ll!!!

 

 

77’ 77777777 777

1293198275

 

 

   

'77

 

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

thesis entitled

CASE MANAGEMENT: AN EVALUATION OF ITS
IMPACT ON THE PRIMARY CAREGIVER

presented by

SUSAN F . PAULSON

has been accepted towards fulfillment
of the requirements for

MASTERS degreein PSYCHOLOGY

 

 

 

Kajor professor

Date JULY 22, 1987

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

MSU

LIBRARIES

 

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

CASE MANAGEMENT: AN EVALUATION OF ITS
IMPACT ON THE PRIMARY CAREGIVER

By

Susan F. Paulson

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

1987

Copyright by
SUSAN FAYE PAULSON
1987

ABSTRACT

CASE MANAGEMENT: AN EVALUATION OF ITS
IMPACT ON THE PRIMARY CAREGIVER

BY

Susan F. Paulson

The present study was an attempt to assess the degree
to which case management has impacted the lives of the
primary caregivers of frail elderly individuals. The sample
consisted of 76 caregivers to frail elderly clients. A 20 to
30 minute telephone interview was administered which
assessed six areas of the caregiver’s life, which were as
follows: social involvements, psychological health,
physical health, the relationship between the caregiver and
the care-receiver, stress the caregiver may be experiencing,
and activities of caregiving. The case management group was
compared to a control group of primary caregivers whose
care-receiver received the information and referral
condition. No differences were detected between the
treatment and control groups. Step-wise regression
procedures were performed and four scales (social, physical
health, psychological health, and relationship) and two
variables were found to be significant predictors. The
results of this study indicate a need for future research to

address the needs of caregivers to the frail elderly.

In Memory of my Grandmothers,

Helmi Anna Eleanora Moyryla Paulson
and

Helena Elizabeth Hukkala Helbacka

"Friendship is the shadow
of the evening,
which strengthens
with the setting sun of life".

-LaFontaine

iv

ACKNOWLEDGEMENTS

Throughout the course of this project, several
individuals provided their time, energy. and support which
helped facilitate this finished product. A personal note of
thanks to my chair, Bill Davidson, who has provided me with
both the freedom and guidance needed to help cultivate my
professional and personal growth here at Michigan State.
Additionally, without the support of my other committee
members, Carol Mowbray and Ellen Strommen, this project
would not be completed. Both of them provided
exceptional technical and personal support prior to and
during the oral defense meeting which helped bring this
project and thesis to completion.

There are many individuals who have indirectly and
directly supported me throughout my professional development
whom I would like to acknowledge. To my dearest friend and
teacher, Grandmother Paulson, kiitos for being who you were.
I miss you so. To each and every member of my family, thank
you for your sisu! My dear friend Sandee Cargill, thank you
for your endearing friendship. A generous thank you goes
to: Marianne "Mern" Beutner, my friend and savant
conversationalist, for the many dreams you have helped me

formulate; Gwen Jolivette, for your unfaltering assistance

during my years at AT&T which lead me back to Michigan in
pursuit of my goals; members of my Thesis Support Group,
Susan "Yeich-oo" Yeich, Jennifer Stanley, and Nancy
McCrohan; De’ Bryant, for her understanding and "listening
ear"; my "family" of friends near and far who are always a
part of me; and Zennee the computer for her patient
temperament and her astounding capabilities!

To my kind friend, Leah "Genzy-doo" Gensheimer, whom
I am indebted not only for the technical advice she gave me
on this manuscript but also for her kindness during the
detours this project unintentionally took.

My "sister" D. Kay Taylor, Joe Bornstein, and Robin
Redner were all very helpful in preparing the early versions
of this manuscript. Thank you all for your ideas and
comments on the early drafts.

Several individuals have been instrumental in laying
the groundwork for my career over the years: The late
Dr. Jean Rutherford, Dr. John Renfrew, Dr. Haryln Hamm,

Dr. Pryse Deurfeldt, Dr. Howard Eisner, and Dr. W. Andrew
Smith. I feel not only thankful for their help but also
grateful for having had an opportunity to work with them.

Finally, thank you to the caregivers who participated
in this study and who gratuitously gave their time, I am in
awe by the spirit in which they live their lives and the
abiding love they share with the individuals they care for.

Thank you all for allowing me to learn from you.

vi

TABLE OF CONTENTS

LIST OF TABLES O O O O O O O 0 O O O O O O 0

INTRODUCTION . . . . . . . . . . . . . . . .

Research on Case Management . . . . . .
Research on the Effects of Caregiving .
Emotional and Physical Effects of
Caregiving. . . . . . . . . .

Types of Caregivers. . . . . . . .
Types of Caregiving Relationships.
General Criticisms of Research on
Caregiving. . . . . . . . . .

Social Support Networks and Health. . .
Background Research . . . . . . . . . .
Specific Research Goals . . . . . . . .
Hypotheses. . . . . . . . . . . . . . .

METHODS O O O O O O O O O O O O O O O O O O 0

Overview. . . . . . . . . . . . . . . .
Present Study . . . . . . . . . . . . .
Subjects . . . . . . . . . . . . .
Procedure. . . . . . . . . . . . .
Instrument . . . . . . . . . . . .
Interviewers . . . . . . . . . . .

Scale Construction . . . . . . . .
Social Scale. . . . . . . . .
Stress Scale. . . . . . . . .
Relationship Scale. . . . . .
Caregiving Scale. . . . .
Psychological Health Scale. .
Physical Health Scale . . . .

RESULTS 0 O O O O O O O I 0 O O O O O O O O O
Homogeneity of Variance . ... . . . . .
Multivariate Analysis of Variance . . .

HypothESiS one. 0 O O O O O I O O O 0 O
Hypothesis Two. . . . . . . . . . . . .

vii

Page

vii

10

10
12
14

17
18
23
29
30

32

32
34
34
38
41
46
47
48
51
51
51
52
52

57

58
58
60
6O

Hypothesis Three. . . . . . . . . . . . . . . . 61
Hypothesis Four . . . . . . . . . . . . . . . . 61
Hypothesis Five . . . . . . . . . . . . . . . . 61
Hypothesis Six. . . . . . . . . . . . . . . . . 62
Regression Analysis . . . . . . . . . . . . . . 62

DISCUSSION 0 O I 7. O O O O I O I O O O O O O O O 0 O 72

An Evaluation of the Treatment. . . . . . . . . 73
Methodological Considerations . . . . . . . . . 74
The Measuring Instrument . . . . . . . . . 74
The Prescreening Tool. . . . . . . . . . . 75
The Role Of the Caregivers. . . . . . . . . . . 75
Final Considerations. . . . . . . . . . . . . . 76

APPENDIX A . . . . . . . . . . . . . . . . . . . . . 81
APPENDIX B . . . . . . . . . . . . . . . . . . . . . 82
APPENDIX C . . . . . . . . . . . . . . . . . . . . . 84
APPENDIX D . . . . . . . . . . . . . . . . . . . . . 85
APPENDIX E . . . . . . . . . . . . . . . . . . . . . 87

REFERENCES 0 O O O O O O O O O O O O O I O O O O O O 102

viii

LIST OF TABLES

Table Page
1. Demographics . . . . . . . . . . . . . . . . 36
2. Factor Analysis Summary. . . . . . . . . . . 49
3. Factor Analysis Summary. . . . . . . . . . . 5O
4. Scale Reliability Summary. . . . . . . . . . 54
5. Pearson Product-Moment Correlations

Between Scales. . . . . . . . . . . . . 55
6. Univariate Homogeneity of Variance

Summary . . . . . . . . . . . . . . . . 59
7. Multivariate Summary . . . . . . . . . . . . 63
8. ANOVA Summary. . . . . . . . . . . . . . . . 63
9. Regression Summary . . . . . . . . . . . . . 67
10. Regression Summary . . . . . . . . . . . . . 67
11. Regression Summary . . . . . . . . . . . . . 7O
12. Regression Summary . . . . . . . . . . . . . 7O

ix

CHAPTER 1

Introduction

The notion of caregiving to aged parents is certainly
not new. For example, in the 1900’s, the care given to
ailing aged parents was usually done by their adult children
in their homes (Dahlin, 1980). Thus, the family bore full
responsibility for the care of incapacitated relatives.
Since the advent of nursing or convalescent homes and public
funding, an argument has ensued which alleges that the
elderly have been abandoned and dumped into institutions
(Dahlin, 1980; Lasch, 1977; Sussman, 1976). Two equally
persistent and pervasive contentions which have also been
generated are: formal support systems, which refer to
governmental and community services, provide the bulk.of
care received by the dependent elderly; and these services
undermine the family's sense of responsibility and encourage
the family to shirk those responsibilities (Brody, 1981).
Brody (1981) has referred to these arguments as "the myth of
family alienation". Shanas (1979) has made reference to
them as a "hydra-headed monster" due to their permeating
nature. Research over the past several decades has
invalidated these myths and consistently documented the
dominant role the family has in providing health care to
impaired noninstitutionalized elderly (Branch & Jette, 1983;

Brody, 1981; Brody, 1987; Cantor, 1983; Chatters, Taylor, &

2
Jackson, 1986; Kohen, 1983; Shanas, 1979; Solomon, 1982;
Tress, 1977; Ward, 1984).

Caregiving involves at least a two-person dyad, the
person receiving assistance, the care-receiver; and the
individual providing care, the caregiver (Cantor, 1983).

A primary caregiver is defined as someone principally
responsible for providing personal care and/or
coordinatining resources required by the impaired elderly
individual (Zarit, Reever, & Peterson, 1980). An example of
various types of services might include: household chores.
transportation, meal preparation, and financial assistance.

A primary caregiver is part of or may be the only
member of the informal support system which is available to
the care-receiver. Informal support systems are defined as
an elder’s spouse, living children, siblings, other
relatives, friends, and neighbors (U.S. Office of Management
and Budget, 1980).

Within the informal support system, two patterns of
familial network use have been recognized. First, a
hierarchical, compensatory model in which kin, especially
spouse and children, are preferred by care-receivers over
friends and neighbors as the cornerstone of the support
system (Cantor, 1983). When a crisis occurs, the elderly
will turn to the immediate family first, then to other
relatives, friends, neighbors, and finally formal supports
(Cantor, 1979). When family members are not available, the
role of friends and neighbors is more dominant. The

existence of a hierarchy has been documented by various

3
researchers studying caregiving (Chatters, Taylor, &
Jackson, 1985; Kulys & Tobin, 1980; Stack, 1974).

An alternative model proposed by Shanas (1979) is the
principle of substitution whereby general patterns of family
support are available to support members, in serial
progression. The individual selected for the primary
caregiver role is done so by the nature of the kin
relationship: spouses before children, children before
siblings, siblings before other extended family members.
This model emphasizes the primacy of spousal and filial
support and only when this is not available, will the older
member turn to distance kin and non-kin for support (Shanas,
1979). This model has been elaborated on by other
researchers (Brody, 1981; Chatters et al., 1985; Johnson,
1983; Johnson & Catalano, 1981; Ward, 1978).

Both the hierarchical and the substitution model place
caregiving responsibilities directly on immediate family
members. The use of non-kin and formal support systems is
generally less frequent by older individuals who are ill
(Cantor, 1979; Chatters et al., 1986; Johnson & Catalano,
1981).

Research on Case Management

Over the past decade, discussions on caregiving have
increased, but more research is needed on the consequences
of long-term care on the primary caregiver (Crossman,
London, & Barry, 1981; Marcus & Jaeger, 1984; Rakowski &
Clark, 1985; Townsend & Poulshock, 1986). At this time, two

questions arise which must be considered: how long can

4
caregivers carry the heavy responsibilities associated with
caregiving and what can be done to ease the burden of
caregiving. By directing our attention towards the latter
of the two, we may be able to eliminate the need to ask the
first question.

What appears to be the overwhelming consensus among
researchers is the need for more collaborative work between
informal and formal support networks for the frail elderly
(Biegel, Shore, & Gorden, 1984; Brody, 1981; Horowitz, 1985;
Johnson, 1983; Jones & Vetter, 1984; Rakowski & Clark, 1985;
Scott & Roberto, 1985; Stoller, 1982, 1983; Stoller & Earl,
1983). Clark and Rakowski (1983) and Zarit, Reever, and
Peterson (1980) stated that alternative ways to reduce
burden of home care have not been given sufficient
attention.

Case management is one alternative which provides a
supportive link between informal support networks and those
of the community. It is a service that assists clients in
identifying needs (physical, emotional, and economic),
explores solutions, and provides a connection between
informal and formal supports in order to achieve the highest
level of independence for the at-risk elderly (Steinberg &
Carter, 1983). Although components of case management
programs vary, the core characteristics typically shared by
all programs include:

1) Prescreening: preliminary assessment of potential

clients to determine eligibility.

2) Assessment/Reassessment: multidimensional

5
assessment (physical, mental, and social
evaluation) done at the initial phase of the
program with periodic reassessments done every 60
to 90 days to determine client progress and
highlight needed changes in service utilization.

3) Care Plan: a plan which documents needed
services. Elements of the plan include
information regarding the frequency, duration, and
goals of services utilized by clients. It also
documents meetings with the informal support
system and various consultations with specialists
and/or other members of the formal support system.

4) Coordination: contacting and contracting with a
variety of providers to arrange for care plan
implementation.

5) Monitoring: maintaining ongoing contact with the
client between scheduled assessments to check on
client progress and satisfaction. In addition,
contact is also made with service providers to
assure that the services prescribed in the care
plan are being delivered.

A case manager is usually a registered nurse or a social

worker and may be working individually or as part of a team
comprised of several nurses and/or social workers. The

primary objective of case management is to maintain the at-risk
elderly person in his/her own home through coordination of
in—home and community-based services, thus reducing

premature institutionalization.

6

Institutionalization is considered premature when an
individual is placed in an institution despite the fact that
the individual could maintain him or herself in the
community if they had supplemental assistant (Ward, 1984).
Those individuals who cannot obtain special assistance or
additional health care services are more likely to be
admitted to an instititution (Wachtel, Fulton. & Goldfarb,
1987).

Institutionalization for the elderly has been referred
to as a life—or-death issue (Ward, 1984). Coinciding with
the growth of institutions has been the concern about the
impact institutionalization has on the elderly. A number of
studies have reported the consequences of being
institutionalized which include: depersonalization,
physical and other abuses, demoralization, relocation stress
or transfer trauma, as well as the financial costs (Kimmel,
1980; Shanas & Maddox, 1976; Ward, 1984; Williams, 1985).
Thus, the gravest decision that spouses, family members, and
long-term care providers face involve recommending
institutional placement.

The obvious benefits of case management for the formal
support system are: containing long-term care costs,
reducing service fragmentation, and eliminating service
duplication (Steinberg & Carter, 1983). But more
importantly, increased assistance from formal supports will
reduce the burden on informal caregivers (Brill & Horowitz,
1984; Callahan & Stanley, 1981; Kodner, Mossey, & Dapello,

1984).

7

With Medicaid funding cuts pushing greater financial
responsibility for care of the frail elderly on the family,
the increase in the old-old (age 75+) rising rapidly, and a
shortage of potential caregivers predicted in the future,
case management has been looked upon as a viable solution to
a possible crisis situation (Callahan & Stanley, 1981).
Although the literature on case management is sparse, what
does exist presents very promising outcomes.

