenses? 16. Do you feel that you will be unable to take care of him/her much longer? 17. Do you feel you have lost control of your life since your spouse's illness? 1 8. Do you feel you could just leave the care of your spouse to someone else? 19. Do you feel uncertain about what to do about your spouse? 20. Do you feel you should be doing more for your spouse? 21. Do you feei you could do a better job in caring for your spouse? 22. Overall, how burdened do you feel in caring for your spouse? 1. Not at all 2. A little 3. Moderately 4. Quite a bit  5. Extremely  139  Appendix Four Geriatric Depression  Scale  1.  Are you basically satisfied with your life?  Yes  No  2.  Have you dropped many of your activities and interests?  Yes  No  3.  Do you feel that your life is empty?  Yes  No  4.  Do you often get bored?  Yes  No  5.  Are you hopeful about the future?  Yes  No  6.  Are you bothered by thoughts you can't get out of your head?  Yes  No  7.  Are you in good spirits most of the time?  Yes  No  8.  Are you afraid something bad is going to happen to you?  Yes  No  9.  Do you feel happy most of the time?  Yes  No  10. Do you often feel helpless?  Yes  No  1 1. Do you often get restless and fidgety?  Yes  No  Yes  No  13. Do you frequently worry about the future?  Yes  No  14. Do you feel you have more problems with memory than most?  Yes  No  15. Do you think it is wonderful to be alive now?  Yes  No  16. Do you often feel downhearted and blue?  Yes  No  17. Do you feel pretty worthless the way you are now?  Yes  No  18. Do you worry a lot about the past?  Yes  No  19. Do you find life very exciting?  Yes  No  20. Is it hard for you to get started on new projects?  Yes  No  21 . Do you feel full of energy?  Yes  No  22. Do you feel that your situation is hopeless?  Yes  No  23. Do you think that most people are better off than you are?  Yes  No  24. Do you frequently get upset about little things?  Yes  No  25. Do you frequently feel like crying?  Yes  No  26. Do you have trouble concentrating?  Yes  No  27. Do you enjoy getting up in the morning?  Yes  No  28. Do you prefer to avoid social gatherings?  Yes  No  29. Is it easy for you to make decisions?  Yes  No  30. Is your mind as clear as it used to be?  Yes  No  12. Do you prefer to stay at home rather than going out and doing new things?  140  Appendix Five Beck Hopelessness Scale  1.  I look forward to the future with hope and enthusiasm  T  F  2.  I might as well give up because there is nothing I can do T  F  they cannot stay that way forever  T  F  4.  I can't imagine what my life will be like in ten years  T  F  5.  I have enough time to accomplish the things I want to do  T  F  6.  In the future, I expect to succeed in what concerns me most  T  F  7.  My future seems dark to me  T  F  8.  I happen to be particularly lucky, and I expect to get more of the good things in life than the ordinary person  T  F  9.  I just can't get the breaks, and there's no reason I will in the future  T  F  10.  My past experiences have prepared me well for the future  T  F  11.  All I can see ahead of me is unpleasantness rather than pleasantness  T  F  12.  I don't expect to get what I really want  T  F  13.  When I look ahead to the future, I expect that I will be happier than I am now  T  F  14.  Things just don't work out the way I want them to  T  F  15.  I have great faith in the future  T  F  16.  I never get what I want, so it's foolish to want anything  T  F  17.  It's very unlikely that I will get any real satisfaction in the future  T  F  18.  The future seems vague and uncertain to me  T  F  19.  I can look forward to more good times than bad times  T  F  20.  There's no use in really trying to get anything I want T  F  about making things better for myself 3.  When things are going badly, I am helped by knowing that  because I probably won't get it  141  Appendix Six Attributional Style Questionnaire  -  Revised  You have been attempting to repair something unsuccessfully for some time  1.  Write down one major cause  2.  Is the cause of your unsuccessful job due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  3.  1  2  3  4  5  6  7  Totally due to me  In the future when making a repair, would this cause again be present? (Circle one number) Will never again be present  1  2  3  4  5  6  7  Will always be present  Is the cause something that just influences making this repair or would it also influence other areas of your life? (Circle one number) Influences all situations in my live  Influences just this particular situation 5a.  How important would this situation be if it happened to you? (Circle one number) Not at all important  5b.  1  Extremely important  How certain are you that this would be the cause of your unsuccessful repair job? (Circle one number)  Not at all certain  1  2  3  4  5  6  7  Extremely certain  142  Appendix Six (Cont.) Attributional  Style Questionnaire - Revised  Your income level significantly decreases  6.  