Brill and Horowitz (1984) have assisted in implementing
a project in New York City called Home Care Project (HCP).
The project provides home care and other maintenance level
services to homebound. frail elderly in four selected areas
of New York City. Longitudinal data are being collected
from a research sample of 504 HCP clients and 200 matched
comparisons.

Preliminary results show a decrease in the length of
stay per hospitalization by 29% and overall positive.
satisfaction with services provided expressed by caregivers.
Kodner et al. (1984) reports similar results with another
project in New York City called "Nursing Homes Without
Walls". Clients and their families appear almost
universally satisfied with the project. Data collected in
the client interview survey indicate that the project has
helped to foster an increase in informal supports and
caregiving by family and friends.

Triage, a project directed to meet the needs of the
frail elderly in Connecticut, was also an experiment in

changing the service delivery mechanism for long-term care

8
(Quinn & Hodgson, 1984). This seven-year project showed a
positive impact on caregivers through the use of counseling
and supportive services to alleviate confusion, despair, and
even crisis situations.

The South Carolina Community Long Term Care Project
(CLTC) began accepting client referrals in July of 1980
(Brown & Learner, 1984). The control group of this project
had 337 participants and the experimental group had 282
participants. The mean age of each group was 74.87 and
73.23 respectively. The majority of the clients (82%) were
at nursing home levels of care and women represented 69% of
the clients.

The preliminary findings in the CLTC study indicated
that the experimental intervention, case management, was
successful in providing community-based care for the frail
elderly with chronic health conditions. The experimental
group used significantly less nursing home days at an
intermediate care facility (ICF) level than the control
group. For ICF controls, 42% of the participation days were
spent in nursing homes. In addition, among the patients who
were intitially at nursing home care levels, a greater
proportion of experimental patients remained in their homes
than did the controls. Similar results have been reported
with other projects by various researchers (Eggert &
Brodows, 1984; Jette, Branch, Wentzel, Carney, Dennis, &
Heist, 1981; Skellie, Favor, Tudor, & Strauss, 1984; Weiss &

Sklar, 1984).

9

These findings, although optimistic, should be
interpreted with some caution due to serious methodological
limitations in some of the studies. For example, Eggert and
Brodows (1984) used a nonequivalent control group design to
analyze their data which was collected over a three year
period. In addition, due to differences in research
designs, data collection systems, and the populations
served, generalizability from project to project is
difficult.

For instance, none of the studies on case management
randomly selected their participants from a general
population, which places a limitation on external validity
(Campbell & Stanley, 1968). Additionally, Brill and
Horowitz (1984) evaluated their program impacts utilizing a
quasi-experimental research design with no randomization.
The researchers did attempt to use a matched comparison
sample of 200 individuals, but their design was weakened by
the extreme levels of disability found with in the treatment
group as compared to the individuals in the comparison group
who were significantly less impaired.

However, several researchers (Brown & Learner, 1984;
Skellie et al., 1984; Weiss & Sklar, 1984) did utilize an
experimental design with randomization of participants to
experimental and control groups.

Clearly what is needed in this area is research which
adheres to rigorous experimental methodology. Only then can
the effects of the experimental innovation be accurately

evaluated against what presently exists in the system

10
(Fairweather & Tornatzky, 1977). Finally, once this type of
research design has been incorporated into the project, then
the magnitude of the impact of case management on
alleviating caregiver burden can then be assessed.
Research on the:§ffect§¥of Caregiving

Along with the mounting evidence which supports the
active involvement of the family in caregiving to the
elderly, there is a growing awareness of the demands, risks,
and costs associated with caregiving. These demands, risks,
and costs have been labeled the "burden" of caregiving which
the informal support network must cope with (Kraus, 1984;
Poulshock & Deimling, 1984; Rokowski & Clark; 1985; Ward,
1984; Zarit et al., 1980). The toll of caregiving to the
old-old, often referred to as the frail elderly when ill,
is high in terms of physical, psychological, and financial
resources (Clark & Rakowski, 1983; Lowenthal, Berkman, &
Associates, 1967). Research has shown that family members
go to great lengths to avoid institutionalization of frail,
homebound elderly family members even to the detriment of
their own health (Brody, 1981; Deimling & Bass, 1986;
Rabowski & Clark, 1985; Shanas & Maddox, 1976).
Emotional and Physical Effects of Caregiving

Farkas (1980) examined the degree of burden associated
with caring for a chronically ill spouse and reported the
spousal caregivers often described themselves as physically
and emotionally drained. Fengler and Goodrich (1979)
reported loneliness, a sense of isolation, and role overload

experienced by wives of elderly disabled men. Similar

11
findings were reported by Crossman et al. (1981) while
utilizing a Wives Support Group for the spouses caring for
disabled men.

Those comprising the informal support network have
expressed feelings of guilt and anger (Crossman et al.,
1981; York & Calsyn, 1977) towards the care-receiver and
other family members who are not involved with caregiving
responsibilities. Additionally, caregivers often experience
low morale and depression (Crossman et al., 1981; Fengler &
Goodrich, 1979; Poulshock & Deimling, 1984). Family members
and those making up the informal support network have been
known to make personal sacrifices even to the point of
giving up their jobs to care for the frail, elderly relative
in need (Brocklehurst, Morris, Andrews, Richards, & Laycock,
1981; Robinson & Thurnher, 1979).

Several studies (Cantor, 1983; Fengler & Goodrich,
1979; Smith & Bengston, 1979) report that caregiving may be
7accompanied by a suspension or even abandonment of one’s
future plans. Rakowski and Clark (1985) investigated future
outlook by contacting the care-receivers and primary
caregivers and found restricted future outlook for both
groups to be associated with situations of greater
impairment, more extensive assistance, and greater stress.
Although the authors report a cause-effect association
cannot be identified due to cross-sectional data, they do
hypothesize that a limited future outlook may in turn

exacerbate feelings of stress.

12
Types of Carggivers

Mcauley, Jacobs, and Carr (1984) studied patterns of
caregiving assistance in 389 older married couples. The
couples were interviewed and asked if they had acted as a
caregiver to anyone within the past six months, and if so,
to describe the relationship between themselves and the
care-receiver. It was found that wives are more likely to
provide assistance outside of the marital relationship,
whereas men are more likely to help their spouse.

Cantor (1983) researched the factors associated with
stress among four types of informal caregivers: 33% were
spouses (49% women), 36% were children (75% women), 19% were
other relatives (86% women), and 12% were friends and
neighbors (92%‘women). The results indicated that the
amount of stress and disruption of daily lives of the
caregiver is different for different groups of caregivers.
Closer bonds make caregiving more stressful and the two
groups of caregivers at highest risk of stress are spouses
and children.

Virtually all of the literature substantiates the fact
that women predominate as providers of direct services to
the frail elderly (Brody, 1981; Cantor, 1983; Crossman et
al., 1981; Horowitz, 1985; Johnson, 1983; Jones & Vetter,
1984; Perspective on Aging, 1985; Rakowski & Clark, 1985;
Shanas, 1979; Stoller, 1982, 1983; Treas, 1977). However,
the data do not reflect the lack of responsibility on the
part of men as caregivers; rather, they represent the
cultural assignment of gender-appropriate roles (Brody,

1981).

13

The sex of the adult child is one of the most important
and consistent predictors of caregiving involvement
(Horowitz, 1985). But due to increased involvement in the
labor market by women, society may see different patterns of
caregiving in the future as well as more equally distributed
responsibility among male and female adult children and
siblings (Horowitz, 1985). At this time, research supports
the fact that adult daughters report considerably greater
stress than adult sons when caring for an older parent
(Horowitz, 1985). Numerous researchers have found this
relationship between gender and caregiving stress (Brody,
1985; Cantor, 1983; Cicirelli, 1981; Johnson, 1983; Marcus &
Jaeger, 1984; Robinson & Thurnher, 1979; Scharlach, 1987).
Robinson and Thurnher (1979) have hypothesized that because
of the more intimate emotional relationship which often
exists between mothers and daughters, greater stress
results. This hypothesis is strengthened by the research
done by Cantor (1983) and the fact that elderly women are
more likely than men to be in the position of the care—
receiver over the age of 75 (Ward, 1984; Wingard, Jones, &
Kaplan, 1987).

The association between female caregivers and greater
reported stress is equally likely to be explained by what
Brody (1981) has named "woman in the middle". Such women
are in middle age, are middle generation, and are in the
middle of competing demands on their time and energy. Added
to their traditional roles as wives, homemakers, mothers,

and grandmothers are roles such as paid workers, caregiving

14
daughters, and daughter-in-laws. These multiple demands may
explain why the experience is more stressful for women.

Bowling (1984) supports the work done by Brody in a
study of the lives of 213 relatives who were caregivers to
elderly widowed people. Sixty-nine percent of these
caregivers were children of the widowed person and of these,
46% were daughters and 33% were sons. It was reported that
daughters were more likely to suffer life restrictions
before and after the death because of the care and support
given to the deceased and surviving spouse. These women
were also committed to full and part time work and most had
their own families to support.

Recently, it has been assumed that women who are
employed outside of the home reduce the amount of time spent
as caregivers. Research on this issue has found that
employment outside of the home is not a deterrent to the
level of involvement by daughters as caregivers (Brody,
Kleban, Johnson, Hoffman, & Schoonover, 1987). Instead of
reducing the amount of time as caregivers, daughters give up
their own free time and opportunities for socializing. A
recent study reported that regardless of their work status
(working or nonworking) daughters provided the most help to
their dependent mothers (Brody & Schoonover, 1986).

Types of Caregiving Relationships

The degree of burden felt by caregivers is also
affected by the relationship which exists between the
caregiver and care-receiver (Kulys & Tobin, 1980). The

long-term quality of their relationship, the degree of

15
physical impairment of the care-receiver, and the individual
personalities combine to affect the degree of burden
experienced (Biegel et al., 1984).

Kulys and Tobin (1980) also report that a close
affectionate bond between a child caregiver and a cared-for-
parent is said to lessen the strain of caregiving.
Supporting the work done by Cantor (1983), Kulys and Tobin
(1980) also found that the intimate bonds established in a
spouse caregiving relationship tends to heighten stress and
perceived sense of burden. This finding may be explained by
a number of factors which are unique to the spousal
relationship and increase stress (Cantor, 1983; Crossman et
al., 1981; Fitting, Rabins, Lucas, & Eastham, 1986; Gibson,
1982; Marcus & Jaeger, 1984). First, the spousal caregiver
often provides care to a severely impaired individual who
can no longer communicate verbally, thus depriving the
caregiver of the companionship and support once derived from
the confidant relationship they once enjoyed. Second, the
spouse’s sexual and affectional needs are frustrated due to
the loss of psychological, physical, and mental capacities
for interpersonal sensitivity and communication. Finally,
the spouse must also adjust to issues of role reversal as he
or she assumes both roles in order to maintain the
household. Therefore, in addition to the demands and costs
of caregiving experienced by caregivers, the spousal
caregiver must also cope with the loss of a lifelong
companion. Although the familiar body remains, the
capacities to nurture a mutually satisfying relationship

have deteriorated.

16

Zarit et al. (1980), while working with caregivers of
persons with senile dementia, reported stress was not
related to behavior problems of the care-receiver but
associated with the social supports available. As the
number of visits to the impaired elderly person increased
from family members, the degree of burden experienced by the
primary caregiver decreased. Marcus and Jaeger (1984) also
found a "high" burden score to be associated with fewer
visits to the household.

Gilleard, Gilleard, Gledhill, and Whittick (1984)
investigated the relationship between the sex of the
dependent person and the amount of stress and burden
reported by the caregiver. Interviews were conducted with
129 caregivers of an elderly mentally infirmed relative.

The results indicated that male dependents were perceived
as being more stressful to live with than female dependents.

Several researchers working with dementia patients
have drawn the connection between stress and the amount
of support given throughout the caregiving relationship
(Caserta, Lund, Wright, & Redburn, 1987; Zarit, Orr, &
Zarit, 1985; Zarit, Todd, & Zarit, 1986). Their findings
indicated that as the disease progressed, the dependence
upon family resources for care also increased to proportions
which are considered to be overwhelming. Thus, research has
focused on the effects of planned interventions such as
respite services and programs designed to improve coping
with everyday problems. Preliminary results indicated that

such programs do have a beneficial impact on caregivers by

17
reducing the stress associated with caregiving (George &
Gwyther, 1986; Zarit el al., 1986). But knowledge and
access to services does not always lead to utilization
because the presence of other informal support reduces the
impact of providing home care to a level that the caregiver
considers manageable (Caserta et al., 1987). The absence of
additional informal support and the caregiver’s feeling that
he or she simply could not continue to provide 24-hour care
needed are two major factors leading to institutional
placement (Chenoweth & Spencer, 1986).
General Criticisms of Research on Caregiving

A variety of methodological weaknesses have been cited
in the literature on caregiving (Clark & Rakowski, 1983).
Many of the published articles are small-scale surveys, case
studies, program reports, and narrative accounts of the
practitioner’s experience. Due to nonrandom sampling
procedures employed in many of the studies and small sample
sizes, the generalizability to populations outside of the
sample used is restricted.

Another shortcoming in the caregiving literature is the
range of definitions for burden; the effects of caregiving,
and its measurement (Poulshock & Deimling, 1984). For
example, the range of definitions for the concept burden
have ranged from role strain, emotional costs, financial
difficulties to disruption of the day—to-day life of the
caregiver. Thus there exists no unified definition of
burden and because burden has been defined and measured in

various ways, cross-study comparisons are very difficult to

18
make. Attempts to derive a specific measure from a concept
with varying definitions creates a lack of consistency in
the results making them difficult to interpret.

The effects of caregiving are detrimental to the
primary caregiver psychologically, physically, and
financially. The literature on caregiving has emphasized
the need to increase the available support systems of the
primary caregiver. Social support systems play a
significant role in the lives of primary caregivers.
Therefore, it is important to examine network theory as an
aid to understanding the functioning of informal and formal
social support networks and how they influence health. Due
to the nature and content of this review, a comprehensive
review of the material on social support networks will not
be presented. 'For an all-inclusive review on this topic,
the reader is referred to Leavy (1983).

Social Support Networks and Health

The past decade has seen the emergence of the term
social support network which has been defined as that set of
personal contacts through which the individual maintains his
or her social identity and receives emotional support,
material aid, services, new social contacts, and information
(Pilisuk & Parks, 1981). Investigators have hypothesized
about the link between an individual’s social support
network and his or her psychological state ever since Cassel
(1974) noted that altered social ties affect the body’s
defense system, so that people became more susceptible to

disease. The precise mechanism of action linking social

19
support and health is still unknown (Minkler, 1981). But
during the past 20 years, social support has been shown to
be a mediator or buffer against the harmful effects of life
events (Cobb, 1976; Gottlieb, 1981; Snow & Gordon, 1980;
Thoits, 1982; Tolsdorf, 1976).

The buffering hypothesis asserts that the support
provided by social networks increases coping ability.
Thus, individuals are better able to cope with major life
changes and this may increase the individuals resistance to
disease (Minkler, 1981). In turn, a person with little or
no social support network is more susceptible to negative
effects from life changes.

Luke, Norton, and Denbigh (1981) compared
psychologically impaired-to psychologically healthy
individuals. The sample consisted of 200 elderly
individuals, with an age range of 80 to 104 years.

Social isolation was reported markedly different in the
psychologically impaired group and the results indicated
that social interaction was the strongest predictor of
psychological well-being in a multiple regression analysis
that included physical health and socioeconomic variables.
This finding suggests that by increasing the social
interaction of primary caregivers, social isolation will
decrease which will have a positive effect on the
psychological well-being of the primary caregiver. This
finding has been found by various researchers (Chatters et
al., 1985; Marcus & Jaeger, 1984). For example, Zarit et'

a1. (1980) studied primary caregivers of elderly individuals

20
with senile dementia. When informal support was increased,
the amount of burden experienced by the primary caregivers
decreased.

An important influence on the literature concerning
social support has been the work done by Caplan (1974,
1976). Caplan places a premium on the collaborative nature
of the relationship between the professional and local
citizens in working with support systems and community
mental health. Focusing his research on how an individual
responds to crisis in his or her life, Caplan has concluded
that response to a crisis by an individual is dependent on
three factors:

1. the nature and variation of stress

2. the present ego strength of the individual

3. the value of emotional support and task-oriented

assistance which the individual’s social support
network is able to provide.
The results of the work done by Caplan indicated that the
last factor is the most important.

Caplan has put forth the concept of mobilization, which
refers to how potential supporters in a network become
supporters in a stressful situation. It is important not
only to examine the context in which mobilization takes
place but also to examine the forces which inhibit the
mobilization process from becoming activated. These forces
can take many forms which are classified into two
categories: subjective and objective, which are either

rooted in the individual or the support network.

21

Subjective factors include beliefs which an individual
may possess regarding when someone should be called on for
help and who the helper might be. Objective factors
include barriers affecting the ability to help someone,
such as physical distance or levels of demand on the
potential caregiver. Measuring levels of stress within the
support network allows for a more concise understanding of
the stress the individual is experiencing and gives one a
complete overview for analyzing social support processes
within the support network.

The vast majority of research findings indicate that
women have more supportive relationships than men. Elderly
women report larger, more intimate network ties than same-
aged men (Ward, 1984). They also are more likely to have
confidant relationships, which appear to be crucial for
adapting to old age (Lowenthal & Haven, 1968; Ward, 1984).
The components of social support for men and women also
differ between sexes. In a study of network characteristics
done by Phillips (1981), male’s happiness was best predicted
by the size of the social network, whereas women’s happiness
was best predicted by the number of social activities in
which they participated over the preceding three months.
Phillips (1981) suggested the traditional sex-role
stereotypes may explain these differences, with men more
connected to non-kin relationships and women to kin and
social activities. Other researchers have reported similar
findings regarding relationships to sex roles (Holahan &

Moos, 1981; House, 1981; Roy, 1978).

22

These findings may help to explain why women report a
greater degree of stress associated with caregiving.

Because of the sexual division of labor which exists in the
care of elderly individuals, women provide more "hands-on"
assistance such as personal care and household chores which
limits their ability to maintain social involvements outside
of the home (Brody et al., 1987; Horowitz, 1985; Stoller,
1983). Thus as the number of social involvements decrease,
social isolation increases and as noted earlier, an increase
in social isolation is associated with psychological
impairment (Luke et al., 1981). The male role in caregiving
is less involved with personal care tasks and tends to be
geared towards financial management and dealing with
bureaucratic organizations (Brody, 1981). Thus their social
network is not altered since their social support network is
traditionally connected with non-kin relationships
(Phillips, 1981).

An important point to consider in addition to the
amount and kind of support, is the quality of support. Not
all ties are supportive and a small proportion of those that
are provide significant support (Wellman, 1981). An overall
theme which needs to be considered is the fit between one’s
ideal type of support and what type of support is actually
available. Research indicates that individuals should take
steps to unify what exists in the informal and formal
networks and mobilize them in order to achieve the highest
quality of support available to thoSe in need of such

support (Porritt, 1979).

23

Problems in the literature on social support parallel
those problems found in the literature on caregiving. In
their review of this topic, Rabkin and Streuning (1976)
point out the psychometric weaknesses of measurement
instruments. Currently, there is no assessment instrument
which measures the components of social support with great
precision (Leavy, 1983). Most support questionnaires are
ad hoc measures with unknown reliability and validity. In
addition, a lack of agreement exists on how to define social
support.

In summary, social support networks can act as a buffer
which minimizes the negative effects of life stressors.
This increases coping ability which may increase an
individuals resistance to disease (Minkler, 1981). As a
result, increasing the informal and formal support networks
of the primary caregiver may decrease the burden associated
with caregiving.

At the present time, it is necessary to infix the
current literature, which has just been reviewed, into the
policy and demographic changes which are now taking place
which affect caregiving. As a result, the global picture
acquired will enable the reader to fully understand the
extent of the present and future problems associated with
caregiving to the elderly.

Background Research

Interaction with trusted others and supportive

relationships with family and friends is essential for

physical and emotional well-being throughout life (Ward,

24

1984). As we age, the importance of social support takes
on even greater significance due to the effects of aging
over time. The effects of aging and disease compound one
another (Kimmel, 1980). While disease may be defined as a
disorder of bodily functions that may occur within a broad
age range; aging involves changes that are caused solely by
the passage of time (Kimmel, 1980).

These changes may be described as a continuum of loss.
Minute changes begin to occur around thirty years of age
(Ward, 1984). But as aging continues, changes occurring in
body functioning, sensory functioning, mental functioning,
family and peer supports, income, self-image, self-esteem,
control, and power begin to accelerate (Biegel et al.,
1984).

If the changes an individual must cope with are
perceived as threatening, potentially harmful, unpleasant,
or overwhelming, they are said to be stressful. Stressful
changes place pressure or strain on a system due to an
imbalance which occurs between the perceived demand placed
upon the individual and the perceived response capability
(Ward, 1984).

Stress activates the body’s entire mental and
physiological systems, precipitating more than 1400
physiological changes. The physiologic response to stress
involves the homeostatic regulatory mechanisms of the body.
As aging continues, there is a decrease in the physiologic
adaptability of the older individual. Thus, the lowered

efficiency of the response mechanisms causes greater

25
sensitivity to stress (Biegel et al., 1984; Kastenbaum,
1969).

At the present time, deaths attributed to diseases
associated with stress account for over 75% of all deaths at
ages 65 and over (Ward, 1984). Because of the changes in
the leading causes of death; from infectious communicable
diseases (e.g. tuberculosis, pneumonia) to chronic
degenerative diseases (e.g. cancer, heart disease), there
has been a marked increase in life expectancy (Ward, 1984).
There have been only small increases in life expectancy at
age 65; the declines in infant mortality have contributed
most to the aging of the population. As life expectancy
increases and the fertility rate decreases, the proportion
of older individuals begins to rise. As a result, the
nation’s "fastest growing minority" are those 65 and older
(Brotman, 1974).

Over 95% of Americans age 65 and over reside in the
community, and even among those 80 and above, 7 of every 8
individuals live in their own homes (Shanas, 1979). About 8
to 10% of the noninstitutionalized elderly are as
functionally impaired as those in institutions (Brody,
1981). Overall estimates of the proportion of the
population who need supportive services hover about the one-
third mark, or about 8 million people (Brody, 1981). Eighty
percent of this long-term care is provided by family and
friends (Brody, 1987; Shanas, 1979).

The likelihood of institutionalization increases with

age. Only about 5% of persons age 65 and over are in

26
institutions, but this proportion rises from 1% of those 65
to 74 to 6% of those 75 to 84, and to 20% among those 85 and
over (Williams, 1985). Because men often marry women
younger than they are, they are more likely to have a spouse
to care for them. In addition, women have a longer life
expectancy than men (Wingard, et al., 1987), thus women
outnumber men in institutions in a ratio of about three to
one (Brody, 1987). The fastest growing age group
is comprised of the old-old; those aged 85 and above
(Manton, Blazar, & Woodbury, 1987).

The increased risk of institutionalization is reflected
in the dramatic increase in Medicaid expenditures in recent
years. The number of nursing home patients in the United
States has increased rapidly to 1.3 million in 1978 and is
projected to approach 2 million by the year 2000 (Schneider,
1983). Almost 40% of total Medicaid expenditures and about
75% of Medicaid expenditures for the elderly are spent on
nursing home care (Kane & Kane, 1981). The increase in
Medicaid expenditures has sparked a debate about increasing
the responsibility of the informal support network for the
care of noninstitutionalized elderly (Branch & Jette, 1983;
Brody, 1981; Callahan, Diamond, Giele, & Morris, 1980). An
untested assumption that providing home care can influence
the rate and timing of nursing home use has further fueled
this debate (Branch & Jette, 1983).

As the need for informal support networks and the
proportion of elderly who need home care continues to grow,

the amount in federal funds for formal programs is

27
decreasing (Stoller, 1983). At the same time, we are
experiencing a decline in the number of available caregivers
(Horowitz, 1985; Stoller, 1983; Stoller & Earl, 1983; Tress,
1977). There are several demographic trends which may be
responsible for this decrease in available caregivers.
First, an increasing proportion of women, ages 45 to 64, are
participating in the labor market. This age group is most
likely to provide caregiving to the elderly and now
comprises 60% of the labor force (Brody, 1981; Horowitz,
1985). Secondly, among the old-old there is a
disproportionate number of widows who must depend on someone
other than their spouse for caregiving assistance (Stoller,
1983; Tress, 1977). This group is less likely to have
living adult children due to declines in fertility rates
during the depression (Tress, 1977). Finally, the adult
women of this group; women in today’s middle generation, are
more likely to have married and have had children then their
predecessors, thus creating a shortage of "maiden aunts" to
devote themselves to aging parents and a reduction of the
pool of potential caregiving adult children (Brody, 1981;
Stoller, 1983; Tress, 1977). Combining these factors with a
higher life expectancy markedly changes the dependency
ratio. If these trends continue, there will be two older
people for every three working Americans in the next century
(Johnson & Catalano, 1981). In addition, the cumulation of
these factors creates a tremendous sense of responsibility
on present and future caregivers in meeting the needs of the

frail elderly individual (Cantor, 1983).

28

Although the family has an active role in the care of
their frail, elderly members, the toll of caregiving is high
physically, psychologically, and financially on the primary
caregiver. Thus, there is a growing need to identify ways
to alleviate the burden of caregiving experienced by the
caregiver. It is hypothesized that increasing informal and
formal social supports will decrease the burden associated
with caregiving. This investigation examined the
effectiveness of case management as a means of alleviating
the stress associated with caregiving to frail elderly

individuals.

29
Specific Research Goals

The present study was an attempt to assess the impact
case management had on the primary caregiver of the frail
elderly individual in comparison to traditional services
which were available in the community (i.e., information and
referral). The purpose was to assess whether case
management reduced the burden of the caregiver through the
incorporation of formal supports into the informal support
network.

The investigation was part of a larger research
project, which is being conducted to evaluate case
management in sites throughout Michigan. Five case
management projects have been implemented throughout the
state. The specific research objective is to evaluate case

management in terms of:

1. nursing home and acute care utilization
2. community service utilization patterns
3. health care costs

4. client life satisfaction/quality of life.

The project is a longitudinal study which employs a
true experimental design. This design should overcome some
of the methodological flaws present in prior studies.

The present study utilized a subset of three of the
five sites of the larger study in an attempt to assess
whether case management is effective in reducing the stress

associated with caregiving to the elderly.

30

Hypotheses

The following hypotheses were tested through use of a

telephone interview which was implemented after completion

of the twelve month follow-up interview from the larger

project.
E.
I.
II.
III.
IV.

The measurement instrument is located in Appendix

The quality of the relationship between the

primary caregiver in the case management group

will be significantly better than the comparable
relationships of caregivers in the information

and referral group.

The physical health status of the primary caregiver
in the case management group will be significantly
better than the primary caregiver in the information
and referral group.

The emotional/psychological health of the primary
caregiver in the case management group will be
significantly better than that of the primary
caregiver in the information and referral group as
measured by negative emotional reactions (anger,
frustration, helplessness, and guilt).

The social contacts of the primary caregiver in the
case management group will be significantly greater
than those of the primary caregiver in the
information and referral group as measured by
interactions with friends, relatives, children,

and other social activities not associated with

caregiving activities.

VI.

31
The perceived sense of burden of the primary
caregiver in the case management group will be
significantly lower than that of the caregiver
in the information and referral group.
The number of informal and formal supports of the
primary caregiver in the case management group
will be significantly greater than those of the
primary caregiver in the information and referral

group.

CHAPTER II

Methods

W

This study was part of a larger ongoing longitudinal
research project conducted at Michigan State University in
collaboration with the Michigan Office of Services to the
Aging (OSA) designed to evaluate case management with the

frail elderly. The project consists of five sites

throughout Michigan. Clients1 are referred to the case
management team from traditional sources. Such sources
include: hospitals, community organizations, nursing homes,

family members, friends, physicians, or themselves.
Although research intake ended in July 1986, the case
management program is still currently operating in the five
demonstration sites.

Once referred, the client was evaluated by a pre-
screening tool to determine whether he/she would be eligible

for the program. If eligible, the client constituted part

of the subject pool. Verbal and written consent was than
obtained prior to the initial baseline interview. The
1

The term client refers to the elderly individual of the
larger project in collaboration with OSA. The future use of
the term participant makes reference to the individuals who
participated in the caregiver study.

32

33

number of clients in the treatment and control groups were
328 and 324 respectively for a total of 652 clients.

Following the completion of the baseline interview,
random assignment was exercised to assign the client
to either the treatment or control group. The treatment
group received case management which consisted of: staff
members assessing client needs, creating a care plan and
helping the client arrange services. The control group
received information and referral which included providing
the client with written information and/or information over
telephone about how to obtain needed services. The client
was responsible for contacting and arranging the services
that he or she needed. These are services typically
provided in the client’s community.

A second and third follow-up interview was conducted
at six and twelve months after the initial interview. Data

collection procedures included self report and archival

records. The five dependent variables were:
1. facility and service utilization
2. costs (health and other support services)

3. client life satisfaction/quality of life

4. client health (includes mortality)

5. informal support
To ensure adequate data collection, inter-rater reliability
checks and retraining of interviewers was conducted
periodically throughout the three data collection periods by

the research evaluation team and the project managers.

34
Present Study
Subjects

Subjects for the study on caregiving were primary
caregivers of the frail elderly clients of the larger
project. Sampling occurred from September 1986 through
March 1987. A total of 101 people were contacted and of
these, 15 refused, 13 couldnot be contacted, and 6 were in
nursing homes. Thus, a total of 76 people completed the
interview. The sample consisted of individuals considered
to be the primary person providing instrumental and
emotional support to the frail elderly client. The primary
caregiver was nominated by the elderly client during the
twelve month follow-up interview.

All of the participants resided in the Grand Rapids,
Flint, and Southfield Michigan areas. Their mean age was
57.01 with a range of 23 to 93 years (See Table 1).
Consistent with the literature on caregiving, women
predominated as caregivers. Seventy-five percent of the
sample was women. Whites comprised 84.2% of the sample,
slightly over 10% of the sample was Black, and the
remaining 5% represented other races.

The majority of the participants were married (71.1%)
and stated that they had children (86.8%) with the most
frequently reported number of children being three (26.3%).
The income levels reported by the participants were well
distributed throughout the income scale (See Table 1).
Twenty—two point five percent of the participants reported

incomes between $5,000 to $9,999 and another 22.5% reported

35
incomes falling in the $30,000 or above category. In
addition, 21.1% fell within the $15,000 to $24,999 range.