Write down one major cause  7.  Is the cause of your lower income due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  8.  4  5  6  7  Totally due to me  1  2  3  4  5  6  7  Will always be present  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  10b.  3  Is the cause something that just affects loosing money or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  10a.  2  In your financial future, would this cause again be present? (Circle one number) Will never again be present  9.  1  1  2  3  4  5  6  7  Extremely important  How certain are you that this would be the cause of your lowered income? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  143  Appendix Six (Cont.) Attributional  Style Questionnaire - Revised  A friend comes to you with a problem and you don't try to help  11.  Write down one major cause  12.  Is the cause of your not helping your friend due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  13.  4  5  6  7  Totally due to me  1  2  3  4  5  6  7  Will always be present  1  2  3  4  5  Influences all situations in my live  6  How important would this situation be if it happened to you? (Circle one number) Not at all important  15b.  3  Is the cause something that just occurs when a friend comes to you with a problem or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  15a.  2  In the future when a friend comes to you with a problem, would this cause again be present? (Circle one number) Will never again be present  14.  1  1  2  3  4  5  Extremely important  6  How certain are you that this would be the cause for not helping your friend? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  144  Appendix Six (Cont.) Attributional Style Questionnaire - Revised  You speak in a social setting and others react negatively  16.  Write down one major  17.  Is the cause of this negative reaction due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  18.  4  5  6  7  Totally due to me  1  2  3  4  5  6  7  Will always be present  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  20b.  3  Is the cause something that just influences speaking in social settings or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  20a.  2  .  In the future when speaking in social settings, would this cause again be present? (Circle one number) Will never again be present  19.  1  cause  1  2  3  4  5  6  7  Extremely important  How certain are you that this would be the cause of others' negative reaction? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  145  Appendix Six (Cont.) Attributional  Style Questionnaire - Revised  Y o u do a project which is received very badly  21.  Write down one major  22.  Is the cause of this reaction due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  23.  2  3  4  5  6  7  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  25b.  Will always be present  Is the cause something that just affect doing projects or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  25a.  Totally due to me  In the future when doing a project, would this cause again be present? (Circle one number) Will never again be present  24.  1  cause  1  Extremely important  How certain are you that this would be the cause of your unsuccessful repair job? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  146  Appendix Six (Cont.) Attributional Style Questionnaire - Revised  Y o u meet a friend who acts hostilely toward you  26.  Write down one  27.  Is the cause of this reaction due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  28.  1  2  3  4  5  6  1  2  3  4  5  6  7  Will always be present  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  30b.  Totally due to me  Is the cause something that just influences interacting with friends or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  30a.  7  In the future when interacting with friends, would this cause again be present? (Circle one number) Will never again be present  29.  major cause  1  Extremely important  How certain are you that this would be the cause of your friends hostile reaction? (Circle one number) Not at all certain  1  Extremely certain  147  Appendix Six (Cont.) Attributional  Style Questionnaire - Revised  You can't get all the work done that others expect of you  31.  Write  down one major  32.  Is the cause of your not getting the work done due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  33.  2  3  4  5  6  7  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  35b.  Will always be present  Is the cause something that just affect doing work others expect or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  35a.  