Spouses (28.9%) and children (31.6%) dominated as
caregivers in this sample. Other relatives made up the
next largest percentage (18.4%). Somewhat surprising is
the low percentage of siblings who acted as caregivers,
only 1.3% of the sample.

A little over one-third of the participants were
employed (35.5%). This finding is somewhat low when
compared to other studies which report employment figures
over 60%. Of those that reported the number of hours
worked, the most frequently reported number of hours was
40. When asked how many months that the participant had
been working, 24 months was reported most often. However,
there was a wide variation in the number of hours
caregivers spent working as can be depicted by the range of
work hours (See Table 1). With respect to living status,
53.9% of the participants stated that they lived with the
individual they were caring for (See Table 1).

The participants were also divided into a treatment
group (case management) and a control group (information
and referral). Forty-one participants were in the case
management group. Thirty-five were in the information and
referral group. Case management provides for the
coordination of health care services into the informal
support system, by the case manager. Information and
referral refers to those services which are typically

available in the client’s community. Thus, this study

36
Table 1

Demographics

 

 

Demographic Adjusted Percent
Age: Mean 2 57.01
Range = 23 to 93
(n=73)
Sex: Female 75.0
Male 25.0
(n=76)
Race: White 84.2
Black 10.5
Other 5.3
(n=76)
Marital
Status: Married 71.1
Widowed 10.5
Divorced 13.2
Never.married 5.3
(n=76)
Children: Yes 86.8
No 13.2
(n=76)
Number of
Children: One 13.2
Two 18.4
Three 26.3
Four 10.5
Five or more 18.4
Missing 13.2
(n=76)
Income: 0 to 4,999 12.7
5,000 to 9,999 22.5
10,000 to 14,999 9.9
15,000 to 19,999 14.1
20,000 to 24,999 7.0
25,000 to 29,999 11.3
30,000 or above 22.5

(n

:71)

37

Table 1 con’d

 

Demograhic

Adjusted Percent

 

Relationship: Spouse
Child
Sibling
Other relative
Neighbor or friend
Paid helper

(n=76)
Work: Yes
No
(n=76)
Number of
Work Hours: Four to 20
25 to 35
40
42 to 60
Missing
(n=26)
Number of
Working months: one to 12
24 to 36
66 to 660
Missing
(n=74)
Living Status: Yes
No
(n=76)
Program Status: Case Management
Information and
Referral

(n=76)

38

compared the effects of a experimental intervention to
treatment as usual..
Procedure

During the twelve month follow-up interview associated
with the larger project, each elderly client was asked to
nominate the one person who he/she considered to provide the
most help and support throughout the day. (See Appendix A
for a copy of the question format). Once the primary
caregiver was identified, the client was presented a consent
form which was read and clarified by the interviewer. The
elderly client was then asked for permission to contact the
nominated primary caregiver. If the client agreed, the
client signed the consent form and the primary caregiver’s
address and phone number were obtained. (See Appendix B for
a copy of the consent form).

Once verbal and written permission were obtained from
the elderly client, the primary caregiver was then
contacted by letter, which informed him/her of the nature
of the study, how his/her name was obtained, and what
his/her involvement would entail if he/she agreed to
participate, i.e., would be called by the research staff in
approximately one week. (See Appendix C for the letter).
In addition, two copies of a consent form were sent; one
copy was for the caregiver and the second copy was for the
project’s records. A self-addressed stamped envelope was
enclosed for the purpose of the caregiver returning the
latter copy to the project staff. (See Appendix D for the

consent form).

39

Approximately one week after the letter was sent to the
caregiver, a telephone call was placed to the caregiver to
explain the study, and to see if the caregiver was
interested in participating. It was explained that written
permission to contact him/her was obtained from the elderly
client of the larger project. In addition to describing the
study to the caregiver, the time commitment and the nature
of the questions were specified so that the caregiver had a
complete understanding of what participation in the study
would entail. The caregiver was free to ask questions about
the study. The caregiver was also assured of anonymity,
confidentiality, and the right to stop at any time while the
interview was in process.

During this telephone conversation the caregiver was
asked if it was a convenient time to talk and if so, the
telephone interview was conducted. The interview took
approximately 45 minutes. If it was not a convenient time
for the caregiver, an interview day and time was set up and
the interview was conducted at that future time. In those
instances when the caregiver refused to participate, the
caregiver’s right to refuse was respected.

The interview began with a section entitled "opening
the interview". This section is a brief introduction
paraphrased by the interviewer which explains the
interviewing process to the caregiver. (See Appendix E).7

Interviewers maintained a call record on each of the
participants, which kept track of the date and time of each

contact, whether or not the interview was completed, if the

40
participant refused to participate, and a future appointment
date if necessary. The call record can be found on page 89
of the interview.

The participant identification sheet can be found on
page 90 of the interview. All pertinent information
necessary to contact the caregiver was compiled in this
section for future use. This sheet was a record of the
participant’s identification number, site location, the
interviewer, interview date, the participant’s name,
address, phone number, the elderly client’s status in the
program (case management or information and referral), the
beginning and ending time of the interview, and whether or
not the participant was currently living with the elderly
client. In addition, how the twelve month follow-up
interview from the larger project was completed (with a
proxy or the combination of the elderly client and a
significant other) was recorded along with the elderly
client’s name and address only if the information was
necessary.

When the interview was completed, the interviewer
followed the standard closing procedure specified in the
section entitled "closing the interview" found on page 100
in the interview. This was an important final step in the
interviewing process for two reasons: (1) The caregiver
may have been upset by the nature of some of the questions;
therefore it was important for the interviewer to try to
assess and if necessary, address such concerns. (2) It was

equally important to reassure the caregiver of anonymity and

41
confidentiality following the interview in case he/she may
have been feeling some regrets for sharing some of the
personal information that he/she disclosed during the course
of the interview. The reassurance may have enabled him/her
to feel more comfortable about discussing issues relating to
the care-receiver.

Part of the closing procedure involved offering the
caregiver a summary of the final report. If the caregiver
expressed an interest in obtaining the report, his/her name
was taken down and added to the master list of names for
future mailing. At this time, the interviewer determined
whether or not the caregiver was living with the elderly
care-receiver, if not already ascertained. Finally, the
interviewer thanked the caregiver for his/her time, allowed
the caregiver to ask any additional questions, and closed
the interview. The completion time of the interview was
then recorded.

Instrument

 

The measuring instrument was designed to assess the
caregiver’s life to determine which areas may have been
negatively affected by caregiving responsibilities.

The structured interview consisted of 47 questions with
five fixed response alternatives. The questions were
divided into seven categories:

1. the quality of the relationship between the
primary caregiver and the care-receiver

2. the physical health of the caregiver

42

3. the emotional and psychological health of the

caregiver
4. the number of social involvements of the caregiver
5. the perceived sense of burden of the caregiver
6. the number of additional informal and formal

support people assisting the caregiver
7. demographic information
The interview is located in Appendix E and provides
quantitative indices of the effects of case management on
reducing stress associated with caregiving.

The items which comprise the interview were rationally
derived after an extensive review of the literature on
caregiving to the elderly, and reviewing previously designed
instruments used with caregivers (e.g., Horowitz (1984),
Kane and Kane (1981), Robinson (1983), and Zarit et al.
(1980)). The work done by lung (1967, 1972) and Yesavage,
Adey, and Werner (1981) on depression was also reviewed to
generate items relating to depression and psychological
health.

Section A of the interview was comprised of questions
which regard the relationship between the caregiver and the
care-receiver. Section A begins on page 91 of Appendix B.
These six questions assessed how the caregiver and the care-
receiver spent their time together. For instance, how often
did they spend time together in conversation, watching
television, and/or laughing. Several items assessed the
underlying feelings the caregiver may or may not have

towards the care-receiver. Thus, the questions focused on

43
both positive and negative aspects of the relationship. All
six questions utilized a response category of five
alternatives: 1-never, 2-rarely, 3-sometimes, 4-quite
frequently, 5-always.

Section B consisted of six questions which focused on
the caregiver’s physical health. An indirect way of
assessing the effects of stressful situations is by
measuring changes which may have occurred in the health of
an individual. Therefore these six questions were directed
towards: 1) changes in the caregiver’s sleeping habits,

2) appetite changes, 3) how quickly he/she tires, 4) whether
or not he/she feels overwhelmed with responsibility, 5)
general mood state upon waking in the morning, and 6) if
he/she experiences feeling restless during the day. The six
questions in this section use the same five point response
scale as the previous section. Section B is found in
Appendix E.

Questions which asked about the caregiver’s emotional
and psychological health are found in Section C of Appendix
E. These six questions utilize the same response alternatives
as the previous two sections of the interview. The overall
theme of this section was to discover what type of emotional
feelings the caregiver may or may not be experiencing during
the time spent assisting the care-receiver. Thus, the
questions revolve around emotional feelings which have been
frequently stated in the literature as common feelings
shared by caregivers. Feelings such as anger, guilt, fear

about the future which exists for the care-receiver, feeling

44
happy around the care-receiver, and whether or not the
caregiver considers his/her life to be interesting were
asked. Again, these questions attempted to reveal the
positive and negative feelings which the caregiver may or
may not be experiencing.

The next section concerned the social involvements of
the caregiver, and how often he/she engaged in activities
which were purely for personal pleasure (e.g., a hobby). The
first four questions determined how many times a week the
caregiver spent socializing with friends, participated in
activities such as hobbies, being with his/her children or
other family members, and how many hours a day were spent
caring for the care—receiver. The first three questions had
five response alternatives which were: 1-daily, 2-3 to 4
times a week, 3-at least twice a week, 4-at least once a
week, 5-less than once a week. The fourth question
concerning the number of hours spent caring for the care-
receiver used the following response alternatives: 1-less
than 5 hours, 2—6 to 10 hours, 3-11 to 15 hours, 4-16 to 20
hours, 5-more than 20 hours. The fifth question asked how
satisfied the caregiver felt about the amount of contact
he/she had with friends and relatives for social purposes in
the past month. A five point Likert scale was used with "1"
representing very dissatisfied and "5" representing very
dissatisfied. The last question asked how often the
caregiver attended meetings sponsored by various
organizations such as: social groups, clubs, churches, or

civic groups. The following five response alternatives

45
were used: l-never, 2-rarely, 3-sometimes, 4-occasionally,
5—regularly.

The caregiver’s perceived sense of stress was the
theme for the next six questions. These questions
attempted to assess the feeling of stress associated with
the caregiver’s role and responsibilities. Six stressors
were the basis for the next set of questions: physical
strain, family adjustments or changes to the caregiver’s
lifestyle, feeling stressed, financial strain, feeling the
care-receiver is demanding, and the convenience of
caregiving. All the questions used the same 5 point scale:
1-never, 2-rarely, 3-sometimes, 4-quite frequently,
5-always.

Section F examined the amount of additional support
which was available to the caregiver (See Appendix E). Two
questions asked how many people from the informal network
were helping the caregiver with caregiving responsibilities
and how much time did he/she spend helping, on the average,
per visit. Another set of questions asked about the formal
service providers that were coming into the home to assist
the caregiver.

The last section of the interview collected
demographic information on the caregivers. This section
included items on age, sex, race, marital status, the
number of children the caregiver has, socioeconomic status,
the caregiver’s relationship to the care-receiver, and

work status.

46
Interviewers

All of the interviews in the Grand Rapids and
Southfield sites were conducted by the project director.

An interviewer from the larger project was recruited to
conduct interviews in the Flint site. Because intake
was based on the completion time of the twelve month
interviews of the larger project, the rate of intake was
moderate and therefore no other additional staff

members were needed.

Because of the extensive experience the interviewer
had with in—person interviews, telephone interviews and
data collection procedures, the interviewer only needed
minimal training regarding interviewing procedures.

Training was conducted at the Flint site, and consisted
of one three hour session. The project, the interview
process, and data collection procedures pertinent to this
study were reviewed. Issues of anonymity of the
participants and confidentiality were also discussed.

The consent forms from Flint were processed by the
project director and copies of the consent forms and the
interviews were sent to the interviewer in Flint.

The completed interviews were returned to the project
director through the mail for processing.

Weekly contact was made with the interviewer by phone
to check on the completion time of interviews and to problem
solve with the interviewer when needed. In person meetings
were conducted when necessary. Because the interviewer
volunteered to conduct interviews, the interviewer did not

receive a set fee for completing interviews.

47

Sgale construction

Because this was an exploratory procedure with no
preset notion of how the variables should go together,
scale construction set out to detect patterns of variables
which make conceptual sense and to possibly reduce the
data. The analysis of the data obtained was done in several
steps utilizing the SPSSQ Statisical Program. First, a
principal factoring method was done on 32 variables
(excluding the demographics and variables 32 and 34 which
asked how many minutes the informal and formal care
provider spent assisting the caregiver respectively).
Principal factoring without iteration was performed
followed by a varimax rotation. This data-reduction
procedure allows one to see whether there are underlying
pattern of relationships among the data which can be
rearranged into a smaller set of components accounting for
the observed interrelations in the data. Using Cattell’s
IScree Criteria, rationality, and independence, a four
factor solution was used. The variables making up factor
one asked about the social involvements of the caregivers
life and was therefore called social (variables 19, 20, 23,
and 24). Variables 11, 14, 16, and 18 were later dropped
from the scale called social because they did not make
conceptual sense within the scale. Variables 11, 14, and
16 were used later in two different scales called
psychological health and physical health (See Table 4).

Variable 18 (how often have you felt afraid about what the

 

future holds for ?) was left as a single item.

48
Factor two was comprised of variables which focused on
stress and its effect on the caregiver. The name
relationship was given to factor three which was made up of
variables which regard the relationship between the
caregiver and the care-receiver. Factor four was called
caregiving. These variables looked at how the caregiver’s
time was spent with the care-receiver. The four factor
solution is reported in Tables 2 and 3.

Second, once the number of factors was determined and
the variables which made up those factors examined by
setting a cut-off factor loading of .50 for inclusion in
a scale, the items were combined to create a scale based on
the factor solution. If an item did not make conceptual
sense with in the context of the other items, it was taken
out of the scale and retained for later use. Thus, the
next step was to test each scale to determine how
homogeneous they were by computing Cronbach’s alpha.

Social Scale. The scale social was made up of

 

variables 19, 20, 23, and 24. The content of these
variables were as follows: variable 19) how often did you
spend time socializing with friends; variable 20) how often
did you have time to do things you like to do; variable 23)
how satisfied have you been with the amount of contact with
friends and relatives for social purposes; and variable 24)
how often have you attended meetings of social groups,
clubs, churches, or civic groups. These variables focused
on the social involvements of the caregiver and time spent
engaging in pleasurable activities. Cronbach’s alpha = .74

(See Table 4).

49
Table 2

Factor Analysis Summary

 

 

Factor Eigenvalue Percent
Variance
1 7.57 23.7
2 2.67 8.3
3 2.16 6.8
4 1.91 6.0
5 1.66 5.2
6 1.42 4.4
7 1.40 4.4
8 1.25 3.9
9 1.14 3.6
10 1.04 3.2
11 .99 3.1
12 .89 2.8
13 .82 2.6
14 .74 2.3
15 .67 2.1
16 .66 2.1
17 .57 1.8
18 .51 1.6
19 .47 1.5
20 .44 1.4
21 .42 1.3
22 .40 1.3
23 .36 1.1
24 .33 1.0
25 .32 1.0
26 .23 .7
27 .22 .7
28 .19 .6
29 .18 .6
30 .15 .5
31 .12 .4

 

50
Table 3

Factor Analysis Summary

 

 

Variable Factor 1 Factor 2 Factor 3 Factor 4 h2
Name Social Stress Relation— Caregiving
ship

1 - 21 .34 -.09 .61 .58
2 - 11 -.09 .11 —.17 .64
3 - 28 -.10 .13 .66 .47
5 27 .00 .16 -.09 .66
6 -.31 .29 -.61 .17 .66
10 .00 -.61 .15 -.05 .35
11 .59 -.29 .15 .30 .55
12 -.17 .66 -.12 -.19 .42
13 .24 .01 .66 .16 .39
14 .66 .02 .41 .13 .58
15 -.10 .66 .04 -.01 .37
16 .66 -.08 .47 .22 .61
19 —.11 .19 -.04 14 .57
20 «.66 .15 .01 11 .49
21 -.27 -.07 -.36 .53 .49
22 -.29 .23 -.08 16 .75
23 61 -.35 .08 -.15 .49
24 61 -.05 .01 -.28 .37
26 11 -.61 .34 -.37 .68
27 —.29 .11 -.21 .09 .66
28 52 .01 .17 -.34 .42

 

51

Stress Scale. Five variables (10, 12, 15, 26, and
and 27) made up the next scale which was called stress and
the content of these variables were as follows: 10) how
often were there times when you didnot feel overwhelmed by
responsibility; 12) how often do you find yourself
restless; 15) how often do you find yourself angry with
other relatives and friends; 26) how often have you been
able to avoid having to make family adjustments due to your
caregiving role; and 27) how often have you felt stressed
while caring for the care-receiver. Cronbach’s alpha was

.73 (See Table 4). 7

Relationship Scale. The relationship between the
caregiver and care—receiver was the content of the next
scale named relationship. The scale was comprised of
questions (2, 5, 6, 13) which asked how often the caregiver:
2) spent time enjoying conversation with the care-receiver;
5) spent time laughing with the care-receiver; 6) felt that
he/she couldnot cope with the care-receiver; and 13) felt
happy around the care-receiver. Cronbach’s alpha was

.76 (See Table 4).

Caregiving Scale. An alpha of .73 was reported for
the last scale, caregiving, which was made up of variables
1, 3, 21, and 22. These questions focused on how the
caregiver’s time was spent with the care-receiver and the
amount of time spent as a caregiver. The content of these
four questions were as follows: variable 1, how often have
you felt that the care-receiver was dependent upon you;

variable 3, how often have you and the care-receiver spent

52
time together watching television; variable 21, how often
have you spent time with children and/or other family
members; and variable 22, in the past day how many hours
per day were spent caring for the care-receiver. (See
Table 4).

The remaining variables were then examined to
determine if any other scales existed. A reliability
procedure was run on these variables and a factor analysis
procedure, as described previously, which resulted in two
additional scales called psychological health and physical
health.

Psychological Health Scale. The variables in the

 

scale psychological health were as follows: 14, 16, 29, 30.
This scale tapped the caregiver’s outlook on life in
relation to his/her attitude towards caregiving. The
content of these variables were as follows: 14) how often
have you felt that your life is full of things that keep
you interested, 16) how often have you felt satisfied with
life, 29) how often have you felt that the care—receiver
asks for more help than he/she really needs, and 30) how
often would you say that you have felt that caring for the
care-receiver is an inconvenience. Cronbach’s alpha was

.60. (See Table 4).

Physical Health Scale. The variables in the

physical health scale are: 7, 8, 9, 11, 25. The nature of
these questions revolve around physical health: question 7
asks if the caregiver has trouble getting to sleep,

question 8 asks if how often the caregiver would say

53

his/her appetite was good, and question 9 concerns how
often the caregiver feels tired after little or no strenous
activity. Questions 11 and 25 ask how often the caregiver
feels fresh and rested in the morning and how often he/she
has experienced physical strain while caring for the care-
receiver respectively. The reported alpha for this scale
was .68. (See Table 4).

The fourth step in the data analysis process, was an
attempt to address the issue of independence of the
measure. Each of the six scales were correlated against one

another using the Pearson project-moment correlation

coefficient (r). The scale social correlated with the scale
stress, -.40; relationship, .28; caregiving, -.44; physical
health, -.47; and psychological health; ~.48. The scale

stress correlated with the scale relationship, -.38;
caregiving, .29; physical health, .58; and psychological
health, .46. The relationship scale correlated with the
7caregiving scale, -.26; with the physical health scale, -

.36; and the psychological health scale, -.51. Caregiving
correlated with the scales physical and psychological health

.31 and .18, respectively. Finally, physical health and
psychological health correlated .38. The results of this
procedure were somewhat disappointing. It was hoped that
the scales would be as independent from one another as
possible with each scale measuring its own unique construct.
Thus, a cut-off level for the correlation between the scales
was set at .30 and below. As Table 5 indicates, some

of the scales correlated above the cut-off point which can

54
Table 4

Scale Reliability Summary

 

 

 

 

 

 

 

 

Variable Corrected r Alpha

Social Scale .74

19 .54573

20 .57640

23 .49260

24 .53449
Stress Scale .73

10 .43869

12 .47583

15 .46235

26 .52677

27 .59280
Relationship Scale .76

2 .53977

5 .66178

6 .58062

13 .44676
Caregiving Scale .73

1 .51804

3 .52320

21 .47286

22 .66980
Psychological Health Scale .60

14 .30747

16 .45437

29 .29433

30 .48390
Physical Health Scale .68

7 .40480

8 .45260

9 .44950

11 .40726

25 .47381

 

55

 

mm.
ma.

Hm.l
we.

w¢.|
Househmm

mm.
mm.- mm..
mm. mm. mm.7
as.. as.. mm. o«..
Hoozmmsm >Hmwmo Qwsmaom mmmwpm Amfloom

Hmmnmmfim
>«mnmo
assmamm
mmmsum

Hmfloom

 

wmfinom cmmzuom
mcoflumamapoo ucosozluospoam comamom

m magma

56
be taken as an indication that those scales shared a
relationship between one another and may be measuring the

same construct.

CHAPTER III

Results

This section presents the results of the statistical
analysis of the data to test the seven hypotheses. The
purpose of the study was to determine if case management was
effective in reducing the stress associatedwith caregiving
to the frail elderly through the incorporation of formal
supports into the informal support system. The study was a
post test design and data was collected using a telephone
questionnaire.

The first statistical procedure tested the assumption
of homogeneity of variance which was then followed by a
multivariate analysis of variance (MANOVA) procedure which
entered all of the six scales as dependent variables to
test the hypotheses about the relationship between the set
of interrelated dependent variables and one grouping
variable, program status.

Each of the seven hypotheses were tested using a one-
way analysis of variance (ANOVA) procedure which uses the
F-ratio to determine whether or not two variances differ
significantly from one another. In addition, regression
procedures were employed to assess the relationship between
the dependent variable stress; and several independent
variables to determine indicators of stress in a caregiving

relationship.

57

58
Homogeneity of Varignce

Because the F test is not robust to violations of the
homogeneity of variance assumption when the sample sizes are
not the same, the assumption of equal variances was tested.
Each of the six following scales were tested univariately:
social, stress, relationship, caregiving, physical health,
and psycholgical health. The results of the Bartlett-box
and Cochran's C tests indicate the assumption was not
violated. The results of these two tests are presented in
Table 6.

Manova

A one by six multivariate analysis of variance (MANOVA)
utilizing a full factorial design was performed on the six
dependent variables: social, stress, relationship,
caregiving, physical health, and psychological health. The
independent variable was the experimental condition. This
analysis was done to test the hypothesis that there is an
univariate difference on the six scales in two experimental
groups of individuals (case management and information and
referral) and to possibly reveal differences in the data not
shown in separate ANOVA procedures.

The MANOVA procedure was performed using SPSSQ, and
the effect was tested at the p<.05 level for a two—tailed
test. Seventy—six cases were used. Wilk’s Lambda was used
to determine if there was a main effect, and if so, to find
a measure of the strength of association between the
independent effect and the combination of the dependent

variables. There was no main effect detected (Lambda:

59

Table 6

Univariate Homogeneity

of Variance Summary

 

 

Variable DF F P
Social Scale
Cochran’s C 37/2 .55120 .535 (Approx.)
Bartlett-Box 1/16142 .38430 .535
Stress Scale
Cochran’s C 37/2 .52889 .727 (Approx.)
Bartlett-Box 1/16142 .12084 .728
Relationship Scale
Cochran’s C 37/2 .58883 .279 (Approx.)
Bartlett-Box 1/16142 1.17446 .279
Caregiving Scale
Cochran’s C 37/2 .53232 .696 (Approx.)
Bartlett-Box 1/16142 .15240 .696
Physical Health
Scale
Cochran’s C 37/2 .50031 .997 (Approx.)
Bartlett-Box 1/16142 .00001 .997
Psychological
Health Scale
Cochran’s C 37/2 .57001 .395 (Approx.)
Bartlett-Box 1/16142 .72346 .395

 

60
.96355, F=.43497, significance of F=.85306). (See
Table 7). Although there was no main effect detected,
each of the dependent variables was tested individually,
since the nature of the study was exploratory.
Hypothegis One

The first hypothesis stated that the quality of the
relationship between the primary caregiver in the case
management group would be significantly better than the
primary caregiver in the information and referral group.
The scale relationship was entered as the dependent
variable. Program status defined whether the client was
in the case management group or in the information and
referral group. The item program status was the
independent variable. There was no significant difference
between the two groups (F=.064, DF=1 p<.801). (See Table
8).

HypothesigfiTwo

The second hypothesis stated that the physical health
status of the primary caregivers in the case management
group would be significantly better that than caregivers in
the information and referral group. The scale physical
health was entered into the ANOVA procedure as the dependent
variable and program status was the independent variable.
The results of the ANOVA procedure were not significant
(F=.495, DF=1 p<.484) therefore, the null hypothesis cannot

be discarded. (See Table 8).

61
HypothesigfiThree
The next ANOVA procedure tested the third hypothesis;
the emotional/psychological health of the primary caregiver
in the case management group would be significantly better
than the primary caregiver in the information and referral
group. The independent variable was program status. The
results were not significant (F=.020, DF=1 p<.887),
therefore, the null hypothesis that there would be a
difference between conditions on emotional/psychological
health cannot be rejected. (See Table 8).
Hypothesi§_Four 7
The following hypothesis addressed the social contacts
of the primary caregivers; the social contacts of the
primary caregivers in the case management group would be
significantly greater that those of the primary caregivers
in the information and referral group. No significant
difference was found between the two groups when using the
scale social as the dependent variable as compared to
program status (F21.528, DF=1 p<.220). (See Table 8).
Hypothesi§,Five
The next analysis was run to test the hypothesis; the
perceived sense of burden of the primary caregiver in the
case management group would be significantly lower than the
primary caregiver in the information and referral group.
The scale stress was entered as the dependent variable and
program status was the independent variable. The mean
score for the case management group on the stress scale was

14.93 and the mean for the information and referral group

62
was 15.17. The results of the ANOVA were non significant,
(F=.058, DF:1 p<.811), the null hypothesis that there would
be a difference between the conditions cannot be rejected.
(See Table 8).
Hypothesis Six

The following hypothesis focused on the number of
informal and formal supports of the caregivers; the number
of informal and formal supports of the primary caregiver in
the case management group would be significantly greater than
those of the primary caregiver in the control group. There
was no significant effect between the two groups using
variable 31, which asked the caregiver how many additional
family members or friends were assisting with caregiving
responsibilities, as the dependent variable for informal
supports and program status as the independent variable
(F=.040, DF:1 p<.841) and variable 33, which asked for the
number of formal service providers who were coming into the
home, as the dependent variable for formal supports and
program status as the independent variable (F:2.958, DF:1
p<.090). (See Table 8).

In summary, none of the six hypotheses could be
confirmed by the either the multivariate or univariate
analysis of variance procedures. Thus, there was no
significant treatment effect and none of the six null
hypotheses could be rejected.

Eggression Analysis
Four separate regression procedures were done to assess

the relationships between the scale stress and various

63
Table 7

Multivariate Summary

 

 

 

 

 

Test Name DF Value F Sig. of F
Wilk’s Lambda 6/69 .96355 .43497 .85306
Table 8
ANOVA Summary
Source of Variation DF MS F Sig. of F
Relationship Scale 1/75 .668 .064 .801
Physical Health Scale 1/75 8.878 .495 .484
Psychological Health 1/75 .195 .020 .887
Scale

Social Scale 1/75 28.209 1.528 .220
Stress Scale 1/75 1.130 .058 .811
Variable 31 1/75 .238 .040 .841
(additional informal

supports)
Variable 33 1/75 2.121 2.958 .090

(additional formal
supports)

 

64
independent variables to determine whether or not they might
be predictors of stress. All four procedures were stepwise
regression with forward inclusion.

In the literature on caregiving, various demographic
variables have been found to be significant predictors of
stress, such as the sex of the caregiver and socioeconomic
status, therefore the first procedure regression procedure
were was done to confirm the hypothesis that the
demographics, variables: living status, program status, the
five scales and the remaining variables 4, 17, 18, 28, 31,
and 33 could be indicators of stress. The scale stress was
entered as the dependent variable and the demographics,
living status, and program status were entered as the
predictors. Listwise deletion was used to handle missing

data therefore; only sixty-seven cases were included in this

analysis. The variable living status was significant with a
significance value 2.005, R square =.11709, and simple R =
.34219. (See Table 9). According to this result, if the

caregiver lived with the care-receiver, there was an
increase in stress reported by the caregiver. This makes
rational sense since living with another individual involves
more opportunities for contact between individuals.
Increased contact can mean increased stress in the
relationship if one of the individuals perceives

the relationship as restricting and the responsibilities
associated with the relationship overwhelming (Cantor, 1983;
Crossman et al., 1981; Rakowski & Clark, 1985; Robinson &

Thurnher, 1979).

65

In the equation, living status correlated with
relationship (r =.50) which indicated that if you lived with
the person you were caring for, you were more likely to be a
relative e.g., spouse, child, other relative. This finding
is consistent with the literature which has documented the
role of the family in providing care for
noninstitutionalized elderly (Brody, 1981; Cantor, 1983;
Tress, 1977). In addition, those caregivers who had a
frail elderly member living with them also reported lower
incomes (r :-.47).

The next regression procedure was done with stress as
the dependent variable and the five scales as predictors.
Seventy-six cases were used in this procedure. Four of the
scales were significant; social, physical health,
relationship, and psychological health. The significance
values were: .000, .007, .006, and .025, respectively.
(See Table 10). The R square values were: social =

.15609, Physical health =.23616, relationship =.31275,
and psychological health =.35984. Simple R values were as
follows: social =-.39508, physical health =.34582,
relationship =—.37535, and psychological health =.14598.

These results indicate that as caregivers reported
more physical strain related to caregiving, stress scores
increased. Similarly, as psychological health scores
increased, indicating negative effects on psychological
health, stress scores increased. Parallel results have
been found in the literature on the physical and emotional

costs of caregiving (Clark & Rakowski, 1983). Thus, stress

66
has a detrimental effect on the physical and psychological
health of the caregiver.