Totally due to me  In the future when doing work others expect, would this cause again be present? (Circle one number) Will never again be present  34.  1  cause  Extremely important  1  How certain are you that this would be the cause of your not getting the work done? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  148  Appendix Six (Cont.) Attributional  You have members  experienced  Style  Questionnaire  increased  difficulty  -  Revised  relating  to  other  family  36.  Write down one major  37.  Is the cause of this difficulty with your relatives due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  38.  2  3  4  5  6  7  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  40b.  Will always be present  Is the cause something that just affects how your relatives interact with you or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  40a.  Totally due to me  In the future when interacting with relatives, would this cause again be present? (Circle one number) Will never again be present  39.  1  cause  1  Extremely important  How certain are you that this would be the cause of difficulty between you and your family members? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  149  Appendix Six (Cont.) Attributional Style Questionnaire - Revised  You apply for a position you want badly (eg., important don't get it  job)  and you  41.  Write down one major  42.  Is the cause of your not getting the position due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  43.  4  5  6  7  Totally due to me  1  2  3  4  5  6  7  Will always be present  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  45b.  3  Is the cause something that just applying for a position or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  45a.  2  In the future when applying for a position, would this cause again be present? (Circle one number) Will never again be present  44.  1  cause  1  2  3  4  5  6  7  Extremely important  How certain are you that this would be the reason you did not get the position? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  150  Appendix Six (Cont.) Attributional  Style Questionnaire - Revised  You must change your residence to a less favourable location  46.  Write down one major cause  47.  Is the cause of your moving due to something about you or something about other people or circumstance? (Circle one number) Totally due to other people or circumstance  48.  2  3  4  5  6  7  1  2  3  4  5  6  7  Influences all situations in my live  How important would this situation be if it happened to you? (Circle one number) Not at all important  50b.  Will always be present  Is the cause something that just affects this change of residence or would it also influence other areas of your life? (Circle one number) Influences just this particular situation  50a.  Totally due to me  In the future, would this cause again be present? (Circle one number) Will never again be present  49.  1  1  2  3  4  5  6  7  Extremely important  How certain are you that this would be the cause of your moving? (Circle one number) Not at all certain  1  2  3  4  5  6  7  Extremely certain  151  Appendix Seven Demographic Questionnaire  Your  Gender  Partner's  Age  Gender  Number of Years  Age  Married  Number of Previous Marriages:  Self  Spouse  Prior to onset of your partner's present health concerns, how would you rate the quality of your relationship:  L V e r y poor 2.Somewhat Poor 3.Poor 4.Satisfactory 5.Good 6.Very good 7. Excellent  How would you rate your current health: L V e r y poor 2.Somewhat Poor 3.Poor 4.Satisfactory 5.Good 6.Very good  7.Excellent  Have you been treated in the past (or are you now being treated) for depression or a related condition? Yes  No  If yes, are you currently receiving treatment  (specify)  If yes, are you currently taking anti-depressant medication: Name of medication:  Dosage:  Frequency:  Duration:  Prescribing  Physician and  Number:  What is (or do you have) a religious affiliation? Self Spouse  152  Appendix Seven (Cont.) Demographic Questionnaire  H o w often have you attended religious services in the past year (if at all) Self Spouse  How would you best describe your current residence: • Urban  • Rural  H o w would you best describe your ethnicity or ancestry: Self Spouse  W h a t are (were) your work or occupations: Self Spouse Your If  current  retired,  employment  number  status  of y e a r s  Total family income (all sources) for the past year: 9,999  $ 50,000 - 59,999  $ 10,000 - 19,999  $ 60,000 - 69,999  $ 20,000 - 29,999  $ 70,000 - 79,999  $ 30,000 - 39,999  $ 80,000 - 99,999  $ 40,000 - 49,999  $ 100,000 +  $  0 -  W o u l d y o u be willing to be contacted in future to answer additional question? • Yes  • No  153  Appendix Eight Marlowe-Crowne  Social Desirability Scale  Listed below are a number of statements concerning personal attitudes and traits. Listen to each item and decide whether the statement is true or false as it pertains to you personally.  