Stress is also heightened by a decrease in social
involvements and a decrease in satisfaction with the
caregiving relationship. It could be explained that the
caregiver becomes more and more dissatisfied with the
caregiving relationship because it means lack of social
involvements. The two scales were correlated (r =.28) which
indicated as social isolation increased, dissatisfaction
increased. Similar results have been found by various
researchers examining the relationship between social
isolation and psychological well-being (Chatters et al.,
1985; Fengler & Goodrich, 1979; Marcus & Jaeger, 1984).

The variables which were retained after the factor
analysis procedures and which did not fit into any of the
scales were combined and utilized in a regression
procedure. Since stepwise regression can be viewed as a
model-building procedure, this procedure was done to find a
subset of independent variables that were most useful in
predicting the dependent variable, and to eliminate those
independent variables that did not provide additional
prediction. Seventy-six cases were used in this procedure.
The variables used as predictors were: V4, how often have
you felt you should be doing more for the care-receiver;
V17, how often have you been bothered by feelings of guilt;
V18, how often have you felt afraid about what the future
holds for the care-receiver; V28, how often have you felt

that you have enough money to care for the care-receiver in

67
Table 9

Regression Summary

 

Variable Significance R Square Simple R

 

Living Status .005 .11709 .34219

 

Table 10

Regression Summary

 

 

Variable Significance R Square Simple R
Social .000 .15609 -.39508
Physical Health .007 .23616 .34582
Relationship .006 .31275 -.37535
Psychological .025 .35984 .14598

Health

 

68

addition to your other expenses; V31, how many additional
family members or friends have helped the care-receiver to
take care of him/herself or have done things around the
house for the care-receiver; V33, how many formal service
providers (i.e., nurses, therapists, chore workers) have
come to the care-receiver’s home, on the average. Once
again, the dependent variable was the scale stress. Of
these variables, three were significant, which are: V18,
V4, and V31. Corresponding significance values are: .004,

.019, .074. R Square values were: .10881, .17436, and

.21030 respectively. Simple R values were as follows:
-.32987, .28325, and -.21586 respectively. (See Table 11).

These results indicated that there were additional
variables which were important indicators of stress based
on the responses given by the caregivers in this sample.
Question 18 refers to the fear caregivers often express
about the future of the care-receiver. This fear is related
to such issues as possibly institutionalizing the care-
receiver, especially if something would happen to the
caregiver. Fear is often mentioned in regards to the care-
receiver’s safety and future deterioration of the
care—receiver’s physical and mental health (Cantor, 1983).
This variable did not correlate significantly with any of
the other variables in this equation.
The second significant variable was number four which

asks "how often have you felt you should be doing more for
care-receiver?" This variable correlated with variable 17

(how often have you been bothered by feelings of guilt) in

69
this equation (r =.28) which is not unusual since most
caregivers feel their "job" is never done and feel they
they should be doing more, whether it is motivated by love
or a sense of obligation (Brody, 1981; Shanas & Maddox,
1976).

The last variable in this equation was number 31 which
asked about the number of additional informal supports
(family members and friends) who may be assisting with
caregiving responsibilities. This variable correlated with
stress r =-.22, which indicated that as informal assistance
increased, stress decreased. This variable did not
correlate with any of the other variables significantly.

The final regression procedure combined the variables
found to be significant indicators in the previous
procedures and entering them as predictors with the scale
stress as the dependent variable. Seventy-six cases were
utilized in this procedure. The variables found to be
significant were the scale social, variable 4; how often
have you felt you should be doing more for the care-
receiver, the scales physical health, psychological health,
and relationship, and variable 18; how often have you
felt afraid about what the future holds for the care-
receiver. The corresponding significance levels were .000,

.005, .009, .013, .013, .036. (See Table 12). R Square
values were: .15609, .24467, .31402, .37094, .42400, and

.45975, respectively. The Simple R values associated with
each of the variables were: -.39508, .28325, .34582,

.14598, -.37535, and —.32987. These results indicated that

70

 

 

 

 

 

Table 11
Regression Summary
Variable Significance R Square Simple R
Variable 18 .004 .10881 -.32987
Variable 4 .019 .17436 .28325
Variable 31 .074 .21030 -.21586
Table 12
Regression Summary
Variable Significance R Square Simple R
Social Scale .000 .15609 -.39508
Variable 4 .005 .24467 .28325
Physical Health .009 .31402 .34582
Scale
Psychological .013 .37094 .14598
Health Scale
Relationship Scale .013 .42400 -.37535
Variable 18 .036 .45975 -.32987
Adjusted R2 .10212

 

71
living with the care-receiver results in decrease social
involvement, more physical health problems, decreased
satisfaction in the caregiving relationship, and increased
fear regarding the care-receiver’s future. Once again,
increased social involvements correlated relationship with
the care-receiver (r =.28) and the social scale correlated
with variable 18 (r =.27) indicating decreased fear
regarding the care-receiver’s future associated with
increased social involvement. Finally, a correlation
between variable 4 and the relationship scale (r =-.22)
indicated as the caregiver felt he/she should be doing more
for the care-receiver, satisfaction in the relationship
decreases.

Because R square tends to overestimate in the positive
direction, the value obtained from the SPSS9 procedure,
which utilizes the Wherry Formula, was entered into the
Cattin Formula which is more advantegeous to use when
~sample sizes are small (Tabachnick & Fidell, 1983).

Adjusted R square =.10212 and is reported in Table 12.

CHAPTER IV

Discussion

The purpose of the present study was to examine the
effect case management had on primary caregivers in hopes of
identifying a means to reduce the stress associated with
caregiving to the elderly. In this study, case management
was compared to a service normally available in the
community called information and referral through the
responses given by caregivers utilizing a telephone
interview.

The consequences of caregiving have been shown to be
deterimental to the psychological and physical health of the
caregiver (Brody, 1985). Therefore, of major interest was
the relationship between stress and several aspects of the
caregiver’s life. These included: social involvements,
physical health, psychological health, the relationship with
the care-receiver, and caregiving itself. The experimental
analyses, MANOVA and ANOVA, used to examined these
relationships found no significant effects by the treatment
case management. Several possible explanations arise which
might account for these results. Let us now take a look at
what has been presented regarding case management, what we
have learned from the present study, and examine some of the
explanations associated with the failure of the present

study to detect any treatment effects.

72

73
An Evaluation of the Treaaaent

First, case management as a treatment provides for the
coordination of available services to assist the frail
elderly to remain as independent as possible in their own
homes, thus reducing the nursing home usage. This study
found no significant difference between the control and
experimental groups regarding the number of formal service
providers. Since a core part of case management is the
coordination of formal services into the informal network,
this may explain why case management, as a treatment, did
not significantly impact the caregivers in this study. In a
preliminary study based on the larger project associated
with this study, it was found that a problem did exist
regarding implementation of services throughout the five
sites (Bornstein, 1986). This study questioned the
effectiveness of case management which may have been
insufficient to produce the expected outcomes because case
management teams were found to be restricted in their
ability to broker services into the client’s home. This was
due to the lack of medicaid and medicare waivers which were
not available to enable the treatment teams to purchase
services for their clients. In addition, the treatment
group did not differ significantly from the control group in
formal service levels.

Although the target group of case management was the
frail elderly, it was hypothesized that increased assistance
to the frail elderly client would have an indirect effect on

the caregiver by easing some of the responsibilities of

74

caregiving. This would in turn, reduce the stress the
caregiver was experiencing. This hypothesis was not found
to be true. It could be that even though a caregiver may

not be providing instrumental support as often because of
increased formal assistance, he/she is still psychologically
involved. Psychologically, it is difficult to separate
oneself completely from someone you care about even if you
aren’t there physically. Horowitz and Dobrof (1982) found
that psychologically related problems were most often cited
by caregivers when asked questions which determined the
consequences of caregiving based on the relationship of the
caregiver (spouse, children, relative) to the care-receiver.
In addition, family supports continued to remain actively
involved despite the addition of formal services into their
network (Brody, 1986). Thus, there is a need for future
research which focuses on programs that are aimed solely for
the caregiver such as respite services and coping strategy
education (Miller, Gulle, & McCue, 1986).

Methodological Consideragions

The Measuring Instrument

 

The second possible explanation forthe results
obtained is that the measuring instrument may not have been
sensitive enough to detect differences which may have
occurred as a result of the intervention. Since the
measuring instrument was newly designed, there was no
previous reliability and validity data on it. Moderate
reliability was reported for each of the six scales. Future

efforts should include examining the psychometric properties

75

of this instrument further, the use of instruments with
established reliability and validity data, in the framework
of longitudinal research which utilizes a true experimental
design.
The ngacreeningiTool

The question arises about the sample whether it was
appropriate for the study. The caregivers who participated
were based on a sample of frail elderly individuals who
were involved in the study on case management. The elderly
participants were selected based the use of a prescreening
instrument which has unknown reliability and validity data.
Just as in other studies on case management which has
questioned whether they were actually targeting the high
risk group in fear of institutionalization (Capitman, 1986),
this study has also questioned the use of a instrument which
has not been shown to actually target the group it is
intended to due to lack of consistency in its use and the
degree of "clinical judgement" associated with it. Thus,
its effectiveness as a measuring instrument is questionable.

The Role of the Caregiyers

The final explanation regards the sense of commitment
many of the caregivers felt for the individual they were
caring for. If a spouse could do what a paid helper could
do, then he/she would do it. So the problem also exists
concerning how to get certain types of caregivers to utilize
services which they are eligible for. In this study, many
family members could not afford to pay for services or they

preferred, as the elderly care-receiver did, a family member

76
to assist them instead of a formal helper. Thus, it is was
not surprising that spouses and children were willing to do
whatever was necessary out of a sense of familial
responsibility regardless of possible adverse consequences
which may occur over time. This result has been found
consistently in the literature on caregiving to the elderly
by family members (Deimling & Bass, 1986; Rabowski & Clark,
1985). Spouses had a deep sense of commitment to a husband
or wife and were willing to accept what life brought to them
whether it was good or bad. On the other hand, children
stated that their parents had taken care of them and now it
was their turn to reciprocate. Children felt indebted to
their parents with a feeling of affection and love towards
them. They cared for their parents because "it was the
thing to do and you did not ask why, you just did it". Most
children stated that they did not want to institutionalize
their parents and would only consider it if they were
physically unable to continue providing assistance. Thus,
these individuals were providing the care necessary and
managing sufficiently. They found no need for assistance
from outside even when it was available to them because they
were able to respond to the needs of the elderly sooner and
were more flexible than the formal system. Thus they were
willing to make the adjustments in their lifestyles that
were necessary in order to care for their frail relative.
Final Considerations
As we move into a era in which the trend is to keep the

frail elderly at home as long as possible, what this

77
actually means is that there is an informal person(s) who is
actively involved with keeping that person at home. The
effects of caregiving need to be taken into consideration
when implementing programs designed to reduce the costs
associated with providing health care to the elderly. The
current trend in this country is to look for alternatives to
the rising costs of providing institutionalized care
(Capitman, 1986). Are we attempting to substitute the
institution for the family in an attempt to reduce
expenditures for health care or because it benefits the
frail elderly population? What is the motivating force
behind the push for case management? Whatever that force
may be, we must not ignore those who have been called the
"hidden victims", the caregivers who continue to make
sacrifices and do the best they can to cope with the
loneliness, isolation, frustration, anger and depression
that accompanies caregiving to the frail elderly. As
attempts are made to modify the health care system, what
seems to be overlooked is that if you change or modify one
part of a system, you affect all other parts of that
system (Miller, 1979). On one hand, we may be helping the
elderly, but we could also be creating another problem.
Because there is a lack of family-oriented services to
support caregivers, they are at risk of experiencing a
negative impact on their mental and physical health (Brody,
1986). If we do not recognize it, we will soon be creating
another agency to deal with the health problems individuals

are experiencing associated with caregiving and spending

78
millions of dollars to fund it, thus defeating the purpose
of what care management was meant to do; save money.

Case management has been shown to have some positive
effects on caregivers in previous studies (Quinn & Hodgson,
1984), thus it should not be cast off as ineffective. But,
it does need to be evaluated in a more stringent manner in
regards to both its influence on the lives of the frail
elderly and its possible effects on caregivers. From the
policy maker’s perspective, reducing or controlling hospital
and/or nursing home costs is the primary objective. This
objective is going to become even more important in the
future as our elderly population continues to grow. But we
also need to incorporate some additional characteristics
into programs designed for the elderly, such as case
management, which take into account the changes which effect
the caregivers as a result of implementing these programs
into the health care system. At the present time, social
policy in the 0.8. typically ignores the needs of the
families providing home care (Brody, 1986).

Before considering widespread implementation of case
management, policy makers facing the decision regarding
implementation, such as the State of Michigan, need to
consider not only the financial aspects of case management
programs but also the-social implications of incorporating
these programs into the health care system. A balance is
needed between what the formal support system can do and
what the informal support system can provide. Each of these

systems are a vital part of the social support network.

79
Ideally, the two system should complement one another, not
increase the responsibilities on one of the systems, such as
the informal system, until it can longer accept the heavy
burden of care and the system itself begins to dissipate.

We must begin to take a hard look at the present and
future results of the evaluations of case management
programs for the elderly and the impact these programs may
have on caregivers for several reasons. First, it is our
tax money which funds these programs. If they are not
effective, then we should look at other alternatives
otherwise, it is just another example of wasteful government
spending. Second, someday you may be in the position of
being a caregiver to a relative or a friend. If this occurs
in your life time, wouldn’t you want support programs which
actually worked or would you be willing to accept the help
from a program which did not meet your needs? Finally, as
you read this now, you are aging. Unfortunately, we are
never in a position to know what our later years will be
like; for instance, you do not know if you will be
financially secure or if you will having surviving family
members to help you. Therefore, if you reached old age,
wouldn’t it be more comforting to know that you can rely on
an effective program to assist you regardless of the type
of assistance you needed and whether or not you already had
informal supports readily available to you?

As in other areas of life, we must be informed
consumers relating to services being offered in the area of

health care to the elderly. And as aging individuals, we

80
all have an vested interest in seeing that the present and

future elderly receive the best services possible.

APPENDICES

APPENDIX A

81

E11. PRIMARY CAREGIVER

TODAY, WE’VE TALKED ABOUT SOME PEOPLE WHO HELP YOU. OF THE
PRIENDS, RELATIVES, AND NEIGHBORS THAT YOU HAVE MENTIONED
TO ME, WHO DO YOU FEEL USUALLY PROVIDES YOU WITH THE MOST
HELP AND SUPPORT?

SITE (01-02)
CLIENT (03-05)
BLANK (06) BLANK

 

NAME: RELATIONSHIP

 

(07-10)

 

BLANK (ll-78) BLANK
CARD (79-80) 50

APPENDIX B

82

Primary Caregiver Study Consent Form

During this interview, you gave us the name of

as the person who provides you with the most

 

help and support throughout your day.

In order to determine how caregiving affects

life, we would like to talk to him/her about:

 

 

 

 

 

 

1. Age, employment, and income etc.

2. Relationship with you.

3. Health.

4. Social activities and who she/he talks to.

5. Things that may trouble .

6. Other people who help care for you.

This should take about 20 to 30 minutes and we will be
talking to over the phone.

May we have permission to contact ?

By allowing the research team to contact (primary
caregiver’s name), I understand that:

1. I am willingly granting my permission for the
research team to contact my primary caregiver.

2. My decision to withdraw from this study will not
affect my health care, support services, or
benefits in any way.

3. My participation is voluntary and I may
discontinue my involvement with this project at
any time.

4. All information disclosed between my primary
caregiver and the research team will be kept
strictly confidential.

5. In no way can any information disclosed be used
against me.

6. My name will not be associated with the
information my primary caregiver may give and will
not be included in any of the research findings.

7. I have a right to a summary of the research report

that is a product of this study.

83

8. The reason for why my permission was needed was
fully explained to me and the purpose of the study
was also explained.

 

Signature

 

Witness

 

Date

Primary Caregiver Address

 

Full Name

 

Street

 

City State Zip

Telephone Number ( ) -

APPENDIX C

84

Date

Caregiver’s name
Street address
City, State, Zip

Dear (Caregivers name),
I am writing to you about our study which is

interested in learning about how caring for an older
individual influences a person’s life. We have been given

 

 

permission by to contact you about your role
as a caregiver. We are interested in your opinions about
how helping and caring for has affected your
life.

The study consists of a brief 20 to 30 minute
telephone interview which will be arranged with you at your
convenience. All the answers you give will be strictly
confidential and anonymous. In addition, you will be free
to skip any questions or stop at anytime throughout the
interview. Therefore, your participation in this study is
completely voluntary.

You will find two copies of a consent form with this
letter. One copy is yours to keep, the other copy is for
you to sign and return in the self-addressed stamped
envelope, should you choose to participate.

I will be contacting you, or one of my staff members,
in the next two weeks to talk with you and answer any
questions you may have about the study. At that time, we
could also conduct the interview if it is a convenient time
for you.

I look forward to talking with you very soon.

Sincerely,

Susan Paulson
Project Director

APPENDIX D

85

Primary Caregiver Study
Client Consent Form
Currently, a study is being conducted at Michigan
State University which is interested in learning about how

caring for an older individual affects a person’s life.

In hopes of learning more about this process, we have

 

been given permission by to contact you. If you
agree to participate, we would like to talk with you for 20
to 30 minutes. We would like your opinions about how
helping and caring for affects your life. In

 

addition, we are interested in your point of view
concerning the needs of primary caregivers.

We would like to talk to you about the following

areas:
1. Age, employment, and work etc.
2. Your relationship with .

 

3. Your health.

4. Your social activities and who you enjoy talking
with.’

5. The things that may trouble you in your life.

6. The other people who may be helping you care for

 

Would you like to participate?
Once the interview is over, your participation in the study
has been completed. I understand that by participating in
this research:

1. My participation is voluntary and I may
discontinue at any time during the interview.

2. My decision to withdraw from this project will not
affect my health care, support services, or
benefits in any way.

3. All information disclosed will be kept strictly
confidential.

4. My name will not be associated with the
information and will not be included in any of the
study’s findings.

5.

6.

86

I can skip any questions I do not want to answer.

I have a right to request a summary of the report
that is a product of this study.

In no way will the information disclosed be used
against me.

The study has been explained to me and I have had
an opportunity to ask questions about the study.

 

Participant’s Signature

 

Witness

 

Date

APPENDIX E

CAREGIVER TELEPHONE INTERVIEW

Opening the Interview

Hello, my name is . I am a interviewer
with a research team at Michigan State University which is
interested in talking with people who care for older
individuals in their homes. This study is being done to
better understand how caregiving to an older individual
influences the life of the caregiver.

Your name has been given to us by , who
has identified you as the one person who provides him/her
with the most help and support throughout his/her day.
Therefore, you were sent a letter explaining the study and
notifying you that someone would be calling you. Have you
received that letter? Do you have any questions about the
letter or the study that I might answer for you at this
time?

At this time, I would like to give you some
information about the interview. The interview will be
conducted over the telephone and will take about 20 to 30
minutes to complete. The questions will concern your
overall health, the number of your social involvements and
informal/formal supports, your relationship with ,
and the amount of stress you may or may not be experiencing
as a caregiver. All the answers you give will be strictly
confidential. No one but the research team will have

access to your answers and they will be anonymous. In

88
addition, you will be free to skip any questions or stop at
anytime throughout the interview. Do you have any
questions you might like to ask at this time?

Would this be a good time for you to answer some
questions about yourself and your role as a caregiver? [IF
YES, THANK THE CAREGIVER USING HIS/HER NAME AND PRECEDE TO
THE INTERVIEW. IF NO, ASK THE CAREGIVER IF YOU CAN ANSWER
ANY QUESTIONS HE OR SHE MAY HAVE CONCERNING THE INTERVIEW,
HIS OR HER INVOLVEMENT, CONFIDENTIALITY, OR HOW THE
INFORMATION WILL BE USED. IF AFTER TALKING WITH THE
CAREGIVER, HE OR SHE AGREES TO PARTICIPATE, THANK HIM OR
HER USING HIS OR HER NAME AND PRECEDE WITH THE INTERVIEW.
IF HE OR SHE REFUSES, THANK THE CAREGIVER FOR HIS OR HER
TIME AND ASK IF WE COULD CALL ON ANOTHER DAY. IF YES, TRY
TO SET UP A DAY OR TELL HIM OR HER YOU WILL CALL IN A FEW
DAYS. IF NO, THANK THE CAREGIVER FOR HIS OR HER TIME AND
CLOSE THE INTERVIEW BY EXTENDING THE WISH FOR THE CAREGIVER

TO HAVE A GOOD DAY OR EVENING WHICH EVER IS APPROPRIATE].

NOTE: Before beginning the interview, inform the caregiver
that he or she is free to ask questions if he or she
doesnot understand something. In addition, remind

the caregiver to return the consent form in the
self—addressed stamped envelope.

89

 

CALL RECORD

 

DATE CALL IS TIME COMPLETED: YES/NO REFUSED FUTURE
MADE (BUSY, NOT IN, CALL
NO ANSWER) DATE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

90

CLIENT IDENTIFICATION SHEET

Client ID Number

 

Interviewer

 

Site Location

 

Interview Date

 

 

Client Name

 

Address

 

 

Home Telephone

 

it: Is the caregiver currently living with the
care-receiver?? YES or NO

 

Status in the study:

1. Case Management

2. Information and Referral
Time of Interview:

1. Beginning Time

 

2. End Time

 

Information Source from the Parent Project:

1. Proxy
2. Combination
3. Client

 

Client of Parent Project for reference in interview:

 

Address, only if needed:

 

91

 

(01-04)
Client I.D. Number

A. QUALITY OF THE RELATIONSHIP BETWEEN
THE PRIMARY CAREGIVER AND THE CARE—RECEIVER
The first group of questions concern your relationship to
I would like to ask some questions

regarding you relationship and how the two of you spend
your time together. The first question is:

 

1. In the past month, how often have you felt that
is dependent upon you?

 

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
(05)

2. In the past month, how often have you and
spent time enjoying conversation together?

 

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
06)

3. In the past month, how often have you and
spent time together watching T.V.?

 

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
(07)

4. In the past month, how often have you felt you should
be doing more for ?

I. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal

(08)

92

5. In the past month, how often have you and
spent time together laughing?

 

1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

 

(09)—
6. In the past month, how often have you felt that you
cannot cope at all with ?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(10)

B. PHYSICAL HEALTH

The next few qestions are about your physical health. The
first question is:

1. In the past month, how often have you had trouble
getting to sleep or staying asleep?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(11)____
2. In the past month, how often would you say your
appetite was good?
1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
(12)____

3. In the past month, how often have you felt tired after
little or no strenous activity?

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal

(13)

93
In the past month, how often were there times when you
didnot feel overwhelmed by responsibility?
1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
(14)____

In the past month, how often have you felt fresh and
rested when you awoke in the morning?

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
(15)

In the past month, how often do you find yourself
restless where you cannot sit still for very long?

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal

(16)

C. EMOTIONAL/PSYCHOLOGICAL HEALTH

The next group of questions are about feelings you may have
at one time or another as you care for and go
about your day.

In the past month, how often have you felt happy when
you are around ?

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal
(17)

In the past month, how often have you felt that your
life is full of things that keep you interested?

I. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal

(18)

94
3. In the past month, how often do you find yourself angry
with other relative and friends?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(19)____
4. In the past month, how often have you felt satisfied
with life?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal
(20)____

5. In the past month, how often have you been bothered by
feelings of guilt?

I. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal

(21)

 

6. In the past month, how often have you felt afraid about
what the future holds for ?

 

1. never 2. rarely 3. sometimes
4. quite frequently 5. always 6. refusal

(22)

D. SOCIAL INVOLVEMENT

The next few questions are about your social involvements.
This group of questions concern how often you are with
other people or are doing things you like to do, for
example, a hobby. The first question is:

1. In the past month, how often did you spend time
socializing with friends? (i.e., being with other
people for enjoyment)

1. daily 2. 3-4 times a week 3. at least twice
a week
4. at least once 5. less than once 6. refusal

a week a week
(23)

95

In the past month, how often did you have time to do
things you like to do? (i.e., hobbies, going for
drives, reading, or going to the movies)

1. daily 2. 3-4 times a week 3. at least twice
a week
4. at least once 5. less than once 6. refusal
a week a week
(24)—

In the past month, how often did you spend time with
children and/or other family members?

1. daily 2. 3-4 times a week 3. at least twice
a week
4. at least once 5. less than once 6. refusal
a week a week
(25)

In the past day, how many hours were spent caring for
9

 

1. less than 5 2. 6—10 3. 11-15 4. 16-20
5. more than 20 6. refusal

(26)____
In the past month, how satisfied have you been with the

amount of contact you have had with friends and
relatives for social purposes?

1. very dissatisfied 2. dissatisfied 3. neutral
4. satisfied 5. very satisfied 6. refusal
(27)—

In the past month, how often have you attended meetings
of social groups, clubs, churches, or civic groups?

1. never 2. rarely 3. sometimes

4. occasionally 5. regularly 6. refusal

(28)

96

E. PERCEIVED SENSE OF STRESS

The next group of questions concern your perception i.e.,
your feeling of stress associated with your caregiving role
and responsibilities.

1.

In the past month, how often have you experienced

 

physical strain (i.e., because of lifting) while caring
for ?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(29)
In the past month, how often have you been able to
avoid having to make family adjustments due to your
caregiving role (i.e., sense of privacy lost,
disruption of daily routines)?

1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

 

' . (30)
In the past month, how often have you felt stressed
while caring for ?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(31)
In the past month, how often have you felt that you
have enough money to care for in addition
to your other expenses?
1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(32)

97

In the past month, how often have you felt that
asks for more help than he/she really

 

needs?

1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal
(33)

In the past month, how often would you say that you
have felt that caring for is an
inconvenience? (i.e., because helping takes so much
time)

1. never 2. rarely 3. sometimes

4. quite frequently 5. always 6. refusal

(34)

F. AMOUNT OF ADDITIONAL SUPPORT

This group of questions is about those people who may or
may not be helping you with your caregiving
responsibilities. The first question is:

1.

In the past month, how many additional family members
or friends have helped to take care of

him/herself or have done things around the house for
G?

 

number of informal care provider
code 66 for refusal / code 99 for nonapplicable

(35-36)

 

In the past month, if you have had an additional
informal care provider helping you care for

, how much time did he or she spend
helping you, on the average?

 

number of minutes
code 6666 for refusal / code 9999 for nonapplicable

(37-40)

 

98

3. In the past month, how many formal service providers
(i.e., nurses, therapists, chore workers) have come to
’8 home, on the average?

 

number of formal service providers

code 66 for refusal / code 99 for nonapplicable

(41-42)

 

4. In the past month, if you have had an additional formal
care provider helping you care for , how
much time did he or she spend helping you, on the
average?

number of minutes

code 6666 for refusal / code 9999 for nonapplicable

(43-46)

 

G. DEMOGRAPHIC INFORMATION

The last group of questions is about background information
on the caregivers that we are talking to. The first
question is:

1. Would you mind telling me how old you are?

(47-48)

 

2. Sex (only if not distinguishable) 1. Male 2. Female

(49)

3. The third question concerns race. Could you tell me
which category you consider yourself in?

1. White 2. Black (50)

3. Indian 4. Hispanic 5. Other

4. Are you currently:
1. Married 2. Widowed 3. Divorced
4. Separated 5. Have never been married

(51)

99

Do you have children? 1. Yes 2. No
(52)
If yes, the number of children

(53-54)

 

For purposes of the interview, we need a rough
indication of your total income. Would you mind giving
me an estimate of your yearly income?

1. below $5,000 a year

2. $5,000 to 9,999 a year

3. $10,000 to 14,999 a year . (55)

4. $15,000 to 19,999 a year

5. $20,000 to 24,999 a year

6. $25,000 to 29,999 a year

7. Above $30,000 a year

Could you please tell me what your relationship is to
from the following categories?

 

1. Husband/Wife

2. Child

3. Sister/Brother (56)

4. Other relative

5. Neighbor/Friend

6. Paid helper

The last question regards whether or not you have a job
in addition to your being a caregiver. Could you
please tell me if you are working and if so, how many

hours per week and how long have you been working?

1. Yes 2. No (57)

100

If yes, how many hours per week (58-59)
If yes, how long have you been working?
Number of months

(60-62)

 

NOTE: Code 9’s for nonapplicable
Interviewer, code Living Status and Program Status as
follows:

Program Status:
1. Case Management
2. Information and Referral
(63)
Caregiver lives with the care-receiver:
1. Yes ____ 2. No ____

(64)

Closing the Interview

 

I would like to thank you for participating in this
interview. Are there any questions you may have that I
could answer for you at this time? Hopefully, by your
answers and the answers of other caregivers, we’ll be able
to determine what the needs and concerns of primary
caregivers are and how caring for another individual
affects the caregiver’s life.

I would like to mention to you again that your
responses to the questions that were asked are confidential
and no one but the research team will have access to them.
In addition, you also have a right to a copy of a summary

of the final report that is a product of this study. If

101

you would like a copy of this report, I will be happy to

send you a copy. YES or NO

Again,
the time you
questions or
interview by
possible and

appropriate.

I would like to say how very much I appreciate

spent with me today. Do you have any
comments at this time? If not, close the
saying, thank you again for making this study

have a nice day or evening whichever is

**** If you have not already done so, determine if the
caregiver is living with the care-receiver.

COMPLETION TIME OF INTERVIEW

 

REFERENCES

REFERENCES

Biegel, D., Shore, B., & Gordon, E. (1984). Building
gupport gupport networks for the elderly: Theory and
applicatioaa. Beverly Hills: Sage.

Bornstein, J. (1986). Coordinagion of cogaunity long7tera
care: An evaluation of case management for the frail
elderly. Unpublished doctoral dissertation, Michigan
State University.

Bowling, A. (1984). Caring for the elderly widowed-the
burden on their supporters. Britiah Journal of Socia1
Work, 11(5), 435-455.

Branch, L., & Jette, A. (1983). Elder’s use of informal
long-term care assistance. The Gerontologist, 16(1),
51—56 0

Brill, R., & Horowitz, A. (1984). The New York Home Care
Project: A demonstration in coordination of health and
social services. gage Health Care Services Quarterly,
1(3/4), 91-106.

Brocklehurst, J. C., Morris, P., Andrews, R., Richards, B.,
& Laycock, P. (1981). Social effects of stroke.
Social Science and Medicine, 16(A), 35-39.

Brody, E. M. (1981). "Women in the middle" and family help
to older people. The Gerontologist, 11(5), 471-480.