1.  Before voting, I thoroughly investigate the qualifications of all candidates  T  F  2.  I never hesitate to go out of my way to help someone in trouble  T  F  3.  It is sometimes hard for me to go on with my work if I am not encouraged  T  F  4.  I have never intensely disliked anyone  T  F  5.  On occasion I have doubts about my ability to succeed in life  T  F  6.  I sometimes feel resentful when I don't get my own way  T  F  7.  I am always careful about my manner of dress  T  F  8.  My table manners at home are as good as when I eat out in a restaurant  T  F  9.  If I could get into a movie without paying and be sure I was not seen T  F  T  F  T  F  T  F  1 3. No matter who I'm talking to, I'm always a good listener  T  F  14. I can remember playing sick to get out of something  T  F  1 5. There have been occasions when I took advantage of someone  T  F  16. I'm always willing to admit when I make a mistake  T  F  I would probably do it 10. On a few occasions, I have given up doing something because I thought too little of my ability 1 1 . 1 like to gossip at times 12. There have been times when I felt like rebelling against people in authority even though I knew they were right  1 54  Appendix Eight (Cont.) Marlowe-Crowne  Social Desirability Scale  17. I always try to practice what I preach  T  F  T  F  19. I sometimes try to get even rather than forgive and forget  T  F  2 0 . When I don't know something, I don't at all mind admitting it  T  F  2 1 . I am always courteous, even to people who are disagreeable  T  F  22. At times, I have really insisted on having things my own way  T  F  2 3 . There have been occasions when I felt like smashing things  T  F  T  F  T  F  T  F  T  F  T  F  2 9 . I have almost never felt the urge to tell someone off  T  F  3 0 . I am sometimes irritated by people who ask favours of me  T  F  3 1 . 1 have never felt that I was punished without cause  T  F  T  F  T  F  18. I don't find it particularly difficult to get along with loud-mouthed, obnoxious people  24. I would never think of letting someone else be punished for my wrong-doings 2 5 . I never resent being asked to return a favour 2 6 . I have never been irked when people expressed ideas very different from my own 2 7 . I never make a long trip without checking the safety of my car 28. There have been times when I was quite jealous of the good fortune of others  32. I sometimes think when people have a misfortune they only got what they deserved 3 3 . I have never deliberately said something that hurt someone's feelings  1 55  Appendix Nine Edmonds Marital Conventionality Scale  1.  There are times when my spouse does things that make me unhappy.  T  F  2.  M y marriage has not been a perfect s u c c e s s .  T  F  3.  Thoughts of separation or divorce occur to me very frequently, as often as once a week or more.  T  F  4.  M y s p o u s e has all the qualities I ever wanted in a mate.  T  F  5.  If my s p o u s e has any faults, I a m not aware of them.  T  F  6.  M y s p o u s e and I understand each other perfectly.  T  F  7.  W e are as well adjusted as any two people in the world c a n be.  T  F  8.  I have s o m e needs that are not being met by my marriage.  T  F  9.  I have d i s c u s s e d separation or divorce with my s p o u s e recently.  T  F  1 0. 11.  Every new thing I have learned about my s p o u s e has pleased me.  T  F  T h e r e are times when I don't feel a great deal of love or affection for my s p o u s e .  T  F  12.  I have filled for, and a m proceeding with, legal separation or divorce.  T  F  13.  I don't think anyone could possibly be happier than my s p o u s e a n d I w h e n we are with one another.  T  F  14.  M y marriage could be happier than it is.  T  F  15.  I don't think any couple could live together with greater harmony than my s p o u s e and I.  T  F  16.  M y s p o u s e completely understands and sympathizes with my every m o o d .  T  F  17.  I have never regretted my marriage, not even for a moment.  T  F  18.  I have contacted a lawyer recently regarding divorce.  T  F  19.  If every person in the world of the opposite sex had been available a n d willing to marry me, I could not have made a better c h o i c e .  T  F  156  Appendix Ten Concurrent Validity of the Geriatric Depression Scale  During  the  neuropsychological assessment,  administered  the  Multi-Focus  Assessment Scale  Crockett, Holliday & Koch, 1985). depression  measure  with  (euthymic to depressed). a  technician  or  patients (MAS;  are  Coval,  This measure contains a 14 point  scores  extending  from  +14  to  -14  This instrument is orally administered by  psychology  intern  trained  to  administer  this  measure as well as all others in the neuropsychology battery. This  interview  is separate  from that in which  patients  are  orally administered the Geriatric Depression Scale (GDS; Yesavage et al., 1983). or  two  This interview generally occurs either two days prior  days  appointment. member  of  subsequent  to  the  patient's  In addition, this interview clinic  staff  and  both  neuropsychology  is conducted by another  interviewers  are  most  often  unaware of the patient's responses to the other depression measure. Neither staff member was aware that comparisons were to be made between the MAS-Mood and the GDS. Among  the  patients  whose  study, the correlation between  caregiver  participated  measures is -.68  in  (p < .001).  this This  would suggest adequate concurrent validity for the G D S .  It should  be  as  noted  that  this  sample  totaled  59  of  constraints did not allow for administration instance.  70  patients  of the  time  G D S in each  157  Appendix Eleven Orthogonal Contrasts Among Caregivers of Demented Spouses  As earlier  discussed, it was first proposed that  in this study be limited probable that  or possible Alzheimer disease.  this  criterion  diathesis-stress Though orthogonal style  participating  would  provide  diagnosed with  It was  the  initially  strongest  believed  test  of  this  model. reevaluated  contrasts  within  this  as  were  in this study, 41  basis of cut-off  discussed  computed  subgroup.  with a demented spouse. On the  to caregivers of patients  participation  As of the  to  earlier  assess  determined 70  in  the  the  text,  attributional  subsequent  caregivers currently  to live  This amounts to 59% of the total sample. scores on the  two  depression  measures  (GDS > 11; BHS > 9), these subjects were categorized as either hopeless and depressed (N = 3), depressed by not hopeless (N = 8) or neither hopeless nor depressed (N = 30).  No subjects fall into the  hopeless but not depressed grouping. As shown in Table significantly  different  11, globality  between  and stability  subjects who  present  levels  remain  as hopeless  and depressed (HD) as compared to both those neither hopeless nor depressed (-H-D). (D-H), however,  As compared to those depressed but not hopeless the differences  in attributional  style versus those  hopeless and depressed (HD) approaches significance.  158  Appendix Eleven (Cont.) Orthogonal Contrasts Among Caregivers of Demented Spouses  It  is  significantly  noteworthy between  that  those  construct  depressed  of but  internality not  differs  hopeless  (D-H)  relative to those neither hopeless nor depressed (-H-D). The  reasons for these differences from orthogonal  contrasts  with the full sample are unclear. This may be due solely to the smaller  sample.  159  Table 12. Orthogonal  Contrasts Comparing Subjects  Caring for a  Demented  Spouse (N = 40)  Grouping (N)  Stability  (SD)  Globality (SD)  Internality (SD)  H D  (3)  52.7 (7.51)  48.3 (3.06)  43.3 (10.7)  D-H  (7)  40.0 (9.70)  37.0 (8.52)  37.3 (6.99)  -H-D (30)  35.8 (9.73)  34.5 (9.80)  43.7 (5.75)  T-Value  P  Contrast 1:  1.91  0.06  Contrast 2:  -2.90  0.01  Contrast 3:  1.05  0.30  T-Value  P  Contrast 1  1.75  0.09  Contrast 2  -2.44  0.02  Contrast 3  0.63  0.53  STABILITY  GLOBALITY  T-Value  P  Contrast 1:  -1.39  0.17  Contrast 2:  0.10  0.94  Contrast 3:  -2.43  0.02  INTERNALITY  Contrast 1: Hopeless and d e p r e s s e d (HD) v s . depressed, not hopeless (D-H) Contrast 2: H o p e l e s s and d e p r e s s e d (HD) vs. neither hopeless nor d e p r e s s e d (-H-D) Contrast 3: D e p r e s s e d ,  not h o p e l e s s (D-H) v s . neither h o p e l e s s nor d e p r e s s e d (-H-D)  160  Appendix Twelve Balanced MANOVA Comparing Attributional Style Among Selected Caregivers  Multivariate  analysis  balanced design. persons  in  of  variance  each  multivariate  violations.  It  presumes  a  More specifically, this requires that the number of grouping  is  equal  (or  As discussed previously, Pillais' criterion of  ordinarily  significance is  uncertain,  is  more  however,  approximately  equal).  was selected as this test robust  to  it  procedure  this  assumption fully  compensated for the degree of imbalance among the full grouping of subjects. For this reason, the MANOVA was recalculated with an equal number of persons in each grouping. (HD)  As the hopeless and depressed  quadrant contains the fewest caregivers (N = 4), the MANOVA  was rerun with only 12 subjects.  Four subjects were chosen from  both the depressed but not hopeless (D-H)  and the neither hopeless  nor depressed (-H-D) groupings with the aid of a table of  random  numbers (Hopkins, Glass & Hopkins, 1987). As shown in Table 13, univariate for stability .017).  