Brody, E. M. (1985). Parent care as a normative family
stress. The Gerontologist, 11, 471-480.

Brody, E. M. (1986). Institutional versus community health
care of the elderly: The delicate balance of social
policy position paper. Home Heagth Cage Servicea

Quarterly, 1(3/4), 113-129.

Brody, E. M., Kleban, M., Johnsen, P., Hoffman, C., &
Schoonover, C. (1987). Work status and parent care: A
comparison of four groups of women. The Gerontologist,
11(2), 201-208. 7

Brody, E. M., & Schoonover, C. (1986). Patterns of care

for the dependent elderly when daughters work and when
they do not. ‘The Gerontolagist, 16, 372-381.

102

103

Brotman, H. (1974). The fastest growing minority: The
aging. American Journal of Public Health, 61, 349-352.

Brown, T., & Learner, R. M. (1984). The South Carolina
Community Long Term Care Project. Home Health Care
ServiceSLQuarterly, 1(3/4), 73-89.

Callahan, J., Diamond, L., Giele, J., & Morris, R. (1980).
Responsibility of families for their severely disabled
elders. Health Cage Fiaanciag Review, 1, 29-48.

Callahan, J., & Stanley, S. (Eds.). (1981). Reforming the
long term careggystea. Lexington, MA: D. C. Heath and
Co.

Campbell, D., & Stanley, J. (1968). Experiaental and
guasi-experiaengal design for reseagch. Chicago: Rand
McNally.

Cantor, M. (1979). Neighbors and Friends: A overlooked
resource in the informal support system. Research and

Cantor, M. (1983). Strain among caregivers: A study of
experience in the United States. The Gerontologist,
16(6), 597-604.

Capitman, J. (1986). Community based long-term care
models, target groups, and impacts on service use. The
Gerontologist, 16, 389-397.

Caplan, G. (1974). Social supportgand coamunitypaental
health. New York: Basic Books.

Caplan, G., & Killilea, M. (Eds.). (1976). Support systems
and mutual help. New York: Grune & Stratton.

 

Caserta, M., Lund, D., Wright, S., & Redburn, D. (1987).
Caregivers to dementia patients: The utilization of
community services. The Gerontologist, 11(2), 209-
214.

Cassel, J. (1974). An epidemiological perspective of
psychosocial factors in disease etiology. American
Journal of Public Health, 61(11), 1040-1043.

Chatters, L., Taylor, R., & Jackson, J. (1985). Size and
composition of the informal helper networks of elderly
blacks. Journal of Gerontology, 16(5), 605-614.

Chatters, L., Taylor, R., & Jackson, J. (1986). Aged
blacks’ choices for and informal helper network.
Journal of Gerontolggy, 11(1), 94-100.

104

Chenoweth, B., & Spencer, B. (1986). Dementia: The
experience of family caregivers. The Gerontologist,
16(3), 267-272.

 

Circirelli, V. (1981). Helpigg elderly parents: The role
of adult children. Boston: Auburn House.

Clark, N., & Rakowski, W. (1983). Family caregivers of
older adults: Improving helping skills. The
Gerontolo ist, 16(6), 637-642.

Cobb, S. (1976). Social support as a moderator of life
stress. Psychosoaatic Medicine, 66, 300-314.

Crossman, L., London, C., & Barry, C. (1981). Older women
caring for disabled spouses: A model for supportive
services. The Gerontongist, 11(5), 464-470.

Dahlin, M. (1980). Perspectives on the family life of the
elderly in 1900. The Gerontologist, 16(1), 99-107.

Deimling, G., & Bass, D. (1986). Symptoms of mental
impairment among elderly adults and their effects on
family caregivers. Journal of Gerontology, 11(6), 778-
784.

Eggert, G., Brodows, B. (1984). Five years of ACCESS:
What have we learned? Home Health Cage Services

Quarterly, 1(3/4), 27-48.

Fairweather, G. W., & Tornatzky, L. G. (1977).
Experimental methods for apcial policy research. New
York: Pergamon Press.

Farkas, S. (1980). Impact of chronic illness on the
patient’s spouse. Health and Social Work, 6, 39-46.

Fengler, A., & Goodrich, N. (1979). Wives of elderly
disabled men: The hidden patients. The Gerontologist,
16, 175-183.

Fitting, M., Rabins, P., Lucas, M. J., & Eastham, J.
(1986). Caregivers for dementia patients: A

comparison of husbands and wives. The Gerontologist,
16(3), 248-252.

George, L. K., & Gwyther, L. P. (1986). Caregiver well-
being: A multidimensional examination of family

caregivers of demented adults. The Gerontologist,
16(3), 253-259.

Gibson, M. (1982). Family support for the elderly in
international perspective: Part 1. Ageing
International, 1, 12-17.

 

105

Gilleard, C., Gilleard, E., Gledhill, K., & Whittick, J.
(1984). Caring for the elderly mentally infirm at
home: A survey of the supporters. Journal of
Epidepiology and Coppunity Health, 66(4), 319-325.

Gottlieb, B. (1981). Social networks and social support.
Beverly Hills: Sage.

Holahan, C. J., & Moos, R. H. (1981). Social support and
psychological distress: A longitudinal analysis.
Journal of Abnormal Paychology, 66, 365-370.

Horowitz, A. (1985). Sons and daughters as caregivers to
older parents: Differences in role performance and
consequences. The Gerontologist, 16(6), 612-617.

Horowitz, A., Dobrof, R. (1982). The role of families in
providipg loag:term care to the frailpand chronically
ill elderly livipg in the coppunity. Final report
submitted to the Health Care Financing Administration
Department of Health and Human Services. New York, New
York.

 

House, J. S. (1981). Work stress and social support.
Reading MA: Addison-Wesley.

Jette, A., Branch, L., Wentzel, R., Carney, W., Dennis, D.,
& Heist, M. (1981). Home care service diversification:
a pilot investigation. The Gerontolo ist, 11(6), 572-579.

Johnson, C. L. (1983). Dyadic family relations and social
support. The Gerontologist, 1614), 377-383.

 

Johnson, C. L., & Catalano, D. L. (1981). Childless
elderly and their family supports. The Gerontologist,
11(6), 610-618.

Jones, D., & Vetter, N. (1984). A survey of those who care
for the elderly at home: Their problems and their
needs. Social Science and Medicine, 16(5), 511-514.

Kane, R. L., & Kane, R. A. (1981). A guide through the
maze of long term care. Western Journal of Madicine,

Kastenbaum, R. (1969). The foreshortened life perspective.
Geriatrics, 11, 126-133.

 

Kimmel, D. C. (1980). Adulthood and Aging. New York:
John Wiley & Sons.

Kodner, D., Mossey, W., & Dapello, R. (1984). New York’s
"Nursing Home Without Walls": A provider-based
community care program for the elderly. Home Health
Care Services Quarterly, 1(3/4), 107-126.

106

Kohen, J. A. (1983). Old but not alone: Informal social
supports among the elderly by marital status and sex.
The Gerontologist, 16(1), 57-63.

Kraus, A. S. (1984). The burden of care for families of
elderly persons with dementia. Canadian Jouraal of

Kulys, R., & Tobin, S. (1980). Older people and their
’responsible others’. Social Work, 16, 138-145.

Lasch, C. (1977). Haven in a heartleaa world: The family
besieged. New York: Basic Books.

Leavy, R. (1983). Social support and psychological
disorder: A review. Journal of Copmunity Paychology,

 

 

_1_0.(4). 3-21-

Lowenthal, M., Berkman, P., & Associates. Aging and pental
disorder in San Francisco. San Francisco: Jossey-
Bass.

Lowenthal, M., Haven, C. (1968). Interaction and
adaptation: Intimacy as a critical variable. In B.
Neugarten (Eds.), Middlepage and aging. Chicago:
University of Chicago Press.

Luke, E., Norton, W., & Denbigh, K. (1981). Medical and
social factors associated with psychological distress

in a sample of community aged. Canadian Journal of
Psychiatry, 16(4), 244-250.

Manton, K. G., Blazar, D., & Woodbury, M. (1987). Suicide
in middle age and later life: Sex and race specific
life table and cohort analyses. Journal of
Gerontology, 11(2), 219-227.

 

Marcus, L., & Jaeger, V. (1984). The elderly as family
caregivers. Canadian Journa1 of Aging, 6(1), 33-43.

Mcauley, W., Jacobs, M., & Carr, C. (1984). Older couples:
Patterns of assistance and support. Journal of
Gerontological Social Work, 6(4), 35-48.

Miller, D. B., Gulle, N., & McCue, F. (1986). The
realities of respite for families, clients, and
sponsors. The Gerontologist, 16, 467-470.

Miller, G. T. (1979). Livipg in the Environpent, Second
Edition. Belmont, CA: Wadsworth Publishing Co.

 

Minkler, M. (1981). Applications of social support theory
to health education: Implication for work with the
elderly. Health Education:Quarterly, 6(2), 147-165.

107

Perspective on Aging. (1985). State support for respite
care: Report of an exploratory survey. March/April.

Phillips, S. L. (1981). Network characteristics related to
the well-being of normals: A comparative base.
Schizophrenia Bulletin, 1, 117-124.

Pilisuk, M., & Parks, S. (1981). The place of network
analysis in the study of supportive social

associations. 6asic and Applied Social Ps cholo , 1,
121-135.

Porritt, D. (1979). Social support in crisis: Quantity or
quality? Social Sciapceiand Medicina, 16(A), 715-721.

Poulshock, S. W., & Deimling, G. T. (1984). Families

 

caring for elders in residence: Issues in the
measurement of burden. Journal of Gerontology, 66(2),
230-239. ‘

Quinn, J., & Hodgson, J. (1984). Triage: A long term care
study. Home Hea1th Caregfiarvices Quarterly, 1(3/4),
171-191.

Rabkin, J., & Streuning, E. (1976). Life events, stress,
and illness. Science, 194, 1013-1020.

Rakowski, W., & Clark, N. (1985). Future outlook,
caregiving, and care-receiving in the family context.
The Gerontolo ist, 16(6), 618-623.

 

Robinson, B., & Thurnher, M. (1979). Taking care of aged
parents: A family cycle transition. The
Gerontologist, 16, 586-593.

 

Roy, A. (1978). Vulnerability factors and depression in
women. British Journal of Psych1atry, 133, 106-110.

Scharlach, A. E. (1987). Relieving feelings of strain
among women with elderly mothers. Psychology and
Aging, 1(1), 9-13.

Schneider, E. L. (1983). Teaching nursing homes. New
England Journal of Medicine, 308, 336-342.

Scott, J. P., Roberto, K. A. (1985). Use of informal and
formal support networks by rural elderly poor. The
Gerontologist, 16(6), 624-630.

Shanas, E. (1979). The family as a social support system
in old age. The Gerontologist, 16, 169-174.

 

108

Shore, E., & Maddox, G. (1976). Aging, health and the
Organization of health resources. In R. H. Binstock &
E. Shanas (Eds.), flandbook of Aging and the Social
Sciences. New York: Van Nostrand, Reinholt Co.

Skellie, A., Favor, F., Tudor, C., & Strauss, R. (1984).
The Georgia Alternative Health Services Project: Cost-
effectiveness depends on population selection. Home
Health Care Services Quarterly, 1(3/4), 49-72.

Smith, L., & Bengtson, V. (1979). Positive congruence of
institutionalization: Solidarity between elderly
parents and their middle-aged children. 166
Gerontologist, 12» 438-447.

 

Snow D., & Gordon, J. (1980). Social network analysis and
intervention with the elderly. The Gerontologist, 16,
463-467.

Solomon, R. (1982). Gerontological social work practice in
long-term care: Serving families of the
institutionalized aged: The four crises. Journal of

 

Gerontological Social Work, 6(1-2), 83-96.
Stack, C. (1974). All our kin. New York: Harper & Row.

Steinberg, R., & Carter,.G. (1983). Case Management and
the elderly. Lexington, MA: D. C. Heath and Co.

 

Stoller, E. P. (1982). Sources of support for the elderly
during illness. Health and Social Work, 1, 111-122.

Stoller, E. P. (1983). Parental caregiving by adult
children. Journal of Marriage and Fapily, 16, 851-858.

Stoller, E. P., Earl, L. (1983). Help with activities of
everyday life: Sources of support for the
noninstitutionalized elderly. The Gerontologist,
16(1), 64-70.

Strong, C. (1984). Stress and caring for elderly
relatives: Interpretations and coping strategies in an
American Indian and White sample. The Gerontologist,
11(3), 251-256.

Sussman, M. (1976). The family life of old people. In R.
Binstock & E. Shanas (Eds.), Handbook of aging and the
social sciences. New York: Van Nostrand, Reinholt Co.

 

Tabachnick, B., & Fidell, L. (1983). UsingaMultivariate
Statistics. New York: Harper & Row.

 

Thoits, P. A. (1982). Conceptual, methodological, and
theoretical problems in studying social support as a
buffer against life stress. Journal of Health and
Social Behavior, 16, 145-159.

 

109

Tolsdorf, C. C. (1976). Social networks, support, and
coping: An exploratory study. Family Process, 16,

Townsend, A., & Poulshock, W. (1986). Intergenerational
perspectives on impaired elders’ support networks.
Journal of Gerontology, 11(1), 101-109.

Tress, J. (1977). Family support systems for the aged:
Some social and demographic considerations. The
Gerontologist, 11, 486-491.

U. S. Office of Management and Budget. (1980). Report of
the interagency statiatigaj coppittea on long;term care
for the elderly: Data coverage for the functionally
11aited elderly. Washington, D. 0.: Human Resources,
Veterans, and Labor Special Studies Division.

 

Wachtel, T. J., Fulton, J. P., & Goldfarb, J. (1987).
Early prediction of discharge disposition after
hospitalization. The Gerontologist, 11(1), 98-103.

Ward, R. (1978). Limitations of the family as a supportive
institution in the lives of the aged. The Family
Coordinator, 11, 365-373.

 

Ward, R. (1984). The aging experience. New York: Harper
& Row.

Weiss, L., & Sklar, B. (1984). Project OPEN: A hospital-
based long-term care demonstration program for the
chronically ill elderly. Home Health Gaga Services

Quarterly, 1(3/4), 127-145.

Wellman, B. (1981). Applying network analysis to the study
of support. In B. Gottlieb (Eds.), Soc1al networks and
social support. Beverly Hills: Sage.

Williams, C. (1985). Teaching nursing homes: Their impact
on public policy, patient care, and medical education.
Journal of the Aperican Geriatrica1§ociety, 66(3), 189-
195.

Wingard, D. L., Jones, D. W., & Kaplan, R. M. (1987).
Institutional care utilization by the elderly: A
critical review. The Gerontologist, 11(2), 156-163.

York, J., & Calsyn, R. (1977). Family involvement in
nursing homes. The Gerontologist, 11, 500-505.

Zarit, S. H., Orr, N. K., & Zarit, J. M. (1985). The
hidden victims of Alzheiper’a diaease: Fapjlies under
stress. New York: New York University Press.

110

Zarit, S. H., Reever, K., & Peterson, J. (1980). Relatives
of the impaired elderly: Correlates of feelings of
burden. The Gerontologist, 16(6), 649-655.

zarit, So Ho. TOdd, PO A., & zarit, J. M. (1986).
Subjective burden of husbands and wives as caregivers:
A longitudinal study. The Gerontologist, 16(3), 260-
266.

 

 

   

"777777777777