using the  stringent  alpha  significance remains level  earlier  evident  employed  (3  At this time, globality approaches significance (3 = .056)  multivariate significance is no longer apparent ( F =1.85, p = .17). 29  = yet  161  Appendix Twelve (Cont.) Balanced MANOVA Comparing Attributional Style Among Selected Caregivers  This analysis would appear supportive of the earlier findings. Considering the  small sample size derived for this calculation, it  would seem violation of the assumption of balanced cells has not significantly biased earlier  results.  The trend which emerges  this MANOVA is similar to earlier findings. degrees  of  freedom  (i.e.,  df  =  4-1)  are  from  Considering all available necessary to  compute  significance levels with three dependent variables, it is  remarkable  that  given  statistical  significance  markedly reduced sample size.  would  still  be  apparent  the  162  Table 13. Multivariate Analysis of Variance Attributional  Grouping  (MANOVA)  Comparing  Constructs Among Selected Caregivers  (N)  Globality  (SD) Stability  (SD)  Levels of  (N= 12)  Internality  (SD)  H D  (4)  50.3 (2.22)  55.0 (3.92)  39.8 (6.90)  D-H  (4)  35.5 (9.57)  34.5 (4.04)  39.8 (8.18)  -H-D  (4)  42.3 (8.06)  36.3 (13.7)  41.0 (5.72)  Univariate  Between  Group  F Tests with 2,9 df  Variable  Hypoth. S S  Error MS  F Value  Stability  1033.17  72.86  7.09  0.014  Globality  436.17  53.83  4.05  0.056  Internality  Multivariate  Test Hotelling's  49.06  4.17  T  Test  of  0.04  Significance  0.959  Significance  Value  Approx. F  df  Significance  1.85  1.85  6.0  0.17  163  Appendix Thirteen Factor Structure of the Edmonds Marital Conventionality Scale  As  noted  earlier,  current  suggests social desirability distinct  factors.  management),  In  wisdom within social psychology  responding may be composed of  contrast  to conscious distortion  self-deception  may  also  cause  two  (impression  respondents  to  underreport various beliefs and behaviours with limited awareness. Because of oral administration, should items in the Edmonds Scale (1967) fall  within these  two  groupings, it was  hypothesized that  the influence of this latter construct would be more pronounced. For analyzed  this for  reason, the this  Paulhus (1991).  study  factor  consistent  structure with  of  the  the  EMCS  previous  was  work  of  Using maximum likelihood extraction and varimax  rotation procedures, two separate factors were sought by retaining items with loadings above .40.  It should be noted that the items  from the Marital Status Inventory (Weiss & Cerreto, 1980), included to obfuscate the  intent of the  EMCS,  were excluded from  these  analyses. Initial support for the assumption of two primary constructs was provided by the scree test as there was a break in eigenvalues between the second and third factors (Tabachnick & Fidell, 1989). Using S P S S F A C T O R (SPSS Inc., 1988), two separate groupings of items emerged from this sample.  The  first  was  composed  of  six  164  Appendix Thirteen (Cont.) Factor Structure of the Edmonds Marital Conventionality Scale  items (3 = .80) and the second was composed of four (3 = .59).  Table  14 shows the respective loadings of all items on each factor. pattern  was  evident  in the  content  of  the  respective  No  items  to  suggest the nature or function of either factor. It is noteworthy that seven of 15 items load on both factors at .30  or greater.  Also, the correlation coefficient  resulting factors was strongly significant  between  (r = .48, p < .001).  the This  would suggest the construct(s) measured by the E M C S may be more cohesive as compared to standard social desirability measures such as  the  Marlowe-Crowne  indication, was  internal  markedly  (O'Rourke  et  al.,  1994).  As a  further  consistency as measured by Cronbach's alpha  higher for the  full scale of the  compared to the M C - S D S (3 = .77).  EMCS  (3  =  .86)  For all analyses within this  study, only the full scale of the E M C S was utilized.  1 6 5  Table 1 4 . Factor Loadings of Edmonds Scale Items on Factors One and Two *  Factor  One  Factor  E M C S  1  .05  E M C S  2  .42  .56  E M C S  4  .50  .15  E M C S  5  .26  .06  E M C S  6  .24  E M C S  7  .68  E M C S  8  .03  Two  .45  .40 .17 .78  E M C S 10  .58  .41  E M C S 11  .33  .31  E M C S 13  .78  .11  E M C S 14  .32  E M C S 15  .68  E M C S 16  .4  6  .36  E M C S 17  .34  .35  E M C S 19  .51  .32  .51 .17  Bold numbers indicate that the item loads significantly (> .40) on that factor only.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0054141/manifest

Comment

Related Items